Datasets:

bzantium

/

LongPerplexity

like 0

the network - layer performance analysis seeks to provide estimates on the delays experienced by traffic traversing the elements of a network , as well as the corresponding buffer requirements . for wireless networks , a question of interest is how the stochastic properties of wireless channels impact delay and backlog performance . wireless channels are characterized by rapid variation of channel quality caused by the mobility and location of communicating devices . this is due to _ fading _ , which is the deviation in the attenuation experienced by the transmitted signal when traversing a wireless channel . the term _ fading channel _ is used to refer to a channel that experiences such effects . in this paper we explore the network - layer performance of a multi - hop network where each link is represented by a fading channel . we model the wireless network by tandem queues with variable capacity servers , where each server expresses the random capacity of a fading channel . we ignore the impact of coding by assuming that transmission rates over the fading channels are equal to their information - theoretic capacity limit , @xmath0 , which is expressed as a function of the instantaneous signal - to - noise ratio ( snr ) at the receiver , @xmath1 , by @xmath2 , where @xmath3 is the channel bandwidth ( in _ hz _ ) . numerous models are available to describe the gain of fading channels depending on the type of fading ( slow or fast ) , and the environment ( e.g. , urban or rural ) . the instantaneous , information - theoretic channel capacity of a fading channel can be represented as the logarithm of @xmath1 by ( see chp . 14.2 in @xcite ) @xmath4 where @xmath5 is a constant and the function @xmath6 is used to characterize the fading channel . we are interested in finding bounds on the end - to - end delay and on buffer requirements for a cascade of fading channels , with store - and - forward processing at each channel . the analysis in this paper takes a system - theoretic stochastic network calculus approach @xcite , which describes the network properties using a @xmath7 dioid algebra . arrival and departure processes at a network element are described by bivariate stochastic processes @xmath8 and @xmath9 , respectively , denoting the cumulative arrivals ( departures ) in the time interval @xmath10 . a network element is characterized by the service process @xmath11 , denoting the available service in @xmath10 . the input - output relationship at the network element is governed by @xmath12 where the convolution operator ` @xmath13 ' is defined as @xmath14 . if network traffic passes through a tandem of @xmath15 network elements with service processes @xmath16 , the service of the network as a whole can be expressed by the convolution @xmath17 . the stochastic properties of fading channels present a formidable challenge for a network - layer analysis since the service processes corresponding to the channel capacity of common fading channel models such as rician , rayleigh , or nakagami-@xmath18 , require to take a logarithm of their distributions . as discussed in the next section , researchers frequently turn to higher - layer abstractions to model fading channels . widely used abstraction are the two - state gilbert - elliott model and its extensions to finite - state markov channels ( fsmc ) @xcite . fsmc models simplify the analysis to a degree that the network model becomes tractable , at least at a single node . extensions to multi - hop settings encounter a rapidly growing state space . as of today , a general multihop analysis that is applicable to models of fading channels , such as rician , rayleigh , or nakagami-@xmath18 , remains open . in this paper , we pursue a novel approach to the analysis of multi - hop wireless networks . we develop a calculus for wireless networks that can be applied to fading channel models from the wireless communication literature to provide network - layer performance bounds . we view the network - layer model with arrival , departure and service processes as residing in a _ bit domain _ , where traffic and service is measured in bits . we view the fading channel models used in wireless communications as residing in an alternate domain , which we call the _ snr domain _ , where channel properties are expressed in terms of the distribution of the signal - to - noise ratio at the receiver . we then derive a method to compute performance bounds from these traffic and service characterizations . a key observation in our work is that service elements in the snr domain obey the laws of a dioid algebra . we devise a suitable dioid , referred to as _ algebra _ , where the minimum takes the role of the standard addition , and the second operation is the usual multiplication , and use it for analysis in the snr domain . in this domain multi - hop descriptions of fading channels become tractable . in particular , we find that a cascade of fading channels can be expressed in terms of a convolution in the new algebra of the constituting channels . the key to our analysis is that we derive performance bounds entirely in the snr domain . observing that the bit and snr domains are linked by the exponential function , we transfer arrival and departure processes from the bit to the snr domain . then , we derive backlog and delay bounds in the transfer domain using the algebra . the results are mapped back to the original bit domain to finally give us the desired performance bounds . our derivations in the snr domain require the computation of products and quotients of random variables . here , we take advantage of the _ mellin transform _ to facilitate otherwise cumbersome calculations . then , the computational problem is reduced to finding the mellin transform for service and traffic processes . the main contribution of this paper is the development of a framework for studying the impact of channel gain models on the network - layer performance of wireless networks . for the purposes of this paper , the snr domain is used solely as a transfer domain that enables us to solve an otherwise intractable mathematical problem . on the other hand , the ability to map quantities that appear in network - layer models and concepts found in a physical - layer analysis may prove useful in a broader context , e.g. , for studying cross - layer performance issues in wireless communications . moreover , the algebra and the mellin transform form a tool set that can be applied more generally in wireless communications for studying the channel gain of cascades of fading channels . as the first paper on the network calculus algebra , our paper only considers simple network scenarios and makes numerous convenient assumptions ( which are made explicit in sec . [ sec : model ] ) . there is room for significant future work on extensions of the model and a relaxation of the presented assumptions . the remainder of the paper is organized as follows . in sec . [ sec : related ] we discuss related work . we describe the system model in sec . [ sec : model ] , where we also motivate the use of the snr domain . in sec . [ sec : calculus ] we present the algebra and derive performance bounds . in sec . [ sec : rayleigh ] we apply the analysis to a cascade of rayleigh channels , and present numerical examples . in sec . [ sec : rayleigh ] , we investigate a network with cross traffic at each node . we discuss brief conclusions in sec . [ sec : conclusions ] . analytical approaches for network - layer performance analysis of wireless networks include queueing theory , effective bandwidth and , more recently , network calculus . since the service processes corresponding to the channel capacity of common fading channel models such as rician , rayleigh , or nakagami-@xmath18 , require to take a logarithm of their distributions , researchers often turn to higher - layer abstractions to model fading channels , which lend themselves more easily to an analysis . a widely used abstraction is the two - state channel model developed by gilbert @xcite and elliott @xcite , and subsequent extensions to a finite - state markov channel ( fsmc ) @xcite . markov channel models are well suited to express the time correlation of fading channel samples . we refer to @xcite for a survey of the development and applications of fsmc models . zorzi et al . @xcite evaluated the accuracy of first - order markov channel models of fading channels , where the next channel sample depends only on the current state of the markov process , and higher order processes that can capture memory extending further back in the process history . the authors found that a first - order markov model is a good approximation of the fading channel , and that using higher order markov processes does not significantly improve the accuracy of the model . queue - based channel ( qbc ) @xcite is an alternative model for fading channels , which models a binary additive noise channel with memory based on a finite queue . here , a queue with size @xmath19 contains the last @xmath19 noise symbols , and the noise process is an @xmath20-order markov chain . the model was found to provide a better approximation to the rayleigh and rician slow fading channels compared to the gilbert - elliot model @xcite . an extension of the qbc model , called weighted qbc @xcite permit queue cells ( i.e. , channel samples ) to contribute with different weights to the noise process . queueing theoretic studies of fading channels generally apply approximations to reduce the complexity of multi - hop models . le , nguyen , and hossain @xcite apply a decomposition approximation to analyze the loss probability and average delay of a multi - hop wireless network with slotted transmissions for a batch bernoulli arrival process , and with independent cross traffic at each node . the wireless link is assumed to employ adaptive modulation and coding with multiple modes , where each mode corresponds to a given link rate , which is chosen based on the snr of the channel . the channel state of a link is assumed to be stationary , and channel states in consecutive time slots are independent . another decomposition approximation is presented by le and hossain @xcite , who consider a multi - hop tandem network with a batch arrival process and multi - rate transmissions , to develop a routing scheme that can meet given delay and loss requirements . the analysis obtains end - to - end loss rates and delays with a decomposition analysis , and feeds the results as metrics to the routing algorithm . bisnik and abouzeid @xcite model a multi - hop wireless network as an open network of g / g/1 queueing systems . using diffusion approximation , they obtain closed - form expressions for average end - to - end delays . ishizaki and hwang @xcite studied the impact of multiuser diversity assisted packet scheduling on the packet delay performance in a wireless network with nakagami - m fading channels . the network is modeled by a multi - queue system , where each channel is described by an fsmc . under assumptions of stationarity , homogeneity , and independence of the channel processes , they approximate the tail distribution of packet delays , and compare it to that of a round - robin scheduler . the results indicate that the delay performance of multiuser diversity assisted scheduling algorithms is not necessarily superior to that of round - robin scheduling . since the application of queuing theoretical methods to study cascade of fading channels requires many assumptions on arrival and service distributions , and simplifications of the model , the use of classical queueing theoretic methods for the performance analysis of multi - hop wireless networks has been put into question @xcite . an effective bandwidth @xcite analysis seeks to develop ( asymptotic ) bounds on performance metrics , e.g. , an exponential decay of the backlog . wu and negi @xcite have adapted an effective bandwidth analysis to the analysis of fading channels . they introduce the concept of effective capacity , which characterizes a wireless channel by a log - moment generating function ( log - mgf ) of the channel capacity . they obtain an asymptotic approximation of the delay bound violation probability of a rayleigh fading channel . due to the difficulty of computing the moment generating function ( mgf ) of the rayleigh distribution , they assume non - correlated distributions with low snr and estimate channel parameters from measurements . the work has been extended to correlated rayleigh and correlated nakagami-@xmath18 channels , and to cascades of fading channels @xcite . a closely related concept is the effective channel capacity presented by li et al . @xcite , which describes the available channel capacity by a first order markov chain and computes the log - mgf of the underlying markov process . taking advantage of methods developed in @xcite , they compute statistical delay bounds for nakagami - m fading channel . hassan , krunz , and matta @xcite use an effective bandwidth analysis to study delay and loss performance at a single wireless link , which is modeled by an fsmc . for fluid on - off traffic and fifo buffering , they obtain a closed form expression for the effective bandwidth required to guarantee bounds on delay and packet loss . there is a collection of recent works that apply stochastic network calculus methods @xcite to wireless networks with fading channels . the stochastic network calculus is closely related to the effective bandwidth theory , in that it seeks to develop bounds on performance metrics under assumptions also found in the effective bandwidth literature . different from effective bandwidth literature , stochastic network calculus methods seek to develop non - asymptotic bounds . an attractive element of a network calculus analysis is that sometimes it is possible to extend a single node analysis to a tandem of nodes , using the convolution operation seen in the introduction . fidler @xcite presents a network calculus methodology for a two - state fsmc model of a single - hop fading channel . he applies the mgf network calculus , which was suggested in the problem sets of chapter 7 in @xcite , and which has been developed in @xcite . the mgf network calculus takes its name from the extensive use of moment generating functions in the derivation of performance bounds . mahmood , rizk , and jiang @xcite apply the mgf network calculus to mimo channels and derive delay bounds for periodic traffic sources . @xcite also use an mgf network calculus to study the performance of two - hop relay networks . a similar methodology is applied by mahmod , vehkaper , and jiang @xcite to compute the throughput of a multi - user ds - cdma system with delay constraints . in the works above that apply the mgf network calculus , models for a cascade of multiple fading channels become complex , so that multi - node results for networks with more than two nodes have not been obtained . the network calculus developed in this paper uses similar descriptions and assumptions for traffic and service as the mgf network calculus . by performing computations in a transfer domain , where fading channel models take a simpler form , we are able to compute multi - node service descriptions for an arbitrarily large number of nodes . the mgf network calculus assumes that arrivals and service at each node are independent . these assumptions can be relaxed using statistical envelope descriptions for traffic ( effective envelopes ) and service ( statistical service curve ) @xcite . jiang and emstad @xcite have applied an approach with envelopes to a fading channel where the wireless channel is characterized by two stochastic processes : an ideal service process and an impairment process , where the impairment process captures effects due to fading , noise , and cross traffic . verticale and giacomazzi @xcite obtain a closed form expression for the variance of a service curve , which describes the available service by a markov chain . this is used for the analysis of an fsmc model of a rayleigh fading channel . for computing the bounds for markovian arrivals , they apply the bounded - variance network calculus introduced in @xcite , which is an extension of the central limit theorem methods by choe and shroff @xcite to multi - hop paths . verticale @xcite has applied the same methodology to constant bit rate traffic . ciucu , pan , and hohlfeld @xcite and ciucu @xcite present non - asymptotic ( i.e. , finite number of hops ) , closed - form expressions for the delay and throughput distributions for multi - hop wireless networks . as many of the works discussed above , the fading channel is modeled by an abstraction that uses a link layer model of the transmission channel . here , the channel is assumed to be governed by a slotted - aloha system in half - duplex mode . the model of this channel is a two - state on - off server , where a node can transmit ( i.e. , is in the on state ) only when other all nodes in the interference range are not transmitting . there is also a literature on physical - layer performance metrics of fading channels in multi - hop wireless networks . hasna and alouini @xcite have presented a framework for evaluating the end - to - end outage probability of a multi - hop wireless relay network with independent , non - regenerative relays , i.e. , amplify - and - forward ( af ) , over nakagami fading channels . they provide a closed - form expression for the mgf of the reciprocal of the equivalent end - to - end snr for independent nakagami fading channels . for the same af relay network , tsiftsis @xcite obtains a closed - form bound for the average error probability . this bound is reportedly tight at low snr , but may become loose for higher snr values and for more severe fading environment , e.g. , rayleigh fading . amarasuriya , tellambura , and ardakani @xcite present an alternative bound to @xcite on the end - to - end snr in multi - hop af relay networks . they derive the distribution function and the mgf for i.i.d . nakagami - m fading and for independent , but non - identically distributed rayleigh fading . the works above study physical layer performance bounds of channel - assisted , amplify - and - forward relaying over a multi - hop fading channels . they do not consider buffering or traffic burstiness , and are not concerned with network performance metrics addressed in this paper . delay and backlog analysis and optimization of multihop wireless networks remains an open research problem @xcite . we consider a wireless @xmath15-node tandem network as shown in fig . [ fig : tandem ] , where each node is modeled by a server with an infinite buffer . we are interested in the performance experienced by a ( through ) flow that traverses the entire network and may encounter cross traffic at each node . one can think of the cross traffic at a node as the aggregate of all traffic traversing the node that does not belong to the through flow . the service given to the through flow at a node is a random process , which is governed by the instantaneous channel capacity as well as the cross traffic at the node . we consider a fluid - flow traffic model where the flow is infinitely divisible . we will work in a discrete - time domain @xmath21 , where @xmath22 is the set of integers and @xmath23 is length of the time unit . setting @xmath24 allows us to replace @xmath25 by @xmath26 , which we interpret as the index of a time slot . we assume that the system is started with empty queues at time @xmath27 . different nodes and different traffic flows will be distinguished by subscripts . the cumulative arrivals to , the service offered by , and the departures from the node are represented by random processes @xmath28 , @xmath29 , and @xmath30 that will be described more precisely below , with @xmath31 for @xmath32 . we denote by @xmath33 and @xmath34 the arrivals to and the departures from the tandem network . throughout , we assume that arrival and service processes satisfy stationary bounds . consider for the moment a single node . dropping subscripts , we write @xmath35 for the cumulative arrivals and departures , respectively , at the node in the time interval @xmath36 , where @xmath37 denotes the arrivals and @xmath38 the departures in the @xmath26-th time slot . due to causality , we have @xmath39 . the processes lie in the set @xmath40 of non - negative bivariate functions @xmath41 that are increasing in the second argument and vanish unless @xmath42 . the backlog at time @xmath43 is given by @xmath44 and the delay at the node is given by @xmath45 the service of the node in the time interval @xmath36 is given by a random process @xmath46 , such that eq . holds for every arrival process @xmath47 and the corresponding departure process @xmath48 . this service description with bivariate functions is referred to as _ dynamic server_. initially defined for non - random service @xcite , dynamic servers have been extended to random processes in @xcite . the above model is a typical network - layer model . where traffic is measured in bits , and service is measured in bits per second . we thus refer to this model of arrivals , departures , and service as residing in a _ bit domain_. the network calculus exploits that networks which satisfy the input - output relation of eq . with equality can be viewed as linear systems in a dioid algebra @xcite . in the ( @xmath49 ) dioid , the minimum and addition take the place of the standard addition and multiplication operations . the network calculus is based on the fact that @xmath50 is again a dioid @xcite . note that the min - plus convolution , which provides the second operation in the dioid , is not commutative in @xmath40 . to compute a service model for a wireless channel , we assume that the channel state information is sampled at equal time intervals @xmath23 . with @xmath51 , let @xmath52 denote the instantaneous signal - to - noise ratio observed at the receiver in the @xmath26-th sampling epoch . then , @xmath52 is a nonnegative random variable that has the probability distribution of the underlying fading model . we assume that the random variables @xmath52 are independent and identically distributed . this assumption is justified when @xmath23 is longer than the channel coherence time . otherwise , the assumption will give optimistic bounds . we emphasize that the network calculus in this paper applies to settings without independence , however , the derivation of performance bounds will proceed differently . using eq . , the instantaneous service offered by the channel in the @xmath26-th slot is given by @xmath53 . and the corresponding service process is given by @xmath54 where we haven chosen units such that the constant in eq . takes the value @xmath55 . the service description in eq . requires us to work with the logarithm of fading distributions , which presents a non - trivial technical difficulty via the usual network calculus or queueing theory . on the other hand , observe that the exponential @xmath56 is described more simply by @xmath57 this motivates the development of a system model that allows us to exploit the more tractable service representation in eq . . in this alternative model , arrivals , departures , and service reside in a different domain , where we can work directly with the distribution functions of the fading channel gain and the corresponding snr at the receiver . we now proceed by mapping the network model from fig . [ fig : tandem ] into a transfer domain , which we refer to as _ snr domain_. we will seek to derive performance bounds in the transfer domain , and then map the results to the bit domain to obtain network - layer bounds for backlog and delays . the relationship of the network models in bit domain and snr domain is illustrated in fig . [ fig : domains ] . in the previous subsection , we constructed the service process for a wireless link in the snr domain in eq . as @xmath58 by analogy , we describe the arrivals and departures in the snr domain by @xmath59 throughout this paper , we use calligraphic upper - case letters to represent processes that characterize traffic or service as a function of the instantaneous snr in the sense of eq . . due to the monotonicity of the exponential function , @xmath60 and @xmath61 are increasing in @xmath62 , and satisfy the causality property @xmath63 . the backlog process is accordingly described by @xmath64 since time is not affected by this transformation , the delay is given by @xmath65 to interpret these processes in the transfer domain , let @xmath66 be the instantaneous channel snr required to transmit @xmath37 in a single time slot , assuming transmission at the rate of the capacity limit . the arrival process in the snr domain can then be expressed in terms of these variables as @xmath67 here , we are treating channel quality expressed in terms of the instantaneous snr as a commodity . an arrival in a time unit represents a workload , where @xmath68 expresses the amount of resources that will be consumed by the workload . the backlog can similarly be expressed in terms of the instantaneous snr as @xmath69 with the interpretation that a node with backlog @xmath70 at time @xmath62 requires full use of the channel capacity for @xmath71 time units to clear the backlog . most importantly , the concept of the dynamic server translates to the snr domain . in a network system , the service process in the bit domain satisfies eq . if and only if the process @xmath72 in the snr domain satisfies @xmath73 we refer to a network element that satisfies eq . for any sample path as _ dynamic snr server_. in this general setting , we not require that @xmath74 takes the form in eq . , in particular , @xmath75 does not be equal to @xmath76 . traffic aggregation in the snr domain is expressed in terms of a product . when @xmath19 flows have arrivals at a node with arrival processes denoted by @xmath77 , then the total arrival , @xmath78 , are given for any @xmath79 by @xmath80 if we let @xmath81 and @xmath82 denote the corresponding processes in the snr domain , we see that @xmath83 with the above definitions , the usual network description by a dioid algebra in the bit domain can be expressed in the snr domain by a dioid algebra on @xmath40 where the second operator is a multiplication . this enables the development of the network calculus in sec . [ sec : calculus ] . we observe that the exponential function defines a one - to - one correspondence between arrival and departure processes in the bit and snr domains . the physical arrival , departure , service , and backlog processes can be recovered from their counterparts in the snr domain by taking a logarithm ( see fig . [ fig : domains ] ) . this section contains our main contribution : an analytical framework for statistical end - to - end performance bounds for a network , where service is expressed in terms of fading distributions residing in the snr domain . by an _ snr process _ we mean a bivariate process @xmath84 taking values in @xmath85 that is increasing in the second argument , with @xmath86 for all @xmath62 . the space of snr processes will be denoted by @xmath87 . for any pair of snr processes @xmath84 and @xmath88 , set @xmath89 and @xmath90 we refer to ` @xmath91 ' and ` @xmath92 ' as the the convolution and deconvolution operators , respectively . the arrival , departure , and service processes constructed in the previous section are snr processes . with the convolution , we can express the defining property of a dynamic snr server from eq . as @xmath93 for every pair of snr arrival and departure processes @xmath94 and @xmath95 . we note that , in fact , for any system description in the bit domain by the ( @xmath49 ) and the ( @xmath96 ) dioid algebras there exists a corresponding characterization in the snr domain using ( @xmath97 ) and ( @xmath98 ) dioids . the following argument confirm that the properties of a dioid are satisfied . [ lm : mx - dioid ] @xmath99 is a dioid . we show that ( @xmath97 ) satisfies the dioid axioms . for @xmath100 : 1 . _ commutativity of @xmath101 : _ @xmath102 . associativity of @xmath101 : _ @xmath103 . idempotency of @xmath101 : _ @xmath104 . associativity of @xmath105 : _ @xmath106 . _ distributivity of @xmath105 : _ @xmath107 . existence of a null element : _ the null element is @xmath108 . since @xmath109 . _ absorption of the null element : _ _ existence of a unity element : _ the unit of multiplication is 1 , since @xmath111 . @xmath112 is a dioid . given bivariate functions @xmath113 . since the @xmath101 operation is a pointwise minimum , properties of the @xmath101 operation , that is , properties ( 1)(3 ) from the proof of lemma [ lm : mx - dioid ] , follow from the @xmath99 dioid . for the remaining properties we have * _ associativity of @xmath91 : _ @xmath114 * _ distributivity of @xmath91 over finite sums : _ @xmath115 * _ existence of a null element : _ the null element is @xmath116 for all values of @xmath117 and @xmath62 . * _ absorption of the null element : _ @xmath118 note that functions in @xmath87 are strictly positive by definition . * _ existence of a unity element : _ the unity element is @xmath119 , where @xmath120 this gives @xmath121 the existing network calculus in the bit domain allows for the concatenation of tandem service elements using the convolution ( see page 1 ) . as an immediate consequence , single node performance bounds are extended to a multi - hop setting . we now show establish the corresponding result in the network calculus . specifically , the concatenation of dynamic snr servers is again a dynamic snr server . we will prove the result for a tandem network of two nodes , as shown in fig . [ fig : concat ] . [ thm : conc ] let @xmath122 and @xmath123 be two dynamic snr servers in tandem as shown in fig . [ fig : concat ] . then , the service offered by the tandem of nodes is given by the dynamic snr server @xmath124 with @xmath125 using eq . , the departure process @xmath126 can be written as @xmath127 \end{aligned}\ ] ] the extension to networks with more than two nodes follows by iteratively applying lemma [ thm : conc ] . hence , the dynamic network snr server with @xmath15 nodes in tandem is given by @xmath128 performance bounds in the network calculus are computed with the deconvolution operator . this is analogous to role of the deconvolution in the existing network calculus . the bounds are expressed in the following lemma . [ lem : obbound ] given a system with snr arrival process @xmath94 and dynamic snr server @xmath75 . * output burstiness . the snr departure process is bounded by @xmath129 . * backlog bound . the snr backlog process is bounded by @xmath130 . * delay bound . the delay process is bounded by @xmath131 . for the output bound , we fix @xmath117 and @xmath62 with @xmath132 and derive @xmath133 where we used the inequality @xmath134 in the second line . for any fixed sample path , fix an arbitrary @xmath135 . the bound on the backlog is derived by @xmath136 where we used @xmath137 in the second step . recall that the delay is invariant under the transform of domains , that is , @xmath138 . by definition of the delay in eq . , a delay bound @xmath139 satisfies @xmath140 where we used the inequality @xmath141 in the second line . with an algebraic description for network performance bounds in the snr domain in hand , we now turn to the problem of computing the bounds . the concise ( and familiar ) expressions from the previous section for the network service and performance bounds in the snr domain hide the difficulty of computing the expressions . in fact , all expressions of the network calculus contain products or quotients of random variables . the _ mellin transform _ @xcite facilitates such computations , particularly when the arrival and service processes are independent . the mellin transform of a nonnegative random variable @xmath142 is defined by @xmath143\ ] ] for any complex number @xmath144 such that this expected value exists . among its many properties , we will exploit that the mellin transform of a product of two independent random variables @xmath142 and @xmath145 equals the product of their mellin transforms , @xmath146 similarly , the mellin transform of the quotient of independent random variables is given by @xmath147 we will evaluate the mellin transform only for @xmath148 , where it is always well - defined ( though it may take the value @xmath108 ) . when @xmath149 , the mellin transform is order - preserving , i.e. , for any pair of random variables @xmath150 with @xmath151 we have @xmath152 for all @xmath144 . when @xmath153 , the order is reversed . hence bounds on the distribution of a random variable @xmath142 generally imply bounds on its mellin transform . a more subtle question is how to obtain bounds on the distribution of a random variable from its mellin transform . here , the complex inversion formula is not helpful . instead , we will use the moment bounds @xmath154 for all @xmath155 and @xmath156 . for bivariate random processes @xmath84 , we will write @xmath157 . in our calculus , we work with the mellin transform of convolutions and deconvolutions , which not only involves products and quotients , but also requires to compute infimums and supremums . the computation of the exact mellin transform for these operations is generally not feasible . we therefore resort to bounds , as specified in the next lemma . [ lem : mtconvdecon ] let @xmath84 and @xmath88 be two independent nonnegative bivariate random processes . for @xmath153 , the mellin transform of the convolution @xmath158 is bounded by @xmath159 for @xmath149 , the mellin transform of the deconvolution @xmath160 is bounded by @xmath161 note that the function @xmath162 is increasing for @xmath149 and decreasing for @xmath153 . for @xmath153 , the convolution is estimated by @xmath163\\ & \hspace{-2 cm } = e \bigl [ \ \sup_{\tau\le u \le t } \ { ( \x(\tau , u))^{s-1 } \cdot ( \y ( u , t))^{s-1 } \ } \bigr]\\ & \hspace{-2 cm } \le \sum_{u=\tau}^{t } e \big [ ( \x(\tau , u ) ) ^{s-1 } \big ] \cdot e \big [ ( \y ( u , t ) ) ^{s-1 } \big]\,.\end{aligned}\ ] ] in the last line , we have used the non - negativity of @xmath164 and @xmath165 to replace the supremum with a sum , and their independence to evaluate the expectation of the products . . follows by inserting the definition of the mellin transform . the deconvolution is similarly estimated for @xmath149 by @xmath166 \\ & \hspace{-2 cm } = e \bigl [ \ \sup_{0 \le u \le \tau } \bigl\ { \bigl ( \x ( u , t)\bigr)^{s-1 } \cdot \bigl(\y ( u,\tau ) \bigr)^{1-s } \bigr\ } \bigr]\\ & \hspace{-2 cm } \le \sum_{u=0}^{\tau } e \bigl [ \bigl ( \x ( u , t)\bigr)^{s-1}\bigr ] \cdot e\bigl [ \bigl(\y ( u,\tau)\bigr)^{1-s}\bigr]\,,\end{aligned}\ ] ] and eq . follows from the definition of the mellin transform . as an application of lemmas [ thm : conc ] and [ lem : mtconvdecon ] , we compute a bound on the mellin transform of the service process for a cascade of fading channels . we also make the idealizing assumption that the channels are independent . [ cor : cascade ] consider a cascade of @xmath15 independent , identically distributed fading channels , where the service process for each channel is given by eq . , with i.i.d . random variables @xmath52 . let @xmath124 denote the snr service process for the entire cascade . then , the mellin transform for this process satisfies @xmath167 for all @xmath153 . we use the server concatenation formula in eq . to represent the service of the cascade as @xmath168 . by lemma [ lem : mtconvdecon ] , its mellin transform satisfies for @xmath153 @xmath169 where the sum runs over all sequences @xmath170 with @xmath171 and @xmath172 . the assumptions on the service processes of the individual channels imply that each product evaluates to the same function @xmath173 where @xmath1 is a random variable that has the same distribution as the @xmath52 . since the number of summands in eq . is given by @xmath174 , the claim follows . we next obtain network - level performance bounds for the bit domain . this involves a transformation from the snr domain to the bit domain via the relationship in fig . [ fig : domains ] , which provides the translation of the abstract metrics @xmath175 and @xmath176 into processes @xmath48 and @xmath177 , which , along with @xmath3 , are concrete measures for traffic burstiness , buffer requirements , and delay . [ thm : blbound ] given a system where arrivals are described by a bivariate process @xmath8 , and the available service is given by a dynamic server @xmath11 . let @xmath94 and @xmath75 be the corresponding snr processes . fix @xmath178 and define , for @xmath156 , @xmath179 then , we have the following probabilistic performance bounds . @xmath180 output burstiness : @xmath181 , where @xmath182 @xmath180 backlog : @xmath183 , where @xmath184 @xmath180 delay : @xmath185 , where @xmath186 is the smallest number satisfying @xmath187 for the bound on the distribution of the output burstiness , we start from the inequality @xmath188 . it follows from the moment bound and lemma [ lem : mtconvdecon ] that , for any choice of @xmath189 and all @xmath156 @xmath190 to obtain the claim , we set the right hand side equal to @xmath191 , solve for @xmath192 , and optimize over the value of @xmath156 to obtain @xmath193 . the proof of the backlog bound proceeds in the same way , starting from the inequality @xmath194 . the delay bound is slightly more subtle . fix @xmath195 . using lemma [ lem : obbound ] and the moment bound with @xmath196 , we obtain that @xmath197 for every @xmath156 . by lemma [ lem : mtconvdecon ] , the mellin transform @xmath198 satisfies a bound that agrees with the function @xmath199 , except that the upper limit in the summation that defines @xmath199 would have to be replaced by @xmath200 . to obtain a sharper estimate , we use instead eq . from the proof of lemma [ lem : obbound ] . the resulting bound is that @xmath201 satisfies @xmath202 where we have used that the supremum in the definition of @xmath203 extends only up to @xmath204 , and then repeated the proof eq . . the claim follows by optimizing over @xmath144 . corresponding bounds as in theorem [ thm : blbound ] can be obtained using the algebra and the network calculus with moment - generating functions @xcite . the significance of theorem [ thm : blbound ] is that it permits the application of the the network calculus , where traffic is characterized in the bit domain , and service is naturally expressed in the snr domain . this will become evident in the next section , where the theorem gives us concise bounds for delays and backlog of multi - hop networks with rayleigh fading channels . we now apply the techniques developed in the two previous sections to a network of rayleigh channels . consider the dynamic snr server description for a rayleigh fading channel , as constructed in sec . [ sec : model].b . we use eq . , with the function @xmath6 given by @xmath205 where @xmath206 is the average snr of the channel and @xmath207 is the fading gain . for rayleigh fading , @xmath207 is a rayleigh random variable with probability density @xmath208 . in a physical system , @xmath209 , where @xmath210 and @xmath211 are the received signal power and the ( additive white gaussian ) noise power at the receiver , respectively . then , @xmath212 is exponentially distributed , and the mellin transform of @xmath6 is given by @xmath213 where @xmath214 is the upper incomplete gamma function . using the assumption that the @xmath52 are independent and identically distributed , we obtain for the mellin transform of the dynamic server @xmath215 for the arrival process , we use a characterization due to chang @xcite , where the moment - generating function of the cumulative arrival process in the bit domain is bounded by @xmath216 \le \rho(s)\cdot(t-\tau ) + \sigma(s)\end{aligned}\ ] ] for some @xmath156 . in general , @xmath217 and @xmath218 are nonnegative increasing functions of @xmath144 that may become infinite when @xmath144 is large . this traffic class , referred to as bounded arrivals , is broad enough to include markov - modulated arrival processes . the corresponding class of snr arrival processes is defined by the condition that @xmath219 for some @xmath149 . we consider the transmission of bounded arrivals on a rayleigh fading channel . to obtain single - hop performance bounds , we apply theorem [ thm : blbound ] with the expressions for the mellin transforms for the snr service and arrival processes from eqs . and . for the function @xmath220 from the statement of the theorem , we compute @xmath221_+}^ { \infty } \big ( \underbrace{e^{s \rho(s ) } e^{1/\bar\gamma } \bar{\gamma}^{-s } \gamma(1-s,\bar\gamma^{-1})}_{\deq v(s ) } \big ) ^ { u } \,,\end{aligned}\ ] ] where @xmath222_+$ ] is the maximum of @xmath223 and @xmath224 . the sum converges when @xmath225 , which can be interpreted as a stability condition . inserting the result into theorem [ thm : blbound ] , we obtain for the output burstiness the probabilistic bound @xmath226 the backlog bound is obtained by setting @xmath227 , @xmath228 the delay bound is the smallest number @xmath186 such that @xmath229 we also derive end - to - end bounds for a cascade of @xmath15 rayleigh channels with bounded arrivals , using the same parameters as before . let @xmath124 be the service process for the entire cascade . by corollary [ cor : cascade ] , its mellin transform satisfies for @xmath230 @xmath231 we will use again theorem [ thm : blbound ] . for @xmath230 , we compute @xmath232 where @xmath233 is as defined above , and where we applied the combinatorial identity @xmath234 for any @xmath235 and for @xmath236 . inserting @xmath237 into theorem [ thm : blbound ] gives for the end - to - end output bound , denoted by @xmath238 , the value @xmath239 note that for @xmath240 , this bound agrees with the previous bound for a single node . in the same way , we derive the probabilistic backlog bound @xmath241 for the delay bound , we estimate for @xmath242 @xmath243 here , the first term in the minimum is obtained by extending the summation down to @xmath244 , and the second term results from the inequality @xmath245 for @xmath246 . in both cases , the resulting sum can be evaluated with eq . . the delay bound @xmath186 is determined according to theorem [ thm : blbound ] by setting the right hand side of eq . equal to @xmath191 , solving for @xmath139 , and minimizing over @xmath144 . because of the complexity of the bound in eq . , the last two steps can only be performed numerically . it is apparent that the complexity of computing end - to - end bounds is no different than bounds for a single channel . more importantly , observe that the end - to - end bounds scale linearly in the number of nodes @xmath15 . relaxing the strong independence assumptions on the channel properties would give different scaling properties . ) vs. average channel snr ( @xmath247 ) for multihop rayleigh fading channels with @xmath248 , bounded traffic with @xmath249 kb and @xmath250 kbps , and @xmath251 khz.,width=480 ] ) vs. arrival rate ( @xmath217 ) for multi - hop rayleigh fading channels with @xmath248 , bounded traffic with @xmath249 kb , @xmath252 db , and @xmath251 khz.,width=480 ] ) vs. arrival rate ( @xmath217 ) for multi - hop rayleigh fading channels with @xmath248 , bounded traffic with @xmath249 kb , @xmath252 db , and @xmath251 khz.,width=480 ] we next present numerical results , where we assume rayleigh channels with a transmission bandwidth of @xmath253 khz . for traffic , we use bounded arrivals with default values @xmath254 kb and @xmath255 kbps , i.e. , the bounds on rate and bursts are deterministic . hence , the source of randomness in the examples results only from randomness of the channels . we use a violation probability of @xmath256 . in fig . [ fig : backlog - vs - snr - multihop ] we show the end - to - end backlog for cascades of @xmath15 rayleigh channels , as a function of the average snr of each channel . even though the backlog bounds increase only linearly in the number of nodes , the per - node requirements at least for the last node of the cascade must satisfy the end - to - end bounds , since it can not be assumed that backlog is equally distributed across the nodes . when the snr of the nodes is sufficiently high , the backlog remains low even for a large number of hops . we observe that the channel becomes saturated for @xmath257 db . when the number of channels is small , the backlog increases sharply in the vicinity of @xmath257 , but remains low everywhere else . in figs . [ fig : backlog - vs - rho - multihop ] and [ fig : backlog - vs - rho2-multihop ] we present , for values of the average snr of @xmath258 db ( fig . [ fig : backlog - vs - rho - multihop ] ) and @xmath259 db ( fig . [ fig : backlog - vs - rho2-multihop ] ) , how the end - to - end backlog increases as a function of the transmission rate , for different network sizes . we observe that the maximum achievable rate that does not result in a ` blow - up ' of the backlog decreases as the number of nodes is increased . we observe that the blow - up occurs earlier when the channel has a lower snr . ) vs. average channel snr ( @xmath247 ) for multi - hop rayleigh fading channels for @xmath260 and @xmath261 with buffer size @xmath262 kb , bounded traffic with @xmath249 kb and @xmath263 or @xmath264 kbps , and @xmath251 khz.,width=480 ] ) vs. average channel snr ( @xmath247 ) for multi - hop rayleigh fading channels for @xmath260 and @xmath261 with buffer size @xmath262 kb , bounded traffic with @xmath249 kb and @xmath263 or @xmath264 kbps , and @xmath251 khz.,width=480 ] ) vs. end - to - end delay bound for multi - hop rayleigh fading channels for @xmath260 , @xmath265 db , bounded traffic with @xmath249 kb and @xmath250 kbps and @xmath251 khz.,width=480 ] ) vs. end - to - end delay bound for multi - hop rayleigh fading channels for @xmath260 , @xmath265 db , bounded traffic with @xmath249 kb and @xmath250 kbps and @xmath251 khz.,width=480 ] ) vs. average channel snr ( @xmath247 ) for multi - hop rayleigh fading channels for @xmath266 , @xmath267 ms , bounded traffic with @xmath268 kb and @xmath263 kbps and @xmath251 khz.,width=480 ] suppose that buffer sizes are set to satisfy the end - to - end backlog . for a fixed buffer size @xmath269 , we can then use the probability @xmath270 as an estimate of the probability of dropped traffic , and refer to it as the _ loss probability_. in figs . [ fig : loss - vs - snr - multihop ] and [ fig : loss - vs - snr2-multihop ] , we depict the loss probability as a function of the average channel snr , for values of @xmath271 kb ( fig . [ fig : loss - vs - snr - multihop ] ) and @xmath272 kb ( fig . [ fig : loss - vs - snr2-multihop ] ) , for traffic with a rate of @xmath273 and @xmath264 kbps , and for @xmath260 , and @xmath261 nodes . the figure shows the minimum snr needed to support a given loss probability is very sensitive to the number of network nodes . in figs . [ fig : viol - vs - delay - multihop ] and [ fig : viol - vs - delay - multihop2 ] we show the violation probability for given end - to - end delay bounds , where we compare the delays at a single node ( @xmath240 ) with a multi - hop network ( @xmath266 ) for different average channel snr values , where we use eqs . and . the traffic parameters are @xmath249 kb for the burst , and @xmath263 kbps ( in fig . [ fig : viol - vs - delay - multihop ] ) @xmath250 kbps ( in fig . [ fig : viol - vs - delay - multihop2 ] ) . the figures illustrates that at sufficiently high snr values , low delays are achieved even when traffic traverses 10 links . when the snr is decreased , we can observe how the delay performance deteriorates in the multi - hop scenario . note that the curves for the violation probability are essentially piecewise linear with two segments . this is caused by the different exponential decay rates , which follows from eq . . depending on where in the equation the minimum occurs , we obtain a faster or slower decay . next we investigate how the delay violation probability for given end - to - end delay bounds is impacted by the snr of the channel . we consider a network with @xmath266 nodes , with the same parameters as before . the traffic parameters are @xmath249 kb and @xmath263 kbps . we consider end - to - end delay bounds of @xmath267 ms . [ fig : viol - vs - snr - multihop ] presents the results . an interesting observation is that the snr required to support a given violation probability for a short delay bound @xmath274 ms , the snr requirement increase fast for low violation probabilities . delay bounds @xmath275 ms and higher can be supported with low violation probabilities even when the average snr is small . the graphs presented here can be used in the planning of a multi - hop wireless network where predefined qos bounds are desired for a given transmission rate . since the average snr depends largely on the path loss , which , in turn , is a function of the transmission radius , the graphs could help with determining the maximum distance between nodes to support a desired transmission rate and qos . consider a scenario in fig . [ fig : leftover - snr ] where a through flow arriving to a fading channel shares the available bandwidth with other flows . we will refer to the traffic from these other flows as _ cross traffic_. we use @xmath276 and @xmath277 to denote the snr arrival processes of the through flow and the cross traffic , respectively , and let @xmath278 and @xmath279 denote the corresponding departure processes . in the snr domain , cross traffic can be viewed as reducing the channel capacity of the through flow by generating interference . the following lemma states that , in the snr domain , the service available to a through flow that experiences cross traffic at a channel can be expressed by a dynamic snr server . [ lem : leftover ] consider a channel with a through flow and cross traffic as shown in fig . [ fig : leftover - snr ] . assume that the channel provides a dynamic snr server to the aggregate of through flow and cross traffic , with service process @xmath75 , i.e , @xmath280 then @xmath281 is a dynamic snr server satisfying for all @xmath135 that @xmath282 if , moreover , the service to the cross flows satisfies the upper bound @xmath283 , then @xmath284 for any a sample path , and any @xmath195 , we have @xmath285 let @xmath286 be the point where the infimum is assumed . dividing by @xmath287 , we obtain @xmath288 where we used that @xmath289 by causality . this the first claim in eq . . to see the second claim , assume that @xmath283 . then @xmath290 , and therefore @xmath291 combining this with the first part of the proof , we obtain @xmath292 proving the second claim . note that @xmath293 need not be monotone in @xmath62 , i.e. , it may not lie in @xmath87 . we next estimate the service process available to the through flow across a cascade of channels . assuming that @xmath294 and @xmath74 are independent , we obtain for the mellin transform of the service process at a single node @xmath295 these service descriptions may be convolved , using lemmas [ thm : conc ] and [ lem : mtconvdecon ] to obtain a bound for the mellin transform of the service provided by a cascade of fading channels to a flow that experiences cross traffic at each node . [ cor : cascade - leftover ] consider a network as in fig . [ fig : tandem ] , where a through flow experiences cross traffic at each fading channel . let the snr service process at each channel be given by eq . . assume that the arrival process of the cross traffic satisfies eq . with parameters @xmath296 . assume that arrivals from through flow and cross traffic , as well as the service processes at each channel are independent . then , the service provided by the network to the through flow satisfies @xmath297 , where the mellin transform of @xmath298 satisfies , for @xmath153 , @xmath299 for a single node ( @xmath240 ) , we estimate for @xmath230 and @xmath153 @xmath300 where we have used that @xmath301 for @xmath153 . by lemma [ thm : conc ] , the service of the network is given by the convolution , @xmath302 . we use lemma [ lem : mtconvdecon ] to bound its mellin transform by @xmath303 where the sum runs over all sequences @xmath170 with @xmath171 and @xmath172 . collecting terms , we compute , as in the proof of corollary [ cor : cascade ] , that each product evaluates to the same term , @xmath304 since the number of summands is @xmath174 , this proves the claim . we now consider rayleigh fading channels , and assume that both through flow and cross traffic are bounded traffic with parameters @xmath305 and @xmath306 for the through flow , and @xmath307 and @xmath308 for the cross traffic . we next give end - to - end performance bounds , using theorem [ thm : blbound ] . for the function @xmath309 , we compute as in sec . [ sec : rayleigh].b for @xmath310 , @xmath311 where @xmath312 this computation is valid under the stability condition that @xmath313 . thus , we obtain that the output burstiness at the network egress gives @xmath314 \bigg \ } \ , . \nonumber \end{aligned}\ ] ] for the end - to - end backlog of the through flow , we obtain @xmath315 \bigg \ } \ , . \nonumber \end{aligned}\ ] ] similarly , for the delay bound , we estimate for @xmath242 @xmath316 ) vs. arrival rate ( @xmath306 ) for multi - hop rayleigh fading channels with @xmath248 , bounded traffic with @xmath317 kb , bounded cross traffic with @xmath318 kb and @xmath319 , @xmath252 db , and @xmath251 khz.,width=480 ] ) vs. arrival rate ( @xmath306 ) for multi - hop rayleigh fading channels with @xmath248 , bounded traffic with @xmath317 kb , bounded cross traffic with @xmath318 kb and @xmath319 , @xmath252 db , and @xmath251 khz.,width=480 ] ) vs. arrival rate ( @xmath306 ) for multi - hop rayleigh fading channels with @xmath248 , bounded traffic with @xmath317 kb , bounded cross traffic with @xmath318 kb and @xmath319 , @xmath252 db , and @xmath251 khz.,width=480 ] we now present numerical examples for rayleigh fading channels . we use the same traffic and channel model as in the numerical examples in sec . [ sec : numerical ] . the cross traffic is bounded traffic with parameters @xmath307 and @xmath308 . as before , we assume that cross traffic is deterministically bounded by using fixed values for @xmath307 and @xmath308 . consequently , there is no statistical multiplexing between through and cross traffic . throughout , the violation probability is set to @xmath248 . in fig . [ fig : backlogvssnr_wct ] we show the end - to - end backlog bound @xmath320 as a function of the average channel snr @xmath206 . the through flow has fixed parameters @xmath317 kb and @xmath321 kbps , and the cross traffic has parameters @xmath318 kb and @xmath319 . we consider networks with @xmath266 and @xmath322 nodes . the graph illustrates how the bursts of the cross traffic contribute to the backlog bound . there is an additive component for each traversed channel , which explains the difference for the backlog with @xmath323 and @xmath324 channels . the minimum required snr needed for stability appears less sensitive to the number of channels . in figs . [ fig : backlogvsrho_wct1 ] and [ fig : backlogvsrho_wct2 ] we again evaluate the end - to - end backlog bound @xmath325 . here , we keep the channel snr values constant at @xmath258 ( fig . [ fig : backlogvsrho_wct1 ] ) and @xmath259 ( fig . [ fig : backlogvsrho_wct2 ] ) . we vary the rate of the through flow @xmath306 on the x axis , and plot graphs for different values of @xmath319 . we again consider @xmath266 and @xmath322 . the outcomes are as seen in the previous figure . we observe the effects of the additive component contributed by each node . also , we observe , at least for lower transmission rates , that the stability of end - to - end backlogs is not sensitive to the number of traversed channels . we have developed a novel network calculus that can incorporate fading channel distributions , without the need for secondary models , such as fsmc . we use the calculus to compute statistical bounds on delay and backlog of multi - hop wireless networks with fading channels . we took a fresh point of view , where the descriptions of the arrivals and the fading channels reside in different domains , referred to as bit domain and snr domain . we found that by mapping arrival processes to the snr domain , an end - to - end analysis with fading channels becomes tractable . an important discovery was that arrivals and service in the snr domain obey the laws of a dioid algebra . the analytical framework developed in this paper appears suitable to study a broad class of fading channels and their impact on the network - layer performance in wireless networks . even though we computed numerical examples for very simplified networks , in particular , we made strong independence assumptions for the fading channels , our network calculus is applicable to networks where these assumptions are relaxed . generalizing our framework and obtaining a more profound understanding of the dioid algebra and computational methods in the snr domain is the subject of future research . f. ciucu , p. hui , and o. hohlfeld . non - asymptotic throughput and delay distributions in multi - hop wireless networks . in _ proc . 48th annual allerton conf . on communication , control , and computing _ , pages 662669 , sep . 2010 . y. jiang and p. j. emstad . analysis of stochastic service guarantees in communication networks : a server model . in _ proc . iwqos , springer lecture notes in computer science 3552 _ , pages 233245 , june 2005 . f. kelly . notes on effective bandwidths . in _ stochastic networks : theory and applications . ( editors : f.p . kelly , s. zachary and i.b . ziedins ) royal statistical society lecture notes series , 4 _ , pages 141168 . oxford university press , 1996 .

about @xmath0 sec after the big bang , the universe underwent a phase transition in which the electroweak gauge symmetry was reduced from @xmath1 to @xmath2 . in the standard model ( sm ) and many of its extensions , the electroweak phase transition ( ewpt ) is understood in terms of the dynamics of a fundamental scalar field , the higgs @xmath3 : the vacuum @xmath4 preferred at high temperatures becomes unstable at a certain `` critical temperature '' @xmath5 and the higgs develops a vacuum expectation value ( vev ) @xcite . the nature of this transition is a subject of considerable interest . for example , is the ewpt a violent , out - of - equilibrium event accompanied by massive entropy production ( first - order transition ) or a gentle , quasi - adiabatic process ( second - order transition ) ? answering this question would have far - ranging implications : for example , electroweak baryogenesis , which theoretically is arguably the most attractive of the mechanisms proposed to explain the observed matter - antimatter asymmetry , can only occur if the ewpt is strongly first - order @xcite . it is very difficult to obtain direct information on the ewpt from astrophysical observations , due to the long and complex history of the universe since the transition . ( note , however , that gravitational waves produced in a first - order transition might be observable at lisa @xcite . ) luckily , the dynamics of the phase transition can be deduced theoretically , once the fundamental lagrangian governing the physics at the electroweak scale is known . for example , it is well known that if there is no physics beyond the sm relevant at that scale , the transition is never first - order for the higgs boson mass obeying the current experimental lower bound , 114.4 gev . of course , the gauge hierarchy problem and other arguments strongly suggest that physics beyond the standard model ( bsm ) should appear at the electroweak scale , and the nature of the ewpt depends on the ( as yet unknown ) details of that physics . observing all of the weak - scale bsm particles and measuring their masses and interactions ( in particular with the higgs ) would provide sufficient information to determine the order of the transition . in practice , however , this will be a very difficult task for the upcoming large hadron collider ( lhc ) , and even for a more distant next - generation @xmath6 collider such as the proposed international linear collider ( ilc ) , especially in models where some of the new states are gauge singlets . it would therefore be useful to identify simpler observables that are correlated with the order of the transition . in this paper , we propose that the value of the higgs boson cubic self - coupling is one such observable . specifically , we argue that models leading to a first - order phase transition predict a large ( typically order - one ) deviation of the higgs cubic coupling from its sm value . we show that this is the case in three toy models , which capture three basic mechanisms for obtaining a first - order ewpt : loop - induced corrections to the higgs effective potential , nonrenormalizable operators in the potential , and mixing between the weak - doublet higgs and singlet scalars at the weak scale . the only exceptions to this statement that were observed in our analysis are due to accidental cancellations between large corrections to the higgs cubic of different origins . thus , while non - observation of a large deviation of the higgs cubic from the sm prediction would not completely rule out a first - order ewpt , it would make it very unlikely . the correlation between a strong first - order ewpt and a large deviation of the higgs self - coupling from the sm value was already noted , in the specific contexts of low - cutoff models @xcite and the two - higgs doublet model @xcite . here , we will argue that this corellation is in fact much more general , and applies ( with a small caveat requiring no accidental cancellations ) in a very broad class of models . in this paper , we will not restrict ourselves to a specific extension of the sm . we will instead study a series of toy models , which capture the important features of realistic bsm theories exhibiting strong first - order ewpt . the toy models analyzed in this section satisfy two basic conditions . first , the only scalar field which changes its value during the ewpt is the neutral component @xmath3 of the sm higgs doublet with gauge - singlet scalars is allowed . ] @xmath7 . with this assumption , the finite - temperature effective potential @xmath8 completely specifies the dynamics of the ewpt . second , at tree level , the higgs sector is weakly coupled and can be adequately described by a renormalizable lagrangian , i.e. no irrelevant terms are important . many theoretically motivated models satisfy these conditions , including the minimal supersymmetric standard model ( mssm ) @xcite for a broad range of parameters corresponding to the so - called higgs decoupling region , little higgs models @xcite such as the littlest higgs with t - parity , and others . interesting models which violate either one of these conditions and exhibit first - order ewpt also exist ; we will study such examples in sections [ sec:3 ] and [ sec:4 ] . with the above conditions , the tree - level higgs potential has the form v^t(h ) = - h^2 + h^4 , while the masses of the sm and bsm particles are given by m_i^2(h ) = m_0i^2 + a_i h^2 . at the one - loop level , the higgs effective potential , including finite - temperature corrections , can be determined completely once the values of @xmath9 and @xmath10 for every particle in the theory are known . in practice , particles with @xmath11 ( weak coupling to the higgs ) can be neglected . in the sm , these parameters are given by i & = & ( t , w , z , h , g),m_0 ^ 2&=&(0,0,0,-^2,-^2),a & = & ( , , , 3 , ) . the zero - temperature ( coleman - weinberg ) correction to the potential , in the @xmath12 regularization scheme , has the form v^l_0(h ) = _ i m_i^4 ( + c_i ) where @xmath13 is the renormalization scale and @xmath14 and @xmath15 are the multiplicity and fermion number of the field @xmath16 . in the sm , in the basis of eq . , @xmath17 and @xmath18 . the constants @xmath19 are regularization scheme - dependent and are physically irrelevant , since their only effect is an additive renormalization of the parameters @xmath20 and @xmath21 and the cosmological constant . rescaling @xmath13 has the same effect ; we fix @xmath22 gev and interpret @xmath20 and @xmath21 as the values of the couplings at that scale . the parameters are subject to the constraint ( v^t(h = v)+v^l_0(h = v ) ) = 0 where @xmath23 gev . in addition , the higgs mass , m_h^2 = ( v^t(h = v)+v^l_0(h = v ) ) , must satisfy the experimental lower bound , m_h>114.4 gev . the potential defined in this way suffers from an infrared singularity that appears at the minimum of the tree - level potential , @xmath24 , and is due to the massless ( goldstone ) bosons @xmath25 at that point @xcite . while the potential itself remains finite ( @xmath26 ) , the cubic coupling defined in diverges at that point . resummation of the infrared logarithms is required to correctly evaluate @xmath27 in the neighborhood of that point . this resummation is conveniently achieved by switching from the zero - momentum renormalization scheme to the on - shell scheme @xcite . technically , this is equivalent to adding an extra term to @xmath28 , and interpreting the second derivative at the minimum of the corrected potential as the physical ( on - shell ) higgs mass . the additional term eliminates the infrared divergence . we will employ the on - shell scheme throughout this paper . the one - loop finite - temperature potential has the form @xcite v_t(h , t ) & = & _ f_i=0 dk k^2 & & - _ f_i=1 dk k^2 , where @xmath29 . if the model contains bosons with masses small compared to @xmath30 , multi - loop infrared - divergent contributions need to be included to correctly analyze the dynamics of @xmath3 . these contributions can be resummed and written in the form of the `` ring potential '' @xcite : v_r(h , t ) = _ i tr , where the sum runs over all ( light ) bosonic degrees of freedom , and @xmath31 is the zero - momentum polarization tensor . in the sm , _ h(0 ) = _ g(0 ) & = & t^2 ( g^2 + g^2+_t^2+),_gb(0 ) & = & t^2 diag(g^2,g^2,g^2,g^2 ) , and m_gb^2(h)= ( cccc g^2 & 0 & 0 & 0 + 0 & g^2 & 0 & 0 + 0 & 0 & g^2 & -g g^2 + 0 & 0 & -g g^2 & g^2 + ) . these expressions are valid at @xmath32 ; for smaller @xmath30 , the ring terms are unimportant and can be dropped . the full potential , the potential @xmath33 develops an imaginary part , due to the quantum instability of a classical state with spatially uniform value of @xmath3 @xcite . this effect does not affect the dynamics of a first - order transition , as long as the rate of this instability is small compared to the electroweak scale . this condition is always satisfied in our analysis . ] is @xmath34 . a useful `` decoupling '' feature of this potential is that at any @xmath30 , the contributions from particles with @xmath35 are exponentially suppressed . thus , in bsm theories , only weak - scale states need to be included to analyze the ewpt . at high temperature , @xmath36 for all states , the potential has the simple form ( h ; t ) ( _ f_i=0 g_i a_i + _ f_i=1 g_i a_i ) . if the quantity in brackets is positive , @xmath37 is a stable minimum , and the electroweak symmetry is restored . this is the case , for example , in the sm and the mssm , where @xmath38 for all states . we will restrict our analysis to such models , since otherwise the ewpt does not occur . a first - order transition occurs if @xmath39 develops a second , symmetry - breaking local minimum , @xmath40 , while the high - temperature minimum @xmath37 is still classically stable . then , once the symmetry - breaking vacuum becomes energetically preferred , bubbles of the symmetry - breaking vacuum will start rapidly nucleating , growing , and coalescing , eventually encompassing the entire universe and completing the phase transition . the critical temperature @xmath5 at which the phase transition occurs can be found from the condition is strongly suppressed compared to its natural value , @xmath41 . phase transition does not occur until @xmath42 . given the large hierarchy @xmath43 , we will not include this possibility in our analysis . ] ( = 0 ; t_c)=(v_t(t_c ) ; t_c ) . the `` strength '' of the first - order transition ( more precisely , the degree of deviation from quasiadiabatic evolution ) can be characterized by the dimensionless parameter = . for example , it is well known that successful electroweak baryogenesis requires @xmath44 . due to the complicated form of the potential , the problem of finding @xmath5 and @xmath45 in a given model is in general not tractable analytically . we have developed a numerical code which is used to evaluate these quantities in all examples studied in this paper . in the sm , the ewpt is always second - order once the bound is satisfied . below , we analyze the situation in a few simple extensions of the sm . consider a toy model in which a single real scalar state is added to the sm , with the potential v = m_0 ^ 2 s^2 + ^2 |h|^2 s^2 . this is the most general renormalizable potential which respects a symmetry @xmath46 . ( without this symmetry , tree - level mixing between @xmath3 and @xmath47 would generally occur . such models will be discussed below . ) note that while we use the susy - inspired notation @xmath48 for the coefficient in eq . , this coefficient can in general be negative , as long as @xmath47 does not become tachyonic until @xmath49 and the lifetime of the vacuum at @xmath50 is larger than the present age of the universe . ( the runaway direction can be stabilized by the addition of a small @xmath51 term , which does not affect our analysis . ) the scalar contribution to the one - loop effective potential is given by eqs . , , , where @xmath52 , @xmath53 and @xmath54 . in addition , the ring terms due to the @xmath3 and @xmath25 fields have to be modified by adding the scalar contribution to the polarization tensors , @xmath55 . ( we neglect additional contributions to polarization tensors due to @xmath47 self - coupling or coupling to other bsm singlets . ) in order to identify the points with a first - order ewpt , we scanned the parameters of the model : in the scan in order to demonstrate that they exhibit the same correlation between the higgs cubic coupling and the @xmath56 parameter as the positive-@xmath48 models . we imposed the constraint that @xmath47 should not be tachyonic in the range @xmath57 gev , but did not analyze the lifetime of the conventional ewsb vacuum for these models , so some of the negative-@xmath48 points may be phenomenogically unacceptable . a detailed analysis of this issue is beyond the scope of this paper , and its outcome would not affect the main point of our argument in any way . ] @xmath58 gev , @xmath59 . we also scanned the parameters @xmath60 and @xmath21 , making sure that the constraints and are satisfied ; the scanned region corresponds to @xmath61 gev . ( we did not find any points with higher @xmath62 and a strong first - order ewpt . ) for every point exhibiting a first - order ewpt , we compute the one - loop value of the physical higgs boson cubic self - coupling , _ 3 |_h = v . the results are presented in fig . [ fig : scalar ] . in the left panel , we plot the parameters @xmath56 and @xmath27 ( normalized to the sm one - loop value ) for the points in our scan ; in the right panel , we plot @xmath62 and @xmath27 . it is clear from the plots that the higgs self - coupling is significantly larger than the sm value for _ all _ points with a first - order ewpt in this model . if @xmath62 is known from experiment ( as it surely will be , by the time the experiments become sensitive to the higgs self - coupling ) , the correlation between @xmath27 and @xmath56 becomes tighter , as illustrated in fig . [ fig : fixedmh ] . experimental measurement of @xmath27 requires collecting a significant sample of events with two higgs bosons in the final state . at the lhc , this coupling can be determined with 20 - 30 % precision for the higgs mass between 160 and 180 gev , for integrated luminosity of 3000 fb@xmath63 , which can be achieved if the luminosity upgrade to @xmath64 @xmath65s@xmath63 is realized @xcite . if an energy upgrade to a 200 tev vhlc is realized , the coupling can be measured with @xmath66% precision for 150 gev @xmath67 gev @xcite . for lower higgs masses , the vlhc could make a similarly precise measurement of @xmath27 using rare decays @xcite . also , the 500-gev ilc is expected to measure @xmath27 with @xmath68 accuracy for @xmath69 gev and a luminosity of @xmath70 ab@xmath63 @xcite . our analysis shows that if a strong first - order ewpt occurs due to an extra scalar contribution at one loop , and the higgs is sufficiently light , these experiments will discover a deviation of @xmath27 from the sm prediction . higher center - of - mass energy would be required if @xmath62 is larger ; note , however , that strong first - order ewpt in this setup requires @xmath71 gev , so a modest increase in @xmath72 would be sufficient to cover all of the interesting parameter space . , defined in eq . , vs. higgs cubic self - coupling . ( right panel ) higgs cubic self - coupling vs. higgs mass for points exhibiting a strong first - order ewpt , @xmath73 . in both plots , the higgs self - coupling is normalized to the one - loop sm expectation for the same @xmath62.,title="fig:",width=264 ] , defined in eq . , vs. higgs cubic self - coupling . ( right panel ) higgs cubic self - coupling vs. higgs mass for points exhibiting a strong first - order ewpt , @xmath73 . in both plots , the higgs self - coupling is normalized to the one - loop sm expectation for the same @xmath62.,title="fig:",width=264 ] , but with the higgs mass fixed at 160 gev.,width=264 ] in order to better understand this result , consider our model with @xmath74 . in this case , the one - loop higgs potential at zero temperature has the form v_0 h^4 + h^4 , where @xmath75 can be expressed in terms of the coupling constants of the model , and the contributions from higgs loops have been neglected . if @xmath76 ( which occurs for sufficiently large @xmath48 in our model ) , this potential has a stable symmetry - breaking minimum at @xmath77 , and a shallow local maximum at @xmath4 . this is of course just a realization of the purely radiative ewsb via dimensional transmutation ( dt ) _ a la _ coleman and weinberg @xcite . if now a small _ positive _ higgs mass term @xmath78 is added , the potential around @xmath79 is dominated by this mass term , and the maximum at the origin is turned into a minimum . at the same time , if @xmath20 is sufficiently small , the symmetry - breaking minimum is preserved , resulting in a potential with a `` bump '' between the two minima . a bumpy zero - temperature potential essentially guarantees that the ewpt will be of the first order , and most of the points with strong first - order ewpt in our scan ( the blue points in fig . [ fig : scalar ] ) have this feature . the higgs self - coupling for the pure dt potential , eq . , is 66% larger than the ( tree - level ) sm value for the same @xmath80 and @xmath62 : _ 3^dt = = _ 3^sm , tree , which is roughly at the center of the range for the models with strongly first - order ewpt in our scan . even without the bump ( red points in fig . [ fig : scalar ] ) , the models with @xmath81 have zero-@xmath30 potentials whose characteristic feature is a flatter maximum at @xmath4 compared to the sm . this shape is close to the dt potential , again resulting in large deviations of @xmath27 from the sm prediction . the connection with the pure dt case points to a potential concern about our model : in the dt case , the physical higgs mass is given by @xmath82 , so that large values of @xmath75 are required to satisfy the experimental lower bound on @xmath62 . correspondingly , in our scans , fairly large values of @xmath48 ( between 3 and 10 ) are required to obtain phenomenologically consistent points with a strong first - order ewpt . for @xmath83 , one might worry about the validity of perturbation theory that was used in our analysis . even for more moderate @xmath84 , two - loop corrections might be numerically significant . this is the case , for example , in the mssm , where the stops effectively contribute with @xmath85 ( see section [ multi ] ) . on the other hand , the main outcome of our analysis is not the precise values of @xmath27 for each point in the scan , but the fact that the deviation from the sm in this quantity is large for models with strong first - order ewpt . it seems very unlikely that the higher - loop corrections would conspire to cancel the large one - loop correction to @xmath27 and bring it back precisely to the sm value , so our main qualitative conclusion should continue to hold even at strong coupling . it should be noted that it is possible to have a large deviation of @xmath27 from the sm prediction _ without _ a strongly first - order transition . we found such points in our scan . these points have values of @xmath86 above , but not too far from , the higgs vev @xmath80 , and large values of @xmath87 . the zero - temperature effective potential receives a correction of order @xmath88 , giving a substantial correction to @xmath27 ; the correction to the finite - temperature potential decouples exponentially , as @xmath89 , and is not sufficient to yield a first - order transition . thus , while an observation of a large correction to @xmath27 would keep the door open to the possibility of a strongly first - order phase transition scenario , it would not prove it . in our analysis , however , the minimum of the one - loop potential @xmath80 is typically sufficiently far away from the tree - level minimum due to large one - loop renormalization by top and scalar loops , and the @xmath25 contribution to @xmath27 is numerically small , of order a few percent . we discard the points in our scan where the one - loop @xmath25 contribution to @xmath27 is 10% or larger , since the validity of perturbation theory is questionable for these points . no such points appeared in the scan of this section , but this constraint will play a ( minor ) role in the analysis of section [ sfpair ] below . consider the sm with a single additional weyl fermion @xmath90 , with a majorana mass term ( m_0 + + ) , where dots denote higher - order nonrenormalizable terms . the mass has the form with @xmath91 , and the formalism presented above applies , with @xmath92 . adding a single fermion also modifies the ring contributions : @xmath93 , where @xmath87 is the same coefficient appearing in eq . , and @xmath94 for a weyl ( dirac ) fermion . we neglect additional ring contributions that would arise if @xmath90 coupled to other bsm states . in this model , we performed the scan with @xmath95 gev , @xmath58 gev , and @xmath96 . for @xmath97 , @xmath27 is always suppressed with respect to the sm value , while for @xmath98 , @xmath27 is enhanced . ( in the @xmath99 effective potential , an @xmath98 fermion behaves like an @xmath97 boson , and vice versa . ) but in neither case do we find a strong ( @xmath73 ) first - order ewpt . since the only inputs from the model into the ewpt analysis are the coefficients in the mass formula , the same analysis applies in a model with a single dirac bsm fermion ( with @xmath100 if the fermion is chiral , and possibly @xmath101 if it is vector - like ) . the only change is the new multiplicity factor , @xmath102 . this does not affect the conclusions . , defined in eq . , vs. higgs cubic self - coupling . ( right panel ) higgs cubic self - coupling vs. higgs mass for points exhibiting a strong first - order ewpt , @xmath73 . in both plots , the higgs self - coupling is normalized to the one - loop sm expectation for the same @xmath62.,title="fig:",width=264 ] , defined in eq . , vs. higgs cubic self - coupling . ( right panel ) higgs cubic self - coupling vs. higgs mass for points exhibiting a strong first - order ewpt , @xmath73 . in both plots , the higgs self - coupling is normalized to the one - loop sm expectation for the same @xmath62.,title="fig:",width=264 ] any realistic extension of the sm has multiple bosonic and fermionic states , whose contributions to the higgs effective potential must be added . in this subsection , we will explore how the interplay between bosonic and fermionic contributions affects the connection between the higgs cubic coupling and the strength of a first - order ewpt . to address this , consider a toy model with a single fermion ( for concreteness , we assume it to be a vector - like dirac fermion ) , and four identical scalars . the number of scalars is chosen so that the number of bosonic and fermionic degrees of freedom is the same , as would be the case in supersymmetric models . we searched for points with strong first - order ewpt in the following parameter space : @xmath103 , @xmath104 gev , @xmath105 gev , @xmath106 gev . ( we did not find any points with a strong first - order ewpt for higher @xmath62 . ) the results are shown in fig . [ fig : sf ] . in most of the parameter space , the correlation between @xmath56 and @xmath27 , observed above in the purely scalar model , persists , and models with strong first - order ewpt predict observable enhancements of @xmath27 with respect to the sm value . there is also a smaller region where an observable suppression of @xmath27 is predicted ; this is due to the dominance of the fermion loops , which contribute to @xmath27 with a negative sign , for @xmath107 . there is also , however , an exceptional parameter region , in which @xmath27 is close to the sm value . this region has the following origin . in the one - loop zero - temperature potential , fermion and scalar contributions cancel if @xmath108 and @xmath109 , since @xmath47 and @xmath110 form an exact supermultiplet , and @xmath27 has its sm value . the scalar and fermion contributions to the finite - temperature part of the potential , however , do not cancel in this limit . in particular , if @xmath111 is large , a large cubic term can be generated at high temperature , which can lead to a first - order transition . fermion loops do not contribute to this term . this explanation highlights the exceptional nature of the region with sm - like @xmath27 : the relations @xmath108 and @xmath109 can not be enforced by a symmetry other than susy . given that the @xmath47 and @xmath112 fields have significant couplings to the higgs sector , and that susy must be broken in that sector , these relations are not radiatively stable . for example , renormalization group flow will naturally induce a mass splitting between @xmath47 and @xmath110 of order m^2 = higgs - higgsino mass splitting ( typically of order 100 gev or above ) and @xmath113 tev is the messenger scale where logarithmic divergences are cut off . given the large value of @xmath48 in the region where the first - order transition is possible , we expect @xmath114 to be naturally large , of order a few hundred gev . imposing that the mass splitting be at least 100 gev eliminates the region with first - order ewpt and no deviation in @xmath27 . we find that the minimal deviation in the cases @xmath115 gev , is 7% , 10% , 17% and 20% , respectively . we conclude that in the absence of accidental degeneracies in the mass spectrum , the @xmath116 corellation works well in this model . so far , we have observed the correlation between @xmath27 and @xmath56 in simple toy models . will it persist in realistic bsm theories with more complicated spectra and interactions ? while it is clearly impossible to analyze all possible cases , it seems very likely that the answer is positive . several arguments point in this direction . first , note that in a model with @xmath117 real ( or @xmath118 complex ) scalar fields with identical masses and couplings , the zero - temperture potential ( and therefore @xmath27 ) is the same as that for an `` effective '' single real scalar with m_0,eff^2 = m_0 ^ 2 , _ eff^2 = ^2 , g_eff=1 . the same scaling applies for the scalar contribution to the finite - temperature potential ( including rings ) , except one has to also rescale @xmath119 . if the scalar contribution dominates , this implies that the critical temperature in the @xmath117-scalar model is reduced by @xmath120 compared to the effective single - scalar model , and @xmath121 is unchanged . we checked numerically that this scaling is accurate to within 20% even after the sm contributions are added . thanks to this scaling , the approximate correlation between @xmath56 and @xmath27 is preserved for @xmath122 , and the models with @xmath122 and strong first - order ewpt predict _ larger _ minimal enhancement in @xmath27 compared to the @xmath123 case . the same approximate scaling laws apply to models with @xmath117 identical species of fermions as well . since the gauge charges of the added states have no effect on the higgs potential in the one - loop approximation , we conclude that the results of our toy - model analyses in fact apply to _ any _ model where a single multiplet ( or a single supermultiplet ) dominates the bsm contribution to the higgs potential . the mssm with light stops , and models with multiple gauge - singlet scalars as in ref . @xcite , are in this category . even if multiple bsm states with different masses and couplings contribute to the higgs potential , the qualitative picture should remain the same . in fact , we checked that a toy model with two scalars , whose masses and higgs couplings are varied independently , produces results very similar to the single - scalar case . this is reasonable , since the similarity between the properties of the models with strong first - order ewpt and the dt model , noticed in section [ singles ] , should persist independently of the precise origin of the large radative corrections to the higgs potential . based on this evidence , we believe that significant deviation of the higgs cubic self - coupling from the sm value is a generic prediction of weakly - coupled models of ewsb with strong first - order ewpt , provided that no accidental cancellations occur . a measurement of this coupling can serve as a diagnostic tool to glean information about the nature of the ewpt . this would be particularly valuable in models with extra gauge singlets , whose masses and couplings to the higgs are difficult to measure directly , especially if the decay @xmath124 is kinematically forbidden . the effect of new physics at the scale @xmath125 on the higgs potential can be parametrized by adding nonrenormalizable operators , suppressed by powers of @xmath13 , to the potential . grojean , servant and wells ( gsw ) pointed out that in the presence of such terms , strong first - order ewpt may occur @xcite . following gsw , consider a model with a tree - level higgs potential of the form v^t(h ) = ^2 h^h + ( h^h)^2 + ( h^h)^3 . it is reasonable to use this effective theory to describe the ewpt if the electroweak symmetry - breaking vev is small compared to @xmath13 . the strong first - order ewpt occurs for positive @xmath20 and negative @xmath21 . with these choices , it can be shown that the theory is applicable only if @xmath126 . in general , this requires fine - tuning : for example , if the dimension-6 operator arises due to exchanges of a heavy weak - singlet scalar , one can show that @xmath127 will also be generated when the scalar is integrated out . this should be cancelled by a bare term in the lagrangian . however , given the relatively low values of @xmath13 interesting for our analysis ( 500 gev1 tev range ) , the required fine - tuning is numerically quite modest . operator . models with a `` bumpy '' zero - temperature higgs potential are shown in blue , and those without the bump in red . ( left panel ) the strength of the first - order ewpt @xmath56 , defined in eq . , vs. higgs cubic self - coupling . ( right panel ) higgs cubic self - coupling vs. higgs mass for points exhibiting a strong first - order ewpt , @xmath73 . in both plots , the higgs self - coupling is normalized to the one - loop sm expectation for the same @xmath62.,title="fig:",width=264 ] operator . models with a `` bumpy '' zero - temperature higgs potential are shown in blue , and those without the bump in red . ( left panel ) the strength of the first - order ewpt @xmath56 , defined in eq . , vs. higgs cubic self - coupling . ( right panel ) higgs cubic self - coupling vs. higgs mass for points exhibiting a strong first - order ewpt , @xmath73 . in both plots , the higgs self - coupling is normalized to the one - loop sm expectation for the same @xmath62.,title="fig:",width=264 ] we analyzed the dynamics of the ewpt in the gsw model , following the general framework discussed in section [ sec:2 ] . we performed a scan of the following parameters : @xmath128 gev , @xmath129 gev . ( we did not find any points with strong first - order ewpt for higher @xmath13 and @xmath62 . ) our results for the order of the transition are roughly consistent , @xmath130 and @xmath131 , listed in section [ sec:2 ] , which were not included in ref . @xcite . on the other hand , we do not include the corrections to the one - loop potential and ring terms induced by the dimension-6 operator , and do not account for the possibility of delayed nucleation . both effects were included in ref . these differences do not affect the argument of this paper . ] with refs . we also computed the higgs cubic self - coupling ( at the one - loop order ) for each point in the scan . the results are shown in fig . [ fig : h6 ] . in the region where there is a strong first - order ewpt , this model also predicts a large , order - one enhancement of @xmath27 from its sm value , which will be easily observable at the ilc . this is easy to understand analytically : at tree level , we have = 1 + . since strong first - order ewpt can only occur for @xmath13 in the 500 gev1 tev range , the correction term is large for these points . ( the one - loop corrections to eq . are at most about 10% in our scan . ) in addition , as fig . [ fig : h6 ] indicates , there is a clear and strong correlation between this coupling and the strength of the ewpt . in all examples studied above , the sm - like higgs field @xmath3 was the only field that acquired an expectation value during the ewpt . if other scalar fields are present , they may change their value in the same transition , so that the order parameter for the ewpt is effectively a linear combination of @xmath3 and other fields . of particular interest are models with extra gauge - singlet scalars , such as the nmssm , certain little higgs models , and others . at low temperatures , both singlet and higgs vevs are generically non - zero . at high temperatures , the higgs vev is driven to zero , and although the singlet vev typically remains non - zero , the ewpt involves changes in both vevs . since the zero - temperature singlet potential is not restricted to even - degree polynomials by gauge invariance , it may contain cubic terms , which can naturally produce large `` bumps '' and lead to strong first - order ewpt . we will consider a simple model of this kind in this section . our main goal here is to illustrate that the connection between the strong first - order ewpt and observable deviations of the higgs self - coupling from the sm value persists in this model . many studies of models of this type have appeared in the literature , mostly in the context of the nmssm and the nmssm @xcite . we will follow the conventions of a recent analysis by profumo , ramsey - musolf , and shaughnessy ( prs ) @xcite , which applies more generally . ( see also ref . @xcite . ) consider a model with a single real scalar @xmath47 added to the sm . the most general renormalizable scalar potential has the form v^t(h , s ) = -^2 h^h + ( h^h)^2 + h^h s+ h^h s^2+s^2 + s^3+s^4 , where @xmath132 . we assume that the model has a stable vacuum at @xmath133 . ( the conditions for this are listed in ref . @xcite . ) the physical higgs fields are obtained by expanding around this minimum , @xmath134 and @xmath135 , and diagonalizing the mass matrix by a rotation h_1&= & s+ h , h_2&= & s - h. the mixing angle is defined such that @xmath136 . with this convention , @xmath137 is more doublet - like and @xmath138 is more singlet - like . for positive ( negative ) @xmath139 , the eigenmass @xmath140 is greater than ( less than ) @xmath141 . we focus on the state @xmath137 , which for simplicity we will refer to simply as `` the higgs '' , and study the deviation of its cubic self - coupling from the sm prediction for a higgs of the same mass . at tree level , & = & ( 12 v^3 + 3 a_1 ^ 2 + 6 a_2(x+v)+ & & 4 b_3 ^ 3 + 12 b_4x^3 ) . unlike the models studied in section [ sec:2 ] , no large couplings are required to achieve first - order ewpt in this setup . we restrict our analysis to the points where all couplings are weak , and ignore the loop corrections to the relation . , defined in eq . , vs. higgs cubic self - coupling . ( right panel ) higgs cubic self - coupling vs. higgs mass for points exhibiting a strong first - order ewpt , @xmath73 . in both plots , the higgs self - coupling is normalized to the tree - level sm expectation for the same @xmath62.,title="fig:",width=264 ] , defined in eq . , vs. higgs cubic self - coupling . ( right panel ) higgs cubic self - coupling vs. higgs mass for points exhibiting a strong first - order ewpt , @xmath73 . in both plots , the higgs self - coupling is normalized to the tree - level sm expectation for the same @xmath62.,title="fig:",width=264 ] it is straighforward to compute the finite - temperature one - loop potential in this model , and analyze the dynamics of the ewpt numerically . we used the high - temperature approximation for @xmath142 , expanding it up to linear order in @xmath30 . this approximation greatly speeds up the numerical analysis . we performed a partial scan of the parameter space of the model , setting @xmath143 and scanning @xmath144 gev , @xmath145 gev , @xmath146 , @xmath147 gev , and @xmath148 . this parameter region includes a large set of points with strong first - order ewpt , and lies roughly within the range consistent with precision electroweak constraints gev mass range is allowed if new physics produces a positive contribution to the t parameter . this is the case , for example , in the lht model @xcite . ] determined in @xcite . the results of this scan are illustrated in fig . [ fig : svevblob ] . the models with strong first - order ewpt predict a suppression of the higgs cubic self - coupling relative to the sm value , by a factor of 2 or more . the sign of the effect , however , is not unique in this model : for example , we found points with @xmath149 which predict a strong first - order ewpt and an enhanced higgs self - coupling . the important point is that large , order - one deviations in @xmath27 from the sm are typical . this is not surprising , since both these deviations and the possibility of strong first - order ewpt are due to same new tree - level interactions . the corellation between the size of this effect and the strength of the ewpt is less clear than in the examples of sections [ sec:2 ] and [ sec:3 ] . this is simply due to a large number of parameters that affect @xmath27 , but have only a marginal effect on @xmath56 . if those parameters are fixed , and , for example , only the higgs - singlet mixing parameter @xmath150 is varied , a clear corellation between @xmath27 and @xmath56 emerges , as seen in fig . [ fig : sveva1 ] . ( in this figure we fixed @xmath151 gev , @xmath152 , @xmath153 gev , @xmath148 , @xmath143 , and scanned @xmath154 gev . ) vs. higgs cubic self - coupling , in the model of eq . . only the mixing parameter , @xmath150 , is varied . the higgs self - coupling is normalized to the one - loop sm expectation for the same @xmath62.,width=264 ] unlike the models of sections [ sec:2 ] and [ sec:3 ] , where the non - standard higgs cubic coupling might provide the first experimental hint of the existence of extra singlet scalars at the weak scale , the model studied here will provide a variety of deviations in the higgs production cross section and branching ratios from the sm due to the singlet admixture @xcite . many such deviations will probably be observed before the higgs self - coupling measurement is available , especially if @xmath27 is suppressed . nevertheless , it is interesting that the general pattern of large deviations in @xmath27 in all models exhibiting a strong first - order ewpt persists in this class of models . we fixed @xmath151 gev , @xmath152 , @xmath153 gev , @xmath148 , @xmath143 , and scanned @xmath155 gev . the results are presented in fig . [ fig : sveva1 ] . this scenario demonstrates a direct proportionality between @xmath150 and @xmath56 , but an inverse proportionality between @xmath150 and @xmath27 . the origin of this supression is as follows . the terms in the potential involving only the doublet components of the @xmath137 produce a positive contribution to @xmath27 , as in the sm . ( this contribution reduces to the sm value in the no - mixing limit . ) the term involving doublet - singlet mixing , as well as the @xmath156 term , produce a negative contribution for @xmath157 , since their coefficients @xmath158 and @xmath159 are negative . note that the sign of @xmath158 and @xmath159 is effectively determined by the requirement of strong first - order ewpt : it is these terms that introduce the desired `` bump '' in the potential . in this paper , we investigated the link between the dynamics of the ewpt and the deviation of the cubic higgs boson self - coupling in extensions of the sm . we studied this link in a series of toy models . our toy models implement three mechanisms by which the weak - scale new physics could turn the second - order adiabatic transition predicted in the pure sm into a first - order transition . in the first set of models , the dynamics is determined by large quantum corrections to the higgs potential , both zero- and finite - temperature , from loops of new particles ( possibly , but not necessarily , gauge singlets ) with strong coupling to the higgs . in the second model , the first - order transition is due to the presence of a nonrenormalizable operator in the higgs potential . in the third model , the crucial new feature is the tree - level mixing between the higgs and an extra scalar singlet . in all cases , we found that models that exhibit a strong first - order ewpt predict large deviations of the higgs cubic coupling from the standard model value . order - one deviations are typical . the models of the first and second classes typically predict an enhancement in the higgs cubic coupling , while in the models of the third class both signs are equally likely . the only cases where the cubic coupling does _ not _ deviate significantly from the sm prediction are due to accidental cancellations among unrelated contributions to this coupling . thus , we conclude that a measurement of the higgs cubic coupling consistent with the sm prediction would strongly disfavor ( although not completely rule out ) a first - order electroweak phase transition . a discovery of a substantial deviation of this coupling from the sm would keep the possibility of a first - order transition open , but would not conclusively prove it , as discussed at the end of section [ singles ] . while a measurement of the higgs cubic coupling is very difficult at the lhc , a 20%-level measurement appears to be within the reach of the proposed luminosity or energy upgrades of the lhc , as well as the ilc . based on the analyis , we conclude that such a measurement could serve as a sensitive probe of the dynamics of the electroweak phase tranition . if no deviation from the sm is observed , it would be very unlikely that the transition is first - order . note that at least some of the sources of the first - order transition are very difficult to exclude by any other means : for example , a gauge - singlet scalar that does not mix with the higgs at tree level but has a strong quadratic coupling , as in the model of section [ singles ] , is essentially impossible to detect directly as long as the decay @xmath124 is kinematically forbidden . therefore , even with a large body of knowledge about the bsm particles and their couplings from the lhc and ilc experiments , the question of the order of the phase transition would be very difficult to settle definitively . a measurement of the higgs cubic coupling would be crucial in addressing this issue . it is exciting that the next generation of collider experiments will likely contribute to our understanding of cosmology . one example is the possibility that the dark matter particle will be discovered , and its propserties measured , at the lhc and the ilc . this connection has attracted much attention in the literature , and has been analyzed both in specific models of new physics @xcite , and in a model - independent framework @xcite . understanding the dynamics of the electroweak phase transition is another open question in the early universe cosmology where the lhc and ilc experiments will have an important role to play . in this paper , we studied one simple experimentally observable quantity that is tightly correlated with the order of the transition . it is also possible that a combination of mass and coupling measurements could be used to completely reconstruct the higgs potential , providing a much more detailed picture of the transition . it would be interesting to understand what quantities would need to be measured , and how precise the measurements would need to be . we are grateful to sohang gandhi , christophe grojean and geraldine servant for useful discussions . we are also very grateful to david rainwater for a communication regarding the prospects for a measurement of @xmath27 at the lhc . this research is supported by the nsf grant phy-0355005 . d. a. kirzhnits , jetp lett . * 15 * , 529 ( 1972 ) [ pisma zh . eksp . teor . fiz . * 15 * , 745 ( 1972 ) ] ; d. a. kirzhnits and a. d. linde , phys . b * 42 * ( 1972 ) 471 . m. e. shaposhnikov , nucl . b * 287 * , 757 ( 1987 ) ; nucl . b * 299 * , 797 ( 1988 ) . for reviews of more recent developments , see a. riotto and m. trodden , ann . sci . * 49 * , 35 ( 1999 ) [ arxiv : hep - ph/9901362 ] ; m. dine and a. kusenko , rev . phys . * 76 * , 1 ( 2004 ) [ arxiv : hep - ph/0303065 ] . c. grojean and g. servant , phys . d * 75 * , 043507 ( 2007 ) [ arxiv : hep - ph/0607107 ] ; c. caprini , r. durrer and g. servant , arxiv:0711.2593 [ astro - ph ] . c. grojean , g. servant and j. d. wells , phys . d * 71 * , 036001 ( 2005 ) [ arxiv : hep - ph/0407019 ] . s. kanemura , y. okada and e. senaha , phys . b * 606 * , 361 ( 2005 ) [ arxiv : hep - ph/0411354 ] ; `` electroweak baryogenesis and the triple higgs boson coupling , '' _ in the proceedings of 2005 international linear collider workshop ( lcws 2005 ) , stanford , california , 18 - 22 mar 2005 , pp 0704 _ [ arxiv : hep - ph/0507259 ] ; s. w. ham and s. k. oh , [ arxiv : hep - ph/0502116 ] . for reviews , see h. e. haber and g. l. kane , phys . * 117 * , 75 ( 1985 ) ; s. p. martin , arxiv : hep - ph/9709356 . for reviews , see m. schmaltz and d. tucker - smith , ann . nucl . part . sci . * 55 * , 229 ( 2005 ) [ arxiv : hep - ph/0502182 ] ; m. perelstein , prog . part . phys . * 58 * , 247 ( 2007 ) [ arxiv : hep - ph/0512128 ] . g. w. anderson and l. j. hall , phys . d * 45 * , 2685 ( 1992 ) . l. dolan and r. jackiw , phys . d * 9 * , 3320 ( 1974 ) ; s. weinberg , phys . rev . d * 9 * , 3357 ( 1974 ) . m. e. carrington , phys . d * 45 * , 2933 ( 1992 ) . e. j. weinberg and a. q. wu , phys . d * 36 * , 2474 ( 1987 ) . u. baur , t. plehn and d. l. rainwater , phys . lett . * 89 * , 151801 ( 2002 ) [ arxiv : hep - ph/0206024 ] . u. baur , t. plehn and d. l. rainwater , phys . d * 67 * , 033003 ( 2003 ) [ arxiv : hep - ph/0211224 ] . u. baur , t. plehn and d. l. rainwater , phys . d * 69 * , 053004 ( 2004 ) [ arxiv : hep - ph/0310056 ] . a. djouadi , j. lykken , k. monig , y. okada , m. j. oreglia and s. yamashita , arxiv:0709.1893 [ hep - ph ] . s. r. coleman and e. weinberg , phys . d * 7 * , 1888 ( 1973 ) . w. d. goldberger , b. grinstein and w. skiba , arxiv:0708.1463 [ hep - ph ] . j. r. espinosa and m. quiros , phys . d * 76 * , 076004 ( 2007 ) [ arxiv : hep - ph/0701145 ] . s. j. huber and t. konstandin , arxiv:0709.2091 [ hep - ph ] . c. delaunay , c. grojean and j. d. wells , arxiv:0711.2511 [ hep - ph ] . g. f. giudice , c. grojean , a. pomarol and r. rattazzi , jhep * 0706 * , 045 ( 2007 ) [ arxiv : hep - ph/0703164 ] . see , for example : m. pietroni , nucl . b * 402 * , 27 ( 1993 ) [ arxiv : hep - ph/9207227 ] ; s. j. huber and m. g. schmidt , nucl . b * 606 * , 183 ( 2001 ) [ arxiv : hep - ph/0003122 ] ; a. menon , d. e. morrissey and c. e. m. wagner , phys . d * 70 * , 035005 ( 2004 ) [ arxiv : hep - ph/0404184 ] ; c. balazs , m. s. carena , a. freitas and c. e. m. wagner , jhep * 0706 * , 066 ( 2007 ) [ arxiv:0705.0431 [ hep - ph ] ] . s. profumo , m. j. ramsey - musolf and g. shaughnessy , jhep * 0708 * , 010 ( 2007 ) [ arxiv:0705.2425 [ hep - ph ] ] . a. ahriche , phys . d * 75 * , 083522 ( 2007 ) [ arxiv : hep - ph/0701192 ] . j. hubisz , p. meade , a. noble and m. perelstein , jhep * 0601 * , 135 ( 2006 ) [ arxiv : hep - ph/0506042 ] . see , for example , e. a. baltz , m. battaglia , m. e. peskin and t. wizansky , phys . d * 74 * , 103521 ( 2006 ) [ arxiv : hep - ph/0602187 ] . a. birkedal , k. matchev and m. perelstein , phys . d * 70 * , 077701 ( 2004 ) [ arxiv : hep - ph/0403004 ] ; c. bartels and j. list , arxiv:0709.2629 [ hep - ex ] .

living systems as far - from - equilibrium open systems , are essentially cognitive , and conversely , in order to grasp the essence of cognition , one may attempt to understand the ontology of the former processes . often this is viewed in the framework of a cell - like , structured , self - organizing system engaged in a two - way interaction between its functional mechanisms and that of its neighboring environment . there are several theories that revolve around this principle . readers may be familiar with a central relationship as explained by autopoiesis , a seemingly enduring theory as developed by maturana and varela ( 1980a , 1980b ) over several decades . though a much earlier and somewhat different hypothesis of bertalanffy ( 1972 ) had proposed that living systems are maintained in non - equilibrium states by a flow - like patterns when drawing matter and energy from their environment , and adjust accordingly in a `` flowing balance '' ( bertalanffy , 1972 ; capra , 1996 ) . further scientific rigor aimed at understanding this hypothesis was achieved by prigogine ( 1980 ) who formulated similar ideas cast within a theory of dissipative structures . whereas living systems are continuously maintained in - far - from equilibrium , dissipative structures do likewise while being capable of evolving . they are destabilized in the increase of information and energy , though as they remain self - organized and self - perpetuating , the complexity of their structure increases ; often this is simply for the sake of survival within the environment . elucidating a possible synthesis of these separate approaches , commencing from the bertalanffy hypothesis , is the main topic of capra ( 1996 ) . but cognition is also a function of a prevailing culture : by means of social and historical patterns of behavior traditions , etc . , human cognition feeds back into that culture through the course of social interaction , adaptive technologies , policies , memetic trends , etc . , and inevitably alters it in time ( see e.g. clark , 1997 ; hollan et al . , 2000 ; hutchins , 1994 ; richerson and boyd , 2004 ; wallace and fullilove , 2008 ) . there are several common factors at stake here , and in seeking to understand these , the ` immunology language ' viewpoint of atlan and cohen ( 1998 ) ( see also cohen , 2000 ) views human organizations at all levels as perceiving patterns of threat or opportunity , comparing those patterns with some internal , learned or inherited , picture of the world , and then choosing one or a small number of responses from a vastly larger repertory of that which is possible to them . putting it another way , consider a basic observation concerning the immune system : the latter incorporates its own system of options and responds cognitively at the level of information processing such that the _ meaning _ of an antigen is defined by the response of the immune system , somewhat reminiscent of earlier work of jerne ( 1974 ) who had postulated such ` meaning ' . the broader cognitive model that results from this can said to be a ` reactive ' system determined by contextual factors in cohen and harel ( 2007 ) as depicted in figure 1 . starting from this basic perspective , wallace ( 2005 ) has formulated a model of cognition that takes ` meaning ' in the sense of dretske s theory of semantic communication ( dretske , 1981 , 1988 ) claiming that the immune perception / response information networks of any cognitive system must be constrained by shannon s fundamental limit theorems of information theory ( see e.g. ash , 1990 ; berger , 1971 ; cover and thomas , 1991 ) . these networks comprising of cognitive modules interact within the framework of a kind of broadcasting system relaying within ` a theater of consciousness ' the basic operative hypothesis of the _ global ( neuronal ) workspace _ theory as developed in ( baars , 1988 ; baars and franklin , 2003 ) that forms an integral part of the model in ( wallace , 2005 ; wallace and fullilove , 2008 ) . here we take a further step forward that incorporates a number of related factors mainly following the ( generalized ) onsager relations of non - equilibrium thermodynamics combined with the principles of _ rate distortion _ theory which is one of the mainstays of shannon s pioneering work . one particular observation is that distortion in communication between interacting cognitive modules , is patently stochastic , in particular , it is manifestly a process of brownian motion . this observation is developed in several steps relative to a corresponding stochastic differential equation that eventually unfolds to a kind of master equation . by considering critical solutions for the distortion , we argue on a case - by - case basis that the ensuing brownian motion ( viz weiner process ) can be locally one of bounded variation . accordingly , we are motivated to introduce the necessary conditions that permit integration of these equations in order to obtain _ a multidimensional it process _ ; further we discuss those formal conditions that describe an associated stochastic solution flow . thus in a way we have recovered bertalanffy s perception of the `` flow '' of living systems , the kind of which information - tied epigenetic systems are certain representations . whereas such model conditions are mathematically stated for a flow in a smooth manifold geometry , we do not claim such ` smoothness ' in general . indeed , such epigenetic ` flow systems ' in living organisms and brain - environment interactions may be at best continuous over time and are most likely to be ` singular ' in some sense . the latter requires harder topological techniques and possible complications that detract from a more natural interpretation . eventually one seeks a workable common ground between various mathematical abstractions and the actual empirical properties of the systems themselves . another slant emerging from this cognitive paradigm goes towards epigenetic information sources as providing a _ tunable catalyst _ directed to gene expression by which the embedding of information sources can direct developmental pathways within the ontology of enveloped structures via the process of information ( wallace and wallace , 2008 ) . this leads to further developments of several cognitive paradigms for gene expression , in part furthering the scientific reason underlying that of jablonka and lamb ( 2005 ) who consider epigenetic inheritance systems in which information may be transmitted through generations , not just simply through the base sequence of dna , but also transmitted via cultural and behavioral means in higher animals , and by epigenetic means in cell lineages . this is initiated by memory systems that enable somatic cells of differing phenotype , but of identical genotype , to transmit their phenotypes to their descendants , even in the absence of the original stimuli that had engaged these phenotypes . conditional upon an individual s phenotype , environmental factors may trigger alterations in behavior and health , eventually impinging upon the nervous system with the likely consequence of mental disarray . in this respect moffitt et al . ( 2006 ) ( cf caspi and moffitt , 2006 ) make several observations : ( 1 ) a heritable versus environmental influence on phenotype variation across a given environment , ( 2 ) altered gene expressions via epigenetic programming geared in response to subsequent health - behavioral reactions towards the environment , ( 3 ) how an individual s phenotype determines a risk - level towards the environment , and ( 4 ) behavioral effects due to interdependence between specified variations in the dna sequence specific to a measured environment . indispensable to understanding and extending these findings is the parallel question of how gene - cultural interaction , two distinct but interacting hereditary systems , compares psychopathology across oriental and western cultures in basic perceptual processes , often an important piece missing from the larger picture ( nisbett , 2003 ; richerson and boyd , 2004 ; wallace , 2005 ; wallace , 2009 ) . as such distinct interacting systems of information influencing action and behavior , both kinds ( genes and culture ) are claimed in durham ( 1991 ) to create a real and unambiguous symmetry : between genes and phenotypes on the one hand , and culture and phenotypes on the other , whereby genes and culture may be reasonably viewed as two parallel lines of hereditary influence on phenotypes . from the perspective of ( wallace , 2005 ; wallace and wallace , 2008 ; wallace and wallace , 2009 ) both can be realized as generalized languages in the sense they have their own intrinsic recognizable grammar and syntax ( see [ meaningful ] ) . likewise , on reflecting upon the fundamental mechanisms of serial endosymbiotic theory ( see e.g. margulis , 2004 ) , witzany ( 2006 ) argues a case for extending the latter via biosemiotic cell - cell type interactions as ` signed ' language - communication processes subject to a range of syntactic , pragmatic and semantic rules as applied to protein coding dna , rna editing , dna splicing , transcription and other essential functions . a more general perspective is to consider those central relations characterizing the system and its various structural components , and how a state is altered under perturbations by the environment . an autopoietic system is organizationally closed and structurally determined . the system s autopoiesis is preserved within the living state , adaptable only to structural fluctuations for as long as the living entity survives within and is structurally coupled to its environment ; otherwise there is termination . the autopoiesis of the nervous system ( though itself is not strictly autopoietic ) functionally self-(creates ) replicates in order to engage in cognition . though structurally dependent , the nervous system affords an innate plasticity and with appropriate alterations , it is conducive to learning and can adapt itself towards broader interactions and human self - consciousness ( maturana and varela , 1980a , 1980b ) . such broader interpretations of this theory are addressed in mingers ( 1991 ) , such as the family nexus which via its idiosyncratic behavioral and linguistic interactions ( in relationship to hereditary factors , socioeconomic status , environment , culture , etc . ) , creates a peculiar structured reality that for the best part is only accessible to the members themselves , and through which a prospective psychotherapist must explore in order to fathom out those recurrent patterns of conversation and behavior influencing the cognitive malfunctioning of any concerned ( cf laing and esterson , 1964 ) . there are related theories with varying degrees of overlap to the overall concepts of maturana and varela ( 1980a , 1980b ) . one example is a biogenetic - structuralist theory ( laughlin and daquili , 1974 ) where connections within an evolutionary context are made between neural organization / brain functional activity and the environment . the result is `` acquired models of reality '' , again , a cognitive outcome of the response to and engagement with the environment , proposing a concept of ` neurognosis ' , a kind of ` holographic ' model based upon biogenetically induced rudimentary information embedded in various associated regions of the brain , as taken from birth towards the later stages of maturity . the neurognostic model depends upon ontogenetic feedback from an external reality according to which acquired components are continually engaged with sensory input . such modes are claimed to be represented in neural processes by probability expectations which we shall see , lie at the heart of information theory , and which encode the behavior of these process in relationship to the environment . thus it may be reasonable to suppose that enriched environments of some sort reciprocate to a greater advantage than others , and the structures governing the alignment of the self to the environment to an extent evolved in accordance to some degree of mastery over the latter . take for instance , the hypothalamic pituitary adrenal ( hpa ) axis ( as governing the neurophysiology of the `` flight or fight '' mechanism ) is cognitive in the sense of atlan and cohen ( 1998 ) . if there is an arousal of the individual s close environment , then mind , memory and emotional cognition engage , evaluate and select appropriate responses . the hpa accelerates this process and possible malfunctioning may induce hyper reactivity as observed in cases of post traumatic stress disorder . depression , as another example , may be partly viewed as the evolution of a structure conducive to a negative alignment of the self with an external reality . if we think of a child as gradually mastering its environment via a symbiotic relation with its mother , then take away the mother , the inability to further manage the environment activates a negative kind of ( neurognostic ) structure such as those that have been researched in the context of various evolutionary theories of depression that are founded upon the occurrence of attachment - defeat - loss , diminished opportunities , down - regulation of foraging capability , social / professional rank , etc . incurred with varying risk factors within a culturally influenced environment ( see e.g. gilbert , 2006 ; moffitt et al . , 2006 ; wallace , 2009 ) . another example is to consider the body s blood pressure control system consisting as a network of cognitive systems which compare a set of incoming signals with an internal reference structure in order to select a suitable level of blood pressure from possible levels ; hence as claimed in ( wallace and fullilove , 2008 ; wallace and wallace , 2009 ) an elaborate tumor control strategy must be at least as cognitive as the immune system itself . thus granted that most ( if not all ) given classes of cognitive modules interact within their cultural environment , we may proceed to consider what happens when the environment _ is _ the communication medium itself . for instance , in typical ai laboratory models where multi - agent , inter - sensing systems function in local coordinated tasks ( often with no explicit communication between agents ) , the eventual net effect may induce a ` shared memory ' ( cao et al . , 1997 ; krieger et al . , 2000 ) where , for instance , it was claimed that more energetic and efficient foraging tasks were typical of multi - agent ( robotic ) systems compared to individual agents , and the former tended to produce behavioral patterns similar to those of ` ant like ' decentralized control systems . on the other hand , biological regulatory networks besides being susceptible to alterations in the environment and/or intracellular conditions , may operate stochastically in varying degrees as seen along the pathways of neuronal signalling transduction ( manninen et al . , 2006 ) that provides an analogy motivating part of this paper ( see [ ito - process ] ) . a similar scenario is that of neuronal ` trail systems ' ( ts ) in glade et al . ( 2009 ) : single wire and logical gates in a self organized bioprocessor along which self propelled particles communicate via traces etched out in the environment , thus creating the ts . the claim of glade et al . ( 2009 ) is that such systems , although not precisely defined , are capable of programming and function on the basis of a turing machine . it suggests an evolutionary factor by which various living beings , by activating their respective nervous systems , have trained themselves to use models like the ts towards evolution within and possible mastery over their environment , by means of simulating ` trail ' signals . more from an information - theory viewpoint , wallace ( 2009b ) envisages similar ideas to trail systems as reminiscent of ` roman roads ' : decision making and tasking within small local communities , eventually creating ` roads ' , manifestly inter connected cognitive modules having different time constraints , but eventually creating extensions of ` local consciousness ' . a rate distortion argument applies here to account for the mutual crosstalk between different modules using the homology of the rate distortion function with free energy as will be described later . next we proceed with some specifics . as is well known , distortion arises when there is a fast relay of information through some channel which exceeds the latter s capacity . one of the principles of the shannon theory is that in order to reproduce a message transmitted from a source to a receiver , it is necessary to know what sort of information should be transmitted and how . for the purpose of engineering a communication system , one needs to figure out a suitable encoding / decoding system once the nature of the channel is specified . following berger ( 1971 ) , we briefly recall some of the basic principles involved . * source encoder * : we may consider some output @xmath0 emanating from the source as projected to a finite set of preselected images , namely , the space of possible source outputs is partitioned into a set of _ equivalence classes _ and the source encoder informs the channel encoder of that class containing the particular source output observed . once the channel encoder is informed that the source output belongs to say , the @xmath1-th equivalence class , it transforms the corresponding waveform @xmath2 across the channel . * source decoder * : within the system is a cascade of a channel encoder and a source decoder . the channel decoder receives a waveform @xmath3 of a corresponding function @xmath4 over some time interval and decides upon the nature of the message as transmitted . then it sends its approximation @xmath5 of the message number to the source decoder which in turn creates @xmath6 to register the system s estimate of @xmath0 over that time interval . initially , we may think of @xmath0 and @xmath4 as ` waveforms ' , but in our case , we consider these as consisting of a language with its own intrinsic grammar / syntax , as well as ` meaning ' to be made more specific in [ meaningful ] . analogous considerations apply to the channel signals @xmath7 and @xmath3 ( see figure 2 ) . one of shannon s notable results was that a communication system can be designed such that it achieves a level of fidelity @xmath8 once the _ rate distortion _ @xmath9 , where @xmath10 denotes the channel capacity . putting it another way , if the receiver can tolerate an average amount of distortion @xmath8 , the rate distortion @xmath11 is the effective rate at which the source can relay information with that level of tolerance . the rate at which a source produces information subject to insisting upon perfect reproduction , is the _ source entropy _ @xmath12 . given a distortion measure such that perfect reproduction is assigned zero distortion , then we have @xmath13 . as @xmath8 increases , @xmath11 becomes a monotonically decreasing ( convex ) function which eventually is zero , typically at a maximum value for @xmath8 ( see berger , 1971 , chapter 1 ) . this is a very basic observation , and typically in rate distortion theory one seeks a reduction of @xmath12 by either slowing down the emission of coding , or encoding the relevant languages at a lower rate . in view of shannon s theorem , as long as @xmath14 , we will obtain appropriate fidelity in transmission . inherent difficulties are clear , since the source rate may be corrupted due to low memory and coding congestion , hence the need for a communicating system to evolve so as recover the source data at the channel output satisfying the shannon estimate . having mentioned the rate distortion function @xmath11 we now follow berger ( 1971 ) to give its specific definition in terms of _ average mutual information _ ( an alternative , and equivalent definition of @xmath11 and a statement of the rate distortion theorem will be given in appendix [ appendix - b ] ) . firstly , for @xmath15 running over a suitable alphabet , let us write a given conditional probability assignment as @xmath16 such that in the usual way we have an associated joint distribution @xmath17 . we express _ the average distortion _ as @xmath18 where @xmath19 denotes the distortion measure . a conditional probability assignment @xmath20 is said to be _ @xmath8admissible _ if and only if @xmath21 . the set of all @xmath8admissible conditional probability assignments we denote by @xmath22 along with an average distortion @xmath23 , we also have an _ average mutual information _ @xmath24.\ ] ] then for fixed @xmath8 , the rate distortion function is defined as @xmath25 observe that if a parameter @xmath26 represents the slope of the function @xmath11 at a point @xmath27 generated parametrically , we have @xmath28 ( this is not exactly trivial : see berger , 1971 , theorem 2.5.1 ) . more formally , a pattern of sensory input is mixed in an unspecified but systematic algorithmic manner with a pattern of internal ongoing activity to create a path of combined signals @xmath29 . each @xmath30 thus represents some functional composition of internal and external signals . wallace ( 2005 ) provides some neural network examples . this path is fed into a highly nonlinear , but otherwise similarly unspecified , decision oscillator , @xmath31 , which generates an output @xmath32 that is an element of one of two disjoint sets @xmath33 and @xmath34 of possible system responses . let @xmath35 assume a graded response , supposing that if @xmath36 the pattern is not recognized , and if @xmath37 the pattern is recognized , and some action @xmath38 takes place . such oscillators may be influenced by ` forcing ' when a signal is subjected to some impulse such that its frequency , and hence the response , adjusts accordingly with respect to that applied impulse . more familiar oscillating physical models react to this by exhibiting ` beats ' and ` resonance ' for instance . the principal objects of formal interest are paths @xmath39 which , through information flow , trigger pattern recognition - and - response . that is , given a fixed initial state @xmath40 , we examine all possible subsequent paths @xmath39 beginning with @xmath40 and leading to the event @xmath41 . thus @xmath42 for all @xmath43 , but @xmath44 . for each positive integer @xmath45 , let @xmath46 be the number of high probability grammatical / syntactical paths of length @xmath45 which begin with some particular @xmath40 and further leading to the condition @xmath47 . these are paths of combined signals as above , that are structured to some language . for short , we call such paths ` meaningful ' , assuming , not unreasonably , that @xmath46 will be considerably less than the number of all possible paths of length @xmath45 leading from @xmath40 to the condition @xmath41 . one critical assumption which permits an inference on the necessary conditions constrained by the asymptotic limit theorems of information theory , is that the finite limit @xmath48}{n},\ ] ] ( the ` uncertainty ' ) both exists and is independent of the path @xmath39 . the rate distortion principle applies as follows ( wallace , 2005 ) : _ the restriction to meaningful sequences of symbols increases the rate at which information can be transmitted with arbitrary small error , and that the grammar / syntax of the path can be associated with a dual information source_. here we may assume a typical information source @xmath49 to be ` adiabatic ' , ` piece - wise stationary ' and ` ergodic ' ( apse ) , and that a system engaging in a cognitive process is describable as such . we list here the explanations : * ` adiabatic ' means that the changes are slow enough to allow the necessary limit theorems to function . * ` stationary ' means that between pieces the probabilities hardly change , and ` piecewise ' means that these properties hold between phase transitions which are described using renormalization methods ( see wallace , 2005 ) . * ` ergodic ' means that in the long term , correlated sequences of symbols are generated at an average rate equal to their ( joint ) probabilities . more specifically , the essence of ` adiabatic ' is that , when the information source is parametrized according to some appropriate scheme , within continuous ` pieces ' of that parametrization , alterations in parameter values occur slowly enough so that the information source @xmath49 remains as close to stationary and ergodic as needed to put to work the fundamental limit theorems of information theory . in view of , the shannon uncertainty of @xmath49 can be stated more specifically by ( see e.g. cover and thomas , 1991 ) : @xmath50= \lim_{n { { \longrightarrow}}\infty } \frac{\log[n(n)]}{n}.\ ] ] . we recall how the information source uncertainty was defined as in equation . this is quite analogous to the free energy density of a physical system , equation , and the relevance to a cognitive process can be explained by the following steps . for instance , feynman ( 1996 ) provides a series of physical examples ( based in part on the research of c. h. bennett into the thermodynamics of computing ( bennett , 1982 ) where this homology is , in fact , an identity , at least for very simple systems . bennett ( 1982 ) argues , in terms of idealized irreducibly elementary computing machines , that the information contained in a message can be viewed as the work saved by not needing to recompute what has been transmitted , or as feynman ( 1996 ) puts it : the information contained in a message is proportional to the amount of free energy density needed to erase it . the essential argument is that computing , in any form , takes work . thus the more complicated a cognitive process , measured by its information source uncertainty , the greater its energy consumption , and our ability to provide energy to the brain is limited : typically , a unit of brain tissue consumes an order of magnitude more energy than a unit of any other tissue . the less information available to us concerning an event , the higher is its entropy , and information retrieved is not without a cost in expenditure of energy , where ` cost ' may be interpreted as the necessary number of bits needed to encode a message . the thermodynamic minimum of energy in terms of bits of information is @xmath51 erg / bit ( @xmath52 erg / nat ) . so efficiency in an information system is essentially when there is the minimum amount of energy expended in retrieving information . specifically , if @xmath53 is taken to denote the free energy , then taking @xmath54 is to denote the minimum number of nats / sec , the efficiency of the system is given by @xmath55 ( see e.g. berger , 1971 ) . in a similar spirit to bennett s work , li and vitnyi ( 1992 ) consider the thermodynamic costs of computation and how certain thermodynamic considerations can give a recursively invariant notion of ` cognitive distance ' using a kind of billiard dynamics approach . in this case a minimal cognitive distance between two objects will correspond to the minimal amount of work expended for a given cognitive transformation of objects , either by some computational procedure or by some neurocognitive function of the brain . a higher descriptive level leading to more complex and protracted algorithms , then leads to greater kolmogorov complexity ( li and vitnyi , 1993 ) . as a particular application of the bennett / feynman ideas in the global workspace setting , wallace ( 2007 ) argues that the cognitive disorder of ` inattentional blindness ' emerges as a thermodynamic limit on processing capacity in a topologically - fixed global workspace , i.e. one which has been strongly configured about a particular task . institutional and machine generalizations seem clear . many cognitive processes exhibit the patterns of dynamical systems ( see e.g. glazebrook and wallace , 2009a ) . in such systems one aims to unify the internal and external symmetries , and to be able to reduce vast myriad like network configurations into manageable schemes involving the corresponding equivalence classes analogous to those already mentioned in source encoding / decoding , etc . in [ rd - function ] ( see also [ equiv - info ] below ) . a precise way of doing this lies within the categorical concept known as a _ groupoid _ ( see brown , 2006 ; connes , 1994 ; weinstein , 1996 ) . in essence a groupoid @xmath56 consists of both a set of objects @xmath57 and a set of morphisms , or ` arrows ' , each of which project to an object in @xmath57 , and all such morphisms admit an inverse . the most familiar example of a groupoid , as known to students of algebra , is that of a ` group ' where there is a single object ( ` the identity ' ) . hence groupoids can be viewed as extensions of the ` group ' concept to sets of _ multiple identities _ thus providing a wide scope of applications to the dynamics of neurocognitive and socio bioinformatic systems ( see e.g. baianu et al . , 2006 ; glazebrook and wallace , 2009a , 2009b ; golubitsky and stewart , 2006 ; stewart et al . , 2003 ; wallace 2005 ; wallace and fullilove , 2008 ) . a groupoid can be depicted by @xmath58^{{\alpha } } \ar[r]_{{\beta } } & x}\ ] ] where the groupoid morphisms @xmath59 onto objects , are called the _ range _ and _ source maps _ , respectively . informally , the groupoid represents a feature of built in reciprocity between its algebraic structures , internalizing and externalizing the prevailing symmetries . the morphisms @xmath60 satisfy certain algebraic relations of associativity , existence of two - sided identities , etc . ( details can be seen in e.g. brown , 2006 ; connes , 1994 ; weinstein , 1996 ) . a groupoid can here be understood in relationship to a linkage by a meaningful path of an information source dual to a cognitive process for which the underlying principle is that : _ states @xmath61 in a set @xmath62 are related by the groupoid morphism if and only if there exists a high probability grammatical path connecting them to the same base point , and the tuning across the various possible ways in which that can happen the different cognitive languages parametrizes the set of equivalence relations and creates the groupoid . _ [ ex2 ] since we have already mentioned equivalence classes in the context of source encoding / decoding , it seems appropriate to see how an equivalence relation @xmath63 defined on ( a set ) @xmath57 takes shape as a groupoid . here we have the two projections @xmath64 , and a product @xmath65 whenever @xmath66 together with an identity , namely @xmath67 , for each @xmath68 . moreover , the essential equivalence relations ( classes ) derived from a systems space ( network ) arise from the orbit equivalence relation of some groupoid @xmath56 acting on that space ( see e.g. weinstein , 1996 ) . in the context of connected ( sub)networks / graphs which can reduced to equivalence classes , natural groupoid structures come about in accordance with equivalence classes of relations @xmath69 , as above , that is simply interpreted as having an edge linking node @xmath39 to node @xmath70 . conversely , a groupoid ( of equivalence relations ) admits an underlying graph structure via its implicit scheme of objects and morphisms between objects ( for details , see e.g. brown , 2006 ) . thus we have the two - way associations whereby ` objects ' can be identified with ` nodes ' , and ` morphisms ' identified with ` edges ' in groupoids ( of equivalence relations ) and networks , respectively:@xmath71 an important observation about multi - tasking in institutional and distributed cognitive systems concerns how the various submodules interact . when a given subnetwork is represented in its groupoid form , then such interactions naturally can be realized in terms of _ groupoid actions _ , a topic that warrants further attention . then we would like to see the overall cognitive system in terms of such interacting groupoids and designed by an ` atlas ' of the latter , and one , as shown in glazebrook and wallace ( 2009a ) , containing the representation of several possible emergent ` giant components ' induced by the outcome of local group(oid ) actions within the workspace ( see figure 3 ) . a workable concept seems to be that of a _ groupoid atlas _ ( bak et al . , 2006 ) which provides a schematic representation for coupling interactions between multi - agent systems and uses a pasting together of the local dynamic groupoid actions with the net effect of a ` global ' groupoid . one commences from a family of dynamically interacting groupoids @xmath72 where each groupoid has the same set of objects ; this family is called a _ single domain _ or _ multiple groupoid_. _ a groupoid atlas _ is then defined as a set with a covering by patches each of which comprise a single domain with global action , representing the local processing which is then globalized across the atlas . this is a desirable effect and one particularly suited to logically inscribing processors or sensors ( the ` agents ' ) within the cognitive modules of the workspace . as a descriptive mechanism , this atlas has the advantage of admiting a weaker structure compared with that of a conventional manifold since no condition of compatibility between arbitrary overlaps of the patches is necessary . this is a key property relevant to the structure of cognitive modules that can be geared to equivalence class representations where flexibility in the structure is a natural characteristic . in this way , the atlas provides a convenient description of a web of complexity representing the dynamic reciprocity of tightly - knitted functional systems as was applied to small world networks ( glazebrook and wallace , 2009a ) . time and space does not permit including a mathematical outline of the construction ; for the technical details we refer the reader to bak et al . ( 2006 ) and del hoyo and minian ( 2009 ) . however , we intend to apply this concept with some essential details in [ info - atlas ] below . derived from the equivalence relations of their respective networks of multi - agent type cognitive modules . they are viewed as the components of a groupoid atlas . the groupoid actions are indicated by dotted arrows , in this case , between @xmath73 and @xmath74 , and may represent the formation of network linkages as the system is shifted via crosstalk , and for instance , how a ` giant component ' emerges ( see e.g. glazebrook and wallace , 2009a ) . this is a descriptive mechanism that could be seen , for instance , as enveloping the underlying networks of the ` reactive ' system in figure 1 , and is applicable to the content of [ info - atlas ] . , width=113 ] the information supported interactive cognitive modules we have in mind are assumed to possess their own internal metabolisms and mechanisms of self - organization as reflective of vital biochemical processes . just as for the latter , the evolution of these ` cognitive cells ' is characterized by reactive states of thermal nonequilibrium , in accord with the laws of thermodynamics , and the capability to assimilate information which we study in respect of source uncertainty . this we can achieve by applying the onsager relations of nonequilibrium thermodynamics ( kurzynski , 2006 ; landau and lifshitz , 2007 ) . the reasoning starts by observing how a fundamental homology between the information source uncertainty dual to a cognitive process and the free energy density of a physical system can arise mainly from the formal similarity between their definitions in the asymptotic limit . recall that for a thermodynamic state of a given system at fixed temperature @xmath75 with energy @xmath76 and entropy @xmath77 , the _ free energy density _ @xmath53 is defined to be @xmath78 in the hamiltonian formulism one takes the volume @xmath79 and the partition function @xmath80 derived from the system s hamiltonian at inverse temperature @xmath81 ( kurzynski , 2006 ; landau and lifshitz , 2007 ) . the free energy density is then defined to be @xmath82 & = \lim_{v { { \longrightarrow}}\infty } -\frac{1}{k } \frac{\log[z(k , v)]}{v } \\ & = \lim_{v { { \longrightarrow}}\infty } \frac{\log[{\widehat}{z}(k , v)]}{v } , ~ \text{where}~ { \widehat}{z}= z^{-\frac{1}{k}}. \end{aligned}\ ] ] consider now an information source @xmath83 over a corresponding groupoid @xmath84 . the probability of @xmath83 is given by : @xmath85}{\sum_{\beta } \exp[-h_{{{\mathsf{g}}}_{\beta}}k]}~,\ ] ] where the normalizing sum is over all possible subgroupoids of the largest available symmetry groupoid . now let @xmath86.\ ] ] _ the groupoid free energy density ( gfe ) of the system @xmath87 at inverse normalized equivalent temperature @xmath81 _ is defined as @xmath88 = - \frac{1}{k } \log [ z_{{{\mathsf{g}}}}(k)].\ ] ] with each such groupoid @xmath84 of the ( large ) cognitive groupoid , we can associate a dual information source @xmath83 . we recall the rate distortion function between the message sent by the cognitive process and the observed impact , while noting that both @xmath83 and @xmath11 may be considered as free energy density measures . in a sense , @xmath11 constitutes a sort of ` thermal bath ' for the process of cognition . then the probability of the dual information source can be expressed by @xmath89}{\sum_{\beta } \exp[-h_{{{\mathsf{g}}}_{\beta}}/ \kappa r(d ) \tau]}~,\ ] ] where @xmath90 denotes a suitable dimensionless constant characteristic of the system in the context of a fixed machine response time @xmath91 . the sum is over all possible subgroupoids of the largest available symmetry groupoid . accordingly , the term @xmath92 represents a ` rate distortion energy ' , in this case , a kind of ` temperature analog ' . in the context of a fixed @xmath91 , a decline in @xmath11 ( on increase in average distortion ) , acts to ` lower the machine temperature ' , driving it to more simple and less less rich behaviors . the groupoids @xmath84 can indeed be taken to comprise a groupoid atlas @xmath93 on an appropriate set @xmath94 . in bak et al . ( 2006 ) this is motivated by considering a group @xmath95 in the standard algebraic sense , and a family @xmath96 of group actions ` @xmath97 ' on subsets @xmath98 , where the local groups @xmath99 and the corresponding subsets @xmath100 are indexed by an indexing set @xmath101 called _ the coordinate system of @xmath102 _ which is seen to satisfy certain conditions ( bak et al . , now the family of local groups can be replaced by a family of local groupoids @xmath103 defined with respective object sets @xmath100 , and with a coordinate systems @xmath104 that is equipped with a reflexive relation denoted by @xmath105 . this data is to satisfy the following conditions ( bak et al . , 2006 ) : * if @xmath106 in @xmath104 , then @xmath107 is a union of components of @xmath103 , that is , if @xmath108 and @xmath109 acts as @xmath110 , then @xmath111 . * if @xmath106 in @xmath104 , then there is a groupoid morphism defined between the restrictions of the local groupoids to intersections @xmath112 and which is the identity morphism on objects . thus each of the @xmath84 with its associated dual information source @xmath83 constitutes a component of an atlas which incorporates the dynamics of an ( inter)reactive system through information sources , by means of the intrinsic ( groupoid ) actions . for simplicity , let us refer to this atlas as @xmath102 . suppose we have another such atlas @xmath113 representing a separate system which can be related to @xmath102 via a suitable transformation . to make matters precise we consider then a morphism @xmath114 prescribed by a triple @xmath115 satisfying ( del hoyo and minian , 2009 ) : * @xmath116 is a set - theoretic function . * @xmath117 is a function that preserves the relation @xmath105 . * @xmath118 is a ( generalized ) natural transformation of groupoid diagrams over the function @xmath119 which restricts to @xmath120 on objects . these conditions can summarized in the following straightforward way . for each @xmath121 , a function @xmath122 is given , such that for objects @xmath123 , and if @xmath106 , the diagram @xmath124 is commutative . further motivation is provided by considering phases as chemically and thermodynamically homogeneous when formally compared with average mutual information such as applied to living systems that are capable of assimilating and using information reaped either from their environment or from information that is intrinsic to their particular system via genetic characteristics . here we reflect in part upon the theme of [ feynman ] in which we discussed certain physical means by which information runs at the cost of expending free energy . on course with the fundamental homology , it befits us to consider some physical equations that are homologous to the variational calculus of the rate distortion function , and follow to some extent berger ( 1971 , 6.4 ) . the idea is that when a living information system is confronted with an environment , it reacts towards it at an atomic - molecular level . given then some thermodynamic system , let @xmath125 be the number of atomic weights of substance @xmath126 that end up in phase @xmath127 , and let @xmath128 be the number of atomic weights that were introduced originally , so that @xmath129 . the multiphase ( chemical ) equilibrium problem involves determining the @xmath130 . to proceed , let us express the free energy as @xmath131 , where @xmath132 , \end{aligned}\ ] ] where the @xmath133 are _ the free energy constants _ , and where the term contained in the logarithm is the chemical potential of the reactant @xmath126 in phase @xmath127 . let now @xmath134 ( for some @xmath135 ) , @xmath136 and @xmath137 , then @xmath138 \big ] - n \sum_j p_j \log p_j.\ ] ] since in principle the @xmath45 and @xmath139 can be determined , the minimization of @xmath53 reduces to minimizing the double summation on the right side of . letting @xmath140 , then ` nature ' selects the @xmath130 so as to minimize the quantity @xmath141 -s d_{jk } \big].\ ] ] the crucial observation here is that is formally the same as @xmath142 in other words , the function to be minimized is the difference between the average mutual information and @xmath26 times the average distortion . this is minimized by an appropriate choice of @xmath143 such as to determine a point on the @xmath11 curve where , as in [ average ] , we have @xmath144 . the multiphase chemical equilibrium is a local process ; the overall system itself may in general remain in a state far - from - equilibrium . according to berger ( 1971 ) , ` nature ' then automatically performs the analogous minimization in multiphase chemical equilibrium , and when an information system encounters an environment , the reaction on the molecular level follows these principles . thus the free energy of the combined system and the environment is minimized , although this is at the cost of the system s free energy capacity where there will inevitably be some heat dissipation . eventually the system can fine - tune its coding mechanism and re - configures itself for the task of gaining more specialized knowledge about its environment , and thanks to newly acquired information , it may engage the latter to its advantage along with detecting other drifting cells of information . how such optimizing procedures can be realized in an evolutionary context , is of course one of the main tasks of applying rate distortion arguments ( cf wallace and wallace , 1998 , 1999 , 2008 , 2009 ) . for instance , a scenario studied in tlusty ( 2007 , 2008 ) concerns how the genetic code maps the 64 nucleotide triplets ( codons ) to 20 amino acids . in terms of this mapping , the code is viewed as a noisy information channel and the claim is that evolutionary characteristics determine the emergence of the code via appropriate selection of amino acids which minimise the risk of errors , and subsequently the code emerges at a ` supercritical ' phase transition once the mapping ceases to be random . understanding the time dynamics of cognitive systems away from phase transition critical points thus requires a phenomenology similar to the onsager relations . if the dual source uncertainty of a cognitive process is parametrized by some vector of quantities @xmath145 , then , in analogy with nonequilibrium thermodynamics , the gradients in the @xmath146 of the _ disorder _ , defined as @xmath147 become of central interest . equation is similar to the definition of entropy in terms of the free energy density of a physical system , as suggested by the homology between free energy density and information source uncertainty described above . pursuing the homology further , the generalized onsager relations defining temporal dynamics become @xmath148 where the ( kinetic coefficients ) @xmath149 are , in first order , constants interpreted as reflecting the nature of the underlying cognitive phenomena . the partial derivatives @xmath150 are analogous to thermodynamic forces in a chemical system , and may be subject to override by external physiological driving mechanisms as shown in ( wallace , 2005 ; wallace and fullilove , 2008 ) along with further extensions of these dynamical procedures . equation is ` general ' in the sense that we do not necessarily assume the symmetry condition @xmath151 which in this latter case expresses onsager s 4-th law of thermodynamics ( see e.g. ( 3.7 ) in kurzynski , 2006 ) . the matrix @xmath152 $ ] is to be viewed empirically , in the same spirit as the slope and intercept of a regression model , and may have a structure far different when compared to the more basic , more familiar chemical or physical processes . generally , information sources are notoriously one way in time , as exemplified by the patent linguistic scarcity of palindromic structures that do actually make some sense . equations and can be derived in a simple parameter - free covariant manner which relies on the underlying topology of the information source space implicit to a process . different cognitive phenomena have , according to our development , dual information sources , and we are interested in the local properties of the system near a particular reference state . we impose a topology on the system , so that , near a particular ` language ' @xmath62 , dual to an underlying cognitive process , there is ( in some sense ) an open set @xmath153 of closely similar languages @xmath154 , such that @xmath62 and @xmath154 are subsets of @xmath153 . note that it may be necessary to coarse - grain the system s responses to define these information sources . the problem is to proceed in such a way as to preserve the underlying essential topology , while eliminating ` high frequency noise ' . the formal tools for this can be found in e.g. burago et al . ( 2001 ) . since the information sources dual to the cognitive processes are similar , for all pairs of languages @xmath155 in @xmath153 , it is possible to make use of the following : * create an embedding alphabet which includes all symbols allowed to both of them . * define an information - theoretic distortion measure in that extended , joint alphabet between any meaningful ( i.e. high probability grammatical / syntactical ) paths in @xmath62 and @xmath154 , which we write as @xmath156 ( ash , 1990 ; cover and thomas , 1991 ) . note that these languages do not interact , in this approximation . * define a metric on @xmath153 , for example , @xmath157 using an appropriate integration limit argument over the high probability paths . the usual metric properties apply as in burago et al . ( 2001 ) . note that these conditions can be used to define equivalence classes of _ languages _ , where previously we defined equivalence classes of _ states _ which could be linked by meaningful paths to some base point . this led to the characterization of different information sources from which a formal topological manifold , _ which is an equivalence class of information sources _ , can be constructed . as a working hypothesis we may assume this to be a standard differentiable manifold in which the set of such equivalence classes generates the _ dynamical groupoid _ ( cf glazebrook and wallace , 2009a , 2009b ) , and then study those mechanisms , internal or external , which can break that groupoid symmetry . in particular , the imposition of a metric structure on this groupoid , and on its base set , would permit a nontrivial interaction between orbit equivalence relations and isotropy groups , leading to interesting algebraic structures . since @xmath12 and @xmath158 are both scalars , a ` covariant ' derivative can be defined directly as @xmath159 where @xmath160 is the source uncertainty of language @xmath62 . suppose the system is set in some reference configuration @xmath161 . to obtain the unperturbed dynamics of that state , impose a legendre transform using this derivative , defining another scalar @xmath162 the simplest possible onsager relation here seen as an empirical , fitted , equation like a regression model in this case becomes @xmath163 where @xmath164 is the time and @xmath165 represents an analog to the thermodynamic force in a chemical system ( cf [ multiphase ] ) . relevant here are patterns of oscillatory - like behavior where a weak signal is amplified by the presence of noise as result of some synchronized hopping around local extrema . the standard terminology for this phenomenon is _ stochastic resonance _ ( gammaitoni et al . , 1998 ) and we will proceed to give some idea of how this can be related to certain types of cognitive processes . since we are working with stochastic differential equations , the first step is to modify the equation of thermodynamic force accordingly . to this extent equation is rewritten as @xmath166 where @xmath167 is a constant and @xmath168 represents a white noise term . again , the quantity @xmath77 is seen as a function of the parameter @xmath158 . this leads directly to a family of classic stochastic differential equations expressed as differential 1-forms @xmath169 where @xmath170 and @xmath167 are appropriately regular functions of @xmath164 and @xmath158 , and @xmath171 represents the noise structure . such cognitive - epigenetic systems which are driven by stochastic and noise driven diffusion processes , may be suitably conditioned to admitting further noise perturbations that lead to a degree of stochastic resonance capable of amplifying a relatively weak signal or actually reducing the level of randomness in the system . such resonance may function as a catalyst towards the system s self - organization and complexity , in the same way as open systems far - from - equilibrium require internal amplification in order to reach a macroscopic dynamical structure ( gammaitoni et al . , 1998 ; prigogine , 1980 ; west et al . , 2005 ) . recall that the rate distortion function @xmath11 defines the minimum channel capacity necessary for the system to have an average distortion @xmath172 , thus imposing a limit on the information source uncertainty and suggesting how distortion measures can drive information system dynamics . in other words , @xmath11 affords a homological relation to free energy density , very much along the lines of the above relation between free energy density and information source uncertainty . accordingly , it is proposed that the dynamics of cognitive modules interacting in characteristic real time @xmath91 will be constrained by the system as described in terms of @xmath11 , but now we generalize matters as in wallace and wallace ( 2008 ) by producing a vector valued function @xmath173 where in the vector @xmath174 the first component is defined to be the average distortion , and then ( cf ) @xmath175 which leads to the deterministic and stochastic systems of equations analogous to the onsager relations of nonequilibrium thermodynamics @xmath176 together with @xmath177 where the @xmath178 represents often highly structured stochastic noise whose properties may be described in terms of brownian motion and quadratic variation ( see e.g. kunita , 1990 ; protter , 1995 ) . at this stage we introduce several examples for which part of the purpose will be to motivate introduction of the it principle which we will do so below . [ rdd - ex1 ] firstly , for a simple gaussian channel with noise having zero mean and variance @xmath179 , we have @xmath180 the simplest possible onsager relation becomes @xmath181 in which the term @xmath182 represents the force of an ` entropic wind ' which is a kind of internal dissipation inevitably driving the real time system of interacting cognitive information sources toward greater distortion . equation has a solution @xmath183 , showing in this case that the average distortion increases monotonically with time . following wallace ( 2009a , 7.2 ) , this example shows that such a system will inevitably succumb to a relentless entropic force , requiring _ a constant free energy expenditure for maintenance of some fixed average distortion within the system s communication between them_. the distortion in this case will , without free energy input , have a time dependence @xmath184 , with @xmath185 montonically increasing in @xmath164 , eventually leading to the punctuated failure of the system . further , in the einstein diffusion equation , a straightforward argument of wallace ( 2009a , 7.2)shows that the standard deviation of the particle position increases in proportion to @xmath186 . thus , whereas we do not expect the high correlations of an information source to exhibit typical brownian motion , it does seem to be the case that the _ distortion _ in communication between the interacting cognitive modules within the appropriate context of the onsager relations , does display brownian motion which may be of bounded variation in certain cases . here we consider different cognitive developmental subprocesses of gene expression characterized by information sources @xmath187 interacting through chemical or other signals , and assume _ that different processes become each other s principal environments _ which is a suitable hypothesis within a broad coevolutionary context . let @xmath188 where the @xmath189 represent other relevant parameters , and @xmath190 . we regard the dynamics of this system as driven by a recursive network of stochastic differential equations . letting the @xmath191 and @xmath187 all be represented as parameters @xmath192 ( with the caveat that @xmath187 does not depend on itself ) , we follow the generalized onsager formulation of wallace and wallace ( 2009 ) , in terms of the equation @xmath193 to obtain a recursive system of _ phenomenological onsager relations _ , in terms of a system of stochastic differential equations @xmath194,\ ] ] in which , for ease of notation , both the terms @xmath195 and the external @xmath191 s are expressed by the same symbol @xmath192 . as @xmath1 ranges over the @xmath187 we could allow different kinds of ` noise ' @xmath178 , having particular forms of quadratic variation which may represent a projection of environmental factors within the scope of a rate distortion manifold ( glazebrook and wallace , 2009b ) . the next step in extending is to bring in rate distortion functions for mutual crosstalk between a set of interacting cognitive modules by using the homology of @xmath11 itself . to this extent , consider different cognitive processes indexed @xmath196 , and take the mutual rate distortion functions @xmath197 characterizing communication ( and distortion ) between them . at the same time the essential parameters remain the characteristic time constants of each process , @xmath198 , for @xmath199 , together with an overall embedding free energy density @xmath53 . taking the @xmath200 to run over all the relevant parameters and mutual rate distortion functions ( along with the distortion measures @xmath201 ) , then now takes shape as @xmath202 and accordingly becomes @xmath203.\ ] ] this last equation generalizes the treatment in terms of crosstalk , its distortion , the inherent time constants of the different cognitive modules , and _ the overall available free energy density_. [ coev - ex1 ] for a gaussian channel and fixed embedded communication free energy density @xmath53 representing the richness of incoming information from the interacting cognitive modules , we extend to @xmath204 that represents the communication distortion between the modules . the equilibrium solution is @xmath205 . the difference between and is that whereas in the former case , the distortion grows directly as the square root of the elapsed time , equation reveals there is a finite , equilibrium , average distortion that is inversely proportional to the available environmental or informational free energy that the interacting systems can implement in order to navigate their actions . the above situation can be generalized to @xmath206 , where @xmath207 is monotonically increasing in @xmath53 . on introducing a characteristic response time variable @xmath91 , so that @xmath208 where @xmath209 is also monotonically increasing , leads to @xmath210 this example reveals that given a fixed rate of available information free energy , the increasing allowable response time _ decreases _ average distortion in the interaction . [ coev - ex2 ] suppose now that feedback is allowed so that the system actively seeks information in proportion to the distortion between intent and impact , then the onsager relation for a gaussian channel becomes @xmath211 and @xmath212 which is significantly smaller than , and is effectively the classic result for brownian motion in a harmonic central field ( e.g. equation ( 54 ) of wang and uhlenbeck , 1945 ) . together with the multiphase equilibrium problem of [ multiphase ] we have so far pursued a theme of how the stochastic simulation of biochemical systems closely parallels that of evolutionary genetic systems . an initial observation here , and one that will motivate further ideas , is that a stochastic differential equation of the type should in principle model the dynamics of large , intricate networks that are constrained by the costs of actual computational time , which seems relevant in certain senses to many epigenetic processes . the general setting for such processes often involves that of an it stochastic de ( viz _ it process _ ) and this is how we intend to view as a kind of ` master equation ' . for the readers sake , we remark why such a level of mathematical formality is necessary by recalling a basic difficulty : the sample paths @xmath213 of brownian motion are not in general functions of bounded variation , so that @xmath214 is not defined as for that of the usual riemann - stieltjes integral . one may start by supposing that for almost all samples , @xmath215 in is independent of future brownian motion @xmath216 ; otherwise said @xmath217 is an _ adapted _ stochastic process . putting it another way , the information available at a given time includes the history of the process at that time . more generally , the stochastic process may be taken to be _ predictable _ in order to define the it integral of the equation ( see e.g. 2.3 of kunita , 1990 and appendix [ appendix - a ] here ) . on the other hand , there is part justification for making assumptions of ( local ) boundedness of variation in the brownian motion incurred via distortion , and in particular , the adapted condition , points to those examples and observations already quoted that concern the dispensation of available free energy . firstly , in view of example [ rdd - ex1 ] we may expect a constant free energy expenditure for maintenance of some fixed average distortion in communication between the interacting cognitive modules . secondly , given a fixed rate of available information free energy , the increasing allowable response time _ decreases _ average distortion in the interaction ( example [ coev - ex1 ] ) , and thirdly , the possibility that finite , equilibrium , average distortion is actually inversely proportional to the available environmental or informational free energy that the interacting systems can utilize ( example [ coev - ex2 ] ) . however , to make matters precise , it is appropriate to consider the formalities of some filtered probability space @xmath218 , where @xmath219 ( see appendix [ appendix - a ] ) and postulate a multidimensional stochastic process given by the it integral of the master equation : @xmath220.\ ] ] in order to state the conditions for which this process is well - defined , we first express in the simplified form @xmath221 then following harrison ( 1985 , chapter 4 ) : * @xmath222 is measurable with respect to @xmath223 . * @xmath224 is an adapted stochastic process , and @xmath225 , for all @xmath226 . * the integral of ` drift ' , @xmath227 , is a continuous and adapted variation - finite ( vf ) process . granted that our observations about the local boundedness of the ( rate distortion ) brownian motion as essentially fulfilling these conditions , we now have an explicit stochastic model for the role of cross - talk , its distortion , the inherent time constraints of the different cognitive modules , as well as the overall available free energy density , where the @xmath228 parameter structure represents the full scale fragmentation of the system in the presence of some weiner noise . a further possible generalization of is to introduce into that expression a matrix valued function @xmath229 to describe the intrinsic reaction rates ( cf manninen et al . , 2006 ) : @xmath230.\ ] ] towards the possibility of a stochastic flow generated by ( such as a brownian flow of diffeomorphisms ) , we opt to simplify accordingly . following kunita ( 1990 ) we write in a simplified form as @xmath231 typically , we would seek a solution starting from some @xmath39 at some time @xmath26 ; let us call these solutions @xmath232 . then a stochastic flow can be represented as the solution of a stochastic differential equation of the type @xmath233 where in this case , brownian motion @xmath234 valued in vector fields , is given by @xmath235 the formal conditions for to produce a brownian flow of diffeomorphisms are ( kunita , 1990 ) : * @xmath236 is continuous in @xmath237 . * the map @xmath238 ( for suitable @xmath239 ) is a diffeomorphism for any @xmath240 . * @xmath241 , for any @xmath242 . note that conditions ( 1)-(3 ) are ` almost everywhere ( surely ) ' conditions . also , such a flow generates a _ holonomy _ or _ geometric phase ( transition ) _ which can be explained by the process of tracking internal states in relationship to a spatiotemporal orientation . in more precise differential - geometric terms , holonomy results from the parallel transport of vectors around a closed path , thus leading to a representation of the space of the latter into a group of global symmetries . the procedure for constructing a holonomy groupoid associated to this flow concerns some mathematical technicalities , but is nevertheless standard ( see e.g. connes , 1994 ; moerdijk and mrun , 2003 ) . whereas this case is fairly well tempered , we would in general expect ` singularities ' in the flow . the holonomy groupoid can be still be constructed , but this involves deeper mathematics outside of the scope of this paper ( see e.g. debord ( 2001 ) for details ) . we have given here a descriptive account of cognitive modules as components of epigenetic - evolutionary systems ( as far - from - equilibrium open systems ) embedded within the context of environment and culture . using dynamic groupoids of network equivalence classes we have put into atlas form the various constituent ( inter)reactive systems based on rate distortion principles of the shannon theorems and the groupoid free energy density . the rate distortion function @xmath11 determines a channel capacity that is measurable in an analogous way to a free energy that regulates many of the ( inter)reactive /reciprocating processes that have been described . rate distortion arguments suggest that if an external information source is pathogenic , then sufficient exposure to it within a developmental stage will likely result in a image inscribed on mind and body in a punctuated fashion , subsequently causing a developmental dysfunction . in this analogy , the reduction of the @xmath11 amounts to ` lower temperature ' which in turn directs the system to behavioral patterns which are less enriched and are less complex . further , accurate and efficient communicating systems require a greater channel capacity , and keeping in mind the analogy with free energy density , a higher rate of metabolism is necessary and further costs are incurred ( cf [ feynman ] ) . failure to provide such resources equates to a decline in processing , possibly to a point of disintegration . on the other hand , increased communication between the system s cognitive modules , depending on the availability of free energy , will usually be followed up by a phase transition ( in essence , this is what the holonomy groupoid encodes as shown in glazebrook and wallace ( 2009a ) ) inducing further complexity into the systems behavior . in principle one might envisage an associated _ holonomy groupoid atlas _ for interacting cognitive - dynamical systems based on synchronous ( geometric ) phase transitions of the various constituents . this would amount to a broad - scale descriptive artifact for the purpose of understanding the cumulative transitional mechanisms of living processes that may eventually uncover some even deeper conceptual issues . stochastic processes are perhaps more in keeping with what the world expects compared to a strictly deterministic approach , though the former are likely to entail higher computational costs and some neuroscientific work in this area is aimed at reducing such costs ( cf manninen et al . , 2006 ) . integration of the onsager stochastic differential equation towards a multidimensional it process , leads naturally to a stochastic flow which in our formulation diffuses across the atlas through mainly noisy channels ( cf [ multiphase ] ) . such evidence is provided by tlusty s rate - distortion analysis of the genetic coding map where it is shown that the evolution of the code unfolds as a smooth flow on the codon space as realized in tlusty ( 2007 ) . with such examples in mind , we have presented here a novel development , very much in tune with the `` flow '' nature of living systems as once envisaged by bertalanffy and prigogine , while at the same time our approach using rate distortion principles embraces many central features of related theories of cognition such as those proposed in atlan and cohen ( 1998 ) , baars ( 1988 ) , baars and franklin ( 2003 ) , cohen and harel ( 2007 ) and maturana and varela ( 1980a , 1980b ) . we state some basic details as to be found in harrison ( 1985 ) , kunita ( 1990 ) and protter ( 1985 ) . let @xmath243 be a set . a collection @xmath244 of subsets of @xmath243 is called a _ @xmath167-field _ if it contains an empty set and it is closed under the operations of countable unions and complements . it is customary to call @xmath245 a _ measurable space _ in which members of @xmath243 are called _ samples _ and those of @xmath244 are called _ events_. let @xmath246 be a @xmath167-additive measure on @xmath247 . it is called a _ probability _ if @xmath248 . the triple @xmath249 is then called a _ probability space_. let @xmath250 be a family of @xmath167-algebras on @xmath243 such that a ) @xmath251 , for all @xmath252 , and b ) @xmath253 , if @xmath254 . then @xmath255 is said to be a _ filtration _ ( an increasing sequence of sub-@xmath167-algebras ) of @xmath245 . the filtration @xmath255 characterizes how information arises ( how uncertainty is resolved ) and @xmath256 may be interpreted as the set of all events whose occurrence or nonoccurrence will be determined at time @xmath164 . in a filtered probability space it is usually understood that @xmath257 . a stochastic process @xmath258 , is said to be _ adapted _ ( relative to @xmath218 ) if @xmath259 is measurable with respect to @xmath256 , for all @xmath252 . loosely speaking , this means that the information available at time @xmath164 includes the history of @xmath260 up to that point . the _ predictable @xmath167-field _ is the least @xmath167-field in the product space @xmath261 \times \omega$ ] for which all continuous @xmath262-adapted processes are measurable . _ a predictable process _ is then defined as a process that is measurable with respect to the predictable @xmath167-field . an example is a continuous @xmath256-adapted process . recall that a process of random variables @xmath263 is brownian motion ( viz a weiner process ) if and only if * @xmath264 with probability @xmath265 . * for @xmath266 the increment @xmath267 is normally distributed @xmath268 . * for @xmath269 , the set of increments @xmath270 is a set of independent random variables ( that is , the increments are independent of the past ) . a process @xmath228 is called a _ @xmath271-brownian motion _ if it has the form @xmath272 where @xmath260 is a weiner process and @xmath273 is independent of @xmath260 . then we have @xmath274 . invoking the spirit of the shannon - mcmillan theorem , it is possible to define an apse information source @xmath49 associated with stochastic variates @xmath275 having joint and conditional probabilities @xmath276 and @xmath277 such that appropriate joint and conditional shannon uncertainties satisfy the classic relations @xmath278 & = \lim_{n { { \longrightarrow}}\infty } \frac{\log[n(n)]}{n } \\ & = \lim_{n { { \longrightarrow}}\infty}h(x_{n}|x_{0 } , \ldots , x_{n-1})\\ & = \lim_{n { { \longrightarrow}}\infty } \frac{h(x_{0},\ldots , x_{n})}{n}. \end{aligned}\ ] ] this information source is defined as _ dual _ to the underlying ergodic cognitive process ( wallace , 2005 ) . recall that the shannon uncertainties @xmath279 are cross - sectional law - of - large - numbers sums of the form @xmath280 $ ] , where the @xmath281 constitute a probability distribution ( for the basic details , see ash,1990 ; berger , 1971 ; cover and thomas , 1991 ; khinchin , 1957 ) . messages from an information source , seen as symbols @xmath282 from some alphabet , each having probabilities @xmath283 associated with a random variable @xmath57 , are ` encoded ' into the language of a ` transmission channel ' , a random variable @xmath284 with symbols @xmath285 , having probabilities @xmath281 , possibly with error . someone receiving the symbol @xmath285 then retranslates it ( without error ) into some @xmath286 , which may or may not be the same as the @xmath282 that was sent . more formally , the message sent along the channel is characterized by a random variable @xmath57 having the distribution @xmath287 the channel through which the message is sent is characterized by a second random variable @xmath284 having the distribution @xmath288 let the joint probability distribution of @xmath57 and @xmath284 be defined as @xmath289 and the conditional probability of @xmath284 given @xmath57 as @xmath290 then the shannon uncertainty of @xmath57 and @xmath284 independently and the joint uncertainty of @xmath57 and @xmath284 together are defined respectively as @xmath291 the _ conditional uncertainty _ of @xmath284 given @xmath57 is defined as @xmath292.\ ] ] for any two stochastic variates @xmath57 and @xmath284 , we have the inequality @xmath293 , as the knowledge of @xmath57 generally gives some knowledge of @xmath284 . equality occurs only in the case of stochastic independence . since @xmath294 , it is deduced @xmath295 the information transmitted by translating the variable @xmath57 into the channel transmission variable @xmath284 possibly with error and then retranslating without error the transmitted @xmath284 back into @xmath57 , is defined as @xmath296 where we refer to berger ( 1971 ) , cover and thomas ( 1991 ) and khinchin ( 1957 ) for details . the essential point is that if there is no uncertainty in @xmath57 given the channel @xmath284 , then there is no loss of information through transmission . in general this will not be true , and herein lies the essence of the theory . following wallace ( 2005 ) , suppose we have an ( ergodic ) information source @xmath284 with output from a particular alphabet generating sequences of the form @xmath297 ` digitalized ' in some sense , and induce a chain of ` digitalized ' values @xmath298 where the @xmath299alphabet is considered more restricted than the @xmath70alphabet . in this way , @xmath300 is deterministically retranslated into a reproduction of the signal @xmath301 . that is , each @xmath300 is mapped onto a unique @xmath45length @xmath70sequence in the alphabet of @xmath284 : @xmath302 we remark that many @xmath301 sequences may be mapped onto the same retranslation sequence @xmath303 , the set of which is denoted @xmath304 ; this may be interpreted as a loss of information . a distortion measure @xmath305 , between paths @xmath301 and @xmath303 is defined as @xmath306 for some suitable distance function @xmath307 ( such as the hamming distance ) . suppose that with each path @xmath308 and each @xmath300path retranslation @xmath309 into the @xmath70language , we consider the associated individual , joint , and conditional probability distributions @xmath310 the average distortion is then defined to be @xmath311 for the corresponding strings @xmath284 ( incoming ) , @xmath304 ( outgoing ) , applying the shannon uncertainty rule of gives @xmath312 the _ information rate distortion function _ @xmath11 for a source sequence @xmath284 , retranslated sequence @xmath304 with distortion measure @xmath313 , is defined as follows . let @xmath314 . then @xmath315 to explain this notation , the minimization is over all conditional distributions @xmath316 , for which the joint distribution @xmath317 satisfies average distortion less than or equal to @xmath8 . the rate distortion theorem ( see e.g. berger , 1971 ; cover and thomas , 1991 ) states that _ @xmath11 is the minimum necessary rate of information transmission ( effectively the channel capacity ) so that the average distortion does not exceed the distortion @xmath8_. ash , r. , 1990 , _ information theory _ , dover publications , new york . baianu , i. c. , brown , r. , georgescu , g. , and glazebrook , j. f. , 2006 , complex nonlinear biodynamics in categories , higher dimensional algebra and lukasiewicz moisil topos : transformations of neuronal , genetic and neoplastic networks , _ axiomathes _ , * 16 * nos . 12 , 65122 . glade , n. , ben amor , h. m. , and bastien , o. , 2009 , trail systems as fault tolerant wires and their use in bio processors . in amar , p. et al . , ( eds . ) _ modeling complex biological sytems in the context of genomics , proceedings of the spring school , nice 2009 _ , 85119 . moffitt , t. e. , caspi , a. , and rutter , m. , 2006 , measured gene - environment interactions in psychopathology : concepts , research strategies , and implications for research , intervention , and public understanding of genetics , _ perspectives on psychological science _ , * 1*(1 ) , 527 . wallace , r. , and wallace , d. , 2008 , punctuated equilibrium in statistical models of generalized coevolutionary resilience : how sudden ecosystem transitions can entrain both phennotype expression and darwinian selection , _ transactions on computational systems biology ix _ , lnbi 5121 : 23 85 . west , r. l. , stewart t. c. , lebiere c. , and chandrasekharan , s. ( 2005 ) , stochastic resonance in human cognition : act - r versus game theory , associative neural networks , recursive neural networks , q - learning and humans . in bara , b. , barsalou , l. , & bucciarelli , m. ( eds . ) , _ proceedings of the 27th annual conference of the cognitive science society - stresa , italy , 2005 _ 2353 - 2358 ) . mahwah , nj : lawrence erlbaum associates .

networks of balls and springs or frames of nodes connected by compressible struts provide realistic models for physical systems from bridges to condensed solids . their elastic properties depend on their coordination number @xmath0 the average number of nodes each node is connected to . if @xmath0 is large enough , the networks are elastic solids whose long - wavelength mechanical properties are described by a continuum elastic energy with non - vanishing elastic moduli . if @xmath0 is small enough , the networks have deformation modes of zero energy they are floppy . as @xmath0 is increased from the floppy side , a critical value , @xmath4 , is reached at which springs provide just enough constraints that the system has no zero - energy floppy " modes @xcite ( or mechanisms @xcite in the engineering literature ) , and the system is _ isostatic_. the phenomenon of rigidity percolation @xcite whereby a sample spanning rigid cluster develops upon the addition of springs is one version of this floppy - to - rigid transition . the coordination numbers of whole classes of systems , including engineering structures @xcite ( bridges and buildings ) , randomly packed spheres near jamming @xcite , network glasses @xcite , cristobalites @xcite , zeolites @xcite , and biopolymer networks @xcite are close enough to @xmath4 that their elasticity and mode structure is strongly influenced by those of the isostatic lattice . ( a ) section of a kagome lattice with @xmath5 and @xmath6 three - site unit cells . nearest - neighbor bonds , occupied by harmonic springs , are of length @xmath7 . the rotated row ( second row from the top ) represents a floppy mode . next - nearest neighbor bonds are shown as dotted lines in the lower left hexagon . the vectors @xmath8 , @xmath9 , and @xmath10 indicate symmetry directions of the lattice . the numbers in the triangles indicate those that twist together under pbcs in zero modes along the three symmetry direction . note that there are only @xmath11 of these modes . ( b ) section of a square lattice depicting a floppy mode in which all sites along a line are displaced uniformly . ( c ) twisted kagome lattice , with lattice constant @xmath12 , derived from the undistorted lattice by rigidly rotating triangles through an angle @xmath13 . a unit cell , bounded by dashed lines , is shown in violet . arrows depict site displacements for the zone - center , i.e. , zero wavenumber , @xmath14 mode which has zero ( nonzero ) frequency under free ( periodic ) boundary conditions . sites @xmath15 , @xmath16 , and @xmath17 undergo no collective rotation about their center of mass whereas sites @xmath15 , @xmath18 , and @xmath19 do . ( d ) superposed snapshots of the twisted lattice showing decreasing areas with increasing @xmath13 . ] though the isostatic point always separates rigid from floppy behavior , the properties of isostatic lattices are not universal ; rather they depend on lattice architecture . here we explore the the unusual properties of a particular class of periodic isostatic lattices derived from the two - dimensional kagome lattice by rigidly rotating triangles through an angle @xmath13 without changing bond lengths as shown in fig . [ fig : periodic - isostatic ] . the bulk modulus @xmath20 of these lattices is rigorously zero for all @xmath21 . as a result , their poisson ratio acquires its limit value of @xmath22 ; when stretched in one direction , they _ expand _ by an equal amount in the orthogonal direction : they are maximally auxetic @xcite . these modes represent collapse pathways @xcite of the kagome lattice . modes of isostatic systems are generally very sensitive to boundary conditions @xcite , but the degree of sensitivity depends on the details of lattice structure . for reasons we will discuss more fully below , modes of the square lattice , which is isostatic , are in fact insensitive to changes from free boundary conditions ( fbcs ) to periodic boundary conditions ( pbcs ) , whereas those of the undistorted kagome lattice are only mildly so . the modes of both , however , change significantly when rigid boundary conditions ( rgbs ) are applied . we show here that in all families of the twisted kagome lattice , modes depend sensitively on whether fbcs , pcbs or rgbs are applied : finite lattices with free boundaries have floppy surface modes that are not present in their periodic or rigid spectrum or in that of finite undistorted kagome lattices . in the long wavelength limit , the surface floppy modes , which are present in any @xmath23 material with @xmath24 , reduce to surface rayleigh waves @xcite described by a conformally invariant energy whose analytic eigenfunctions are fully determined by boundary conditions . at shorter wavelengths , the surface waves become sensitive to lattice structure and remain confined to within a distance of the surface that diverges as the undistorted kagome lattice is approached . in the simplest twisted kagome lattice , all floppy modes are surface modes , but in more complicated lattices , including ones with uniaxial symmetry , we construct , there are both surface and bulk floppy modes . arguments due to j.c . maxwell @xcite provide a criterion for network stability : networks in @xmath1 dimensions consisting of @xmath25 nodes , each connected with central - force springs to an average of @xmath0 neighbors , have @xmath26 zero - energy modes when @xmath3 ( in the absence of redundant bonds - see below ) . of these a number , @xmath27 , which depends on boundary conditions , are trivial rigid translations and rotations , and the and the remainder are floppy modes of internal structural rearrangement . under fbcs an pbcs , @xmath28 equals @xmath29 and @xmath1 , respectively . with increasing @xmath0 , mechanical stability is reached at the isostatic point at which @xmath30 . the maxwell argument is a global one ; it does not provide information about the nature of the floppy modes and does not distinguish between bulk or surface modes . the kagome lattice of central force springs shown in fig . [ fig : periodic - isostatic](a ) is one of many locally isostatic lattices , including the familiar square lattice lattice in two dimensions [ fig . [ fig : periodic - isostatic](b ) ] and the cubic and pyrochlore lattices in three dimensions , with exactly @xmath2 nearest - neighbor ( @xmath31 ) bonds connected to each site not at a boundary . under pbcs , there are no boundaries , and every site has exactly @xmath23 neighbors . finite , @xmath25-site sections of these lattices have surface sites with fewer than @xmath32 neighbors and of order @xmath33 zero modes . the free kagome lattice with @xmath34 and @xmath35 unit cells along its sides [ fig . [ fig : periodic - isostatic](a ) ] has @xmath36 sites , @xmath37 bonds , and @xmath38 zero modes , all but three of which are floppy modes . these modes , depicted in fig . [ fig : periodic - isostatic](a ) , consist of coordinated counter rotations of pairs of triangles along the symmetry axes @xmath8 , @xmath9 and @xmath10 of the lattice . there are @xmath34 modes associated with lines parallel to @xmath8 , @xmath35 associated with lines parallel to @xmath10 , and @xmath39 modes associated with lines parallel to @xmath9 . in spite of the large number of floppy modes in the kagome lattice , its longitudinal and shear lam coefficients , @xmath40 and @xmath41 , and its bulk modulus @xmath42 are nonzero and proportional to the nearest neighbor ( @xmath31 ) spring constant @xmath43 : @xmath44 , and @xmath45 . the zero modes of this lattice can be used to generate an infinite number of distorted lattices with unstretched springs and thus zero energy @xcite . we consider only periodic lattices , the simplest of which are the twisted kagome lattices obtained by rotating triangles of the kagome unit cell through an angle @xmath13 as shown in figs . [ fig : periodic - isostatic](c ) and ( d ) @xcite . these lattices have @xmath46 rather than @xmath47 symmetry and , like the undistorted kagome lattice , three sites per unit cell . as fig . [ fig : periodic - isostatic](d ) shows , the lattice constant of these lattices is @xmath12 , and their area @xmath48 decreases as @xmath49 as @xmath13 increases . the maximum value that @xmath13 can achieve without bond crossings is @xmath50 so that the maximum relative area change is @xmath51 . since all springs maintain their rest length , there is no energy cost for changing @xmath13 , and as a result , @xmath20 is zero for every @xmath52 , whereas the shear modulus @xmath53 remains nonzero and unchanged . thus , the poisson ratio @xmath54 attains its smallest possible value of @xmath22 . for any @xmath21 , the addition of next - nearest - neighbor ( @xmath55 ) springs , with spring constant @xmath56 ( or of bending forces between springs ) stabilizes zero - frequency modes and increases @xmath20 and @xmath57 . nevertheless , for sufficiently small @xmath56 , @xmath57 remains negative . figure [ fig : auxetic - phased ] shows the region in the @xmath58 plane with negative @xmath57 . phase diagram in the @xmath59 plane showing region with negative poisson ratio @xmath57 . ] we now turn to the linearized phonon spectrum of the kagome and twisted kagome lattices subjected to pbcs . these conditions require displacements at opposite ends of the sample to be identical and thus prohibit distortions of the shape and size of the unit cell and rotations but not uniform translations , leaving two rather than three trivial zero modes . the spectrum @xcite of the three lowest frequency modes along symmetry directions of the undistorted kagome lattice with and without @xmath55 springs is shown in fig . [ fig : kagome - spectrum](a ) . when @xmath60 , there is a floppy mode for each wavenumber @xmath61 running along the entire length of the three symmetry - equivalent straight lines running from @xmath62 to @xmath63 to @xmath62 in the brillouin zone [ see inset to fig . [ fig : kagome - spectrum ] ] . when @xmath64 , there are exactly @xmath65 wavenumbers with @xmath61 along each of these lines for a total of @xmath66 floppy modes . in addition , there are three zero modes at @xmath67 corresponding to two rigid translations , and one floppy mode that changes unit cell area at second but not first order in displacements , yielding a total of @xmath68 zero modes rather than the @xmath69 modes expected from the maxwell count under fbcs . this is our first indication of the importance of boundary conditions . the addition of @xmath55 springs endows the floppy modes at @xmath60 with a characteristic frequency @xmath70 and causes them to hybridize with the acoustic phonon modes [ fig . [ fig : kagome - spectrum](a ) ] @xcite . the result is an isotropic phonon spectrum up to wavenumber @xmath71 and gaps at @xmath63 and @xmath62 of order @xmath72 . remarkably , at nonzero @xmath13 and @xmath60 , the mode structure is almost identical to that at @xmath73 and @xmath74 with characteristic frequency @xmath75 and length @xmath76 . in other words , twisting the kagome lattice through an angle @xmath13 has essentially the same effect on the spectrum as adding @xmath55 springs with spring constant @xmath77 . thus under pbcs , the twisted kagome lattice has no zero modes other than the trivial ones : it is collectively " jammed in the language of references @xcite , but because it is not rigid with respect to changing the unit cell size , it is not strictly jammed . ( a)phonon spectrum for the undistorted kagome lattice . dashed lines depict frequencies at @xmath60 and full lines at @xmath74 . the inset shows the brillouin zone with symmetry points @xmath63 , @xmath62 , and @xmath78 . note the line of zero modes along @xmath79 when @xmath60 , all of which develop nonzero frequencies for wavenumber @xmath80 when @xmath74 reaching @xmath81 on a plateau beginning at @xmath82 defining a length scale @xmath83 . ( b ) phonon spectrum for @xmath84 and @xmath60 . the plateau along @xmath79 defines @xmath85 and its onset at @xmath86 defines a length @xmath87 . ] to understand the origin of the differences in the zero - mode count for different boundary conditions , we turn to an elegant formulation @xcite of the maxwell rule that takes into account the existence of redundant bonds ( i.e. , bonds whose removal does not increase the number of floppy modes @xcite ) and states in which springs can be under states of self - stress . consider a ring network in two dimensions shown in fig . [ fig : self - stress ] with @xmath88 nodes and @xmath89 springs with three springs of length @xmath7 and one spring of length @xmath90 . the maxwell count yields @xmath91 zero modes : two rigid translations , one rigid rotation , and one internal floppy mode all of which are finite - amplitude " modes with zero energy even for finite - amplitude displacements . when @xmath92 , the maxwell rule breaks down . in the zero - energy configuration , the long spring and the three short ones are colinear , and a prestressed state in which the @xmath90-spring is under compression and the three @xmath7-springs are under tension ( or vice versa ) but the total force on each node remains zero becomes possible . this is called a state of self - stress . the system still has three finite amplitude zero modes corresponding to arbitrary rigid translations and rotations , but the finite - amplitude floppy mode has disappeared . in the absence of prestress , it is replaced by two infinitesimal " floppy modes of displacements of the two internal nodes perpendicular of the now linear network . in the presence of prestress , these two modes have a frequency proportional to the square root of the tension in the springs . thus , the system now has one state of self stress and one extra zero mode in the absence of prestress , implying @xmath93 , where @xmath94 is the number of states of self stress . this simple count is more generally valid as can be shown with the aid of the equilibrium and compatibility matrices @xcite , denoted , respectively , as @xmath95 and @xmath96 . @xmath95 relates the vector @xmath97 of @xmath98 spring tensions to the vector @xmath99 of @xmath100 forces at nodes via @xmath101 , and @xmath102 relates the the vector @xmath103 of @xmath100 node displacements to the vector @xmath104 of @xmath98 spring stretches via @xmath105 . the dynamical matrix determining the phonon spectrum is @xmath106 . vectors @xmath107 in the null space of @xmath95 , ( @xmath108 ) , describe states of self - stress whereas vectors @xmath109 in the null space of @xmath102 represent displacements with no stretch @xmath104 , i.e. , modes of zero energy . thus the nullspace dimensions of @xmath95 and @xmath102 are , respectively , @xmath94 and @xmath110 . the rank - nullity theorem of linear algebra @xcite states that the rank @xmath111 of a matrix plus the dimension of its null space equals its column number . since the rank of a matrix and its transpose are equal , the @xmath95 and @xmath102 matrices , respectively , yield the relations @xmath112 and @xmath113 , implying @xmath114 . under pbcs , locally isostatic lattices have @xmath2 exactly , and the maxwell rule yields @xmath115 : there should be no zero modes at all . but we have just seen that both the square and undistorted kagome lattices under pbcs have of order @xmath33 zero modes as calculated from the dynamical matrix , which , because it is derived from a harmonic theory , does not distinguish between infinitesimal and finite - amplitude zero modes . thus , in order for there to be zero modes , there must be states of self - stress , in fact one state of self - stress for each zero mode . in the square lattice under fbcs , @xmath116 and @xmath117 , there are no states of self stress , and @xmath118 zero modes depicted in fig . [ fig : periodic - isostatic](b ) . under pbcs , the dimension of the nullspace of @xmath95 is @xmath119 , and there are also @xmath120 zero modes that are identical to those under fbcs . we have already seen that there are @xmath121 zero modes in the free undistorted kagome lattice . direct evaluations @xcite ( see text s1 ) of the dimension of the null spaces of @xmath95 and @xmath102 for the undistorted kagome lattice with pbcs yields @xmath122 when @xmath64 . the zero modes under pbcs are identical to those under fbcs except that the @xmath123 modes associated with lines parallel to @xmath9 under fbcs get reduced to @xmath34 modes because of the identification of apposite sides of the lattice required by the pbcs as shown in fig . [ fig : periodic - isostatic](a ) . thus the modes of both the square and kagome lattices do not depend strongly on whether fcbs or pbcs are applied . under rbc s , however , the floppy modes of both disappear . the situation for the twisted kagome lattice is different . there are still @xmath124 zero modes under fbcs , but there are only two states of self stress under pbcs and thus only @xmath125 zero modes , as a direct evaluation of the null spaces of @xmath95 and @xmath102 verifies ( see appendix for details ) , in agreement with the results obtained via direct evaluation of the eigenvalues of the dynamical matrix @xcite . all of the floppy modes under fbcs have disappeared . ( a ) ring - network with @xmath126 showing internal floppy mode . ( b ) ring - network with @xmath92 showing one of the two infinitesimal modes . ] an effective long - wavelength energy @xmath127 for the low - energy acoustic phonons and nearly floppy distortions provides insight into the nature of the modes of the twisted kagome lattice . the variables in this theory are the vector displacement field @xmath128 of nodes at undistorted positions @xmath129 and the scalar field @xmath130 describing nearly floppy distortions within a unit cell . the detailed form of @xmath127 depends on which three lattice sites are assigned to a unit cell . figure [ fig : periodic - isostatic](c ) depicts the lattice distortion @xmath14 for the nearly floppy mode at @xmath63 ( with energy proportional to @xmath131 ) along with a particular representation of a unit cell , consisting of a central asymmetric hexagon and two equilateral triangles , with @xmath132 sites on its boundary . if sites @xmath15 , @xmath16 , and @xmath17 are assigned to the unit cell , then the distortion @xmath14 involves no rotations of these sites relative to their center of mass , and the harmonic limit of @xmath127 depends only on the symmetrized and linearized strain @xmath133 and on @xmath14 : @xmath134 , \label{eq : full - elastic}\end{aligned}\ ] ] where @xmath135 is the symmetric - traceless stain tensor , @xmath136 , @xmath137 , @xmath138 , @xmath139 , and @xmath140 . the last term in which @xmath141 is a third - rank tensor , whose only non - vanishing components are @xmath142 , is invariant under operations of the group @xmath47 but not under arbitrary rotations . the @xmath143 term is the only one that reflects the @xmath46 ( rather than @xmath47 ) symmetry of the lattice . there are several comments to make about this energy . the gauge - like coupling in which the isotropic strain @xmath144 appears only in the combination @xmath145 guarantees that the bulk modulus vanishes : @xmath14 will simply relax to @xmath146 to reduce to zero the energy of any state with nonvanishing @xmath144 . the coefficient @xmath78 can be calculated directly from the observation that under @xmath14 alone , the length of every spring changes by @xmath147 , and this length change is reversed by a homogenous volume change @xmath148 . in the @xmath149 limit , @xmath150 , and the energy reduces to that of an isotropic solid with bulk modulus @xmath151 if the @xmath152 and @xmath153 terms , which are higher order in gradients , are ignored . the @xmath153 term gives rise to a term , singular in gradients of @xmath154 , when @xmath14 is integrated out that is responsible for the deviations of the finite - wavenumber elastic energy from isotropy . at small @xmath13 , the length scale @xmath155 appears in several places in this energy : in the length @xmath156 and in the ratios @xmath157 , @xmath158 , and @xmath159 . at length scales much larger than @xmath155 , the @xmath152 and @xmath153 terms can be ignored , and @xmath14 relaxes to @xmath146 leaving only the shear elastic energy of an elastic solid proportional @xmath160 . at length scales shorter than @xmath155 , @xmath14 deviates from @xmath146 and contributes significantly to the form of the energy spectrum . if @xmath15 , @xmath18 and @xmath19 in fig . [ fig : periodic - isostatic](d ) are assigned to the unit cell , then @xmath14 involves rotations relative to the lattice axes , and the energy develops a cosserat - like form @xcite that is a function of @xmath161 rather than @xmath14 . the modes of our elastic energy in the long - wavelength limit ( @xmath162 ) are simply those of an elastic medium with @xmath24 . in this limit , there are transverse and longitudinal bulk sound modes with equal sound velocities @xmath163 and @xmath164 , where @xmath165 is the particle mass at each node and @xmath166 is the mass density . in addition there are rayleigh surface waves @xcite in which there is a single decay length ( rather than the two at @xmath167 ) , and displacements are proportional to @xmath168 $ ] with @xmath169 for a semi - infinite sample in the right half plane so that the penetration depth into the interior is @xmath170 . these waves have zero frequency in two dimensions when @xmath24 , and they do not appear in the spectrum with pbcs . thus this simple continuum limit provides us with an explanation for the difference between the spectrum of the free and periodic twisted kagome lattices . under fbcs , there are zero - frequency surface modes not present under pbcs . further insight into how boundary conditions affect spectrum follows from the observation that the continuum elastic theory with @xmath24 depends only on @xmath171 . the metric tensor @xmath172 of the distorted lattice is related to the strain @xmath173 via the simple relation @xmath174 ; and @xmath175/2 $ ] , which is zero for @xmath176 , is invariant , and thus remains equal to zero , under conformal transformations that take the metric tensor from its reference form @xmath177 to @xmath178 for any continuous function @xmath179 . the zero modes of the theory thus correspond simply to conformal transformations , which in two dimensions are best represented by the complex position and displacement variables @xmath180 and @xmath181 . all conformal transformations are described by an analytic displacement field @xmath182 . since by cauchy s theorem , analytic functions in the interior of a domain are determined entirely by their values on the domain s boundary ( the holographic " property @xcite ) , the zero modes of a given sample are simply those analytic functions that satisfy its boundary conditions . for example , a disc with fixed edges ( @xmath183 ) has no zero modes because the only analytic function satisfying this fbc is the trivial one @xmath184 ; but a disc with free edges ( stress and thus strain equal to zero ) has one zero mode for each of the analytic functions @xmath185 for integer @xmath186 . the boundary conditions @xmath187 and @xmath188 on a semi - infinite cylinder with axis along @xmath189 are satisfied by the function @xmath190 when @xmath191 , where @xmath192 is an integer . this solution is identical to that for classical rayleigh waves on the same cylinder . like the rayleigh theory , the conformal theory puts no restriction on the value of @xmath192 ( or equivalently @xmath193 ) . both theories break down , however , at @xmath194 beyond which the full lattice theory , which yields a complex value of @xmath195 , is needed . figure [ fig : ky - kx](a ) shows an example of a surface wave . at the bottom of this figure , @xmath196 is an almost perfect sinusoid . as @xmath197 decreases toward the surface , the amplitude grows , and in this picture reaches the nonlinear regime by the time the surface at @xmath198 is reached . figure [ fig : ky - kx](b ) plots @xmath199 as a function of @xmath193 obtained both by direct numerical evaluation and by an analytic transfer matrix procedure @xcite for different values of @xmath13 ( text s1 ) . the rayleigh limit @xmath200 is reached for all @xmath13 as @xmath201 . interestingly the rayleigh limit remains a good approximation up to values of @xmath193 that increase with increasing @xmath13 . the inset to fig . [ fig : ky - kx ] , plots @xmath202 as a function of @xmath203 and shows that in the limit @xmath204 ( @xmath205 ) , @xmath199 obeys an @xmath13-independent scaling law of the form @xmath206 . the full complex @xmath207 obeys a similar equation . this type of behavior is familiar in critical phenomena where scaling occurs when correlation lengths become much larger than microscopic lengths . the function @xmath208 approaches @xmath209 as @xmath210 and asymptotes to @xmath211 for @xmath212 . thus for @xmath213 , @xmath214 and for @xmath215 , @xmath216 . as @xmath13 increases , @xmath217 is no longer much larger than one , and deviations from the scaling law result . the situation for surfaces along different directions ( e.g. , along @xmath218 rather than @xmath198 ) is more complicated and will be treated in a future publication @xcite . ( a)lattice distortions for a surface wave on a cylinder , showing exponential decay of the surface displacements into the bulk . this figure was constructed by specifying a small sinusoidal modulation on the bottom boundary and propagating lattice - site positions upward to the free boundary at the top under the constraint of of constant lengths and periodic boundary conditions around the cylinder . distortions near and at the top boundary , which have become nonlinear , are not described by our linearized treatment . ( b ) @xmath219 as a function of @xmath220 for lattice rayleigh surface waves for @xmath221 , in order from bottom to top . smooth curves are the analytic results from a transfer matrix calculation , and dots are from direct numerical calculations . the dashed line is the continuum rayleigh limit @xmath200 . curves at smaller @xmath13 break away from this curve at smaller values of @xmath207 than do those at large @xmath13 . at @xmath222 , @xmath223 diverges at @xmath224 . the inset plots @xmath202 as a function of @xmath225 for different @xmath13 . the lower curve in the inset(black ) is the @xmath13-independent scaling function of @xmath226 reached in the @xmath227 limit . the other curves from top to bottom are for @xmath228 , @xmath229 , @xmath230 , and @xmath231 ( chosen to best present results rather than to match the curves in the main figure ) . curves for @xmath232 are essentially indistinguishable from the scaling limit . the curve at @xmath233 stops because @xmath234 . ] the @xmath46 twisted kagome lattice is the simplest of many lattices that can be formed from the kagome and other periodic isostatic lattices . figures [ fig : otherauxetic ] ( a ) and ( b ) show two other examples of isostatic lattices constructed from the kagome lattice . most intriguing is the lattice with pgg symmetry . its geometry has uniaxial symmetry , yet its long - wavelength elastic energy is identical to that of the @xmath46 twisted kagome lattice , i.e , it is isotropic with a vanishing bulk modulus , and its mode structure near @xmath67 is isotropic as shown in fig . [ fig : otherauxetic ] ( c ) . thus , this system loses long - wavelength zero - frequency bulk modes of the undistorted kagome lattice to surface modes . however , at large wavenumber , lattice anisotropy becomes apparent , and ( infinitesimal ) floppy bulk modes appear . thus in this and related systems , a fraction of the zero modes of the under fbcs are bulk modes that are visible under pbcs , and a fraction are surface modes that are not . randomly packed spheres above the jamming transition with average coordination number @xmath235 exhibit a characteristic frequency @xmath236 and length @xmath237 and a transition from a debye - like ( @xmath238 ) to a flat density of states at @xmath239 @xcite . the square and kagome lattices with randomly added @xmath55 springs have the same properties @xcite . a general cutting " argument @xcite provides a procedure for perturbing away from the isostatic limit and an explanation for these properties . however , it only applies provided a finite fraction of the of order @xmath240 floppy modes of a sample with sides of length @xmath241 cut from an isostatic lattice with pbcs are extended , i.e. , have wave functions that extend across the sample rather than remaining localized either in the interior or at the surface the sample . clearly the twisted kagome lattice , whose floppy modes are all surface modes , violates this criterion ; and indeed , the density of states of the lattice with @xmath242 shows debye - behavior crossing over to a flat plateau at @xmath243 . adding next nearest neighbor bonds gives rise to a length @xmath244 and crossover to the plateau at @xmath245 . the pgg lattice in fig . [ fig : otherauxetic](a ) , however , has both extended and surface floppy modes , so its crossover to the a flat plateau occurs at @xmath246 rather than at @xmath247 or @xmath248 . ( a ) kagome - based lattice with pgg space group symmetry and uniaxial @xmath249 point group symmetry . ( b ) lattice with p6 space symmetry but global @xmath250 point - group symmetry . ( c ) density plot of the spectrum of the lowest frequency branch of the pgg uniaxial kagome lattice . the spectrum is absolutely isotropic near the origin point @xmath63 , but it has a zero modes on two symmetry related continuous curves at large values of wave - number . ] our study highlights the rich and remarkable variety of physical properties that isostatic systems can exhibit . under fbcs , floppy modes can adopt a variety of forms , from all being extended to all being localized near surfaces to a mixture of the two . under pbc s , the presence of floppy modes depends on whether the lattice can or can not support states of self stress . when a lattice exhibits a large number of zero - energy edge modes , its mechanical / dynamical properties become extremely sensitive to boundary conditions , much as do the electronic properties of the topological states of matter studied in quantum systems @xcite . the zero - energy edge modes observed in our isostatic lattices are collective modes whose amplitudes decay exponentially from the edge with a finite decay length , in direct contrast to the very localized and trivial floppy modes arising from dangling bonds . we focussed primarily on high - symmetry lattices derived from the kagome lattice , but the properties they exhibit , namely a deficit of floppy modes in the bulk and the existence of floppy surface modes , are shared by any two - dimensional system with a vanishing bulk modulus ( or the equivalent in anisotropic systems ) . three - dimensional analogs of the twisted kagome lattice @xcite can be constructed by rotating tetrahedra in pyrochlore and zeolite lattices @xcite and in cristobalites @xcite . these lattices are anisotropic . with @xmath31 forces only , they exhibit a vanishing modulus for compression applied in particular planes rather than isotropically , but we expect them to exhibit many of the properties the two - dimensional lattices exhibit . finally , we note that maxwell s ideas can be applied to spin systems such as the heisenberg anti - ferromagnet on the kagome lattice @xcite , and the possibility of unusual edge states in them is intriguing . we are grateful for informative conversations with randall kamien , andea liu , and s.d . guest . this work was supported in part by nsf under grants dmr- 0804900 ( t.c.l . and x.m . ) , mrsec dmr-0520020 ( t.c.l . and a.s . ) and jqi - nsf - pfc ( k. s. ) . in this section , we will provide some of the details for calculating the equilibrium and compatibility matrices @xmath95 and @xmath102 and the dynamical matrix @xmath251 under periodic boundary conditions . the equilibrium matrix @xmath95 relates the @xmath252 dimensional vector of bond tensions @xmath97 to the @xmath100 dimensional vector @xmath99 of forces at the nodes via @xmath101 . we label unit cells by the their position vectors @xmath253 , where @xmath254 and @xmath255 are integers and @xmath256 are the primitive lattice vectors . ( here we take the lattice spacing to be one , i.e. , @xmath257 . ) each unit cell has three sites , labeled @xmath15 , @xmath16 , and @xmath17 as shown in fig . [ fig : lattice1 ] , so we label the @xmath258 sites by @xmath259 , where @xmath260 and @xmath261 is the number of unit cells . forces at the nodes are denoted by @xmath262 and their @xmath263 components as @xmath264 , where @xmath265 . they have to be balanced by stretching forces in the springs located on the bonds . there are six bonds per cell , which we can take to be the six bonds in the distorted hexagon shown in fig . [ fig : lattice1 ] , oriented parallel to the unit vectors @xmath266 for @xmath267 , where @xmath13 is the twist angle of the twisted kagome lattice . when @xmath73 , these vector reduce to the edge vectors of a symmetric hexagon with @xmath268 . the @xmath263 bonds , labeled with @xmath269 , are occupied by springs under tension @xmath270 that exert pulling ( for @xmath271 ) forces , direct along the bond vectors @xmath272 , on nodes . we define vectors @xmath273 . because we have periodic boundary conditions , we can express both @xmath274 and @xmath275 in terms of their fourier components , @xmath276 the equilibrium matrix is diagonal in @xmath277 , so it breaks up into independent @xmath278 blocks for each of the @xmath6 independent vectors @xmath277 . for simplicity , we consider only the case @xmath64 for which @xmath279 and for which the number of sites is @xmath280 . there are four bonds incident on each site . the equations relating @xmath262 to @xmath275 follow from fig . [ fig : lattice1 ] : @xmath281 we map @xmath282 to a single index @xmath165 with @xmath283 and @xmath284 and then fourier transform . the result is that @xmath95 breaks up into independent @xmath278 blocks @xmath285 for each @xmath277 . this leads to @xmath286 where @xmath287 the null space of this matrix is easily calculated with the aid of mathematica . * case 1 , @xmath73 * : this is the untwisted kagome lattice . when @xmath61 , the null space of @xmath288 is empty unless @xmath277 is perpendicular to one of the primitive lattice vectors : @xmath289 and @xmath290 , in which case there is a single vector , @xmath291 where @xmath292 is the number of cells , for each @xmath192 and @xmath293 in the null space . when @xmath67 , there are three vectors , @xmath294 , @xmath295 , and @xmath296 in the null space . these are the @xmath297 limits of @xmath298 , @xmath299 , and @xmath300 , respectively . we can , therefore , take the set of null - space vectors of @xmath95 to be @xmath298 , @xmath299 , and @xmath299 for the @xmath34 values of @xmath301 , which include @xmath297 . linear combinations of these vectors are also null - space vectors , and we can use them to construct vectors confined to a single line of bonds . for example the vector @xmath302 corresponds to a state in which springs on all @xmath15-bond in cells with centers at @xmath303 for any @xmath254 and on all @xmath11-bonds in cells with centers at @xmath304 are all under the same tension . this is a state of self - stress in which all bonds along a particular straight line parallel to the @xmath8-axis are under tension . similar states of self - stress for any of the @xmath305 lines of bonds can be constructed . * case 2 , @xmath306 * : this is the twisted lattice . the null space of @xmath288 is empty for _ all _ @xmath307 , and it contains only two vectors when @xmath308 . these correspond to two states of self - stress in which the stress the stress on the six bonds take on both positive and negative values that are identical in all unit cells . the compatibility matrix @xmath102 relates the @xmath100 displacements @xmath103 to the @xmath309 stretches @xmath310 . the displacements are labeled the same way as the forces , i.e. as @xmath311 and the stretches in the same way as the bond tensions , i.e. as @xmath312 . the equations relating @xmath311 and @xmath313 are @xmath314 fourier transforming and using the same @xmath315 notation as for the equilibrium matrix , we obtain @xmath316 where @xmath317 as required by the constraint @xmath318 . finally , the dynamical matrix is @xmath319 , whose components are @xmath320 \nonumber\\ d_{12}({\mathbf{q } } ) & = & = -\frac{1}{2 } \sqrt{3 } \cos ( 2 \alpha ) \nonumber\\ d_{13}({\mathbf{q } } ) & = & = d_{31}^*({\mathbf{q } } ) = - e^{-i \left(\frac{q_x}{2}+\frac{\sqrt{3 } q_y}{2}\right ) } \sin ^2\left(\frac{\pi } { 6}-\alpha \right ) -e^{-i q_x}\sin ^2\left(\alpha + \frac{\pi}{6}\right ) \nonumber\\ d_{14}({\mathbf{q } } ) & = & d_{41}^*({\mathbf{q } } ) = \frac{1}{2 } e^{-i q_x}\cos \left(\frac{\pi } { 6}-2 \alpha \right ) + \frac{1}{2}e^{-i \left(\frac{q_x}{2}+\frac{\sqrt{3 } q_y}{2}\right ) } \cos \left(2 \alpha + \frac{\pi } { 6}\right ) \nonumber\\ d_{15}({\mathbf{q } } ) & = & d_{51}^*({\mathbf{q } } ) = -2e^{-\frac{1}{2 } i \sqrt{3 } q_y } \cos ^2(\alpha ) \cos \left(\frac{q_x}{2}\right ) \nonumber\\ d_{16}({\mathbf{q } } ) & = & d_{61}^*({\mathbf{q } } ) = i e^{-\frac{1}{2 } i \sqrt{3 } q_y } \sin ( 2 \alpha ) \sin \left(\frac{q_x}{2}\right ) \nonumber\\ d_{22}({\mathbf{q } } ) & = & 2 \sin ^2(\alpha ) + \cos ^2\left(\frac{\pi } { 6}-\alpha \right ) + \cos ^2\left(\alpha + \frac{\pi } { 6}\right ) \nonumber\\ d_{23}({\mathbf{q } } ) & = & d_{32}^*({\mathbf{q } } ) = \frac{1}{2}e^{-i q_x } \cos \left(\frac{\pi } { 6}-2 \alpha \right ) + \frac{1}{2 } e^{-i\left(\frac{q_x}{2}+\frac{\sqrt{3 } q_y}{2}\right ) } \cos \left(2 \alpha + \frac{\pi } { 6}\right ) \nonumber\\ d_{24}({\mathbf{q}})&= & d_{42}^*({\mathbf{q } } ) = -e^{-i \left(\frac{q_x}{2}+\frac{\sqrt{3 } q_y}{2}\right ) } \cos ^2\left(\frac{\pi } { 6}-\alpha \right ) -e^{-i q_x}\cos ^2\left(\alpha + \frac{\pi}{6}\right ) \nonumber\\ d_{2,5}({\mathbf{q } } ) & = & d_{5,2}^*({\mathbf{q } } ) = ie^{-\frac{1}{2 } i \sqrt{3 } q_y } \sin ( 2 \alpha ) \sin \left(\frac{q_x}{2}\right ) \nonumber\\ d_{2,6}({\mathbf{q } } ) & = & d_{6,2}^*({\mathbf{q } } ) = -2e^{-\frac{1}{2 } i \sqrt{3 } q_y } \sin ^2(\alpha ) \cos \left(\frac{q_x}{2}\right ) \nonumber\\ d_{33}({\mathbf{q } } ) & = & 2 \sin ^2\left(\frac{\pi } { 6}-\alpha \right)+ 2 \sin ^2\left(\alpha + \frac{\pi } { 6}\right ) \nonumber\\ d_{34}({\mathbf{q } } ) & = & d_{43}^ * ( { \mathbf{q } } ) = 0 \nonumber\\ d_{3,5}({\mathbf{q } } ) & = & d_{5,3}({\mathbf{q } } ) = -e^{-i \left(\frac{\sqrt{3 } q_y}{2}-\frac{q_x}{2}\right ) } \sin ^2\left(\frac{\pi } { 6}-\alpha \right ) -e^{i q_x}\sin ^2\left(\alpha + \frac{\pi}{6}\right ) \nonumber\\ d_{3,6}({\mathbf{q } } ) & = & d_{6,3}^*({\mathbf{q } } ) = -\frac{1}{2 } e^{i q_x}\cos \left(\frac{\pi } { 6}-2 \alpha \right ) -\frac{1}{2}e^{i\left(\frac{q_x}{2}-\frac{\sqrt{3 } q_y}{2}\right ) } \cos \left(2 \alpha + \frac{\pi } { 6}\right ) \nonumber\\ d_{4,4}({\mathbf{q } } ) & = & 2 \cos ^2\left(\frac{\pi } { 6}-\alpha \right)+ 2 \cos ^2\left(\alpha + \frac{\pi } { 6}\right ) \nonumber\\ d_{4,5}({\mathbf{q } } ) & = & d_{5,4}^*({\mathbf{q } } ) = -\frac{1}{2 } e^{iq_x}\cos \left(\frac{\pi } { 6}-2 \alpha \right ) -\frac{1}{2}e^{i\left(\frac{q_x}{2}-\frac{\sqrt{3 } q_y}{2}\right ) } \cos \left(2 \alpha + \frac{\pi } { 6}\right ) \nonumber\\ d_{4,6}({\mathbf{q } } ) & = & d_{6,4}^*({\mathbf{q } } ) = -e^{i \left(\frac{q_x}{2}-\frac{\sqrt{3 } q_y}{2}\right ) } \cos ^2\left(\frac{\pi } { 6}-\alpha \right ) -e^{i q_x}\cos ^2\left(\alpha + \frac{\pi}{6}\right ) \nonumber\\ d_{5,5}({\mathbf{q } } ) & = & \sin ^2\left(\frac{\pi } { 6}-\alpha \right)+ \sin ^2\left(\alpha + \frac{\pi } { 6}\right)+2 \cos ^2(\alpha ) \nonumber\\ d_{5,6}({\mathbf{q } } ) & = & d_{6,5}^*({\mathbf{q } } ) = \frac{1}{2}\cos \left(\frac{\pi } { 6}-2 \alpha \right ) + \frac{1}{2}\cos \left(2 \alpha + \frac{\pi } { 6}\right ) \nonumber\\ d_{6,6}({\mathbf{q } } ) & = & 2 \sin ^2(\alpha ) + \cos ^2\left(\frac{\pi } { 6}-\alpha \right ) + \cos ^2\left(\alpha + \frac{\pi } { 6}\right)\end{aligned}\ ] ] its eigenvalue spectrum can easily be calculated with the aid of mathematica . the results are shown in fig . 3 of the main text . surface modes decay exponentially into the bulk . they are characterized by a wavevector @xmath321 parallel to the plane of the surface , their frequency @xmath322 , their decay length @xmath323 perpendicular to the surface . in lattices with nearest - neighbor forces only , @xmath322 and @xmath323 can be determined by setting @xmath324 , where @xmath277 is the wavevector that appears in the dynamical matrix and @xmath325 is the component of @xmath277 perpendicular to the surface ( @xmath326 is the unit vector perpendicular to the surface ) , setting @xmath327 , requiring @xmath328 = 0,\ ] ] where @xmath329 is the unit matrix , and that the equation of motion of the surface layer be satisfied . in the case of surface modes of with zero frequency , the relation @xmath330=0 $ ] determines @xmath331 as a function of @xmath321 . the surface modes of the twisted kagome lattice have zero frequency , so we need only solve this equation . we consider only the case in which the surface is perpendicular to the @xmath197-direction and set @xmath332 and @xmath333 . we find @xmath334 = \frac{9}{64 } a(q)\left[r+r^{-1 } - 2g(q)\right ] \left[r+r^{-1 } - 2g^*(q)\right ] , \label{eq : d(q , r)}\ ] ] where @xmath335 \\ g(q ) & = & \frac{\cos(q/2)[3 - 2 \cos p + \cos p \cos ( 4 \alpha ) ] + 2 i \sqrt{3 } \sin^3 ( 3 q/2 ) \sin ( 4 \alpha)}{2 - \cos q + ( 1 - 2 \cos q ) \cos ( 4 \alpha)}\end{aligned}\ ] ] solving for @xmath111 in eq . ( [ eq : d(q , r ) ] ) and to a lattice spacing of @xmath336 , we obtain @xmath337 .\ ] ] this is the function that is plotted in fig . 5(b ) in the text . note that @xmath338 has both a real and an imaginary part , but that in the limit @xmath339 , @xmath340 .

the class of interacting binaries known as the classical algols " are semi - detached systems that contain a late - type giant or subgiant that fills its roche lobe and is transferring mass onto a much more luminous and more massive b a main sequence star . a new class of binaries has been found with properties that are similar to those of the classical algols , except that the mass gaining component is a late - type giant or subgiant like its companion . these systems were first recognized by @xcite and are now referred to as the cool algols . " they typically display signs of magnetic activity in the form of ca ii h@xmath10k emission , h@xmath11 emission , star spots , and strong x - ray emission . only about a dozen cool algols are known , and many of them were mistakenly assigned to the larger group of rs cvn binaries which contain a luminous k - type primary with a fainter f g type companion . the main difference between the cool algols and the rs cvn systems is that the former are undergoing roche lobe overflow , while the latter are detached ( see , e.g. , * ? ? ? * ; * ? ? ? bd@xmath1706 ( rxj0441.9@xmath100537 , ppm 147827 , @xmath12 , @xmath13 , j2000 , @xmath14 ) is one of the recent additions to this rare category of cool algols , and was discovered in the course of spectroscopic follow - up observations of a sample of x - ray sources selected from the rosat all - sky survey south of the taurus - auriga star - forming region @xcite . although that x - ray sample was originally designed to favor the detection of t tauri stars , which are also quite active , bd@xmath1706 was quickly found to be of an entirely different nature . @xcite ( hereafter paper i ) reported a double - lined spectroscopic orbital solution with a period of 18.9 days , a very small and possibly spurious eccentricity , and a mass ratio of @xmath15 ( @xmath16 ) , typical of the algol systems . the system contains two cool giants . the primary " ( the more massive star ) is detached , while the secondary " has filled its critical lobe or may overflow it ( paper i ) . the star we have called the secondary " is currently the larger , less massive , and cooler component , as in the classical algols . spectroscopic convention is used throughout this paper instead of the notation used in papers on stellar evolution , which refer to the initially more massive star as the primary . the general properties of the system were described in paper i. although photometry for bd@xmath1706 was not available at the time of that study , eclipses were estimated to be very likely . subsequent photometric monitoring in the optical confirmed this prediction @xcite . the general characteristics of the cool algols are fairly well defined ( see * ? ? ? * and paper i ) , but accurate information on individual systems is still far from complete . only five of the members , aside from bd@xmath1706 , have spectroscopic orbits for both components ( rz cnc , ar mon , rt lac , ad cap , and rv lib ; * ? ? ? * ; * ? ? ? * ) , and light curve analyses for the first three of these have yielded the absolute masses and radii necessary in order to understand their properties . the remaining five objects identified as possible members of this class ( uz cnc , v1061 cyg , av del , gu her , and v756 sco ; * ? ? ? * ) have very little in the way of observations and in some cases even the orbital periods are poorly known . the high - quality spectroscopic orbit presented for bd@xmath1706 in paper i , combined with the fact that the system is eclipsing , offers the opportunity to determine the absolute dimensions of the components very accurately . to that end , photometric observations in the @xmath0 system were continued after the initial report by @xcite . we present here a full analysis of those data , and also a re - analysis of the spectroscopic material in paper i with improved techniques and the benefit of the additional information provided by the light - curves . the latter offer a means for evaluating subtle distortions in the doppler measurements due to the close proximity of the stars . as a result , we are able to determine here the absolute masses of both components with uncertainties smaller than 2% , and the radii with errors smaller than 3% , which are now the best determinations for any system of the cool algol class . the global x - ray properties and the activity level of bd@xmath1706 ( h and k emission , variable h@xmath11 line strength ) were studied in some detail in paper i , where it was established that the hotter primary star is most likely the site of the bulk of the activity in the binary . additional x - ray observations for the system were secured by us with the rosat satellite essentially over a complete 18.9-day cycle ( yielding an x - ray light curve ) , and clearly show the primary eclipse but not the secondary eclipse . an analysis of these data is presented here as well , allowing us to obtain information about the geometry of the x - ray emitting region surrounding the active star . the photometric observations of bd@xmath1706 in the optical were conducted at the gettysburg college observatory ( gettysburg , pa ) with a 16-inch f/11 ealing cassegrain reflector . the detector was a photometrics thermoelectrically - cooled ccd camera with a front - illuminated thompson 7896 chip with @xmath17 pixels . bd@xmath1706 was observed on 94 nights during the 19971999 observing seasons . a total of 80 , 93 , 133 and 137 observations were made in the johnson @xmath18 , @xmath19 , @xmath20 , and @xmath21 filters , respectively . the comparison and check stars used are tyc 95 - 1495 - 1 ( gsc 00095 - 1495 ) and hd 29764 ( gsc 00091 - 00262 , bd@xmath1704 , sao 111980 ) , respectively . basic properties for these stars are listed in table [ tab : compcheck ] . coverage of the primary eclipse in the @xmath18 and @xmath19 filters was very poor during the first two seasons ( before jd 2,451,000 ) , and therefore those data were not used in the analysis below . the precision of an individual differential photometric measurement is estimated to be 0.010 mag , 0.010 mag , 0.009 mag , and 0.010 mag in the @xmath18 , @xmath19 , @xmath20 , and @xmath21 passbands , respectively , derived from the comparison and check stars under the assumption that they do not vary . the observations in the sense @xmath22variable minus comparison@xmath23 are listed in table [ tab : obsb ] , table [ tab : obsv ] , table [ tab : obsr ] , and table [ tab : obsi ] . the depth of the eclipses in the @xmath19 band is approximately 0.52 mag for the primary minimum and 0.34 mag for the secondary . changes in the overall brightness at the quadratures are fairly obvious from season to season . they are not unexpected given the activity level displayed by the system , and will be discussed in more detail in [ sec : phot ] . the spectroscopic material used here is the same as that reported in paper i. briefly , it consists of 41 single - order echelle spectra obtained from 1994 march to 1996 march with the 1.5-m wyeth reflector at the oak ridge observatory ( harvard , massachusetts ) , the 1.5-m tillinghast reflector at the fred l. whipple observatory ( mount hopkins , arizona ) , and the multiple mirror telescope ( also atop mount hopkins , arizona ) prior to its conversion to a monolithic 6.5-m mirror . the spectra cover 45 with a central wavelength of 5187 , and the resolving power is @xmath24 . the reductions to obtain radial velocities for both components are slightly different from those used in paper i , and are described later . x - ray observations of bd@xmath1706 were obtained with the high resolution imager instrument ( hri ; * ? ? ? * ) aboard the rosat satellite over a period of 18 days covering essentially a full orbital cycle of the binary , from 1997 august 18 to september 5 . the measurements were made at intervals of one day , with the exception of a 3-day gap due to scheduling constraints . the exposure times were typically about 3.5 ksec each , achieving signal - to - noise ( s / n ) ratios ranging from 6 to 15 per broad - band observation ( 0.1 to 2.4 kev ) , with a mean s / n of 10 . the observations were reduced with the extended scientific analysis software ( exsas , * ? ? ? * ) version 2001 running under eso - midas version 01feb . in addition to considering the full energy range of rosat , we subsequently divided the recorded counts into a soft band including detector channels 0 through 3 ( 0.1 to 0.5 kev ) and a hard band including channels 4 through 15 ( 0.6 to 2.4 kev ) in order to obtain additional information from the analysis of these data . we performed standard local and map source detection in all three spectral bands . the s / n ratios range from 4 to 10 in the soft band and from 4 to 11 in the hard band . this allows us to define a hardness ratio " @xmath25 , where @xmath26 and @xmath27 are the hard and soft count rates , respectively ( see * ? ? ? * ) . the individual x - ray observations are listed in table [ tab : xray ] . the columns give the heliocentric julian date and the exposure times , followed by the count rates , the corresponding errors , and the maximum likelihood ( _ ml _ ) estimator for the soft , hard , and broad bands @xcite . the quantity _ ml _ provides a measure of the existence of the source above the local background . for example , a maximum likelihood of 7.4 ( 14.3 ) corresponds to a 3.5@xmath28 ( 5@xmath28 ) detection . the last three columns give the hardness ratio and its corresponding uncertainty , as well as the phase in the orbit of the binary do not exactly add up to the broad - band counts , although the difference is minimal . the mean local background is estimated separately in each band and then subtracted from the corresponding total counts , which contributes to the slight discrepancy . ] . figure [ fig : hist ] shows the time history of the optical photometry , the spectroscopy , and the x - ray observations . because of the variability of the object , the time coverage of the optical light curves and the radial velocities is relevant for the modeling of the system , as we describe in [ sec : spec ] . in the analyses below all data were phased with the ephemeris @xmath29 where the epoch refers to the time of primary eclipse ( more massive star occulted by the other star ) , and @xmath30 is the number of cycles counted from this epoch . this ephemeris was determined from the radial velocity observations described in [ sec : spec ] , and supersedes the one given in paper i. examination of the raw light curves for bd@xmath1706 revealed significant changes over the three observing seasons that precluded their merging into a single data set for analysis . the extent of the changes is such that , for example , during the 19971998 season the light level in the @xmath19 band at the first quadrature was lower than at the second quadrature , whereas by the following season the situation had reversed . variations of this nature are common in active systems , and are well explained by the presence of surface features ( spots ) on one or both stars . in order to minimize the scatter , a compromise between the number of observations , the completeness of the phase coverage ( particularly at the minima ) , and the intrinsic variations was made by dividing the observations into two data sets at julian date @xmath31 ( see figure [ fig : hist ] ) . as mentioned above , the measurements in the @xmath18 and @xmath19 filters in data set # 1 do not cover the primary eclipse , and they were therefore not used for the final fits . light curve analyses were carried out separately for each bandpass in each data set ( @xmath32 for data set # 1 , and @xmath0 for data set # 2 ) , and also for the combined data . the light curve solutions were performed with the wilson - devinney code @xcite in a mode appropriate for semi - detached systems , in accordance with the information available for the stars from paper i. specifically , we used mode 5 " for a secondary that is filling its limiting lobe . detached and overcontact configurations were tried as well , but the solutions were never as satisfactory as those in the semi - detached mode . they always converged towards the secondary star filling its roche lobe and the primary well within its critical surface , in agreement with the assessment in paper i. solutions were found for the modified gravitational potential of the primary ( @xmath33 , in the wilson - devinney usage ) , the inclination angle ( @xmath34 ) , the relative monochromatic luminosity of the primary ( @xmath35 ) , the mean temperature of the secondary ( @xmath36 ) , and a phase offset ( @xmath37 ) . the temperature of the primary was held fixed at the spectroscopic value of @xmath38 k , as was the mass ratio @xmath39 ( see [ sec : spec ] ) . we have assumed the orbit to be circular , consistent with the results of the new spectroscopic analysis below . the monochromatic luminosity of the secondary ( @xmath40 ) was computed by the program directly from the temperatures of the primary and secondary , the luminosity of the primary , the radiation laws , and the geometry of the system . limb - darkening coefficients were interpolated from the tables by @xcite for the appropriate temperatures and for surface gravities of @xmath41 and @xmath42 ( see [ sec : spec ] ) . a value of 0.5 was adopted for the bolometric albedo of both components , appropriate for stars with convective envelopes . gravity brightening coefficients were interpolated from the models by @xcite . aside from the usual light elements , the distortions mentioned above in the light curves suggested the presence of one or more spots requiring additional free parameters . the wilson - devinney code is capable of accounting for spots ( assumed to be circular in shape ) based on a simple representation with four additional parameters : the longitude ( @xmath43 , in degrees ) measured counter - clockwise ( as seen from above ) from the line joining the centers of the stars , the latitude ( @xmath44 , in degrees ) , the angular radius ( @xmath45 , in degrees ) , and a temperature factor ( @xmath46 ) representing the contrast in temperature between the spot and the surrounding photosphere . numerical experiments showed that in our case the fits can not usually distinguish the component on which the spots are located , since similarly good solutions resulted with the spots on either star . in the rs cvn systems the spots are typically located on the cooler star , which is usually the more active component . there are a number of cases , however , in which the spots have instead been shown to be on the hotter star ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? in bd@xmath1706 there are strong indications that the activity is in fact associated with the hotter star ( the primary ) , as evidenced by the h@xmath11 emission ( see paper i ) and also the x - ray properties discussed later in [ sec : xray ] . it is therefore not unreasonable to assume in the following that the spottedness in this system is primarily associated with the hotter star . more complicated models with spots on both stars are possible , but are probably not justified given the data available . the maximum brightness outside of eclipse is much the same for data sets # 1 and # 2 , although it occurs at phase 0.75 in the first set and at phase 0.25 in the second . the lower brightness levels at the other quadrature in each data set are not quite as similar to each other and show more scatter in the photometric observations . this suggests that the lower light levels may be due to one or more cool spots ( which tend to change with time in other similarly active stars ) , as opposed to hot spots being the cause of the higher brightness at the other quadratures . spots that are cooler than the photosphere have typically been found also in virtually all rs cvn systems ( although warm features have been reported in a few cases ; see , e.g. , * ? ? ? thus we will assume here that the spots are cool . the situation for bd@xmath1706 obviously calls for the cool spots to be located at roughly opposite longitudes in the two epochs , and we found that a single spot at each epoch gave an adequate representation of the light curves , within the observational errors . further tests showed that , due to the strong correlation between the size and temperature of the spots , it was not possible to determine both simultaneously from the observations at our disposal . grids of preliminary solutions at fixed values of _ tf _ from 0.5 to 1.0 indicated a slight preference for a spot in data set # 1 with @xmath47 ( 800 k cooler than the photosphere ) , and a similar spot in data set # 2 with @xmath48 ( 650 k cooler than the photosphere ) . these values are within the typical range seen in other rs cvn binaries @xcite . therefore , in the following we have held _ tf _ fixed at these values , and the spot radius was left free . we also found that the sensitivity of the solutions to the latitude of the spot is very weak . experiments with fixed values of @xmath44 from 0 to 90 indicated only a very slight preference for mid - latitude features , with fits at higher or lower latitudes being almost indistinguishable . although spots on the sun are usually found at low latitudes , doppler imaging studies have shown that rs cvn systems tend to display high - latitude features and even spots that straddle the pole ( e.g. , * ? ? ? * ; * ? ? ? while bd@xmath1706 is similar to the rs cvn binaries in many respects , the latter systems typically rotate much more rapidly ( see , e.g. , * ? ? ? theoretical studies suggest that rapid rotation in convective stars leads to the emergence of spots at higher latitudes @xcite , and there is some observational support for this trend ( see * ? ? ? therefore , for this study we have adopted a fixed latitude of 45 , in agreement with the above and also with the theoretical work by @xcite that focuses specifically on evolved stars similar to those in bd@xmath1706 . initially light curve solutions were obtained separately for each bandpass in each data set ( @xmath32 for data set # 1 , and @xmath0 for data set # 2 ) , to evaluate the consistency between them . the results were found to be reasonably similar , except that the second data set showed a larger scatter in the temperature of the secondary ( @xmath36 ) and in the potential of the primary ( @xmath33 ) . solutions were also obtained from the combined @xmath32 and @xmath0 observations from data sets # 1 and # 2 , respectively , which were also quite similar to each other . since the main geometric properties ( @xmath33 and @xmath34 ) are not expected to change from season to season , the results for these two parameters from the two data sets were averaged ( with weights proportional to the internal errors ) , and fixed in subsequent analyses . we then repeated the fits and computed the average secondary temperature ( @xmath49 k ) from the simultaneous multi - band analyses . finally , we fixed @xmath36 to this value , and we solved for the remaining parameters including the size ( @xmath45 ) and longitude ( @xmath43 ) of the spots , which clearly do change with time . the results are given in table [ tab : ds1 ] and table [ tab : ds2 ] for data set # 1 and # 2 , respectively . we show the separate fits for each filter before fixing any of the parameters , and also the simultaneous multi - band analyses in the last column calculated with @xmath33 , @xmath34 , and @xmath36 set to their average values . also listed are the fractional radii ( @xmath50 , @xmath51 , @xmath52 , @xmath53 , in terms of the separation ) as well as the relative radius of a spherical star with the same volume as the distorted stars ( @xmath54 ) , for the primary and secondary . the uncertainties given in these tables are standard errors as reported by the wilson - devinney code . experiments were also carried out to explore the significance of third light ( @xmath55 ) , but in all cases the results were consistent with @xmath56 . both eclipses in bd@xmath1706 are partial , with approximately 65% of the light of the primary being blocked at phase 0.0 . the latter star is nearly spherical in shape since the difference between its polar radius and the radius directed towards the inner lagrangian point is less than 0.5% . parameters @xmath50 , @xmath51 , @xmath52 , and @xmath53 in table [ tab : ds1 ] and table [ tab : ds2 ] confirm that the secondary is distorted because it fills its roche lobe . a graphical representation of the observations and light curve solutions is given in figure [ fig : lc1 ] and figure [ fig : lc2 ] for data sets # 1 and # 2 . the @xmath57 residuals are shown at the bottom for each passband . the @xmath18 and @xmath19 light curves in data set # 1 lack coverage at the primary minimum , and do not allow one to determine any of the parameters reliably . although we show them here for completeness , these observations were not used in any of the fits described above . for these @xmath18 and @xmath19 solutions all geometric and radiative parameters as well as the spot parameters were adopted from the combined @xmath32 solution , and only the primary luminosity was allowed to vary . the fits are consequently poorer . the spots that were found to provide a good fit to the observations in both data sets are represented in figure [ fig : spotpic ] , with the size and separation of the stars shown to scale . the difference in longitude between one epoch and the other ( separated by approximately 1.4 years ) is about 140 . the surface area covered by the spots is in line with what is seen in other active systems , and reaches @xmath5820% of one hemisphere of the primary star in data set # 1 , and @xmath5817% in data set # 2 . the procedures used here for the reduction of the spectroscopic material are similar to those described in paper i in that we derived radial velocities for the two components using todcor @xcite , a two - dimensional cross - correlation technique well suited to our relatively low s / n spectra . however , a number of improvements were introduced that justified a re - analysis of the original data . first , we now have at our disposal a new library of synthetic spectra based on more recent model atmospheres by r. l.kurucz , which provide a somewhat better match to the spectra of real stars . the use of these new templates has a small effect on the temperature and rotational velocity ( @xmath59 ) derived for the components of bd@xmath1706 , and can potentially affect the radial velocities as well . we re - determined the @xmath59 values by running extensive grids of correlations as described in paper i. the results were @xmath60 and @xmath61 , with estimated errors of 1 and 2 , respectively . the primary value is 1 larger than in paper i , and the secondary value is unchanged . the new effective temperatures , derived from similar grids , were @xmath38 k and @xmath62 k , with uncertainties of 100 k. these differ by less than 100 k from the determinations in paper i. the new primary value was adopted for the light curve solutions described above . a second improvement has to do with residual systematic errors in the raw velocities , mostly due to the relatively narrow spectral window of these observations . experience with similar spectroscopic material for other binary systems has shown that these errors can be quite significant in some cases @xcite . we have investigated these errors for the present system by means of numerical simulations that were discussed in detail by @xcite . briefly , we generated a set of artificial binary spectra by combining the primary and secondary templates in the appropriate ratio and applying velocity shifts for both components as computed from a preliminary orbital solution at the actual times of observation of each of our spectra . these artificial spectra were then processed with todcor in exactly the same way as the real observations , and the resulting velocities were compared with the input ( synthetic ) values . the differences @xmath22todcor minus synthetic@xmath23 are typically under 0.5 , but they are systematic in nature . even though the effect on the masses is relatively small in this particular case , we have nevertheless applied these differences as corrections to the measured velocities of bd@xmath1706 . the corrected velocities are listed in table [ tab : rvs ] ( columns 2 and 3 ) . the differences with the values reported in paper i are generally below 1 . as a result of the proximity and distorted shapes of the stars ( see , e.g. , figure [ fig : spotpic ] ) , the radial velocities measured for the components of bd@xmath1706 are biased to some extent because of the loss of symmetry in the intensity distribution across the disks of the stars . in addition , the observations collected during the eclipses suffer from the well - known rossiter effect @xcite , which depends on the rotation rate . the combined magnitude of these two effects on the measured velocities , computed with the aid of the wilson - devinney program , is illustrated in figure [ fig : rvdist ] . the change in the primary velocities is essentially confined to the eclipses ( and is mostly due to the rossiter effect since that star is not far from the spherical shape ) , but the secondary velocities show changes throughout the entire orbital cycle because of the highly distorted shape of the roche - lobe filling star . the corrections for the case of bd@xmath1706 can reach values as large as 4 , and are listed in columns 4 and 5 of table [ tab : rvs ] . the final values for the radial velocities used below , after applying these adjustments , are given in columns 6 and 7 . while the corrections described in the preceding paragraph ( which depend only on the geometry of the system ) do not change with time , the surface features modeled in the light curves , which can also affect the measured velocities , do change significantly as seen in [ sec : phot ] . spots have an effect on the measured velocities because they too contribute to break the symmetry of the intensity distribution on the stellar disks . in this case the effect will be on the primary velocities because the spots are assumed to be on that star . the phase dependence of the effect is seen in figure [ fig : rvspot ] separately for data sets # 1 and # 2 , and was derived once again using the wilson - devinney code . specifically , the effect is calculated as the difference between the velocities from purely keplerian motion and those computed by integration over the visible disk of the star , with its non - uniform brightness distribution due to the spots . it is unclear whether the velocities we have measured by cross - correlation are affected by the full amplitude of the correction computed by this program , but presumably they will be biased to some extent in the direction indicated . unfortunately , as seen in figure [ fig : hist ] our spectroscopic observations were not simultaneous with either of the photometric data sets , so that corrections for this effect can not be applied . the magnitude of the distortion in the velocities can reach values as large as 1 ( figure [ fig : rvspot ] ) . nevertheless , as a test we applied these corrections to our velocities separately for both data sets and recomputed the spectroscopic orbit to explore the impact on the elements . for data set # 1 the changes in all derived quantities including the minimum masses are of the order of 1@xmath28 or smaller . for data set # 2 they are less than 0.5@xmath28 . our final spectroscopic orbital solution was obtained using the measured velocities corrected for proximity effects and for the rossiter distortions only . the minimum masses and the mass ratio we determine are only slightly different from those reported in paper i , but we now find that the eccentricity is indistinguishable from zero . the original solution in paper i gave a value of @xmath63 , more than 5@xmath28 different from zero . this presented something of a puzzle given that tidal forces for stars with deep convective envelopes such as those in bd@xmath1706 are expected to circularize the orbit very quickly for the relatively short period in this system , considering that both components are giants . the distortion corrections applied here appear to have solved this dilemma , and a circular orbit is adopted hereafter . the revised elements of the spectroscopic orbit are given in table [ tab : specelem ] . the ephemeris for the primary eclipse that we derive from this orbit is that given in eq.(1 ) , and was adopted for the light - curve solutions . the light ratio at the wavelength of our observations ( 5187 ) , computed as described by @xcite , is @xmath64 . a slight correction of @xmath65 is required to convert this to the visual band ( see paper i ) . the residuals of the individual observations from the orbital fit are listed in columns 8 and 9 of table [ tab : rvs ] . the light elements in the last column of table [ tab : ds1 ] and table [ tab : ds2 ] combined with the spectroscopic elements in table [ tab : specelem ] lead to the absolute masses and radii of the components , along with other derived properties . as mentioned earlier , the uncertainties listed for the light elements are strictly internal errors reported by the wilson - devinney code . more realistic errors for the inclination angle , the primary and secondary potentials , and for the secondary temperature were adopted based in part on the agreement between the solutions in the different passbands and on the sensitivity of the fits to the various parameters : @xmath66 , @xmath67 , @xmath68 , and @xmath69 k. the physical properties of the stars are summarized in table [ tab : absdim ] . the absolute masses are determined with errors of 1.0% and 1.7% for the primary and secondary , respectively , while the radii are good to 2.6% and 1.9% . since the difference in the depths of the minima is a very sensitive indicator of the temperature difference between the components , the later quantity is typically best determined from the light curves . in this case the result ( @xmath70 k ) is excellent agreement with our spectroscopic determination of @xmath71 k inferred in [ sec : spec ] , based on grids of cross - correlations with synthetic spectra . the ratio of the luminosities in the visual band from the light curve solutions is @xmath72 , whereas the estimate derived from the bolometric luminosities and bolometric corrections ( table [ tab : absdim ] ) is @xmath73 . the first of these values is larger than the spectroscopic estimate [ @xmath74 , a discrepancy possibly related to the presence of spots . the measured projected rotational velocity of the secondary component , @xmath75 , is consistent with the predicted value for synchronous rotation within the errors , as expected from the fact that the star is filling its critical surface . on the other hand , the rotation of the primary is formally some 15% faster than the predicted value , a 2.8@xmath28 difference that may perhaps be significant . the primaries of many other algol systems have also been found to rotate supersynchronously , and in some cases even close to the centrifugal limit ( see , e.g. , * ? ? ? this spin - up is usually explained by angular momentum transfer through accretion @xcite . the inferred distance to bd@xmath1706 of 595 @xmath76 51 pc relies on the apparent brightness of the system ( @xmath77 ) , the interstellar extinction adopted in paper i ( @xmath78 ) , and bolometric corrections from @xcite . the count rates over the full spectral range ( 0.12.4 kev ) of the x - ray observations of bd@xmath1706 carried out with the rosat satellite are shown as a function of orbital phase in figure [ fig : xraycurve]a . two features of this broad - band light curve are immediately obvious : ( a ) while the primary eclipse is clearly present , there is no sign of a secondary eclipse near phase 0.5 ; ( b ) one measurement immediately preceding phase 0.5 and another one following it are about a factor of two higher than the average level of the neighboring points . the first of these features supports the notion that the primary is the x - ray active star in the system , and that the secondary contributes little , if anything , to the total x - ray luminosity of the binary . this observation is consistent with the activity evidence presented in paper i , where the primary was shown to be the dominant source of the h@xmath11 emission . as reported there , several red spectra of the binary outside of eclipse displayed relatively weak h@xmath11 absorption ( residual intensity @xmath79 ) , whereas one spectrum obtained during primary eclipse displayed a much deeper h@xmath11 line ( residual intensity @xmath80 ) . this change was interpreted as an indication that the emission from the active primary fills in the normal absorption line of the secondary at most phases in the orbit making it appear weak , except near phase 0.0 , where the primary is blocked and the secondary dominates the spectrum . the second feature unusually large x - ray flux of the two observations near phase 0.5 could possibly be due to two separate flaring events , which would not be uncommon in this type of system . we discuss this further below . a comparison with the light curves in the optical reveals two other notable characteristics of the x - ray light curve : ( c ) the duration of the primary eclipse in x - rays is much longer than in the optical ; ( d ) the primary minimum is considerably deeper in x - rays than it is in the optical . to illustrate these two features we have superimposed on figure [ fig : xraycurve]a the light curve in the @xmath20 band from data set # 1 , which is contemporaneous with the rosat observations . the optical light curve is normalized to agree with the average level of the x - ray curve outside of eclipse . we note that at the epoch of these observations the configuration of the system inferred in [ sec : phot ] includes a cool spot on the primary that is on the far side of the star at phase 0.75 and therefore does not affect the @xmath20-band light curve at the second quadrature . the solid line in figure [ fig : xraycurve]a represents the modeled @xmath20-band light curve without the effect of the spot , and the dotted line includes the distortion due to the spot , which makes the system fainter at the first quadrature . it can be seen from the figure that the total duration of the primary eclipse in the optical is approximately 0.11 in phase ( @xmath582.1 d ) , while the duration in x - rays is roughly 2.9 times longer ( @xmath580.32 in phase , or @xmath586.0 d ) . this implies that the size of the x - ray emitting region around the primary must be larger than the star itself by approximately the same factor . since the radius of the star in terms of the separation is @xmath81 ( parameter @xmath54 in table [ tab : ds1 ] ) , it follows that the radius of the associated x - ray region is @xmath82 . we note also that the mass ratio of the binary leads independently to a mean radius for the critical surface of the primary of @xmath83 . thus , the size of the x - ray emitting region around the primary happens to be very close to that of its roche lobe possibly not a coincidence and one may speculate that the x - ray emitting plasma is perhaps contained within the critical surface . since there is no evidence from the x - ray light curve of any significant emission from the secondary , one might also speculate that the coronal region around that component represents a relatively thin layer that does not extend significantly above the photosphere of the star , which already fills its roche lobe . as a result , bd@xmath1706 would appear to exhibit the curious property of being essentially an overcontact system in x - rays , and a semidetached system in the optical . at optical wavelengths ( @xmath20 band ) the brightness of bd@xmath1706 during the primary minimum drops to about 62% of the level outside of eclipse ( phase 0.75 ) . in x - rays the eclipse is much deeper , reaching down to @xmath5830% of maximum . since blocking by the same star ( the secondary ) produces significantly different depths in the optical and in x - rays , and given that the apparent size of the primary in x - rays is much larger , one possible explanation for the deeper minimum than in the optical may be that the x - ray brightness distribution on the primary is considerably more concentrated towards the center of the disk ( or toward the inner lagrangian point , which is projected not far from the center ) than in the optical . in order to simulate this situation with light - curve solution programs such as wilson - devinney , not originally intended for this type of problem , the relative sizes of the stars must be changed compared to the geometry inferred from the optical light curves , effectively by altering the mass ratio . furthermore , variations in the predicted light curve near phase 0.5 due to the changing aspect ratio of the primary ( if it is assumed to fill its critical surface in x - rays ) must be ignored as they probably do not apply at x - ray wavelengths . we find a reasonably good agreement between such a model " and the observed light curve when the effective size of the x - ray emitting region of the primary is reduced to approximately the same size as the photosphere of the lobe - filling secondary ( see figure [ fig : xraycurve]b ) . of course , any parameters derived in this way are physically meaningless and are only intended to achieve a qualitatively good fit . the reference provided by this very crude model does seem to suggest , nevertheless , that the x - ray flux preceding the primary eclipse ( beginning at phase 0.75 , and perhaps including the descending branch ) is somewhat lower than the flux after the minimum at symmetrical phases . as mentioned above , the two high x - ray points near phase 0.5 might be due to two separate flare events ( since an observation intermediate between these two shows a normal flux level ) . x - ray flares are common in rs cvn systems ( see , e.g. , * ? ? ? * ; * ? ? ? on the other hand , the semidetached configuration in the optical is an indication that mass transfer by roche - lobe overflow may be occurring . due to the relatively large size of the primary , a stream of material from the secondary to the primary would actually impact the primary star directly rather than sweep past it and form a disk ( * ? ? ? * see also figure 11 in paper i ) . given the orbital phases at which the discrepant x - ray points in figure [ fig : xraycurve]a occur ( 0.423 and 0.539 ) , just before and just after central blocking of the secondary by the primary , it is possible that the sudden increase in x - ray flux is related to the viewing angle of the stream associated with mass transfer , which would be roughly aligned with the line of sight at these phases . detailed hydrodynamical modeling is required to investigate this further . in figure [ fig : hardx ] we have split the rosat x - ray observations of bd@xmath1706 into a soft band ( 0.10.5 kev ) and a hard band ( 0.62.4 kev ) in order to investigate the possibility that there might be a difference in the behavior as a function of orbital phase . the run of the count rates turns out to be quite similar , however , with the soft band being slightly lower throughout ( figure [ fig : hardx]a ) . the changes in the hardness ratio as defined in [ sec : obs ] are shown in figure [ fig : hardx]b . within the errors the _ hr _ values are fairly constant , the decrease toward the primary minimum being only marginally significant . in particular , the observations near phase 0.5 pointed out in the broad - band light curve ( figure [ fig : xraycurve]a ) that are possibly associated with flaring episodes show no sign that the x - ray energy becomes significantly harder ( observations marked with circles ) , as would be expected if the temperature were to increase due to heating of the plasma . the combination of our new @xmath0 photometry with a reanalysis of the spectroscopy reported in paper i has allowed us to obtain highly precise absolute masses and radii for both components of the semidetached system bd@xmath1706 , as well as the effective temperatures and other properties . among the dozen or so known cool algols , this is now by far the one with the best determined physical parameters . only three others have published light - curve solutions to date ( rz cnc , ar mon , and rv lib ) . in paper i the eclipsing nature of bd@xmath1706 was not yet known , so there was no direct information on the orbital inclination angle although some constraints were available based on other evidence . qualitative conclusions on the history and evolutionary state of the system were drawn there based upon the assumption that the inclination angle was large , and therefore that the true values of the masses and the radii were not far from the lower limits derived for those quantities . we can now confirm the validity of that assumption in paper i , and consequently also of those conclusions : the present configuration of bd@xmath1706 was probably reached as a result of case b mass transfer @xcite , and the initial mass ratio of the system is likely to have been close to unity ( see * ? ? ? * ; * ? ? ? * ) so that the original secondary had enough time to begin its evolution off the main sequence before the original primary reached the point of roche - lobe overflow and started transferring mass onto it ( eventually leading to a mass ratio reversal ) . the minimum amount of mass exchanged ( under conservative mass transfer ) is @xmath84 m@xmath4 . thus , the secondary appears to have lost at least 2/3 of its original mass . as noted in our earlier study of this system , although the current mass ratio ( @xmath85 ) is quite typical of classical algol systems the relatively cool temperature of the current primary ( 5000 k ) is not . this difference with the much hotter primaries of the classical algols is at least qualitatively well explained by the ideas about the evolution of cool algols mentioned above and discussed at length in paper i. evidence from our optical light curves suggests the presence of large surface inhomogeneities ( spots ) which are quite common in similarly active systems . significant shifts in the location ( longitude ) of these features are seen in bd@xmath1706 over a period of 1 - 2 years , which are also consistent with changes seen in other rs cvn binaries . our x - ray light curve showing only the primary eclipse confirms the report in paper i that the hotter and more massive primary star is the dominant site of activity in this binary . it is interesting to note that this is not typically the case in rs cvn systems , where the cooler star is usually the more magnetically active component . the x - ray observations also offer interesting clues about the geometry and other properties of the x - ray emitting region , which may be confined to the size of the critical lobe of the current primary star . photometric and spectroscopic studies of other binaries systems identified as cool algols are currently underway as part of a long - term project to study this very interesting class of active stars . detailed analyses of those cases along the lines of the one presented here for bd@xmath1706 are expected to provide further glimpses into their properties and evolutionary histories . we thank the anonymous referee for numerous comments and helpful suggestions . this research has made use of the simbad database , operated at cds , strasbourg , france , and of nasa s astrophysics data system abstract service . rn wishes to acknowledge financial support from the bundesministerium fr bildung und forschung through the deutsche zentrum fr luft- und raumfahrt e.v . ( dlr ) under grant number 50 or 0003 . lm would like to acknowledge the help of julia lynch and akbar rizvi in making these observations . the gettysburg college observatory is supported by gettysburg college with additional funding for the ccd camera provided by the national science foundation . garca - alvarez , d. , foing , b. h. , montes , d. , oliveira , j. , doyle , j. g. , messina , s. , lanza , a. f. , rodon , m. , abbott , j. , ash , t. d. c. , baldry , i. k. , bedding , t. r. , buckley , d. a. h. , cami , j. , cao , h. , catala , c. , cheng , k. p. , domiciano de souza jr . , a. , donati , j. f. , hubert , a. m. , janot - pacheco , e. , hao , j. x. , kaper , l. , kaufer , a. , leister , n. v. , neff , j. e. , neiner , c. , orlando , s. , otoole , s. j. , schfer , d. , smartt , s. j. , stahl , o. , telting , j. , & tubbesing , s. 2003 , , 397 , 285 c@c@ c@c@ c@c@ c@c 772.6936 & @xmath920.412 & 827.7321 & @xmath920.457 & 903.5477 & @xmath920.486 & 1208.6844 & @xmath920.282 + 772.7038 & @xmath920.400 & 828.5234 & @xmath920.447 & 914.5239 & @xmath920.328 & 1209.5255 & @xmath100.113 + 773.6906 & @xmath920.315 & 834.5772 & @xmath920.396 & 915.5189 & @xmath920.233 & 1209.5601 & @xmath100.095 + 777.6886 & @xmath920.341 & 840.5560 & @xmath920.156 & 1133.7158 & @xmath100.017 & 1209.5601 & @xmath100.120 + 777.7616 & @xmath920.341 & 844.5362 & @xmath920.459 & 1136.7110 & @xmath920.408 & 1209.6537 & @xmath100.125 + 778.6654 & @xmath920.418 & 845.5979 & @xmath920.492 & 1140.6862 & @xmath920.421 & 1209.7217 & @xmath100.133 + 784.6604 & @xmath920.085 & 846.5413 & @xmath920.491 & 1142.6528 & @xmath920.223 & 1210.5415 & @xmath920.256 + 784.7649 & @xmath920.089 & 853.6187 & @xmath920.402 & 1148.6430 & @xmath920.434 & 1216.5883 & @xmath920.362 + 795.6793 & @xmath920.307 & 854.6542 & @xmath920.445 & 1149.6591 & @xmath920.411 & 1219.5979 & @xmath920.104 + 798.6679 & @xmath920.450 & 855.5462 & @xmath920.435 & 1157.6063 & @xmath920.492 & 1220.6234 & @xmath920.230 + 799.6417 & @xmath920.421 & 857.5909 & @xmath920.328 & 1162.6686 & @xmath920.069 & 1221.6098 & @xmath920.306 + 800.6508 & @xmath920.342 & 859.5101 & @xmath920.137 & 1165.6690 & @xmath920.371 & 1225.5947 & @xmath920.388 + 802.6055 & @xmath920.172 & 859.6675 & @xmath920.094 & 1166.6970 & @xmath920.440 & 1229.5403 & @xmath920.332 + 802.7780 & @xmath920.158 & 860.6232 & @xmath920.200 & 1170.6638 & @xmath920.285 & 1230.5336 & @xmath920.420 + 803.7220 & @xmath920.114 & 870.6699 & @xmath920.275 & 1173.6636 & @xmath920.382 & 1231.5202 & @xmath920.459 + 809.5834 & @xmath920.473 & 878.5792 & @xmath920.091 & 1180.6680 & @xmath920.187 & 1232.5722 & @xmath920.504 + 814.5945 & @xmath920.409 & 889.5291 & @xmath920.309 & 1190.6157 & @xmath100.114 & 1268.5297 & @xmath920.405 + 824.6819 & @xmath920.397 & 896.5880 & @xmath920.273 & 1198.6265 & @xmath920.303 & 1275.5187 & @xmath920.083 + 825.5523 & @xmath920.462 & 897.5484 & @xmath920.107 & 1207.6736 & @xmath920.346 & 1276.5165 & @xmath920.163 + 827.5643 & @xmath920.492 & 899.5315 & @xmath920.351 & 1208.5180 & @xmath920.303 & 1277.5134 & @xmath920.237 + c@c@ c@c@ c@c@ c@c 772.6907 & @xmath920.432 & 827.5608 & @xmath920.524 & 899.5290 & @xmath920.356 & 1208.5156 & @xmath920.331 + 773.6792 & @xmath920.385 & 827.7283 & @xmath920.494 & 903.5449 & @xmath920.500 & 1208.6819 & @xmath920.312 + 777.6699 & @xmath920.392 & 828.5199 & @xmath920.476 & 914.5212 & @xmath920.352 & 1209.5229 & @xmath100.046 + 777.6784 & @xmath920.393 & 834.5746 & @xmath920.423 & 915.5230 & @xmath920.301 & 1209.5577 & @xmath100.031 + 777.7580 & @xmath920.401 & 840.5508 & @xmath920.167 & 1117.7249 & @xmath920.405 & 1209.6508 & @xmath100.041 + 778.6590 & @xmath920.449 & 840.5621 & @xmath920.182 & 1118.7028 & @xmath920.468 & 1209.7102 & @xmath100.035 + 784.6547 & @xmath920.102 & 842.6729 & @xmath920.344 & 1129.6644 & @xmath920.468 & 1209.7191 & @xmath100.073 + 784.7536 & @xmath920.107 & 844.5336 & @xmath920.496 & 1130.7223 & @xmath920.434 & 1210.5387 & @xmath920.273 + 785.6335 & @xmath920.304 & 845.5953 & @xmath920.523 & 1133.7128 & @xmath920.040 & 1215.6181 & @xmath920.456 + 795.6681 & @xmath920.350 & 846.5346 & @xmath920.508 & 1136.7083 & @xmath920.421 & 1216.5854 & @xmath920.367 + 797.6278 & @xmath920.458 & 853.6160 & @xmath920.428 & 1140.6837 & @xmath920.444 & 1219.5952 & @xmath920.122 + 798.6644 & @xmath920.459 & 854.6436 & @xmath920.473 & 1142.6502 & @xmath920.247 & 1220.6178 & @xmath920.273 + 798.7559 & @xmath920.480 & 854.6458 & @xmath920.471 & 1148.6445 & @xmath920.451 & 1221.6071 & @xmath920.351 + 799.6356 & @xmath920.458 & 855.5436 & @xmath920.470 & 1149.6565 & @xmath920.429 & 1225.5921 & @xmath920.407 + 800.6471 & @xmath920.402 & 857.5882 & @xmath920.369 & 1157.6036 & @xmath920.507 & 1229.5377 & @xmath920.347 + 802.6019 & @xmath920.142 & 859.5074 & @xmath920.162 & 1162.6661 & @xmath920.083 & 1230.5310 & @xmath920.433 + 803.7185 & @xmath920.140 & 859.6649 & @xmath920.126 & 1165.6664 & @xmath920.394 & 1231.5176 & @xmath920.474 + 809.5799 & @xmath920.491 & 860.6207 & @xmath920.193 & 1166.6945 & @xmath920.464 & 1232.5692 & @xmath920.529 + 813.7879 & @xmath920.273 & 870.6672 & @xmath920.331 & 1173.6611 & @xmath920.393 & 1268.5270 & @xmath920.425 + 813.7911 & @xmath920.292 & 878.5685 & @xmath920.134 & 1180.6655 & @xmath920.179 & 1274.5501 & @xmath920.310 + 813.7931 & @xmath920.265 & 889.5291 & @xmath920.346 & 1190.6129 & @xmath100.045 & 1275.5214 & @xmath920.090 + 814.5910 & @xmath920.419 & 896.5854 & @xmath920.299 & 1198.6206 & @xmath920.326 & 1276.5194 & @xmath920.191 + 824.6780 & @xmath920.425 & 897.5454 & @xmath920.144 & 1207.6708 & @xmath920.361 & 1277.5160 & @xmath920.305 + 825.5623 & @xmath920.491 & & & & & & + c@c@ c@c@ c@c@ c@c 474.5480 & @xmath920.417 & 528.5478 & @xmath920.346 & 824.6762 & @xmath920.472 & 1142.6482 & @xmath920.327 + 474.5930 & @xmath920.434 & 528.5859 & @xmath920.323 & 825.5468 & @xmath920.538 & 1148.6430 & @xmath920.535 + 474.6310 & @xmath920.442 & 529.5529 & @xmath920.069 & 827.5594 & @xmath920.570 & 1149.6551 & @xmath920.500 + 474.6720 & @xmath920.436 & 529.5683 & @xmath920.083 & 827.7268 & @xmath920.571 & 1157.6019 & @xmath920.577 + 477.5058 & @xmath920.489 & 529.5919 & @xmath920.096 & 828.5184 & @xmath920.529 & 1162.6646 & @xmath920.147 + 477.5638 & @xmath920.499 & 530.5133 & @xmath920.406 & 834.5729 & @xmath920.481 & 1163.6604 & @xmath920.336 + 477.6088 & @xmath920.496 & 530.5567 & @xmath920.399 & 840.5600 & @xmath920.238 & 1165.6650 & @xmath920.468 + 477.6338 & @xmath920.497 & 530.5831 & @xmath920.414 & 844.5315 & @xmath920.543 & 1166.6930 & @xmath920.531 + 477.6518 & @xmath920.497 & 546.5490 & @xmath920.457 & 845.5935 & @xmath920.575 & 1170.6589 & @xmath920.399 + 477.6718 & @xmath920.497 & 547.5586 & @xmath920.250 & 846.5327 & @xmath920.573 & 1173.6598 & @xmath920.461 + 483.4803 & @xmath920.384 & 548.5461 & @xmath920.096 & 853.6143 & @xmath920.483 & 1180.6640 & @xmath920.251 + 483.4933 & @xmath920.396 & 549.5267 & @xmath920.397 & 854.6396 & @xmath920.526 & 1190.6113 & @xmath920.043 + 483.6473 & @xmath920.417 & 552.5314 & @xmath920.497 & 855.5412 & @xmath920.532 & 1198.6206 & @xmath920.397 + 483.6563 & @xmath920.409 & 772.6879 & @xmath920.494 & 857.5866 & @xmath920.433 & 1207.6685 & @xmath920.435 + 489.6607 & @xmath920.470 & 773.6775 & @xmath920.442 & 859.5056 & @xmath920.222 & 1208.6799 & @xmath920.383 + 489.6847 & @xmath920.469 & 777.6683 & @xmath920.454 & 859.6624 & @xmath920.208 & 1209.5563 & @xmath920.074 + 491.5536 & @xmath920.032 & 777.7565 & @xmath920.462 & 859.6700 & @xmath920.192 & 1209.6492 & @xmath920.059 + 491.5636 & @xmath920.048 & 778.6568 & @xmath920.506 & 860.6190 & @xmath920.251 & 1209.7085 & @xmath920.078 + 491.5736 & @xmath920.041 & 784.6533 & @xmath920.159 & 870.6653 & @xmath920.421 & 1209.7176 & @xmath920.061 + 491.6055 & @xmath920.027 & 784.7522 & @xmath920.170 & 878.5668 & @xmath920.197 & 1210.5372 & @xmath920.346 + 497.6100 & @xmath920.454 & 785.6314 & @xmath920.364 & 889.5257 & @xmath920.408 & 1215.6097 & @xmath920.538 + 497.6280 & @xmath920.462 & 795.6657 & @xmath920.392 & 896.5838 & @xmath920.349 & 1216.5839 & @xmath920.454 + 497.6800 & @xmath920.461 & 797.6207 & @xmath920.508 & 897.5429 & @xmath920.190 & 1219.5937 & @xmath920.195 + 504.5271 & @xmath920.544 & 798.6629 & @xmath920.532 & 899.5276 & @xmath920.417 & 1220.6126 & @xmath920.356 + 504.5450 & @xmath920.549 & 799.6332 & @xmath920.525 & 903.5432 & @xmath920.561 & 1221.6048 & @xmath920.405 + 504.5686 & @xmath920.553 & 800.6455 & @xmath920.460 & 1117.7233 & @xmath920.473 & 1225.5905 & @xmath920.474 + 505.5372 & @xmath920.572 & 802.6005 & @xmath920.288 & 1118.7001 & @xmath920.522 & 1229.5362 & @xmath920.406 + 505.5740 & @xmath920.586 & 802.7744 & @xmath920.241 & 1118.7126 & @xmath920.524 & 1232.5674 & @xmath920.580 + 507.4900 & @xmath920.525 & 803.7233 & @xmath920.201 & 1129.6626 & @xmath920.527 & 1268.5255 & @xmath920.505 + 507.5100 & @xmath920.524 & 809.5782 & @xmath920.545 & 1130.7208 & @xmath920.508 & 1274.5253 & @xmath920.390 + 507.6310 & @xmath920.523 & 809.7652 & @xmath920.545 & 1133.7111 & @xmath920.136 & 1275.5227 & @xmath920.161 + 508.5222 & @xmath920.474 & 813.7806 & @xmath920.365 & 1136.7057 & @xmath920.494 & 1276.5209 & @xmath920.254 + 514.5140 & @xmath920.494 & 814.5891 & @xmath920.458 & 1140.6819 & @xmath920.506 & 1277.5194 & @xmath920.384 + 514.6029 & @xmath920.499 & & & & & & + c@c@ c@c@ c@c@ c@c 474.5550 & @xmath920.489 & 514.5256 & @xmath920.575 & 814.5870 & @xmath920.524 & 1140.6806 & @xmath920.565 + 474.5950 & @xmath920.501 & 514.6011 & @xmath920.561 & 824.6744 & @xmath920.544 & 1142.6471 & @xmath920.389 + 474.6380 & @xmath920.497 & 528.5456 & @xmath920.425 & 825.5444 & @xmath920.594 & 1148.6419 & @xmath920.586 + 474.6740 & @xmath920.497 & 528.5838 & @xmath920.385 & 827.5575 & @xmath920.625 & 1149.6539 & @xmath920.556 + 477.5108 & @xmath920.552 & 529.5488 & @xmath920.169 & 827.7251 & @xmath920.624 & 1157.6004 & @xmath920.626 + 477.5908 & @xmath920.571 & 529.5664 & @xmath920.165 & 828.5168 & @xmath920.599 & 1162.6634 & @xmath920.214 + 477.6108 & @xmath920.569 & 529.5902 & @xmath920.184 & 834.5717 & @xmath920.547 & 1163.6587 & @xmath920.416 + 477.6368 & @xmath920.571 & 530.5581 & @xmath920.465 & 840.5587 & @xmath920.315 & 1166.6909 & @xmath920.577 + 477.6558 & @xmath920.583 & 530.5850 & @xmath920.491 & 842.6694 & @xmath920.480 & 1170.6574 & @xmath920.472 + 477.6758 & @xmath920.567 & 546.5476 & @xmath920.545 & 844.5297 & @xmath920.603 & 1173.6584 & @xmath920.527 + 483.4833 & @xmath920.474 & 548.5441 & @xmath920.180 & 845.5922 & @xmath920.636 & 1180.6629 & @xmath920.333 + 483.4903 & @xmath920.476 & 549.5287 & @xmath920.476 & 846.5297 & @xmath920.614 & 1190.6102 & @xmath920.142 + 483.6453 & @xmath920.481 & 552.5296 & @xmath920.541 & 853.6131 & @xmath920.545 & 1190.6541 & @xmath920.135 + 483.6603 & @xmath920.478 & 772.6852 & @xmath920.554 & 854.6380 & @xmath920.578 & 1198.9575 & @xmath920.466 + 489.6417 & @xmath920.528 & 773.6760 & @xmath920.510 & 855.5386 & @xmath920.569 & 1207.6672 & @xmath920.504 + 489.6827 & @xmath920.534 & 777.6661 & @xmath920.523 & 857.5851 & @xmath920.493 & 1208.5128 & @xmath920.469 + 491.5516 & @xmath920.132 & 777.7550 & @xmath920.533 & 859.5043 & @xmath920.299 & 1208.6774 & @xmath920.449 + 491.5616 & @xmath920.129 & 778.6511 & @xmath920.573 & 859.6609 & @xmath920.269 & 1208.6787 & @xmath920.466 + 491.5716 & @xmath920.145 & 784.6513 & @xmath920.229 & 860.6173 & @xmath920.307 & 1209.5551 & @xmath920.156 + 491.6105 & @xmath920.137 & 784.7507 & @xmath920.236 & 870.6636 & @xmath920.450 & 1209.6478 & @xmath920.150 + 497.6050 & @xmath920.532 & 785.6289 & @xmath920.447 & 878.5655 & @xmath920.255 & 1209.7063 & @xmath920.130 + 497.6250 & @xmath920.541 & 795.6635 & @xmath920.485 & 889.5226 & @xmath920.454 & 1209.7164 & @xmath920.145 + 497.6740 & @xmath920.551 & 797.6187 & @xmath920.565 & 896.5817 & @xmath920.432 & 1210.5362 & @xmath920.433 + 498.6169 & @xmath920.476 & 798.6614 & @xmath920.597 & 897.5411 & @xmath920.238 & 1215.6081 & @xmath920.590 + 498.6339 & @xmath920.483 & 799.6308 & @xmath920.579 & 899.5256 & @xmath920.475 & 1219.5845 & @xmath920.261 + 498.6569 & @xmath920.467 & 799.6353 & @xmath920.588 & 903.5375 & @xmath920.575 & 1220.6091 & @xmath920.416 + 504.5253 & @xmath920.611 & 800.6440 & @xmath920.519 & 1117.7217 & @xmath920.525 & 1221.6032 & @xmath920.496 + 504.5426 & @xmath920.605 & 802.5989 & @xmath920.367 & 1118.6984 & @xmath920.590 & 1225.5884 & @xmath920.521 + 504.5669 & @xmath920.615 & 802.7705 & @xmath920.332 & 1118.7112 & @xmath920.574 & 1229.5347 & @xmath920.466 + 505.5347 & @xmath920.643 & 803.7137 & @xmath920.261 & 1129.6608 & @xmath920.583 & 1232.5661 & @xmath920.615 + 505.5722 & @xmath920.649 & 803.7275 & @xmath920.278 & 1129.6707 & @xmath920.583 & 1268.5243 & @xmath920.530 + 507.4912 & @xmath920.578 & 809.5767 & @xmath920.604 & 1130.7195 & @xmath920.550 & 1275.5238 & @xmath920.227 + 507.5083 & @xmath920.573 & 809.7627 & @xmath920.622 & 1133.7098 & @xmath920.222 & 1276.5225 & @xmath920.319 + 507.6290 & @xmath920.607 & 813.7789 & @xmath920.410 & 1136.7032 & @xmath920.538 & 1277.5205 & @xmath920.456 + 508.5199 & @xmath920.524 & & & & & & + ccrrrcrrrcrrrcccc 50678.6448 & 4456 & 52.43 & 7.42 & 165.2 & & 51.85 & 7.38 & 143.7 & & 104.14 & 10.47 & 299.2 & & @xmath920.01 & 0.10 & 0.900 + 50679.6073 & 2940 & 17.34 & 4.24 & 61.8 & & 16.22 & 4.15 & 43.2 & & 32.98 & 5.90 & 101.3 & & @xmath920.03 & 0.18 & 0.951 + 50680.6024 & 3621 & 20.34 & 4.57 & 64.3 & & 15.48 & 4.10 & 32.2 & & 35.65 & 6.13 & 91.4 & & @xmath920.14 & 0.17 & 0.004 + 50681.6007 & 2871 & 18.55 & 4.36 & 62.0 & & 34.06 & 5.95 & 108.1 & & 52.60 & 7.38 & 169.8 & & @xmath100.29 & 0.13 & 0.056 + 50682.5965 & 3969 & 48.01 & 7.03 & 176.5 & & 64.45 & 8.16 & 195.2 & & 112.64 & 10.78 & 370.0 & & @xmath100.15 & 0.09 & 0.109 + 50683.5920 & 3262 & 49.84 & 7.11 & 224.7 & & 60.94 & 7.89 & 227.6 & & 111.23 & 10.64 & 443.0 & & @xmath100.10 & 0.10 & 0.162 + 50684.5872 & 4190 & 64.31 & 8.10 & 254.5 & & 75.47 & 8.79 & 260.0 & & 139.26 & 11.94 & 508.8 & & @xmath100.08 & 0.09 & 0.214 + 50687.5392 & 2783 & 49.43 & 7.07 & 196.7 & & 47.69 & 6.97 & 161.8 & & 97.25 & 9.94 & 349.3 & & @xmath920.02 & 0.10 & 0.371 + 50688.5344 & 3066 & 88.18 & 9.43 & 434.4 & & 119.38 & 10.99 & 497.5 & & 207.54 & 14.49 & 928.3 & & @xmath100.15 & 0.07 & 0.423 + 50689.5924 & 3351 & 52.98 & 7.33 & 232.9 & & 66.57 & 8.24 & 242.0 & & 119.40 & 11.02 & 470.5 & & @xmath100.11 & 0.09 & 0.479 + 50690.7236 & 2993 & 102.41 & 10.15 & 490.3 & & 122.37 & 11.12 & 530.9 & & 224.66 & 15.06 & 1015.0 & & @xmath100.09 & 0.07 & 0.539 + 50691.5826 & 3615 & 40.56 & 6.41 & 173.4 & & 72.82 & 8.64 & 237.2 & & 113.82 & 10.81 & 409.3 & & @xmath100.28 & 0.09 & 0.585 + 50692.6469 & 3087 & 55.64 & 7.53 & 220.6 & & 62.93 & 8.03 & 212.1 & & 118.36 & 11.01 & 427.8 & & @xmath100.06 & 0.09 & 0.641 + 50693.5719 & 3521 & 42.45 & 6.56 & 169.2 & & 72.87 & 8.62 & 241.5 & & 114.89 & 10.82 & 407.0 & & @xmath100.26 & 0.09 & 0.690 + 50694.6045 & 4054 & 55.74 & 7.54 & 164.9 & & 54.91 & 7.53 & 175.6 & & 110.07 & 10.65 & 333.2 & & @xmath920.01 & 0.10 & 0.745 + 50695.5993 & 4000 & 53.56 & 7.39 & 187.0 & & 68.96 & 8.47 & 199.5 & & 123.10 & 11.26 & 379.6 & & @xmath100.13 & 0.09 & 0.797 + 50696.5934 & 3850 & 56.17 & 7.58 & 187.5 & & 63.95 & 8.11 & 187.0 & & 119.43 & 11.07 & 373.3 & & @xmath100.06 & 0.09 & 0.850 + cccc + @xmath34 ( @xmath93 ) & 79.55 @xmath76 0.18 & 79.46 @xmath76 0.15 & 79.470 @xmath76 0.074 + @xmath94 & 0.2055 & 0.2055 & 0.2055 + @xmath33 & 6.05 @xmath76 0.25 & 5.89 @xmath76 0.16 & 6.03 @xmath76 0.10 + @xmath95 & 2.246 @xmath76 0.011 & 2.246 @xmath76 0.011 & 2.246 @xmath76 0.011 + @xmath50 ( primary , secondary ) & 0.1718 , 0.3439 & 0.1767 , 0.3439 & 0.1724 , 0.3439 + @xmath51 ( primary , secondary ) & 0.1710 , 0.2348 & 0.1758 , 0.2348 & 0.1716 , 0.2348 + @xmath52 ( primary , secondary ) & 0.1715 , 0.2442 & 0.1764 , 0.2442 & 0.1721 , 0.2442 + @xmath53 ( primary , secondary ) & 0.1718 , 0.2767 & 0.1768 , 0.2767 & 0.1724 , 0.2767 + @xmath54 ( primary , secondary ) & 0.1714 , 0.2513 & 0.1763 , 0.2513 & 0.1720 , 0.2513 + @xmath37 & @xmath920.00006 @xmath76 0.00058 & @xmath100.00097 @xmath76 0.00064 & @xmath100.00040 @xmath76 0.00040 + @xmath96 ( k ) & 5000 & 5000 & 5000 + @xmath36 ( k ) & 4612 @xmath76 47 & 4677 @xmath76 40 & 4638 @xmath76 11 + @xmath97 & 0.3956 @xmath76 0.0091 & & 0.3911 @xmath76 0.0004 + @xmath98 & & 0.3810 @xmath76 0.0077 & 0.3764 @xmath76 0.0004 + albedo & 0.5 , 0.5 & 0.5 , 0.5 & 0.5 , 0.5 + gravity brightening & 0.365 , 0.365 & 0.365 , 0.365 & 0.365 , 0.365 + + @xmath43 ( @xmath93 ) & 257.77 @xmath76 0.09 & 254.44 @xmath76 0.11 & 254.39 @xmath76 0.04 + @xmath44 ( @xmath93 ) & 45 & 45 & 45 + @xmath45 ( @xmath93 ) & 36.33 @xmath76 1.35 & 37.29 @xmath76 1.63 & 37.05 @xmath76 0.28 + _ tf _ & 0.84 & 0.84 & 0.84 + + @xmath99 & 0.589 & & 0.589 + @xmath100 & & 0.481 & 0.481 + @xmath101 & 0.527 & 0.527 & 0.527 + @xmath102 & 0.639 & & 0.635 + @xmath103 & & 0.511 & 0.515 + @xmath104 & 0.528 & 0.527 & 0.528 + + @xmath105 ( mag ) & 0.0126 & & 0.0127 + @xmath106 ( mag ) & & 0.0131 & 0.0133 + @xmath107 & 92 & 95 & 187 + cccccc + @xmath34 ( @xmath93 ) & 79.24 @xmath76 0.21 & 79.41 @xmath76 0.25 & 79.38 @xmath76 0.36 & 79.55 @xmath76 0.30 & 79.470 @xmath76 0.074 + @xmath94 & 0.2055 & 0.2055 & 0.2055 & 0.2055 & 0.2055 + @xmath33 & 6.39 @xmath76 0.53 & 5.83 @xmath76 0.34 & 5.83 @xmath76 0.35 & 6.07 @xmath76 0.32 & 6.03 @xmath76 0.10 + @xmath95 & 2.246 @xmath76 0.011 & 2.246 @xmath76 0.011 & 2.246 @xmath76 0.011 & 2.246 @xmath76 0.011 & 2.246 @xmath76 0.011 + @xmath50 ( primary , secondary ) & 0.1623 , 0.3439 & 0.1787 , 0.3439 & 0.1787 , 0.3439 & 0.1712 , 0.3439 & 0.1724 , 0.3439 + @xmath51 ( primary , secondary ) & 0.1616 , 0.2348 & 0.1777 , 0.2348 & 0.1777 , 0.2348 & 0.1704 , 0.2348 & 0.1716 , 0.2348 + @xmath52 ( primary , secondary ) & 0.1620 , 0.2442 & 0.1783 , 0.2442 & 0.1783 , 0.2442 & 0.1710 , 0.2442 & 0.1721 , 0.2442 + @xmath53 ( primary , secondary ) & 0.1622 , 0.2767 & 0.1786 , 0.2767 & 0.1786 , 0.2767 & 0.1712 , 0.2767 & 0.1724 , 0.2767 + @xmath54 ( primary , secondary ) & 0.1620 , 0.2513 & 0.1782 , 0.2513 & 0.1782 , 0.2513 & 0.1709 , 0.2513 & 0.1720 , 0.2513 + @xmath37 & @xmath920.0025 @xmath76 0.0011 & @xmath920.00331 @xmath76 0.00096 & @xmath920.00329 @xmath76 0.00088 & @xmath920.00201 @xmath76 0.00077 & @xmath920.00272 @xmath76 0.00044 + @xmath96 ( k ) & 5000 & 5000 & 5000 & 5000 & 5000 + @xmath36 ( k ) & 4565 @xmath76 67 & 4671 @xmath76 52 & 4715 @xmath76 56 & 4701 @xmath76 59 & 4638 @xmath76 11 + @xmath108 & 0.443 @xmath76 0.013 & & & & 0.4353 @xmath76 0.0007 + @xmath109 & & 0.418 @xmath76 0.016 & & & 0.4102 @xmath76 0.0006 + @xmath110 & & & 0.390 @xmath76 0.020 & & 0.3911 @xmath76 0.0006 + @xmath111 & & & & 0.361 @xmath76 0.015 & 0.3764 @xmath76 0.0007 + albedo & 0.5 , 0.5 & 0.5 , 0.5 & 0.5 , 0.5 & 0.5 , 0.5 & 0.5 , 0.5 + gravity brightening & 0.365 , 0.365 & 0.365 , 0.365 & 0.365 , 0.365 & 0.365 , 0.365 & 0.365 , 0.365 + + @xmath43 ( @xmath93 ) & 110.29 @xmath76 0.17 & 113.14 @xmath76 0.14 & 117.81 @xmath76 0.14 & 112.06 @xmath76 0.14 & 112.73 @xmath76 0.07 + @xmath44 ( @xmath93 ) & 45 & 45 & 45 & 45 & 45 + @xmath45 ( @xmath93 ) & 35.0 @xmath76 1.2 & 33.0 @xmath76 2.0 & 33.4 @xmath76 2.2 & 34.7 @xmath76 1.6 & 33.83 @xmath76 0.65 + _ tf _ & 0.87 & 0.87 & 0.87 & 0.87 & 0.87 + + @xmath112 & 0.867 & & & & 0.867 + @xmath113 & & 0.714 & & & 0.714 + @xmath99 & & & 0.589 & & 0.589 + @xmath100 & & & & 0.481 & 0.481 + @xmath101 & 0.527 & 0.527 & 0.527 & 0.527 & 0.527 + @xmath114 & 0.944 & & & & 0.931 + @xmath115 & & 0.768 & & & 0.774 + @xmath102 & & & 0.624 & & 0.635 + @xmath103 & & & & 0.508 & 0.515 + @xmath104 & 0.529 & 0.527 & 0.527 & 0.527 & 0.528 + + @xmath116 ( mag ) & 0.0169 & & & & 0.0180 + @xmath117 ( mag ) & & 0.0167 & & & 0.0171 + @xmath105 ( mag ) & & & 0.0134 & & 0.0145 + @xmath106 ( mag ) & & & & 0.0104 & 0.0112 + @xmath107 & 37 & 42 & 41 & 42 & 162 + cc@cc@cc@cc@cc 49437.5420 & @xmath9233.96 & @xmath1079.12 & @xmath920.04 & @xmath101.16 & @xmath9234.00 & @xmath1080.28 & @xmath102.91 & @xmath102.70 & 0.229 + 49614.0033 & @xmath9210.38 & @xmath9255.19 & 0.00 & @xmath101.44 & @xmath9210.38 & @xmath9253.75 & @xmath920.95 & @xmath102.34 & 0.567 + 49644.9114 & @xmath9237.77 & @xmath1072.73 & @xmath920.05 & @xmath101.06 & @xmath9237.82 & @xmath1073.79 & @xmath921.64 & @xmath920.24 & 0.202 + 49648.9088 & @xmath9227.64 & @xmath1035.10 & @xmath920.00 & @xmath921.50 & @xmath9227.64 & @xmath1033.60 & @xmath920.04 & @xmath101.27 & 0.413 + 49653.8368 & @xmath101.52 & @xmath92101.99 & @xmath100.02 & @xmath920.03 & @xmath101.54 & @xmath92102.02 & @xmath101.44 & @xmath100.43 & 0.674 + 49680.9384 & @xmath9229.45 & @xmath1043.32 & @xmath920.05 & @xmath100.30 & @xmath9229.50 & @xmath1043.62 & @xmath100.24 & @xmath100.89 & 0.108 + 49681.7665 & @xmath9233.04 & @xmath1060.85 & @xmath920.05 & @xmath100.62 & @xmath9233.09 & @xmath1061.47 & @xmath100.36 & @xmath100.73 & 0.152 + 49696.7771 & @xmath929.42 & @xmath9246.10 & @xmath920.76 & @xmath920.11 & @xmath9210.18 & @xmath9246.21 & @xmath100.66 & @xmath103.04 & 0.946 + 49700.7398 & @xmath9233.59 & @xmath1059.35 & @xmath920.05 & @xmath100.67 & @xmath9233.64 & @xmath1060.02 & @xmath100.08 & @xmath922.07 & 0.156 + 49704.7351 & @xmath9234.58 & @xmath1053.31 & @xmath920.01 & @xmath920.94 & @xmath9234.59 & @xmath1052.37 & @xmath922.61 & @xmath921.25 & 0.367 + 49708.6896 & @xmath928.72 & @xmath9263.23 & 0.00 & @xmath101.49 & @xmath928.72 & @xmath9261.74 & @xmath920.42 & @xmath920.12 & 0.576 + 49732.7773 & @xmath921.03 & @xmath9293.39 & @xmath100.05 & @xmath920.59 & @xmath920.98 & @xmath9293.98 & @xmath100.54 & @xmath100.58 & 0.851 + 49734.8029 & @xmath929.50 & @xmath9240.34 & @xmath922.56 & @xmath920.08 & @xmath9212.06 & @xmath9240.42 & @xmath100.20 & @xmath101.88 & 0.958 + 49766.6378 & @xmath921.68 & @xmath9292.32 & @xmath100.01 & @xmath100.71 & @xmath921.67 & @xmath9291.61 & @xmath100.34 & @xmath100.59 & 0.643 + 49767.6563 & @xmath100.53 & @xmath92109.05 & @xmath100.02 & @xmath920.49 & @xmath100.55 & @xmath92109.54 & @xmath920.65 & @xmath921.71 & 0.697 + 49815.6120 & @xmath9237.30 & @xmath1078.13 & @xmath920.04 & @xmath101.15 & @xmath9237.34 & @xmath1079.28 & @xmath920.36 & @xmath101.35 & 0.234 + 49964.0085 & @xmath9227.93 & @xmath1030.63 & @xmath920.04 & @xmath100.20 & @xmath9227.97 & @xmath1030.83 & @xmath920.43 & @xmath921.17 & 0.086 + 49969.9892 & @xmath9229.01 & @xmath1039.59 & 0.00 & @xmath921.44 & @xmath9229.01 & @xmath1038.15 & @xmath920.32 & @xmath100.53 & 0.403 + 49971.0119 & @xmath9224.69 & @xmath108.75 & 0.00 & @xmath922.71 & @xmath9224.69 & @xmath106.04 & @xmath922.02 & @xmath922.27 & 0.457 + 50002.9115 & @xmath9235.21 & @xmath1057.25 & @xmath920.05 & @xmath100.55 & @xmath9235.26 & @xmath1057.80 & @xmath922.34 & @xmath920.40 & 0.145 + 50006.9548 & @xmath9232.05 & @xmath1058.81 & @xmath920.01 & @xmath920.75 & @xmath9232.06 & @xmath1058.06 & @xmath100.61 & @xmath101.11 & 0.359 + 50031.9081 & @xmath920.28 & @xmath92104.85 & @xmath100.02 & @xmath920.12 & @xmath920.26 & @xmath92104.97 & @xmath920.64 & @xmath921.16 & 0.679 + 50033.8891 & @xmath102.26 & @xmath92110.16 & @xmath100.04 & @xmath921.13 & @xmath102.30 & @xmath92111.29 & @xmath100.44 & @xmath920.26 & 0.784 + 50051.7831 & @xmath102.68 & @xmath92110.90 & @xmath100.03 & @xmath920.97 & @xmath102.71 & @xmath92111.87 & @xmath100.55 & @xmath100.61 & 0.731 + 50060.8041 & @xmath9235.47 & @xmath1074.57 & @xmath920.05 & @xmath101.10 & @xmath9235.52 & @xmath1075.67 & @xmath100.88 & @xmath100.57 & 0.208 + 50061.7975 & @xmath9237.73 & @xmath1077.49 & @xmath920.03 & @xmath101.04 & @xmath9237.76 & @xmath1078.53 & @xmath920.73 & @xmath100.33 & 0.261 + 50062.7964 & @xmath9235.10 & @xmath1071.98 & @xmath920.02 & @xmath100.28 & @xmath9235.12 & @xmath1072.26 & @xmath100.41 & @xmath101.35 & 0.313 + 50064.6757 & @xmath9227.02 & @xmath1035.36 & @xmath920.00 & @xmath921.49 & @xmath9227.02 & @xmath1033.87 & @xmath100.62 & @xmath101.38 & 0.413 + 50083.6864 & @xmath9226.41 & @xmath1033.18 & @xmath920.00 & @xmath921.50 & @xmath9226.41 & @xmath1031.68 & @xmath100.59 & @xmath102.27 & 0.419 + 50085.7517 & @xmath9213.29 & @xmath9235.19 & 0.00 & @xmath103.64 & @xmath9213.29 & @xmath9231.55 & @xmath100.64 & @xmath102.64 & 0.528 + 50088.7862 & @xmath100.62 & @xmath92105.51 & @xmath100.02 & @xmath920.32 & @xmath100.64 & @xmath92105.83 & @xmath920.22 & @xmath100.32 & 0.689 + 50090.7706 & @xmath101.38 & @xmath92107.20 & @xmath100.05 & @xmath921.09 & @xmath101.43 & @xmath92108.29 & @xmath920.14 & @xmath101.32 & 0.794 + 50092.7725 & @xmath925.88 & @xmath9273.48 & @xmath100.05 & @xmath920.26 & @xmath925.83 & @xmath9273.74 & @xmath920.06 & @xmath100.17 & 0.900 + 50117.6706 & @xmath9236.09 & @xmath1075.44 & @xmath920.04 & @xmath101.13 & @xmath9236.13 & @xmath1076.57 & @xmath100.53 & @xmath100.21 & 0.217 + 50141.6862 & @xmath9219.15 & @xmath9212.00 & 0.00 & @xmath922.62 & @xmath9219.15 & @xmath9214.62 & @xmath920.25 & @xmath924.60 & 0.488 + 50143.6293 & @xmath927.21 & @xmath9270.78 & @xmath920.00 & @xmath101.48 & @xmath927.21 & @xmath9269.30 & @xmath920.43 & @xmath920.29 & 0.591 + 50144.6373 & @xmath923.42 & @xmath9293.00 & @xmath100.01 & @xmath100.70 & @xmath923.41 & @xmath9292.30 & @xmath921.50 & @xmath100.38 & 0.644 + 50145.6448 & @xmath101.11 & @xmath92107.04 & @xmath100.02 & @xmath920.48 & @xmath101.13 & @xmath92107.52 & @xmath920.10 & @xmath100.45 & 0.697 + 50148.6641 & @xmath923.71 & @xmath9291.19 & @xmath100.05 & @xmath920.54 & @xmath923.66 & @xmath9291.73 & @xmath921.68 & @xmath100.62 & 0.857 + 50149.5956 & @xmath926.31 & @xmath9269.74 & @xmath100.04 & @xmath920.23 & @xmath926.27 & @xmath9269.97 & @xmath100.18 & @xmath100.63 & 0.906 + 50153.6415 & @xmath9231.32 & @xmath1045.86 & @xmath920.05 & @xmath100.36 & @xmath9231.37 & @xmath1046.22 & @xmath920.47 & @xmath922.13 & 0.120 + lc @xmath118 ( days ) & 18.8988 @xmath76 0.0011 + @xmath119 ( ) & @xmath120 @xmath76 0.14 + @xmath121 ( ) & 19.69 @xmath76 0.22 + @xmath122 ( ) & 95.80 @xmath76 0.33 + @xmath123 & 0 ( fixed ) + @xmath124 ( hjd ) & 2,449,919.857 @xmath76 0.011 + + @xmath125 ( 10@xmath126 km ) & 5.117 @xmath76 0.059 + @xmath127 ( 10@xmath126 km ) & 24.897 @xmath76 0.088 + @xmath128 ( r@xmath4 ) & 43.12 @xmath76 0.15 + @xmath129 ( m@xmath4 ) & 2.502 @xmath76 0.026 + @xmath130 ( m@xmath4 ) & 0.5143 @xmath76 0.0085 + @xmath94 & 0.2055 @xmath76 0.0025 + + @xmath107 & 41 + @xmath131 ( ) & 1.05 + @xmath132 ( ) & 1.57 + lc@ c @xmath118 ( days ) & + @xmath133 ( r@xmath4 ) & + @xmath134 ( m@xmath4 ) & 2.633 @xmath76 0.028 & 0.5412 @xmath76 0.0093 + @xmath20 ( r@xmath4 ) & 7.55 @xmath76 0.20 & 11.02 @xmath76 0.21 + @xmath135 ( k ) & 5000 @xmath76 100 & 4640 @xmath76 150 + @xmath136 & 3.103 @xmath76 0.023 & 2.087 @xmath76 0.018 + @xmath137 ( @xmath138 g @xmath139 ) & 8.63 @xmath76 0.68 & 0.569 @xmath76 0.034 + @xmath140 ( km s@xmath141 ) & 19.9 @xmath76 0.5 & 29.0 @xmath76 0.6 + @xmath59 ( km s@xmath141 ) & 23 @xmath76 1 & 31 @xmath76 2 + @xmath142 ( l@xmath143 ) & 1.504 @xmath76 0.041 & 1.704 @xmath76 0.059 + @xmath144 ( ergs s@xmath141 ) & + @xmath145 & + @xmath146 ( mag ) & 0.97 @xmath76 0.10 & 0.47 @xmath76 0.15 + @xmath147 ( mag ) & 1.26 @xmath76 0.11 & 0.98 @xmath76 0.18 + @xmath148 ( mag ) & + distance ( pc ) & +

an _ orbitope _ is the convex hull of an orbit of a compact algebraic group @xmath2 acting linearly on a real vector space . the orbit has the structure of a real algebraic variety , and the orbitope is a convex , semi - algebraic set . thus , the study of algebraic orbitopes lies at the heart of _ convex algebraic geometry _ the fusion of convex geometry and ( real ) algebraic geometry . orbitopes have appeared in many contexts in mathematics and its applications . orbitopes of finite groups are highly symmetric convex polytopes which include the platonic solids , permutahedra , birkhoff polytopes , and other favorites from ziegler s text book @xcite , as well as the coxeter orbihedra studied by mccarthy , ogilvie , zobin , and zobin @xcite . farran and robertson s regular convex bodies @xcite are orbitopal generalzations of regular polytopes , which were classified by madden and robertson @xcite . orbitopes for compact lie groups , such as @xmath0 , have appeared in investigations ranging from protein structure prediction @xcite and quantum information @xcite to calibrated geometries @xcite . barvinok and blekherman studied the volumes of convex bodies dual to certain @xmath0-orbitopes , and they concluded that there are many more non - negative polynomials than sums of squares @xcite . this paper initiates the study of orbitopes as geometric objects in their own right . the questions we ask about orbitopes originate from three different perspectives : _ convexity _ , _ algebraic geometry _ , and _ optimization_. in convexity , one would seek to characterize all faces of an orbitope . in algebraic geometry , one would examine the zariski closure of its boundary and identify the components and singularities of that hypersurface . in optimization , one would ask whether the orbitope is a spectrahedron or the projection of a spectrahedron . spectrahedra are to semidefinite programming what polyhedra are to linear programming . more precisely , a _ spectrahedron _ is the intersection of the cone of positive semidefinite matrices with an affine space . it can be represented as the set of points @xmath3 such that @xmath4 where @xmath5 are symmetric matrices and @xmath6 denotes positive semidefiniteness . from a spectrahedral description many geometric properties , both convex and algebraic , are within reach . furthermore , if an orbitope admits a representation ( [ eq : lmi ] ) then it is easy to maximize or minimize a linear function over that orbitope . here is a simple illustration . [ ex : hankelsmall ] consider the action of the group @xmath7 on the space @xmath8 of binary quartics and take the convex hull of the orbit of @xmath9 . the four - dimensional convex body @xmath10 is a _ carathodory orbitope_. this orbitope is a spectrahedron : it coincides with the set of all binary quartics @xmath11 such that @xmath12 this representation ( [ 3by3hankel ] ) will be derived in section 5 , where we will also see that it is equivalent to classical results from the theory of positive polynomials @xcite . the hankel matrix shows that the boundary of @xmath10 is an irreducible cubic hypersurface in @xmath13 , defined by the vanishing of the hankel determinant . it also reveals that this four - dimensional carathodory orbitope is @xmath14-neighborly : the extreme points are the rank one matrices , and any two of them are connected by an edge . the typical intersection of @xmath10 with a three - dimensional affine plane looks like an inflated tetrahedron . this three - dimensional convex body is bounded by _ cayley s cubic surface _ , shown in figure [ f : fat_tetrahedron ] . alternative pictures of this convex body can be found in ( * ? ? ? * fig . 3 ) and ( * ? ? ? * fig . 4 ) . the four vertices of the tetrahedron lie on the curve @xmath15 , and its six edges are inclusion - maximal faces of @xmath10 . @xmath16{figures / free - spec}\ ] ] -0.3 cm [ f : fat_tetrahedron ] this article is organized as follows . we begin by deriving the basic definitions and a few general results about orbitopes , and we formulate * ten key questions * which will guide our subsequent investigations . these are organized along the themes of convex geometry ( subsection 2.1 ) , algebraic geometry ( subsection 2.2 ) and optimization ( subsection 2.3 ) . these questions are difficult . they are meant to motivate the reader and to guide our investigations . we are not yet able to offer a complete solution to any of these ten problems . section 3 is concerned with the action of @xmath1 and @xmath0 by conjugation on @xmath17-matrices . these decompose into the actions by conjugation on symmetric and skew - symmetric matrices . the resulting _ schur - horn orbitopes _ are shown be spectrahedra , their algebraic boundary is computed , and their face lattices are derived from certain polytopes known as permutahedra . the spectrahedral representations of the schur - horn orbitopes , stated in theorems [ thm : symspec ] and [ thm : skewsymspec ] , rely on certain schur functors also known as the _ additive compound matrices_. given a compact real algebraic group of @xmath17-matrices , we can form its convex hull in @xmath18 . the resulting convex bodies are the _ tautological orbitopes_. in subsection 4.2 we study the tautological orbitope and its dual coorbitope for the group @xmath1 . both are spectrahedra , characterized by constraints on singular values , and they are unit balls for the operator and nuclear matrix norm considered in @xcite . subsections 4.1 and 4.4 are devoted to the tautological orbitope for @xmath0 . a characterization of its faces is given in theorem [ thm : sontauto ] . the orbitopes of the group @xmath19 are the convex hulls of trigonometric curves , a classical topic initiated by carathodory in @xcite , further developed in @xcite , and our primary focus in section 5 . these @xmath19-orbitopes can be represented as projections of spectrahedra in two distinct ways : in terms of hermitian toeplitz matrices or in terms of hankel matrices . a natural generalization of the rational normal curves in section 5 are the veronese varieties . their convex hulls , the _ veronese orbitopes _ , are dual to the cones of polynomials that are non - negative on @xmath20 , as seen in @xcite . in section 6 we undertake a detailed study of the @xmath21-dimensional veronese orbitope and its coorbitope which arise from ternary quartics . in section 7 we mesh our investigations with a line of research in differential geometry . the _ grassmann orbitope _ is the convex hull of the oriented grassmann variety in its plcker embedding in the unit sphere in @xmath22 . this vector space is the @xmath23-th exterior power of @xmath24 and it is isomorphic to @xmath25 . the facial structure of grassmann orbitopes has been studied in the theory of calibrated manifolds @xcite . we take a fresh look at these orbitopes from the point of view of convex algebraic geometry . theorem [ thm : g2n ] furnishes a spectrahedral representation and the algebraic boundary in the special case @xmath26 , while theorem [ thm : g36bad ] shows that the grassmann orbitopes fail to be spectrahedra in general . let @xmath2 be a real , compact , linear algebraic group , that is , a compact subgroup of @xmath27 for some @xmath28 given as a subvariety . prototypic examples are the _ special orthogonal group _ @xmath29 and the _ unitary group _ @xmath30 . a real representation of @xmath2 is a group homomorphism @xmath31 for some finite dimensional real vector space @xmath32 . we will write @xmath33 for @xmath34 and @xmath35 . the representation is _ rational _ if @xmath36 is a rational map of algebraic varieties . by choosing an inner product @xmath37 on @xmath32 , we may define a @xmath2-invariant inner product as @xmath38 here @xmath39 denotes the _ haar measure _ , which is the unique @xmath2-invariant probability measure on @xmath2 . choosing coordinates on @xmath32 so that @xmath40 becomes the standard inner product on @xmath41 , we can identify the matrix group @xmath2 with a subgroup of the _ orthogonal group _ @xmath42 . the _ orbit _ of a vector @xmath35 under the compact group @xmath2 is the set @xmath43 . this is a bounded subset of @xmath32 . the orbit @xmath44 is a smooth , compact real algebraic variety of dimension @xmath45 here @xmath46 is the _ stabilizer _ of the vector @xmath47 . in particular , the orbit is isomorphic as a @xmath2-variety to the compact homogeneous space @xmath48 . the _ orbitope _ of @xmath2 with respect to the vector @xmath35 is the semialgebraic convex body @xmath49 we tacitly assume that the group @xmath2 and its representation @xmath36 are clear from the context and sometimes we write @xmath50 or @xmath51 for @xmath52 . the dimension of an orbitope @xmath53 is the dimension of the affine hull of the orbit @xmath54 . for small @xmath55 , the connected subgroups of the orthogonal group @xmath1 are known . this leads to the following census of low - dimensional orbitopes of connected groups . @xmath56 we identify @xmath2 with a subgroup of @xmath0 . we assume @xmath57 is an @xmath55-dimensional orbitope in @xmath24 . this implies that @xmath2 fixes no non - zero vector . here is our census : _ @xmath58 : _ there are no one - dimensional orbitopes because @xmath59 is a point . it is known that every proper connected subgroup of @xmath19 or @xmath60 fixes a non - zero vector , so for @xmath61 , the subgroup @xmath2 must be equal to @xmath0 . this establishes the next two cases : _ @xmath62 : _ the only orbitopes in @xmath63 are the discs @xmath64 . _ @xmath65 : _ the only orbitopes in @xmath66 are the balls @xmath67 . the four - dimensional case is where things begin to get interesting : _ @xmath68 : _ the group @xmath69 has connected subgroups @xmath2 of dimension @xmath70 , @xmath14 , @xmath71 and @xmath72 . * if @xmath73 then @xmath74 and the orbitopes are balls in @xmath13 . * if @xmath75 then @xmath76 , acting as the unit quaternions on all quaternions , @xmath77 , by either left or right multiplication . here the orbitopes are also balls in @xmath13 . * if @xmath78 then @xmath79 . these tori act on @xmath13 through an orthogonal direct sum decomposition @xmath80 and their orbitopes are products of two discs . * if @xmath81 then @xmath82 and we obtain four - dimensional orbitopes that are isomorphic , for some positive integers @xmath83 , to the carathodory orbitopes @xmath84\bigr\}\,.\ ] ] these orbitopes were introduced one century ago by carathodory @xcite . their study was picked up in the 1980s by smilansky @xcite and recently by barvinok and novik @xcite . we note that @xmath85 is affinely isomorphic to the hankel orbitope in example [ ex : hankelsmall ] . to compute the dimensions of orbitopes in general we shall need a pinch of representation theory @xcite . a representation @xmath32 of the group @xmath2 is _ irreducible _ if its only subrepresentations are @xmath86 and @xmath32 . if @xmath32 and @xmath87 are irreducible representations , then the space @xmath88 of equivariant linear maps between them is zero unless @xmath89 . schur s lemma states that @xmath90 is a division algebra over @xmath91 , that is , either @xmath91 , @xmath92 , or @xmath93 . if @xmath94 is a complete list of distinct irreducible representations of @xmath2 , and @xmath32 is any representation of @xmath2 , then we have a canonical decomposition into isotypical representations : @xmath95 this is an isomorphism of @xmath2-modules . the map ( [ eq : isotypical ] ) on each summand is @xmath96 . the image of the @xmath97th summand in @xmath32 is called the _ @xmath98-isotypical component of @xmath32 _ , and when it is non - zero , we say that the irreducible representation @xmath98 _ appears _ in the @xmath2-module @xmath32 . we say that @xmath32 is _ multiplicity - free _ if each irreducible representation @xmath98 appears in @xmath32 at most once , so that @xmath99 has rank 1 or 0 over @xmath100 . suppose that @xmath32 contains the trivial representation and write @xmath101 , where @xmath102 is the trivial isotypical component of @xmath32 and @xmath103 does not contain the trivial representation . any vector @xmath104 can be written as @xmath105 , where @xmath106 and @xmath107 . thus we lose no geometric information in assuming that @xmath32 does not contain the trivial representation . this property ensures that the linear span of an orbit coincides with its affine span . hence the affine span of an orbitope decomposes along its isotypical components : @xmath109 to determine the dimension of @xmath110 for @xmath47 in a single isotypical component @xmath32 we proceed as follows . let @xmath111 with @xmath87 irreducible . then @xmath112 and the affine span of @xmath54 is isomorphic to @xmath113 , where @xmath114 is the rank of the @xmath115-module spanned by @xmath116 . hence , the dimension of @xmath117 over @xmath91 equals @xmath118 . in particular , @xmath119 . if @xmath32 is multiplicity free and @xmath47 has a nonzero projection into each isotypical component of @xmath32 , then @xmath120 . we see this in the carathodory orbitopes for the group @xmath19 of @xmath121 rotation matrices . its nontrivial representations are @xmath122 , where a rotation matric acts through its @xmath123th power , for @xmath124 . let @xmath125 be the orbitope of @xmath19 with respect to a general vector @xmath126 . if @xmath127 , then @xmath32 has two isotypical components and @xmath125 has dimension four . if @xmath128 , then @xmath129 consists of a single isotypical component and @xmath130 is two - dimensional , as @xmath131 , the span of @xmath132 is complex one - dimensional . orbitopes are convex bodies , and it is natural to begin their study from the perspective of classical convexity . a point @xmath133 in a convex body @xmath134 is an _ extreme point _ if @xmath135 . thus , the set @xmath136 of extreme points of @xmath137 is the minimal subset satisfying @xmath138 . an extrinsic description of @xmath137 is given by its _ support function _ @xmath139 in terms of the support function , the convex body @xmath137 is the set of points @xmath140 such that @xmath141 for every @xmath142 . each linear functional @xmath142 defines an _ exposed face _ of @xmath137 : @xmath143 an exposed face @xmath144 is itself a convex body of dimension @xmath145 . an exposed face of dimension @xmath146 is called an _ exposed point _ of @xmath137 . it follows that every exposed point is extreme , but the inclusion is typically strict . however , for orbitopes , these two notions coincide . every point in the orbit @xmath147 is exposed in its convex hull . in particular , every extreme point of the orbitope @xmath148 is an exposed point . since @xmath2 acts orthogonally on @xmath32 , the orbit @xmath147 lies entirely in the sphere in @xmath32 that is centered at @xmath146 and contains the point @xmath47 . as every point of the sphere is exposed , the entire orbit consists of exposed points and hence extreme points . a closed subset @xmath149 is a _ face _ if @xmath150 contains the two endpoints to any open segment in @xmath137 it intersects . this includes @xmath151 and @xmath137 itself . an inclusion - maximal proper face of @xmath137 is called a _ facet_. by separation , every face is contained in an exposed face and thus facets are automatically exposed . in general , every exposed face is a face but not conversely . when does an orbitope have only exposed faces ? the exposed faces of a convex body form a partially ordered set with respect to inclusion , called the _ face lattice_. the face lattice is atomic but in general not coatomic as was pointed out to us by stephan weis . a necessary condition is that the polar body ( see below ) has all faces exposed ( cf . @xcite ) . also , it is generally not graded because `` being a face of '' is not a transitive relation . for example , the four - dimensional barvinok - novik orbitope in section [ s : caratheodory ] has triangular exposed faces for which the three vertices are exposed but the edges are not . similar behavior is seen in the convex body on the right of figure [ f : trig_curves ] , which has two triangular exposed facets whose edges and two of three vertices are not exposed . [ question2 ] describe the face lattices of orbitopes . for an orbitope @xmath152 , the faces come in @xmath2-orbits and these @xmath2-orbits come in algebraic families . in particular , the zero - dimensional faces come in a family parametrized by @xmath2 . the description of these families is the point of question [ question2 ] . for instance , the orbitope in example [ ex : hankelsmall ] is a four - dimensional , two - neighborly convex body . its exposed points are parametrized by the circle @xmath153 and the edges come in a two - dimensional family . the _ polar body _ @xmath154 is called the _ coorbitope _ of @xmath2 with respect to @xmath104 . our assumption that @xmath32 does not contain the trivial representation ensures that @xmath146 is the centroid of @xmath51 , and this implies @xmath155 . we shall also make use of the cone over the coorbitope @xmath156 . this is the _ coorbitope cone _ @xmath157 for a convex body @xmath137 the assignment @xmath158 defines an _ ( asymmetric ) norm _ on @xmath32 with unit ball @xmath137 . if @xmath137 is centrally - symmetric with respect to the origin , then @xmath159 is an actual norm . in that case the polar body @xmath160 is also centrally - symmetric and @xmath161 is the _ dual norm_. norms and support functions are related by @xmath162 in particular , if @xmath137 is an orbitope , then @xmath159 and @xmath161 are @xmath2-equivariant norms . every point @xmath133 in a convex body @xmath137 is in the convex hull of finitely many extreme points . we denote by @xmath163 the least cardinality of a set @xmath136 of extreme points with @xmath164 . we call @xmath165 the _ carathodory number _ of @xmath137 . carathodory s theorem ( see e.g. @xcite ) asserts that @xmath166 is bounded from above by @xmath167 . fenchel showed that @xmath168 when the set of extreme points of @xmath137 is connected @xcite . note that the carathodory number of an orbitope @xmath169 in general depends on @xmath47 ( cf . * theorem 4.9 ) ) whereas , for multiplicity - free representations , the dimension of @xmath50 does not . what are the carathodory numbers of orbitopes ? we refer to the recent work of barvinok and blekherman @xcite for more information about the convex geometry of orbitopes and coorbitopes , especially with regard to their volumes . here we look at orbitopes as objects in real algebraic geometry . fix a rational representation @xmath170 of a compact connected algebraic group @xmath2 . every orbit @xmath44 is an irreducible real algebraic variety in @xmath171 , and we may ask for its prime ideal . by the tarski - seidenberg theorem @xcite , the orbitope is a semi - algebraic set . which orbitopes are _ basic _ semi - algebraic sets , i.e. for which triples @xmath172 can @xmath10 be described by a finite conjunction of polynomial equations and inequalities ? the boundary @xmath173 of an orbitope @xmath51 in @xmath171 is a compact semi - algebraic set of codimension one in its affine span @xmath174 . the zariski closure of @xmath173 is denoted by @xmath175 . we call it the _ algebraic boundary _ of @xmath51 . if @xmath176 then the algebraic boundary @xmath175 is the zero set of a unique ( up to scaling ) reduced polynomial @xmath177 whose coefficients lie in the field of definition of @xmath172 . that field of definition will often be the rational numbers @xmath178 . since scalars in @xmath179 have an exact representation in computer algebra , but scalars in @xmath91 require numerical floating point approximations , we seek to use @xmath179 instead of @xmath91 wherever possible . [ ques : algbound ] how to calculate the algebraic boundary @xmath180 of an orbitope @xmath181 ? the irreducible factors of the polynomial @xmath177 that cuts out @xmath182 arise from various singularities in the boundary @xmath183 of the coorbitope @xmath184 . we believe that a complete answer to question [ ques : algbound ] will involve a whitney stratification of the real algebraic hypersurface @xmath185 . recall that a _ whitney stratification _ is a decomposition into locally closed submanifolds ( strata ) in which the singularity type of each stratum is locally constant along the stratum . the faces of a polytope form a whitney stratification of its boundary , which is dual to the stratification of the polar polytope . we expect a similar duality between the whitney stratification of the boundary of an orbitope and of the boundary of its coorbitope . how to compute and study the algebraic boundary @xmath186 of the coorbitope @xmath187 ? is every component of @xmath188 the dual variety to a stratum in a whitney stratification of @xmath185 ? recall that the _ dual variety _ @xmath189 of a subvariety @xmath190 in @xmath171 is the zariski closure of the set of all affine hyperplanes that are tangent to @xmath190 at some regular point . when studying this duality , algebraic geometers usually work in complex projective space @xmath191 rather than real affine space @xmath192 . in some of the examples for @xmath193 seen in this paper , the algebraic boundary @xmath185 of the coorbitope @xmath184 coincides with the dual variety @xmath189 of the orbit @xmath194 . a good example for this is the discriminantal hypersurface in corollary [ cor : discrorbi ] . more generally , we have the impression that the hypersurface @xmath185 is often irreducible while @xmath182 tends to be reducible . for further appearances of dual varieties in convex algebraic geometry see @xcite . @xmath114-th secant variety _ of @xmath15 is the zariski closure of all @xmath195-flats spanned by points of @xmath54 . the study of secant varieties leads to lower bounds for the carathodory number : [ prop : secant ] if @xmath196 then the @xmath114-th secant variety of @xmath15 is the ambient space @xmath171 . let @xmath197 . the set of points that lie in the convex hull of @xmath198 points of @xmath15 is dense in @xmath50 and hence is zariski dense in @xmath171 . the @xmath114-th secant variety contains that set . the lower bound for @xmath199 from proposition [ prop : secant ] usually does not match fenchel s upper bound @xmath200 . for instance , consider the carathodory orbitope @xmath50 in example [ ex : hankelsmall ] . its algebraic boundary @xmath201 equals the second secant variety of the orbit @xmath15 , so proposition [ prop : secant ] implies @xmath202 . this orbitope satisfies @xmath203 but @xmath204 . for which orbitopes @xmath205 is the boundary @xmath206 one of the secant varieties of @xmath15 ? when is the lower bound on the carathodory number @xmath207 in proposition [ prop : secant ] tight ? a fundamental object in convex optimization is the set @xmath208 of positive semidefinite symmetric real @xmath17-matrices . this is the closed basic semi - algebraic cone defined by the non - negativity of the @xmath209 principal minors . it can also be described by only @xmath55 polynomial inequalities , namely , by requiring that the elementary symmetric polynomials in the eigenvalues , i.e. the suitably normalized coefficients of the characteristic polynomial , be non - negative . the algebraic boundary @xmath210 of the cone @xmath208 is the symmetric @xmath17-determinant . all faces of @xmath208 are exposed , isomorphic to @xmath211 for @xmath212 , and indexed by the lattice of linear subspaces ordered by reverse inclusion . spectrahedra inherit these favorable properties . recall that a _ spectrahedron _ is the intersection of the cone @xmath208 with an affine - linear subspace in @xmath213 . if we know that an orbitope is a spectrahedron then this either answers or simplifies many of our questions . [ ques : isspec1 ] characterize all @xmath214-orbitopes that are spectrahedra . polytopes are special cases of spectrahedra : they arise when the affine - linear space consists of diagonal matrices . one major distinction between polytopes and spectrahedra is that the class of spectrahedra is not closed under projection . that is , the image of a spectrahedron under a linear map is typically not a spectrahedron . see section 5 for orbitopal examples . characterizing projections of spectrahedra among all convex bodies is a major open problem in optimization theory ; see e.g. @xcite . here is a special case of this general problem : [ ques : isspec2 ] is every orbitope the linear projection of a spectrahedron ? in questions [ ques : isspec1 ] and [ ques : isspec2 ] , it is important to keep track of the subfield of @xmath91 over which the data @xmath172 are defined . frequently , this subfield is the rational numbers @xmath178 , and in this case we seek to write the orbitope as a ( projected ) spectrahedron over @xmath178 and not just over @xmath91 . semidefinite programming is the problem of maximizing a linear function over a ( projected ) spectrahedron , and there are efficient numerical algorithms for solving this problem in practice . in our context of orbitopes , the optimization problem can be phrased as follows : what method can be used for maximizing a linear function @xmath215 over an orbitope @xmath51 ? equivalently , how to evaluate the norm @xmath216 associated with the coorbitope @xmath156 ? this is equivalent to a non - linear optimization problem over the compact group @xmath2 . we seek to find @xmath34 which maximizes @xmath217 . this maximum is an algebraic function of @xmath47 . in this section we study two families of orbitopes for the orthogonal group @xmath218 . this group acts on the lie algebra @xmath219 by restricting the adjoint representation of @xmath27 . the @xmath1-module @xmath220 decomposes into two distinguished invariant subspaces , namely @xmath221 and @xmath222 . these correspond to the normal and tangent space of @xmath223 at the identity . in matrix terms , the spaces of symmetric and skew - symmetric matrices form two natural representations of @xmath1 for the action @xmath224 with @xmath225 and @xmath226 . for a symmetric matrix @xmath227 we define the _ symmetric schur - horn orbitope _ @xmath228 we shall see that these orbitopes are intimately related to certain polytopes which govern their boundary structure and spectrahedral representation . this connection arises via the classical schur - horn theorem @xcite . the material in the sections below could also be presented in symplectic language , using the moment maps of atiyah - guillemin - sternberg @xcite . the @xmath229-dimensional space @xmath221 decomposes into the trivial @xmath1-representation , given by multiples of the identity matrix , and the irreducible representation of symmetric @xmath17-matrices with trace zero . every symmetric matrix @xmath227 is orthogonally diagonalizable over @xmath91 . the ordered list of eigenvalues of @xmath230 is denoted @xmath231 . the orbit @xmath232 equals the set of matrices @xmath233 that satisfy @xmath234 . we shall see that its convex hull @xmath235 is the set of matrices @xmath233 for which @xmath236 is majorized by @xmath237 . for @xmath238 we say that @xmath239 is _ majorized _ by @xmath133 , written @xmath240 , if @xmath241 , and , after reordering , @xmath242 and @xmath243 , we have @xmath244 for a fixed point @xmath245 , the set of all points @xmath239 majorized by @xmath133 is the convex polytope @xmath246 here @xmath247 denotes the symmetric group , and @xmath248 is the _ permutahedron _ with respect to @xmath133 . this is a well - studied polytope @xcite and is itself an orbitope for @xmath247 . the permutahedron @xmath248 for @xmath249 consists of all points @xmath250 such that @xmath251 and @xmath252$.}\ ] ] let @xmath253 be the linear projection onto the diagonal . let @xmath227 and @xmath254 its symmetric schur - horn orbitope . then the diagonal @xmath255 is majorized by the vector of eigenvalues @xmath237 . in fact , the orbitope @xmath254 maps linearly onto the permutahedron : @xmath256 [ cor : sh1 ] the schur - horn orbitope equals @xmath257 . we have shown that the right hand side equals @xmath258 . we claim that a matrix @xmath259 is in this set if and only if @xmath259 lies in @xmath260 . this is clear if @xmath259 is a diagonal matrix . it follows for all matrices since both sets are invariant under the @xmath1-action . our next goal is to derive a spectrahedral characterization of @xmath254 . consider the natural action of the lie group @xmath27 on the @xmath114-th exterior power @xmath261 . if @xmath262 is any basis of @xmath24 , then the induced basis vectors of @xmath261 are @xmath263 for @xmath264 . a matrix @xmath265 acts on a basis element by sending it to @xmath266 . we denote by @xmath267 the induced map of lie algebras . the linear map @xmath268 is defined by the rule @xmath269 the @xmath270-matrix that represents @xmath271 in the standard basis of @xmath261 is known as the _ @xmath114-th additive compound matrix _ of the @xmath272-matrix @xmath273 . it has the following main property : [ lem : compoundeigen ] let @xmath274 with eigenvalues @xmath275 . then @xmath271 is symmetric and has eigenvalues @xmath276 for @xmath264 . let @xmath277 be an eigenbasis for @xmath273 . then the formula ( [ eq : addcomprule ] ) says @xmath278 hence @xmath279 is an eigenvector of @xmath271 with eigenvalue @xmath280 . this leads to the result that each symmetric schur - horn orbitope @xmath254 is a spectrahedron . [ thm : symspec ] let @xmath227 with ordered eigenvalues @xmath281 . then @xmath282 a matrix @xmath233 is in @xmath254 if and only if @xmath236 is in the permutahedron @xmath283 . from the inequality representation in ( [ eq : permineq ] ) , in conjunction with lemma [ lem : compoundeigen ] , we see that this is the case if and only if the largest eigenvalue of @xmath284 is at most @xmath285 . we shall now derive the description of all faces of the schur - horn orbitope @xmath254 . since @xmath254 is a spectrahedron , by theorem [ thm : symspec ] , we know that all of its faces are exposed faces . hence a face of @xmath254 is the set of points maximizing a linear function @xmath286 . the canonical @xmath1-invariant inner product on @xmath221 is given by @xmath287 and , by identifying spaces , a linear function may be written as @xmath288 . note that the dual space @xmath289 is equipped with the contragredient action , that is , @xmath290 . [ thm : facelatticesh ] every @xmath1-orbit of faces of @xmath254 intersects the pullback of a unique @xmath247-orbit of faces of the permutahedron @xmath283 . in particular , the faces of @xmath254 are products of symmetric schur - horn orbitopes of smaller dimensions corresponding to flags in @xmath24 . let @xmath150 be a face of @xmath254 and let @xmath291 be a linear function maximized at @xmath150 . then the face @xmath292 is given by @xmath293 and we may identify @xmath294 with the orbit @xmath295 . since @xmath273 is orthogonally diagonalizable , the orbit @xmath295 contains the diagonal matrices @xmath296 for @xmath297 . the corresponding faces are the pullbacks of the orbit of faces of @xmath283 given by the linear function @xmath298 . faces of the permutahedron correspond to flags of coordinate subspaces , and @xmath1 translates these to arbitrary flags of subspaces . facets of the permutahedron @xmath248 correspond to coordinate subspaces . replacing these with arbitrary subspaces yields supporting hyperplanes for the schur - horn orbitope @xmath254 . [ cor : sh2 ] the schur - horn orbitope @xmath299 is the set of matrices @xmath233 such that @xmath300 the following three examples show how theorem [ thm : facelatticesh ] translates into explicit face lattices . [ ex : freespec ] let @xmath301 be the diagonal matrix with diagonal @xmath302 . the orbitope @xmath260 is the convex hull of all symmetric rank @xmath70 matrices with trace @xmath70 , and hence @xmath303 . this orbitope plays the role of a `` simplex among spectrahedra '' because every compact spectrahedron is an affine section . the face @xmath304 of the orbitope @xmath254 in direction @xmath274 is isomorphic to @xmath305 where @xmath306 is the unit sphere and @xmath307 is the eigenspace of @xmath273 with maximal eigenvalue . thus , @xmath308 is isomorphic to a lower dimensional schur - horn orbitope for a rank one matrix . we conclude that the face lattice of @xmath254 consists of the linear subspaces of @xmath24 ordered by inclusion . this fact is well known ; see @xcite . the dimension of a face corresponding to a @xmath114-subspace is @xmath309 . the projection @xmath310 is the standard @xmath311-dimensional simplex @xmath312 whose faces correspond to the coordinate subspaces . we now describe continuous analogs to hypersimplices , extending the simplices in example [ ex : freespec ] . fix @xmath313 and let @xmath314 be the diagonal matrix with @xmath114 ones and @xmath315 zeros . the orbitope @xmath316 of the _ @xmath317-spectrahedral hypersimplex _ , as its diagonal projection @xmath318 is the classical @xmath317-hypersimplex ; cf . * example 0.11 ) . for instance , if @xmath68 and @xmath319 then @xmath299 is nine - dimensional and @xmath320 is an octahedron . up to @xmath321-symmetry , the octahedron has one orbit of vertices and edges but two orbits of triangles . the pullback of any edge is a circle , and the pullbacks of the triangles are five - dimensional symmetric schur - horn orbitopes @xmath322 for @xmath323 and @xmath324 . both facets are isomorphic to free spectrahedra . let @xmath227 be a symmetric matrix with distinct eigenvalues , e.g. @xmath325 . the image of @xmath254 under the diagonal map is the classical permutahedron @xmath326 . its face lattice may be described as the collection of all flags of coordinate subspaces in @xmath24 ordered by refinement . we may associate to every @xmath274 the complete flag whose @xmath114-th subspace is the direct sum of the eigenspaces of the first @xmath114 largest eigenvalues . thus , @xmath327 may be identified with the complete flag variety over @xmath91 . as for the facial structure , the face @xmath308 is isomorphic to the convex hull of the orbit @xmath328 . here , the stabilizer decomposes into a product of groups isomorphic to @xmath329 where @xmath330 is the dimension of the @xmath97-th eigenspace of @xmath273 . hence , the face @xmath308 is isomorphic to a cartesian product of generic schur - horn orbitopes and is of dimension @xmath331 . the face only depends on the flag associated to @xmath273 . this implies that the face lattice of @xmath299 is isomorphic to the set of partial flags ordered by refinement . again , in every orbit of flags there is a flag consisting only of coordinate subspaces . these special flags form the face lattice of the standard permutahedron @xmath332 . we regard @xmath299 as a continuous analog of the permutahedron . we conclude this subsection with a discussion of the algebraic boundary @xmath333 of the schur - horn orbitope . let @xmath334 be the smallest subfield of @xmath91 that contains the eigenvalues @xmath335 , and suppose that the @xmath336 are sufficiently general . then the hypersurface @xmath333 is defined in the affine space @xmath337 by the following polynomial of degree @xmath338 in @xmath229 unknowns over the field @xmath334 : @xmath339 however , from a computer algebra perspective , this is not what we want . assuming that @xmath340 has entries in @xmath178 , we prefer not to pass to the field extension @xmath334 , but we want the algebraic boundary @xmath333 to be the @xmath178-zariski closure of the above hypersurface @xmath341 . for instance , suppose that the characteristic polynomial of @xmath340 is irreducible over @xmath178 . then we must take the product of @xmath342 over all permutations of the eigenvalues @xmath335 , and the polynomial @xmath343 that defines @xmath333 over @xmath178 is the reduced part of that product . it equals @xmath344 where @xmath345 denotes the _ tensor sum _ of two square matrices of the same size ( see e.g. @xcite ) . here the product goes only up to @xmath346 because the matrices @xmath259 and @xmath340 have the same trace . for special matrices @xmath340 , the characteristic polynomial may factor over @xmath178 , and in this case the algebraic boundary @xmath333 is cut out by a factor of the polynomial @xmath342 or @xmath343 . the space @xmath222 consists of skew - symmetric @xmath17-matrices @xmath347 . the eigenvalues of @xmath347 are purely imaginary , say @xmath348 , where @xmath349 , with @xmath350 and an additional @xmath146 eigenvalue if @xmath55 is odd . thus @xmath347 is not diagonalizable over @xmath91 , but the adjoint @xmath1-action brings the matrix @xmath347 into the normal form @xmath351 here @xmath352 is a suitable matrix in @xmath353 , @xmath354 is the diagonal matrix with diagonal @xmath355 , and we denote @xmath356 . let @xmath357 be the linear map such that @xmath358 if @xmath347 is in normal form as above , then @xmath359 . we call @xmath360 the _ skew - diagonal _ of @xmath347 . in analogy to the symmetric case , the set @xmath361 of all skew - diagonals arising from @xmath299 is nicely behaved ; in fact , the necessary changes are rather modest . for a point @xmath362 we denote by @xmath363 the vector of absolute values . for @xmath364 let @xmath365 be the set of points @xmath362 such that @xmath366 is _ weakly majorized _ by @xmath367 . this means that @xmath367 and @xmath366 satisfy the majorization conditions except that @xmath368 is allowed . the polytope @xmath369 is the _ @xmath370-permutahedron_. it is the convex hull of the orbit of @xmath133 under the action of the coxeter group @xmath370 , the group of all @xmath371 signed permutations . the @xmath370-permutahedron @xmath365 for @xmath372 consists of all points @xmath362 with @xmath373 $ with $ i \cap j = \emptyset$.}\ ] ] as expected , we have the following analog of the symmetric schur - horn theorem . let @xmath374 with @xmath375 and @xmath376 the skew - symmetric schur - horn orbitope of @xmath347 . then @xmath377 is weakly majorized by @xmath378 . in particular , we have @xmath379 . the same arguments as in the symmetric case yield the following results . [ thm : facelatticeskewsh ] every @xmath1 orbit of faces of the skew - symmetric schur - horn orbitope @xmath376 contains the pullback of a unique @xmath370-orbit of faces of the @xmath370-permutahedron @xmath380 . the skew - symmetric schur - horn orbitope @xmath376 coincides with the set of skew - symmetric matrices @xmath259 such that @xmath381 is weakly majorized by @xmath377 . fix @xmath382 and @xmath383 , and @xmath384 . then the system ( [ eq : bpermineq ] ) consists of @xmath385 linear inequalities , namely , six inequalities @xmath386 , twelve inequalities @xmath387 , and eight inequalities @xmath388 . their solution set is the @xmath389-permutahedron @xmath390 , commonly known as the _ truncated cuboctahedron _ , and it has @xmath391 vertices , @xmath392 edges and @xmath385 facets ( six octagons , twelve squares and eight hexagons ) . a picture is shown in figure [ f : b3permuta ] . @xmath393{figures / bpermut}\ ] ] -0.3 cm let @xmath347 denote a general skew - symmetric @xmath394-matrix . theorem [ thm : facelatticeskewsh ] implies that the facets of the @xmath21-dimensional orbitope @xmath376 come in three families , corresponding to @xmath395-orbits of the octagons , squares , and hexagons in figure [ f : b3permuta ] . the facets arising from the octagons are skew - symmetric schur - horn orbitopes for @xmath69 with skew - diagonal @xmath396 and therefore have dimension six . the facets arising from the squares are the product of a line segment and a disc , coming from @xmath397 with @xmath398 acting by the determinant . the facets arising from the hexagons are @xmath399-orbitopes isomorphic to symmetric schur - horn orbitopes with eigenvalues @xmath400 and therefore have dimension five . we next present a spectrahedral description of an arbitrary skew - symmetric schur - horn orbitope @xmath376 . to derive this , we return to symmetric matrices and their real eigenvalues . let @xmath374 be a matrix with eigenvalues @xmath401 and let @xmath402 then @xmath403 has eigenvalues @xmath404 . for any @xmath405 , the additive compound matrix @xmath406 has largest eigenvalue @xmath407 . we conclude that each skew - symmetric skew - horn orbitope @xmath380 is a spectrahedron : [ thm : skewsymspec ] let @xmath374 with @xmath408 . then @xmath409 from this theorem we can derive a description of the algebraic boundary as before , and again the issue arises that the @xmath410 lie in an extension @xmath334 over the field of definition of @xmath347 , which will usually be @xmath178 . at present we do not know whether theorems [ thm : symspec ] and [ thm : skewsymspec ] can be extended to obtain spectrahedral representations of the respective orbitopes over @xmath178 . we close this section with one more example of a skew - symmetric schur - horn orbitope . consider the skew - symmetric schur - horn orbitope @xmath376 for some @xmath411 with @xmath412 . according to theorem [ thm : skewsymspec ] , a spectrahedral representation is @xmath413 the projection @xmath361 to the skew diagonal is the _ crosspolytope _ @xmath414 . this a regular polytope with symmetry group @xmath370 , and it has only one orbit of faces in each dimension . the orbitope @xmath376 is the @xmath26 instance of the _ grassmann orbitope _ @xmath415 . these are important in the theory of calibrated manifolds , and we shall study them in section 7 . we argued in section [ sec : setup ] that , given a compact group @xmath2 acting algebraically on @xmath416 , we can identify @xmath2 with a subgroup of @xmath1 , or even of @xmath0 when @xmath2 is connected . the ambient space @xmath417 is itself an @xmath418-dimensional real representation of the group @xmath2 . the action of @xmath2 on @xmath419 is by left multiplication . the orbit of the identity matrix @xmath420 under this action is the group @xmath2 itself . we call the corresponding orbitope @xmath421 the _ tautological orbitope _ for the pair @xmath422 . this orbitope lives in @xmath423 , and it serves as an initial object because it maps linearly to every orbitope @xmath52 in @xmath32 . tautological orbitopes of finite permutation groups have been studied under the name of _ permutation polytopes _ ( see @xcite ) . the most famous of them all is the _ birkhoff polytope _ for @xmath424 , which was studied for other coxeter groups by mccarthy , ogilvie , zobin , and zobin @xcite . in this section we investigate the tautological orbitopes for the full groups @xmath1 and @xmath0 . similar to the schur - horn orbitopes in section [ sec : schurhorn ] , the facial structure is governed by polytopes arising from the projection onto the diagonal . we begin with the example @xmath425 . the group @xmath60 of @xmath426 rotation matrices has dimension three . its tautological orbitope is a convex body of dimension nine . the following spectrahedral representation was suggested to us by pablo parrilo . [ prop : pabloso3 ] the tautological orbitope @xmath427 is a spectrahedron whose boundary is a quartic hypersurface . in fact , a @xmath426-matrix @xmath428 lies in @xmath427 if and only if @xmath429 we first claim that @xmath427 coincides with the set of all @xmath426-matrices @xmath430 where @xmath431 runs over all positive semidefinite @xmath432-matrices having trace @xmath70 . positive semidefinite @xmath432-matrices with both trace @xmath70 and rank @xmath70 are of the form @xmath433 their convex hull is the free spectrahedron of example [ ex : freespec ] . the image of the above rank @xmath70 matrices @xmath434 under the linear map ( [ eq : pablosu2 ] ) is precisely the group @xmath60 . this parametrization is known as the _ cayley transform_. geometrically , it corresponds to the double cover @xmath435 . the claim follows because the linear map commutes with taking the convex hull . the symmetric @xmath432-matrices @xmath431 with @xmath436 form a nine - dimensional affine space , and this space is isomorphic to the nine - dimensional space of all @xmath426-matrices @xmath428 under the linear map given in . we can express each @xmath437 in terms of the @xmath438 by inverting the linear relations @xmath439 , @xmath440 , etc . the resulting symmetric @xmath432-matrix @xmath441 is precisely the matrix in the statement of proposition [ prop : pabloso3 ] . the ideal of the group @xmath399 is generated by the entries of the @xmath426-matrix @xmath442 , while the prime ideal of @xmath60 is that same ideal plus @xmath443 . we can check that the prime ideal of @xmath60 coincides with the ideal generated by the @xmath444-minors of the matrix . thus the group @xmath60 is recovered as the set of matrices of rank one . proposition [ prop : pabloso3 ] implies that @xmath427 is affinely isomorphic to the free spectrahedron for @xmath68 , that is , to the set of positive semidefinite @xmath432-matrices with trace equal to @xmath70 . this implies a characterization of all faces of the tautological orbitope for @xmath60 . first , all faces are exposed because @xmath427 is a spectrahedron . all of its proper faces are free spectrahedra , for @xmath445 , so they have dimensions @xmath146 , @xmath14 and @xmath446 , as seen in example [ ex : freespec ] . we now examine @xmath447 . as before , let @xmath448 denote the projection of the @xmath17-matrices onto their diagonals . [ lem : ontodiagonal ] the projection @xmath449 of the tautological orbitope for the orthogonal group @xmath1 to its diagonal is precisely the @xmath55-dimensional cube @xmath450^n$ ] . the columns of a matrix @xmath451 are unit vectors . thus every coordinate @xmath452 in bounded by @xmath70 in absolute value and @xmath449 is a subset of the cube . for the reverse inclusion , note that all @xmath453 diagonal matrices with entries @xmath454 are orthogonal matrices . the cube @xmath455^n$ ] is the special @xmath456-permutahedron @xmath457 . as with schur - horn orbitopes , the projection onto this polytope reveals the facial structure . as general endomorphisms are not normal , the key concept of diagonalizability is replaced by that of _ singular value decomposition_. recall that for any linear map @xmath226 there are orthogonal transformations @xmath458 such that @xmath459 is diagonal with entries @xmath460 . these entries are the _ singular values _ of @xmath259 . we shall see in ( [ eq : onspec ] ) that @xmath461 is a spectrahedron , hence all of its faces are exposed faces . the following result recursively characterizes all faces of this tautological orbitope . [ thm : tautoon ] let @xmath462 be a linear function on @xmath18 with @xmath463 . then the face of @xmath461 in direction @xmath215 is isomorphic to @xmath464 where @xmath465 . let @xmath466 be a linear function with @xmath467 , so that @xmath468 . we fix a singular value decomposition @xmath469 of the matrix @xmath273 . here @xmath470 is a diagonal matrix @xmath17 with its first @xmath471 entries positive and remaining @xmath472 entries zero . this matrix also defines a linear function @xmath473 on @xmath18 . cyclic invariance of the trace ensures that the faces @xmath474 and @xmath475 are isomorphic . the subset of @xmath1 at which @xmath476 is maximized is the subgroup @xmath477 . the convex hull of this subgroup equals @xmath475 . it coincides with the tautological orbitope for @xmath478 . we interpret theorem [ thm : tautoon ] geometrically as saying that the tautological orbitope for @xmath1 is a continuous analog of the @xmath55-dimensional cube . every face of the cube is a smaller dimensional cube and the dimension of a face maximizing a linear functional @xmath215 is determined by the support of @xmath215 . the role of the support is now played by the rank of the matrix @xmath273 . this behavior yields information about the carathodory number of the tautological orbitope . the carathodory number of the orbitope @xmath479 is at most @xmath480 . by ( * ? * lemma 3.2 ) , the carathodory number of a convex body @xmath137 is bounded via @xmath481 since every proper face is isomorphic to @xmath482 for some @xmath483 the result follows by induction on @xmath55 . the base case is @xmath484 for which @xmath485 is a @xmath70-simplex . note that the orbit @xmath486 coincides with the orbit of the identity matrix @xmath420 under the action of the product group @xmath487 by both right and left multiplication . hence the tautological orbitope @xmath461 is also an @xmath487-orbitope for that action . we shall now digress and study these orbitopes in general . after we have seen ( in theorem [ thm : ononspec ] ) that these are spectrahedra , we shall resume our discussion of @xmath461 . the group @xmath488 acts on @xmath18 by simultaneous left and right translation . the action is given , for @xmath489 and @xmath226 , by @xmath490 ky fan proved in @xcite that the schur - horn theorem for symmetric matrices under conjugation by @xmath1 generalizes to arbitrary square matrices under this @xmath491 action . now , singular values play the role of the eigenvalues . the following is a convex geometric reformulation : [ lem : fan ] for a square matrix @xmath226 let @xmath492 denote its orbitope under the action ( [ eq : fanact ] ) of the group @xmath493 . then the image @xmath494 of the projection to the diagonal is the @xmath456-permutahedron with respect to the singular values @xmath495 . we shall refer to @xmath496 as the _ fan orbitope _ of the matrix @xmath259 . from lemma [ lem : fan ] , one easily deduces the analogous results to theorems [ thm : facelatticesh ] and [ thm : facelatticeskewsh ] . the facial structure of the fan orbitope @xmath492 is determined by the facial structure of the @xmath456-permutahedron @xmath497 specified by the singular values of the matrix @xmath259 . the description of the @xmath456-permutahedron in terms of weak majorization was stated in . rephrasing these same linear inequalities for the singular values , and using lemma [ lem : fan ] , now leads to a spectrahedral description of the fan orbitopes . for that we make use of the alternative characterization of singular values as the square roots of the ( non - negative ) eigenvalues of @xmath498 . using schur complements , it can be seen that the @xmath499-matrix @xmath500 has the eigenvalues @xmath501 . we form its additive compound matrices as before . [ thm : ononspec ] let @xmath259 be a real @xmath17-matrix with singular values @xmath502 . then its fan orbitope @xmath492 equals the spectrahedron @xmath503 fix an integer @xmath504 . the _ ky fan @xmath133-norm _ is defined by @xmath505 this function is indeed a norm on @xmath18 , and its unit ball is the fan orbitope @xmath492 where @xmath259 is the diagonal matrix with @xmath133 diagonal entries @xmath506 and @xmath507 diagonal entries @xmath146 . theorem [ thm : ononspec ] shows that the unit ball in the ky fan @xmath133-norm is a spectrahedron . two norms of special interest in applied linear algebra are the _ operator norm _ @xmath508 and the _ nuclear norm _ @xmath509 . indeed , these two norms played the key role in work of fazel , recht and parrilo @xcite on compressed sensing in the matrix setting . both the operator norm and the nuclear norm are closely tied to their vector counterparts . the unit balls in the operator and nuclear norm are both fan orbitopes . the projection to the diagonal yields the unit balls for the @xmath510 and the @xmath511-norm on @xmath24 . these two unit balls are polytopes in @xmath24 , namely , the @xmath55-cube and the @xmath55-crosspolytope respectively . fazel _ et al . _ showed in ( * ? ? ? 2.1 ) that the nuclear norm ball has the structure of a spectrahedron , and semidefinite programming duality allows for linear optimization over the operator norm ball . the spectrahedral descriptions of both unit balls in @xmath18 are special instances of theorem [ thm : ononspec ] , to be stated explicitly once more . the operator norm ball consists of all matrices @xmath190 whose largest singular value is at most @xmath70 . this is equivalent to @xmath512 the nuclear norm ball consists of all matrices @xmath190 for which the sum of the singular values is at most @xmath70 . this is equivalent to saying that the sum of the largest @xmath55 eigenvalues of the symmetric @xmath513 matrix @xmath514 is at most @xmath70 . hence the nuclear norm ball in @xmath18 is the spectrahedron defined by the linear matrix inequality @xmath515 we are now finally prepared to return to the main aim of this section , which is the study of tautological orbitopes . the proof of theorem [ thm : tautoon ] implies that the convex hull of @xmath1 equals the set of @xmath516-matrices whose largest singular value is at most @xmath70 . in other words : the operator norm ball in @xmath18 is equal to the tautological orbitope of the orthogonal group @xmath1 . the coorbitope @xmath517 is the nuclear norm ball . both of these convex bodies are spectrahedra . they are characterized in ( [ eq : onspec ] ) and ( [ eq : onspec2 ] ) respectively . it would be worthwhile to explore implications of our geometric explorations of orbitopes for algorithmic applications in the sciences and engineering , such as those proposed in @xcite . we next discuss the faces of the tautological orbitope of the group @xmath0 . the relevant convex polytope is now the convex hull @xmath518 of all vertices @xmath519 with an even number of @xmath520-entries . this polytope is known as the _ demicube _ or _ halfcube_. it is a permutahedron for the coxeter group of type @xmath521 , i.e. signed permutations with an even number of sign changes . the facet hyperplanes of @xmath518 are derived by separating infeasible vertices of @xmath455^n$ ] by hyperplanes through the @xmath55 neighboring vertices : @xmath522^n \,:\ , \sum_{i \not\in j } x_i - \sum_{i \in j } x_i \le n-2 \,\,\,\ , \hbox{for all $ j \subseteq [ n]$ of odd cardinality } \bigr\}.\ ] ] while @xmath523 is only a line segment , we have @xmath524 for @xmath525 . for example , the halfcube @xmath526 is the tetrahedron with vertices @xmath527 , @xmath528 , @xmath529 and @xmath530 . note that its facet inequalities appear as the diagonal entries in the symmetric @xmath432-matrix ( [ eq : magic ] ) : @xmath531 this observation is explained by results of horn ( cf . @xcite ) on the diagonals of special orthogonal matrices . these imply the following lemma about the tautological orbitope of @xmath0 : the projection of @xmath532 onto the diagonal equals the halfcube @xmath518 . just like in lemma [ lem : ontodiagonal ] , it is clear that @xmath518 is a subset of @xmath533 . the converse is derived from the linear algebra fact that the trace of any matrix in @xmath534 is at most @xmath535 . for @xmath536 $ ] let @xmath537 be the diagonal matrix with @xmath538 if @xmath539 and @xmath540 if @xmath541 . let @xmath542 . then @xmath543 for all @xmath544 of odd cardinality . this means that @xmath545 satisfies the linear inequalities in ( [ eq : sonfacets ] ) and hence lies in @xmath518 . there is a variant of singular value decomposition with respect to the restricted class of orientation preserving transformations . for every matrix @xmath226 there exist rotations @xmath546 such that @xmath547 is diagonal . the diagonal entries are called the _ special singular values _ and denoted by @xmath548 . the main difference to the usual singular values is that @xmath549 may be negative ; only the first @xmath550 entries of @xmath551 are non - negative . we need to make this distinction in order to understand the faces of @xmath552 . [ thm : sontauto ] the tautological orbitope of @xmath0 has precisely two orbits of facets . these are the tautological orbitopes for @xmath553 and the free spectrahedra of dimension @xmath554 . up to @xmath521-symmetry , the halfcube @xmath518 has only two distinct facets , namely an @xmath311-dimensional halfcube and an @xmath311-dimensional simplex . a typical halfcube facet @xmath555 arises by maximizing the linear function @xmath556 over @xmath518 , and a typical simplex facet @xmath557 arises by maximizing the linear function @xmath558 . pulling back @xmath215 along the diagonal projection @xmath559 , we see that the facet of @xmath552 corresponding to the halfcube facet is the convex hull of all rotations @xmath542 that fix the first standard basis vector @xmath560 . pulling back @xmath476 along @xmath559 , we see that the facet of @xmath552 corresponding to the simplex facet is the convex hull of @xmath561 . this facet is isomorphic to the convex hull of all @xmath562 such that @xmath563 . since @xmath564 is orientation reversing , one eigenvalue of @xmath564 is @xmath565 , and @xmath566 forces all other eigenvalues to be equal to @xmath70 . hence the facet in question is the symmetric schur - horn orbitope for the diagonal matrix @xmath567 . example [ ex : freespec ] implies that this is a free spectrahedron . in subsection 4.1 we exhibited a spectrahedral representation for the tautological orbitope @xmath427 , and in that case , the two facet types of theorem [ thm : sontauto ] collapse into one type . at present , we do not know how to generalize the representation ( [ eq : magic ] ) to @xmath0 for @xmath568 . orbitopes for @xmath19 were first studied by carathodory @xcite . the coorbitope cone ( [ eq : coorbitopecone ] ) dual to such a _ carathodory orbitope _ consists of non - negative trigonometric polynomials . this leads to the toeplitz spectrahedral representation of the universal carathodory orbitope in theorem [ th : toeplitz ] , which implies that the convex hulls of all trigonometric curves are projections of spectrahedra . the universal carathodory orbitope is also affinely isomorphic to the convex hull of the compact even moment curve , whose coorbitope cone consists of non - negative univariate polynomials . this leads to the representation by hankel matrices in theorem [ th : hankel ] . the irreducible representations @xmath569 of @xmath19 are indexed by non - negative integers @xmath123 . here , @xmath570 is the trivial representation . when @xmath571 is positive , the representation @xmath569 of @xmath19 acts on @xmath63 , and it sends a rotation matrix to its @xmath123th power : @xmath572 for a vector @xmath573 we consider the direct sum of these representations @xmath574 the _ carathodory orbitope _ @xmath575 is the convex hull of the orbit @xmath576 under the action @xmath577 on the vector space @xmath578 . this orbit is the _ trigonometric moment curve _ @xmath579 this curve is also identified with the matrix group @xmath580 lying in the space @xmath581 of block - diagonal @xmath582-matrices with @xmath23 blocks of size @xmath583 . thus @xmath575 is isomorphic to the convex hull of @xmath580 , and can therefore also be thought of as a tautological orbitope . we distinguish between isomorphisms of orbitopes that preserve the @xmath19-action , and the weaker notion of affine isomorphisms that preserve their structure as convex bodies . [ l : all_cara ] any orbitope of the circle group @xmath19 is isomorphic to a carathodory orbitope @xmath575 , where @xmath584 has distinct coordinates , and it is affinely isomorphic to a carathodory orbitope where the coordinates of @xmath259 are relatively prime integers . let @xmath51 be an orbitope for @xmath19 . we may assume that its ambient @xmath19-module @xmath32 has no trivial components and is the linear span of @xmath51 . then @xmath32 has the form @xmath577 for @xmath585 with distinct non - zero components , that is , @xmath32 is multiplicity free . this is because @xmath586 , where we identify @xmath63 with @xmath92 and @xmath19 with the unit circle in @xmath92 . the orbitope @xmath51 is generated by a vector @xmath587 with non - zero coordinates . by complex rescaling , we may assume that @xmath588 , showing that @xmath51 is isomorphic to @xmath575 . lastly , if the coordinates of @xmath589 have greatest common divisor @xmath123 , then @xmath577 is the composition @xmath590 , where @xmath591 and @xmath51 is affinely isomorphic to an orbitope for the module @xmath592 . we henceforth assume that @xmath593 where the @xmath594 are relatively prime . when @xmath595 , carathodory @xcite studied the facial structure of @xmath575 . an even - dimensional cyclic polytope is the convex hull of finitely many points on carathodory s curve ( see @xcite ) . smilansky @xcite studied the four - dimensional carathodory orbitopes ( @xmath26 ) , and this was recently extended by barvinok and novik @xcite who studied @xmath596 . the corresponding curve ( [ caracurve ] ) is the _ symmetric moment curve _ which gives rise to a remarkable family of centrally symmetric polytopes with extremal face numbers . many questions remain about the facial structure of the barvinok - novik orbitopes @xmath596 . see @xcite for details . we now focus on the _ universal carathodory orbitope _ @xmath597 in @xmath598 . this convex body has the following spectrahedral representation in terms of hermitian toeplitz matrices . [ th : toeplitz ] the universal carathodory orbitope @xmath599 is isomorphic to the spectrahedron consisting of positive semidefinite hermitian toeplitz matrices with ones along the diagonal : @xmath600 we note that the complex spectrahedron ( [ toeplitz ] ) can be translated into a spectrahedron over @xmath91 as follows . consider a hermitian matrix @xmath601 where @xmath150 is real symmetric and @xmath2 is real skew - symmetric . the hermitian matrix @xmath602 is positive definite if and only if @xmath603 a _ trigonometric curve _ is any curve in @xmath24 that is parametrized by polynomials in the trigonometric functions sine and cosine , or equivalently , any curve that is the image under a linear map of the universal trigonometric moment curve where @xmath595 . the convex hull of any trigonometric curve is a projected spectrahedron . in particular , all carathodory orbitopes are projected spectrahedra . @xmath604{figures / trig_curve_1.eps } } \put(135,0){\includegraphics[height=120pt]{figures / trig_curve_2.eps } } \put(280,0){\includegraphics[height=120pt]{figures / trig_curve_3.eps } } \put ( 0,-17){$(\cos(\theta),\sin(\theta),\cos(2\theta))$ } \put(135,-17){$(\cos(2\theta),\sin(2\theta),\cos(\theta))$ } \put(280,-17){$(\cos(\theta),\sin(2\theta),\cos(3\theta))$ } \end{picture}\ ] ] -0.3 cm figure [ f : trig_curves ] shows the convex hulls of three trigonometric curves in @xmath66 . the left and middle convex bodies are each the intersection of two convex quadratic cylinders ( @xmath605 and @xmath606 for the former ; @xmath607 and @xmath608 for the latter ) and hence are spectrahedra . the rightmost convex body is visibly not a spectrahedron . its exposed points are @xmath609 for @xmath610 , and it has two algebraic families of one - dimensional facets . in addition , there are two two - dimensional facets , namely the equilateral triangles for @xmath611 and @xmath612 . the six edges of these two triangles are one - dimensional non - exposed faces , as there is no linear function which achieves its minimum on this body along these edges . also , exactly one vertex of each triangle at @xmath613 and @xmath614 is exposed . we conclude that this convex body is a projected spectrahedron but it is not a spectrahedron . the coorbitope cone dual to @xmath615 consists of all affine - linear functions that are non - negative on @xmath615 . these correspond to non - negative trigonometric polynomials : @xmath616 we identify each point @xmath617 in @xmath618 with the laurent polynomial @xmath619\ ] ] with @xmath620 , @xmath621 , @xmath349 , and @xmath622 . then @xmath623 and @xmath624 if and only if @xmath625 is non - negative on the unit circle @xmath153 of @xmath92 . roots of @xmath625 occur in pairs @xmath626 and those on @xmath153 have even multiplicity . choosing one root from each pair gives the factorization @xmath627 where @xmath628 $ ] has degree @xmath23 and the coefficient vectors of @xmath602 and @xmath629 are complex conjugate . this factorization is the classical fejr - riesz theorem . utilizing the monomial map @xmath630 with @xmath631 , this is equivalent to the following : a trigonometric polynomial @xmath632 is non - negative on the unit circle if and only if there is a non - zero vector @xmath633 such that @xmath634 a point @xmath635 belongs to the carathodory orbitope @xmath636 if and only if @xmath637 the sum on the left equals the hermitian inner product in @xmath638 of the coefficient vector @xmath639 of the polynomial @xmath632 and the vector @xmath640 with @xmath641 as in ( [ toeplitz ] ) . the formula ( [ circlesos2 ] ) expresses @xmath639 as the image of the hermitian matrix @xmath642 under some linear projection @xmath643 . if @xmath644 denotes the linear map dual to @xmath643 then @xmath645 is precisely the hermitian toeplitz matrix in ( [ toeplitz ] ) . we conclude that the sum in ( [ circlepairing ] ) equals @xmath646 thus the point @xmath647 represented by a hermitian toeplitz matrix @xmath190 lies in @xmath615 if and only if @xmath648 for all @xmath649 if and only if @xmath190 is positive semidefinite . the proof of theorem [ th : toeplitz ] elucidates the known results about the facial structure of @xmath615 . [ neighborly ] the universal carathodory orbitope @xmath650 is a neighborly simplicial convex body . its faces are in inclusion - preserving correspondence with sets of at most @xmath23 points on the circle . a laurent polynomial @xmath625 as in lies in the boundary of the coorbitope cone @xmath651 if and only if it is non - negative on the unit circle @xmath153 but not strictly positive . it supports the face of @xmath615 spanned by the points of the trigonometric moment curve corresponding to its zeros in @xmath153 . each zero has multiplicity at least @xmath14 , so there are at most @xmath23 such points , and conversely any subset of @xmath652 points supports a face . since any fewer than @xmath653 points on the curve are affinely independent and since all faces are exposed , these faces are simplices . corollary [ neighborly ] implies that the carathodory number @xmath654 equals @xmath655 , as no point in the interior of @xmath615 lies in the convex hull of @xmath23 points of the orbit , but @xmath654 is at most one more than the maximal carathodory number of a facet , by lemma 3.2 of @xcite . carathodory orbitopes are generally not spectrahedra because they can possess non - exposed faces . smilansky @xcite showed that if we write @xmath656 for a point on the trigonometric moment curve with weights @xmath70 and @xmath71 then the faces of @xmath657 are exactly the points @xmath658 of the orbit , the line segments @xmath659 , where @xmath660 , and the triangles @xmath661 in particular , each line segment @xmath662 is a non - exposed edge of @xmath657 . we conclude that the barvinok - novik orbitope @xmath657 is not a spectrahedron . the toeplitz representation ( [ toeplitz ] ) of the universal carathodory orbitope @xmath615 reveals complete algebraic information . for example , the algebraic boundary @xmath663 is the irreducible hypersurface of degree @xmath655 defined by the determinant of that @xmath664-matrix . the curve itself is the set of all positive definite hermitian toeplitz matrices of rank one . the @xmath583-minors of the matrix ( [ toeplitz ] ) generate the prime ideal @xmath665 of this rational curve . the union of the @xmath666-dimensional faces of @xmath615 is the set of positive definite hermitian toeplitz matrices of rank @xmath667 , as a point lies on a @xmath666-dimensional face if and only if it is the convex combination of @xmath667 points of the curve . the zariski closure of this stratum is the set of all rank @xmath667 hermitian toeplitz matrices which is defined by the vanishing of the @xmath668 minors of that matrix . this is also the @xmath667th secant variety of the carathodory curve . lastly , this rank stratification is a whitney stratification of the algebraic boundary . the derivation of the algebraic description of @xmath575 for arbitrary @xmath669 requires the process of elimination . for instance , the ideal @xmath670 of the trigonometric moment curve ( [ caracurve ] ) can be computed from the ideal of @xmath444-minors for @xmath671 by eliminating all unknowns @xmath672 with @xmath673 . the equation of the algebraic boundary @xmath674 is obtained by the same elimination applied to a certain ideal of larger minors of the toeplitz matrix ( [ toeplitz ] ) . we refer to recent work of vinzant @xcite for a detailed study of the edges of barvinok - novik orbitopes . an analysis of the algebraic boundary the orbitope @xmath675 can be found in @xcite . the cone over the degree @xmath23 moment curve is the image of @xmath63 in its @xmath23th symmetric power @xmath676 under the map @xmath677 this map is naturally @xmath19-equivariant . we define the _ compact moment curve _ to be the image @xmath678 of the unit circle under the map @xmath679 . this restricted map equals @xmath680 the convex hull of the curve @xmath678 is an orbitope . by lemma [ l : all_cara ] , it is isomorphic to some carathodory orbitope @xmath575 . the following lemma makes this identification explicit . [ l : identification ] if @xmath681 is odd , then @xmath682 is isomorphic to the barvinok - novik orbitope @xmath683 . if @xmath681 is even , then @xmath682 is isomorphic to @xmath684 , which is affinely isomorphic to the universal carathodory orbitope @xmath685 . complexifying the @xmath19-module @xmath577 where @xmath589 gives the @xmath686-module with symmetric weights @xmath687 . thus the underlying real @xmath19-module of this @xmath686-module is @xmath688 . the lemma follows because the complexified representation @xmath689 of @xmath690 has weights @xmath691 , and this representation is spanned by the pure powers of linear forms , so every weight appears in the linear span of the orbit . suppose now that @xmath692 is even . we describe the moment curve and the carathodory orbitope in coordinates @xmath693 for @xmath694 . fix the @xmath695-hankel matrix @xmath696 [ th : hankel ] the even moment curve consists of all vectors @xmath697 such that the hankel matrix @xmath698 has rank one , is positive semidefinite , and satisfies the linear equation @xmath699 its convex hull consists of all @xmath700 such that @xmath698 is positive semidefinite and satisfies . thus the carathodory orbitope @xmath701 has a second _ hankel representation _ as a spectrahedron . the proof of this theorem follows from the well - known fact that every non - negative polynomial in one variable is a sum of squares of polynomials . it uses the duality in @xcite . observe that the points of the even compact moment curve satisfy , which comes from the polynomial identity @xmath702 . as @xmath703 has just one copy of the trivial representation , this is the only affine equation that holds on @xmath704 . the dual to @xmath703 is the space @xmath705_{2n}$ ] of real homogeneous polynomials of degree @xmath706 in @xmath707 and @xmath708 . the coefficients of such a polynomial gives coordinates for @xmath705_{2n}$ ] . the coorbitope cone @xmath709 dual to the orbitope @xmath710 is the cone of homogeneous polynomials of degree @xmath706 that are non - negative on @xmath63 . thus a point @xmath711 lies in the orbitope @xmath710 if and only if it satisfies and @xmath712 for every non - negative polynomial @xmath713 . since non - negative homogeneous polynomials in @xmath707 and @xmath708 are sums of squares ( cf . @xcite ) , we only need these inequalities to hold when @xmath714 is a square . writing @xmath715 for the coefficient vector of the polynomial @xmath716 , the sum becomes @xmath717 where @xmath698 is the hankel matrix . this proves the theorem . the hankel representation of the universal carathodory orbitope arose by considering the image of the circle @xmath718 in @xmath703 and its relation to non - negative binary forms . generalizing from @xmath63 to @xmath719 gives the veronese orbitopes whose coorbitope cones ( [ eq : coorbitopecone ] ) consist of non - negative @xmath23-ary forms . when @xmath720 , non - negative forms are not necessarily sums of squares , except for quadratic forms and the exceptional case of ternary quartics . the set of decomposable symmetric tensors is the image of the veronese map @xmath721 the @xmath722-orbits through any two non - zero decomposable tensors are scalar multiples of each other and are thus isomorphic . we define the _ veronese orbitope _ @xmath723 to be the convex hull of the orbit through the specific decomposable tensor @xmath724 . that orbit is also the image @xmath725 of the unit @xmath726-sphere under the @xmath472-th veronese embedding of @xmath719 . suppose that @xmath727 is even . then the orbit @xmath725 can be identified with @xmath728 since @xmath729 is two - to - one with @xmath730 . the dual vector space to @xmath731 is the space of homogeneous forms of degree @xmath706 on @xmath719 . the only invariant forms are those proportional to the form @xmath732 , so both @xmath731 and its dual space contain one copy of the trivial representation , and @xmath733 lies in the hyperplane of @xmath731 defined by @xmath734 . the dual cone to the veronese orbitope @xmath735 is the cone of non - negative forms of degree @xmath706 in @xmath736 . see also ( * ? ? ? * example ( 1.2 ) ) . we write @xmath737 for the veronese coorbitope cone consisting of non - negative forms . the cone @xmath738 of non - negative forms contains the cone @xmath739 of sums of squares , but when @xmath720 , @xmath740 , and @xmath741 , hilbert @xcite showed that the inclusion is strict . we refer to @xcite for a recent study which compares the dimension of the faces of these cones . the cone @xmath739 is naturally a projection of the positive semidefinite cone . hence its dual cone @xmath742 is a spectrahedron : it can be realized as the intersection of the positive semidefinite cone with a certain linear space of generalized hankel matrices , discussed in detail in reznick s book @xcite . this spectrahedron @xmath743 is strictly larger than the veronese orbitope @xmath744 when @xmath720 , @xmath740 , and @xmath741 . in fact , @xmath744 is precisely the convex hull of the subset of extreme points in @xmath743 that have rank one . we present a detailed case study of the exceptional case of ternary quartics , when @xmath745 . the veronese orbitope @xmath746 is a @xmath747-dimensional convex body . let @xmath748 be the @xmath21-dimensional cone over the veronese orbitope @xmath749 . as all non - negative ternary quartics are sums of squares , we have the following identities of cones : @xmath750 we next present reznick s spectrahedral representation of @xmath749 . for this , we identify @xmath751 with its dual , and we introduce coordinates @xmath752 where the indices @xmath753 are the exponents of monomials in variables @xmath754 of degree @xmath755 . the ternary quartic corresponding to @xmath700 is @xmath756 where @xmath757 is the multinomial coefficient . the inner product @xmath758 is @xmath60-invariant . given a ternary quartic @xmath759 in the notation , we associate to it the following symmetric @xmath760-matrix with hankel structure as in ( * ? ? ? ( 5.25 ) ) : @xmath761 the veronese orbitope @xmath762 is a spectrahedron . it consists of all positive semidefinite hankel matrices @xmath763 as in @xmath764 that satisfy the equation @xmath765 it is shown in ( * ? ? ? 5 ) that the quartic @xmath759 is non - negative if and only if the hankel matrix @xmath763 is positive semidefinite . furthermore , the equation is the affine equation @xmath766 which defines the hyperplane containing the orbit @xmath767 . by contrast , the veronese coorbitope is not a spectrahedron . [ th : sos ] the convex cone of non - negative ternary quartics @xmath768 is not a spectrahedron . its facets have dimension twelve and the intersection of any two facets is an exposed face of dimension nine . it also has maximal non - exposed faces of dimension nine . we thank greg blekherman who explained this to us . the cone @xmath769 is full - dimensional in the @xmath21-dimensional space @xmath736 . its facets come from its defining linear inequalities @xmath770 for @xmath771 , let @xmath772 be the facet exposed by the inequality @xmath773 , which consists of those non - negative ternary quartics @xmath239 that vanish at @xmath133 . since the boundary of @xmath769 is @xmath747-dimensional and we have a two - dimensional family of isomorphic facets , we see that each facet @xmath772 is @xmath774-dimensional . a non - negative form that vanishes at @xmath771 must also have its two partial derivatives ( in local coordinates at @xmath133 ) vanish , which gives three linear conditions on the facet @xmath772 . concretely , if we take @xmath775 $ ] with local coordinates @xmath776 , then the constant and linear terms of the inhomogeneous quartic @xmath777 must vanish . consequently , @xmath778 where @xmath602 , @xmath779 , and @xmath780 are , respectively , the terms of degrees @xmath14 , @xmath71 , and @xmath755 in @xmath239 . these are binary forms . their @xmath781 coefficients parametrize the linear span of the facet @xmath772 , showing again that @xmath772 is 12-dimensional . the quadratic form @xmath782 is the hessian of @xmath777 at @xmath775 $ ] in these coordinates . a form @xmath239 lies in the relative interior of the facet @xmath772 if and only if , given any @xmath783 which vanishes at @xmath133 along with its partial derivatives , so that it lies in the linear span of @xmath772 , there is an @xmath784 such that @xmath785 also lies in @xmath772 . these conditions are equivalent to 1 . @xmath239 has no other zeros in @xmath786 , and 2 . the hessian of @xmath239 at @xmath133 is a positive definite quadratic form . a form @xmath787 lies in the boundary of @xmath772 when one of these conditions fails , that is , either 1 . @xmath239 vanishes at a second point @xmath788 , or 2 . the hessian of @xmath239 has a double root at some point @xmath789 . faces of type ( 1 ) have the form @xmath790 . these are nine - dimensional and occur in a four - dimensional family parametrized by pairs of distinct points in @xmath786 . the union of all such faces is a semialgebraic subset of dimension @xmath791 in the boundary of @xmath769 . faces of type ( 2 ) also have dimension nine . the condition that the hessian has a double root at a point @xmath792 gives two linear conditions on the coefficients of @xmath239 . there is an additional condition that the cubic part @xmath779 of @xmath239 in also vanishes at @xmath793 , for otherwise @xmath239 takes negative values along the line through @xmath133 corresponding to @xmath793 . a face of type ( 2 ) is the limit of faces @xmath790 of type ( 1 ) as @xmath794 approaches @xmath133 along the line corresponding to @xmath793 . these faces form a three - dimensional family on which @xmath60 acts faithfully and transitively . let us now examine the exposed faces of @xmath769 . let @xmath795 be a symmetric tensor in the dual cone to @xmath769 . then @xmath215 is the sum of decomposable symmetric tensors , @xmath796 and so it supports the face @xmath797 . thus the exposed faces of @xmath769 are intersections of facets , and they consist of non - negative ternary quartics that vanish at a given set of points . since the faces of type ( 2 ) are not of this form , they are not exposed . our next agenda item is a discussion of the algebraic boundaries of the convex bodies and cones discussed above . the algebraic boundary of the cone @xmath748 is characterized by the vanishing of the determinant of the hankel matrix @xmath763 in . from this we conclude : the algebraic boundary of the veronese orbitope @xmath749 is the variety of dimension @xmath798 and degree six which is defined by the linear equation and the hankel determinant @xmath799 . the extreme points of @xmath749 are precisely the hankel matrices @xmath763 of rank @xmath70 . we observed in the proof of theorem [ th : sos ] that the boundary of @xmath769 consists of non - negative quartics @xmath239 that vanish at some point @xmath133 of @xmath786 , and that the partial derivatives of @xmath239 necessarily also vanish at @xmath133 . that is , the plane quartic curve defined by @xmath800 is singular at @xmath133 . thus the algebraic boundary of @xmath769 consists of singular ternary quartics . working in @xmath801 , and its dual space of ternary quartics , this algebraic boundary is seen to be the dual variety to the veronese surface which consists of rank @xmath70 hankel matrices @xmath763 . [ cor : discrorbi ] the algebraic boundary of the coorbitope cone @xmath769 is an irreducible hypersurface of degree @xmath802 . its defining polynomial is the discriminant @xmath803 of the ternary quartic @xmath804 the discriminant @xmath803 is a homogeneous polynomial of degree @xmath802 in the @xmath21 indeterminates @xmath805 . in what follows we shall present an explicit expression for @xmath803 . that expression will be derived from a beautiful classical formula due to sylvester which can be found in section 3.4.d , starting on page 118 , of the book by gel@xmath806fand , kapranov , and zelevinsky @xcite . according to ( 1.7 , page 434 ) , the discriminant @xmath803 is proportional to the resultant @xmath807 of the three partial derivatives of the quartic @xmath239 . here @xmath808 denotes the resultant of three ternary cubics , and the precise relation is @xmath809 . we write @xmath810 for the space of linear forms on @xmath66 , and we introduce the linear map @xmath811 next , for an exponent vector @xmath812 of degree @xmath813 and any variable @xmath814 , we choose a decomposition of the cubic partial derivative @xmath815 where @xmath816 , @xmath817 , and @xmath818 are forms of degree @xmath819 , @xmath820 , and @xmath821 , respectively . then @xmath822 is a quartic polynomial . finally , we define a linear map @xmath823 by sending @xmath824 , where @xmath825 is the basis dual to the monomial basis of @xmath826 . ( sylvester @xcite ) the discriminant @xmath803 is proportional to the resultant of the ternary cubics @xmath827 which is equal to the determinant of the linear map @xmath828 this is an irreducible homogeneous polynomial of degree @xmath802 in the @xmath21 coefficients @xmath805 . we write this map explicitly as a @xmath829 matrix @xmath830 whose rows are indexed by the @xmath21 monomials @xmath831 of degree @xmath832 and whose columns are indexed by @xmath21 auxiliary quartics , and whose entry in a given row and column is the coefficient of that monomial in that auxiliary quartic . nine of the quartics come from the map @xmath833 in . they are @xmath834 the other six are the polynomials @xmath835 , @xmath836 , @xmath837 , @xmath838 , @xmath839 , and @xmath840 from @xmath841 . we only describe @xmath835 and @xmath838 as the others may be recovered from these by symmetry . for @xmath835 , note that each partial derivative of @xmath239 has six terms divisible by @xmath707 , three divisible by @xmath708 and not @xmath707 , and a unique term involving @xmath842 . this leads to a decomposition , and @xmath835 is the determinant of the @xmath843 matrix : @xmath844 by a similar reasoning , we find that @xmath838 is the determinant of the @xmath843 matrix : @xmath845 this concludes our discussion of the algebraic boundary of the coorbitope cone @xmath769 . we close with the remark that the notations @xmath846 and @xmath847 are consistent with those used in the paper @xcite where these cones consist of concentration matrices and sufficient statistics of a certain gaussian model . the _ grassmann orbitope _ @xmath415 is the convex hull of the grassmann variety of oriented @xmath23-dimensional linear subspaces of @xmath24 in its plcker embedding in the unit sphere in @xmath22 . equivalently , this is the highest weight orbitope for the group @xmath0 acting on @xmath22 : @xmath848 faces of the grassmann orbitope are of considerable interest in differential geometry since , according to the fundamental theorem of calibrations , they correspond to area - minimizing @xmath23-dimensional submanifolds of @xmath24 . references to this subject include the seminal article on calibrated geometries by harvey and lawson @xcite and the beautiful expositions by morgan @xcite . in this section we review basic known facts about @xmath415 and we initiate its study from the perspectives of combinatorics , semidefinite programming , and algebraic geometry . vectors in @xmath22 are written in terms of plcker coordinates relative to the standard basis : @xmath849 the plcker vector @xmath133 lies in the _ oriented grassmann variety _ if and only if it is decomposable , i.e. @xmath850 for some pairwise orthogonal subset @xmath851 of @xmath852 . this happens if and only if @xmath133 lies in the unit sphere in @xmath22 and satisfies all quadratic plcker relations , the relations among the @xmath853-minors of a @xmath854-matrix . these relations generate the prime ideal called the _ plcker ideal _ @xmath855 . thus the oriented grassmann variety is the algebraic subvariety of @xmath22 defined by the ideal @xmath856 the convex hull of that real algebraic variety is the @xmath857-dimensional grassmann orbitope @xmath415 . [ ex : g24 ] the grassmann orbitope @xmath858 is the convex hull of the variety of @xmath859 as suggested by ( * ? ? ? * proposition 3.2 ) , we perform the orthogonal change of coordinates @xmath860 this is simultaneous rotation by @xmath861 in each of the coordinate planes spanned by the pairs @xmath862 , @xmath863 , and @xmath864 . in these new coordinates , the prime ideal ( [ eq : pluckerideal24 ] ) equals @xmath865 this reveals that @xmath858 is the direct product of two three - dimensional balls of radius @xmath866 . we next examine the case @xmath867 and arbitrary @xmath55 . the vectors @xmath133 in @xmath868 can be identified with skew - symmetric @xmath17-matrices , and this brings us back to the orbitopes in section 3.2 . if @xmath133 is a real skew - symmetric matrix whose eigenvalues are @xmath872 , where @xmath873 , then the matrix @xmath874 is hermitian and its eigenvalues are the real numbers @xmath875 . recall that the operator @xmath876 computes the @xmath114-th additive compound matrix of a given matrix . [ thm : g2n ] let @xmath877 and @xmath878 . the grassmann orbitope equals the spectrahedron @xmath879 its algebraic boundary @xmath880 is an irreducible hypersurface of degree @xmath881 , defined by a factor of the determinant of the matrix @xmath882 . the proper faces of @xmath869 are @xmath883-orbitopes for @xmath884 . every face @xmath150 is associated with an even - dimensional subspace @xmath885 equipped with an orthogonal complex structure and the extreme points of @xmath150 correspond to complex lines in @xmath885 . everything in this theorem is also true for the small cases @xmath886 , with the exception that the quartic hypersurface @xmath887 is not irreducible , as was seen in example [ ex : g24 ] . in light of theorem [ thm : facelatticeskewsh ] , the reducibility of @xmath858 arises because the two - dimensional crosspolytope equals the square , which decomposes as a minkowski sum of two line segments . this may also be seen as the stabilizer of a decomposable tensor in @xmath69 is @xmath888 , where @xmath889 is the identity matrix , and @xmath890 , so @xmath858 is also an orbitope for @xmath891 . we begin with the last statement about the face lattice of @xmath869 . this result is well - known in the theory of calibrations , where it is usually phrased as follows : _ every face of the grassmannian of two - planes in @xmath24 consists of the complex lines in some @xmath892-dimensional subspace of @xmath24 under some orthogonal complex structure_. see @xcite . to derive the spectrahedral representation ( [ eq : spectrag2n ] ) , we note that the eigenvalues of @xmath893 are the sums of any @xmath114 distinct numbers of the eigenvalues of the skew - symmetric matrix @xmath133 . these are @xmath894 if @xmath55 is even and @xmath895 if @xmath55 is odd . in light of corollary [ rmk : grassinsec3 ] , we can apply the results in section 3.2 to conclude that @xmath133 lies in @xmath869 if and only if @xmath896 for all choices of signs . in terms of polyhedral geometry , this condition means that the vector @xmath897 lies in the crosspolytope @xmath898 . since all @xmath899 eigenvalues of @xmath893 are bounded above by the maximum of the @xmath881 special eigenvalues @xmath900 , we conclude that @xmath901 if and only if @xmath902 . to compute the algebraic boundary @xmath903 we consider the expression @xmath904 this is a symmetric polynomial of degree @xmath905 in the squared eigenvalues @xmath906 , and hence it can be written as a polynomial in the coefficients of the characteristic polynomial @xmath907 the resulting polynomial has degree @xmath881 in the entries @xmath908 of the matrix @xmath909 , and it vanishes on the boundary of the orbitope @xmath869 . it can be checked that it is irreducible for @xmath877 . let @xmath910 and consider the characteristic polynomial of our hermitian matrix : @xmath911 the algebraic boundary of the grassmann orbitope @xmath912 is derived from the polynomial @xmath913 we rewrite this expression in terms of the @xmath21 unknowns @xmath914 to get an irreducible polynomial of degree @xmath915 with @xmath916 terms . this is the defining polynomial of the hypersurface @xmath917 . for @xmath918 we use the same polynomial in @xmath919 but now there is one more eigenvalue @xmath146 . the defining polynomial of @xmath920 has @xmath921 terms of degree @xmath915 in the @xmath922 matrix entries @xmath914 . we now come to the harder case @xmath923 . the grassmann orbitope @xmath924 is a @xmath925-dimensional convex body . its facial structure was determined by dadok and harvey in @xcite , and independently by morgan in @xcite . we present their well - known results in our language . 1 . one - dimensional faces from pairs of subspaces that satisfy the angle condition ( [ eq : threeangles ] ) ; 2 . three - dimensional faces that arise as @xmath60-orbitopes and these are round @xmath71-balls ; 3 . @xmath774-dimensional faces that are @xmath926-orbitopes , from the lagrangian grassmannian . harvey and morgan @xcite extended these results to @xmath927 . we will here focus on @xmath923 . our first goal is to explain the faces described in theorem [ thm : morgan36 ] , starting with the edges . let @xmath928 and @xmath929 be three - dimensional linear subspaces in @xmath930 with corresponding unit length plcker vectors @xmath133 and @xmath794 . we define @xmath931 to be the minimum angle between any unit vector @xmath932 and any unit vector @xmath933 . next , @xmath934 is the minimum angle between two unit vectors @xmath935 and @xmath936 such that @xmath937 and @xmath938 . finally , we define @xmath939 to be the angle between two unit vectors @xmath940 and @xmath941 such that @xmath942 and @xmath943 . we refer to ( * ? ? ? * lemma 2.3 ) for the fact that @xmath944 and @xmath945 are well defined by this rule . the angle condition referred to in part ( 1 ) of theorem [ thm : morgan36 ] is the inequality @xmath946 the convex hull of @xmath133 and @xmath794 is an exposed edge of @xmath924 if and only if the condition ( [ eq : threeangles ] ) holds . the maximal facets in part ( 3 ) of theorem [ thm : morgan36 ] are known to geometers as _ special lagrangian facets _ , and we represent them as orbitopes as follows . the _ special unitary group _ @xmath926 consists of all complex @xmath426-matrices @xmath434 with @xmath947 and @xmath948 . if @xmath949 and @xmath950 , so that @xmath951 , then we can identify @xmath434 with the real @xmath394-matrix @xmath952 note that this matrix lies in @xmath953 if and only if @xmath954 , @xmath955 , and @xmath956 . hence the transformation @xmath957 realizes @xmath926 as a subgroup of @xmath953 . the orbit @xmath958 is known as the _ special langrangian grassmannian_. this is a five - dimensional real algebraic variety in the unit sphere in @xmath959 . we call its convex hull @xmath960 the _ special lagrangian orbitope_. it is obtained from the @xmath925-dimensional grassmann orbitope @xmath924 by maximizing the linear function @xmath961 , which takes value @xmath70 on that face . the face @xmath962 is @xmath774-dimensional because it satisfies the two affine equations @xmath963 plus the six independent linear equations that cut out the langrangian grassmannian : @xmath964 the exposed faces of type ( 2 ) in theorem [ thm : morgan36 ] are not facets . each of them is the intersection of two special lagrangian facets . for example , consider the two linear functionals @xmath965 which were discussed in @xcite . each of them supports a special lagrangian facet . the intersection of these two @xmath747-dimensional facets is a three - dimensional ball . indeed , the linear functional @xmath966 is bounded above by @xmath70 on @xmath924 , and the subset at which the value equals @xmath70 is contained in the linear span of the six basis vectors @xmath967 where @xmath968 . in the intersection of this linear space with @xmath924 we find the grassmann orbitope @xmath858 from example [ ex : g24 ] , with our linear functional @xmath969 being represented by the scaled coordinate @xmath970 . the face of @xmath858 where @xmath970 attains its maximal value @xmath70 is a @xmath71-ball . this concludes our discussion of the census of faces given in theorem [ thm : morgan36 ] . the first assertion was proved by dadok and harvey in ( * ? ? ? * theorem 7 ) . for the second assertion we proceed as follows . we apply the technique in ( * ? ? ? * lemma 2.2 ) and restrict to the linear subspace @xmath973 of @xmath959 . the intersection of @xmath974 is the @xmath975-orbitope of @xmath976 , where the action is by unitary diagonal @xmath426-matrices , while the intersection @xmath977 is the @xmath978-orbitope of @xmath979 , where the action is by special unitary diagonal @xmath426-matrices , as described in @xcite . we claim that the @xmath978-orbitope @xmath977 is @xmath14-neighborly . this can be seen by examining the bivariate trigonometric polynomials of the form @xmath980 indeed , for any choice of @xmath981 and @xmath982 , we can find eight real coefficients @xmath983 such that @xmath984 and its derivatives vanish at @xmath981 and @xmath982 but @xmath984 is strictly positive elsewhere . the results in @xcite imply that every exposed edge of @xmath974 is also an exposed edge of @xmath924 . we believe that morgan s technique can be adapted to show that _ every _ exposed edge of @xmath977 is an exposed edge of @xmath962 . for the purpose of proving theorem [ thm : g36bad ] , however , we only need to identify _ one _ exposed edge of @xmath977 that is an exposed edge of @xmath962 . the claim that such exposed edges exist can be derived from ( * ? ? ? * theorem 12 ( i ) ) . with the help of philipp rostalski , we also obtained a computational proof of that claim . this was done as follows . we selected various random choices of points @xmath981 and @xmath985 in @xmath63 . each choice specifies two three - dimensional subspaces @xmath928 and @xmath929 of @xmath930 for which equality holds in the angle condition ( [ eq : threeangles ] ) . the corresponding line segment is not an exposed edge of @xmath924 . to show that the line segment between @xmath928 and @xmath929 is exposed in @xmath962 , we use a technique from semidefinite programming . first we compute the eight coefficients @xmath986 of the supporting function @xmath984 as above . this gives a linear function @xmath987 on @xmath959 . we then run a first - order lasserre relaxation ( cf . @xcite ) to minimize this linear function subject to the linear and quadratic constraints that cut out the special lagrangian grassmannian . the optimal value is zero , the optimal lasserre moment matrix has rank two , and its image in @xmath959 lies in the relative interior of the line segment between @xmath928 and @xmath929 . we then re - optimize for various perturbations of the linear function @xmath987 . the output of each run is a rank one moment matrix which certifies either @xmath928 or @xmath929 as optimal solution of the optimization problem . this proves that the face of @xmath962 exposed by @xmath987 is the line segment between @xmath928 and @xmath929 . nicholas mccarthy , david ogilvie , nahum zobin , and veronica zobin , _ convex geometry of coxeter - invariant polyhedra _ , trends in banach spaces and operator theory ( memphis , tn , 2001 ) , contemp . math . 321 , amer . soc . , providence , ri , 2003 , pp . 153179 . kristian ranestad and bernd sturmfels , _ the convex hull of a variety _ , to appear in the julius borcea memorial volume , ( eds . petter brndn , mikael passare and mihai putinar ) , trends in mathematics , birkhuser verlag , arxiv:1004.30 .

gauge / gravity duality with non - relativistic scale invariance has drawn considerable interest in recent years . an interesting direction of studies was based on the schrdinger geometry @xcite . see also @xcite . schrdinger symmetry is a scale invariant extension of the galiean symmetry by the particle number symmetry , anisotropic scale symmetry with dynamical exponent @xmath4 and special conformal symmetry . the geometry of @xcite realizes this symmetry of a qft in @xmath5 spacetime dimension as isometries of a @xmath6 dimensional spacetime of the following form has been studied in @xcite , prior to @xcite . ] : @xmath7\ \ , \ \ \ \vec{x}\in\mathbb{r}^d\ .\ ] ] in particular , the shift symmetry of the coordinate @xmath8 realizes the particle number symmetry . there are many works on schrdinger solutions in string / m - theory , including @xcite , which also discuss schrdinger solutions with general dynamical exponent @xmath9 . see also many references below that we mention in more detailed contexts . @xmath10 dimensional non - relativistic cft duals of 5 dimensional schrdinger geometries ( which we call sch@xmath11 ) were studied in @xcite with string theory embedding , as deformations of @xmath12 dimensional relativistic field theories by irrelevant operators which partly break the 4 dimensional conformal symmetry . see @xcite for more discussions on the holography in this context . on the other hand , a very natural way of obtaining a non - relativistic theory is to start from a relativistic cft living in a spacetime with _ same dimension _ , and to obtain the non - relativistic theory in the ir after a suitable mass deformation and performing the standard non - relativistic limit of redefining the energy and discarding ` anti - particles . ' indeed , it has been shown concretely that a class of @xmath10 dimensional mass - deformed cft gives a non - relativistic theory with schrdinger symmetry @xcite . contrary to the other realization obtained by starting from a relativistic cft in one higher dimension , the schrdinger symmetry can not be embedded into the conformal symmetry of the relativistic theory . it is thus interesting to see if there are mass - deformed cft s admitting gravity duals which show ` holographic rg flows ' from ads@xmath13 to sch@xmath14 . to sch@xmath14 will be clarified later , as the latter lacks @xmath15 dimensional poincare symmetry of ads@xmath13 . we shall obtain sch@xmath11 after combining an internal direction with four directions , performing the coordinate transformation ( [ coordinate - transform ] ) to have a time @xmath16 conjugate to the non - relativistic energy , and finally taking the scaling limit ( [ scaling ] ) . ] the case with @xmath17 will be our main interest in this paper , although similar studies could be done in other dimensions . in particular , @xcite discussed the non - relativistic superconformal field theory obtained by deforming the @xmath18 scft @xcite by a mass term preserving all poincare supersymmetries , considering a particular vacuum in which all scalar expectation values are zero . the classical lagrangian of this non - relativistic field theory preserves @xmath19 supercharges , including @xmath20 extra non - relativistic conformal supercharges . there have been some efforts to construct the gravity dual of this theory in the ir based on the geometric realization of schrdinger symmetry @xcite . none of them found it , at least with all @xmath19 killing spinors . soon it was argued @xcite with a witten index calculation that the corresponding field theory vacuum does not exist , at least in the subset of supersymmetric vacua . the gravity dual of this theory is still ill - understood to date . see more comments about it in the discussion . in this paper , we find a very simple class of flow solutions from ads@xmath0 to sch@xmath11 after deforming a class of relativistic cft@xmath21 with a mass operator . the 3 dimensional relativistic cft@xmath21 can be regarded as those living on parallel m2-branes in m - theory . our construction is based on the relativistic uv cft dual to the so - called skew - whiffed @xmath22 , where @xmath23 denotes 7 dimensional sasaki - einstein manifold . the skew - whiffing in this context corresponds to changing the sign of m2-brane charge , generically breaking supersymmetry . for the particular case of @xmath24 , maximal superconformal symmetry is preserved and the uv cft is relatively well understood . in this case , we explicitly identify the mass operator , which turns out to be non - supersymmetric and invariant under an @xmath25 subgroup of @xmath26 . our finding shows how schrdinger geometries can appear by deforming relativistic cft s by mass operators and taking the non - realtivistic limit . it could find applications in many different gauge / gravity dual models containing massive degrees only . we briefly comment on some possible applications in the discussion . the solution we consider in this paper can actually be understood as the zero - temperature black brane solution found and discussed in @xcite from 4 dimensional gauged supergravity . the 4 dimensional metric in the einstein frame in the ` ir regime ' ( with small @xmath27 ) is given by @xmath28 which has zero entropy . this type of geometry with poincare symmetry and broken ( isotropic ) scale invariance was considered in @xcite , and a similar system was considered in @xcite . this type of geometry is different from those considered in @xcite which have approximate lifshitz symmetry . from our model , it is clear that the broken isotropic scale invariance originates from the mass deformation of the uv cft . the remaining part of this paper is organized as follows . in section 2 , we explain our ansatz and obtain the analytic solution in the ir in the small radius expansion . we explain the gravity dual operation of taking the non - relativistic limit of qft , the appearance of sch@xmath11 geometry , and the exact solution which interpolates our ir solution with asymptotic ads@xmath0 . in section 3 , we explain the mass term of the dual field theory when the internal 7-manifold is @xmath29 or orbifolds thereof . section 4 concludes with discussions . we seek for flow solutions from @xmath22 to 5 dimensional schrdinger geometries in the consistent truncation ansatz discovered in @xcite . it will turn out that we can realize a schrdinger solution in the infrared if we advocate the so - called skew - whiffed truncation . this truncated theory contains an @xmath22 vacuum , generically with broken supersymmetry . for the special case of @xmath24 , the @xmath30 vacuum is maximally supersymmetric . for the case with @xmath31 orbifold , the @xmath30 vacuum after orbifold is non - supersymmetric with skew - whiffing . the nature of the mass term driving the flow of our solution will be discussed in section 3 after studying the gravity solutions . the full solution is actually available in an analytic form @xcite , but to illustrate how the solution looks like and to explain why skew - whiffed setting is necessary to have nontrivial solutions , we explain some details of the derivation . the consistent truncation ansatz of @xcite is @xmath32\ , \end{aligned}\ ] ] where @xmath33 is the khler 2-form of the khler - einstein base @xmath34 , @xmath35 is the reeb 1-form with @xmath36 , @xmath37 . @xmath38 , @xmath39 , @xmath40 are 1-form , 2-form , 3-form in 4 dimensions , while @xmath41 and @xmath42 are real / complex scalars , respsectively . the truncation with @xmath43 contains supersymmetric and generically non - supersymmetric @xmath30 vacuum . the case with @xmath44 is the skew - whiffed truncation . we would like to study the flow solutions preserving the @xmath10 dimensional poincare symmetry on the m2-brane world - volume . this is because the flow itself after a mass deformation does not break poincare symmetry on the worldvolume . it is only the non - relativistic limit , discarding ` anti - particles ' after suitably redefining non - relativistic energy , which breaks it . we thus consider the poincare invariant ansatz @xmath45 from poincare symmetry , the gauge fields @xmath39 , @xmath46 are zero . at this stage , it appears that @xmath40 may be nonzero if proportional to the volume form @xmath47 . from the equations of motion and the bianchi identities for @xmath48 , one obtains @xmath49 and @xmath50 from the @xmath39 equation of motion on the second line , one should set @xmath51 . so either @xmath52 or @xmath40 should be vanishing . as we shall see later , nonzero @xmath52 is crucial to have a schrdinger solution in the ir , similar to those in @xcite . so we set @xmath53 . the field @xmath42 can always be consistently set to zero , which we do from now on . then , only the @xmath52 equation of motion ( third line ) remains nontrivial among the above equations . as we shall explain shortly , we want an ir solution with @xmath54 , @xmath55 as @xmath56 , to get 5 dimensional schrdinger solutions . it turns out that one can set @xmath57 , which is the value for the ads@xmath0 vacuum , all along the flow . we shall later explain that this is a consistent ansatz , which actually comes from the existence of a further consistent truncation found in @xcite . with @xmath58 , the 11 dimensional einstein equation becomes @xmath59 with @xmath60 . the last equation is the @xmath61 equation of motion . note that @xmath62 . with our 4 dimenisional metric ( [ 4d - metric ] ) , one obtains @xmath63 for @xmath64 , where prime denotes @xmath27 derivative . therefore , the two nontrivial components ( @xmath65 and @xmath66 ) of the einstein equations and the @xmath67 equations become @xmath68\ .\end{aligned}\ ] ] apparently , there are 5 equations for 3 functions @xmath69 . one more equation appears due to our poincare invariant ansatz with @xmath27 dependent functions only , and another one due to our ansatz @xmath58 . the former extra equation will be eliminated from the consistency of poincare invariant ansatz , and the second one from the consistency of @xmath57 ansatz . we first explain some structures of these equations . as briefly explained above , we want @xmath70 for some constant @xmath71 as @xmath56 . then the terms containing @xmath72 and @xmath73 in the first three equations all yield @xmath74 , which are of order @xmath75 . also , the last factors @xmath76 become very large due to @xmath77 , unless @xmath78 is small . especially , in the second equation , considering the signs of other terms , this term can not be canceled at all if it is indeed large . the only way to have the desired ir solution is to take @xmath79 with skew - whiffing and demand @xmath80 as @xmath56 . in section 3 , we will also provide a group theoretical argument from uv perspective on why the desired flow solutions exist only with skew - whiffing in this truncation . firstly , the equation for @xmath61 becomes to @xmath81[e^v+(-1+h^2)e^{-v}]\ .\ ] ] this relates @xmath52 and @xmath82 in an algebraic manner . it will turn out that @xmath83 is the solution that we want . it is a special case of an algebraic relation found in the truncation of @xcite with @xmath84 . we are left with 4 equations for 2 variables @xmath85 . one can easily show that the @xmath52 equation is equivalent to the @xmath86 equation . to show this , we multiply @xmath52 to the last equation of ( [ 5-equations ] ) and obtain @xmath87\ ] ] after inserting ( [ h - v ] ) . as the expression in the last parenthesis is the @xmath86 equation , the two equations are equivalent as long as @xmath88 . in our flow solution , @xmath52 will only be asymptotically zero in uv , where we can easily check separately that the uv fixed point solves all equations . so the last constraint will not obstruct the above elimination of one equation . so far , the three equations for the two variables @xmath85 ( with ( [ h - v ] ) assumed ) are @xmath89 where the last equation may be replaced by @xmath90\ , \ ] ] where @xmath91 $ ] denotes the expression in the @xmath86 equation of motion , on the last line of ( [ 3-eqns ] ) . also , we denote by @xmath92 $ ] and @xmath93 $ ] the expressions on the first and second lines of ( [ 3-eqns ] ) . to show that one of these three equations is redundant , we find it useful to consider the following two combinations of three independent equations : @xmath94-\frac{1}{3}[v]&:&0=a^{\prime\prime}+(a^\prime)^2-a^\prime v^\prime-4e^{2v}\\ \left[r_{rr}\right]-[r_{\mu\nu}]&:&2a^{\prime\prime}+v^{\prime\prime}+(v^\prime)^2 -a^\prime v^\prime+\frac{3}{2}e^{-2v}(h^\prime)^2\ .\nonumber\end{aligned}\ ] ] let us take 3 combinations of equations to be @xmath95-[v]}{3}=0 $ ] , @xmath93=0 $ ] and @xmath96\right)-2e^{2v}a^\prime\left([r_{rr}]-[r_{\mu\nu}]\right ) -2e^{2v}v^\prime\left(\frac{1}{3}[r_{rr}]-\frac{1}{3}[v]\right)\nonumber\\ & = & e^{2v}\left[\frac{1}{2}e^{-2v}(h^\prime)^2 - 2a^\prime v^\prime-2(a^\prime)^2\right]^\prime.\end{aligned}\ ] ] the last total derivative can be obtained after some straightforward algebra . so this equation can be integrated to yield @xmath97-\frac{1}{3}[v]\right)-[r_{\mu\nu}]\ , \ ] ] where the last equation holds for @xmath98 . thus , the first two equations guarantee this last equation with the integration constant chosen as @xmath98 . so we have two independent equations for two functions in our ansatz . to get ir solutions near @xmath99 , we define @xmath100 with a positive number @xmath71 and expand @xmath101 we can set @xmath102 without losing generality by shifting the variable @xmath27 , as all equations have translation symmetry in @xmath27 . @xmath103 can also be eliminated to , say @xmath104 , by rescaling the poincare coordinates @xmath105 . we leave it as it is as we want to identify it with a mass parameter in the uv later . plugging these in , we find @xmath106 at @xmath107 terms of the equations . continuing the iteration , one finds the following small radius expansion in @xmath108 @xmath109 with @xmath57 and @xmath110 . the solution with leading and next - to - leading terms is given by @xmath111 where vol@xmath112 . to obtain the schrdinger solution , we first perform a coordinate transformation to a rotating frame . taking @xmath35 , where @xmath36 is twice the khler 2-form of @xmath34 , we define @xmath113 to be the coordinates in the rotating ( or boosted ) frame . for later use , we note that @xmath114 is @xmath115 periodic for @xmath116 in our normalization . from the dual field theory perspective , the shift of @xmath114 is identified as the particle number symmetry , or more precisely the symmetry conjugate the total rest mass as there could be more than one particle species . so the conserved charge for @xmath16 takes the form of the non - relativistic energy @xmath117 , where @xmath118 and @xmath119 are conserved charges for the translation of @xmath120 . the above metric can be written as @xmath121 note that @xmath122 terms cancels by the coordinate transformation , leaving the leading @xmath123 term . taking the scaling limit @xmath124 with @xmath125 , one obtains @xmath126 the 5 dimensional metric apart from the @xmath34 part is the schrdinger geometry dual to @xmath10 dimensional non - relativistic field theory . the coordinate transformation and scaling explained above are to be understood as the gravity dual of the non - relativistic limit . sch@xmath11 solutions in m - theory of this type was found in @xcite . however , the solution found there has different coefficients in various terms . in fact , one can find a generalized solution @xmath127 which includes our solution ( [ ir - solution ] ) and that of @xcite as special cases . as before , one can eliminate one of the two parameters @xmath128 by scaling @xmath129 in same ratio , after which one parameter remains . setting @xmath130 yields our solution ( [ ir - solution ] ) , while setting @xmath131 yields that of @xcite . ( with @xmath132 which is our normalization ) , correcting a minor typo in their expression @xmath133 . relating their quantities with ours as @xmath134 , @xmath135 , @xmath136 , @xmath137 with same @xmath16 , their eqn.(4.4 ) takes the form of ( [ generalized ] ) above . ] the ` leading behavior ' of the ir solution ( [ leading ] ) before coordinate redefinition , keeping @xmath138 only , might look like @xmath139 fibered over @xmath34 with a compact longitudinal direction along @xmath114 . however , this @xmath139 fibered over @xmath34 _ does not _ satisfy the equations of motion ( [ 5-equations ] ) itself , although making the leading @xmath107 equations satisfied . namely , there are subleading terms in the equations caused by the leading terms of the functions @xmath140 , to be canceled by the subleading terms of the functions . there exist @xmath141 solutions in m - theory without any fibration in the @xmath114 angle , which is related to our ` leading behavior ' by replacing @xmath142 with same radii for @xmath139 and @xmath34 @xcite . another point to mention is that this @xmath139 can not be obtained by a scaling limit similar to the one we explained above for @xmath143 . naively , it might appear that the scaling @xmath144 , @xmath145 , @xmath146 , @xmath147 with @xmath148 would do this job . however , the last @xmath114 direction is compact . this scaling limit demands one to zoom out to the long wavelength along @xmath149 , which does not make sense with a compact direction with a maximal wavelength . note that the scaling ( [ scaling ] ) to the schrdinger geometry leaves @xmath8 unchanged . before studying the full flow solution which interpolates the above ir solution with ads@xmath0 uv fixed point , we review the asymptotic expansion in uv @xcite . we obtain @xmath150 with two possible solutions for the coefficients @xmath151 defining @xmath152 , one obtains the asymptotic ads@xmath0 metric @xmath153 upon defining @xmath154 . the general asymptotic solution is @xmath155 with @xmath156 and @xmath157 . both modes are normalizable with these coefficients and there are two possible dual cft s in which one mode is the source of a boundary operator whose expectation value is given by the coefficient of the other mode . we stick to the cft in which the field @xmath52 sources the fermion mass term at the boundary as we review in detail in section 3 , in which the mode with @xmath158 is taken to be the source . the fermion mass operator comes with dimension @xmath159 . from the gravity solution , one can get the precise values of the fermion and boson masses , which we explain in section 3 . . the thick / thin lines are plots of @xmath160 and @xmath82 , respectively . @xmath82 approaches @xmath75 as @xmath161 ( the value for the @xmath30 vacuum).,width=302 ] we now study if the ir solution can be interpolated to the asymptotic ads@xmath0 solution at @xmath162 . we chose the first equation of ( [ basis ] ) and the @xmath52 equation ( [ h - equation ] ) for numerics . the numerical analysis of the differential equations exhibits the desired interpolation . the result for @xmath163 is shown in fig [ flow ] . one can also vary the values of @xmath103 and numerically relate it with the variable @xmath164 in the uv expansion . in doing so , we should remember that both in the ir and uv solutions we implicitly tuned a coefficient by using the translational symmetry of @xmath27 in the equations . in matching the two solutions , only one tuning is allowed . so we relax our uv asymptotic behavior by admitting the @xmath27 translation degrees and write the solution as @xmath165 etc . , where @xmath166 is the same coefficient as above . we fit the two coefficients @xmath167 by comparing the expansion ( [ uv - coefficients ] ) with the numerical solution at two large values of @xmath27 , which we chose as @xmath168 , and find @xmath169 , @xmath170 as functions of @xmath103 . we find the relation to be @xmath171 @xmath172 appearing as the leading coefficient of @xmath52 in uv is the mass for fermions , consistent with the ir expectation from ( [ coordinate - transform ] ) that @xmath103 should be proportional to the mass parameter . surprisingly , one can find an exact analytic solution for the whole flow @xcite , @xmath173 where the subscript @xmath174 denotes the 4 dimensional metric in the einstein frame . ( note that their @xmath27 above is not our @xmath27 used previously . ) it is related to our 4 dimensional metric as @xmath175 . we managed to check this exact solution by a heavy use of mathematica . we also checked that our ir expansion ( [ small - radius ] ) is completely reproduced from the above exact solution , after relating our ir radial variable @xmath176 with their @xmath27 as @xmath177 according to @xmath178 . when the 7 manifold is @xmath24 or its @xmath31 orbifolds , we can identify the mass term which drives our flow . at @xmath179 , the m2-brane worldvolume scalars and fermions transform under @xmath26 as a vector @xmath180 and a chiral spinor @xmath181 , respectively . the @xmath182 supercharge as well as the conformal supercharge transform as @xmath183 . our skew - whiffed truncation ansatz has a preferred @xmath184 symmetry @xcite : in particular , the ansatz is invariant under @xmath185 isometry of the @xmath186 base . the three @xmath26 representations decompose under @xmath184 as @xmath187 the charges of @xmath188 are chosen so that @xmath189 is possible , correcting a minor sign error in @xcite . the m2-brane theory without skew - whiffing can be obtained by a parity transformation on @xmath190 , interchanging @xmath181 and @xmath183 representations . after this change , the supercharges in @xmath191 of @xmath181 are the @xmath18 supersymmetries of @xcite . with skew - whiffing , all the supercharges transforming in @xmath192 are broken for @xmath193 . at @xmath179 , monopole operators with @xmath188 charge @xmath194 make them gauge - invariant @xcite , enabling maximal supersymmetry . we would like to explain the nature of the mass term triggering our flow . the mass term should be invariant under the @xmath185 symmetry of the @xmath195 in our truncation , in the skew - whiffed frame . decomposing @xmath181 ( matter fermions ) into @xmath196 , we call them @xmath197 , @xmath198 , @xmath199 with @xmath200 , following @xcite . the fermions @xmath198 ( @xmath200 ) in @xmath191 are majorana spinors , proposed to be in the adjoint representation of the first @xmath201 , while the complex fermion @xmath197 in @xmath202 is in the bi - fundamental representation of @xmath203 . it seems that the parity symmetry of this system is not manifest in the field theory . in our discussions below , the role of these gauge group representations will be rather minor . the mass for real fermions @xmath204 , with @xmath205 in @xmath181 of @xmath26 , is given by ( at least on a single m2-brane ) @xmath206 with an @xmath207 symmetric mass matrix @xmath208 . the symmetric matrix has @xmath209 real components and decomposes under @xmath26 as @xmath210 , where @xmath211 is the ` traceless ' part with respect to the charge conjugation matrix , and @xmath212 is the ` identity ' part . the spinor bi - linear in @xmath211 of @xmath26 is equivalent to self - dual 4-forms in @xmath190 , and will turn out to be dual to the self - dual 4-form modes in the gravity dual @xcite as we explain in some detail below . for now , we just mention that the last two terms on the second line of ( [ truncation ] ) with nonzero @xmath52 provides the relevant 4-form flux . the mass matrix decomposes under @xmath185 as @xcite @xmath213 more concretely , the possible mass terms ( on a single m2-brane ) look like @xmath214+\frac{m_{ij}}{2}\psi^i\psi^j\ ] ] with real masses @xmath215 and @xmath216 , real traceless symmetric mass matrix @xmath217 , complex masses @xmath218 , @xmath219 . all but one combination of the first two mass terms proportional to @xmath215 , @xmath216 belong to @xmath211 . a combination of the first two becomes an @xmath26 invariant . this can not be dual to the 4-form flux that we turn on in our solution , as 4-form fluxes in @xmath190 always break @xmath26 . so we do not consider this combination . the other ` traceless ' combination @xmath220 , with canonically normalized kinetic terms for the fermions , corresponds to @xmath221 in the above decomposition of @xmath211 . this is our mass term , as our gravity solution is invariant under @xmath184 . for multiple m2-branes , the first two mass terms proportional to @xmath222 can be made gauge invariant by taking traces , as @xmath198 is an adjoint and @xmath223 are in conjugate representations . we emphasize that , as fermions are in @xmath183 without skew - whiffing , the mass matrix , now given by anti - self - dual 4-forms , decomposes under @xmath185 as @xmath224 without singlets . this explains why our consistent truncation only has solutions with skew - whiffing . the bosonic mass term is also @xmath185 invariant , which is unique up to coefficient : @xmath225 with @xmath226 . this mass term is actually invariant under @xmath26 , at least for the single m2-brane for which we can decompose complex scalars to real scalars in @xmath180 . even at @xmath179 in which the uv cft is supersymmetric , it is easy to see that the mass - deformed theory is always non - supersymmetric . this is simply because the mass is same for all bosonic fields , while there appear two different nonzero masses for fermions . one can compute the precise values of the fermion / boson masses with the known form of the flux . since we have a cft before mass deformation , physically meaningful quantity is just the ratio @xmath227 of the two masses . to conveniently match factors , it is useful to compare the normalization with the well known maximally supersymmetric mass deformation . as a warming up , let us first consider an illustrative example of single m2-brane coupled to a background 4-form field @xmath228 in ( almost ) flat space . we shall later explain that the same result is obtained from the large @xmath229 calculation in @xmath230 . the coupling to this field is given by @xmath231_{012}++\beta\bar\psi\gamma^{ijkl}\psi[g_4]_{ijkl } + \cdots\ , \ \ \ \ ( i , j , k , l=1,2,\cdots,8)\ ] ] with coefficients @xmath232 determined in @xcite . we turn on small constant self - dual @xmath48 on @xmath190 , where smallness is to keep the back reaction to the flat space to be small . the bosonic mass @xmath233 is related to the constant flux as @xcite @xmath234 this comes partly from the first term of ( [ wv - fermion - mass ] ) after the constant self - dual 4-form back - reacts to @xmath235 , and also partly from the back - reaction to the metric . for the two cases @xmath236 and ( @xmath237 below is the kahler 2-form on @xmath190 , not on @xmath195 ) @xmath238 on @xmath190 , one obtains @xmath239 respectively . all quantities with the subscript @xmath240 are for the case with the maximally supersymmetric mass deformation , to be compared with the case of our interest . in each case , expanding the fermion mass term in ( [ wv - fermion - mass ] ) , one obtains @xmath241 where the last two masses are those for the complex fermion @xmath197 in @xmath242 and the real fermions @xmath243 in @xmath191 . their mass ratio @xmath244 is what we anticipated from group theory . in the former case , since the masses for bosons and fermions are equal due to supersymmetry , @xmath245 can be used to conveniently fix the normalization as @xmath246 . inserting this back , one obtains @xmath247 these masses happen to satisfy a supertrace constraint @xmath248 . and the relation ( [ boson - mass ] ) to the boson mass , one can show @xmath249 . ] now we explain how one can obtain the same masses from the asymptotic @xmath230 gravity solution in the large @xmath229 limit . contrary to the simple case with single m2-branes , which is described by a free quantum field theory , the masses that we calculate below with the large @xmath229 gravity solution is the bare mass , or the uv mass , of the field theory . since the mass - deformed field theory preserves no supersymmetry , there can be nontrivial quantum corrections for the actual mass of particles in the ir . the latter masses will be important quantities when one discusses the non - relativistic limit from the field theory . the procedure of obtaining fermionic masses @xcite is similar to the above calculation for single m2-branes . the 4-form @xmath250 in @xmath190 defined by @xmath251 and @xmath252 $ ] plays the role similar to the constant 4-form which appeared in the mass matrix for m2-branes in flat space , where @xmath253 is the radius of 7-sphere which we set @xmath75 in our normalization . @xmath48 above is the flux on the @xmath29 and the radial part only . working with the asymptotic solution of the form , @xmath254 one obtains @xmath255 . the resulting fermion masses are @xmath256 as for the overall normalization , we matched with the convention used in @xcite for the maximally supersymmetric mass deformation , in which @xmath257 from the gravity solution leads to the boson / fermion masses @xmath258 . the bosonic mass can be quickly computed if one takes a probe m2-brane in the mass - deformed uv solution . using the radial coordinate @xmath27 appearing in ( [ uv - coordinate ] ) , one obtains a quadratic potential in @xmath27 after partly canceling the contributions from the nambu - goto and wess - zumino terms . in the maximally supersymmetric case and for our case , we obtain @xmath259 and @xmath260 , respectively , consistent with ( [ single - mass ] ) . suppose we have a system with particles having various values of mass and global @xmath188 charge . we consider non - relativistic limit with the @xmath188 symmetry as we did in our gravity solution . the bosonic and two fermionic fields have charges @xmath261 , @xmath20 , @xmath240 , respectively from ( [ skew - whiffed - rep ] ) . there could be other particles coming from bound states carrying nonzero global charge . we take the non - relativistic energy to be @xmath262 , where @xmath118 is the relativistic energy and @xmath263 is some constant . the constant @xmath263 has to be chosen such that the condition @xmath264 is met for all particles , while being saturated by a subset of them . from the gravity solution , as we made a coordinate transformation @xmath265 , @xmath266 , the conserved charge for @xmath16 is given by @xmath117 , with @xmath267 . the non - relativisitic field theory only keeps a sector of particles which satisfy @xmath268 when they are at rest . on the other hand , it discards at low energy all the other particles with @xmath269 at rest , as such particles have large non - relativistic energies . this is exactly same as discarding anti - particles with non - relativistic energy @xmath270 . so we have @xmath271 for some constant @xmath263 , where the modes which saturate the inequality survive the non - relativistic limit . therefore , the non - relativistic theory would be keeping the particles with the largest value of @xmath272 , using the ir masses of the particles . let us start from the theory on a single m2-brane , described by a free qft . considering the three elementary particles only , one obtains the ratio @xmath273 thus , on single m2-branes , the @xmath197 mode survives in the non - relativistic theory if one uses @xmath188 as the particle number charge . for the case with large @xmath229 , had all the uv masses been the physical masses of elementary particles without any quantum corrections , one would have obtained the same ratio as the single m2-brane . the actual values for @xmath274 from our gravity solution are @xmath275 , @xmath276 from ( [ uv - modes ] ) and ( [ uv - ir - relation ] ) , which are both larger than @xmath267 which we obtained from the coordinate transformation . there could be two possible reasons for this . it could be that the physical masses for these particles in ir acquire nontrivial quantum corrections causing smaller mass - to - charge ratios : at least nonzero quantum corrections are natural for our non - supersymmetric theories . and/or , there can also appear bound states of these particles with nonzero binding energies , lowering the ratio . as the schrdinger geometry comes with an infinite tower of kaluza - klein states on @xmath8 , representing infinitely many particle species @xcite , the last issue of bound states should anyway be an essential ingredient for understanding this system . it should be interesting to have a better understanding on these quantum aspects of the spectrum . in this paper , we considered the gravity solutions for a class of mass - deformed cft s in 3 dimensions , and showed that there exist 5 dimensional schrdinger geometries in the ir after taking the non - relativistic limit with a global @xmath277 symmetry . as far as we are aware of , this is the first occasion in which non - relativistic conformal geometries are explicitly derived by taking the natural non - relativistic limit of massive relativistic systems . one could think of some generalizations and extensions of this idea to various directions . firstly , one may ask if a similar story holds for the maximally supersymmetric mass deformation of the same uv theory . as the 4-form flux preserves @xmath278 , the flow solution would have two round 3-spheres . we should be generalizing @xcite and seek for a solution with broken supersymmetry if the witten index calculation of @xcite is implying that the vacuum spontaneously breaks supersymmetry . let us try to see if a mechanism explained in this paper could be working , namely , if the ` particle number circle ' can shrink in ir to be part of the 5 dimensional schrdinger geometry . as the @xmath277 symmetry is taken from the two round 3-spheres in @xcite , any shrinking circle necessarily demands that the whole 3-spheres shrink as well , presumably making the ir geometry very singular . it may also be worth considering the possibility that this vacuum could be metastable , or does not exist at all . as explained before , the nonrelativistic cft of @xcite is based on the ` symmetric vacuum ' of the supersymmetric mass - deformed theory , with zero scalar vev . the vacuum preserves supersymmetry only when @xmath279 . its gravity dual will be highly curved , as the t hooft coupling is small . still , formal gravity solutions with large curvature are identified in @xcite . it may be possible to learn something useful from these solutions . let us also mention at this point that , from the viewpoint of the non - relativistic limit of mass - deformed cft , the appearance of schrdinger geometries as the gravity duals of non - relativistic cft s seems to demand a strong restriction . to illustrate this point , note that there are many ways of obtaining non - relativistic theories with a given mass - deformed cft if there exist more than one global @xmath277 symmetries , simply by choosing different @xmath277 s as particle number symmetries . indeed , for the maximally supersymmetric mass deformation , different non - relativistic cft s obtained by choosing different @xmath277 s in the @xmath278 global symmetry were studied @xcite . in our mass deformed theory , one can also consider using a @xmath277 in the @xmath185 isometry as the particle number symmetry , rather than @xmath188 along the @xmath114 or @xmath8 fiber of @xmath23 as we did in this paper . note that a crucial ingredient which allowed sch@xmath11 of this paper was the shrinking circle for @xmath188 symmetry , @xmath280 , in ir . this can not happen for @xmath281 as the @xmath34 part of the metric is finite in ir . this seems to be implying that , for the geometry of @xcite to emerge as the gravity dual of schrdinger invariant qft obtained from mass - deformed relativistic cft , there should be nontrivial conditions on the quantum dynamics . it would be interesting to further explore this issue . our solutions reduced down to 4 dimensional spacetime of the gauged supergravity may not look so well - behaved @xcite . it should be interesting to get a low dimensional intuition for the solutions with schrdinger symmetry to explore a more general class of such solutions , maybe using attractor mechanisms @xcite . similar studies have been done in @xcite , which attempts to realize the particle number symmetry without using internal isometries . following the idea of this paper , it might be possible to find flow solutions from ads@xmath11 to sch@xmath282 by studying the consistent kaluza - klein truncation of @xcite based on @xmath283 . there also exists a holographic flow solution for the 4 dimensional @xmath284 yang - mills theory with an @xmath285 supersymmetric mass deformation , which is singular in the ir @xcite . perhaps it might be worth mentioning that the original motivation in @xcite of using this geometry to holographically study cold fermionic atoms at unitarity may be demanding alternative realizations of the schrdinger symmetry without using the isometry of @xmath8 as the particle number . this question has been raised due to an exotic thermodynamics of schrdinger black holes @xcite , probably caused by the presence of an infinite tower of kaluza - klein particle species @xcite . of course our system would be showing same phenomena , simply due to the appearance of the geometry of the form ( [ schrodinger ] ) . the non - relativistic limit of @xcite is based on using the d0-brane charge ( geometrizable to m - theory isometry ) as the particle number , which is also the case for our system . the non - relativistic cft discussed in @xcite essentially has this @xmath8 direction as it is obtained from a @xmath286 dimensional qft in the uv , where the @xmath8 isometry simply comes from the extra direction of the uv field theory . on the other hand , non - relativistic qft s obtained by the standard non - relativistic limit of mass - deformed qft do not necessarily have such an infinite tower of bound states . from the field theory perspective , the special example of m2-brane cft discussed in this paper seems to be rather exceptional , containing geometrizable d0-brane global charge and exhibiting bound states for all d0-brane numbers . thus , it could be possible to seek for a study along the line of @xcite , to have a gravity solution invariant under schrdinger symmetry with non - geometrizable particle number symmetry in the context of the mass - deformed cft . we would also like to comment that in @xcite , the phases of skew - whiffed field theory with relevant deformations ( i.e. coupled to nonzero @xmath52 field in the gravity dual ) have been studied extensively from the gravity dual , after considering more modes than what we did in this paper . more precisely , a charged scalar and a massless gauge field were kept , apart from @xmath52 . in particular , there is a phase transition which appears at nonzero chemical potential @xmath287 and mass , the latter one being the mass that we considered in this paper . when @xmath287 is sufficiently larger than the mass , the system is shown to be in a superconducting phase with spontaneously broken @xmath277 symmetry . on the other hand , if the chemical potential is small compared to the mass , the system is supposed to be in a normal phase which is gapped , without the goldstone boson associated with spontaneously broken @xmath277 . in particular , the line @xmath288 at zero temperature on the parameter space that we have considered all belongs to this normal phase . at zero temperature in the normal phase , with all the propagating degrees being massive , one may ask if there are interesting non - relativistic solutions like what we considered on the @xmath288 line . although the numerical solution was explained to be singular at zero temperature , this could be physically acceptable . in string theory , there are examples of four - dimensional extremal black holes described by singular geometries , yet being supersymmetric @xcite . furthermore if we are interested in condensed matter systems with unique ground states , the dual geometries have zero entropy , which often lead to singular geometries . also , uplifting the 4 dimensional geometry to 11 dimensions could reveal more interesting structures , as was the case on the @xmath288 line we studied in this paper . finally , it could be desirable to investigate the stability of the solutions discussed in this paper . an instability was reported for a solution in the skew - whiffed truncation with nonzero values of the charged field @xcite . although the solution we discussed is more complicated than the ads solution studied there , at least studies of the ir solution ( [ ir - solution ] ) would be doable and worthwhile . perhaps this problem could be related to the stability analysis of a class of @xmath141 solutions @xcite . we would like to thank aleksey cherman , aristomenis donos , shamit kachru , sangmin lee , sungjay lee , eoin o colgain , sandip trivedi , daniel waldram , toby wiseman , jackson wu and especially jerome gauntlett and julian sonner for helpful discussions . this work is supported by the bk21 program of the ministry of education , science and technology ( hk , sk ) , kosef grant r01 - 2008 - 000 - 20370 - 0 ( jp ) , the national research foundation of korea grants no . 2006 - 0093850 ( kl ) , 2009 - 0084601 ( kl ) , 2009 - 0085995 ( jp ) , 2010 - 0007512 ( hk , sk ) , 2009 - 0072755 , 2009 - 0084601 ( hk ) and 2005 - 0049409 through the center for quantum spacetime ( cquest ) of sogang university ( kl , jp ) . sk thanks the theory group of imperial college london and damtp for hospitality during his visit , where this work was completed . jp appreciates apctp for its stimulating environment for research . d. t. son , `` toward an ads / cold atoms correspondence : a geometric realization of the schrodinger symmetry , '' phys . rev . d * 78 * , 046003 ( 2008 ) [ arxiv:0804.3972 [ hep - th ] ] . k. balasubramanian and j. mcgreevy , `` gravity duals for non - relativistic cfts , '' phys . * 101 * , 061601 ( 2008 ) [ arxiv:0804.4053 [ hep - th ] ] . c. duval , g. w. gibbons and p. horvathy , phys . d * 43 * , 3907 ( 1991 ) [ arxiv : hep - th/0512188 ] ; c. duval , m. hassaine and p. a. horvathy , annals phys . * 324 * , 1158 ( 2009 ) [ arxiv:0809.3128 [ hep - th ] ] . d. israel , c. kounnas , d. orlando and p. m. petropoulos , fortsch . * 53 * , 73 ( 2005 ) [ arxiv : hep - th/0405213 ] ; d. orlando , s. reffert and l. i. uruchurtu , j. phys . a * 44 * , 115401 ( 2011 ) [ arxiv:1011.1771 [ hep - th ] ] . s. a. hartnoll and k. yoshida , jhep * 0812 * , 071 ( 2008 ) [ arxiv:0810.0298 [ hep - th ] ] ; a. donos and j. p. gauntlett , jhep * 0903 * , 138 ( 2009 ) [ arxiv:0901.0818 [ hep - th ] ] ; jhep * 0907 * , 042 ( 2009 ) [ arxiv:0905.1098 [ hep - th ] ] ; jhep * 0910 * , 073 ( 2009 ) [ arxiv:0907.1761 [ hep - th ] ] ; n. bobev , a. kundu and k. pilch , jhep * 0907 * , 107 ( 2009 ) [ arxiv:0905.0673 [ hep - th ] ] ; e. o colgain and h. yavartanoo , jhep * 0909 * , 002 ( 2009 ) [ arxiv:0904.0588 [ hep - th ] ] . c. p. herzog , m. rangamani and s. f. ross , jhep * 0811 * , 080 ( 2008 ) [ arxiv:0807.1099 [ hep - th ] ] ; j. maldacena , d. martelli and y. tachikawa , jhep * 0810 * , 072 ( 2008 ) [ arxiv:0807.1100 [ hep - th ] ] ; a. adams , k. balasubramanian and j. mcgreevy , jhep * 0811 * , 059 ( 2008 ) [ arxiv:0807.1111 [ hep - th ] ] . m. guica , k. skenderis , m. taylor and b. c. van rees , jhep * 1102 * , 056 ( 2011 ) [ arxiv:1008.1991 [ hep - th ] ] ; p. kraus and e. perlmutter , jhep * 1105 * , 045 ( 2011 ) [ arxiv:1102.1727 [ hep - th ] ] . y. nakayama , m. sakaguchi and k. yoshida , jhep * 0904 * , 096 ( 2009 ) [ arxiv:0902.2204 [ hep - th ] ] . k. m. lee , s. lee and s. lee , jhep * 0909 * , 030 ( 2009 ) [ arxiv:0902.3857 [ hep - th ] ] . o. aharony , o. bergman , d. l. jafferis and j. maldacena , jhep * 0810 * , 091 ( 2008 ) [ arxiv:0806.1218 [ hep - th ] ] . h. ooguri and c. s. park , nucl . phys . b * 824 * , 136 ( 2010 ) [ arxiv:0905.1954 [ hep - th ] ] . j. jeong , h. c. kim , s. lee , e. o colgain and h. yavartanoo , jhep * 1003 * , 034 ( 2010 ) [ arxiv:0911.5281 [ hep - th ] ] . h. c. kim and s. kim , `` supersymmetric vacua of mass - deformed m2-brane theory , '' nucl . b * 839 * , 96 ( 2010 ) [ arxiv:1001.3153 [ hep - th ] ] . s. cheon , h. c. kim and s. kim , `` holography of mass - deformed m2-branes , '' arxiv:1101.1101 [ hep - th ] . j. p. gauntlett , j. sonner and t. wiseman , `` quantum criticality and holographic superconductors in m - theory , '' jhep * 1002 * , 060 ( 2010 ) [ arxiv:0912.0512 [ hep - th ] ] . c. charmousis , b. gouteraux , b. s. kim , e. kiritsis and r. meyer , jhep * 1011 * , 151 ( 2010 ) [ arxiv:1005.4690 [ hep - th ] ] . n. iizuka , n. kundu , p. narayan and s. p. trivedi , arxiv:1105.1162 [ hep - th ] . k. goldstein , s. kachru , s. prakash and s. p. trivedi , jhep * 1008 * , 078 ( 2010 ) [ arxiv:0911.3586 [ hep - th ] ] . j. p. gauntlett , s. kim , o. varela and d. waldram , jhep * 0904 * , 102 ( 2009 ) [ arxiv:0901.0676 [ hep - th ] ] . e. o colgain , o. varela and h. yavartanoo , `` non - relativistic m - theory solutions based on kaehler - einstein spaces , '' jhep * 0907 * , 081 ( 2009 ) [ arxiv:0906.0261 [ hep - th ] ] . b. dolan , phys . b * 140 * ( 1984 ) 304 . a. imaanpur and m. naghdi , phys . rev . d * 83 * , 085025 ( 2011 ) [ arxiv:1012.2554 [ hep - th ] ] . d. forcella and a. zaffaroni , arxiv:1103.0648 [ hep - th ] . s. kim , nucl . b * 821 * , 241 ( 2009 ) [ arxiv:0903.4172 [ hep - th ] ] ; m. k. benna , i. r. klebanov and t. klose , jhep * 1001 * , 110 ( 2010 ) [ arxiv:0906.3008 [ hep - th ] ] ; d. bashkirov and a. kapustin , jhep * 1105 * , 015 ( 2011 ) [ arxiv:1007.4861 [ hep - th ] ] . i. bena , phys . d * 62 * , 126006 ( 2000 ) [ arxiv : hep - th/0004142 ] . n. lambert and p. richmond , `` m2-branes and background fields , '' jhep * 0910 * , 084 ( 2009 ) [ arxiv:0908.2896 [ hep - th ] ] . a. hashimoto , arxiv:1105.3687 [ hep - th ] . j. l. f. barbon and c. a. fuertes , `` ideal gas matching for thermal galilean holography , '' phys . d * 80 * , 026006 ( 2009 ) [ arxiv:0903.4452 [ hep - th ] ] . k. balasubramanian and j. mcgreevy , `` the particle number in galilean holography , '' jhep * 1101 * , 137 ( 2011 ) [ arxiv:1007.2184 [ hep - th ] ] . n. halmagyi , m. petrini and a. zaffaroni , arxiv:1102.5740 [ hep - th ] ; s. kachru , r. kallosh and m. shmakova , arxiv:1104.2884 [ hep - th ] . d. cassani , g. dallagata and a. f. faedo , jhep * 1005 * , 094 ( 2010 ) [ arxiv:1003.4283 [ hep - th ] ] ; j. t. liu , p. szepietowski and z. zhao , phys . d * 81 * , 124028 ( 2010 ) [ arxiv:1003.5374 [ hep - th ] ] ; j. p. gauntlett and o. varela , jhep * 1006 * , 081 ( 2010 ) [ arxiv:1003.5642 [ hep - th ] ] . l. girardello , m. petrini , m. porrati and a. zaffaroni , nucl . b * 569 * , 451 ( 2000 ) [ arxiv : hep - th/9909047 ] . d. garfinkle , g. horowitz and a. strominger , phys . d * 43 * , 3140 ( 1991 ) ; m. duff and j. rahmfeld , phys . b * 345 * , 441 ( 1995 ) [ arxiv:9406105 [ hep - th ] ] .

enumeration of tilings on the square lattice with southwest - to - northeast diagonals drawn in was first studied in the 1990s ( see @xcite and the list of references therein ) . chris douglas @xcite proved a conjecture posed by james propp about the number of tilings of an analog of the aztec diamond on the square lattice with every second diagonal drawn in ( see figure [ douglas ] ) . in particular , douglas showed that the region of order @xmath0 has exactly @xmath1 tilings . ( 0,0 ) , @xmath2 and @xmath3 . the figure was first introduced in @xcite.,title="fig : " ] # 1#2#3#4#5 @font ( 6283,2966)(402,-2292 ) ( 2363,-1641)(0,0)[lb ] ( 6260,-1878)(0,0)[lb ] ( 828,-696)(0,0)[lb ] the author @xcite generalized the result of douglas by considering a common generalization of the douglas region and the aztec diamond on the square lattice with arbitrary diagonals drawn as in figure [ douglasgen ] , called a _ generalized douglas region _ , and showing that the tilings of the region are also enumerated by a power of @xmath4 ( see theorem 4 in @xcite ) . # 1#2#3#4#5 @font james propp @xcite also investigated a hybrid between the aztec diamond and a hexagon on the square lattice with every third diagonal drawn in , called _ quasi - hexagons_. the author @xcite solved a long standing problem of propp ( problem 16 in @xcite ) by giving an explicit formula for the number of hybrid domino - lozenge tilings of a quasi - hexagon . especially , in the case of symmetry quasi - hexagon , the tiling formula is a simple product formula , which is the product of a power of @xmath4 and the number of tilings of a semi - regular hexagon given by macmahon s formula @xcite . the author also proved a simple product tiling formula for an @xmath5-vertex counter part of the quasi - hexagons in @xcite . inspired by the work of b.y . yang @xcite and m. ciucu @xcite on the symmetric tilings of the aztec diamond , we consider the number of @xmath6 cyclically symmetric tilings of a generalized douglas region . moreover , we also consider a more general case when certain portions of the region have been removed along the symmetry diagonal as in figure [ douglashole ] . we show that the number of cyclically symmetric tilings of a generalized douglas region is given by a closed form product formula ( see theorem [ symthm1 ] in section 2 ) . the study symmetric tilings of the hexagons on the triangular lattice date back to the 1980s when richard stanley published his classical paper @xcite on 10 classes of symmetric plane partitions . as a quasi - hexagon can be viewed as a generalization of a hexagon , we investigate cyclically symmetric tilings of the quasi - hexagons . we use the subgraph replacement method to show that the number of symmetric tilings of a quasi - hexagon is the same as the symmetric tiling of a hexagon , up to a multiplicative factor ( see theorem [ hexsym ] in section 2 ) . the rest of this paper is organized as follows . we give detailed definition of the generalized douglas regions and the quasi - hexagons , and the statement of our main results ( theorems [ symthm1 ] and [ hexsym ] ) in section 2 . section 3 is devoted to several fundamental results in the subgraph replacement that will be employed in our proofs . in section 4 , we enumerate the perfect matchings of an aztec rectangle graph with holes , that itself can be considered as a generalization of the related work of b.y . yang @xcite and m. ciucu @xcite in the case of aztec diamonds . we will use this enumeration in our proof of theorem [ symthm1 ] is shown in section 5 . finally , section 6 presents the proof of theorem [ hexsym ] . a lattice divided the plane into disjoint fundamental regions , called _ cells_. a ( lattice ) _ region _ is a finite connected union of fundamental regions of that lattice . tile _ is the union of any two fundamental regions sharing an edge . a _ tiling _ of the region @xmath7 is a covering of @xmath7 by tiles with no gaps or overlaps . the number of tilings of a region @xmath7 is denoted by @xmath8 . let @xmath9 be a fixed drawn - in ( southwest - to - northeast ) diagonal on the square lattice . assume that @xmath10 diagonals have been drawn in above @xmath9 with the distances from between two consecutive ones from the top are @xmath11 , @xmath12, ,@xmath13 , and @xmath14 diagonals have been drawn in below @xmath9 with the distances between two consecutive ones from the bottom are @xmath15 , @xmath16, ,@xmath17 ( see figure [ hexagonregion ] ) . next , we color black and white the dissection obtained from the above set - up of drawn - in diagonals on the square lattice , so that two fundamental regions sharing an edge have different colors . we define the _ quasi - hexagon _ @xmath18 as follows . pick a lattice point @xmath19 on the the top drawn - in diagonal . start from @xmath19 we go south or west in each step so that the black fundamental region stays on the left . the resulting lattice path from @xmath19 intersects the diagonal @xmath9 at a lattice point @xmath20 . from @xmath20 , we go south or west so that the white fundamental region stays on the left in each step . our lattice path stops when reaching the bottom drawn - in diagonal at a lattice point @xmath21 . the described lattice path passing @xmath19 , @xmath20 and @xmath21 is the southwest boundary of the region . next , we pick a lattice point @xmath22 on the top drawn - in diagonal so that @xmath22 is @xmath23 unit on the right of @xmath19 . the northeast boundary is obtained from the southwest one by reflecting about the perpendicular bisector of the segment @xmath24 and changing the steps from south to north and from west to east . assume that the northeast boundary intersects @xmath9 and the bottom drawn - in diagonal by @xmath25 and @xmath26 , respectively . we complete the boundary of the region by connecting @xmath21 and @xmath26 , and @xmath22 and @xmath19 along the drawn - in diagonal . the six lattice points @xmath27 and @xmath22 are called the _ vertices _ of the region , and the diagonal @xmath9 are called the ( southwest - to - northeast ) axis of the region . ( 0,0 ) . the figure first appeared in @xcite.,title="fig : " ] # 1#2#3#4#5 @font ( 5276,5190)(406,-4616 ) ( 2200,-4276)(0,0)[lb ] ( 3200,-229)(0,0)[lb ] ( 3500,-793)(0,0)[lb ] ( 2600,158)(0,0)[lb ] ( 4400,-646)(0,0)[lb ] ( 800,-586)(0,0)[lb ] ( 1400,-3778)(0,0)[lb ] ( 0,-1704)(0,0)[lb ] ( 1400,-3076)(0,0)[lb ] ( 3400,-4141)(0,0)[lb ] ( 4600,-2986)(0,0)[lb ] ( 2000,344)(0,0)[lb ] ( 3700,-1329)(0,0)[lb ] the cells in a quasi - hexagon are unit squares or triangles . the triangular cells only appear along the drawn - in diagonals . a _ row of cells _ consists of all the triangular cells of a given color with bases resting on a fixed lattice diagonal , or consists of all the square cells ( of a given color ) passed through by a fixed lattice diagonal . define the _ ( generalized ) douglas region _ @xmath28 to be the region obtained from the portion of @xmath29 above the axis @xmath9 by replacing the triangles running along the top and the bottom draw - in diagonals by squares of same color ( see figure [ douglasgen ] ) . the ( generalized ) douglas region was first investigated in @xcite , and also in @xcite and @xcite as a common generalization of the original douglas regions @xcite and the aztec diamonds @xcite . as mentioned in @xcite ( theorem 2.1(a ) and theorem 2.3(a ) ) , if the triangular cells running along the bottom drawn - in diagonal of a quasi - hexagon or a douglas region are black , then the region has no tilings . _ therefore , from now on , we assume that the triangular cells running along the bottom drawn - in diagonal are white . _ this is equivalent to the fact that the last step of the southwest boundary is an east step . for any finite set of integers @xmath30 , @xmath31 , we define four functions @xmath32 @xmath33 @xmath34 and @xmath35 where the empty products are equal 1 by convention . the functions @xmath36 and @xmath37 were first introduced by jockusch and propp in @xcite , and @xmath38 and @xmath39 were first introduced by the author in @xcite as the tiling numbers of the so - called _ quartered aztec rectangles_. given a douglas region @xmath40 admitting the southwest - to - northeast symmetry axis @xmath41 . it is easy to see that we must have @xmath42 , for @xmath43 , @xmath10 is odd , and @xmath41 is _ not _ a drawn - in diagonal ( so all the cells running along @xmath41 are squares ) . we label those squares passed by @xmath41 as follows . if the symmetry center of @xmath26 stays inside on of the squares , we call this square _ center cell _ , and label it by @xmath44 . next we label the two squares closest to the center cell by @xmath45 , we label the two squares that are second closest to the center cell by @xmath4 , and so on . a cell of @xmath26 is said to be _ regular _ is it is either a black square or a black triangle pointing away from @xmath41 . we define the _ upper height _ @xmath46 of @xmath26 to be the number of rows of regular cells above @xmath41 or passed by @xmath41 , and the _ lower height _ @xmath47 of @xmath26 to be the number of rows of regular cells below @xmath41 or passed by @xmath41 . the number of regular cells above @xmath41 is denoted by @xmath48 , and the number of regular cells below @xmath41 is denoted by @xmath49 . finally , the total number of regular cells is denoted by @xmath50 . we have @xmath51 if @xmath41 passes white squares , and @xmath52 otherwise , where @xmath53 is the number of squares passed through by @xmath41 . we call @xmath53 the _ width _ of @xmath26 . we call a row of an odd number of black triangles pointing toward @xmath41 and above @xmath41 a _ singular row_. the statistic @xmath54 is called the _ defect _ of @xmath26 . we remove all squares having label in @xmath55 . denote by the resulting region by @xmath56 ( see figure [ douglashole ] for an example ; the black squares indicate the ones that have been removed ) . assume that @xmath57-\mathcal{s}=\{i_1,i_2,\dotsc , i_k\}$ ] , for @xmath58 . we define @xmath59 and @xmath60 . # 1#2#3#4#5 @font we notice that if @xmath56 admits a tiling , then the number of squares removed equals @xmath61 if @xmath41 passes white squares , and equals @xmath62 otherwise . moreover , in the latter case the center cell must be removed . in this case the number of cyclically symmetric tilings of the region @xmath63 is given by the following theorem . [ symthm1 ] given a positive @xmath64 and a sequence of positive integers @xmath65 so that @xmath66 admits a southwest - to - northeast symmetry axis @xmath41 . assume that @xmath54 is the number of rows of an odd number of black triangles that are pointing toward @xmath41 and above @xmath41 . assume that we remove all squares running along @xmath41 with labels in @xmath67 , and that the number of squares removed equals @xmath61 if @xmath41 passes white squares , and equals @xmath62 otherwise . \(a ) if @xmath41 passes white square cells of @xmath68 and @xmath69 , then the number of cyclically symmetric tilings of @xmath56 equals @xmath70 when @xmath53 is even , and equals @xmath71 when @xmath53 is odd . \(b ) if @xmath41 passes black square cells of @xmath68 and @xmath72 , then the number of cyclically symmetric tilings of @xmath56 equals @xmath73 when @xmath53 is odd , and equals @xmath74 when @xmath53 is even . we consider next the cyclically symmetric tilings of a symmetric quasi - hexagon @xmath75 define the polynomial of 3 variables @xmath76 this is exactly the number of transposed - complementary plane partition fit in the box @xmath77 ( see theorem 6.1 in @xcite ) . the cyclically symmetric tilings of a quasi - hexagon is given by the following theorem . a _ regular _ cell of a quasi - hexagon @xmath78 is either the square or a triangle pointing away from the axis @xmath9 . denote by @xmath79 and @xmath80 the number of rows of black regular cells above @xmath9 and the number rows of white regular cells below @xmath9 , respectively . denote @xmath81 and @xmath82 the number of black regular cells above @xmath9 and the number of white regular cells below @xmath9 . the _ width _ @xmath83 of @xmath78 is the number of cells running along @xmath9 . we use the notation @xmath84 for the number of the @xmath6 cyclically symmetric tilings . [ hexsym ] let @xmath85 be positive integers , so that for which the symmetric quasi - hexagon @xmath86 has the heights @xmath87 less than the width @xmath83 . + ( a ) if the @xmath88 and @xmath83 are even , then @xmath89 ( b ) if the @xmath88 is even and @xmath83 is odd , then @xmath90 ( c ) if the @xmath88 is odd and @xmath83 is even , then @xmath91 this section shares several preliminary results and definitions with the prequels @xcite of our paper . the first result not reported in @xcite is ciucu s lemma [ 4cycle ] . a _ perfect matching _ of a graph @xmath92 is a collection of edges such that each vertex of @xmath92 is adjacent to exactly one edge in the collection . the tilings of a region @xmath7 can be naturally identified with the perfect matchings of its _ dual graph _ ( i.e. , the graph whose vertices are the fundamental regions of @xmath7 , and whose edges connect two fundamental regions precisely when they share an edge ) . in view of this , we denote the number of tilings of a graph @xmath92 by @xmath93 . more generally , if the edges of @xmath92 have weights on them , @xmath93 denotes the sum of the weights of all perfect matchings of @xmath92 , where the weight of a perfect matching is the product of the weights on its constituent edges . a _ forced edge _ of a graph @xmath92 is an edge contained in every perfect matching of @xmath92 . let @xmath92 be a weighted graph with weight function @xmath94 on its edges , and @xmath95 is obtained from @xmath92 by removing forced edges @xmath96 , and removing the vertices incident to these edges . then one clearly has @xmath97 we present next three basic preliminary results stated below . ( 0,0 ) # 1#2#3#4#5 @font ( 4188,1361)(593,-556 ) ( 1336,591)(0,0)[lb ] ( 3549,621)(0,0)[lb ] ( 3757 , 89)(0,0)[lb ] ( 3999,621)(0,0)[lb ] ( 820,-541)(0,0)[lb ] ( 1840,-535)(0,0)[lb ] ( 3031,-535)(0,0)[lb ] ( 4426,-484)(0,0)[lb ] [ vertex - splitting lemma , @xcite lemma 2.2][vs ] let @xmath92 be a graph , @xmath98 be a vertex of it , and denote the set of neighbors of @xmath98 by @xmath99 . for an arbitrary disjoint union @xmath100 , let @xmath95 be the graph obtained from @xmath101 by including three new vertices @xmath102 , @xmath103 and @xmath104 so that @xmath105 , @xmath106 , and @xmath107 ( see figure [ vertexsplitting ] ) . then @xmath108 . [ star ] let @xmath92 be a weighted graph , and let @xmath98 be a vertex of @xmath92 . let @xmath95 be the graph obtained from @xmath92 by multiplying the weights of all edges that are adjacent to @xmath98 by @xmath109 . then @xmath110 . part ( a ) of the following result is a generalization due to propp of the urban renewal " trick first observed by kuperberg . parts ( b ) and ( c ) are due to ciucu ( see lemma 2.6 in [ 5 ] ) . ( 0,0 ) # 1#2#3#4#5 @font ( 4054,1735)(340,-948 ) ( 355,-156)(0,0)[lb ] ( 1064,-933)(0,0)[lb ] ( 1891,-106)(0,0)[lb ] ( 1182,603)(0,0)[lb ] ( 2717,-189)(0,0)[lb ] ( 3426,-933)(0,0)[lb ] ( 4253,-106)(0,0)[lb ] ( 3426,603)(0,0)[lb ] ( 904,-382)(0,0)[lb ] ( 1396,-388)(0,0)[lb ] ( 1418,130)(0,0)[lb ] ( 946,130)(0,0)[lb ] ( 2968,284)(0,0)[lb ] ( 3934,311)(0,0)[lb ] ( 3964,-544)(0,0)[lb ] ( 2965,-526)(0,0)[lb ] ( 2197,-817)(0,0)[lb ] [ spider lemma][spider ] ( a ) let @xmath92 be a weighted graph containing the subgraph @xmath111 shown on the left in figure [ spider1 ] ( the labels indicate weights , unlabeled edges have weight 1 ) . suppose in addition that the four inner black vertices in the subgraph @xmath111 , different from @xmath112 , have no neighbors outside @xmath111 . let @xmath95 be the graph obtained from @xmath92 by replacing @xmath111 by the graph @xmath113 shown on right in figure [ spider ] , where the dashed lines indicate new edges , weighted as shown . then @xmath114 . \(b ) consider the above local replacement operation when @xmath111 and @xmath113 are graphs shown in figure [ spider2](a ) with the indicated weights ( in particular , @xmath115 has a new vertex @xmath26 , that is incident only to @xmath19 and @xmath21 ) . then @xmath116 . \(c ) the statement of part ( b ) is also true when @xmath111 and @xmath113 are the graphs indicated in figure [ spider2](b ) ( in this case @xmath95 has two new vertices @xmath21 and @xmath26 , they are adjacent only to one another and to @xmath20 and @xmath19 , respectively ) . we quote the following useful result due to ciucu @xcite . ( 0,0).,title="fig : " ] # 1#2#3#4#5 @font ( 1725,1926)(931,-1107 ) ( 1418 , 12)(0,0)[lb ] ( 1655,603)(0,0)[lb ] ( 2245,603)(0,0)[lb ] ( 2363 , 12)(0,0)[lb ] ( 969 , 36)(0,0)[lb ] ( 946,-696)(0,0)[lb ] ( 1479,-819)(0,0)[lb ] [ 4cycle ] let @xmath92 be a weighted graph having a @xmath117-vertex subgraph @xmath78 consisting of two @xmath118-cycles that share a vertex . let @xmath119 , @xmath120 , @xmath121 , @xmath122 and @xmath119 , @xmath123 , @xmath124 , @xmath125 be the vertices of the 4-cycles ( listed in cyclic order ) and suppose @xmath122 and @xmath125 are only the vertices of @xmath78 with the neighbors outside @xmath78 . let @xmath95 be the subgraph of @xmath92 obtained by deleting @xmath120 , @xmath121 , @xmath123 and @xmath124 , weighted by restriction . then if the product of weights of opposite edges in each @xmath118-cycle of @xmath78 is constant , we have @xmath126 ( 0,0 ) with symmetric axis ; ( b ) the resulting graph after the cutting procedure.,title="fig : " ] # 1#2#3#4#5 @font ( 7111,2868)(225,-2253 ) ( 710,-460)(0,0)[lb ] ( 1477,-519)(0,0)[lb ] ( 2304,-492)(0,0)[lb ] ( 2540,-814)(0,0)[lb ] ( 3131,-519)(0,0)[lb ] ( 4490,-460)(0,0)[lb ] ( 5257,-519)(0,0)[lb ] ( 5494,-755)(0,0)[lb ] ( 5848,-610)(0,0)[lb ] ( 6379,-756)(0,0)[lb ] ( 6911,-519)(0,0)[lb ] ( 4253,189)(0,0)[lb ] ( 4312,-1287)(0,0)[lb ] ( 6200,-492)(0,0)[lb ] ( 532,-933)(0,0)[lb ] ( 296,-460)(0,0)[lb ] ( 4135,-401)(0,0)[lb ] next , we introduce a useful tool due to ciucu @xcite to investigate the number of perfect matchings of reflective symmetric graphs . let @xmath92 be a weighted planar bipartite graph that is symmetric about a horizontal line @xmath9 . assume that the set of vertices lying on @xmath9 is a cut set of @xmath92 ( i.e. , the removal of these vertices disconnects @xmath92 ) . one readily sees that the number of vertices of @xmath92 on @xmath9 must be even if @xmath92 has perfect matchings , let @xmath127 be half of this number . let @xmath128 be the vertices lying on @xmath9 , as they occur from left to right . let us color vertices of @xmath92 by black or white so that any two adjacent vertices have opposite colors . without loss of generality , we assume that @xmath129 is always colored white . delete all edges above @xmath9 at all white @xmath130 s and black @xmath131 s , and delete all edges below @xmath9 at all black @xmath130 s and white @xmath131 s . reduce the weight of each edge lying on @xmath9 by half ; leave all other weights unchanged . since the set of vertices of @xmath92 on @xmath9 is a cut set , the graph obtained from the above procedure has two disconnected parts , one above @xmath9 and one on below @xmath9 , denoted by @xmath132 and @xmath133 , respectively ( see figure [ verticalfactor ] ) . [ factorthm ] let @xmath92 be a bipartite weighted symmetric graph separated by its symmetry axis . then @xmath134 consider a @xmath135 rectangular chessboard and suppose the corners are black . aztec rectangle graph _ @xmath136 is the graph whose vertices are the white unit squares and whose edges connect precisely those pairs of white unit squares that are diagonally adjacent . the _ odd aztec rectangle graph _ @xmath137 is the graph whose vertices are the black unit squares whose edges connect precisely those pairs of black unit squares that are diagonally adjacent . figures [ twoar](a ) and ( b ) illustrate an example of @xmath138 and @xmath139 respectively . if one removes the bottom row of the board @xmath20 and applies the same procedure the definition of @xmath136 , the resulting graph is denoted by @xmath140 , and called a _ baseless aztec rectangle_. we also consider the graph @xmath141 that is obtained from the aztec rectangle @xmath136 by removing all its leftmost vertices . it is worth noticing that when @xmath142 , the aztec rectangle graph @xmath136 become the _ aztec diamond graph _ @xmath143 . elkies , kuperberg , larsen and propp @xcite showed that the number of perfect matchings of @xmath143 is exactly @xmath144 . the aztec rectangle graph @xmath136 does _ not _ have a perfect matching in general , however , where certain vertices have been removed from one side of its , the perfect matchings are enumerated by a simple product formula ( see proposition 2.1 in @xcite ) . next , we consider several variants of aztec rectangle graphs as follows . label the vertices on the left side of the aztec rectangle @xmath136 from bottom up by @xmath145 . denote by @xmath146 and @xmath147 the graphs obtained from @xmath136 by removing all odd - label and all even - label vertices , respectively ( see figures [ symt5](b ) and ( d ) for @xmath148 and @xmath149 , and figures [ symt6](b ) and ( d ) for @xmath150 and @xmath151 ) . we call @xmath146 and @xmath147 the _ odd- _ and _ even - trimmed versions _ of @xmath136 , respectively . apply similarly process we obtain the odd- and even - trimmed versions of the graphs @xmath137 , @xmath152 , and @xmath141 . figures [ symt2](b ) and ( d ) illustrate the graph @xmath153 and @xmath154 ; while the graphs @xmath155 and @xmath156 are shown in figures [ symt1](b ) and ( d ) . see figures [ symt3](b ) and ( d ) for @xmath157 and @xmath158 , and figures [ symt4](b ) and ( d ) for @xmath159 and @xmath160 . finally , examples of @xmath146 and @xmath147 are illustrated in figures [ symt8](b ) and ( d ) f and in figures [ symt7](b ) and ( d ) , respectively . similar to the case of the aztec rectangle graphs , the above trimmed aztec graphs " do not have perfect matching in general ; and we are interested in the case some bottom vertices of them have been removed . label the bottom vertices of @xmath147 , @xmath146 , @xmath161 , and @xmath162 by @xmath163 form left to right . for @xmath164 and @xmath165 , define @xmath166 ( resp . , @xmath167 ) to be the graph obtained from @xmath147 ( resp . , @xmath146 ) by removing all bottom vertices , except for the ones at the positions @xmath168 . define @xmath169 ( resp . , @xmath170 ) to be the graph obtained from @xmath161 ( resp . , @xmath162 ) by removing the bottom vertices at the positions @xmath168 . similarly , we label the bottom vertices of @xmath171 and @xmath172 by @xmath173 form left to right ; and we also label the bottom vertices of @xmath174 , and @xmath175 by @xmath163 from left to right . for @xmath164 and @xmath165 , define @xmath176 ( resp . , @xmath177 ) to be the graph obtained from @xmath171 ( resp . , @xmath174 ) by removing all bottom vertices , except for the ones at the positions @xmath168 . the graph @xmath178 ( resp . , @xmath179 ) is the graph obtained from @xmath172 ( resp . , @xmath175 ) by removing the bottom vertices at the positions @xmath44 and @xmath168 ( resp . , at the positions @xmath168 ) . the author showed that perfect matchings of our above graphs are always enumerated by a simple product formula ( see theorems 1.2 and 1.3 in @xcite ; our graphs here are the dual graphs of the regions mentioned in these theorems ) . [ qarthm ] for any @xmath180 and @xmath165 @xmath181 @xmath182 @xmath183 @xmath184 @xmath185 @xmath186 @xmath187 and @xmath188 in his ph.d . thesis @xcite , bo - yin yang proved a conjecture posed by jockush on the number of cyclically symmetric tilings of aztec diamond . ciucu reproved the result in @xcite using his factorization theorem [ factorthm ] and a result of jockush and propp @xcite on the so - called _ quartered aztec diamonds_. this is worth noticing that the gave a new proof for the result of jockush and propp in @xcite , and also generalized the result ( see theorems 1.2 and 1.3 in @xcite ) . in this section we enumerate cyclically symmetric perfect matchings of an aztec rectangle graph with several vertices removed along the symmetry axis . our result deduces ciucu and jockush s result ass a special case when the set of removed vertices is empty . consider an aztec rectangle @xmath136 with the horizontal symmetry axis @xmath9 and the vertical symmetry axis @xmath189 . we label the vertices of @xmath190 on @xmath9 as follows . if the symmetry center of the graph is a vertex on @xmath9 , then we label it by @xmath44 . label two vertices that is closest to the center by @xmath45 , label the second closest vertices by @xmath4 , and so on . we remove @xmath191 vertices so that the resulting graph still admits the vertical symmetry axis @xmath189 . denote by @xmath192 the label set of removed vertices , which are not the center ; and denote by @xmath193 the resulting graph . assume that @xmath194 is label set of the vertices of @xmath193 on @xmath9 . it is easy to see that if @xmath195 is even , then @xmath193 admits a perfect matching when @xmath196 , and if @xmath195 is odd , then we must have @xmath197 ( the latter inequality comes from the fact that the number of removed vertices , @xmath198 , must less than or equal to the number of vertices on @xmath9 ) . ciucu showed that if @xmath195 is even and @xmath196 , then the total number of perfect matchings of @xmath193 is given by a simple product formula ( see theorem 4.1 in @xcite ) . we are also interested in the perfect matchings of @xmath193 , which are invariant under the @xmath199 rotation around the symmetry center of the graph . denote by @xmath200 the number of cyclically perfect matchings a graph @xmath92 . we separate the label set of the vertices of @xmath193 on @xmath9 into two subsets : @xmath201 and @xmath202 . the number of cyclically symmetric perfect matchings of an aztec rectangle graph with holes is given by simple products in the following theorem . [ sym ] ( a ) for any @xmath203 and @xmath204 @xmath205 ( b ) for any @xmath206 and @xmath207 @xmath208 ( c ) for any @xmath206 and @xmath209 @xmath210 ( d)for any @xmath203 and @xmath211 @xmath212 .,width=510 ] we present only the proof of part ( a ) , the other parts can be obtained similarly . we will show that the number of @xmath199 cyclically symmetric perfect matchings of our graph is given by the product of the numbers of perfect matchings of two graphs in theorem [ qarthm ] . consider the holey aztec rectangle @xmath213 with the horizontal and vertical symmetry axes @xmath9 and @xmath189 ( see figure [ aztechole](a ) for @xmath214 ) . in this case consider the subgraph @xmath78 of @xmath92 that is induced by vertices on the right @xmath189 or on @xmath189 . label the vertices of @xmath92 on @xmath189 and above @xmath9 by @xmath216 from bottom to top ; and label the vertices of @xmath92 on @xmath189 and below @xmath9 by @xmath217 from top to bottom . it is readily to see each cyclically symmetric perfect matching @xmath218 of @xmath92 is determined uniquely by its sub - matching in the edge set of @xmath78 , @xmath219 . on the other hand , by the symmetry of @xmath218 , exactly one of two vertices @xmath220 and @xmath221 is covered by @xmath222 . therefore , given a cyclically symmetric perfect matching @xmath218 of @xmath92 , @xmath223 corresponds to a perfect matching of @xmath224 , the graph obtained from @xmath78 by identifying @xmath220 and @xmath221 , for any @xmath225 . this implies that the cyclically symmetric perfect matchings of @xmath92 are in bijection with the perfect matchings of @xmath224 . moreover , we can put the new vertices in @xmath224 on @xmath9 , so that @xmath224 admits the horizontal symmetry axis @xmath9 ( see figure [ aztechole](b ) ) . by factorization theorem [ factorthm ] , we have @xmath226 where @xmath224 has exactly @xmath227 vertices on @xmath9 . for even @xmath195 , we have @xmath228 is isomorphic to @xmath229 and @xmath230 is isomorphic to @xmath231 , while @xmath228 is isomorphic to @xmath231 , and @xmath230 is isomorphic to @xmath229 when @xmath195 is odd . therefore , the theorem follows from theorem [ qarthm ] . next , we present serval new transformations that we will employ in the proof of theorem [ symthm1 ] . the _ connected sum _ @xmath232 of two disjoint graphs @xmath92 and @xmath233 along the ordered sets of vertices @xmath234 and @xmath235 is the graph obtained from @xmath92 and @xmath95 by identifying vertices @xmath221 and @xmath236 , for @xmath237 . in lemmas [ newt1 ] , [ newt2 ] , [ newt3 ] , and [ newt4 ] , we always assume that @xmath92 is a graph , and @xmath238 is an ordered set of vertices in @xmath92 . moreover , all connected sums act on @xmath92 along @xmath238 and on other summands along their bottommost vertices ordered from left to right . ) of lemma [ newt1].,width=226 ] ) of lemma [ newt1].,width=226 ] [ newt1 ] @xmath239 and @xmath240 where @xmath241 and @xmath242 are the graphs obtained from @xmath146 and @xmath147 by appending @xmath0 vertical edges to their bottommost vertices , respectively ( see figure [ symt1 ] for examples of the transformation in ( [ t1a ] ) , and figure [ symt2 ] for examples of the transformation in ( [ t1b ] ) ) . .,width=377 ] we only prove here the transformation in ( [ t1a ] ) , based on figure [ prooft1 ] , for @xmath243 ; and the transformation in ( [ t1b ] ) can be obtained in the same way . first we apply the vertex - splitting lemma [ vs ] to all vertices of @xmath244 that are incident to a shaded diamond or a partial diamond as in figure [ prooft1](a ) . we get the graph @xmath245 on figure [ prooft1](b ) . next , we apply spider lemma [ spider ] around @xmath246 shaded diamonds and partial diamond ( the dotted edges have weight @xmath247 ) , and remove all edges incident to a vertex of degree 1 , which are forced . we obtain a weighted graph @xmath248 obtained from @xmath249 by assigning to each edge of @xmath161 a weight @xmath247 . finally , we get back the graph @xmath249 by applying the star lemma [ star ] with factor @xmath250 at @xmath251 shaded vertices in figure [ prooft1](c ) . by lemma [ vs ] , [ spider ] , and [ star ] , we have @xmath252 which implies ( [ t1a ] ) . ) of lemma [ newt2].,width=226 ] ) of lemma [ newt2].,width=226 ] by applying the transformations in lemma [ newt1 ] ( in reverse ) , and then vertex - splitting lemma , one can get the following transformations . [ newt2 ] @xmath253 and @xmath254 where @xmath255 and @xmath256 are the graphs obtained from @xmath162 and @xmath161 by appending @xmath0 vertical edges to their bottommost vertices , respectively ( figure [ symt6 ] shows the transformation in ( [ t2a ] ) , and figure [ symt5 ] illustrates the transformation in ( [ t2b ] ) ) . ) of lemma [ newt3].,width=226 ] ) of lemma [ newt3 ] . red edge has weight @xmath247.,width=226 ] by using lemma [ 4cycle ] we can get the following lemma . [ newt3 ] @xmath257 and @xmath258 where @xmath259 is the graph obtained from @xmath175 by appending @xmath0 vertical edges to its bottommost vertices ; and where @xmath260 is the graph obtained from @xmath261 by appending @xmath0 vertical edges to its bottommost vertices , the leftmost vertical edge is weighted @xmath247(the transformation in ( [ t3a ] ) is shown in figure [ symt4 ] , and the transformation in ( [ t3b ] ) is illustrated in figure [ symt3 ] ) . ) in lemma [ newt3].,width=377 ] ) in lemma [ newt3].,width=377 ] we only need to prove ( [ t3a ] ) for even @xmath195 , and the case of odd @xmath195 follows by removing forced edges . our proof is illustrated in figure [ prooft3 ] , for @xmath262 and @xmath263 . we follow the typical procedure in our proof of lemma [ newt1 ] : first apply vertex - splitting lemma [ vs ] to the vertices in @xmath264 that are incident to a shaded diamond or partial diamond ( see figure [ prooft3](a ) and ( b ) ) . second , apply suitable replacement in spider lemma [ spider ] around @xmath246 shaded diamonds and partial diamonds . third , apply lemma [ 4cycle ] to remove @xmath265 7-vertex subgraph consisting of two shaded 4-cycles ( see figure [ prooft3](c ) ; the dotted edges are weighted by @xmath247 ) . finally , apply the star lemma [ star ] with factor @xmath250 to all @xmath266 shaded vertices as in figure [ prooft3](c ) . the resulting graph is exactly @xmath267 . by lemmas [ vs ] , [ spider ] , [ 4cycle ] , and [ star ] , we obtain @xmath268 which implies ( [ t3a ] ) . next , we show the proof of ( [ t3b ] ) for odd @xmath195 , the case of even @xmath195 follows by removing forced edges . our proof is shown in figure [ prooft3b ] , for @xmath269 and @xmath263 . we apply vertex - splitting lemma to the vertices in @xmath270 incident to a shaded diamond or partial diamond as in figures [ prooft3b](a ) and ( b ) . then apply spider lemma to @xmath271 shaded diamonds and partial diamonds . next , we apply lemma [ 4cycle ] to remove @xmath272 subgraphs consisting of two shaded 4-cycles ( see figure [ prooft3b](c ) ; the dotted edges have weight @xmath247 ) , and apply vertex - splitting lemma [ vs ] ( in reverse ) to eliminate the two solid edges in the resulting graph . finally , apply star lemma [ star ] ( for the factor @xmath250 ) to all @xmath266 shaded vertices . by lemmas [ vs ] , [ spider ] , [ 4cycle ] , and [ star ] , we get @xmath273 which yields ( [ t3b ] ) . ) of lemma [ newt4],width=226 ] ) of lemma [ newt4 ] . red edge has weight @xmath247.,width=226 ] similar to lemma [ newt3 ] , we have the following lemma . the proof of the next lemma is essentially the same as that of lemma [ newt3 ] , and will be omitted . [ newt4 ] @xmath274 and @xmath275 where @xmath276 is the graph obtained from @xmath174 by appending @xmath0 vertical edges to its bottommost vertices ; and where @xmath277 is the graph obtained from @xmath278 by appending @xmath0 vertical edges to its bottommost vertices , the leftmost vertical edge is weighted @xmath247 ( the transformation in ( [ t4a ] ) is shown in figure [ symt8 ] , and the transformation in ( [ t4b ] ) is illustrated in figure [ symt7 ] ) . we are now ready to prove theorem [ symthm1 ] . we will prove only for the case when @xmath41 passes white square cells and @xmath279 is even ( the other cases can be obtained in the same manner ) . we recall that @xmath41 is not a drawn - in diagonal , and that @xmath10 is odd . since @xmath41 is always the symmetry axis of the region , for convenience , in this proof , we use notation @xmath83 , @xmath88 , @xmath21 , @xmath280 for @xmath53 , @xmath46 , @xmath281 , @xmath54 when the holey douglas region @xmath282 is determined . the dual graph @xmath283 of the region @xmath56 ( rotated @xmath284 ) has the vertical and horizontal symmetry axes @xmath189 and @xmath9 . denote by @xmath285 the intersection of the two axes . consider the subgraph @xmath78 of @xmath283 that is induced by the vertices staying on the right or lying on @xmath189 . let @xmath224 be the graph obtained from @xmath78 by identifying two vertices on @xmath189 of the same distance to @xmath285 . we can put the new vertices in @xmath224 ( that obtained from identifying two vertices in @xmath78 ) on the @xmath9 ( see figure [ symf1 ] ) . we call @xmath224 the _ orbit graph _ of @xmath283 , denoted by @xmath286 . and its orbit graph.,width=453 ] the number of cyclically symmetric tilings of @xmath56 is equal to the number of cyclically symmetric perfect matchings of @xmath283 ; and , similar to proof of theorem [ qarthm ] , the latter number equals the number of perfect matchings of the obit graph @xmath286 . by theorem [ qarthm ] , we only need to show that @xmath287 we prove ( [ symrefine ] ) by induction on the number of layers @xmath10 ( @xmath10 is odd ) of @xmath288 . if @xmath289 , then the theorem follows from theorem [ qarthm ] . assume that ( [ symrefine ] ) is true for all holey douglas region with less than @xmath10 layers , @xmath290 ; we need to show that ( [ symrefine ] ) also holds for any holey douglas region having @xmath10 layers @xmath56 . .,width=453 ] consider the dual graph @xmath291 of region ( see figure [ symf1](a ) for an example ) . there are four cases to distinguish , based on the parities of @xmath292 and @xmath119 . @xmath292 and @xmath119 are even . _ define a holey douglas region @xmath293 by @xmath294 and @xmath295 assume that @xmath296 . apply the transformation ( [ t1b ] ) in lemma [ newt1](1 ) to the top part of @xmath297 , we get the graph @xmath298 , where @xmath95 is the dual graph @xmath293 , and obtain @xmath299 this process is illustrated in figures [ symf2](a ) and ( b ) . similarly , we apply the transformation in ( [ t1a ] ) of lemma [ newt1 ] to the bottom part of @xmath300 , we get the graph @xmath301 and obtain @xmath299 this process is shown in figures [ symf2](c ) and ( d ) . multiply the two above equalities , we get @xmath302 by factorization theorem , we deduce @xmath303 then thus @xmath304 since we are assuming that @xmath41 passes white squares of @xmath68 , the number of squares removed from @xmath41 is @xmath305 . it means that , the number vertices of @xmath92 on @xmath9 is @xmath306 . moreover , it is easy to see that the number of vertices of @xmath92 running along the vertical symmetry axis @xmath189 is also @xmath306 . thus , @xmath307 . one readily sees that @xmath26 and @xmath293 have the same height , so @xmath308 in this case . it means that ( [ case1a ] ) can be simplified to @xmath309 similarly , if @xmath310 , then we apply transformation ( [ t1b ] ) in lemma [ newt1 ] to the top part of @xmath300 , we get the graph @xmath301 and obtain @xmath311 next , we apply the transformation ( [ t1a ] ) to the bottom part of @xmath297 , we get the graph @xmath298 and @xmath312 therefore , similar to ( [ case1b ] ) , we have @xmath313 into the orbit graph of @xmath314.,width=491 ] by ( [ case1b ] ) and [ case1c ] ) , we always have @xmath315 denote by @xmath316 the @xmath41-width , the height , the number of black regular cells above @xmath41 , and defect of @xmath293 , respectively . one readily sees that @xmath317 , @xmath318 , @xmath319 , and @xmath320 by induction hypothesis for the region @xmath293 having @xmath321 layers @xmath322 and ( [ symrefine ] ) follows from ( [ case1d ] ) . _ @xmath292 and @xmath119 odd . _ do similarly to case 1 by using suitable transformations in lemma [ newt2 ] to the top of @xmath300 or the bottom of @xmath297 ( as opposed to using the transformations in lemma [ newt1 ] ) . if @xmath323 , then we get @xmath324 and if @xmath325 , then @xmath326 where @xmath95 is the dual graph of the region @xmath327 defined by @xmath328 and @xmath329 moreover , we can again the @xmath26 and @xmath327 have the same height ( with respect to the southwest - to - northeast axis @xmath41 ) , so @xmath330 . then thus , we always have in this case @xmath331 denote by @xmath332 the @xmath41-width , the @xmath41-height , the number of black @xmath41-regular cells in the upper part , and the @xmath41-defect of @xmath327 , respectively . one can check that the number of rows of black regular rows above @xmath9 is unchanged . we also have @xmath333 , @xmath334 , @xmath335 . moreover , @xmath336 thus ( [ symrefine ] ) follows from the induction hypothesis for @xmath327 and ( [ case2d ] ) . @xmath292 is odd and @xmath119 is even . _ apply the same procedure in case 1 and case 2 by using suitable transformations in lemma [ newt3 ] to the top of @xmath300 or the bottom of @xmath297 . we also get @xmath337 where @xmath338 is the dual graph of the region @xmath327 defined as in case 2 . similarly to case 2 , we have ( [ symrefine ] ) . @xmath292 is even and @xmath119 is odd . _ again , we apply the same procedure as that in the three previous cases by using suitable transformations in lemma [ newt4 ] to the top of @xmath300 or the bottom of @xmath297 . if @xmath296 , we get @xmath339 and if @xmath310 , then @xmath340 where @xmath95 is the dual graph of the region @xmath293 defined as in case 1 . then thus , we always have @xmath341 similarly , to case 1 , we have @xmath342 , @xmath317 , and @xmath343 . however , in this case @xmath344 , since we have a bad row staying right below the first layer of @xmath26 , but it does not appear in @xmath293 . thus , by induction hypothesis for @xmath293 and ( [ case4d ] ) , we have @xmath345 then ( [ symrefine ] ) follows . assume that the triangles having base on the axis of @xmath86 are black . we consider the dual graph @xmath92 of the region @xmath86 ( rotated @xmath346 ) with the horizontal and vertical axes @xmath9 and @xmath189 . similar to the proof of theorem [ symthm1 ] , we consider the orbit graph @xmath347 obtained from the portion @xmath348 of @xmath92 on the right of its vertical symmetry axis @xmath189 by identifying any two vertices on that axis which have the same distance to @xmath9 . again , the cyclically symmetric perfect matchings of @xmath92 are in bijection with the perfect matchings of @xmath347 ( see figure [ symmetricfig5 ] ) . moreover , we can arrange the identified vertices on @xmath9 so that the orbit graph @xmath347 admit @xmath9 as its symmetry axis . if @xmath289 , then ( [ thm3eq1 ] ) is trivial . assume that ( [ thm3eq1 ] ) holds for any symmetric quasi - hexagon with less than @xmath10 ( @xmath352 ) layers in each part , we need to show that the equality holds also for any symmetric quasi - hexagon @xmath353 . we use similar argument to that in the proof of theorem [ symthm1 ] . in particular , we transform the orbit graph of @xmath86 into the orbit of a symmetric quasi - hexagon having less layers by using the suitable transformations in lemmas [ newt1][newt4 ] . similar to the theorem 7.1 in @xcite , we notice that it can not apply the factorization theorem directly here , since the vertices of @xmath347 on @xmath9 do _ not _ form a cut set . however , the lemma 2.1 in @xcite still applies , it means that all @xmath354 graphs obtained from @xmath347 by cutting edges from above or below each of @xmath130 s have the same number of perfect matchings . we now consider a cutting procedure at the vertices @xmath130 s as follows . first , we color the vertices of @xmath347 inductively from left to right : color @xmath129 by white , then color the next vertex the same color as its left if there is not an edge connecting them , otherwise we color it by the opposite color . assume that @xmath355 is the graph obtained from @xmath347 by cutting above all white @xmath130 s and below all black @xmath130 s . we claim that all perfect matchings of @xmath355 have the white @xmath131 s matched upward , and black @xmath131 s matched downward . we consider the collection @xmath356 of @xmath354 graphs obtained from @xmath355 by cutting at all @xmath131 s from above or below . the set of perfect matchings of @xmath355 is in bijection to of disjoint union of sets of perfect matchings of the members in @xmath356 . all members @xmath357 of @xmath356 are bipartite graphs , and it easy to check that its two vertex classes have the same size only if @xmath357 is obtained from cutting below all white @xmath131 s and above all black @xmath131 s . this implies our claim . now denote by @xmath358 is the graph obtained from @xmath347 by cutting above all white @xmath130 s and black @xmath131 s , and below all black @xmath130 s and white @xmath131 s . then we get @xmath359 apply the transformations in [ newt1 ] to the top and bottom layer of @xmath358 , we get the graph isomorphic to @xmath360 , where @xmath95 is the dual graph of the region @xmath361 defined by @xmath362 and obtain @xmath363 by induction hypothesis and explicit calculating the statistic of the region @xmath361 , we obtain @xmath364 where @xmath365 refer to @xmath361 corresponding to their unprimed counterparts in @xmath78 . then ( [ thm3eq1 ] ) follows . consider the dual graph @xmath338 of the region @xmath366 apply the equation ( [ thm3eq1 ] ) to the graph @xmath338 , we have @xmath367 thus , by ( [ thm3eq1 ] ) and ( [ thm3eq2 ] ) , we obtain @xmath368 then the number of perfect matchings of @xmath369 is given by theorem 7.1 in @xcite , and the theorem follows in the current case . next , we apply the vertex - splitting lemma to all vertices at the bottom of the upper part of @xmath373 , and apply the suitable transformations in lemmas [ newt1][newt4 ] , we transform @xmath373 to @xmath374 , where @xmath350 is the dual graph of @xmath375 . then this case follows from the case we treated before . n. elkies , g. kuperberg , m.larsen , and j. propp , _ alternating - sign matrices and domino tilings ( part i ) _ , j. algebraic combin . * 1 * ( 1992 ) : 111132 . n. elkies , g. kuperberg , m.larsen , and j. propp , _ alternating - sign matrices and domino tilings ( part ii ) _ , j. algebraic combin . * 1 * ( 1992 ) : 219234 . t. lai , _ enumeration of tilings of quartered aztec rectangles _ , electron . j. combin . * 21*(4 ) ( 2014 ) , p4.46 . t. lai , _ a generalization of aztec diamond theorem , part i _ , electron . j. combin . * 21*(1 ) ( 2014 ) , p1.51 .

our universe starts with hydrogen ( h ) , helium ( he ) and a small amount of lithium ( li ) . these elements are synthesized during the big bang and all the other elements equal to or heavier than carbon ( c ) are synthesized in stars and during supernovae . in astrophysics , the history of heavy element generation via the nucleosynthesis in stars and succeeding pollution of the interstellar medium ( ism ) are expressed as `` chemical evolution '' . during stellar evolution , nucleosynthesis progresses from h to iron ( fe ) where the final products depend on the mass of a star . the massive stars heavier than @xmath0 are thought to explode as type ii supernovae ( hereafter sne ii ) . in this phase , heavy elements ( @xmath1 fe ) are synthesized . if a star is not so massive , say @xmath2 , it experiences an asymptotic giant branch ( agb ) phase after the main sequence phase . during the agb phase , the neutron capture process takes place with a slow speed ( slow means that the timescale of the @xmath3 decay , @xmath4 , is comparable to that of the neutron capture , @xmath5 ) . this slow process ( @xmath6-process ) generates elements heavier than fe efficiently . in addition , a part of the intermediate mass stars binary systems is thought to explode as a type ia sne ( hereafter sne ia ) . sne ia are triggered by the roche - lobe overflows or the final coalescences of binary systems . in the former situation , one of the progenitor star in a binary is a degenerated star and , in the latter , both are degenerated stars . thus , these two scenarios are called single - degenerate or double - degenerate scenarios @xcite . again , heavier elements are synthesized , such as fe and nickel ( ni ) . numerical studies tell us that neutron star - neutron star mergers ( nsms ) could be a possible site of @xmath7-process ( @xmath8 ) elements , in particular , the heavier r - process elements ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? each process has its own characteristic time - scale and abundance pattern . thus , they work as the chronometer for the formation history of the galactic stellar system . for instance , in an early phase ( @xmath9 ) , the chemical evolution progresses mainly through the pollution by sne ii . with this process , the @xmath10 elements of which the masses are multiples of he , such as oxygen ( o ) , magnesium ( mg ) , silicon ( si ) , and calcium ( ca ) , are ejected to the interstellar medium ( ism ) . on the other hand , in sne ia , fe , and ni are efficiently released to the ism . the combination of these events makes the observed features of the plateau ( flat part ) of the [ @xmath10/fe ] as a function of [ fe / h ] \equiv \log_{10 } ( n_{\rm a}/n_{\rm b } ) - \log_{10 } ( n_{\rm a}/n_{\rm b})_{\odot}$ ] , where @xmath11 and @xmath12 are the number abundance of element a and b , respectively . ] in a low metallicity part , and causes the distribution of [ @xmath10/fe ] as a function of [ fe / h ] ( e.g. , * ? ? ? the flat part represents the abundance pattern of sne ii . the decreasing part is composed of the mixture of sne ii and sne ia where the contribution of sne ii is gradually decreasing , reflecting its time - scale , and that of sne ia is gradually increasing . the breaking point of the flat part reflects the time scale of the enrichment of the chemical composition in the ism . the breaking point is [ fe / h ] @xmath13 in the milky way galaxy ( e.g. , * ? ? ? * ) , whereas it is much lower in the local dwarf galaxies ( e.g. , * ? ? ? agbs do not mainly contribute to the distributions of [ @xmath10/fe]-[fe / h ] relations while they can contribute rather low mass elements such as c and nitrogen ( n ) . there are many galactic surveys which resolve the chemical compositions and kinematic information of a lot of stars , e.g. , rave @xcite , apogee @xcite , gaia - eso @xcite , hermes ( grash ) @xcite . the data obtained by these surveys can make a strong constraint on the formation and evolution of the milky way galaxy and thus , it becomes a clue to understanding the formation and evolution of galaxies in general . it has been almost a quarter of a century since the first 3-dimensional simulations of galaxy formation , including dark matter and baryon . the first galaxy formation simulation including chemical evolution was carried out by @xcite . in this simulation , they only solved the evolution of the metallicity @xmath14 , the mass fraction of the heavy elements , synthesized by sne ii . then , @xcite introduced both sne ii and ia . the models have improved further and the current concordance simulations include about 10 elements ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) , as well as the mixing of them in the ism @xcite . semi - analytical models are also considered , in which various elements are released from different feedback models ( e.g. , * ? ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? unlike solving the evolution of individual stars , simulations of galaxy formation use a simple stellar populations ( ssp ) approximation , where star particles consist of a cluster of stars sharing the same age and metallicity , and whose mass function follows a certain initial mass function ( hereafter imf ) . in order to solve chemical evolution , the amount of newly synthesized elements in stars , `` yields '' , and the return mass fraction of ssp particles are necessary . these data are obtained from the studies of stellar evolution and explosions . there are flexibilities to choose yields tables as well as the functional forms of imf and its mass range . the most popular yields for sne ii are @xcite and its improved version supplied by @xcite . for sne ia , the yields of @xcite and its updated version given by @xcite are well - used . for agbs , the yields tables of @xcite , @xcite , and @xcite are widely used . the combinations of yields tables can lead to different chemical abundance patterns and thus it is intensively studied ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? it is pointed out by the galactic chemical evolution model of @xcite that the fe yields of @xcite are slightly large in order to obtain good agreement with observations . is too small . ] usually the fe yield is reduced by a factor of 2 from the original values ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? the yields tables of @xcite , which is an extended version of @xcite , also have the same problem . sometimes ad hoc modifications are used to halve the yield of fe and double those of mg and c in order to fit the observations ( see * ? ? ? since these yield tables are widely used in simulations of galaxy formation , these modifications are commonly used . an essential solution to solve this problem is to update the yields tables following the progress of stellar evolution models and reaction rates . however , it might be difficult to update yields tables frequently since , in simulation code , the chemical evolution part is deeply connected to them . moreover , the size and range of mass and metallicity grids of the yields tables are generally different from each other . this inhibits the smooth replacement of yields tables , although it is a theoretically straightforward task . if the chemical evolution part is implemented separately in a software library which can be linked from simulation codes directly and it has clearly defined application interfaces ( apis ) , the chemical evolution part is developed independently and it becomes easy to take the latest models . insofar as we know , such software library does not exist . there are several softwares which can generate stellar yields tables . starburst99 is a widely - used software to model spectrophotometric properties of star - forming galaxies @xcite . this software contains a huge database of stellar evolutions and thus , it can output yields . since this software is not designed to link from other codes , it is necessary to pre - generate a yields table if one uses it in one s simulation code . thus , when parameters change , e.g. , the functional form and the mass range of imf , the reconstruction of the yields table is necessary . the nugrid collaboration has developed a python code , sygma , which can deal with the yields of the nugrid project and can be used in chemical evolution of galaxies @xcite . since this is implemented as an ipython notebook , users need to pre - generate yields in order to use the outcomes from sygma . the aim of this paper is to describe a software library which deals with chemical evolution via apis and to demonstrate the capability of this library . this library is named `` celib '' after the term `` chemical evolution library '' . since the primary purpose of this library is to use it with simulation codes of galaxy formation , this library works under the ssp approximation . note that its main function is to return yields ( and energy ) of various feedback types from ssp particles . redistribution of these quantities is not within its scope because there are a number of redistribution manners and the implementation of them is deeply coupled with simulation codes . models and parameters of this library are selectable at the runtime and the imf weighted yields are automatically generated in the initialization process . celib is an open - source software library and is released under the mit license . the whole source codes including some examples are distributed through the following website : https://bitbucket.org/tsaitoh/celib . a documentation is also available in this site . the structure of this paper is as follows . first , we briefly describe the design concept of this library in [ sec : design ] . then we describe the notations and definitions in [ sec : notations ] and we introduce components which this library consists of in [ sec : components ] . the reference feedback models of sne ii / ia , agbs and nsms are described in [ sec : modeling ] . in [ sec : implementation ] , we explain our implementation of this library . in [ sec : applications ] , we show the results of the simple one - zone model and galaxy simulation using this library . finally , we provide the summary in [ sec : summary ] . the key concepts of the design of celib are as follows . * celib returns yields and energy from ssp particles , based on the adopted imf and yields tables . all returned quantities are imf weighted ones . useful data , such as the functional form of imf and stellar lifetimes , are also implemented . * celib supplies reference feedback models of sne ii / ia , agbs and nsms . reference feedback models can provide the event time and the amount of released mass , metal and energy . a redistribution manner of these quantities is outside its scope . * the functional form of the imf , its mass ranges , yields tables , etc . are changed by using control flags and parameters . all changes of flags and parameters are reflected when the initializer of celib is called . * all communication between a user simulation code and celib is carried out through celib apis . figure [ fig : concept ] shows the schematic picture of the relation between a user s simulation code and celib . the simulation code deals with the time integration of a system solving gravitational interactions among dm , gas , and star ( ssp ) particles and hydrodynamics , involving baryon physics such as radiative cooling / heating , star formation , and feedbacks . celib is responsible for the chemical evolution part . event time , released energy and ejected metals are evaluated when the simulation code sends data of an ssp particle to celib via apis . what the simulation code needs to do is to add the released energy and the ejected metals to the surrounding ism and to reduce the mass of the ssp particle so that the mass is conserved . before explaining details of celib , we need to confirm the notations and what quantities are used and obtained by using the library . we describe them in this section . the primary purpose of this library is to calculate the stellar yields from various evolution phases of stars under the ssp approximation . the stellar yield , the ejected mass of the @xmath15-th element from a star whose mass is @xmath16 , is expressed as @xmath17 where @xmath18 is the _ net yield _ which represents that the newly synthesized mass of @xmath15-th element , @xmath19 is the metallicity of the @xmath15-th element originally presented in the progenitor star and @xmath20 is the total ejecta mass of the star . this quantity , @xmath21 , is always larger than or equal to zero . the total ejecta mass is @xmath22 where @xmath23 is the remnant mass of a star whose mass is @xmath16 . it is essential to use the net yields for the study of the galactic chemical evolution . rearranging eq . , we can obtain @xmath24 we gather this quantity from the literature . however , some studies provide not @xmath25 but @xmath21 or @xmath26 where @xmath27 is the ejected mass of the @xmath15-th element via stellar winds . in such cases , we reconstruct @xmath28 carefully with several reasonable assumptions . this quantity can be positive and negative , unlike the stellar yield @xmath29 . we use @xmath28 for sne ii and agbs , whereas we adopt @xmath29 for sne ia and nsms . since this library is used under the ssp approximation , the imf weighted yield is useful . here , we use @xmath30 where @xmath31 is the imf function ( see below for the concrete functional forms ) . with this quantity and the return mass fraction , @xmath32 we can evaluate the yield of an ssp particle as @xmath33 by multiplying the mass of the ssp particle to eq . , we can obtain the amount of the released mass of @xmath15-th element from a feedback event . the imf weighted released amount of energy from massive stars can be evaluated by the following equation : @xmath34 where , @xmath35 is the released energy from a star whose mass is @xmath16 . strictly speaking , the released mass of @xmath15-th element changes depending on the amount of the other elements . the @xmath36 formulation @xcite is known as a model which considers the effect of the change in amounts of other elements . we do not implement the @xmath36 formulation here in celib , for simplicity . in this section , we describe key components of celib . imfs which are pre - implemented are shown in [ sec : imf ] . the metallicity dependent stellar lifetimes are explained in [ sec : lifetime ] . then , thirteen elements in celib are explained in [ sec : elements ] . yields for sne ii / ia , agbs and nsms are described in [ sec : snii ] , [ sec : snia ] , [ sec : agb ] , and [ sec : nsm ] , respectively . released energy and return mass from an ssp particle with various imfs are summarized in [ sec : misc ] . in [ sec : abundance ] , the implemented solar abundance patterns are shown . celib pre - implements seven popular imfs for ordinary population stars ( population i / ii ; hereafter pop i / ii ) and users can choose which one to use . moreover , celib supports an imf for population iii ( hereafter pop iii ) stars , which is utilized only for extremely low metal ssps . celib does not allow the functional form of the imf to be changed depending on the situation in a single simulation , except for the case using a pop iii imf . when the contribution of pop iii stars is adopted , the functional form of the imf and mass range are changed depending on metallicity . the definition of the imf is @xmath37 where @xmath38 is the number of stars in a given mass bin ( @xmath39 ) . the explicit functional forms of the imf users can choose will be explained below . the normalization of the imf is done by the following equation : @xmath40 here , @xmath41 and @xmath42 are the lower and upper mass boundaries of the imf . substituting eq . into eq . , we have @xmath43 thus , the normalization is `` the mass weighted number of stars '' . because of this definition , the number of stars per @xmath44 of a given mass range ( @xmath45 ) can be calculated as @xmath46 the current version of celib supports seven well used imfs for pop i / ii and one imf for pop iii stars . they are the salpeter , diet salpeter , miller - scalo , kennicutt , kroupa(2001 & 1993 ) , chabrier , and susa imfs . we describe the functional form of each imf below . the salpeter imf @xcite is the most fundamental one which has a power law form with a single segment : @xmath47 the second imf is the diet salpeter imf . the lower part of the imf is reduced to be a flat profile , whereas the higher part is the same as the original salpeter imf . this imf is originally introduced in order to explain the observed mass - to - light ratio of disk galaxies @xcite . the form of the diet salpeter imf is @xmath48 the miller - scalo imf @xcite has a power law form with three segments : @xmath49 the high ( low ) mass end has a deeper ( shallower ) power index when we compare it with that of the salpeter imf . the functional form of the kroupa imf @xcite celib adopts is as follows : @xmath50 note that the minimum mass of a star which is followed by the original form of the kroupa s imf is @xmath51 , but we do not take them and adopt an imf greater than @xmath52 . in addition to the kroupa imf @xcite , celib also has the imf of @xcite ( hereafter kroupa1993 imf ) . the functional form of this imf is @xmath53 as is obvious , the slope of this imf in the mass regime at @xmath54 is steeper than that of the @xcite s imf . @xcite has provided the imf with the following form : @xmath55 the final imf for ordinary populations of stars is the chabrier imf @xcite . this imf has a rather complicated form since the lower part of it is expressed by a log - normal function . the values of the normalization parameters are given by @xcite . the functional form of the imf is @xmath56 ^ 2}{2 \sigma^2 } \right \ } \\ \hspace{1 cm } 0.1~{m_{\odot } } < m < 1~{m_{\odot } } , \\ m^{-1.3 } \\ \hspace{1 cm } 1~{m_{\odot } } < m < 100~{m_{\odot } } , \end{cases}\ ] ] where @xmath57 and @xmath58 . for the imf of pop iii stars , we use the following function : @xmath59 ^ 2 } { 2 \sigma_{\rm popiii}^2 } \right \ } \\ & \hspace{1 cm } 0.7~{m_{\odot } } < m < 300~{m_{\odot } } , \nonumber\end{aligned}\ ] ] where @xmath60 and @xmath61 . this functional form is evaluated by using the eye - ball fitting of the mass distribution of @xcite ( their figure 9 ) . this imf is used only when the pop iii mode turns on . it should be noted that there is no consensus on the shape of the pop iii imf and other researchers obtain different functional forms of pop iii imfs ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? it is formally not difficult to extend this library to use other functional forms for pop iii stars . the normalization coefficient of each imf for the given mass range is evaluated at the initialization phase of the library following eq . . the contributions of the lower and upper mass boundaries of pop i / ii imfs are not crucial for the whole evolution of a system . for the lower mass boundary , we adopted @xmath52 for all these imfs for simplicity . there are some variations , but the contribution of the lower mass limit is almost negligible because of their long lifetimes ( see [ sec : lifetime ] ) . for the upper mass boundary , we assumed @xmath62 for the chabrier imf and @xmath63 for the others . this difference is , again , negligible , because the contribution of massive stars is insignificant due to steeply decreasing profiles . note that the lower and upper mass ends of the imf are changeable from the fiducial ones in this library . for pop iii stars , we adopt an entirely different mass range , @xmath64@xmath65 , reflecting the potential difference in formation of pop iii stars from ordinary stars due to their inherent cooling processes ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? ? * ; * ? ? ? * ) . in figure [ fig : imfs ] , we compare the shapes of these seven imfs . for ordinary populations , we can see that stars with masses of @xmath66 dominate in the six imfs except for the salpeter imf ; it has many more low - mass stars . we also find a substantial variation at the high mass end ; high mass end pop i / ii stars make the largest contribution in the chabrier imf , and the smallest in the miller - scalo and kroupa1993 imfs . these differences reflect the amount of energy released by sne ii ( see [ sec : modeling : snii ] and table [ tab : sniienergy ] ) . the pop iii imf has a peak at @xmath67 which surely follows the mass spectrum of the pop iii in @xcite . it is noted that the very massive stars ( @xmath68 ) with @xmath69 results in pair - creation instability sne @xcite and so far there is no clear relic which predicts the existence of the pair - creation instability sne @xcite . with our pop iii imf , the contribution of the pair - creation instability sne is negligible and thus , there is no discrepancy . there is a possibility that the imf is not universal . for instance , the top - heavy imf is favored to obtain a sufficient number of faint sub - mm galaxies in semi - analytical models @xcite ( see also * ? ? ? * which does not require a top - heavy imf to reproduce the number count of sub - mm galaxies ) . also , it is used in galaxy formation simulations @xcite . as is described in the above studies , the top - heavy imf is utilized for a case with violent events such as mergers and when strong shocks take place . recent observations indicate that imf slopes alter depending on host galaxy mass @xcite . @xcite reported that there is a tension between binary population synthesis results and observed agb stars in the galactic halo if the present - day imf is applied . in the current version of celib , the imf does not allow situation - dependent changes to its slopes and mass range , except for in the case of extremely metal poor stars . in this library , we adopt the metallicity dependent lifetime table provided by @xcite ( see their table 14 ) . the definition of the lifetime in @xcite is the sum of the h - burning and he - burning time scales of the stellar tracks of the padova library . the lifetimes depend slightly on the stellar metallicity . their table covers stars from @xmath70 to @xmath71 and from @xmath72 to @xmath73 . note that we use the linear interpolation in @xmath14 and the lifetime of @xmath72 ( @xmath73 ) if the metallicity is below ( above ) the value . it is also worth noting that there are other stellar lifetime data : both analytical forms ( e.g. , * ? ? ? * ; * ? ? ? * ) and grid data @xcite . the differences between them are not so large . for pop iii stars , we need the lifetime of stars @xmath74 . unfortunately , the lifetime table of @xcite did not provide the lifetime of this mass range . here , we use the lifetime data obtained by @xcite . he calculated the lifetime of the pop iii stars within a mass range of @xmath75 . to make the lifetime data of @xmath69 stars , we combine the @xcite s lifetime data of @xmath72 and the @xcite s lifetime data of @xmath69 . we employ the @xcite s lifetime data to less than @xmath71 stars whereas we adopt the @xcite s lifetime data of @xmath76@xmath65 stars . with these two sets of data , we can cover the full range of pop iii stars ( @xmath64@xmath65 ) . figure [ fig : lifetime ] shows the stellar lifetime as a function of mass . here , we interpolate both mass and metallicity and the linear interpolation is used for this process . apparently , the curve of the lifetime is not smooth . these `` bumps '' come from the original data , and they will propagate other quantities . thus , instead of using this data , we make polynomial functions using the least square method and use them as the fiducial ones . the procedure is described below . we assume that the lifetime distribution function can be described using the following polynomial function : @xmath77 ^k , \label{eq : lifetime : polynomial}\ ] ] where @xmath78 is the mass of stars in units of @xmath79 , @xmath80 is the lifetime of a star with @xmath78 in the unit of year , and @xmath81 , is the coefficient of @xmath82-th power of @xmath83 . coefficients are summarized in table [ tab : lifetime : polynomial ] . the first three coefficients for @xmath84 ( 9.99 , -3.55 , 2.60 ) are not far from those found in the other fitting formula for the stellar lifetime of eq . ( 5 ) in @xcite ( 10 , -4.319 , 1.543 ) which was based on @xcite and @xcite . using the five fitted functions and the results of their interpolations in the direction of the metallicity ( we use the linear interpolation ) , we obtain smoothed lifetime distribution functions . we show the lifetime distributions of five representative metallicities in figure [ fig : lifetime : polynomial ] . the typical difference between the original and the smoothed lifetimes at a given mass is less than @xmath85 . celib adopts these smoothed lifetime distribution functions as the fiducial ones . .coefficients of stellar lifetime distribution functions used in eq . . [ cols="^,^,^,^,^,^,^ " , ] + @xmath86 note that the modifications for the yields of c , mg , and fe are applied . see also section [ sec : snii ] . figure [ fig : chemodyn : disks ] displays the face - on and edge - on maps of the surface stellar density of runs a , b , c and d at @xmath87 . when we compare the results with the yields tables of @xcite and @xcite and without agbs , i.e. , runs a and b , we see that the stellar disk of run a is more compact and thicker than that of run b. the reason for this difference is that there is a relatively smaller amount of energy release from sne ii in run a compared to run b , as is expected from table [ tab : sniienergy ] . with this less efficient feedback , a dissipation - less nature is emphasized further . to clarify the contribution of agbs , we turn on the agb feedback , which is mainly the mass recycling processes involving newly synthesized metals ( runs c and d ) . we see from figure [ fig : chemodyn : disks ] that the stellar disk becomes larger in the radial direction . the return mass via agbs enhances the late epoch star formation , resulting in the extended disk . reflecting the fact that run c forms more stars , the contribution of agbs is also more prominent , resulting in a larger disk compared to that in run d. the impacts of the pop iii imf ( run e ) and hne ( runs f and g ) on galactic structures are found in figure [ fig : chemodyn : disks : efg ] . with the pop iii imf and @xcite s yields table , the released energy due to sne ii increases @xmath88 times larger than that without the pop iii imf ( see table [ tab : sniienergy ] ) . thus there are striking effects on the galactic structure . when we use the pop iii imf , the initial burst of star formation is suppressed significantly ( we argue it below ) . in this case , the stellar disk becomes smaller because the initial burst induced by the pop iii feedback removes the gas component from the galaxy . as is expected , the model with hne ( @xmath89 ) is almost no difference from the run without hne ( run c ) . the model with hne ( @xmath90 ) , on the other hand , has large impacts on the galactic structure since the amount of the released energy with @xmath90 is about four times larger than that with @xmath91 ( see table [ tab : sniienergy ] ) . this larger feedback energy results in the strong suppression of star formation and the removal of the log - angular momentum gas ( e.g. , * ? ? ? the disk becomes much smaller compared to runs c and f. figure [ fig : chemodyn : disks : hijk ] shows the runs involving both hne ( @xmath92 and @xmath93 ) and the pop iii imf , without and with metal mixing ( runs h , i , j , and k ) . in the model with the metal mixing ( run j ) , metals spread to the surrounding ism and the radiative cooling becomes more efficient compared to that without mixing ( run h ) . thus , the star formation is significantly enhanced and the disk size becomes larger when the metal mixing is involved . in the models with @xmath90 ( runs i and k ) , the difference is hard to see because of their strong feedback . note that these models are not models of a realistic galaxy formation through hierarchical mergers and hence , the contribution of metal mixing might be more complicated in the realistic model . we will study this in the future . star formation rates ( sfrs ) as a function of time for all runs are shown in figure [ fig : chemodyn : sfr ] . all of the models shown the upper panel have strong initial bursts of star formation and then they rapidly quench . the duration of the initial starburst depends on the adopted yields tables and model parameters . models whose released energy are large have shorter duration times : the duration time of the initial starburst of run b ( d ) is shorter than that of run a ( c ) . we find the contribution of agbs is easily seen in run c while not in run d. the reason why the run d does not show the enhancement of star formation at the late stage would be non - linear effects . the bottom panel of figure [ fig : chemodyn : sfr ] summarizes star formation histories with pop iii / hne / metal mixing . we see that the initial burst of the star formation in the model with pop iii ( run e ) is strongly enhanced : this enhancement of star formation is due to a large amount of return mass and metals . when the released energy reaches the typical potential energy of the central region , the gas component leaks from the galaxy and thus the star formation quenches . this effect is also found in the models involving both pop iii and hne ( for instance , runs h and j ) . when we turn on the hn mode ( runs f and g ) , the peaks of sfrs decrease . in the case with @xmath90 , the star formation is strongly suppressed and the peak of the sfr becomes less than @xmath94 , because of an extremely large release energy . the metal mixing model enhances star formation in the whole time . the peak sfr increases from @xmath95 ( run h , without mixing ) to @xmath96 ( run j , with mixing ) . the star formation in the late stage is kept a slightly higher sfr of @xmath97 if the metal mixing model is used . for the models with @xmath90 , the contribution of metal mixing is hard to see . figures [ fig : chemodyn : xfefeh : n13 ] and [ fig : chemodyn : xfefeh : p98 ] show [ x / fe]-[fe / h ] relations of nine elements for runs a , b , c , and d. we draw the median and 10% and 90% values as a function of [ fe / h ] in these figures . the evolution of [ x / fe ] in these figures is basically similar to those obtained by our one - zone models ( see figures [ fig : oz : afefeh : n ] and [ fig : oz : afefeh : p ] ) , although the metal redistribution scale is completely different . initially , there are plateaus consisting of the sne ii yields . then the values of [ x / fe ] become mixtures of sne ii , sne ia and agbs yields . in the cases with agbs in 3-dimensional simulations , the contribution of agbs are more prominent compared to the one - zone models and they are not only limited in light elements . this effect is more prominent in the run with the @xcite s yields table . this is because of the localization effect . the breaking points are shifted to + 0.5 dex from those of one - zone models due to the rapid evolution of sfr ( see figure [ fig : chemodyn : sfr ] ) . when the agbs are taken into account , not only the median but also the scatter has changed . in particular in the low metal stars , the scatter increases because of the early agbs ( recall figure [ fig : agbs : cumulativereturnmass ] ) . figure [ fig : chemodyn : xfefeh : popiiihne ] displays the [ x / fe]-[fe / h ] relations with the pop iii stars ( run e ) and with hne ( runs f and g ) . in these runs , the @xcite s yields table is used . the impact of the pop iii stars on [ x / fe]-[fe / h ] relations are significant . the effect of the pop iii stars is observed up to [ fe / h ] = @xmath98 ( note that @xmath99 ) . with the pop iii stars , the amounts of [ @xmath10/fe ] increase ( see figure [ fig : snii : yields : n13 ] ) and thus , [ @xmath10/fe]s at [ fe / h ] @xmath100 shift upward . this behavior is comparable to that found in the one - zone model in figure [ fig : oz : afefeh : n ] . however , these rapid changes in [ x / fe]-[fe / h ] relations are not observationally confirmed . we will see in the next figures , figures [ fig : chemodyn : xfefeh : nsms:005 ] and [ fig : chemodyn : xfefeh : nsms ] , that these gaps diminish / disappear when we adopt both the pop iii imf and hne . the contributions of hne to [ x / fe]-[fe / h ] relations can be seen as the timescale of the fe pollution . comparing [ x / fe]-[fe / h ] relations of runs f and g to those of run c , we find that the breaking points of [ x / fe]-[fe / h ] relations change from @xmath101 ( run c ) to @xmath102 ( runs f and g ) . this is because the larger fe yield of hne ( recall figure [ fig : snii : yields : n13hn ] ) . in the case with @xmath90 , the values of [ x / fe ] at sne ii plateaus decrease @xmath103 dex in [ o / fe ] , [ ne / fe ] , [ mg / fe ] , [ si / fe ] , [ s / fe ] and [ ca / fe ] , while that increases @xmath103 dex in [ ni / fe ] . we can see similar tendencies in the case with @xmath89 , although the changes are much smaller than the case with @xmath90 . from figure [ fig : chemodyn : xfefeh : nsms:005 ] , we can see the impact of metal mixing on the [ x / fe]-[fe / h ] relations . as is expected , the metal mixing greatly reduces scatters , which is identical to the result obtained in @xcite . the transition points from the pop iii sne to the normal sne ii and the breaking points of sne ii / ia are slightly moved toward high [ fe / h ] regions when the mixing is adopted . this is because that the self - enrichment is more efficient in the run with the mixing model . with a small amount of hne , the values of [ x / fe ] of the most metal - rich stars are almost identical to those without hne ( runs a and c ) . figure [ fig : chemodyn : xfefeh : nsms ] shows the [ x / fe]-fe / h relations with @xmath90 and with / without the metal mixing . although the star formation histories are completely different between cases with @xmath104 and @xmath89 , the positions of breaking points with @xmath105 are comparable to those found in the case with @xmath89 . the scatters in the run i are much wider than those in the run h. this would be that stars in the run i distribute farther from the galactic center and agbs affect more in the outer fresh gas , due to the more energetic feedback . when the metal mixing turns on , the scatters become narrow ( run k ) . the black curves ( run k ) have sudden rises at @xmath106 . their values of [ x / fe]s are identical to those expected by yields of sne ii . hence , the contribution of sne ii dominates in stars at the metallicity range . this might be an accidental case . the [ eu / fe]-[fe / h ] relations are shown in figures [ fig : chemodyn : xfefeh : nsms:005 ] and [ fig : chemodyn : xfefeh : nsms ] . the median values of [ eu / fe ] rise at [ fe / h ] @xmath102 and they saturate [ eu / fe ] @xmath107 at [ fe / h ] @xmath108 . these evolution tracks are expected by one - zone simulations . while we adopted @xmath109 and @xmath110 , the evolution tracks on the plane in figures [ fig : chemodyn : xfefeh : nsms:005 ] and [ fig : chemodyn : xfefeh : nsms ] are similar to the case with @xmath109 and @xmath111 in figure [ fig : oz : nsms : eu ] . this is because star - formation time scales in our three - dimensional simulations are shorter than those used in one - zone simulations . as a result , stars with low - z and high [ eu / fe ] are hard to see in our model , whereas stars with [ eu / fe ] @xmath112 and [ fe / h ] @xmath113 are found in observations ( see * ? ? ? the gap between our results and observations originate from the models we used in this paper . our models are too simple to express detailed distributions of chemical composition generated by rare events . we expect that the scatter will revert in a cosmological simulation because of its hierarchical nature . we will investigate the evolution of r - process elements in the cosmological context elsewhere in the near future . we developed a software library for the chemical evolution simulation of galaxy formation , named `` celib '' . this library adopts the ssp approximation and , under this approximation , it provides the return mass of each element and released energy from an ssp particle depending on feedback type . how these quantities are redistributed is left to the user s simulation code . all of the necessary data , such as imfs , stellar lifetime data , and yields are implemented in celib , as well as reference feedback models . the data is obtained from the literature . most functions are selectable at the runtime , and hence it is easy to compare the contribution of each model . the use of celib is easy if users follow the standard way because it consists of a limited number of apis . since celib is a simple software library , it is easy to carry out simulations of chemical evolution even for a beginner of this field . using internal functions , it is also possible to construct one s favorite model of chemical evolution . as demonstrations of celib , we showed the results of our simple one - zone models and three - dimensional chemodynamical simulations in a dark matter halo . the yields for sne ii have a large impact on chemical evolution , as is pointed out in previous studies . when we compare the results with the yields table of @xcite and that of @xcite with slight modifications , these two results are almost comparable except for the light elements , c and n. on the other hand , if we use the original yields of @xcite , the [ x / fe]-[fe / h ] relations become inconsistent with other results . sne ia affect the position of the breaking point of the plateaus . a different yields table gives different final amounts of metals released by sne ia . this implies that we need to be careful with models and yields of sne ia when we compare results with observations . agbs affect the evolutions of relatively light elements , such as c and n. these effects can be seen in both one - zone and three - dimensional simulations . the feedback from agbs is sometimes not taken into account even in current simulations of galaxy formation . however , they can not be ignored since they are very common in the ism . we provide the community with this library ( https://bitbucket.org/tsaitoh/celib ) and it is also archived on zenodo ( 10.5281/zenodo.190830 ) . we believe that this library will accelerate the understanding of galaxy formation from the perspective of chemical evolution . following are the future prospects of this library . * further flexibility would be necessary for imf shapes . it is more convenient if each stellar particle can have its own imf and mass range . the importance of the top - heavy imf in galaxy formation is well understood ( e.g. , * ? ? ? * ; * ? ? ? * further extension of yields tables is desirable . for instance , recently , the nugrid collaboration published their yields tables @xcite which are based on the mesa and genec codes @xcite . the published yields are @xmath114 and @xmath115 . we will take their yields when all data is published . in the current version of this library , all yields tables are obtained from models without stellar rotations . it is pointed out that stellar rotation changes yields ( e.g. , * ? ? ? * ; * ? ? ? the effect of the rotating zero metal stars was studied in @xcite and they showed these stars have significant effects on [ c / fe]-[fe / h ] and [ n / fe]-[fe / h ] relations and a moderate effect on the [ o / fe]-[fe / h ] relation . * further sophisticated treatment of the chemical enrichment might be important . strictly speaking , the yield of each element depends on the abundance of other elements . however , the current formulation implemented on celib ignores this effect . as such , the @xmath36 formalism has been proposed @xcite . with and without this formulation , there might be certain differences in chemical enrichment @xcite . note , however , that they assumed the solar proportions , which is not ideal for understanding the @xmath36 formulation . * further extension of available yields and isotopes are important to connect studies of galaxy formation and planet formation . in the current version of celib , the distributions of isotopes are not considered . it is also important to distinguish the contributions from different types of feedback . generally , only the data of long lived isotopes are provided . . this data is sufficient for galactic chemical evolution studies . however , in the case that the distributions of the short lived isotopes are important , the current treatment of the isotopes is insufficient . for example , it is pointed out that the radio isotope of aluminum , @xmath116al , is abundant during the early age of the solar system ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) and the decay heat of @xmath116al , whose half - time period is @xmath117 , is considered to be a primary source of the earth s early evolution ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) , as well as these of @xmath118u , @xmath119u , @xmath120th , and @xmath121k . to deal with the decays of all of radioactive isotopes is unrealistic . at least it is necessary to deal with the decays of some of the important radioactivate isotopes to understand the formation history of stars and planets from the perspective of galactic chemodynamical simulations . it is certain that star - by - star simulations are the next breakthrough of galaxy formation simulation since there are an enormous amount of evidence that massive stars have crucial impacts on the galaxy formation and evolution . in some first star and first galaxy simulations , pop iii stars are dealt with discrete stars sampled from a pop iii imf instead of using the ssp approximation ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) or its formation is directly followed ( e.g. , * ? ? ? * ; * ? ? ? while such treatment of stars is beyond the original scope of celib , the all necessary data used for the chemical evolution of star - by - star simulations have been implemented . hence , celib provides apis which can be used by star - by - star simulation ( see appendix [ sec : apis : starbystar ] ) . the author thanks the anonymous referee who gave constructive and helpful comments that improved this study . the author also thanks yutaka hirai , takashi okamoto , junichi baba , daisuke kawata , ko nakamura , takuma suda and yutaka katsuta who gave important input for this study . a part of numerical simulations was carried out on the cray xc30 system in the center for computational astrophysics at the national astronomical observatory of japan . this work is supported by a grant - in - aid for scientific research ( 26707007 ) of japan society for the promotion of science and strategic programs for innovative research of the ministry of education , culture , sports , science and technology ( spire ) . here we describe the major functions of celib . all available functions are defined in celib.h . this library is initialized by just calling this function ; .... void celibinit(void ) ; .... when this function is called , the imf and its mass range are fixed . then under these conditions , the imf weighted yields of the adopted yield tables are computed . the smoothed lifetime functions and necessary data for reference feedback models are also computed . this function should be called at the beginning of a simulation using this library . every model s parameters are fixed at this time . if a user wants to use different model parameters , the user needs to call this function again after resetting new parameters . with reference feedback models , the user can easily obtain event times . in order to obtain an event time of an ssp particle , the user has to use the following function ; .... double celibgetnexteventtime(struct celibstructnexteventtimeinput input , const int type ) ; .... this function returns the event time of a target event in units of year . this function requires two arguments . the first argument is a structure defined in ` celib.h ` and the definition of it is .... struct celibstructnexteventtimeinput { double r ; // a random real number in [ 0,1 ) double initialmass_in_msun ; // an initial mass of the target ssp particle // in units of the solar mass double metallicity ; // a metallicity of the target ssp particle int count ; // a counter for a target event // this is used in sne ia / agbs / nsms // count should start zero . } ; .... the second argument of this reference api is used to specify the feedback type . feedback types are defined as ` enum ` and the user should select one out of four ; .... enum { celibfeedbacktype_snii , celibfeedbacktype_snia , celibfeedbacktype_agb , celibfeedbacktype_nsm , celibfeedbacktype_number , } ; .... for example , we consider a case in which the user wants to obtain the explosion time of sne ii using the reference model . in this case , the user has to call the reference api like this ; .... struct celibstructnexteventtimeinput input = { .r = a_r , // a random real number in [ 0,1 ) .initialmass_in_msun = m_ssp , // the initial mass of the target ssp particle .metallicity = z_ssp , // the metallicity of the target ssp particle } ; double t_snii = celibgetnexteventtime(input , celibfeedbacktype_snii ) ; .... ` count ` is not used to obtain the feedback time of sne ii and thus it is ignored in this case . the return value ` t_snii ` is the explosion time which follows figures [ fig : snii : rate : n13 ] or [ fig : snii : rate : p98 ] . if the simulation time is @xmath122 , the feedback event takes place at @xmath123 . the released masses of metals and energy are also easily obtained by using a reference api . when the time in a simulation reaches the event time , we need to call the following function ; .... struct celibstructfeedbackoutput celibgetfeedback(struct celibstructfeedbackinput input , const int type ) ; .... the first argument is the structure defined in ` celib.h ` and it holds all the necessary data to evaluate the feedback event . the structure is .... struct celibstructfeedbackinput { double mass ; // the mass of the target ssp particle in simulation unit double metallicity ; // the metallicity of the target ssp particle double massconversionfactor ; // a factor to convert elements [ ] from // the simulation mass unit to the solar mass double * elements ; // the pointer to the array of elements for // the target ssp particle in the simulation mass unit int count ; // a counter of the target event } ; .... the second argument is the type of feedback and it is the same as that used to obtain the feedback time . the results are loaded to a structure whose type is ` struct celibstructfeedbackoutput ` . the definition of it is as follows : .... struct celibstructfeedbackoutput { double energy ; // the released energy in units of erg double ejectamass ; // the ejecta mass in units of the solar mass double remnantmass ; // the remnant mass in units of the solar mass double elements[celibyield_number ] ; // the mass of released metals in units of the solar mass } ; .... note that ` celibyield_number ` is defined in celib.h and is thirteen in the current version . for example , we show the way to obtain the results of sne ii feedback . first the user needs to put all necessary data to the structure ` struct celibstructfeedbackinput ` and then call ` celibgetfeedback ` with ` celibfeedbacktype_snii ` . .... struct celibstructfeedbackinput input = { .mass = m_ssp , .metallicity = z_ssp , .massconversionfactor = mass_solar_mass / mass_sim_unit , .elements = elements_ssp , } struct celibstructfeedbackoutput snii = celibgetfeedback(input , celibfeedbacktype_snii ) ; .... units of energy and mass in ` snii ` are erg for energy and @xmath79 . if the simulation units are different from these units , the user needs to convert them into the simulation units . all of the control parameters are stored in the structure ` struct celibstructrunparameters ` . celib prepares a structure ` celibrunparameters ` and it is used to manage this library . the fiducial model is the same as that used in [ sec : chemodyn ] as run f ( see table [ tab : chemodyn : runs ] ) . when the values in ` celibrunparameters ` are changed and the initializer is called , celib recomputes all data and thus a new simulation is ready to start . celib provides apis for star - by - star simulations . so far , celib supports only feedback from massive stars . this function returns the lifetime of a star in units of year , referring the lifetime tables built in [ sec : lifetime ] . the structure of the first argument , which is also defined in ` celib.h ` , is as follows : the second argument is the type of feedback . although only feedback from massive stars is supported in the current version of celib , we prepare this argument for the future extension . in the current version , ` type ` should be ` celibfeedbacktype_snii ` . [ [ get - released - masses - of - metals - and - energy - for - star - by - star - simulations ] ] get released masses of metals and energy for star by star simulations ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ .... struct celibstructfeedbackstarbystarinput { double mass ; // mass of the star in simulation unit . double metallicity ; // metallicity of the star z double massconversionfactor ; // a factor to convert elements [ ] from // the simulation mass unit to msun . double * elements ; // star particle 's elements composition in simulation unit . } ; .... .... struct celibstructfeedbackstarbystaroutput { double energy ; // the released energy in units of erg double ejectamass ; // the ejecta mass in units of the solar mass double remnantmass ; // the remnant mass in units of the solar mass double elements[celibyield_number ] ; // the mass of released metals in units of the solar mass } ; ....

given the current landscape in experimental high - energy physics , these lectures are focused on applications of event generators for hadron colliders like the tevatron and lhc . much of it would also be relevant for @xmath0 machines like lep and ilc or @xmath1 machines like hera , but with some differences not discussed here . heavy - ion physics is not at all addressed , since it involves rather different aspects , specifically the potential formation of a quark gluon plasma . further , within the field of high - energy @xmath2 collisions , the emphasis will be on the common aspects of qcd physics that occurs in all collisions , rather on those aspect that are specific to a particular physics topic , such as @xmath3 production or supersymmetry . both heavy ions and other physics topics are instead covered by other lectures at this school . section 2 contains a first overview of the physics picture and the generator landscape . thereafter section 3 describes the usage of matrix elements , section 4 the important topics of initial- and final - state showers , and section 5 how showers can be matched to different hard processes . the issue of multiple interactions and their role in mimimum - bias and underlying - event physics is introduced in section 6 , followed by some brief comments on hadronization in section 7 . the article concludes with an outlook on the ongoing generator - development work in section 8 . slides for these and other similar lectures @xcite are complementary to this writeup in style and contents , including many ( colour ) illustrations absent here . other useful resources include the `` les houches guidebook to monte carlo generators for hadron collider physics '' @xcite and a recent review on qcd physics at the tevatron and lhc @xcite . in real life , machines produce events that are stored by the data acquisition system of a detector . in the virtual reality , event generators like herwig @xcite and pythia @xcite play the role of machines like the tevatron and lhc , and detector simulation programs like geant 4 @xcite the role of detectors like atlas or cms . the real and virtual worlds can share the same event reconstruction framework and subsequent physics analysis . it is by understanding how an original physics input is distorted step - by - step in the better - controlled virtual world that an understanding can be gained of what may be going on in the real world . for approximate studies the detector simulation and reconstruction steps can be shortcut , so that generators can be used directly in the physics studies . a number of physics analyses would not be feasible without generators . specifically , a proper understanding of the ( potential ) signal and background processes is important to separate the two . the key aspect of generators here is that they provide a detailed description of the final state so that , ideally , any experimental observable or combination of observables can be predicted and compared with data . thereby generators can be used at various stages of an experiment : when optimizing the detector and its trigger design to the intended physics program , when estimating the feasibility of a specific physics study , when devising analysis strategies , when evaluating acceptance corrections , and so on . however , it should always be kept in mind that generators are not perfect . they suffer from having to describe a broad range of physics , some of which is known from first principles , while other parts are modelled in different frameworks . ( in the latter case , a generator actually acts as a vehicle of ideology , where ideas are disseminated in prepackaged form from theorists to experimentalists . ) given the limited resources , different authors may also have invested more or less time on specific physics topics , and therefore these may be more or less well modelled . it always pays to shop around , and to compare several approaches before drawing too definite conclusions . blind usage of a generator is not encouraged : then you are the slave rather than the master . why then _ monte carlo _ event generators ? basically because einstein was wrong : god does throw dice ! in quantum meachanics , calculations provide the _ probability _ for different outcomes of a measurement . event - by - event , it is impossible to know beforehand what will happen : anything that is at all allowed could be next . it is only when averaging over large event samples that the expected probability distributions emerge provided we did the right calculation to high enough accuracy . in generators , ( pseuo)random numbers are used to make choices intended to reproduce the quantum mechanical probabilities for different outcomes at various stages of the process . the buildup of the structure in an event occurs in several steps , and can be summarized as follows : * initially two hadrons are coming in on a collision course . each hadron can be viewed as a bag of partons quarks and gluons . * a collision between two partons , one from each side , gives the hard process of interest , be it for physics within or beyond the standard model : @xmath4 , @xmath5 , @xmath6 , etc . ( actually , the bulk of the cross section results in rather mundane events , with at most rather soft jets , or of a simple elastic or diffractive character that is not easily described as partonic processes . such events usually are filtered away at an early stage , however . ) * when short - lived `` resonances '' are produced in the hard process , such as the top , @xmath7 or @xmath8 , their decay has to be viewed as part of this process itself , since e.g. spin correlations are transferred from the production to the decay stages . * a collision implies accelerated colour ( and often electromagnetic ) charges , and thereby bremsstrahlung can occur . emissions that can be associated with the two incoming colliding partons are called initial - state radiation ( isr ) . as we shall see , such emissions can be modelled by so - called space - like parton showers . * emissions that can be associated with outgoing partons are instead called final - state radiation ( fsr ) , and can be approximated be time - like parton showers . often the distinction between a hard process and isr and fsr is ambiguous , as we shall see . * so far we only extracted one parton from each incoming hadron to undergo a hard collision . but the hadron is made up of a multitude of further partons , and so further parton pairs may collide within one single hadron hadron collision multiple interactions ( mi ) . ( not to be confused with pileup events , when several hadron pairs collide during a bunch bunch crossing , but with obvious analogies . ) * each of these further collisions also may be associated with its isr and fsr . * the colliding partons take a fraction of the energy of the incoming hadrons , but much of the energy remains in the beam remnants , which continue to travel essentially in the original directions . these remnants also carry colours that compensate the colour taken away by the colliding partons . * at short timescales , when partons are close to each other , the principle of asymptotic freedom tells us that we can think of each parton as freely moving along its trajectory . however , as time goes by and the partons recede from each other , confinement forces become significant . the structure and time evolution of these force fields can not be described from first principles within any calculational technique currently at our disposal , so models have to be introduced . one common approach is to assume that a separate confinement field is stretched between each colour and its matching anticolour , with each gluon considered as a simple sum of a colour and an anticolour , and all colours distinguishable from each other ( the @xmath9 limit ) . * such fields can break up by the production of new quark antiquark pairs that screen the endpoint colours , and where a quark from one break ( or from an endpoint ) can combine with an antiquark from an adjacent break to produce a primary hadron . this process is called hadronization . * many of those primary hadrons are unstable and decay further at various timescales . some are sufficiently long lived that their decays are visible in a detector , or are ( almost ) stable . thereby we have reached scales where the event - generator description has to be matched to a detector - simulation framework . * it is only at this stage that experimental information can be obtained and used to reconstruct back what may have happened at the core of the process , e.g. whether a higgs particle was produced or not . the monte carlo method allows these steps to be considered sequentially , and within each step to define a set of rules that can be used iteratively to construct a more and more complex state , maybe ending with hundreds of particles moving out in different directions . since each particle contains of the order of ten degrees of freedom ( flavour , mass , momentum , production vertex , lifetime , ) we realize that thousands of choices are involved for a typical event . the aim is to have a sufficiently realistic description of these choices that both the average behaviour and the fluctuations around this average are well decribed . schematically , the cross section for a range of final states is provided by @xmath10 properly integrated over the relevant phase - space regions and summed over possible `` paths '' ( of showering , hadronization , etc . ) that lead from a hard process to the final state . that is , the dimensional quantities are associated with the hard process ; subsequent steps are handled in a probabilistic approach . the spectrum of event generators is very broad , from general - purpose ones to more specialized ones . herwig and pythia are the two most commonly used among the former ones , with isajet @xcite and sherpa @xcite as the other two main programs in this category . among more specialized programs , many deal with the matrix elements for some specific set of processes , a few with topics such as parton showers or particle decays , but there are e.g. no freestanding programs that handle hadronization . in the end , many of the specialized programs are therefore used as `` plugins '' for the general - purpose ones . from the lagrangian of a theory the feynman rules can be derived , and from them matrix elements can be calculated . combined with phase space it allows the calculation of cross sections . as a simple example consider the scattering of quarks in qcd , say @xmath11 , a process similar to rutherford scattering but with gluon exchange instead of photon ditto . the mandelstam variables are defined as @xmath12 , @xmath13 and @xmath14 . in the cm frame of the collision @xmath15 is the squared total energy and @xmath16 where @xmath17 is the scattering angle . the differential cross section is then @xmath18 which diverges roughly like @xmath19 for transverse momentum @xmath20 . we will come back to this issue when discussing multiple interactions ; for now suffice to say that some lower cutoff @xmath21 need to be introduced . similar cross sections , differing mainly by colour factors , are obtained for @xmath22 and @xmath23 . a few further qcd graphs , like @xmath24 , are less singular and give smaller contributions . these cross sections then have to be convoluted with the flux of the incoming partons @xmath25 and @xmath26 in the two incoming hadrons @xmath27 and @xmath3 : @xmath28 the parton density functions ( pdfs ) of gluons and sea quarks are strongly peaked at small momentum fractions @xmath29 , @xmath30 . this further enhances the peaking of the cross section at small @xmath31 values . nevertheless , with high machine luminosity the jet cross section can be studied out to quite high values . the cross section of other processes can be suppressed by two main effects . firstly , for massive particles the @xmath31 spectrum is strongly dampened below the respective mass scale , and it is only above it that these precesses have a chance to stand up above the qcd background . secondly , the processes may involve electroweak ( or other small ) couplings rather than strong ones . in order to address the physics of interest a large number of processes , both within the standard model and in various extensions of it , have to be available in generators . indeed many can also be found in the general - purpose ones , but by far not enough . further , often processes are there available only to lowest order , while experimental interest may be in higher orders , with more jets in the final state , either as a signal or as a potential background . so a wide spectrum of matrix - element - centered programs are available @xcite , some quite specialized and others more generic . ( 320,210)(0,0 ) ( 100,185)(220,205 ) ( 160,195)[]process selection ( 100,165)(220,185 ) ( 160,175)[]resonance decays ( 160,165)(160,152 ) ( 100,130)(220,150 ) ( 160,140)[]parton showers ( 100,110)(220,130 ) ( 160,120)[]multiple interactions ( 100,90)(220,110 ) ( 160,100)[]beam remnants ( 160,90)(160,77 ) ( 100,55)(220,75 ) ( 160,65)[]hadronization ( 100,35)(220,55 ) ( 160,45)[]ordinary decays ( 160,35)(160,22 ) ( 100,0)(220,20)(160,10)[]detector simulation ( 0,185)(80,205 ) ( 40,195)[]me generator ( 40,185)(40,172 ) ( 0,150)(80,170 ) ( 40,160)[]me expression ( 80,165)(98,195 ) ( 80,160)(98,180 ) ( 0,85)(80,135 ) ( 40,125)[]susy/ ( 40,110)[]spectrum ( 40,96)[]calculation ( 40,135)(40,148 ) ( 80,125)(98,170 ) ( 240,170)(320,205 ) ( 280,195)[]phase space ( 280,180)[]generation ( 230,195)(238,190 ) ( 230,195)(222,200 ) ( 240,115)(320,135 ) ( 280,125)[]pdf library ( 240,130)(222,190 ) ( 240,125)(222,140 ) ( 240,120)(222,120 ) ( 240,60)(320,80 ) ( 280,70)[]@xmath32 decays ( 240,70)(222,50 ) ( 240,25)(320,45 ) ( 280,35)[]@xmath3 decays ( 240,35)(222,40 ) ( 100,28.75)(13.75,270,90)5 ( 100,33.75)(28.75,270,90)5 ( 80,155)(98,120 ) the way these programs can be combined with a general - purpose generator is illustrated in fig . [ fig : combgen ] . in the study of supersymmetry ( susy ) it is customary to define a model in terms of a handful parameters , e.g. specified at some large grand unification scale . it is then the task of a spectrum calculator to turn this into a set of masses , mixings and couplings for the physical states to be searched for . separately , the matrix elements can be calculated with these properties as unknown parameters , and only when the two are combined is it possible to speak of physically relevant matrix - element expressions . these matrix elements now need to be combined with pdf s and sampled in phase space , preferable with some preweighting procedure so that regions of phase space with high cross sections are sampled more frequently . the primarily - produced susy particles typically are unstable and undergo sequential decays down to a lightest supersymmetric particle ( lsp ) , again with branching ratios and angular distributions that should be properly modelled . the lsp would be neutral and escape undetected , while other decay products would be normal quarks and leptons . it is at this stage that general - purpose programs take over . they describe the showering associated with the above process , the presence of additional interactions in the same hadron hadron collision , the structure of beam remnants , and the hadronization and decays . they would still rely on the externally supplied pdf s , and potentially make use of programs dedicated to @xmath32 and @xmath3 decays , where spin information and form factors require special encoding . even after the event has been handed on to the detector - simulation program some parts of the generator may be used in the simulation of secondary interactions and decays . several standards have been developed to further this interoperability . the les houches accord ( lha ) for user processes @xcite specifies how parton - level information about the hard process and sequential decays can be encoded and passed on to a general - purpose generator . originally it was defined in terms of two fortran commonblocks , but more recently a standard les houches event file format @xcite offfers a language - independent alternative approach . the les houches accord parton density functions ( lhapdf ) library @xcite makes different pdf sets available in a uniform framework . the susy les houches accord ( slha ) @xcite allows a standardized transfer of masses , mixings , couplings and branching ratios from spectrum calculators to other programs . finally , the hepmc c++ event record @xcite succeeds the hepevt fortran one @xcite as a standard way to transfer information from a generator on to the detector - simulation stage . one of the key building blocks for several of these standards is the pdg codes for all the most common particles @xcite , also in some scenarios for physics beyond the standard model . the @xmath33 processes we started out with above are about the simplest one can imagine at a hadron collider . in reality one needs to go on to higher orders . in @xmath34 two new kind of graphs one kind is where one additional parton is present in the final state , i.e. @xmath35 processes . the cross section for such processes is almost always divergent when one of the parton energies vanish ( soft singularities ) or two partons become collinear ( collinear singularities ) . the other kind is loop graphs , with an additional intermediate parton not present in the final state , i.e. a correction to the @xmath33 processes . strictly speaking , at @xmath34 one picks up the interference between the lowest - order graph and the loop graph , and this interference has negative divergences that exactly cancel the positive ones above , with only finite terms surviving . for inclusive event properties such next - to - leading order ( nlo ) calculations lead to an improved accuracy of predictions , but for more exclusive studies the mathematical cancellation of singularities has to be supplemented by more physical techniques , which is far from trivial . the tricky part of the calculations is the virtual corrections . nlo is now state - of - the - art , with nnlo still in its infancy . if one is content with born - level diagrams only , i.e. without any loops , it is possible to go to quite high orders , with up to something like eight partons in the final state . these partons have to be kept well separated to avoid the phase - space regions where the divergences become troublesome . in order to cover also regions where partons become soft / colliner we therefore next turn our attention to parton showers . to iterate , the emission rate for a branching such as @xmath36 diverges when the gluon either becomes collinear with the quark or when the gluon energy vanishes . the qcd pattern is similar to that for @xmath37 in qed , except with a larger coupling , and actually a coupling that also increases for smaller relative @xmath31 in a branching , thereby further enhancing the divergence . furthermore the non - abelian character of qcd leads to @xmath38 branchings with similar divergences , without any correspondence in qed . the third main branching , @xmath39 with its @xmath40 qed equivalence , does not have the soft divergence and is less important . now , if the rate for one emission of a gluon is big , then also the rate for two or more will be big , and thus the need for high orders and many loops in matrix - element - based descriptions . with showers we introduce two new concepts that make like easier , + ( 1 ) an iterative structure that allows simple expressions for @xmath36 , @xmath38 and @xmath39 branchings to be combined to build up complex multiparton final states , and + ( 2 ) a sudakov factor that offers a physical way to handle the cancellation between real and virtual divergences . + neither of the simplifications is exact , but together they allow us to provide sensible approximate answers for the structure of emissions in soft and collinear regions of phase space . ( 240,160)(0,-10 ) ( 20,20)(100,60)(15,20)[r]@xmath41 ( 20,140)(100,100)(15,140)[r]@xmath42 ( 100,60)(100,100)45 ( 100,60)(180,20)(185,20)[l]@xmath41 ( 100,100)(180,140)(185,140)[l]@xmath42 ( 100,80)(35,0,355)5 ( 115,80)[]@xmath43 ( 100,-10)[]@xmath33 ( 50,35)(100,20)46 ( 50,50)(68,45)(45,50)[r]@xmath44 ( 50,125)(100,140)46 ( 50,110)(68,115)(45,110)[r]@xmath45 ( 50,-10)[]isr ( 140,40)(190,55)46 ( 130,27)(130,43)(130,22)[t]@xmath46 ( 140,120)(190,105)46 ( 130,133)(130,117)(130,138)[b]@xmath47 ( 150,-10)[]fsr the starting point is to `` factorize '' a complex @xmath48 process , where @xmath49 represents a large number of partons in the final state , into a simple core process , maybe @xmath33 , convoluted with showers , fig . [ fig : factorize ] . to begin with , in a simple @xmath50 process the incoming and outgoing quarks must be on the mass shell , i.e. satisfy @xmath51 , at long timescales . by the uncertainty principle , however , the closer one comes to the hard interaction , i.e the shorter the timescales considered , the more off - shell the partons may be . ( in the oldfashioned perturbative language particles were always on mass - shell , and the uncertainty relation allowed energy not to be conserved temporarily . in the modern feynman - graph language four - momentum is conserved at each vertex , but intermediate `` propagator '' particles need not be on the mass shell . the final physics is the same in both languages . ) thus the incoming quarks may radiate a succession of harder and harder gluons , while the outgoing ones radiate softer and softer gluons . one definition of hardness is how off - shell the quarks are , @xmath52 , but we will encounter other variants later . in the initial - state radiation ( isr ) part of the cascade these virtualities are spacelike , @xmath53 , hence the alternative name spacelike showers . correspondingly the final - state radiation ( fsr ) is characterized by timelike virtualities , @xmath54 , and hence also called timelike showers . to see where this distinction comes from , consider the kinematics of an arbitrary branching @xmath55 , with @xmath56 defined to be moving along the @xmath57 axis . then it is useful to introduce lightcone momenta @xmath58 , so that the relation @xmath59 translates to @xmath60 . now define the splitting of @xmath61 by @xmath62 and @xmath63 . obviously @xmath64 so @xmath65 . remains to ensure conservation of @xmath66 : @xmath67 in an initial - state branching the incoming @xmath56 should be ( essentially ) massless , and if @xmath68 does not interact any further it should also be massless . this gives @xmath69 , a virtuality that is acceptable if @xmath70 is on its way in to a hard scattering , i.e.is destined only to live for a short while . in isr the @xmath71 virtualities , such as @xmath45 and @xmath44 in fig . [ fig : factorize ] , are usually defined as @xmath72 to keep them positive definite . for a final - state branching , assume that @xmath70 and @xmath68 will not branch any further and thus are massless , while @xmath56 is an intermediate particle coming from the hard interaction . then @xmath73 , cf . @xmath47 and @xmath46 in fig . [ fig : factorize ] , here with @xmath74 . the cross section for the whole @xmath48 graph is associated with the cross section of the hard subprocess , with the approximation that the other @xmath71 virtualities can be neglected in the matrix - element expression . in the limit that all the @xmath75 this should be a good approximation . in other words , first the hard process can be picked without any reference to showers , and only thereafter are showers added with unit probability . but , of course , the showers do modify the event shape , so at the end of the day the cross section is affected . for instance , the total transverse energy @xmath76 of an event is increased by isr , so the cross sections of events with a given @xmath76 is increased by the influx of events that started out with a lower @xmath76 in the hard process . it is important that the hard - process scale @xmath43 is picked to be the largest one , i.e. @xmath77 in fig . [ fig : factorize ] . if e.g. @xmath78 then instead the @xmath4 subgraph ought to be chosen as hard process , and the gluon of virtuality @xmath43 ought to be part of the isr off the incoming @xmath41 . without such a criterion one might doublecount a given graph , or even count it once for every possible subgraph inside the complete @xmath48 graph . in addition , the approximation of neglecting virtualities in the hard - scattering matrix elements obviously becomes worse the more the incoming and outgoing partons are off - shell , another reason not to put a larger scale than necessary in the shower part . ( 180,90)(0,-50 ) ( 0,0)(40,0)44(20,15)[]0 ( 40,0)(110,-40)(120,-40)[l]@xmath79 ( 1 ) ( 110,40)(40,0)(120,40)[l]@xmath80 ( 2 ) ( 55,-18)[]@xmath25 ( 75,-20)(110,-10)44(120,-10)[l]@xmath81 ( 3 ) ( 70,-50)[](a ) ( 150,90)(0,-50 ) ( 0,0)(40,0)44(20,15)[]0 ( 40,0)(110,-40)(120,-40)[l]@xmath79 ( 1 ) ( 110,40)(40,0)(120,40)[l]@xmath80 ( 2 ) ( 55,18)[]@xmath25 ( 75,20)(110,10)44(120,10)[l]@xmath81 ( 3 ) ( 70,-50)[](b ) let us next turn to a more detailed presentation of the showering approach , and begin with the simpler final - state stage . this is most cleanly studied in the process @xmath82 . the first - order correction here corresponds to the emission of one additional gluon , by either of the two feynman graphs in fig . [ fig : qqg ] . neglect quark masses and introduce energy fractions @xmath83 in the rest frame of the process . then the cross section is of the form @xmath84 where @xmath85 is the @xmath86 cross section , i.e.without the gluon emission . now study the kinematics in the limit @xmath87 . since @xmath88 we see that this corresponds to the `` collinear region '' , where the separation between the @xmath79 and @xmath81 vanishes . equivalently , the virtuality @xmath89 of the intermediate quark propagator @xmath25 in fig . [ fig : qqg]a vanishes . although the full answer contains contributions from both graphs it is obvious that , in this region , the amplitude of the one in fig . [ fig : qqg]a dominates over the one in fig . [ fig : qqg]b . we can therefore view the process as @xmath90 followed by @xmath36 . define the energy sharing in the latter branching by @xmath91 and @xmath92 . the kinematics relations then are @xmath93 so that @xmath94 here @xmath95 corresponds to the `` collinear '' or `` mass '' singularity and @xmath96 to the soft - gluon singularity . the interesting aspect of eq . ( [ eq : pqtoqg ] ) is that it is universal : whenever there is a massless quark in the final state , this equation provides the probability for the same final state except for the quark being replaced by an almost collinear @xmath97 pair ( plus some other slight kinematics adjustments to conserve overall energy and momentum ) . that is reasonable : in a general process any number of distinct feynman graphs may contribute and interfere in a nontrivial manner , but once we go to a collinear region only one specific graph will contribute , and that graphs always has the same structure , in this case with an intermediste quark propagator . corresponding rules can be derived for what happens when a gluon is replaced by a collinear @xmath98 or @xmath99 pair . these rules are summarized by the dglap equations @xcite @xmath100 furthermore , the rules can be combined to allow for the successive emission in several steps , e.g. where a @xmath36 branching is followed by further branchings of the daughters . that way a whole shower develops , fig . [ fig : fsrshower ] . ( 240,160)(-70,-80 ) ( -60,0)(0,0)46 ( 0,0)(160,80 ) ( 0,0)(160,-80 ) ( 32,-16)(80,-8)45 ( 80,-8)(160,-16)48(80,-8)(160,0)48 ( 64,-32)(112,-40)45 ( 112,-40)(160,-32)45(112,-40)(160,-48)45 ( 96,-48)(160,-64)47 ( 48,24)(96,24)45 ( 96,24)(160,16)(96,24)(160,32 ) ( 80,40)(160,48)48 ( 120,60)(160,64)44 such a picture should be reliable in cases where the emissions are strongly ordered , i.e. @xmath101 . showers would not be useful if they only could be applied to strongly - ordered parton configurations , however . a further study of the @xmath102 example shows that the simple sum of the @xmath36 and @xmath103 branchings reproduce the full matrix elements , with interference included , to better than a factor of 2 over the full phase space . this is one of the simpler cases , and of course one should expect the accuracy to be worse for more complicated final states . nevertheless , it is meaningful to use the shower over the whole strictly - ordered , but not necessarily strongly - ordered , region @xmath104 to obtain an approximate answer for multiparton topologies for which the complete matrix elements would be too lengthy . with the parton - shower approach , the big probability for one branching @xmath105 turns into a big probability for several successive branchings . nevertheless we did not tame the fact that probabilities blow up in the soft and collinear regions . for sure , perturbation theory will cease to be meaningful at so small @xmath43 scales that @xmath106 diverges ; there confimenent effects and hadronization phenomena take over . typically therefore some lower cutoff at around 1 gev is used to regulate both soft and collinear divergences : below such a scale no further branchings are simulated . whatever perturbative effects may remain are effectively pushed into the parameters of the nonperturbative framework . that way we avoid the singularities , but we can still have `` probabilities '' well above unity , which does not seem to make sense . this brings us to the second big concept of this section , the _ sudakov ( form ) factor _ @xcite . in the context of particle physics it has a specific meaning related to the properties of the loop diagrams , but more generally we can just see it as a consequence of the conservation of total probability @xmath107 where the former is multiplicative in a time - evolution sense : @xmath108 now subdivide further , with @xmath109 , @xmath110 : @xmath111 \longrightarrow ~~ \mathrm{d}\mathcal{p}_{\mathrm{first}}(t ) & = & \mathrm{d}\mathcal{p}_{\mathrm{something}}(t ) \ ; \exp \left ( - \int_0^t \frac{\mathrm{d } \mathcal{p}_{\mathrm{something}}(t)}% { \mathrm{d } t } \mathrm{d } t \right ) ~.\end{aligned}\ ] ] that is , the probability for something to happen for the _ first _ time at time @xmath112 is the naive probability for this to happen , _ times _ the probability that this did not yet happen . as such it applies to a host of situations . take the example of football ( relevant at the time of the school ) . assume that players are equally energetic and skillful from the first minute of the match to the last . then the chances of scoring a goal is uniform in time , but the probability of scoring the _ first _ goal of the match is bigger at the beginning , because later on any goal could well be the second or third . in physics a common example is that of radioactive decay . if the number of undecayed radioactive nuclei at time @xmath113 is @xmath114 , with initial number @xmath115 at time @xmath116 , then a naive ansatz would be @xmath117 , where @xmath68 parametrizes the decay likelihood per unit of time . this equation has the solution @xmath118 , which becomes negative for @xmath119 , because by then the probability for having had a decay exceeds unity . so what we made wrong was not to take into account that only an undecayed nucleus can decay , i.e. that the equation ought to have been @xmath120 with the solution @xmath121 . this is a nicely well - behaved expression , where the total probability for decays goes to unity only for @xmath122 . if @xmath68 had not been a constant but varied in time , @xmath123 , it is simple to show that the solution instead would have become @xmath124 for a shower the relevant `` time '' scale is something like @xmath125 , by the heisenberg uncertainty principle . that is , instead of evolving to later and later times we evolve to smaller and smaller @xmath43 . thereby the dglap eq . ( [ eq : dglap ] ) becomes @xmath126 where the exponent ( or simple variants thereof ) is the sudakov factor . as for the radioactive - decay example above , the inclusion of a sudakov ensures that the total probability for a parton to branch never exceeds unity . then you may have sequential radioactive decay chains , and you may have sequential parton branchings , but that is another story . it is a bit deeper than that , however . just as the standard branching expressions can be viewed as approximations to the complete matrix elements for real emission , the sudakov is an approximation to the complete virtual corrections from loop graphs . the divergences in real and virtual emissions , so strange - looking in the matrix - element language , here naturally combine to provide a physical answer everywhere . what is not described in the shower , of course , is the non - universal finite parts of the real and virtual matrix elements . the implementation of a cascade evolution now makes sense . starting from a simple @xmath99 system the @xmath79 and @xmath80 are individually evolved downwards from some initial @xmath127 until they branch . at a branching the mother parton disappears and is replaced by two daughter partons , which in their turn are evolved downwards in @xmath43 and may branch . thereby the number of partons increases , until the lower cutoff scale is reached . this does not mean that everything is uniquely specified . in particular , the choice of evolving in @xmath128 is by no means obvious . any alternative variable @xmath129 would work equally well , since @xmath130 . alternative evolution variables therefore include the transverse momentum , @xmath131 , and the energy - weighted emission angle @xmath132 . both these two alternative choices are favourable when the issue of _ coherence _ is introduced . coherence means that emissions do not occur independently . for instance , consider @xmath133 , followed by an emission of a gluon either from 2 or 3 . when this gluon is soft it can not resolve the individual colour charges of @xmath134 and @xmath135 , but only the net charge of the two , which of course is the charge of @xmath136 . thereby the multiplication of partons in a shower is reduced relative to naive expectations . as it turns out , evolution in @xmath31 or angle automatically includes this reduction , while one in mass does not . in the study of fsr , e.g. at lep , three algorithms have been commonly used . the herwig angular - ordered and pythia mass - ordered ones are conventional parton showers as described above , while the ariadne @xcite @xmath31-ordered one is based on a picture of dipole emissions . that is , instead of considering @xmath137 one studies @xmath138 . one aspect of this is that , in addition to the branching parton , ariadne also explicitly includes a `` recoil parton '' needed for overall energy momentum conservation . additionally emissions off @xmath56 and @xmath70 are combined in a well - defined manner . all three approaches have advantages and disadvantages . as already mentioned , pythia does not inherently include coherence , but has to add that approximately by brute force . both pythia and herwig break lorentz invariance slightly . the herwig algorithm can not cover the full phase space with it emissions , but has to fill in some `` dead zones '' using higher - order matrix elements . the ariadne dipole picture does not include @xmath39 branchings in a natural way . and so on . when all is said and done , it turns out that all three algorithms do a quite decent job of describing lep data @xcite , but typically ariadne does best and herwig worst . since ariadne uses pythia for hadronization the difference between those two is entirely due to the shower algorithms , while comparisons with herwig also are complicated by significant differences in hadronization . the structure of initial - state radiation ( isr ) is more complicated than that of fsr , since the nontrivial structure of the incoming hadrons enter the game . a proton is made up out of three quarks , @xmath139 , plus the gluons that bind them together . this picture is not static , however : gluons are continuously emitted and absorbed by the quarks , and each gluon may in its turn temporarily split into two gluons or into a @xmath86 pair . thus a proton is teeming with activity , and much of it in a nonperturbative region where we can not calculate . we are therefore forced to introduce the concept of a parton density @xmath140 as an empirical distribution , describing the probability to find a parton of species @xmath70 in a hadron , with a fraction @xmath141 of the hadron energy momentum when the hadron is probed at a resolution scale @xmath43 . while @xmath140 itself can not be predicted , the change of @xmath142 with resolution scale can , once @xmath43 is large enough that perturbation theory should be applicable : @xmath143 this is actually nothing but our familiar dglap equations . before they were written in an exclusive manner : given a parton @xmath56 , what is the probability that it will branch to @xmath144 during a change @xmath145 ? here the formulation is instead inclusive : given that the probability distributions @xmath146 of all partons @xmath56 are known at a scale @xmath43 , how is the distribution of partons @xmath70 changed by the set of possible branchings @xmath147 ( @xmath148 , here implicit ) . the splitting kernels @xmath149 are the same to leading order , but differ between isr and fsr in higher orders . in higher orders also the concept of @xmath150 as a positive definite probability is lost , additional complications that we will not consider any further here . even though eqs . ( [ eq : dglap ] ) and ( [ eq : dglapisr ] ) are equivalent , the physics context is different . in fsr the outgoing partons have been kicked to large timelike virtualities by the hard process and then cascade downwards towards the mass shell . in isr we rather start out with a simple proton at early times and then allow more and more spacelike virtualities as we get closer to the hard interaction . not that big fluctuations could not happen at early times they do but if they happen too early the uncertainty relation does not allow them to live long enough to be of any intererest to us . that is , the higher the virtuality , the later the fluctuation has to occur . ( 420,120)(-10,-60 ) ( 0,0)(80,0 ) ( 80,0)(160,-10 ) ( 160,-10)(240,10 ) ( 240,10)(320,-20 ) ( 320,-20)(400,20 ) ( 395,15)(405,25 ) ( 395,25)(405,15 ) ( 80,0)(120,10 ) ( 120,10)(160,0)5 ( 120,10)(160,30)5 ( 160,30)(200,20)5 ( 160,30)(200,40)5 ( 160,-10)(200,-30 ) ( 200,-30)(240,-20)5 ( 240,-20)(280,-15)5 ( 240,-20)(280,-30)5 ( 200,-30)(240,-40)5 ( 240,-40)(280,-45)5 ( 240,-40)(280,-60)5 ( 240,10)(280,40 ) ( 280,40)(320,50)5 ( 280,40)(320,20)5 ( 320,20)(360,10)5 ( 320,20)(360,40)5 ( 360,40)(400,30)5 ( 360,40)(400,50)5 ( 320,-20)(350,-40 ) ( 350,-40)(400,-60)5 ( 350,-40)(380,-20)5 ( 380,-20)(400,-35)5 ( 380,-20)(400,-5)5 so , when the hard scattering occurs , in some sense the initial - state cascade is already there , as a virtual fluctuation . had no collision occured the fluctuation would have collapsed back , but now one of the partons of the fluctuation is kicked out in a quite different direction and can no longer recombine with its sister parton from its last branching , nor with its aunt from the last - but - one branching . and so on for each preceding branching in the cascade that lead up to this particular parton . post facto we therefore see that a chain of branchings with increasing @xmath43 values built up an isr shower , fig . [ fig : isrshower ] . the obvious way to simulate this situation would be to pick partons in the two incoming hadrons from parton densities at some low @xmath43 scale , and then use the exclusive formulation of eq . ( [ eq : dglap ] ) to construct a complete picture of partons available at higher @xmath43 scales , event by event . the two sets of incoming partons could then be weighted by the cross section for the process under study . a problem is that this may not be very efficient . we have to evolve for all possible fluctuations , but at best one particular parton will collide and most of the other fluctuations will collapse back . the cost may become prohibitive when the process of interest has a constrained phase space , like a light - mass higgs which has to have the colliding partons matched up in a very narrow mass bin . there are ways to speed up this `` forwards evolution '' approach . however , the most common solution is instead to adopt a `` backwards evolution '' point of view @xcite . here one starts at the hard interaction and then tries to reconstruct what happened `` before '' . to be more precise , the cross - section formula in eq . ( [ eq : sigma ] ) already includes the summation over all possible incoming shower histories by the usage of @xmath43-dependent parton densities . therefore what remains is to pick one exclusive shower history from the inclusive set that went into the @xmath43-evolution . to do this , recast eq . ( [ eq : dglapisr ] ) as @xmath151 then we have defined a _ conditional probability _ : if parton @xmath70 is present at scale @xmath43 , what is the probability that it will turn out to have come from a branching @xmath137 at some infinitesimally _ smaller _ scale ? ( recall that the original eq . ( [ eq : dglapisr ] ) was defined for increasing virtuality . ) like for fsr this expression has to be modified by a sudakov factor to preserve total probability , and this factor is again the exponent of the real - emission expression with a negative sign , integrated over @xmath43 from an upper starting scale @xmath127 down to the @xmath43 of the hypothetical branching . the approach is now clear . first a hard scattering is selected , making use of the @xmath43-evolved parton densities . then , with the hard process as upper maximum scale , a succession of isr branchings are reconstructed at lower and lower @xmath43 scales , going `` backwards in time '' towards the early low - virtuality initiators of the cascades . again some cutoff needs to be introduced when the nonperturbative regime is reached . unfortunately the story does not end there . for fsr we discussed the need to take into account coherence effects and the possibility to use different variables . such issues exist here as well , but also additional ones . for instance , evolution need not be strictly ordered in @xmath43 . ( eq . ( [ eq : ptkin ] ) only gives you @xmath152 , not @xmath153 . ) non - ordered chains in some cases can be important . another issue is that there can be so many partons evolving inside a hadron that they become close - packed , which leads to additional recombinations . see the lectures of g. ingelman for further details @xcite . in summary , isr and fsr share many aspects , but also differ . the dglap evolution with sudakov factors allows a simple probabilistic framework , where an initial parton undergoes successive branchings . for isr the branching ( usually ) is in terms of higher and higher spacelike virtualities as the hard scattering approaches , while for fsr the branchings involve lower and lower timelike virtualities as the hard scattering recedes . in fsr both daughter partons appear on equal footing , in that both can be timelike and branch further . in isr only the daughter parton on its way in to the hard scattering can be spacelike virtual ; its sister will become part of the final state and thus has to be on mass shell , or else be timelike and start an fsr shower of its own . and , last but not least , the whole fsr framework is considerably better understood than the isr one . as we have seen , both matrix elements ( me ) and parton showers ( ps ) have advantages and disadvantages . to recall , me allow a systematic expansion in powers of @xmath154 , and thereby offer a controlled approach towards higher precision . calculations can be done with several ( up to @xmath155 ) partons in the final state , so long as only born - level results are asked for , and it is possible to tailor the phase - space cuts for these partons precisely to the experimental needs . loop calculations are much more difficult , on the other hand , and the mathematically correct cancellation between real- and virtual - emission graphs in the soft / collinear regions is not physically sensible . therefore me can not be used to explore the internal structure of a jet , and are difficult to match to hadronization models , which are supposed to take over in the very soft / collinear region . ps , on the other hand , clearly are approximate and do not come with a guaranteed level of precision for well separated jets . you can not steer the probabilistic evolution of a shower too much , and therefore the efficiency for obtaining events in a specific region of phase space can be quite low . on the other hand , ps are universal , so for any new model you only need to provide the basic hard process and then ps will turn that into reasonably realistic multiparton topologies . the use of sudakov factors ensures a physically sensible behaviour in the soft / collinear regions , and it is also here that the ps formalism is supposed to be as most reliable . it is therefore possible to obtain a good picture of the internal structure of jets , and to provide a good match to hadronization models . in a nutshell : me are good for well separated jets , ps for the structure inside jets . clearly the two complement each other , and a marriage is highly desirable . to do this , without doublecounting or gaps in the phase space coverage , is less trivial , and several alternative approaches have been developed . in the following we will discuss three main options : merging , vetoed parton showers and mc@nlo , roughly ordered in increasing complexity . which of these to use may well depend on the task at hand . the aspiration of merging is to cover the whole phase space with a smooth transition from me to ps . the typical case would be a process where the lowest - order ( lo ) me is known , as well as the next - to - leading - order ( nlo ) real - emission one , say of an additional gluon . the shower should then reproduce @xmath156 starting from a lo topology . if the shower populates phase space according to @xmath157 this implies that a correction factor @xmath158 need to be applied . at first glance this does not apper to make sense : if all we do is get back @xmath159 , then what did we gain ? however , the trick is to recall that the ps formula comes in two parts : the real - emission answer and a sudakov factor that ensures total conservation of probability . what we have called @xmath157 above should only be the real - emission part of the story . it is also this one that we know _ will _ agree with @xmath159 in the soft and collinear regions . actually , with some moderate amount of effort it is often possible to ensure that @xmath158 is of order unity over the whole phase space , and to adjust the showers in the hard region so that the ratio always is below unity , i.e.so that standard monte carlo rejection techniques can be used . what the sudakov factor then does is introduce some ordering variable @xmath43 , so that the whole phase space is covered starting from `` hard '' emissions and moving to `` softer '' ones . at the end of the day this leads to a distribution over phase space like @xmath160 that is , we have used the ps choice of evolution variable to provide an exponentiated version of the me answer . as such it agrees with the me answer in the hard region , where the sudakov factor is close to unity , and with the ps in the soft / collinear regions , where @xmath161 . the method is especially convenient for resonance decays , such as @xmath82 where it was first introduced @xcite . in that case there is an added bonus : the full nlo answer , with virtual corrections included , is known to be @xmath162 . so it is trivial to use the procedure above and rescale everything by @xmath163 to obtain a complete nlo answer . note that the difference between using @xmath164 or @xmath165 in the denominator of eq . ( [ eq : wme ] ) only gives a difference to @xmath166 , i.e. to nnlo . in pythia this approach is used for essentially all resonance decays in the standard model and minimal supersymmetric extensions thereof : @xmath167 , @xmath168 , @xmath169 , @xmath170 , @xmath171 , it is also used in isr to describe @xmath172 @xcite , but here the nlo corrections are more tricky , so the cross section remains as provided by the lo number . merging is also used for several processes in herwig , such as @xmath90 , @xmath167 and @xmath172 @xcite . a special problem here is that the angular - ordered algorithms , both for fsr and for isr , leave some `` dead zones '' of hard emissions that are kinematically forbidden for the shower to populate . it is therefore necessary to start directly from higher - order matrix elements in these regions . a consistent treatment still allows a smooth joining across the boundary . in some sense vetoed parton showers is an extension of the merging approach above . the objective is still to combine the real - emission behaviour of me with the emission - ordering - variable - dependent sudakov factors of ps . while the merging approach only works for combining the lo and nlo expressions , however , the vetoed parton showers offer a generic approach for combining several different orders . therefore it is likely to be a standard tool for many studies in the future . to understand how the algorithm works , consider a lowest - order process such as @xmath173 . for each higher order one additional jet would be added to the final state , so long as only real - emission graphs are considered : in first order e.g.@xmath174 , in second order e.g.@xmath175 , and so on . call these ( differential ) cross sections @xmath85 , @xmath176 , @xmath177 , . it should then come as no surprise that each @xmath178 , @xmath179 , contains soft and collinear divergences . we therefore need to impose some set of me phase - space cuts , e.g. on invariant masses of parton pairs , or on parton energies and angular separation between them . when these cuts are varied , so that e.g. the mass or energy thresholds are lowered towards zero , all of these @xmath178 , @xmath179 , increase without bounds . however , in the me approach without virtual corrections there is no `` detailed balance '' , wherein the addition of cross section to @xmath180 is compensated by a depletion of @xmath178 . that is , if you have an event with @xmath25 jets at some resolution scale , and a lowering of the minimal jet energy reveals the presence of one additional jet , then you should reclassify the event from being @xmath25-jet to being @xmath181-jet . add one , subtract one , with no net change in @xmath182 . so the trick is to use the sudakovs of showers to ensure this detailed balance . of course , in a complete description the cancellation between real and virtual corrections is not completely exact but leaves a finite net contribution , which is not predicted in this approach . a few alternative algorithms exist along these lines . all share the three first steps as follows @xcite : + 1 ) pick a hard process within the me - cuts - allowed phase - space region , in proportions provided by the me integrated over the respective allowed region , @xmath183 . use for this purpose a fix @xmath184 larger than the @xmath154 values that will be used below . + 2 ) reconstruct an imagined shower history that describes how the event could have evolved from the lowest - order process to the actual final state . that provides an ordering of emissions by whatever shower - evolution variable is intended . + 3 ) the `` best - bet '' choice of @xmath154 scale in showers is known to be the squared transverse momentum of the respective branching . therefore a factor @xmath185 , provides the probability that the event should be retained . now the algorithms part way . in the ckkw l approach the subsequent steps are : + 4 ) evaluate sudakov factors for all the `` propagator '' lines in the shower history reconstructed in step 2 , i.e. for intermediate partons that split into further partons , and also for the evolution of the final partons down to the me cuts without any further emissions . this provides an acceptance weight @xmath186 where @xmath187 is the large scale where a parton is produced by a branching and @xmath188 either is the scale at which the parton branches or the me cuts , the case being . + 4a ) in the ckkw approach @xcite the sudakovs are evaluated by analytical formulae , which is fast . + 4b ) in the l approach @xcite trial showers are used to evaluate sudakovs , which is slower but allows a more precise modelling of kinematics and phase space than offered by the analytic expression . + 5 ) now the matrix - element configuration can be evolved further , to provide additional jets below the me cuts used . in order to avoid doublecounting of emissions , any branchings that might occur above the me cuts must be vetoed . the mlm approach @xcite is rather different . here the steps instead are : + 4 ) allow a complete parton shower to develop from the selected parton configuration . + 5 ) cluster these partons back into a set of jets , e.g. using a cone - jet algorithm , with the same jet - separation criteria as used when the original parton configuration was picked . + 6 ) try to match each jet to its nearest original parton . + 7 ) accept the event only if the number of clustered jets agrees with the number of original partons , and if each original parton is sensibly matched to its jet . this would not be the case e.g. if one parton gave rise to two jets , or two partons to one jet , or an original @xmath70 quark migrated outside of the clustered jet . + the point of the mlm approach is that the probability of _ not _ generating any additional fatal jet activity during the shower evolution is provided by the sudakovs used in the step 4. mc@nlo @xcite in some respects is the most ambitious approach : it aims to get not only real but also virtual contributions correctly included , so that cross sections are accurate to nlo , and that nlo results are obtained for all observables when formally expanded in powers of @xmath154 . thus hard emissions should again be generated according to me , while soft and collinear ones should fall within the ps regime . in simplified terms , the scheme works as follows : + 1 ) calculate the nlo me corrections to an @xmath49-body process , including @xmath189-body real corrections and @xmath49-body virtual ones . + 2 ) calculate analytically how a first branching in a shower starting from a @xmath49-body topology would populate @xmath189-body phase space , excluding the sudakov factor . + 3 ) subtract the shower expression from the @xmath189 me one to obtain the `` true '' @xmath189 events , and consider the rest as belonging to the @xmath49-body class . the ps and me expressions agree in the soft and collinear limits , so the singularities there cancel , leaving finite cross sections both for the @xmath49- and @xmath189-body event classes . + 4 ) now add showers to both kinds of events . ( 350,200)(-50,-50 ) ( 0,0)(180,0)(190,0)[]@xmath190 ( 0,0)(0,135)(0,145)[]@xmath191 ( 190,140)[l ] ( 70,98)(40,45 ) ( 60,110)[l]@xmath192 jet me ( 165,35)(90,10 ) ( 175,35)[l]generate as z + 1 jet + shower ( 165,110)(43,26 ) ( 175,110)[l]@xmath192 jet according to shower ( 175,95)[l](first emission , without sudakov ) ( 165,60)(35,10 ) ( 175,60)[l]generate as @xmath193 + shower ( 0,0)(0,70 ) ( -10,40)[r]lo real ( 0,0)(0,-50 ) ( -10,-30)[r]nlo virtual a toy example , for the case of @xmath8 production , is shown in fig . [ fig : mcnlo ] . several more complicated processes have been considered , such as @xmath194 , @xmath195 and @xmath196 production . a technical problem is that , although me and ps converge in the collinear region , it is not guaranteed that me is everywhere above ps . this is solved by having a small fraction of events with negative weights . in summary , mc@nlo is superiour in that it does provide the total cross section for a process to nlo accuracy , and so is essential for a set of precision tests . the real - emission @xmath189-body part is the same as used for merging , however , so for normalized event shapes the merging approach is as valid . ( any differences are of higher order . ) finally , if multijet topologies need to be studied , where several orders may contribute , vetoed showers are more appropriate . each tool to its task . the cross section for @xmath33 qcd parton processes is dominated by @xmath113-channel gluon exchange , as we already mentioned , and thus diverges like @xmath197 for @xmath20 . introduce a lower cut @xmath198 and integrate the interaction cross section above this , properly convoluted with parton densities . at lhc energies this @xmath199 reaches around 100 mb for @xmath200 gev , and 1000 mb at around 2 gev . since each interaction gives two jets to lowest order , the jet cross section is twice as big . this should be compared with an expected _ total _ cross section of the order of 100 mb . ( qcd is a confining theory and thus provides a finite total cross section , unlike qed , where infinitely small scattering angles are allowed at infinitely large distances . ) in addition , at least a third of the total cross section is related to elastic scattering @xmath201 and low - mass diffractive states @xmath202 that could not contain jets . so can it really make sense that @xmath203 ? yes , it can ! the point is that each incoming hadron is a bunch of partons . you can have several ( more or less ) independent parton parton interactions when these two bunches pass through each other . and the point is that an event with @xmath49 interactions above @xmath198 counts once for the total cross section but once _ for each interaction _ when the interaction rate is calculated . that is , @xmath204 where @xmath205 is the cross section for events with @xmath49 interactions . thus @xmath206 is equivalent to @xmath207 , that each event on the average contains more than one interaction . furthermore , if interactions really do occur independently when two hadron pass by each other , then one would expect a poissonian distribution , @xmath208 , so that several interactions could occur occasionally also when @xmath209 , e.g. for a larger @xmath198 cut . momentum conservation ensures that interactions never are truly independent , and also other effects enter ( see below ) , but the poissonian ansatz is still a useful starting point . multiple interactions ( mi ) can only be half the solution , however . the divergence for @xmath210 would seem to imply an infinite average number of interactions . but what one should realize is that , in order to calculate the @xmath211 matrix elements within standard perturbation theory , it has to be assumed that free quark and gluon states exist at negative and positive infinity . that is , the confinement of colour into hadrons of finite size has not been taken into account . so obviously perturbation theory has to have broken down by @xmath212 the nature of the breakdown is also easy to understand : a small-@xmath31 gluon , to be exchanged between the two incoming hadrons , has a large transverse wavelength and thus almost the same phase across the extent of each hadron . the contributions from all the colour charges in a hadron thus add coherently , and that means that they add to zero since the hadron is a colour singlet . what is then the typical scale of such colour screening effects , i.e. at what @xmath31 has the interaction rate dropped to @xmath213half of what it would have been if the quarks and gluons of a proton had all been free to interact fully independently ? that ought to be related to the typical separation distance between a given colour and its opposite anticolour . when a proton contains many partons this characteristic screening distance can well be much smaller than the proton radius . empirically we need to introduce a @xmath198 scale of the order of 2 gev to describe tevatron data , i.e. of the order of 0.1 fm separation . it is not meaningful to take this number too seriously without a detailed model of the space time structure of a hadron , however . the 2 gev number is very indirect and does not really tell exactly how the dampening occurs . one can use a simple recipe , with a step - function cut at this scale , or a physically more reasonable dampening by a factor @xmath214 , plus a corresponding shift of the @xmath154 argument , @xmath215 with @xmath216 a dampening scale that also lands at around 2 gev . this translates into a typical number of 23 interactions per event at the tevatron and 45 at lhc . for events with jets or other hard processes the average number is likely to be higher . the first studies of complete events based on perturbatively generated mi @xcite started out from a mimimally simple model : + 1 ) use a sharp @xmath198 cut as the key tunable parameter . + 2 ) only address inelastic nondiffractive events , i.e. with @xmath217 , so that the average number of interactions per such events is @xmath218 . + 3 ) to first approximation this gives a poissonian distribution in the number of interactions per event , with a fraction @xmath219 of purely low-@xmath31 interactions . + 4 ) the interactions are generated in an ordered sequence of decreasing @xmath31 values : @xmath220 . this is possible with the standard sudakov kind of trick : @xmath221 ~,\ ] ] with a starting @xmath222 . + 5 ) the ordering of emissions allows parton densities to be rescaled in @xmath141 after each interaction , so that energy momentum is not violated . thereby the actual distribution of the number of interactions becomes narrower than poissonian , since the rate of a further interaction is reduced if previous ones already took away energy . + 6 ) for technical reasons the model was simplified after the first interaction , so that there only @xmath98 or @xmath99 outgoing pairs were allowed , and no showers were added to these additional @xmath33 interactions . + already this simple pythia - based model is able to `` explain '' a large set of experimental data . some additional features were then included in a more sophisticated variant . + 1 ) use the smooth turnoff of eq . ( [ eq : smoothmi ] ) all the way down to @xmath223 . then an event has to contain at least one interaction to be an event at all , i.e. @xmath216 has to be selected sufficiently small that @xmath224 . + 2 ) hadrons are extended , and therefore partons are distributed in ( transverse ) coordinates . to allow a flexible parametrization and yet have an easy - to - work - with expression , a double gaussian @xmath225 is used , where @xmath226 and @xmath227 are tunable parameters . + 3 ) the matter overlap during a collision , calculated by @xmath228 directly determines the average activity in events at different impact parameter @xmath70 : @xmath229 . that is , central collisions tend to have more activity , peripheral less , but of course properly normalized so that the @xmath70-integrated interaction cross section agrees with standard perturbation theory ( modulo the already - discussed dampening at small @xmath31 ) . thereby the @xmath230 distribution becomes broader than a poissonian . + as before , several simplifications are necessary . + this is the scenario that has been used in many of the experimental studies over the years . more recently a number of improvements have been included @xcite . + 1 ) the introduction of junction fragmentation , wherein the confinement field between the three quarks in a baryon is described as a y - shaped topology , now allows the handling of topologies where several valence quarks are kicked out , thus allowing arbitrary flavours and showering in all interactions in an event . + 2 ) parton densities are not only rescaled for energy momentum conservation , but also to take into account the number of remaining valence quarks , or that sea quarks have to occur in @xmath99 pairs . + 3 ) the introduction of @xmath31-ordered showers allows the selection of new isr branchings and new interactions to be interleaved in one common sequence of falling @xmath31 values . thereby the competition between these two components , which both remove energy from the incoming beams , is modelled more realistically . fsr is not yet interleaved , but also does not compete for beam energy . + this scenario is not yet as well studied experimentally . the traditional herwig soft underlying event ( sue ) approach to this issue has its origin in the ua5 monte carlo . in it a number of clusters are distributed almost independently in rapidity and transverse momentum , but shifted so that energy momentum is conserved , and the clusters then decay isotropically . the multiplicity distribution of clusters and their @xmath231 and @xmath31 spectra are tuned to give the observed inclusive hadron spectra . no jets are produced in this approach . the jimmy @xcite program is an add - on to herwig . it replaces the sue model with a mi - based one more similar to the pythia ones above , e.g. with an impact - parameter - based picture for the multiple - interactions rate . technical differences exist , e.g. jimmy interactions are not picked to be @xmath31-ordered and thus energy momentum issues are handled differently . the dpmjet / dtujet / phojet family of programs @xcite come from the `` historical '' tradition of soft physics , wherein multiple @xmath232 `` pomeron '' exchanges fill a role somewhat similar to the hard mi above . jet physics was originally not included , but later both hard and soft interactions have been allowed . one strong point is that this framework also allows diffractive events to be included as part of the same basic machinery . how do we know that mi exist ? the key problem is that it is not possible to identify jets coming from @xmath233 gev partons . therefore we either have to use indirect signals for the presence of interactions at this scale or we have to content ourselves with studying the small fraction of events where two interactions occur at visibly large @xmath31 values . an example of the former is the total charged multiplicity distribution in high - energy @xmath2 collision . this distribution is very broad , and is even getting broader with increasing energy , meaured in terms of the width over the average , @xmath234 . by contrast , recall that for a poissonian this quantity scales like @xmath235 and thus is getting narrower . simple models , with at most one interaction and with a fragmentation framework in agreement with lep data , can not explain this : they are way to narrow , and have the wrong energy behaviour . if mi are included the additional variability in the number of interactions per event offers the missing piece @xcite . the variable - impact - parameter improves the description further . another related example is forward backward correlations . consider the charged multiplicity @xmath236 and @xmath237 in a forward and a backward rapidity bin , each of width one unit , separated by a central rapidity gap of size @xmath238 . it is not unnatural that @xmath236 and @xmath237 are somewhat correlated in two - jet events , and for small @xmath238 one may also be sensitive to the tails of jets . but the correlation coefficient , although falling with @xmath238 , still is appreciable even out to @xmath239 , and here again traditional one - interaction models come nowhere near . in a mi scenario each interaction provides additional particle production over a large rapidity range , and this additional number - of - mi variability leads to good agreement with data . direct evidence comes from the study of four - jet events . these can be caused by two separate interactions , but also by a single one where higher orders ( call it me or ps ) has allowed two additional branchings in a basic two - jet topology . fortunately the kinematics should be different . assume the four jets are ordered in @xmath31 , @xmath240 . if coming from two separate interactions the jets should pair up into two separately balancing sets , @xmath241 and @xmath242 . if an azimuthal angle @xmath243 is introduced between the two jet axes this also should be flat if the interactions are uncorrelated . by contast the higher - order graph offers no reason why the jets should occur in balanced pairs , and the @xmath243 distribution ought to be peaked at small values , corresponding to the familiar collinear singularity . the first to observe an mi signal this way was the afs collaboration @xcite at isr ( @xmath244 at 62 gev ) , but with large uncertainties . a more convincing study was made by cdf @xcite , who obtained a clear signal in a sample with three jets plus a photon . in fact the deduced rate was almost a factor of three higher than naive expectations , but quite in agreement with the impact - parameter - dependent picture , wherein correlations of this kind are enhanced . a topic that has quite extensively studied in cdf is that of the jet pedestal @xcite , i.e. the increased activity seen in events with a jet , even away from the jet itself , and away from the recoiling jet that should be there . some effects come from the showering activity , i.e. the presence of additional softer jets , but much of it rather finds its explanation in mi , as a kind of `` trigger bias '' effect , as follows . ( 1 ) central collisions tend to produce many interactions , peripheral ones few . ( 2 ) if an event has @xmath49 interactions there are @xmath49 chances that one of them is hard . combine the two and one concludes that events with hard jets are biased towards central collisions and many additional interactions . the rise of the pedestal with triggger - jet energy saturates once @xmath245 , however , because by then events are already maximally biased towards small impact parameter . and this is indeed what is observed in the data : a rapid rise of the pedestal up to @xmath246 gev , and then a slower increase that is mainly explained by showering contributions . in more detailed studies of this kind of pedestal effects there are also some indications of a jet substructure in the pedestal , i.e.that indeed the pedestal is associated with the production of additional ( soft ) jet pairs . in spite of many qualitative successes , and even some quantitative ones , one should not be lead to believe that all is understood . possibly the most troublesome issue is how colours are hooked up between all the outgoing partons that come from several different interactions . a first , already difficult , question is how colours are correlated between all the partons that are taken out from an incoming hadron . these colours are then mixed up by the respective scattering , in principle ( approximately ) calculable . but , finally , all the outgoing partons will radiate further and overlap with each other on the way out , and how much that may mess up colours is an open question . a sensitive quantity is @xmath247 , i.e. how the average transverse momentum of charged particles varies as a function of their multiplicity . if interactions are uncorrelated in colour this curve tends to be flat : each further interaction adds about as much @xmath31 as @xmath248 . if colours somehow would rearrange themselves , so that the confinement colour fields would not have to run criss - cross in the event , then the multiplicity would not rise as fast for each further interaction , and so a positive slope would result . the embarrassing part is that the cdf tunes tend to come up with values that are about 90% on the way to being maximally rearranged @xcite , which is way more than one would have guessed . obviously further modelling and tests are necessary here @xcite . another issue is whether the @xmath216 regularization scale should be energy - dependent . in olden days there was no need for this , but it became necessary when hera data showed that parton densities rise faster at small @xmath141 values than had commonly been assumed . this means that the partons become more close - packed and the colour screening increases faster with increasing collision energy . therefore an energy - dependent @xmath216 is not unreasonable , but also can not be predicted . currently the default pythia ansatz is @xmath249 , i.e a predicted @xmath250 gev at 14 tev . this gives a minimum - bias multiplicity of about 7 per unit of rapidity in the central region , and a pedestal under jet events of around 30 charged particles per unit @xcite . however , these numbers are model- and parameter - dependent . phojet predicts about half as big a pedestal , and typical jimmy tunes about 50% more , a priori leaving a big range of uncertainty to be resolved once lhc runs begin . the physics mechanisms discussed so far are mainly being played out on the partonic level , while experimentalists observe hadrons . in between exists the very important hadronization phase , where all the outgoing partons end up confined inside hadrons of a typical 1 gev mass scale . this phase can not ( so far ? ) be described from first principles , but has to involve some modelling . the main approaches in use today are string fragmentation and cluster fragmentation . these are described in the lectures of g. ingelman @xcite , so this section will be very brief , only with a few comments . hadronization models start from some ideologically motivated principles , but then have to add `` cookbook recipes '' with free parameters to arrive at a complete picture of all the nitty gritty details . this should come as no surprise , given that there are hundereds of known hadron species to take into account , each with its mass , width , wavefunction , couplings , decay patterns and other properties that could influence the structure of the observable hadronic state , and with many of those properties being poorly or not at all known , either from theory ( lattice qcd ) or from experiment . in that sense , it is sometimes more surprising that models can work as well as they do than that they fail to describe everything . the simpler initial state at an @xmath251 collider , such as lep , implies that this is the logical place to tune the hadronization framework to data @xcite , and thereafter those tunes can be applied to other studies . one such is the internal structure of jets in hadron collider , where the pattern in many respects is surprisingly well described . on the other hand , at the hera @xmath1 collider it has been observed that the relative amount of strange - particle production is only @xmath252 of that at lep , and of ( anti)baryons only @xmath253 . this has no simple explanation within the string fragmentation model , so it acts as a useful reminder that we still do not know as much as we should . also other examples could be provided in these lectures we have followed the flow of generators roughly `` inwards out '' , i.e. from short - distance processes to long - distance ones . at the core lies the hard process , described by matrix elements . it is surrounded by initial- and final - state showers , that should be properly matched to the hard process . multiple parton parton interactions can occur , and the colour flow is tied up with the structure of beam remnants . at longer timescales the partons turn into hadrons , many of which are unstable and decay further . this basic pattern is likely to remain in the future , but many aspects will change . one such aspect , that stands a bit apart , is that of languages . the traditional event generators , like pythia and herwig , have been developed in fortran up until the end of the lep era this was the main language in high - energy physics . but now the experimental community has largely switched to c++ and decisions have been taken e.g.at cern to discontinue fortran altogether . the older generators are still being used , hidden under c++ wrappers , but this can only be a temporary solution , for several reasons . one is that younger experimentalists often need to look into the code of generators and tailor some parts to specific needs of theirs , and if then the code is in an unknown language this will not work . another is that theory students who apply for non - academic positions are much better off if their resums say `` expert on object - oriented programming '' rather than `` fortran fan '' . a conversion program thus has begun on many fronts . sherpa , as the youngest of the general - purpose generators , was conceived from the onset as a c++ package and thus is some steps ahead of the other programs in this respect . @xcite is a complete reimplementation of the herwig , as is pythia 8 of the current pythia 6 . both conversions have taken longer than originally hoped , but progress is being made and first versions exist . thepeg @xcite is a generic toolkit for event generators , used by herwig++ and the upcoming new ariadne . * faster , better and more user - friendly general - purpose matrix - element generators with an improved sampling of phase space . * new ready - made libraries of physics processes , in particular with full nlo corrections included . * more precise parton showers . * better matching between matrix elements and parton showers . * improved models for minimum - bias physics and underlying events . * some upgrades of hadronization models and decay descriptions . in general one would say that generators are getting better all the time , but at the same time the experimental demands are also getting higher , so it is a tight race . however , given that typical hadronic final states at lhc will contain hundreds of particles and quite complex patterns buried in that , it is difficult to see that there are any alternatives . as the same time as you need to use generators , you should remain critical and be on the lookout for bugs and bad modelling . bjorken already many years ago worried about the passive attitude many experimentalists have towards the output of generators ; they `` carry the authority of data itself . they look like data and feel like data , and if one is not careful they are accepted as if they were data . '' do nt you fall into that trap ! s. mrenna , ` http://www.phys.psu.edu/`@xmath213`cteq/schools/summer04/mrenna/mrenna.pdf ` + p. richardson , ` http://www.ippp.dur.ac.uk/`@xmath213`richardn/talks/ ` + m. seymour , ` http://seymour.home.cern.ch/seymour/slides/cernlectures.html ` + t. sjstrand , ` http://www.thep.lu.se/`@xmath213`torbjorn/ ` ( click on `` talks '' ) i.g . knowles et al . , in ` physics at lep2 ' , eds . g. altarelli , t. sjstrand and f. zwirner , cern 9601 ( geneva , 1996 ) , vol . 2 , p. 103 ; + p. bambade et al . , in ` reports of the working groups on precision calculations for lep2 physics ' , eds . s. jadach , g. passarino and r. pittau , cern 2000 - 009 , p. 137

investigation of the magnetization dynamics at the nanoscale is a key task of the modern nanomagnetism.@xcite one of the typical topologically nontrivial magnetic configurations of a nanoscaled magnet is a magnetic vortex , which can form a ground state configuration of submicron sized magnetic disk shaped particles ( nanodots ) . such a vortex is characterized by a curling divergent free in plane configuration with magnetization tangential to the edge surface of the nanoparticle . @xcite the out of plane magnetization appears only in a very thin region around the vortex core with about the size of an exchange length ( typically about 10 nm for magnetically soft materials @xcite ) . the vortex state is degenerated with respect to the upward or downward magnetization of the vortex core ( the vortex polarity @xmath0 ) , hence the vortex polarity can be considered as a bit of information in nonvolatile magnetic vortex random access memories ( vram ) . @xcite that is why one needs to control the vortex polarity switching process in a very fast way . the vortex polarity switching phenomena were predicted originally for the heisenberg 2d magnets . @xcite the interest to this problem was renewed after an experimental detection of the vortex core reversal in nanodots by an excitation with short bursts of an alternating field , @xcite which opened a possibility to use the vortex state dots as the vram . moreover , this motivated numerous fundamental studies of the vortex core switching mechanism itself.@xcite there are two basic scenarios of the vortex polarity switching . in the first , axially symmetric ( or punch through ) scenario , the vortex polarity is switched due to the direct pumping of axially symmetric magnon modes . such a switching occurs , e.g. , under the influence of a dc transversal field . @xcite in the second , axially asymmetric scenario , the switching occurs due to a nonlinear resonance in the system of certain magnon modes with nonlinear coupling . @xcite such a scenario is accompanied by the temporary creation and annihilation of vortex antivortex pairs . @xcite the axially asymmetric switching occurs , e.g. , under the action of different in plane ac magnetic fields or by a spin polarized current , see ref . and references therein . recently the interest to the axially symmetric switching was renewed : using the micromagnetic simulations @xcite demonstrated that the vortex polarity reversal can be realized under the action of an alternating perpendicular magnetic field . in this case the resonant pumping of the radial magnon modes initiates the switching at much lower field intensities than by the dc fields . we have very recently predicted the possibility of the chaotic dynamics of the vortex polarity under the action of the homogeneous transversal ac magnetic field in the 10 ghz range:@xcite @xmath1 in order to describe the switching behavior we proposed in ref . the analytical two parameter cutoff model , which gave us a possibility to describe both deterministic and chaotic behavior of the vortex polarity . the goal of the current work is to study in detail the vortex dynamics under the action of a perpendicular ac magnetic field : we found a rich vortex polarity dynamical behavior , including the _ regular _ and _ chaotic _ regimes of magnetization reversal . in order to analyze the complicated temporal evolution of the vortex polarity we used here the discrete _ reduced core model _ , @xcite which allows us to describe different regimes of vortex polarity dynamics , including the resonant behavior , the weakly nonlinear regimes , the reversal dynamics , and the chaotic regime . the reduced core model is another way to treat the discretness effects . as opposed to the cutoff model , the core model is simpler , hence it allows to go further in analytics . the paper is organized as follows : the full scale micromagnetic simulations are detailed in sec . [ sec : sim ] . our diagram of switching events demonstrates regimes of the regular reversal ( single , multiple and periodic ones ) , intermittent and chaotic regimes . in sec . [ sec : core ] we describe the comprehensive vortex core dynamics using a simple collective coordinate model , which provides all features of the full scale simulations . we propose a way of a unidirectional switching controlled switching in sec . [ sec : control ] . in sec . [ sec : conclusion ] we state our main conclusions . in appendix [ app : core ] we derive the reduced core mode . we use the method of multiple scales to perform a weakly nonlinear analysis of the analytical model in appendix [ app : wna ] . at ( 0,2.5 ) ; ( pointer ) at ( 8.7,0.92 ) ; ( leftcorner ) at ( 0,5.5 ) ; ( leftcorner ) ( @xmath2 ) ( pointer ) ( @xmath3 ) ( @xmath4 ) ( @xmath5 ) ( @xmath6 ) cycle ; at ( @xmath7 ) ; ; ( pointer1 ) at ( 6.35,0.75 ) ; ( pointer2 ) at ( 7.5,0.75 ) ; ( leftcorner ) at ( 3.025,4.475 ) ; ( leftcorner ) ( @xmath8 ) ( pointer1 ) ( @xmath9 ) ( @xmath10 ) ( pointer2 ) (@xmath11 ) ( @xmath4 ) ( @xmath12 ) (@xmath13 ) cycle ; at ( @xmath14 ) ; ; ( pointer ) at ( 8.7,1.2 ) ; ( leftcorner ) at ( 6.05,5.5 ) ; ( leftcorner ) ( @xmath2 ) ( pointer ) ( @xmath3 ) ( @xmath4 ) ( @xmath5 ) ( @xmath6 ) cycle ; at ( @xmath7 ) ; ; ( pointer ) at ( 9.96,1.2 ) ; ( leftcorner ) at ( 9.075,5.5 ) ; ( leftcorner ) ( @xmath2 ) ( pointer ) ( @xmath3 ) ( @xmath4 ) ( @xmath5 ) ( @xmath6 ) cycle ; at ( @xmath7 ) ; ; ( 1.7,3 ) node[left ] @xmath15 ( 2.,3.4 ) ; ( 2.4,2.1 ) node[left ] @xmath16 ( 2.8,2.3 ) ; nowadays the micromagnetic simulations are the inherent tools for the nanomagnetic research.@xcite namely using the numerical simulations it was shown in refs . that the resonant perpendicular field forces the vortex core to reverse . here we perform full scale micromagnetic simulations to study the complicated vortex core dynamics in details . we consider a disk shaped nanoparticle ( 198 nm in diameter and 21 nm in thickness ) under the action of the vertical oscillating field using an oommf@xcite micromagnetic simulator with integration method rk5(4)7fc . the material parameters correspond to permalloy ( @xmath17 ) with exchange constant @xmath18pj , saturation magnetization @xmath19ka / m , zero anisotropy coefficient and the gilbert damping coefficient @xmath20 . the mesh cell was chosen to be @xmath21 nm ( the three dimensional mesh will be discussed at the end of this section ) . for all oommf simulations we use as initial state the relaxed vortex with the polarity directed upward and counter clockwise in plane magnetization direction . we also simulated the dynamics of the vortices for the samples with other geometrical parameters : as we expected the qualitative behavior of the system remains the same . first of all we examined the eigenfrequencies of the lower axially symmetric spin waves by applying a rectangular pulse with the strength of 30mt during 100ps perpendicular to the nanodisk in the vortex state in the same way as in ref . . under the action of such a pulse , the magnetization starts to oscillate : a set of symmetrical magnon modes @xmath22 is excited . using the fast fourier transformation ( fft ) of the @xmath23component of the total magnetization , typically during @xmath24 $ ] , we identified the eigenfrequency of the lowest symmetrical mode @xmath25 ghz . this value defines the lowest threshold for the polarity switching @xcite . the next nearest peaks in the fft spectrum correspond to 16.75 ghz and 27.93 ghz . it is already known@xcite that the vortex polarity switching under the action of the ac field occurs in a wide range of field parameters ( the field intensities @xmath26 and field frequencies @xmath27 ) . the minimal field intensity is reached at about the resonance frequency @xmath28 . in the current study we are interested in the long time vortex dynamics , which is accompanied by the axially symmetric polarity reversal mechanism . in all numerical experiments we calculated the polarity and the position of the vortex as functions of time : the vortex position @xmath29 is determined as cross section of isosurfaces @xmath30 and @xmath31,@xcite and the vortex polarity @xmath32 is determined as the average @xmath23magnetization of four neighbor cells to @xmath29 . to study in details the temporal evolution of the vortex polarity , we simulated the long time system dynamics with the time step of 1ps for a wide range of the field parameters ( the field intensity @xmath26 varies from @xmath33 to @xmath34mt , and the field frequency @xmath27 changes from @xmath35 to @xmath36 ghz).ghz and @xmath37mt , 10mt and 0.5ghz in the range @xmath38ghz and @xmath39mt along the border of the diagram , 50mt and 1ghz for other ranges . for the frequencies @xmath40ghz the full time of simulations was 10ns [ extended till 30ns for some particular points @xmath41 and for frequencies @xmath42ghz the full time of simulations was 5ns . ] the results can be summarized in the diagram of dynamical regimes , see fig . [ fig : sim : homega ] . depending on the field parameters ( @xmath26 , @xmath27 ) , one can separate several different dynamical regimes : ( i ) the absence of the vortex polarity switching , ( ii ) the chaotic polarity oscillations , ( iii ) the regular switchings with frequencies depending on the field frequency , ( iv ) the intermittent switchings , and ( v ) the complex vortex magnon dynamics , where the vortex escapes from the origin . \(i ) we start from a weak field : the field intensity is not strong enough to switch the vortex polarity ; this regime corresponds to the linear or weakly nonlinear oscillations of the vortex polarity ( marked as open boxes in fig . [ fig : sim : homega ] ) . the weak pumping of the system ( field intensities @xmath43mt , see fig . [ fig : sim : nonlinamp ] ) causes the resonance at the frequency @xmath28 . the increase of the field intensity leads to the nonlinear dynamical regime . however , if the field intensity is not strong enough , one has a weakly nonlinear regime , which corresponds to the nonlinear resonance . apart from the nonlinear resonance behavior , the strong pumping causes the vortex polarity instability , it also causes the shift of the main peak in the fft spectrum ( see fig . [ fig : sim : nonlinamp]b ) and the beats in the polarity oscillations ( see fig . [ fig : sim : nonlinamp]a ) . mt , @xmath44ghz during 15ns . arrow indicates the pumping frequency . ] let us consider the case when the magnetization reversal occurs . the switching diagram ( see fig . [ fig : sim : homega ] ) has two well defined minima . the first one corresponds to the resonant excitation of the radially symmetrical mode @xmath28 . the second minimum near 18ghz , probably , corresponds to the dynamics near the higher resonances@xcite . \(ii ) the open circles @xmath45 - 0.81em@xmath46 on the switching diagram ( see fig . [ fig : sim : homega ] ) correspond to the chaotic polarity reversal process . the typical temporal evolution is shown in fig . [ fig : sim : homega]c . to draw a conclusion about chaotic behavior of the vortex polarity , or more accurately , to make quantitative measures of chaotic dynamics , we use two standard ways : the autocorrelation function for the temporal evolution of the vortex polarity and the fourier distribution of its frequency spectra.@xcite first , we define the autocorrelation function of the vortex polarity signal @xmath47 for the discretized time @xmath48 with the step @xmath49ps , with the boundary values assumed as zero . it is well known@xcite from the correlational analysis , when a signal is chaotic , information about its past origins is lost , i.e. the signal is only correlated with its recent part : the autocorrelation function decays very rapidly , @xmath50 as @xmath51.@xcite for a periodical signal , the autocorrelation function is a periodic too . a typical example is presented in fig . [ fig : sim : autocor ] : the autocorrelation function @xmath52 is aperiodic and sharply decays for the applied magnetic field 70mt with the frequency 14ghz , which corresponds to the chaotic dynamics . the autocorrelation for the regular dynamics demonstrates the oscillations under the action of @xmath53mt with @xmath54ghz . the second way is to calculate the fourier spectrum of a chaotic signal . a typical fft signal is presented in the fig . [ fig : sim : fourier ] . it is distinctive for the chaotic regime that the continuous spectrum dominates the discrete spikes ( one can identify in the fig . [ fig : sim : fourier ] only one discrete spike at the pumping frequency ) . the fitting of such a signal demonstrates typical pink noise behavior with a power law decay of the spectrum , @xmath55 with @xmath56 . spectrum , where @xmath57 , see ref . . ] \(iii ) the regular oscillations of the vortex polarity appear in the high frequency regime , see the black diamonds on the switching diagram ( see figs . [ fig : sim : homega ] , [ fig : sim : homega]d ) . we have detected the periodical motion of the vortex polarity using the pumping frequency 18ghz with the field intensities higher than 100mt . the main peak in the fft spectrum corresponds to 6ghz , _ i. e. _ it occurs at @xmath15 of the pumping . other spikes with decaying intensities appear with steps of 6ghz . we compare the autocorrelation functions for the regular and chaotic oscillations , see fig . [ fig : sim : autocor ] . in contrast to the chaotic regime , @xmath52 for periodic oscillations exhibits a high periodicity with a slowly decaying amplitude due to the finite observation time . mt and @xmath54ghz ( blue dashed line ) and @xmath58mt and @xmath59ghz ( red solid line ) . the first process demonstrates a periodical process with periodical @xmath52 and the second one demonstrates a chaotic behaviour with rapidly decaying @xmath52 . ] in order to compare the temporal dynamics of the polarity in chaotic and regular regimes , we calculate the pseudo phase trajectories . the method of the pseudo phase space is usually used when only one variable [ the discretized vortex polarity @xmath60 in our case ] is measured:@xcite the pseudo phase space plot can be made using @xmath60 and its future value @xmath61 , where the absolute value of the time step @xmath62 affects only the shape of the trajectory . in the case of chaotic dynamics , one has open trajectories in pseudo phase space @xmath63 , see the fig . [ fig : sim : pseudofield](a ) . in the regular case , pseudo phase trajectories are closed , see the fig . [ fig : sim : pseudofield](b ) . both trajectories are shown for the first 10ns of the dynamics : in the first case the trajectory every time makes a new loop in a different place and in the second case all loops coincide . below in sec . [ sec : core ] we construct the phase trajectories for the theoretical model of our system ( see figs . [ fig : core : homega]a , [ fig : core : homega]d ) . \(iv ) the green triangles on the switching diagram ( see fig . [ fig : sim : homega ] ) correspond to an intermittent process . the typical example of the temporal dynamics in such a regime is plotted in the fig . [ fig : sim : homega]a : the vortex state can retain its polarity for a relatively long time of a few nanoseconds ; after that multiple reversal processess occur during @xmath64ns . note that in the vicinity of other regimes in the switching diagram we observed that the vortex polarity , after a few switches , can be ` frozen ' for the rest of the observation time . for example , two switching events occur during the first 1.2ns ( @xmath65mt and @xmath59ghz , see fig . [ fig : sim : homega]a ) , after that the dynamical polarity has only weak oscillations . a similar picture occurs for @xmath66mt and @xmath67ghz , where after three reversals during the first nanosecond the resulting polarity remains negative . since the reversals occur only at the beginning , one can conclude that this occurs because the field is not switched on smoothly . \(v ) the last regime corresponds to the field parameters ( @xmath26 , @xmath27 ) , where the vortex escapes from the system origin on a long time scale ( see the red circles in fig . [ fig : sim : homega ] ) . typically , the vortex starts to move during the first 10ns . the detailed analysis shows that the switching scenario differs essentially from the above mentioned one : the magnetization reversal is mediated by the transient creation and annihilation of a vortex antivortex pair @xcite ( for details of the axially asymmetric switching mechanism see ref . and references therein ) . two examples of the possible trajectories are shown in fig . [ fig : sim : homega]b : the trajectory b@xmath68 corresponds to the chaotic motion , which is accompanied by numerous reversal events . in the regular regime the vortex trajectory has a smooth shape ( b@xmath69 ) . when the vortex stays in the center of the sample , polarity switching is accompanied by generation of the radially symmetrical modes . after some time of observation , a new 4-fold symmetry occurs around the vortex , which was mentioned in the ref . and linked with the square mesh symmetry used in the oommf . when the vortex moves from the center , the switching scenario is changed : a pair antivortex new vortex is created and the antivortex annihilates with the old vortex . the further magnon dynamics becomes unpredictable . we performed very long time simulations ( up to 30ns ) for all parameters from the switching diagram , where the vortex does not leave the disk center ( see fig . [ fig : sim : homega ] ) : the vortex motion was found for all parameters with @xmath70ghz . for higher frequencies ( e.g. , for @xmath54ghz and @xmath71mt ) the small oscillations of the vortex position were observed only for @xmath72ns . in the prolonged simulations ( iv ) the vortex polarity does not change its value during the time of observation in agreement with the conclusion made above that the field is sharply switched on . mt , @xmath59ghz ) : solid line corresponds to the numerical data , the dashed line is the fitting to the pink noise . ] [ page : features ] the switching diagram for the low frequency range has several new features . one can identify from the plot two local minima ( 4.5ghz and 6ghz ) , which correspond to resonances for fractional frequencies ( @xmath73 and @xmath74 ) . the strong field causes the vortex polarity reversal , which corresponds to the quasi static regime and the lower fields cause the escape of the vortex from the system origin . we checked the idea about the quasi static regime by computing the threshold value for the static field , which is in our case 611mt , cf . refs . . ) ; ( b ) the same diagram for the regular dynamics.,title="fig : " ] + ( a ) @xmath75mt , @xmath76ghz ( b ) @xmath75mt , @xmath54ghz it should be noted that in all simulations discussed above we used the effective 2d mesh @xmath21 nm . in order to check our assumption about the uniform magnetization distribution along the thickness z coordinate , we also performed 3d oommf simulations with the mesh size @xmath77 nm . one can see that eigenfrequencies and boundaries of dynamical regimes are slightly influenced by the nonhomogeneous distribution along the z coordinate , see fig . [ fig : sim : homegadiagram3d ] . it is known that the vortex reversal under the action of a perpendicular static field is accompanied by the temporal creation and annihilation of a bloch point : the switching process , as a rule , is mediated by the creation of two bloch points , however , the single bloch point scenario was also mentioned . @xcite the bloch point propagation during the polarity reversal under the ac perpendicular field was also mentioned by @xcite . it should be noted that the bloch point is a 3d micromagnetic singularity , hence it does not exist in 2d simulations . the dynamics of the bloch point is not well studied to the moment . during the switching process we observed a new 3d picture of the switching , where some switching events are not completed : the vortex near one face surface rapidly reverses its polarity to opposite and returns back , while the vortex near the second face surface saves its polarity during this time . . ] to gain some insight to the switching mechanism , we need a model which allows the magnetization reversal process . it is worth reminding that in the continuum limit the vortex states with different polarities are separated by an infinite barrier . in the spin lattice the barrier becomes finite @xcite and the reversal can occur . it is already known from our previous paper@xcite that the dominating contribution to the switching mechanism is caused by the exchange interaction inside the vortex core . that is why to describe the polarity reversal process we use here the discrete _ reduced core model _ , which was initially introduced by @xcite for the vortex instability phenomenon . later the vortex core model was developed to analyze the vortex polarity switching in heisenberg magnets.@xcite one has to note that the reduced core model does not pretend a quantitative agreement with simulations . it is the simplest model which allows to describe a rich variety of different regimes of vortex polarity dynamics , including the resonant behavior , the weakly nonlinear regimes , the reversal dynamics , and the chaotic regime . we consider the anisotropic classical heisenberg disk shaped system with thickness @xmath78 and the radius @xmath79 , assuming that the magnetization of the magnet is uniform along the thickness . in terms of the normalized magnetic moment @xmath80 the energy of such a magnet with the account of the interaction with magnetic field reads @xmath81\\ & - a_0 ^ 2 m_s l_z \sum_{{\bm{n } } } { \bm{m}}_{{\bm{n}}}\cdot { \bm{b}}(t ) , \end{split}\ ] ] where the vector @xmath82 connects nearest neighbors of the three dimensional cubic lattice with the lattice constant @xmath83 , @xmath84 is the exchange constant , the parameter @xmath85 is the effective anisotropy constant , and @xmath86 is the saturation magnetization . according to this model the planar vortex is stable when @xmath87 , where @xmath88 for the square lattice.@xcite in a cylindrical frame of reference @xmath89 the planar vortex distribution is described by [ eq : vortex ] @xmath90 where @xmath91 is a vortex chirality ( we use here the positive sign in calculations below ) . when @xmath92 , the nonplanar vortex appears , which is characterized by the well localized out of plane magnetization @xmath93.@xcite ( -0.5,-0.5 ) grid ( 3.5,3.5 ) ; ( 0,0 ) ( 0.9 , 0.9 ) ; ( 0,0 ) ( 1.20344867377942 , 0.4143806095590807 ) ; ( 1,1 ) circle ( 3pt ) ; ( 1,2 ) circle ( 3pt ) ; ( 2,1 ) circle ( 3pt ) ; ( 2,2 ) circle ( 3pt ) ; ( 0,0 ) circle ( 2pt ) ; ( 0,3 ) circle ( 2pt ) ; ( 3,0 ) circle ( 2pt ) ; ( 3,3 ) circle ( 2pt ) ; ( 1,0 ) circle ( 2pt ) ; ( 2,0 ) circle ( 2pt ) ; ( 3,1 ) circle ( 2pt ) ; ( 3,2 ) circle ( 2pt ) ; ( 2,3 ) circle ( 2pt ) ; ( 1,3 ) circle ( 2pt ) ; ( 0,2 ) circle ( 2pt ) ; ( 0,1 ) circle ( 2pt ) ; ( 0,0 ) ( 0.75895,-0.25298 ) ; ( 0,0 ) ( 0.75895,0.25298 ) ; ( 0,0 ) ( 0.25298 , 0.75895 ) ; ( 0,0 ) ( -0.25298 , 0.75895 ) ; ( 0,0 ) ( -0.75895,0.25298 ) ; ( 0,0 ) ( -0.75895,-0.25298 ) ; ( 0,0 ) ( 0.25298 , -0.75895 ) ; ( 0,0 ) ( -0.25298 , -0.75895 ) ; ( 0,0 ) ( 0.56569 , 0.56569 ) ; ( 0,0 ) ( -0.56569 , 0.56569 ) ; ( 0,0 ) ( -0.56569 , -0.56569 ) ; ( 0,0 ) ( 0.56569 , -0.56569 ) ; ( 0,0 ) ( 0.7564148604794535 , 0.26045452356572535 ) ; ( 0,0 ) ( -0.26045452356572535 , 0.7564148604794535 ) ; ( 0,0 ) ( -0.7564148604794536 , -0.2604545235657251 ) ; ( 0,0 ) ( 0.26045452356572507 , -0.7564148604794536 ) ; ( 0,0 ) arc ( 20:42:8 mm ) ; at ( 0,0 ) @xmath94 ; ( 0,0 ) arc ( 100:170:3 mm ) ; ( 0,0 ) arc ( 190:260:3 mm ) ; ( 0,0 ) arc ( 280:350:3 mm ) ; ( 0,0 ) arc ( 10:80:3 mm ) ; at ( 0,0 ) 4 ; at ( 0,0 ) 1 ; at ( 0,0 ) 2 ; at ( 0,0 ) 3 ; at ( 0,0 ) 12 ; at ( 0,0 ) 5 ; at ( 0,0 ) 6 ; at ( 0,0 ) 7 ; at ( 0,0 ) 8 ; at ( 0,0 ) 9 ; at ( 0,0 ) 10 ; at ( 0,0 ) 11 ; in the reduced core approach , we suppose that only the four magnetic moments of the first coordinate shell can vary , forming the vortex core ; all the other moments are fixed in the sample plane in a vortex configuration , see fig . [ fig : core - schematic ] . by symmetry , all four moments are characterized by the same out of plane magnetization @xmath95 and equal in plane phase @xmath94 , which is determined as a deviation from the vortex configuration . therefore , the magnetization distribution of the first coordinate shell is describe as follows : @xmath96 we consider the core magnetization @xmath95 , which has the meaning of the _ dynamical vortex polarity _ , and the in plane _ turning phase _ @xmath94 as two collective variables . the energy of the model , normalized by @xmath97 has the form ( see appendix [ app : core ] for details ) : @xmath98 where we introduced the reduced anisotropy parameter @xmath99 , the reduced field intensity @xmath100 , the reduced field frequency @xmath101 , and the rescaled time @xmath102 . we use @xmath103 ( @xmath104 ) and @xmath105 in a majority of numerical calculations below . note that such a value of the @xmath106parameter is chosen for illustrative purposes , it does not fit to the correct material parameters from simulations . the magnetization dynamics in the reduced core model can be described by the following equations ( see appendix [ app : core ] for details ) : @xmath107,\\ \dot \psi & = \mu - \frac{\lambda \mu \cos\psi}{\sqrt{1-\mu^2 } } + h\sin \omega\tau - \frac{\eta \lambda \sin\psi}{\sqrt{1-m^2 } } , \end{split}\ ] ] where the overdot means the derivative with respect to @xmath108 and @xmath109 is a gilbert damping coefficient . the ground state of the model corresponds to @xmath110 in terms of the core model two opposite values of @xmath111 describe vortices with opposite polarities @xmath111 . , @xmath112 , @xmath113 , @xmath114 , @xmath115 . the effective double well potential @xmath116 , see , is plotted in the inset . ] at ( 0,2.5 ) ; ( pointer ) at ( 1.9,4.12 ) ; ( leftcorner ) at ( 0,5.5 ) ; ( leftcorner ) ( @xmath2 ) ( pointer ) ( @xmath3 ) ( @xmath4 ) ( @xmath5 ) ( @xmath6 ) cycle ; at ( @xmath7 ) ; ; ( pointer ) at ( 1.85,4.17 ) ; ( leftcorner ) at ( 3.025,5.5 ) ; ( leftcorner ) ( @xmath2 ) ( pointer ) ( @xmath3 ) ( @xmath4 ) ( @xmath5 ) ( @xmath6 ) cycle ; at ( @xmath7 ) ; ; ( pointer ) at ( 7.05,4.1 ) ; ( leftcorner ) at ( 6.05,5.5 ) ; ( leftcorner ) ( @xmath2 ) ( pointer ) ( @xmath3 ) ( @xmath4 ) ( @xmath5 ) ( @xmath6 ) cycle ; at ( @xmath7 ) ; ; ( pointer ) at ( 7.,4.15 ) ; ( leftcorner ) at ( 9.075,5.5 ) ; ( leftcorner ) ( @xmath2 ) ( pointer ) ( @xmath3 ) ( @xmath4 ) ( @xmath5 ) ( @xmath6 ) cycle ; at ( @xmath7 ) ; ; ( pointer ) at ( 2.77,3.7 ) ; ( leftcorner ) at ( 0.75,0.8 ) ; ( leftcorner ) ( @xmath117 ) ( pointer ) ( @xmath118 ) ( @xmath119 ) ( @xmath120 ) (@xmath121 ) cycle ; at ( @xmath122 ) ; ; ( pointer ) at ( 4.6,0.87 ) ; ( leftcorner ) at ( 8.9,2.21 ) ; ( leftcorner ) ( @xmath119 ) ( @xmath120 ) ( @xmath121 ) ( @xmath123 ) ( pointer ) ( @xmath124 ) cycle ; at ( @xmath122 ) ; ; let us start our analysis with a system without damping , @xmath125 . supposing that the turning phase is small enough , @xmath126 , one can easily exclude @xmath94 from the consideration . in this case the eqs . correspond to the effective lagrangian @xmath127 this simplification allows us to interpret the complicated dynamics as the motion of a particle with variable mass @xmath128 in the double well potential @xmath129 under a periodic pumping , see the inset in the fig . [ fig : core : nonlinamp ] . the linear oscillations near the equilibrium state correspond to the harmonic oscillations of the effective particle in one of the wells ; the eigenfrequency of such oscillations is @xmath130 let us describe the nonlinear regime of the dynamics . in spite of the small damping in the system , its value can be comparable with the pumping intensity . therefore we consider below the full set of the model equations eqs . . to analyze the weakly nonlinear regime , we use the multiscale perturbation method@xcite . when the field intensity is much less than the frequency detuning ( @xmath131 ) , we can limit ourselves to a three scale expansion in the form @xmath132 using such the expansion , one can derive from eqs . the resonance curve @xmath133 as the solution of the following equation @xmath134 with @xmath135 being the amplitude of oscillations , see appendix [ app : wna ] for details . the typical nonlinear resonance curve is shown in the fig . [ fig : core : nonlinamp ] , cf . fig . [ fig : sim : nonlinamp ] . we analyze the strongly nonlinear regimes solving numerically eqs . in a wide range of parameters ( @xmath136 ) , see the diagram of switching events in the fig . [ fig : core : homega ] . the absolute minimum in the diagram corresponds to the switching in the range near the resonance frequency @xmath137 . other local minima correspond to resonances at double frequency , @xmath138 , and sub harmonics , @xmath139 and @xmath140 . note that all resonance frequencies are shifted in the low frequency direction due to the nonlinear resonance . we classified the dynamical regimes by using the method of poincar maps ( @xmath141 points per map ) . we constructed such maps for each pair ( @xmath142 , @xmath143 ) where the switching takes place . one can separate four oscillation regimes related to the corresponding regimes in the oommf simulations ( sec . [ sec : sim ] ) with vortex dynamics in the center of the sample : ( i ) the absence of switching , ( ii ) the chaotic dynamics , ( iii ) the regular polarity oscillations between two polarities @xmath144 , ( iv ) the switching with final oscillations around one of the points @xmath145 in the coordinates @xmath146 . \(i ) the border between the switching region and the region , where field or frequency are not enough for jumps between @xmath147 , shows a few well - defined resonance minima corresponding to resonances at @xmath140 , @xmath139 , @xmath137 and @xmath138 . 0.32 0.32 0.32 \(ii ) the chaotic dynamics occurs for the low - frequency part of the fig . [ fig : core : homega ] and in the stretched region between resonances at @xmath137 and @xmath138 . an example of the temporal evolution is shown in the fig . [ fig : core : homega]c . the projection of the phase diagram on the @xmath146 plane ( see fig . [ fig : core : homega]d ) looks similarly to the pseudo phase diagram in the fig . [ fig : sim : pseudofield](a ) : the projection of trajectory is not closed and representation point makes a lot of windings around both @xmath144 . the shape of the chaotic poincar maps depends on the frequency . they show the shape of strange attractors , see fig . [ fig : core : poincareevolution ] . their cantor structure is ill defined due to using low damping . note , that they are similar to strange attractors for the duffing oscillator@xcite ( nonlinear oscillator with quadratic and cubic nonlinearities in the double - well potential ) . however , the reduced core model has a more complicated nonlinearity term ; using the mechanical analogy one can speak about the motion of a particle with a variable mass ( 7 ) in a double well potential . \(iii ) the main part of the diagram of switching events [ fig : core : homega ] is occupied by the region of regular dynamics . the most frequently observed poincar maps for this case contain some number of stable focuses . the observed numbers are @xmath148 , 15 , 16 , 18 , 21 , 24 , 30 and 96 . the most frequently observed ones are 1 ( the grey region in the fig . [ fig : core : homega ] ) and 3 ( included into the blue region in the fig . [ fig : core : homega ] ) . some of the points with a higher number of focuses demonstrate a complicated regular dynamics in phase space , see fig . [ fig : core : homega]e . the analogue of quasi rectangular regular polarity oscillations in oommf simulations is found in the @xmath140 region , see phase diagram in the fig . [ fig : core : homega]a and temporal evolution in the fig . [ fig : core : homega]b ( compare with the pseudo phase diagram shown in the fig . [ fig : sim : pseudofield](b ) and the temporal evolution in the fig . [ fig : sim : homega]d ) . \(iv ) the analogue of the intermittent switching linked with perturbation by the applying of the external field are shown by two dark grey color intensities in the fig . [ fig : core : homega ] . the final dynamics is an oscillation around upward or downward polarity ( points @xmath149 and @xmath150 in the phase space projection , respectively ) . an example of the temporal evolution is shown in the fig . [ fig : core : homega]f . as in oommf simulations , such oscillations typically occur near the border of the switching region . as in case of the chaotic dynamics , the resulting polarity is highly dependent on field , frequency and integration conditions . cc [ 41pt ] 0.48 & 0.48 + & 0.48 + as it is shown by analysis of the diagram of switchings events ( fig . [ fig : sim : homega ] ) , the vortex polarity switching under the action of the perpendicular resonant field produces multiple switchings during a short time comparable with one period of the acting field . such a situation is not unique among different polarity switching methods . so , for the axially asymmetric scenario , sufficient strength of gaussian pulse in the sample s plane produces more than one sequential vortex antivortex pair creation and annihilations.@xcite using an in plane rotating field with frequency @xmath151 codirectional with the vortex polarity @xmath152 ( @xmath153 ) stabilizes the vortex in the center of the sample . but when @xmath154 the reversal occurs in a specific range of the field intensities and frequencies , above which multiple switching was observed . @xcite a similar picture was reported in ref . for the current induced switching . however for the further application in contrast to multiple switching a unidirectional vortex polarity reversal , is needed . because the pumping does not select any direction of the vortex polarity , the most natural way to avoid the multiple reversal consists in limiting the pulse duration . we test an influence of a short wave train in the form @xmath155 , \\ 0 , & \text{otherwise } , \end{cases}\ ] ] where @xmath156 is the number of periods of the sinusoidal magnetic field in the wave train . we investigate the vortex dynamics under the action of the field on the resonant frequency @xmath157 . the vortex polarity is observed during a long time @xmath158 in order to damp magnons . the response of the magnetization to the field shows the nonlinear dependency on @xmath26 and @xmath159 , see fig . [ fig : train ] . when field intensity and period numbers are small , there are only small oscillations of the magnetization inside the vortex core . when @xmath159 becomes larger than some critical value , a typical system behaviour looks like a few switchings , which also occurs when the field is already turned off . a unidirectional switching from upward polarity to downward is observed for @xmath65mt and @xmath160 and some higher fields . we check the controllability of the discussed switching method by applying of series of wave trains . the series of wave trains is applied to the relaxed vortex . the time interval between trains is varied with steps of @xmath161 in different series . for @xmath162mt and @xmath163 the first switching occurs in 627ps and the vortex starts to relax ( field is turned off at the time 429ps ) . the sequence containing 3 wave trains allows to get controllable unidirectional vortex polarity reversal with minimal interval between wave trains of @xmath164ns , see fig . [ fig : unidirectional ] . these time intervals correspond speed of changing state of such memory cell about 250mhz . the core model also gives the same qualitative result , see fig . [ fig : unidirectionalcore ] . the axially symmetric switching of the vortex core under the action of periodic pumping was very recenltly predicted in refs . . @xcite were concerned with switching events : for the typical nanodot size the switching at the resonant frequency occurs during 600ps . @xcite computed the diagram of switching events where they noticed the existence of resonances on double and triple harmonics and shown that the exchange energy becomes higher than the threshold value for the vortex core reversal . in this work we study the long and short time vortex dynamics and propose an analytical model which describes the phenomena of full scale simulations . in order to explain the complicated vortex dynamics , we use here the reduced core model . @xcite it should be noted that this model does not pretend a quantitative agreement with simulations . in particular , it does not provide even the eigenfrequencies of the radially symmetric magnon modes , which is rather complicated task . @xcite nevertheless , the model we use is the simplest one which allows to describe a rich variety of different regimes of vortex polarity dynamics . this model provides a simple physical picture of the switching phenomenon in terms of the nonlinear resonance in a double well potential . such a potential arises mainly from the exchange interaction : the presence of two wells corresponds to the energy degeneracy with respect to the direction of the vortex polarity ( up or down ) ; the energy barrier between the wells becomes higher as the discreteness effects become less important . one has to stress that the switching process is forbidden in the continuum theory . in a real magnet the magnetization reversal is possible due to the discreteness of the lattice . that is why to describe the switching analytically we use the discrete core model : the switching can be considered as the motion of an effective mechanical particle with a variable mass in the double well potential . under the action of periodical pumping the particle starts to oscillate near the bottom of one of the wells . when the pumping increases , there appear nonlinear oscillations of the particle ; under a further forcing the particle overcomes the barrier , which corresponds to the magnetization reversal process . the chaotic dynamics of the magnetization is studied for domains walls@xcite and current induced phenomena in monodomain nanoparticles.@xcite very recently the existence of incommensurate chaotic vortex dynamics in spin valves was reported.@xcite in our case the chaos enters in the vortex polarity switching process due to the periodical pumping of the system with two equivalent equilibrium states as it happens in a duffing oscillator.@xcite the periodic pumping does not select the preferable vortex polarity direction which causes multiple switchings under the action of sufficiently high fields and frequencies . however , accurate fitting of the pulse duration and the time interval between sequential pulses allows to obtain a controlled unidirectional core reversal . thus , the chaotic , regular and controlled vortex polarity dynamics could find applications in physical layer data encryption@xcite and memory devices.@xcite in the current study we do not consider thermal effects on the vortex dynamics . the influence of the temperature was found to be not essential for the current induced motion of an individual vortex in py nanodisks @xcite . nevertheless it should be noted while magnetic vortices are stable up to very high temperature,@xcite the heating can influence the gyroscopical vortex dynamics @xcite . the heat can induce the vortex dynamics in the system . @xcite the temperature induced vortex dynamics also can influence the critical fields for vortex nucleation and annihilation.@xcite we expect that the physical picture discussed in the paper with a variety of different dynamical regimes survives with the temperature . the thermal effects will cause the shift of boundaries between different regimes . o.v.p . and d.d.s . thank the university of bayreuth and computing center of the university of bayreuth , where a part of this work was performed , for kind hospitality . o.v.p . acknowledges the support from the bayhost project . d.d.s . acknowledges the support from the alexander von humboldt foundation . f.g.m . acknowledges the support by micinn through fis2011 - 24540 , and by junta de andalucia under project no . all simulations results presented in the work were obtained using the computing cluster of kiev university @xcite and bayreuth university @xcite . taking into account the explicit form of the magnetic field , , the energy reads@xcite @xmath165- a_0 ^ 2 m_s l_z b_0 \sin\left(2\pi f t\right ) \sum_{{\bm{n } } } m_{{\bm{n}}}. \end{split}\ ] ] now we incorporate here the reduce core ansatz . then the energy reads @xmath166 after the renormalization the eq . [ eq : energy-2 ] takes the form , where @xmath167 , @xmath168 . the magnetization dynamics follows the landau gilbert equations @xmath169 with @xmath109 being a gilbert damping coefficient and the rescaled time @xmath102 . substituting eqs . into the eq . we obtain the equations for the @xmath146 : @xmath170 substituting eq . into we get the eq . . let us consider the weakly nonlinear case for eqs . . using the series expansion , the time derivative reads @xmath171 and the equations of motion break into three pairs of equations for the different orders in @xmath172 : @xmath173 the solution of eqs . and reads @xmath174 following the floquet theory @xcite , one needs to omit all secular terms . thus , eqs . and show @xmath175 and eqs . and gives the equation for @xmath176 : @xmath177 by solving the eq . one obtains the eq . . 51ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ] + 12$12 & 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty link:\doibase 10.1080/00018732.2012.663070 [ * * , ( ) ] @noop _ _ ( , , ) link:\doibase 10.1126/science.1075302 [ * * , ( ) ] link:\doibase 10.1063/1.2807274 [ * * , ( ) ] link:\doibase 10.1063/1.3551524 [ * * , ( ) ] http://link.aps.org/abstract/prb/v59/p7010 [ * * , ( ) ] link:\doibase 10.1103/physrevb.61.9449 [ * * , ( ) ] http://dx.doi.org/10.1038/nature05240 [ * * , ( ) ] http://www.sciencedirect.com/science/article/b6tjj-447dcmv-3/2/84e9fdfbf9e0cab3d3182fc4db4b4032 [ * * , ( ) ] http://link.aps.org/abstract/prb/v67/e094410 [ * * , ( ) ] http://dx.doi.org/10.1134/s1063783407100186 [ * * , ( ) ] link:\doibase 10.1103/physrevb.79.172410 [ * * , ( ) ] link:\doibase 10.1103/physrevb.80.100405 [ * * , ( ) ] link:\doibase 10.1103/physrevb.81.094431 [ * * , ( ) ] link:\doibase 10.1063/1.2957013 [ * * , ( ) ] link:\doibase 10.1063/1.3687909 [ * * , ( ) ] link:\doibase 10.1063/1.4705690 [ * * , ( ) ] http://www.ujp.bitp.kiev.ua/files/journals/58/6/580611p.pdf [ * * , ( ) ] link:\doibase 10.1140/epjb / e2003 - 00057-y [ * * , ( ) ] http://stacks.iop.org/0022-3727/43/i=26/a=264004 [ * * , ( ) ] http://math.nist.gov/oommf/ [ `` , '' ] http://link.aps.org/abstract/prl/v98/e117201 [ * * , ( ) ] @noop _ _ ( , ) p. link:\doibase 10.1103/physrevb.49.8780 [ * * , ( ) ] http://books.google.com.ua/books?id=pt3vaaaamaaj[__ ] , applied mathematical sciences ( , ) http://books.google.com.ua/books?id=3zbvaaaamaaj[__ ] , a wiley interscience publication ( , ) http://books.google.com.ua/books?id=eh6rmwz51nic[__ ] , physics textbook ( , ) link:\doibase 10.1063/1.2770819 [ * * , ( ) ] link:\doibase 10.1063/1.2780107 [ * * , ( ) ] link:\doibase 10.1103/physrevb.80.094415 [ * * , ( ) ] http://dx.doi.org/10.1134/s1063776109070103 [ * * , ( ) ] http://dx.doi.org/10.1134/s106377610705010x [ * * , ( ) ] , http://dx.doi.org/10.1038/nmat1237 [ * * , ( ) ] link:\doibase 10.1103/physrevb.71.052403 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.99.134101 [ * * , ( ) ] http://dx.doi.org/10.1038/nphys2362 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.74.5028 [ * * , ( ) ] , ed . , @noop _ _ , ed . ( , ) p. http://link.aps.org/abstract/prb/v74/e014424 [ * * , ( ) ] http://dx.doi.org/10.1029/2002jb002195 [ * * , ( ) ] link:\doibase 10.1103/physrevb.83.224424 [ * * , ( ) ] link:\doibase 10.1016/j.physleta.2012.09.039 [ * * , ( ) ] link:\doibase 10.1063/1.3694757 [ * * , ( ) ] link:\doibase 10.1063/1.3360841 [ * * , ( ) ] http://cluster.univ.kiev.ua/eng/ [ ] http://www.rz.uni-bayreuth.de/ [ ] @noop _ _ , edited by ( , )

to consider the most simple models is sometimes necessary for studying the charge carrier and thermal transport in solids . indeed mathematical approximations used on more sophisticated models may make less evident , or even completely suppress , certain effects . for the transport coefficients the latter statement surely concerns the thermoelectric power and the thermal conductivity . for studying the electrical resistivity theoretical methods are surely simpler , better developed and more reliable . when there is more than a single scattering source the electrical resistivity is sufficiently accurately treated in metals under the assumption of matthiessen rule @xcite ( see exceptions in concluding remarks ) . further assumptions are necessary to find simple and effective expressions for the thermoelectric power and the thermal conductivity . the gorter nordheim rule@xcite , which is usually applied to treat the thermoelectric power , is based on the assumption of the mott(-jones ) formula validity , however , restricted to the elastic scattering @xcite and the adiabatic approximation @xcite . for the thermal conductivity the wiedemann franz law is a usual basis on which the separation of the electronic and lattice contribution to the experimental thermal conductivity is made , whence for comparing experimental and theoretical data . the validity of the wiedemann franz law is also restricted to elastic collisions . clearly to resolve the question of elasticity and inelasticity of the scattering is essential for comparing experimental data and theoretical laws in particular for the thermoelectric power and the thermal conductivity . the most simple inelastic scattering model of the charge carriers by _ phonons _ describes most features @xcite of the thermoelectric power in normal metals . consequently , an approach well accounting for the energy transfer between the charge carriers and the scattering quasiparticles should be applied beyond that based on the mott formula@xcite . it is known that the elastic approximation is sufficient for considering the scattering of the conduction electrons by _ static impurities_. this leads to a constant , temperature independent residual resistivity , when the potential is appropriately chosen and the electron gas is degenerate . the simplest and well founded potential which leads to the residual resistivity in metals is the screened coulomb potential ( or yukawa potential ) provided that impurity concentration is sufficiently low . phenomenological considerations @xcite show that without assuming a behavior for the electrical residual resistivity one can not explain the low temperature behavior of the _ electronic _ contribution to the _ thermal _ conductivity . at moderate temperatures , along with the scattering by phonons the impurity scattering explains the magnitude and the basic features of the temperature dependence of the thermal conductivity of normal nonmagnetic @xcite and even magnetic @xcite metals , provided that the magnetic moment in the latter is periodic and well localized . it was shown indeed in ref . , devoted solely to the thermal conductivity , @xmath0 , that the magnetic scattering was of equal importance to the impurity and the phonon scattering . however , @xmath0 was neither compared to the electrical resistivity @xmath1 , nor to the thermoelectric power @xmath2 , for example calculated in the same temperature interval in ref .. moreover the final results were not presented through standard functions such as the bloch grneisen ones . we shall do so in the present paper , but confining ourselves to non magnetic impurity and phonon scattering only . as in our previous papers we shall apply the ziman variational approach @xcite within the boltzmann equation formalism , an approach in which the notion of `` relaxation time '' is not used . thus , the relationship to the mott formula is not straightforward @xcite . however , terms responsible for elastic and inelastic scattering can be easily selected , at least when the conduction band is parabolic . within this approach one can generalize the matthiessen rule : we consider a superposition of the transition probabilities themselves within the boltzmann equation formalism @xcite instead of the ( integrated ) relaxation time components . therefore the electrical resistivity , the thermoelectric power and the thermal conductivity are considered here with respect to impurity and phonon scattering on the same ground , _ i.e. _ both model and approximations . for this purpose it is sufficient to introduce a single parameter which accounts for the mutual contribution of the phonon and the impurity scattering to the transport coefficients . a similar model accounting for the scattering by phonons and impurities was considered in the fifties @xcite within the kohler variational formalism @xcite . however , the mathematical method was more complicated and less physical than that developed later by ziman @xcite . it did not allow @xcite to get qualitatively correct results for the thermoelectric power in contrast to those for the electrical resistivity and the electronic thermal conductivity . the finding of an accurate solution of the ( bloch)boltzmann equation is not a past problem of the fifties or the seventies but still a present day problem . numerical methods are used for this purpose . recently @xcite the allen method @xcite of solving the equation with a good numerical convergence has been used to study the influence of the electron electron scattering on the electrical and thermal conductivity . a comparative discussion of the approximation made while using the ziman variation method @xcite and the allen one @xcite is given in sec.v . in the following section ii we shortly describe the ziman variational method and the general results when applied to the model here defined . certain general results and formulae are also given in appendix a for completeness . in section iii the matrix elements describing the phonon and impurity scattering are calculated and final formulas for the electrical resistivity , electronic thermal conductivity and thermoelectric power are systematically presented . in section iv the temperature dependence of the transport coefficients and that of the lorenz number is numerically illustrated . the visual presentation is made by using dimensionless functions . an analysis of appropriate material constants allowing to estimate the magnitude of the final transport coefficients is found in appendix b. section v contains the concluding remarks . we assume that the current carriers are free electrons with a parabolic isotropic energy band @xmath3 spectrum . we also assume that the concentration of the electron gas ( of both subbands with spin up or down ) @xmath4 ( or the fermi energy @xmath5 ) is sufficiently high to consider the gas to be degenerate . then one can assume that the chemical potential is @xmath6 , and all relevant functions of @xmath7 can be considered as asymptotic expressions of the argument @xmath8 . in particular the fermi - dirac integrals [ 1 ] f_n(z)=_0^x , z= can be represented as [ 2 ] f_n(z ) z^n+1/(n+1 ) , z / t for sufficiently large @xmath9 . the carriers are assumed to be scattered by static , nonmagnetic impurities , which concentration is sufficiently low to have a linear influence on the electrical resistivity . the scattering of carriers by phonons will be considered on the same footing . however , the phonon ( quasiparticle ) scattering will be treated as inelastic while the impurity scattering is elastic in its nature , since the impurity is usually assumed to be more heavy than the electron effective mass . even though some calculations can be performed without assuming the form of the potential ) we assume the form of the impurity potential to be the screened ( yukawa ) potential [ 3 ] u ( ) = - with @xmath10 , @xmath11 the effective valence and the screening constant @xmath12 . the fourier transform @xmath13 of the above function is responsible for the magnitude of the carrier scattering ; @xmath14 with @xmath15 is the electron wave vector transferred during the scattering . the transport relaxation time following from @xmath16 and consequently the residual resistivity is temperature independent for a degenerate electron gas . their magnitude is related to the fermi wave vector @xmath17 or the energy @xmath5 through the function @xcite [ 4 ] 2 _ 0 ^ 2 = ( 1 + 4 ^ 2 ) - 4 ^ 2/(1 + 4 ^ 2 ) ( 2 ) , where @xmath18 . note that the standard considerations @xcite yield that @xmath19 and thus one can estimate @xmath20 since the effective bohr radius @xmath21 is of the order of @xmath17 . the phonon system and the electron phonon scattering are considered in the bloch approximation , _ i.e. _ phonons in the harmonic approximation and the electron phonon interaction in the deformation potential approximation . we restrict the magnitude of the phonon wave vector , @xmath22 , to be within the debye sphere . then , for sufficiently high values of the fermi energy , @xmath23 ( see , e.g . , appendix b in ref . ) , the phonon limited transport coefficients are expressed in terms of the bloch(grneisen ) functions [ 5 ] f_n(t)=_0 ^ 1/t . a relation between the deformation potential approximation and the method based on the dynamical structure factor introduced by van hove @xcite is described in sec . v. as we mentioned in the introduction we apply the ziman variation method @xcite to consider the combined effect of the impurity and phonon scattering on the transport coefficients . the method consists in calculating the elements of a matrix describing the scattering and two vectors governing the flow of the electric charge and heat . the exact results correspond to an infinite order matrix and infinite dimension current vectors ( see , _ e.g. _ appendix a and sec . iv in ref . ) . here , we truncate the matrix and vectors to four elements and two components , respectively . this is the lowest order approximation , which accounts well for both the electrical and thermal current if the variational trial functions are appropriately chosen@xcite . this choice is known to be [ 6 ] _ i()=()(-)^i-1 , where @xmath24 and @xmath25 , while @xmath26 denotes the unit vector in the direction of the applied temperature gradient or that of the external electric field causing the charge flow , and @xmath27 is the electron velocity . in this approximation the transport coefficients : the electrical resistivity @xmath1 , the thermoelectric power or the seebeck coefficient @xmath2 and the ( electronic ) thermal conductivity @xmath0 for cubic crystals , are given as such [ 7 ] = ( p_11p_22 - p_12 ^ 2)^-1 , [ 8 ] s=[p_22j_1u_2-p_12(j_1u_2+j_2u_1)+p_11j_2u_2](t ) ^-1 [ 9 ] = ( j_1 ^ 2u_2 ^ 2+j_2 ^ 2u_1 ^ 2 -2 j_1j_2u_1u_2)(t ) ^-1 . with = p_22j_1 ^ 2 - 2p_12j_1j_2+p_11j_2 ^ 2 , the elements of the scattering matrix being [ 10 ] p_ij&=&c(,)u_ij(, ) , + u_ij(,)&=&[_i()-_i( ) ] [ _ j()-_j( ) ] , and the components of the trial currents [ 11a ] j_i =- e(- ) ( ) _ i ( ) , [ 11b ] u_i=(- ) ( ) ( -)_i ( ) , where @xmath28 is the transition probability to be determined on the ground of an appropriate hamiltonian describing the scattering . the function @xmath29 denotes the fermi dirac one , @xmath30 $ ] , and @xmath31 is the electron charge . the integrals in ( [ 10 ] ) and ( [ 11a])-([11b ] ) should be taken over the brillouin zone but in our free electron considerations we let [ 12 ] _ 0^ d ( ) , where @xmath32 is an element of the isoenergy surface . the integrals with respect to the electron energy in ( 10 ) for the degenerate electron gas are described in appendix b of ref . 8 . in the case of elastic scattering by impurities , as here , the integrals can be treated as simply . namely , we take advantage of the approximate equalities [ 30 ] _ 0^(-f/)(- ) g ( ) = ( ^2/3 ) ( t)^2 [ g/ ] _ , [ 29 ] _ 0^(-f/)(- ) ^2 g ( ) = ( ^2/3 ) ( t)^2 g ( ) . where @xmath33 is an arbitrary function . the trial currents ( [ 11a])-([11b ] ) can be represented @xcite as acombination of the fermi dirac integrals ( see appendix b ) : @xmath34 , @xmath35 , @xmath36 , @xmath37 , where @xmath38 and @xmath39 are material constants . for the electron phonon system the @xmath40matrix elements can be derived ( see ref . ) without any further approximation . they are expressed by double integrals with respect to the electron energy and the transferred electron momentum ( wave vector ) . for a debye cutoff and in the limit @xmath41 they can be represented used in ref.8 is introduced here directly in the definition of @xmath42 . ] by @xmath43 , where @xmath44 is a material constant and @xmath45 are dimensionless combinations of the the bloch grneisen functions of the argument @xmath46 ( @xmath47 is the debye temperature ) . for the impurity scattering an expression similar to the above is readily obtained . in this case the @xmath42 s are simply related to @xmath48 defined in ( 4 ) . the explicit form of the functions @xmath42 will be given in the next section . the magnitude of @xmath44 and that of other material constants are estimated in appendix b. in view of the above , by calculating the scattering matrix elements in the degenerate limit , retaining only @xmath49 in case of the electrical resistivity and the lowest order terms in the @xmath50 in the trial currents ( 12)-(13 ) , we obtain [ 13 ] = = , [ 14 ] s = s_0(t ) + s_1(t ) , [ 15 ] = = , where @xmath51 in ( [ 13 ] ) and ( [ 15 ] ) is the trial current in its asymptotic form and @xmath52 is the sommerfeld value of the lorenz number . the explicit form of the functions @xmath53 and @xmath54 of ( [ 14 ] ) having a more complicated dependence on the matrix elements @xmath55 are discussed in section iii b. as seen from ( [ 13])-([15 ] ) it remains to calculate the elements of the @xmath40matrix in order to find the final expressions for the transport coefficients . in our case the crux is to take into account the combined effect of the scattering by impurities and phonons . as we mentioned in the introduction we apply the matthiessen rule in a more generalized way @xcite , i.e. to the microscopic transition probabilities c(, ) themselves appearing in eq.([10 ] ) rather than to the final coefficients : the electrical and thermal resistivity . the variational method enables us to use the same approximation to both these coefficients and the thermoelectric power through the integrated probabilities , _ , we assume [ 16 ] c(, ) = c^(imp)(, ) + c^(ph)(, ) . this means that the resultant scattering probability for the phonon and impurity scattering can be simply considered as a sum of the scattering probabilities of the impurities ( the phonon scattering being quenched ) and by phonons ( those of the pure crystal ) . this , of course , implies that the elements of the resultant scattering matrix , ( [ 10 ] ) , are also the sums of the corresponding matrix elements for the phonon and impurity scattering . therefore , we write [ 17 ] p_ij = p_ij^(imp ) + p_ij^(ph ) . in formulae ( [ 7 ] ) - ( [ 9 ] ) . the electron - phonon contribution @xmath56 s have been derived in previous papers , _ e.g. _ , in ref . they are expressed in terms of the bloch(-grneisen ) functions ( [ 5 ] ) . the appropriate expressions for the electron - impurity contribution @xmath57 following from the yukawa potential ( [ 3 ] ) are given in this section , but first we write down the final expression for the electronic thermal conductivity with the influence of the residual electrical resistivity as a parameter . for this purpose we have to separate elastic and inelastic contribution to @xmath58 and therefore we write explicitly the matrices @xmath59 of the kernels in ( 10 ) . for the parabolic band @xmath60 they can be represented as [ 18 ] u_11&=&(/m)^2()^2 , + u_12=u_21&=&(/m)^2[(-)+(-)]()^2 , + u_22&=&(/m)^2[(-)^2 + 2(-)(- ) ] ( ) ^2+(-)^2[()^2+()^2].where ( @xmath61 ) is the transferred energy during the scattering , @xmath62 is the transferred wave vector and terms linear in @xmath63 and @xmath64 , which do not contribute to the integrals , have been neglected . it is seen that the second component in @xmath65 and the second and third one in @xmath66 contribute only when a change of the electron energy in the scattering is taken into account . it means that the latter terms describe the effects of _ inelasticity _ and the first terms in the case of @xmath65 and @xmath66 are expressed by the debye functions in case of pure elastic scattering and by the bloch(-grneisen ) functions in case of inelastic scattering ( see ref.8 ) . the terms proportional to @xmath67 and @xmath68 in @xmath65 and @xmath66 lead to an essential qualitative effect in the thermoelectric power and the thermal conductivity . ] in the scattering . with the definition [ 19 ] p_11^(ph ) = p_0^(ph)p_11^(ph)(t ) , with t = t / t_d and where @xmath69 is a material constant ( see appendix b ) , we get [ 20 ] p_22^(ph ) = p_0^(ph)(t)^2 [ p_22^(1,ph)(t ) + p_22^(2,ph)(t ) ] p_22^(1,ph ) + p_22^(2,ph ) , where [ 21 ] p_22^(1,ph)(t)&=&p_11^(ph)(t ) , + p_22^(2,ph)(t)&=&t^5[f_7(t)+ f_5(t)-f_6(t ) ] , and @xmath70 is a constant ( see appendix b ) . the explicit form of @xmath71 describing the electrical resistivity is [ 22 ] p_11^(ph)(t ) = t^5f_5(t ) . the two terms in @xmath65 , accounting respectively for the elastic and inelastic contributions , after integration with the transition probability for the phonon scattering can be represented by a single term , [ 23 ] p_12^(ph ) = p_21^(ph ) = p_0^(ph ) ( t ) p_12^(ph)(t ) with [ 24 ] p_12^(ph)(t ) = p_21^(ph)(t ) = t^5f_6(t ) . two material constants influence the magnitude of @xmath72 . these are the fermi energy , @xmath5 , and @xmath73 , where @xmath74 is the sound velocity ( see the table i in appendix ) . the latter energy ( or more precisely @xmath75 ) is a threshold energy for the phonon emission in the electron phonon scattering process ( see appendix j in ref . ) . assume now that the charged ( static ) impurity concentration is so low that one can consider them to be statistically independent . then the scattering of the conduction electrons by a single impurity contributes solely to the @xmath76 matrix elements and they are linear in their concentrations . in consequence the electrical resistivity is linear in the electron concentration and for the degenerate electron gas it is also temperature independent . the latter statement can be easily confirmed by the calculation of @xmath77 in the metallic limit . for further convenience we assume @xmath78 and due to the sommerfeld expansion of terms of ( [ 18 ] ) corresponding to an elastic scattering we have @xmath79 . therefore , the final expressions for @xmath80 and @xmath58 describing the impurity and phonon scattering are [ 25 ] p_11= p_0^(ph)p_11(t ) , _ 11(t ) = p_0^(ph)[+ p_11^(ph)(t ) ] + p_22 = p_0^(ph)(t)^2p_22(t ) , _ 22(t ) = [ + p_11^(ph)(t ) ] + p_22^(2,ph)(t ) . the above expressions for the matrix elements of the @xmath40matrix yield the electrical resistivity [ 26 ] = _ 0,ph[+ p_11^(ph)(t ) ] , where @xmath81 is a constant ( see appendix b ) , and the electronic thermal conductivity [ 27 ] = _ 0,ph , _ 0,ph = . note that the lorenz number reads [ 28 ] l = = l_0 and reduces to the sommerfeld value @xmath82 for elastic scattering , when one neglects @xmath83 in ( 25 ) . the results ( 26 ) describing the electrical resistivity are obviously standard ones and do not require any comments . as concerns the electronic thermal conductivity and the lorenz number similar expressions in terms of the bloch grneisen functions as those in the present paper were obtained by wilson @xcite and were numerically examined in ref . manifesting the influence of the impurity scattering . the wilson expression for the thermal resistivity in our notation reads [ 33 ] t^5f_5(t)+t^5 f_5(t)-t^5f_7(t ) . in order to discuss the thermoelectric power , @xmath2 , we introduce , as for the discussion of the thermal conductivity in ref . , a relative order of magnitude parameter @xmath84 , in terms of which for the yukawa potential ( [ 3 ] ) we have @xmath85 . it will be also convenient to introduce here the function @xmath86 in terms of which we express @xmath87_\ef = ( \ef)^{-1}\mit\psi ( 2\kf /\lambda)$ ] . in so doing the off - diagonal component of the @xmath40matrix responsible for the impurity scattering is linear in @xmath88 and reads [ 31 ] p_12^(imp)(t ) = p_0^(imp)(t)p_12^(imp ) , _ 12^(imp ) = ( 2/ ) . when these expressions are inserted into ( [ 8 ] ) and linear terms in @xmath89 are only accounted for , one obtains the following expressions for the coefficients in ( [ 14 ] ) [ 32 ] s_0(t ) = , s_1(t ) = -\ { - ( ) ^2 } . note that the impurity scattering has an influence on the term @xmath53 only through the matrix element @xmath90 in the denominator . if @xmath91 , the above formula corresponds to that obtained in ref . for pure phonon limited thermoelectric power . notice that the pure phonon limited thermoelectric power obtained within the wilson formalism@xcite do not represent any characteristic features of the experimentally observed @xmath2 behavior in simple monovalent metals . such results for @xmath2 obtained in ref . were already discussed there in terms of various approximations . to compute the transport coefficients as a function of @xmath92 we assume the value of @xmath93 and @xmath94 . the temperature dependence of the reduced electrical resistivity following from ( 26 ) , @xmath95 , which is shown in fig . 1 , and the seebeck coefficient in fig . 2 have been computed for @xmath96 and the corresponding values of @xmath97(0.75 ) = 0.0 , 0.0173 , 0.0259 , 0.0431 , 0.0862 , 0.173 respectively . as it is seen the impurity scattering slightly lowers the maximum of @xmath2 following from the transfer of the energy between the electron and phonon system@xcite . the values of @xmath70 and @xmath5 have been chosen to be the same as those in ref . , namely @xmath98k and @xmath99ev . the electronic thermal conductivity is considerably affected by the impurity scattering , what is seen from the data represented in fig . 3 and fig.4 . in fig.3 the reduced thermal conductivity ( @xmath100 ) is shown as a function of @xmath101 in the low temperature range such as in fig . 4.16 of ref . , where the results of wilson s considerations@xcite were already illustrated . in fig . 4 the results are shown in the same temperature range as the thermoelectric power in fig . we note that he dependence of the thermal conductivity as a function of temperature here obtained by the ziman variational method is qualitatively the same as that following from wilson s considerations@xcite . furthermore , at physically realistic temperatures the values of @xmath0 obtained by the latter method approach a saturation , which is characteristic of monovalent metals . under the assumption of the same values of the parameters describing the impurity scattering the saturation of @xmath0 following from wilson formula occurs at a so high temperature that it ceases to be meaningful . according to ziman@xcite , who examined the wilson formula , physically reasonable results for monovalent metals ( saturation and lack of minimum at intermediate temperatures ) are obtained when the umklapp scattering is taken into account . experimental results for rare earth intermetallic compounds exhibit the existence of a minimum at temperatures such as those seen in figs . 3 and 4 ( see e.g. fig . 7 in ref . and references therein ) , in which the magnetic scattering are also taken into account . the reduced lorenz number , @xmath102@xmath103 , as a function of @xmath101 is shown in fig . 5 for the same values of the parameters as @xmath100 in fig . the behavior of this quantity as a function of @xmath101 is qualitatively the same as that following from the wilson considerations @xcite . note that for only phonon scattering @xmath104(0 ) = 0 , whereas @xmath105 tends to infinity at zero temperature . using a simple physical electron model , applying the ziman variational method within the boltzmann equation approach and using the same level of approximations for each property we have calculated the phonon - limited electrical resistivity , thermoelectric power and the electronic thermal conductivity of a simple metal and studied how the presence of charged impurities influences these transport coefficients . the method allows us to determine the temperature dependence of the coefficients in a wide temperature interval and to express them in terms of simple and standard functions already useful for examining experimental data . however we are aware that to take into account a more complicated band structure is surely necessary in many cases . in this purpose the allen method@xcite is surely more appropriate though the task needs to bear upon numerical technicalities from the very beginning . note that the latter method is strictly applicable to metallic systems , when the electron gas is degenerate , while by using the method of this present paper one advantage is that one can easily consider semiconductor or semimetallic systems as well @xcite . a comment is in order on the results of the present paper obtained in the limit of the degenerate electron gas . as concerns the electrical resistivity and the electronic thermal conductivity ( or resistivity ) the final results are qualitatively the same as those obtained previously in refs .. in the case of the thermoelectric power however we can obtain a new evidence as a conclusion , i.e. the maximum at intermediate temperatures is caused by the energy transfer between the conduction electrons and phonons@xcite and is lowered as a result of the elastic scattering of the electrons on impurities . the final formal expression for the electronic thermal resistivity reads as a simple expression : this quantity is proportional to the scattering matrix element @xmath58 , much like the electrical resistivity is proportional to @xmath80 . the result is in fact model independent and is a consequence of confining the scattering matrix to four elements ( see appendix a ) . the range of applicability of the method we have used here above can not be finally discussed without referring to papers using more general and sophisticated methods and/or models . in the first place this concerns the papers in which the electron transport was studied within the formalism of the boltzmann equation@xcite . in the second place a relation to the linear response theory @xcite and the quantum transport equation @xcite will be given . however , by using these methods one can not obtain simple , compact , and useful expressions for the three transport coefficients , moreover being valid in the wide temperature region experimentally studied . further complications have also to be expected when taking into account the phonon drag @xcite , the virtual recoil @xcite or the electron phonon renormalisation @xcite . let us start our discussion with a comparison of our model and results with the other ones obtained by the ziman variational method within the formalism of the boltzmann equation . as we stressed at the beginning we confined ourselves to a model of the parabolic electron band and the debye model of the phonon system . this oversimplified model allowed us to gain a coherent and qualitatively correct description of the three transport coefficients : the electrical resistivity , thermoelectric power and the thermal conductivity ( resistivity ) for a simple and standard model of the scattering . any generalization of these results to an arbitrary band structure requires numerical studies at an initial stage . a compact formula in this general case of the ziman variational method has been obtained only for the electrical resistivity among the three coefficients , a formula which has been called ziman s resistivity formula it has been used to compute the electrical resistivity for a realistic fermi surface . again a more exact method is that of allen@xcite though a numerical one . the electrical resistivity is again simplest to compute but the results for the thermal conductivity @xcite and the thermoelectric power @xcite were also obtained by this method . the essence of the method is the separation of the @xmath63-variation during the scattering into two types : ( i ) an angular variation over the fermi surface , and ( ii ) a radial variation with @xmath106 perpendicular to the fermi surface describing the inelastic electron scattering . it is obvious from our results for the thermoelectric power @xcite that taking into account inelastic electron scattering , which is equivalent to considering a sufficient number of harmonics describing the radial variation within the allen formalism , is of major importance for obtaining a correspondence to experimental results if the scattering of the electrons by excitations such as phonons is to be accounted . in ref . , in which the lowest order radial harmonic was considered and the analysis was in fact restricted to the mott formula , a poor correspondence to the experimental results for the thermoelectric power was obtained . inelastic scattering and the radial harmonics are also important to describe properly the thermal conductivity . in ref . the radial harmonics were applied to discuss the influence of the electron electron scattering on the thermal conductivity . according to our knowledge no such considerations were made to analyze the influence of the electron phonon scattering neither on the thermal conductivity nor the thermoelectric power . the assumed statistical independence between the impurities as scattering centers restrict our considerations to low impurity concentration . this is consistent with the experimental fact that the matthiessen rule is valid below a critical residual resistivity @xcite above which the @xmath107 low temperature dependence ( following , for instance , from the bloch grneisen dependence of the electrical resistivity ) ceases to be valid . one can hardly estimate such a critical concentration without analyzing a specific system and experimental results . to estimate the magnitude of the transport coefficients out of the reduced values illustrated on each figure in sec . iv we present the order of magnitude of the relevant material constants in appendix b. the magnitude of @xmath70 in table i is the same as in ref . and corresponds to the magnitude of the sound velocity and the effective electron mass(es ) of real materials . we start our discussion of more sophisticated methods than the boltzmann equation formalism by noting that in the linear response theory the validity of the matthiesen rule , which was assumed by us in a generalized version , is related to the fact that the self - energy of the conduction electrons is the same as the sum of the phonon and impurity self - energy @xcite . this holds only in the lowest born approximation . when higher order processes are taken into account , _ i.e. _ , when interference between the electron - phonon and electron - impurity interaction is to be considered , the keldysh quantum transport equation formalism @xcite is apparently better . in ref . the same model as in the present paper was considered to study the influence of the electron - phonon - impurity interference on the thermoelectric power . a considerable nonlinearity of @xmath2 as a function of temperature was found , the stronger the higher impurity effective charge . the effect can be hardly compared with that studied in the present paper since it has been obtained under an assumption of elasticity for the electron - phonon scattering whereas the assumption of inelasticity of the electron - phonon scattering is essential @xcite for obtaining the maximum of @xmath2 found in the present paper and ref . note that as a result of a weak elastic impurity scattering ( considered within the born approximation ) the bump in @xmath2 is found to be lowered with the strength of the impurity interaction or the impurity concentration . the electron - phonon renormalization @xcite leads also to a nonlinearity of @xmath2 as a function of @xmath108 . the effect was studied under the assumption of elastic electron - phonon interaction or in the adiabatic approximation . again there can be no discussion between our results and those in ref . and except for the region of very low temperatures , where the effect of elasticity and renormalization are not effective and where @xmath2 is a linear function of @xmath108 with the same slope @xcite as in the pure crystal @xcite . though the impurity scattering is known to play an important role in the thermal conductivity at least in the low temperature region there is a very limited number of papers devoted to this coefficient aspect . the basic difficulties and problems related to applying the methods of the linear response formalism to treating the electronic thermal conductivity problem in the pure electron - phonon system are described in ref .. however , such considerations as those in ref . have not be generalized for a system with impurities . on the other hand the considerations of refs . , devoted to thermoelectric effects for a model of debye phonons with fixed impurities , have not been used by these authors in an appropriate way to consider the electronic thermal conductivity . therefore , a complete theory of the electronic thermal conductivity still requires a thorough consideration of the mutual role of the electron - phonon - impurity interference , phonon renormalization and elasticity of the scattering , much the same as on the thermoelectric power . for comparison to data , in not too good metals , the phonon heat contribution should be also taken into account for the thermal conductivity . a question may arise if our results can be compared to experimental ones in a straightforward , quantitative way . the answer seems that these are the alkali metals , which exhibit essentially spherical fermi surface to within a few per cent , that should be first of all considered as materials for the comparison . it occurs , however , that the phonon system and the electron phonon interaction in these metals are too complicated to be well accounted by the applied approximations . it concerns in the first place , ( i ) , the continuous medium approximation used to treat the phonon system and the deformation potential approximation applied to the electron phonon interaction ( see , e.g. , ref . ) . in the second place , ( ii ) , the debye approximation is insufficient for the alkali metals . finally , in the description of the electron scattering we also neglect the umklapp processes , which occur important even at very low temperatures @xcite in the alkali metals . the approximations ( i ) mean that the longitudinal phonons of long wavelength are only considered , whereas the transverse phonons @xcite and those of short wavelength ( contributing also to the umklapp process ) have to be taken into account @xcite . due to a strong elastic anisotropy the debye approximation ( ii ) is not appropriate even at low temperatures since a constant frequency surface is distorted from the debye spherical form @xcite . furthermore , the strong influence of the electronic system on the behavior of the phonons cause that the basic symmetry relations between the elastic constants ( cauchy relations ) are no longer valid @xcite and , therefore , even the basic results following from the continuous medium approximation of ref . may be questioned . all these factors cause that the simple bloch approach have to be used with a great caution in attempts of the _ quantitative _ description of the electrical and thermal transport in the alkali metals . the most early investigations @xcite of the alkali metals indicated that one can fix such parameters as those of appendix b to fit the experimental electrical resistivity to the bloch grneisen function , provided that the temperature is not too low . thorough analysis @xcite based on the measured phonon spectra of the alkali metals @xcite showed later that the problem is more complicated and the effects mentioned by us above have to be in fact taken into account . therefore , even in case of these simplest metals the correct quantitative theory based on first principles is in fact impossible and a quantitative analysis have to be based on a cross section for the scattering of the conduction electrons written in terms of the phonon spectrum resulting from neutron scattering data @xcite . precise experimental measurements of the transport coefficients of the alkali metals at low temperatures @xcite support this conjuncture and reveals other physical effects which should be taken into account for a proper quantitative analysis in this temperature region . as concerns the thermoelectric power our results indicate that other factors than those in case of the electrical resistivity are important . the impurity scattering occurred less essential than expected . within the assumed model this is the parameter @xmath70 which besides @xmath5 is solely responsible for the strength of the inelastic scattering of electrons by phonons and controls the magnitude and sign of @xmath2 . the obtained temperature dependence of @xmath2 is the same in character as that obtained experimentally @xcite for all the the alkali metals except li ( which exhibits a peculiar behavior @xcite but the fermi surface of which comes closest to the brillouin zone boundaries ) . in search for a quantitative theory of @xmath109 of a specific metal one should also estimate the order of the contribution of the phonon drag @xcite , that following from the electron phonon renormalization @xcite and the electron phonon impurity interference . the described above factors which influence the electrical resistivity in the alkali metals may occur of secondary importance . the effects of the impurity scattering manifest in most pronounced way in the behavior of the thermal conductivity but due to the validity the wiedemann franz law in the limit of low temperatures the effects of the impurity scattering can be easily extracted out of our results at low temperatures for a comparison with experimental ones @xcite . as concerns the comparison of the values of appropriate coefficients following from the theory and experimental data for the alkali metals the same problems as in case of the studies of the electrical resistivity may arise . the umklapp process are expected to be less dominant at low temperatures than in case of the electrical resistivity but the electron electron interactions may occur essential for the thermal conductivity since both the experimental and theoretical investigations indicate @xcite that electron electron scattering in the alkali metals impedes heat flow much more effectively than charge flow . kd thanks fnrs and kbn for financial aid during part of this work . ma thanks the wroclaw pan staff for its always warm welcome during various stays at the institute for low temperature and structure research . thanks also to various european programs , allowing for international exchanges . [ ( a.3 ) ] = _ i , j=1j_iq_ijj_j , l=-1t_i , j=1j_iq_iju_j , + m=_i , j=1u_iq_ijj_j , & & n=-1t_i , j=1u_iq_iju_j , where @xmath113 and @xmath114 ( @xmath115 ) are respectively the electron and thermal trial currents defined by ( [ 11a])-([11b ] ) and @xmath116 is the inverse matrix to the scattering matrix @xmath76 , i.e. @xmath117 . in the approximation consisting in abbreviating the combinations in ( [ ( a.3 ) ] ) to two components ( @xmath118 ) we obtain ( 7)(9 ) from ( [ ( a.2 ) ] ) . for a degenerate electron gas , i.e. for sufficiently large @xmath119 , the final expressions for the transport coefficients are ( 14 ) ( 16 ) . notice that the thermal resistivity in particular is proportional to @xmath58 . the coefficient which determines the dimension and the order of the magnitude of the phonon scattering matrix depends on the effective electron mass , @xmath123 , the volume of the primitive cell , @xmath124 , its total mass , @xmath125 , the sound velocity @xmath74 , the phonon wave vector cutoff , @xmath126 , and the deformation potential constant , @xmath127 . it reads [ a.6 ] _ 0,ph = ( ) ^6 ( v_s q_d)^-1 = ( ) ^6(t_d)^-1 , where @xmath132 is the free electron mass and @xmath133 . for the value of @xmath134 assumed for our representative data we have @xmath135@xmath136 cm . in table i we show some representative data of @xmath137 for @xmath138ev , @xmath139k , @xmath140 and the values of @xmath141 given in the table . they correspond to those in ref . , where numerical relations between @xmath141 and other constants describing the phonon system are presented . the screened coulomb potential scattering contribution to the transport is described in terms of the dimensionless functions @xmath142 and @xmath143 defined by ( [ 4 ] ) and ( [ 31 ] ) . the coefficient @xmath144 of the scattering matrix elements is = = ( 2/ ) ( 2/ ) and shall estimate @xmath153 . it means that @xmath154 ( 31 ) is @xmath155 . for the mass of the unit cell @xmath156 , its volume @xmath157nm@xmath158 and @xmath159ev we have @xmath160 . this is an underestimated value in many cases , thus we may assume @xmath161 with @xmath162 . for @xmath138ev , @xmath139k @xmath140 the value of @xmath163 is : @xmath164 cm . the latter value along with those ones for @xmath81 following from the values of @xmath137 presented the table i yields an estimate of @xmath153 . wilson , _ the theory of metals _ ( cambridge university press , cambridge , 1954 ) . ziman , _ principles of the theory of solids _ ( cambridge university press , cambridge , 1972 ) . s.m sharma , phys . rev.b * 20 * , 1514 ( 1979 ) barnard , _ thermoelectricity in metals and alloys _ ( taylor & francis , london , 1972 ) . k. durczewski and m.ausloos , phys . b * 61 * , 5303 ( 2000 ) . m. jonson and g.d . mahan , phys . b * 21 * , 4223 ( 1980 ) . m. jonson and g.d . mahan , phys . b * 42 * , 9350 ( 1990 ) . k. durczewski and m. ausloos , phys . b * 53 * , 1762 ( 1996 ) ; ibid . * 54 * , 17 224 ( 1996 ) . j.e . parrott and a.d stuckes , _ thermal conductivity of solids _ ( pion ltd , london , 1975 ) . ziman , _ electrons and phonons _ ( clarendon press , oxford , 1962 ) . a.rassili and m. ausloos , j. magn . magn . mater . * 147 * , 341 ( 1995 ) . k. durczewski and m. ausloos , j. magn . magn . mater . * 51 * , 230 ( 1985 ) . e. h. sondheimer , proc.cambridge philos . soc . * 43 * , 571 ( 1947 ) . w.w . schultz and p.b.allen , phys . b * 52 * , 7994 ( 1995 ) . p.b.allen , phys . rev . b * 17 * , 3725 ( 1978 ) . l.van hove , phys . rev . * 95 * , 249 ( 1954 ) . c.kittel , _ introduction to solid state physics _ ( john wiley & sons . inc . , new york , 1996 ) . ziman , proc . soc ( london ) * a226 * , 436 ( 1954 ) . a. rassili , k.durczewski , and m. ausloos , phys . b*58 * , 5665 ( 1998 ) . a.vilenkin , and p.l.taylor , phys . b * 18 * , 5280 ( 1978 ) . y. ono , prog . phys * 46 * , 757 ( 1971 ) . livanov and a.v . sergeev , phys . b * 48 * , 13 137 ( 1993 ) . m.yu . reizer and a.v . sergeev , zh . * 92 * , 2291 ( 1987 ) ; sov . jetp * 65 * , 1291 ( 1987 ) . reizer and a.v . sergeev , zh . . fiz . * 93 * , 2191 ( 1987 ) ; sov . jetp * 66 * , 1250 ( 1987 ) . belashchenkov , and b.v . livanov , zh . . fiz . * 111 * , 2237 ( 1997 ) ; sov . jetp * 84 * , 1221 ( 1997 ) . blatt , p.a . schroeder , c.l . foiles and d. greig , _ thermoelectric power of metals _ ( plenum press , new york and london , 1976 ) . nielsen and p.l.taylor , phys . b * 10 * , 4061 ( 1974 ) . j. opsal , b.j . thale and j. bass , phys . lett . * 36 * , 1211 ( 1976 ) . a.b . kaiser , phys . b * 29 * , 7088 ( 1984 ) . g.grimval , _ the electron phonon interaction in metals _ ( north holland , amsterdam new york oxford , 1981 ) . p.b.allen , w.e.picket and h.krakauer , phys . b * 37 * , 7482 ( 1988 ) . j.j . krempasky and a. schmid , j. low temp . phys . * 34 * , 197 ( 1979 ) . keldysh , zh . . fiz . * 47 * , 1515 ( 1964 ) ; sov . jetp * 20 * , 1018 ( 1965 ) . klemens , physica * 69 * , 171 ( 1973 ) . r.j.m.van vucht , h. van kempen and p.wyder , rep . . phys . * 48*,853 ( 1985 ) . j.k.darby and n.h.march , proc . b * 84*,591 ( 1964 ) . m.p.greene and w.kohn , phys . rev . * 137*,a 514 ( 1965 ) . a.d.b.woods , b.n.brockhouse , r.h . march , a.t.stewart and r.bowers , phys 128 * , 1112 ( 1962 ) . r.a.cowley , a.d.b.woods , and g.dolling , phys . rev . * 150 * , 487 ( 1966 ) . e.g.brovman and y.m . kogan in _ dynamical properties of solids , vol . 1 : crystalline solids , fundamentals . g.h.horton and a.a.maradudin ( north holland , amsterdam new york oxford , 1974 ) . j.s.dugdale and d.gugan , proc . roy.soc . a * 254 * , 184 ( 1960 ) . j.s.dugdale and d.gugan , proc . a * 270 * , 186 ( 1962 ) . .the values of the parameters allowing to estimate the magnitude of the electrical resistivity and electronic thermal conductivity for a free electron effective mass , when the fermi energy is @xmath150 ev and the debye temperature @xmath165k . [ cols="^,^,^,^,^,^ " , ] * figure 1 * temperature dependence of @xmath95 resulting from eq.(26 ) for values of the parameters @xmath93 and the relative scattering strength @xmath166 . they correspond to the values of @xmath97(0.75 ) which are 0.0 , 0.0173 , 0.0431 , 0.0862 , respectively . * figure 2 * temperature dependence of @xmath167 for values of the parameters as in fig.1 with @xmath99 ev , @xmath168 k , and the effective electron mass equal to that of a free electron . @xmath169 = @xmath170 = 86 @xmath171 v / k . * figure 3 * temperature dependence of the reduced thermal conductivity below @xmath47 , for @xmath172 and @xmath173 ; the corresponding values of @xmath152 are 0.0 , 0.0173 , 0.0431 , 0.0862 , 0.173 , respectively . for @xmath152 = 0 , the value of @xmath0 is infinite . note that for the magnitude of @xmath152 as large as 2.0 there is a plateau of @xmath0 instead of the maximum at @xmath174 . * figure 4 * temperature dependence of the reduced thermal conductivity in the same temperature range as that displayed for the thermoelectric power in fig.2 . the value of @xmath175 is the same as in fig.3 , i.e. 0.75 , and @xmath176 . * figure 5 * temperature dependence of the reduced lorenz number in the same temperature range as that of the thermoelectric power in fig.2 . the value of @xmath175 is 0.75 as in fig.3 , and @xmath177 .

outer layers of low - mass stars are unstable to thermal convection due to a rapid increase in opacity resulting from the partial ionization of hydrogen and the dissociation of h@xmath0 . below @xmath1 , main sequence stellar interiors are theorized to become fully convective along the main sequence @xcite . largely characterized by near - adiabatic convection , fully convective stars are considered the simplest stars to describe from a theoretical perspective . in fact , properties of fully convective stars predicted by stellar structure models are largely insensitive to input variables ( e.g. , the mixing length parameter , @xmath2 ) and input physics ( e.g. , nuclear reaction rates , element diffusion ; * ? ? ? * ; * ? ? ? discovery of significant radius discrepancies between observations and stellar model predictions for fully convective stars therefore presents a curious puzzle ( see , e.g. , * ? ? ? * ; * ? ? ? evidence indicating that stellar structure models can not properly predict radii of fully convective stars has been gathered from studies of detached eclipsing binaries ( debs ) . masses and radii can be measured for stars in debs with precisions below 3% provided the observations are of high quality and analyses are performed with care @xcite . presently , there are five debs with at least one fully convective component whose mass and radius has been quoted with precision below 3% : kepler-38 @xcite , kepler-16 @xcite , lspm j1112 + 7626 @xcite , koi-126 @xcite , and cm draconis ( hereafter cm dra ; * ? ? ? * ; * ? ? ? * ; * ? ? ? of these systems , only the fully convective stars of koi-126 can be accurately characterized by stellar evolution models @xcite . every other fully convective star appears to have a radius inflated compared to model predictions . most consequential are the inflated radii of the stars in cm dra . historically , the stars of cm dra are _ the _ fully convective stars against which to benchmark stellar models . as such , cm dra has been well - studied and rigorously characterized . over the years , discrepancies between model radii of cm dra and those determined from observations has grown . initial modeling efforts were optimistic that agreement could be achieved @xcite , but disagreement was quickly identified with the introduction of more sophisticated models @xcite and more precise mass and radius measurements @xcite . this disparity has been increased , yet again , with converging reports of the system s metallicity @xcite . strong magnetic fields maintained by tidal synchronization are presently considered the leading culprit producing the observed radius discrepancies ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? magnetic activity indicators , such as soft x - ray emission , ca ii h & k emission , and h@xmath3 emission , appear to correlate with radius inflation @xcite , providing evidence in favor of the magnetic hypothesis . theoretical investigations also support a magnetic origin of radius inflation for main sequence deb stars @xcite . despite significant evidence in favor of the magnetic hypothesis , several clues suggest otherwise . discovery of the hierarchical triple koi-126 in 2011 introduced a second pair of well - characterized fully convective stars whose masses and radii were measured with better than 2% precision . stellar evolution models are able to reproduce the properties of koi-126 ( b , c ) , as previously mentioned , based only on inferred properties from the more massive primary star @xcite . with orbital and stellar properties similar to cm dra , koi-126 ( b , c ) presents a sharp contrast to known modeling disagreements . adding to the evidence mounting against the magnetic hypothesis , fully convective stars in kepler-16 , kepler-38 , and lspm - j1112 + 7626 show inflated radii despite existing in long period ( @xmath4 17days ) systems . the fact that most inflated stars in debs appeared to exist in short period systems was proffered as strong circumstantial evidence in support of the magnetic hypothesis . however , only a few deb systems were known , all of which had short orbital periods due to inherent observational biases . with the influx of data from long time baseline photometric monitoring campaigns , including _ kepler _ and mearth , fully convective stars in long period debs have been shown to have inflated radii . of course , the presence of inflated low - mass stars in long period systems is only contradictory if the inflated stars are slowly rotating ( @xmath5kms@xmath6 ) . @xcite find that lspm j1112 + 7626 a rotates with a period of 65days , from which we infer an age of order 9gyr assuming the gyrochronology relation of @xcite . this result requires confirmation , but if confirmed , it would be seem likely that the fully convective , low - mass secondary is slowly rotating . a rotation period has also been measured for the primary star in kepler-16 @xcite . it was determined to be rotating with a period of nearly 36days , close to the pseudo - synchronization rotation period @xcite . if we assume the fully convective secondary has a similar rotation period ( from pseudo - synchronization ) , then it would have @xmath7 kms@xmath6 , well below the empirical velocity threshold thought to be required for fully convective stars to maintain a strong magnetic field @xcite . therefore , in at least two cases , it appears that slowly rotating fully convective stars exhibit inflated radii . in this paper , we extend our on - going investigation into the magnetic origin of inflated low - mass stellar radii , initiated in @xcite , to fully convective stars . a brief overview of how we include magnetic effects in our models is presented in section [ sec : dmestar ] . detailed analysis of two well - characterized debs , kepler-16 and cm dra , is given in section [ sec : ind_deb ] with a discussion of the results in section [ sec : disc ] . section [ sec : disc ] also provides comparisons with previous studies and a careful examination of the magnetic hypothesis . a summary of key results and conclusions is then given in section [ sec : summ ] . stellar evolution models used in this investigation are from the dartmouth magnetic evolutionary stellar tracks and relations ( dmestar ) program . dmestar was developed as an extension of the dartmouth stellar evolution program ( dsep ; * ? ? ? * ) , a descendant of the yale rotating evolution code @xcite . the magnetic version of the dartmouth stellar evolution code is described in @xcite , @xcite , and @xcite . we refer the reader to these papers for a thorough overview . the pertinent aspects of the standard , non - magnetic stellar evolution code for modeling fully convective stars are : the equation of state ( eos ) and the surface boundary conditions . all fully convective stars are modeled with the freeeos , a publicly available eos code written by alan irwin and based on the free energy minimization technique . we call freeeos in the eos4 configuration to provide a balance between numerical accuracy and computation time . with this eos , we are able to reliably model stars with masses above the hydrogen burning minimum mass @xcite . surface boundary conditions are prescribed using phoenix ames - cond model atmospheres @xcite . atmosphere structures are used to define the initial gas pressure for our model envelope integration . above @xmath8 , the gas pressure is determined an optical depth where @xmath9 . however , below @xmath8 , the regime where convection is sufficiently non - adiabatic extends deeper into the star @xcite . thus , we specify our boundary conditions at the optical depth @xmath10 in this mass regime . for the present work , we have extended the initial metallicity grid of model atmosphere structures @xcite by interpolating within the original set of structures . care was taken to ensure that the interpolation produced reliable results and that no discontinuities in either @xmath11 or the starting temperature were introduced . this will be discussed in a future publication . a set of atmosphere structures with finer metallicity spacing allows for more accurate predictions of stellar properties at metallicities that lie between the original grid spacings . implementation of a magnetic perturbation is described in detail by @xcite and @xcite . we abstain from providing a mathematical description and refer the reader to those papers . however , it is beneficial to review multiple variations on our basic formulation that arose in @xcite . these variations take the form of different magnetic field strength radial profiles and what we have called different `` dynamos . '' the latter refers not to a detailed dynamo treatment , but a conceptual framework that concerns from where we assume the magnetic field sources its energy . our treatment of magneto - convection depends on how we assume the magnetic field is generated . assuming that rotation drives the dynamo , as in a standard shell dynamo @xcite , leads to perturbations consistent with the idea that magnetic fields can stabilize a fluid against thermal convection ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? this assumption forms the basis of our magneto - convection formulation @xcite . however , permissible magnetic field strengths can reach upward of 6 kg at the model photosphere with exact upper limits determined by equipartition with the thermal gas pressure . this also results in interior magnetic field strengths that can grow nearly without limit owing to large internal gas pressures . models that rely on stabilizing convection with a magnetic field are hereafter referred to as having `` rotational dynamos . '' the form of the perturbation used for a star with an assumed rotational dynamo is not necessarily valid in all stellar mass regimes . this is particularly true in fully convective stars , where the interface region between the radiative core and convective envelope is thought to disappear . to address this , and the problem that magnetic fields in the rotational dynamo can grow without limit , we introduced a `` turbulent dynamo '' mechanism @xcite . this formulation assumes that the magnetic field strength at a given grid point within the model receives its energy from the kinetic energy of convecting material . therefore , the local alfvn velocity can not exceed the local convective velocity . it is a simple approach developed to address zeroth - order effects within the already simplified convection framework of mixing length theory . for low - mass stars , this methods places an upper limit of around 3 kg for surface magnetic field strengths , consistent with observed upper limits of average surface magnetic field strengths @xcite . although rotation is still required for turbulent dynamo action @xcite , magnetic field strengths are more sensitive to properties of convection . we note , again , that the term `` dynamo '' is used loosely . our formulations of magneto - convection do not rigorously solve the equations of magnetohydrodynamics . instead , we seek to capture physically relevant effects on stellar structure in a phenomenological manner consistent with actual dynamo processes . each of the above descriptions rely equally on a prescribed magnetic field strength profile within the star . models of the `` dipole radial field '' variety are the standard sort introduced in @xcite . this profile is characteristic of a magnetic field generated by a single current loop centered on the stellar tachocline . given a surface magnetic field strength , the radial profile of the magnetic field is determined by calculating the peak magnetic field strength at the tachocline . lacking a tachocline , we define fully convective stars to have a peak magnetic field strength at 15% of the stellar radius ( @xmath12 ) . this is loosely based on results from three - dimensional magnetohydrodynamic ( mhd ) models of fully convective stars @xcite . mhd models indicate that the magnetic field reaches a maximum around @xmath13 @xcite . note that this maximum is not a sharply defined peak in the magnetic field strength profile since the profile is based largely on equipartition of the magnetic field with convective flows . still , we adopt @xmath14 knowing this caveat , which we address in a moment . the rest of the interior magnetic field is then calculated by assuming the magnetic field strength falls off steeply towards the core and surface of the star . explicitly , @xmath15 where @xmath16 is the prescribed surface magnetic field strength , @xmath17 is the radius of the tachocline normalized to the total stellar radius , and @xmath18 is the radius within the star normalized to the total stellar radius . to increase the peak magnetic field strength for a given surface magnetic field strength , we implemented a gaussian profile ( section 4.4.1 in * ? ? ? the peak magnetic field strength is still defined at the tachocline in partially convective stars and at @xmath19 in fully convective stars . however , instead of a power - law decline of the interior magnetic field strength from the peak , the peak was set as the center of a gaussian distribution . thus , @xmath20,\ ] ] where @xmath21 controls the width of the gaussian and @xmath22 is defined with respect to the surface magnetic field strength . the width of the gaussian depends on the depth of the convection zone . deeper convection zones have a wider gaussian profile compared to stars with thin convective envelopes @xcite . hereafter , we will refer to these as `` gaussian radial profile '' models . note that this prescription has no physical motivation beyond providing stronger interior magnetic fields . we define a third radial profile for this study : a `` constant @xmath23 radial profile . '' when a turbulent dynamo is invoked , the magnetic field radial profiles described above can cause the alfvn velocity to exceed the convective velocity within the model interior . this happens quite easily in models of fully convective stars where the peak magnetic field strength is defined deep within the model . as a result , perturbed convective velocities become imaginary leading convergence problems when solving the equations of mixing length theory . if we assume that kinetic energy in convective flows generates the local magnetic field , an alfvn velocity exceeding the convective velocity will lead to a decaying magnetic field strength . eventually , equipartition will be reached . we avoid iterating to a solution by using a profile that assumes a constant ratio of magnetic field to the equipartition value , @xmath24 , where @xmath25 . this factor , @xmath23 , was introduced in @xcite as a means of comparing reduced mixing length models ( i.e. , * ? ? ? * ) to our turbulent dynamo models . perturbations to the equations of mixing length theory are now expressed as functions of @xmath23 . removing energy from convection slows convective flows such that @xmath26 where @xmath27 is the convective velocity prior to losing energy to the magnetic field . we restrict @xmath28 to avoid imaginary convective velocities . reduction of convective velocity causes a significant reduction in convective energy flux since @xmath29 . steeping of the background temperature gradient , @xmath30 , occurs as radiation attempts to transport additional energy . this increase , @xmath31 , over the non - perturbed temperature gradient is @xmath32 where @xmath33 is the characteristic squared velocity of an unimpeded convecting bubble over a pressure scale height . this is a sort of terminal convective velocity , where @xmath34 is the local acceleration due to gravity , @xmath2 is the convective mixing length parameter , @xmath35 is the pressure scale height , and @xmath36 is the coefficient of thermal expansion . resulting effects on convection likely represent an upper limit on the effects of such a dynamo mechanism in inhibiting thermal convection . note that when using this radial profile and dynamo mechanism , modifications to the schwarzschild criterion ( i.e. , the formalism outlined in * ? ? ? * ) are neglected . we have chosen to study the debs kepler-16 and cm dra in detail . lspm j1112 + 7626 was not modeled in detail because it lacks a proper metallicity estimate required for a careful comparison with models . kepler-38 was also not selected as initial comparisons suggest that there may be some issues with modeling the primary . finally , koi-126 has been modeled in detail previously and does not appear to require magnetic fields @xcite . as we will show below , magnetic models of kepler-16 b and cm dra produce similar results that can be generalized to all fully convective main sequence stars in debs . l c c d11 mass ( @xmath37 ) & @xmath38 & @xmath39 + b12 mass ( @xmath37 ) & @xmath40 & @xmath41 + radius ( @xmath42 ) & @xmath43 & @xmath44 + @xmath45 ( k ) & @xmath46 & @xmath47 + @xmath48 $ ] & + age ( gyr ) & [ tab : kep_16_prop ] report of the first circumbinary exoplanet was announced by @xcite with their study of kepler-16b . while the planet is interesting in its own right , what made the finding even more interesting was that the two host stars formed a long - period low - mass deb . this enabled a precise characterization of the stars and circumstellar environment . kepler-16 contains a k - dwarf primary with a fully - convective m - dwarf secondary in a 41 day orbit . properties of the two stars are listed in table [ tab : kep_16_prop ] . in a follow - up investigation , @xcite estimated the composition and age of the system . an age of @xmath49gyr was estimated using gyrochronology and an age - activity relation based on ca ii emission . spectroscopic analysis of the primary revealed the system has a solar - like metallicity of [ fe / h ] @xmath50dex . they also compared properties of the kepler-16 stars to predictions of stellar evolution models . @xcite model predictions agreed with properties of the primary at the given age . however , the radius of the fully convective secondary star was larger than model predictions by @xmath51 . independent confirmation of the stellar properties and model disagreements was provided by @xcite , who found a best fit age of 1 gyr with [ fe / h ] @xmath52 using dartmouth models . radial velocity confirmation of the component masses within @xmath53was later obtained by @xcite . they found masses that were 5% lower than the original masses @xcite . of particular note , is that mass ratio is different between the two studies . @xcite attempted to pin - point the origin of this discrepancy , but were unable to do so with complete confidence . despite the disagreement , the spectroscopic masses largely confirm that masses derived using a photometric - dynamical model are reliable @xcite . however , a slight mass difference significantly alters comparisons with stellar evolution models . that is , if we assume that the derived masses do not heavily influence the radius and effective temperature predictions . we therefore opt to treat the two different mass estimates independently to assess how these differences affect our modeling efforts . hereafter , masses originally quoted by @xcite will be referred to as masses , whereas the revised values of @xcite will be referred to as masses . table [ tab : kep_16_prop ] lists masses measured by each group . we first focus our attention on kepler-16 a , as the analysis will directly influence our analysis of the secondary . plotted in figure [ fig : kep16_a_nmag ] are non - magnetic dartmouth models computed at the measured masses of the primary provided by and . three separate tracks are illustrated for each mass estimate , corresponding to [ fe / h ] @xmath54 , and @xmath55 @xcite . in this figure , the horizontal shaded region highlights the observed radius with associated @xmath56 uncertainties . ages are determined by noting where the mass tracks are located within the bounds of the empirical radius constraints . similarly , we confirm that when the mass track has the required radius it also has an appropriate effective temperature in figure [ fig : kep16_a_nmag](b ) . given a primary mass from , we find that standard stellar evolution models match the stellar radius and temperature at an age of @xmath57gyr . when the model radius equals the precise empirical radius ( @xmath58 ) , the associated model effective temperature is 4354 k , which is within 17k of the spectroscopic effective temperature @xcite . we note that this age is also consistent with the estimated age from @xcite . agreement between models and observations does not guarantee the validity of the masses over the masses . in fact , this agreement is rather expected . the effective temperature and metallicity for the primary were determined using spectroscopy made easy ( * ? ? ? * hereafter sme ) , which relies on theoretical stellar atmospheres . our models also rely on theoretical atmospheres , although phoenix model atmospheres , used by our models , adopt a different line list database ( see * ? ? ? * and references therein ) than the theoretical atmospheres used by sme ( vald : vienna atomic line database ; * ? ? ? * ) . in a sense , the fact that we find such good agreement with the effective temperature for a given @xmath59 and metallicity ( i.e. , those of kepler-16 a ) may be a better test of the agreement between different stellar atmosphere models rather than a test of the interior evolution models . what is encouraging , is that we derive appropriate stellar properties at an age consistent with gyrochronology and age - activity relations @xcite . age consistency is not ensured by agreement between stellar atmosphere structures . our previous discussion may be erroneous if we adopt an incorrect mass . suggest this is the case . the effect of adopting the lower b12 masses is displayed in figure [ fig : kep16_a_nmag ] . assuming the radius measurement remains constant , we derive an age of @xmath60gyr for the primary star . the effective temperature associated with the model also appears too cool compared to observations at the measured metallicity ( @xmath61dex ) . relief is found by lowering the metallicity by @xmath62dex , which increases the temperature by @xmath63k . this is enough to bring the model temperature to within @xmath56 of the spectroscopic value @xcite . there is a caveat : the spectroscopic analysis by @xcite relied on fixing the stellar @xmath59 as input into sme . thus , the temperature and metallicity are intimately tied to the adopted @xmath59 . reducing the mass of the primary by 5% , as is done by , but leaving the radius fixed to the value leads to a decrease in @xmath59 of @xmath64 dex . such a change in the fixed value of @xmath59 may decrease the derived effective temperature and bring the model and empirical temperatures into agreement . all things considered , we believe it is safe to assume that a shift in mass does not introduce any significant effective temperature disagreements at a given radius . this is predicated on the fact that the temperature is safely above @xmath65k . below this temperature , theoretical atmosphere predictions start to degrade with the appearance of molecular bands . more simply stated , we have no reason to doubt the dartmouth model predicted temperature for kepler-16 a , regardless of the adopted mass . a model age of 11gyr for the primary mass appears old given the multiple age estimates provided by @xcite . is it possible that the system is actually 11gyr old , but appears from rotation and activity to be considerably younger ? consider that a @xmath66day rotation period of the primary , as measured by @xcite , is nearly equal to the pseudo - synchronization rotation period , predicted to be 35.6days @xcite . one might think that tidal effects are unimportant in a binary with a 41 day orbital period , however , subsequent tidal interactions briefly endured when the components are near periastron can drive the components towards pseudo - synchronous rotation . pseudo - synchronous rotation will keep the stars rotating at a faster rate than if they were completely isolated from one another . the time scale for this to occur is approximately 3gyr @xcite , meaning that the rotation period is not necessarily indicative of the system s age . the primary will have approximately the same rotation period at 11gyr as it will at 3gyr . furthermore , the timescale for orbital circularization is safely estimated to be between @xmath67 @xmath68gyr @xcite . tidal evolution calculations are subject to large uncertainties and should approached as an order of magnitude estimate . but , we are unable to immediately rule out the possibility that kepler-16 has an age of 11gyr . we have so far neglected any remark on the agreement between standard models and kepler-16 b. this comparison is carried out in figure [ fig : kep16_b_nmag ] . no effective temperature estimates have been published , explaining our neglect of the @xmath45-radius plane . as with figure [ fig : kep16_a_nmag ] , mass tracks are shown for multiple metallicities . standard model mass tracks for kepler-16 b are unable to correctly predict the observed radius at an age consistent with estimates from the primary . the disagreement is independent of the adopted metallicity , which introduces @xmath69 variations in the stellar radius at a given age . no evidence is available to support the idea that kepler-16 b is magnetically active . still , we look to magnetic fields to reconcile the model predictions with the observations . all possible scenarios relating to the various stellar mass estimates are considered . explicitly , we compute models for both the and masses and then attempt to fit the observations using the magnetic dartmouth stellar evolution models . the primary mass implies that kepler-16 is approximately 2gyr old , as shown in figure [ fig : kep16_a_nmag](a ) . due to the consistency between the model derived age and the empirically inferred age , we see no reason to introduce a magnetic perturbation into models of kepler-16 a. @xcite observe only moderately weak chromospheric activity coming from the primary , further supporting our decision . thus , we seek to reconcile models of kepler-16 b with masses at 2gyr . magnetic models of kepler-16 b were computed for a range of surface magnetic field strengths . a dipole radial profile was used and the perturbation was applied over a single time step at an age of 1gyr . mass tracks with a @xmath70 kg and @xmath71 kg surface magnetic field strength are shown in figure [ fig : kep16b_dip ] along side a standard mass track for comparison . note , that even though the perturbation is applied at an age close to the age we are trying to fit , the models adjust to the perturbation rapidly . we are unable to produce a radius inflation larger than 1% , even with a strong surface magnetic field strength of @xmath71 kg . the peak field strength in the @xmath71 kg model ( located at @xmath72 ) is approximately 1.8 mg ( @xmath73 ) . discussion about how real such a magnetic field might be is deferred until section [ sec : mag_conv ] . for the moment , we are interested in knowing what magnetic field strength is required to reconcile models with observations . we next constructed magnetic models using a gaussian radial profile , which are shown in figure [ fig : kep16b_d11 ] . the magnetic perturbation was again introduced as a single perturbation at 1gyr . two surface magnetic field strengths were used , @xmath70 kg and @xmath74 kg . the model with a @xmath74 kg surface magnetic field strength causes the model to become over inflated compared to the observed radius at 2gyr . from figure [ fig : kep16b_d11 ] , we see that a magnetic field intermediate between 4 kg and 5 kg is required to produce agreement between the models and observations . in contrast to results for partially convective stars @xcite , dipole and gaussian radial profiles produce different results for a given surface magnetic field strength in fully convective stars . this is caused by a difference in peak magnetic field strengths ( dipole : 1.8 mg , @xmath75 ; gaussian : 30 mg , @xmath76 ) . we will return to this issue in section [ sec : fc_profs ] . finally , figure [ fig : kep_16_beq ] shows the influence of a constant @xmath23 profile on a model of kepler-16 b. recall , this radial profile invokes a turbulent dynamo mechanism . we started with a rather high value of @xmath77 to gauge the model s reaction to this formulation . impact on the radius evolution of a @xmath78 star is effectively negligible . further increasing @xmath23 has no effect on the resulting radius evolution . adopting masses mainly alters the age derived from stellar models . instead of 2gyr , we infer an age of 11gyr from models of kepler-16 a , as was shown in figure [ fig : kep16_a_nmag](a ) . the relative radius discrepancy noted between models and kepler-16 b is increased by approximately 2% over the case . magnetic models were computed for kepler-16 b with the mass estimate using a gaussian radial profile introduced at an age of 1gyr . these models are shown in figure [ fig : kep16b_b12 ] . we did not generate models with a dipole radial profile or with a constant @xmath23 profile given the lack of radius inflation observed for these magnetic field profiles for the masses . we find that a surface magnetic field strength slightly weaker than @xmath71 kg is required to fit the observations . this translates to a nearly @xmath79 mg ( @xmath76 ) peak magnetic field strength . a stronger field strength was required when using the mass instead of the mass because of the 1% increase in the radius discrepancy mentioned above . kepler-16 , although it has components with fundamental properties measured with better than 3% precision , must be approached cautiously when comparing to stellar models . mass differences quoted in the literature obscure how well models perform against the observations . though the masses are determined with high precision by each group , the 3% 5% uncertainty introduced by their disagreement overwhelms the measurement precision . this produces an age difference of 9gyr for the primary star , calling into question inferences drawn about this system from stellar models . it is also unclear how revising the masses would affect estimates of the component radii , the system s metallicity , and the primary star s effective temperature . furthermore , since the primary may be rotating pseudo - synchronously , it is not possible to rule out either of the age estimates . until the mass differences are resolved , _ care must be taken when comparing kepler-16 to stellar models_. however , the disparity between the observed radius of kepler-16 b and model predictions is apparent regardless of the adopted mass estimate . changing the mass estimate simply changes the level of inferred radius inflation . this result appears robust and can be used to test the magnetic hypothesis for low - mass star radius inflation . our magnetic models require magnetic field strengths of similar magnitudes . surface magnetic field strengths are on the order of 4 kg 6 kg with interior field strengths of a few tens of mg . cm dra ( gj 630.1 ac ) contains two fully convective low - mass stars and is arguably one of the most important systems for benchmarking stellar evolution models . shortly after cm dra was discovered by luyten , @xcite uncovered that the star was actually a deb . it was not clear from observations whether the secondary was a dark , very low - mass companion such that no secondary eclipse occurred or whether the two components were of nearly equal mass . evidence was tentatively provided in favor cm dra being a single dme star with a dark companion @xcite , although more observations were encouraged as the author found a possible hint of a secondary eclipse . any speculation that the secondary companion to the dme star of cm dra was a dark , lower - mass object was laid to rest by @xcite who obtained radial velocity measurements to provide the first determination of stellar parameters for both stars . l c c[b ] mass ( @xmath37 ) & @xmath80 & @xmath81 + radius ( @xmath42 ) & @xmath82 & @xmath83 + @xmath45 ( k ) & @xmath84 & @xmath85 + @xmath48 $ ] & + age ( gyr ) & [ tab : cm_dra_prop ] + following lacy s determination of the stellar properties , subsequent studies refined and improved the masses and radii of the cm dra stars , pushing the measurement precision below 2% @xcite . currently accepted values @xcite are listed in table [ tab : cm_dra_prop ] . additional information about the cm dra stars has been revealed in recent years . @xcite provided an analysis of a nearby white dwarf ( ) common proper motion companion and estimated an age of 4.1 @xmath86 0.8 gyr . this age was based upon the cooling time of the white dwarf and its estimated progenitor lifetime , which depends on the initial to final mass relation for white dwarfs . in light of recent advances in our understanding of white dwarf cooling ( e.g. , * ? ? ? * ) and the initial - to - final mass relation ( e.g. , * ? ? ? * ) we intend to re - examine the question of the age of cm dra in a future paper . for the present work , we adopt the aforementioned age . deriving a metallicity for cm dra has proven more difficult than estimating its age . near - infrared ( nir ) spectroscopic studies that fit theoretical model atmospheres to atomic and molecular features have consistently favored a metal - poor abundance ( @xmath87 } \approx -0.6 $ ] ; * ? ? ? * ; * ? ? ? * ; * ? ? ? optical spectroscopy of molecular features ( cah & tio ; * ? ? ? * ) and nir photometric colors @xcite , on the other hand , suggests that the system might have a near - solar metallicity . more recent techniques relying on empirically - calibrated narrow - band nir ( h & k band ) spectral features have converged on a value of [ m / h ] = @xmath88 @xcite . further support for the latter estimate is provided by the photometric color - magnitude - metallicity relation of @xcite , which predicts [ fe / h ] @xmath89 . for this study , we adopt [ fe / h ] = @xmath90 presented by @xcite who controlled for uncertainties introduced by orbital phase variations . it is well documented that the stars of cm dra are inflated compared to standard stellar models @xcite . this fact has become steadily more apparent since an initial comparison was performed by @xcite , which found little disagreement . a precise estimate of the level of disagreement depends on the adopted metallicity ( see , e.g. , * ? ? ? * ; * ? ? ? * ) , but the problem is robust . figure [ fig : cmdra_nmag ] demonstrates the level of disagreement compared to standard dartmouth mass tracks.it also illustrates how metallicity influences the stellar models . given that there is no evidence for polar spots on cm dra ( see section [ sec : spots ] ) , we have elected to use the radii established by @xcite . the level of disagreement observed in figure [ fig : cmdra_nmag ] is between 5% 7% for each star , with cm dra a have a consistently smaller deviation than cm dra b by about 0.5% . @xcite , as a consequence of their metallicity estimate , have essentially doubled the radius disagreement from 3% 4% , noted in previous studies ( e.g. , * ? ? ? * ) , to 5% 7% . cm dra is magnetically active . balmer emission and light curve modulation due to spots have been recognized since very early investigations @xcite . frequent optical flaring has also been continually noted ( e.g. , * ? ? ? * ; * ? ? ? further details on the flare characteristics of cm dra may be found in the work by @xcite . the system is also a strong source of x - ray emission based on an analysis of data in the rosat all - sky survey bright source catalogue @xcite . high levels of magnetic activity and a short orbital period ( @xmath91d ) have been used to justify the need for magnetic perturbations in stellar evolution models of cm dra . such studies were carried out by @xcite , @xcite , and @xcite using various methods ( see section [ sec : comp_study ] of this work ) . in each case , magnetic models were found to provide satisfactory agreement with the observations . cm dra therefore provides a pivotal test of our models and of the magnetic field hypothesis . magnetic model mass tracks are displayed in figures [ fig : cmdra_mag_d ] and [ fig : cmdra_mag_2 ] . we computed magnetic models using all three radial profiles discussed in section [ sec : rad_prof ] and using both dynamo formulations . perturbations were introduced over a single time step at an age of 1gyr . as with kepler-16 b , the magnetic models adjust to the perturbation well before the age where we attempt to perform the fit of models to observations . all magnetic models have [ fe / h ] @xmath92 and a solar - calibrated @xmath2 . since model radius predictions are only affected at the 1% level due to metallicity variations , inflation required of magnetic fields is the dominating factor when attempting to correct radius deviations of 6% . results for cm dra are similar to those for kepler-16 b. models with a dipole radial profile are unable to inflate model at the level needed , despite strong surface magnetic field strengths being applied ( 5.0 kg 6.0 kg ) . this is evidenced in figure [ fig : cmdra_mag_d](a ) , where magnetic - field - induced inflation occurs at the 1% gaussian radial profile models were able to largely reconcile models with observations . figure [ fig : cmdra_mag_d](b ) shows that surface magnetic field strengths of @xmath93 kg are required to provide the necessary radius inflation . note that in [ fig : cmdra_mag_d](b ) the 6.0 kg mass track for cm dra a actually has a 5.7 kg magnetic field for reasons related to model convergence . one can also see in figure [ fig : cmdra_mag_d](b ) that cm dra b requires a slightly stronger magnetic field strength . as with kepler-16 b , the peak interior magnetic field strengths are 1.8 mg ( @xmath75 ) and 40 mg ( @xmath76 ) for the dipole and gaussian profiles , respectively . constant @xmath23 models are shown in figure [ fig : cmdra_mag_2 ] . we only plot a @xmath77 for each mass . model radius inflation induced by these models is negligible , as with the case for kepler-16 b. the different results produced by the three magnetic field profiles introduced in section [ sec : rad_prof ] can be understood in terms of convective efficiency @xcite . convection near the surface of partially convective stars displays a higher level of super - adiabaticity than does convection in the outer layers of fully convective stars . in general , this suggests that convection is less efficient in the outer layers of partially convective stars . the structure of partially convective stars is therefore more sensitive to changes in convective properties . as a result , structural changes induced by modification to convective properties at the stellar surface induce the necessary radius inflation before the interior magnetic field strength becomes large enough to inhibit convection near the base of the convection zone . dipole and gaussian profiles then produce similar results for partially convective stars . for fully convective stars , the situation is reversed . we display the run of @xmath94 in two models of cm dra a in figure [ fig : cmdra_del_dela ] . one standard dartmouth model and one magnetic model are shown . the magnetic model is a gaussian radial profile model with a @xmath71 kg surface magnetic field . this is the same model that was plotted in figure [ fig : cmdra_mag_d](b ) . fully convective stars are largely characterized by near - adiabatic convection from the outer layers down to the core of the star . changes in convective properties have little effect on the flux transported by convection because convection is extremely efficient . changes to the properties of convection do have some structural effects ( see figure [ fig : cmdra_mag_d](a ) ) , but they are minimal . it s not until the deep interior magnetic field strength becomes strong enough to stabilize interior regions of the star against convection ( @xmath95 ) that significant structural changes occur ( as with the gaussian radial profile ) . when this occurs , a large radiative shell appears in the interior , as shown by the inset in figure [ fig : cmdra_del_dela ] . this radiative shell extends over 54% of the star by radius ( between @xmath96 and @xmath97 ) and 78% by mass ( between @xmath98 and @xmath99 ) . we verified this was the dominant reason for structural changes by looking at the profile of a dipole radial profile model with a similar surface magnetic field strength . the @xmath94 profile exhibited in the surface layers by the dipole model is nearly identical to the gaussian model . however , a radiative shell develops deep in the gaussian model . this supports the idea that fully convective stars require radiative zones to be consistent with observations @xcite . a model with the constant @xmath23 profile has a slightly different @xmath94 profile in the super - adiabatic layer . it also produces a marginally larger surface radiative zone than the standard non - magnetic model . convection in the deep interior is relatively unaffected , so a radiative core does not develop . we opted to not display these features in a figure because the overall profile is almost identical to the non - magnetic profile in figure [ fig : cmdra_del_dela ] . given the insensitivity of the overall stellar structure of fully convective stars to the size of the super - adiabatic layer , the constant @xmath23 models have a negligible influence on the stellar radius . surface magnetic field strengths are estimated from x - ray luminosity , @xmath100 , measurements using the relation between total unsigned magnetic flux , @xmath101 , and @xmath100 derived by @xcite . we are unable , however , to estimate a reliable magnetic field strength for kepler-16b as it does not appear in the rosat catalogs . in a moment we will provide at least a reasonable upper bound , but it is easiest to first address cm dra . the rosat bright source catalogue @xcite indicates that cm dra has an x - ray count rate of @xmath102 with a hardness ratio of @xmath103 . this translates into an x - ray luminosity per star of @xmath104 , where we have used a parallax of @xmath105mas @xcite . note that the x - ray luminosities are upper limits due to possible x - ray contamination in the rosat data . there are several stars nearby to cm dra in the plane of the sky , but it is difficult to judge whether they contribute to the rosat count rate . from the x - ray luminosity derived above , we find @xmath106 . errors associated with the surface magnetic field strength estimates are substantial due to the large error on the surface magnetic flux . converting magnetic fluxes to surface magnetic field strengths , we estimate that @xmath107 kg and @xmath108 kg for cm dra a and b , respectively . note that quoted uncertainties are mean uncertainties . these estimates imply that the @xmath71 kg surface magnetic field strengths predicted by our gaussian radial profile models are likely too strong . however , this does not invalidate the magnetic field models , only our choice of radial profile . the magnetic field strength in the deep interior is of greater consequence , so it may be possible to construct a radial profile to greater reflect this fact . models that use a constant @xmath23 formalism predict surface magnetic fields strengths up to @xmath109 kg . this upper limit is set by the magnetic field coming into equipartition with kinetic energy of convective flows @xcite . values of @xmath110 kg are consistent with our x - ray estimated field strengths . additionally , equipartition magnetic field strengths are consistent with typical average magnetic field strengths measured at the surface of m - dwarfs @xcite . although magnetic field strengths are consistent with observations , our models are unable to produce radii consistent with these realistic field strengths . although there are no x - ray measurements from rosat for kepler-16 , we can attempt to derive a reasonable estimate . from photometry , we may estimate a distance to kepler-16 by assuming that the primary contributes to most of the observed flux in the visible . we estimate a distance of about 60pc using the temperature and luminosity provided by @xcite and @xcite in combination with a bolometric correction from the phoenix model atmospheres . if we make the assumption that all of flux at x - ray wavelengths is from the secondary ( recall the primary shows only weak magnetic activity ; * ? ? ? * ) , then taking the rosat sensitivity limit of @xmath111 counts per second @xcite we find @xmath112 erg s@xmath6 . this also assumes @xmath113 , typical for dwarf stars . converting to an unsigned magnetic flux yields @xmath114 , or a magnetic field strength of order 2 kg for kepler-16b . therefore , kepler-16 likely does not have a magnetic field strength any larger than that found on the surface of cm dra . furthermore , if kepler-16b is rotating pseudo - synchronously , like it s companion , then it would have a rotation velocity @xmath7 km s@xmath6 , as mentioned in section [ sec : intro ] . cm dra , on the other hand , has @xmath115 km s@xmath6 . kepler-16 may very well have a weaker magnetic field , owing to its longer orbital ( and presumably rotational ) period . therefore , we believe that the @xmath71 kg surface magnetic field required to bring our models into agreement with observations is too strong . previous attempts to reconcile model radii with the observed radii of fully convective stars have focused solely on cm dra , as it was the only well - studied system known @xcite . in each case , magnetic fields and magnetic activity were found to provide an adequate solution , which is quite the opposite conclusion from results presented in section [ sec : ind_deb ] . this seeming contradiction of previous results can be understood by our neglect of star spots , in particular , the potential for observed radii to be over - estimated on a spotted star . before submitting that spots are the solution and calling this case closed , we will review existing results , placing ours into context , and the provide an assessment of the magnetic hypothesis . methods used in previous studies were , in some respects similar . each used a method for treating magneto - convection . four techniques have been employed thus far : a reduced-@xmath2 approach @xcite , stabilization of convection by a vertical magnetic field @xcite , stabilization by a more general magnetic field ( not specifically vertical ; this work ) , and then a turbulent dynamo approach that is similar to a reduced-@xmath2 ( this work ) . including effects of star spots is inherently difficult in a 1d stellar evolution code . spots are blemishes scattered across the stellar surface that extend a non - fixed distance into the surface convection zone . spots have largely been treated in the same fashion in previous investigations and rely on reducing the total stellar surface flux by a fractional amount , @xmath116 , \label{eq : spot}\ ] ] where @xmath117 is the surface areal coverage of spots and @xmath118 is the spot temperature contrast . this approach is based on the `` thermal plug '' spot model advanced by @xcite and @xcite . the modified surface flux is then @xmath119 , where @xmath120 is the flux of the star if the photosphere is spot free . @xcite performed a detailed analysis to establish how specific star spot properties effect results from both theoretical modeling and light curve analyses . for a given star spot @xmath121 and assumed distribution of spots over the stellar surface ( uniform , clustered at mid - latitudes , and clustered at the poles ) , they evaluated the reliability of light curve analyses in determining stellar radii . if spots are preferentially located at the poles , they showed that stellar radii may be over - estimated by as much as 6% . on the other hand , if spots are more evenly distributed across the surface , or clustered at mid - latitudes , radius determinations proved reliable for @xmath122 . combining the aforementioned results with the influence of spots on stellar evolution models , via equation ( [ eq : spot ] ) , @xcite found that @xmath123 was required to fit the stars of cm dra . magneto - convection using a reduced-@xmath2 was found to be ineffective and was not required . thus , they predict that the stars in cm dra are 35% covered by spots that are 15% cooler than the surface ( this latter value was fixed in their analysis ) . this was deemed sufficient to correct model radii with observations . @xcite , on the other hand , were able to produce agreement between their model and observations using a combination of magneto - convection and star spots . they identified regions of @xmath124 ( @xmath125 in their paper ) parameter space that reconciles their models with cm dra ( see their figures 13 15 ) . their magnetic inhibition parameter is defined as @xmath126 where @xmath127 is the magnetic field strength , @xmath128 is the ratio of specific heats , and @xmath11 is the total gas pressure . this quantity is added directly to the adiabatic gradient in the schwarzschild stability criterion . what strongly distinguishes the @xcite study from the @xcite investigation is that @xcite do not include model inflation due to spots as described by equation ( [ eq : spot ] ) . instead , they only adopt the adjustment made to the observed stellar radii due to the influence of polar spots on the light curve analysis . model radius inflation is then caused by stabilization of convection using their inhibition parameter described above . the full range of values for which they find agreement is @xmath129 with @xmath130 . to further constrain the parameter space , they assume the best fit star spot @xmath123 from @xcite . this narrows the acceptable range for the magnetic inhibition parameter to @xmath131 . these values of @xmath132 correspond to vertical magnetic field strengths of approximately 500 g at the stellar photosphere and a ( capped ) interior vertical magnetic field strength of 1 mg . our modeling results are largely consistent with those of @xcite and @xcite . comparing first with @xcite , we focus on the dipole and gaussian radial profile models with a rotational dynamo ( figures [ fig : cmdra_mag_d](a ) and ( b ) ) . the magnetic perturbation applied in these models is qualitatively similar to magneto - convection method favored by @xcite . we have estimated that our magnetic field strengths should be up to an order of magnitude larger than those of @xcite as a consequence of our formulation @xcite . our `` inhibition parameter '' is primarily controlled by the quantity , @xmath133 where @xmath134 is a magnetic compression coefficient and @xmath135 is the gradient of the magnetic energy per unit mass with respect to the total pressure . we showed that @xmath136 @xcite , but our formulation has the additional term @xmath137 . therefore , to achieve the same results as @xcite , we should need @xmath134 to be about 10 times larger than @xmath132 . thus , our requirement of a 6.0 kg surface magnetic field strength with a roughly 40 mg peak magnetic field strength for the gaussian radial profile is consistent with the values we would expect given the values from @xcite . although the dipole radial profile model has a 5.0 kg surface magnetic field strength , the peak magnetic field strength is only 1.5 mg , a factor of 10 or so too small to impart the necessary structural changes . to compare with @xcite , we refer to our constant @xmath23 models , which closely matches the reduced-@xmath2 formulation . we found that even if the magnetic field were a considerable fraction of the equipartition magnetic field strength ( @xmath77 ) , there is little impact on the stellar radii . this is consistent with @xcite , who found that the reducing @xmath2 has a negligible impact on the stellar model predictions for fully convective stars , especially after star spots were invoked . in that sense , we confirm that reducing convective efficiency does not appear to be a viable method to fully account for inflation among fully convective stars . note that we have not adopted the reduced stellar radius presented by @xcite nor have we explicitly addressed reductions in surface flux due to star spots . models presented in @xcite did not require star spots to provide an adequate fit to the data . this reinforces a known degeneracy between magneto - convection implementations and introducing star spots @xcite . one may not be too surprised that these two issues are so closely connected , as spots are the physical manifestation of suppressed convection near the stellar surface . therefore , it is reasonable to assume that reduction in convective flux from either magneto - convection or star spots should produce similar results . the degeneracy , in a sense , can be broken by the idea that star spots could bias light curve analyses towards larger radii . magneto - convection , strictly speaking , has no influence on the light curve analysis , if we assume global changes to stellar properties . in that sense , particular configurations of spots ( clustered near the pole ) could have a significantly larger impact on the mass - radius problem than magneto - convection . despite the aforementioned results that support the magnetic field hypothesis , we are skeptical of the interpretation that magnetic fields are driving the observed radius inflation . we will now assess the results primarily by considering available observational data . reductions in convective efficiency appear to be inadequate for producing the observed stellar radii ( section [ sec : ind_deb ] of this work ; * ? ? ? * ; * ? ? ? we showed that , for kepler-16 and cm dra , radius inflation induced by the inhibition of convective efficiency was effectively negligible ( @xmath1380.1% ) . this is in agreement with @xcite and @xcite , who find reducing the convective mixing length from @xmath139 to @xmath140 has little effect on the radii of fully convective stars . reducing @xmath2 further could begin to provide reasonable effects on fully convective stellar radii . the question then becomes , `` is it possible for @xmath141 , and if so , by what mechanism ? '' results from the stabilization of convection are encouraging . a proper amount of radius inflation can be achieved with somewhat reasonable surface magnetic field strengths ( section [ sec : ind_deb ] of this work ; * ? ? ? * ) . instead of surface field strengths posing a problem as they did with partially convective stars @xcite , it is the interior magnetic field strengths , which range from 1 mg to 50 mg , that require attention . there is presently no observational evidence to suggest a 1 mg ( or greater ) magnetic field could exist within a fully convective star . to be fair , there is also no direct evidence to rule out the possibility . we must be honest about the lack of observational data concerning interior magnetic field strengths . however , there is some indirect evidence that casts doubt on the existence of super - mg interior magnetic fields . we will now carefully examine the possibility that such fields do exist deep in fully convective stars . from a theoretical perspective , there is concern about how a super - mg magnetic field is generated . simulations suggest that turbulent dynamos reach saturation at field strengths of @xmath13850 kg @xcite , two orders of magnitude below the 1 mg magnetic field required by stellar models . we will return to this a bit later . for now , let us assume that super - mg magnetic fields can not be generated by dynamo action . instead , it could be assumed that super - mg magnetic fields are the result of the amplification of a primordial @xmath142 g seed magnetic field during the proto - stellar collapse . assuming magnetic flux is conserved during collapse of the proto - star , one finds that super - mg magnetic fields could plausibly exist . however , the super - mg magnetic fields must then survive several gyr within the stellar interior without decaying . there are three primary timescales of interest : that given by macroscopic diffusion in a non - convecting medium , the rise time for a buoyantly unstable flux tube , and the advection timescale in a convecting medium with relatively high conductivity . we ll consider the latter , first . if the stellar plasma is highly conducting , which is a valid assumption throughout the interior of stars , then the field lines will be frozen into the plasma . we might then expect that the flux tubes will be carried by convection from deep within the star to the stellar surface , where the they will then dissipate their energy . the timescale for this to occur is roughly @xmath143 , where @xmath144 cm and @xmath145 cm s@xmath6 . we find that @xmath146 s , or about 1 yr . even if we assume @xmath147 cm s@xmath6 , @xmath148 s. thus , the timescale for this type of advection is about 10 yr . although convection can quickly carry a flux tube from the deep interior to the stellar surface , the process can also carry flux tubes from the surface to deep in the interior . a more uni - directional process that can transport magnetic flux to the stellar surface is the magnetic buoyancy instability ( e.g. , * ? ? ? * ; * ? ? ? magnetic flux tubes are assumed to be in pressure equilibrium with their surroundings deep in the interiors of stars , @xmath149 where @xmath150 and @xmath151 are the interior and exterior gas pressures acting on the flux tube , respectively , and @xmath152 is the magnetic pressure within the flux tube . since flux tubes are supported by both the internal gas pressure and the magnetic pressure , the gas density within a flux tube is lower than the surrounding gas pressure , provided that the temperature inside the flux tube is the same as the temperature of the surroundings . this density perturbation creates a buoyancy force towards the surface of the star , but will be counteracted by radially inward hydrodynamic forces resulting from convective down - flows . if we assume a polytropic eos , a reasonable approximation for deep stellar interiors , and that @xmath153 can be expressed as a power - series expansion about @xmath154 , the gas density exterior to the flux tube , then truncating the series to first order gives @xmath155 + \frac{b^2}{8\pi } = p_{{\rm gas},\,e},\ ] ] where @xmath128 is the ratio of specific heats and @xmath156 is the density perturbation . the specific buoyancy force resulting from this density perturbation is @xmath157 balancing the buoyancy force with the hydrodynamic force @xcite yields the condition whereby a magnetic flux tube will be unstable to buoyant rise , @xmath158 where @xmath127 is the magnetic field , @xmath35 is the local pressure scale height , @xmath159 is the flux tube radius , and @xmath160 is equipartition magnetic field given in section [ sec : clambda ] . this is a simplified picture that ignores effects due to curvature and the response of the magnetic tension that may act to prevent the onset of buoyant instability . however , it provides a reasonable order of magnitude approximation . given the propensity for magnetic stellar models to invoke magnetic fields with @xmath161 g , we can estimate a typical flux tube radius needed to maintain this magnetic field strength stable against buoyancy . for m - dwarfs , @xmath162 cm and @xmath163 g. thus , to maintain strong @xmath164 g magnetic fields deep in the interior ( as in * ? ? ? * ) , flux tubes must be no larger than @xmath165 cm , or 1 km . this means the ratio of the pressure scale height to the flux tube radius must be @xmath166 . by comparison , the values estimated for this ratio near the solar tachocline are of order unity @xcite . this ratio rises to @xmath164 ( @xmath167 m ) if the deep interior magnetic field is to be of order @xmath168 g ( this work ) . once the flux tube becomes unstable to buoyant rise , an estimate of the rise time can be approximated assuming that the tube travels at the alfvn velocity @xcite . a @xmath164 g magnetic field will traverse a stellar radius @xmath169 cm in approximately @xmath170 s , or about 10 days , assuming a constant density of @xmath171 g @xmath172 . this rise time will be increased by non - adiabatic heating effects , but remains small compared to evolutionary timescales @xcite . figure [ fig : cmdra_del_dela ] indicates that a radiative shell develops deep within the star in the presence of a super - mg magnetic field . this means that the dissipation time of the magnetic field does not necessarily obey the advection timescale discussed above . instead , we may look at the timescale for diffusion given by the induction equation . in the absence of any favorable current networks , we have that @xmath173 where @xmath174 is the magnetic diffusivity . the approximate timescale for diffusion , assuming @xmath175 @xmath176 s@xmath6 @xcite , is then @xmath177 , where @xmath178 is the size of the radiative shell . the field diffuses through the radiative shell until it reaches the convective boundary , which then efficiently transports magnetic flux to the stellar surface . based on our models of cm dra , we find the size of radiative shell is @xmath179 cm . therefore , @xmath180 yr , assuming the magnetic field fully diffuses out of the radiative zone . however , even within a radiation zone , magnetic buoyancy must be considered and can lead to a rapid rise time for magnetic flux tubes ( e.g. , * ? ? ? these are only order of magnitude approximations , so it is possible that the field could survive for a shorter or longer time in any scenario . instead , this exercise suggests that an amplified seed field , or any super - mg magnetic field , present in a fully convective star will likely decay away rapidly owing to the magnetic buoyancy instability . we note that this is still possible , even if a sizeable radiative zone develops as a result of the strong magnetic field . furthermore , the radiative zone is a product of the magnetic field and therefore exists in an unstable equilibrium . as the magnetic field decays away , so will the radiative zone . @xcite recognized that primordial magnetic fields would likely decay and thus were not a suitable solution for the existence of super - mg magnetic fields within cm dra . instead , they proposed that the dynamo mechanism may be able to generate the necessary magnetic fields . their proposal contradicts what we mentioned earlier , that simulations are unable to generate magnetic fields greater than @xmath13850 kg in fully convective models . @xcite reason that this is a result of those simulations using rotational velocities similar to those in the sun . if one nominally assumes that magnetic field strengths increase with rotation rate , then a star like cm dra will have considerably stronger magnetic fields than does the sun . using a scaling relation between magnetic field strength and rotational angular velocity , @xcite find that cm dra should have a surface magnetic field strength of about 500 g with an internal magnetic field strength anywhere between 0.3 mg and 1.3 mg . this could then explain the existence of super - mg magnetic field strengths . note that the magnetic field strength is predicted to be the large - scale magnetic field strength , not necessarily the total magnetic field strength . however , the peak interior magnetic field strength seen in simulations is moderated by the kinetic energy available in convective flows . this was discussed in section [ sec : clambda ] . simulations find that a dynamo driven by turbulent convection can create both small - scale magnetic fields and large - scale magnetic fields @xcite . it is not immediately clear that rotation would not drive stronger magnetic field strengths , so we will proceed in our discussion assuming that it might to see if we encounter any inherent contradictions . the scaling relation adopted by @xcite would then also explain why koi-126 can be fit by stellar evolution models . koi-126 is a triply eclipsing system found earlier in the _ kepler _ data @xcite that contains two fully convective stars that are nearly identical to cm dra . yet , the properties of koi-126 ( b , c ) are well reproduced by standard stellar evolution models @xcite . assuming a scaling relation between magnetic field strength and rotational angular velocity may support this finding . koi-126 ( b , c ) have an orbital period of 1.77days , to be compared to the 1.27day orbit of cm dra . this 0.5day difference in orbital period means koi-126 ( b , c ) would have weaker magnetic fields than cm dra and would experience considerably less radius inflation @xcite . @xcite predict that koi-126 should be inflated by approximately 2 3% . we find this difficult to understand in context of the agreement between standard stellar evolution models and the observed radii . inflating model radii by 2 3% would mean that models would over - predict the radii of koi-126 ( b , c ) . if one assumes that spots are biasing the observed radius measurements , it would decrease the measured radii , again breaking the agreement between the observations and the models . there seems to be problems with simultaneously increasing model radii and decreasing the measured radii of koi-126 ( b , c ) while still maintaining model - observation agreement . furthermore , the scaling relation suggested to validate super - mg magnetic fields may not be compatible with observations . results seem to show that at a given mass , the large - scale magnetic field strength of fully convective stars do not appear to depend on rotation below a critical threshold ( see , e.g. , figure 3 in * ? ? ? * ) . for low - mass stars , this threshold corresponds to a rotation period of about 3 or 4 days . magnetic field saturation is also observed in data regarding the total magnetic field strength of fully convective stars . magnetic fields saturate around 3 kg for @xmath181kms@xmath6 @xcite . saturation inherently implies that scaling relations are invalid . since both cm dra and koi-126 ( b , c ) have components rotating faster than the threshold for saturation , there is no reason to assume that their magnetic fields are that different . there is the possibility that their large - scale components ( and thus vertical ) magnetic fields are sufficiently different , but this is not immediately obvious . finally , we note that magneto - convection techniques alone have not been sufficient to rectify the model - observations disagreements . effects of star spots must be included and must therefore be assessed on their own basis . the most efficient means of inflating low - mass stellar models of cm dra is by including star spots @xcite . although we did not explicitly include spots , we will take a careful look at the results of these two previous studies . we feel this analysis is required in light of recent observational data , including confirmation that cm dra has a sub - solar metallicity of [ fe / h ] @xmath182 @xcite . we use approximate fits to the @xcite spot - modeling results to estimate how a metallicity reduction alters their conclusions . assuming polar spots , they find that spots tend to bias observations towards larger radii . from the few data points in their figure 7 , we calculate that spots produce larger radii according to the function @xmath183 where @xmath184 is the ratio of the radius derived when polar spots are present to the actual radius . this function can be used to correct for the radius bias produced by spots . in stellar models , spots can also increase radius predictions . given the data in figure 8 of @xcite , model radii can be approximated as , @xmath185 where @xmath186 is the radius of a non - magnetic model in units of solar radii . we note that this formula approximates the effects of spots only for model masses of @xmath187 . the intersection of these two relations in the @xmath121@xmath18 plane provides an estimate of the star spot parameter required to correct models . figure [ fig : spot_corr ] illustrates that a model of cm dra a with [ fe / h ] @xmath188 ( see also figure [ fig : cmdra_nmag ] ) needs a star spot @xmath189 to match observations . small differences in our results from those of @xcite are due to the adopted base model ( dartmouth vs lyon ) and our method of fitting polynomials to the spot radius bias and inflation data in @xcite . changing @xmath121 has proportionally larger effect on correction for radius measurement biases than it does on inflation caused in stellar models . if this correction is not applied , @xmath190 is needed to produce agreement using model radius inflation at sub - solar metallicity . whether the required star spot parameter is realistic is a more difficult question . spot properties for m - dwarfs are relatively unconstrained , as identifying spot sensitive features in spectra is complicated by uncertainties in modeling molecular features . nonetheless , there are three components to the spot parameter that are of interest : spot coverage , spot temperature contrast , and spatial distribution . we will address each separately and then present a unified picture afterward . [ [ temperature - contrast ] ] temperature contrast + + + + + + + + + + + + + + + + + + + + spot temperature contrast is defined , here , to be the ratio of the spot temperature to the temperature of the unblemished stellar photosphere . as seen from equation ( [ eq : spot ] ) , constraining the spot temperature contrast provides insight into the required filling factor predicted by the spot model . although no direct empirical constraints exist for m - dwarfs , there have been studies performed to measure spot temperature contrasts of k stars using tio bandhead features @xcite . we will take a closer look at those results , here . properties of spots were determined by fitting the shape of tio bandheads with a spectrum created using a two temperature spot model . this technique involves convolving two template spectra for which the effective temperature is assumed to be known and finding the best combination of template spectra to produce the molecular bandhead features in spectra of a given star ( see e.g. , * ? ? ? the result is an estimate of the temperatures for the unblemished photosphere and the spotted regions , as well as the fractional surface coverage of each feature ( see below ) . using this technique , it was found that active k stars specifically ii peg and eq vir have spots with a characteristic temperature contrast of 75% ( 25% cooler than the unblemished photosphere ) . this estimate , however , relies on the temperature for the two template spectra being correct . results from interferometric studies ( e.g. , * ? ? ? * ) have revised the temperature estimates for the stars whose spectra were used as templates in the @xcite investigations . these revisions only affect the temperatures for template spectra used as the spot ( cool ) component in the fit , namely the temperatures associated with cooler m - dwarf spectra , and increase the associated effective temperature by 300 k 400 k. taking into account the revised temperature scale for the cooler template spectra , one finds the temperature contrast for k stars inferred from molecular bandhead features is between 82% ( ii peg ) and 90% ( ek vir ) . the warmer spot temperatures are supported by doppler imaging ( di ) studies of ii peg , which indicate temperature contrasts between 80% and 85% @xcite . assuming that spot temperature contrasts become weaker towards later spectra types @xcite , then one might expect m - dwarfs to have spot temperature contrasts around 90% ( 10% cooler than the unspotted surface ) . using this value , one finds that a spot surface coverage of 55% is required to satisfy the condition that @xmath191 for cm dra . of course , slightly warmer or cooler spots may be permitted , depending on the efficiency at which spots temporarily suppress surface convection . for the sake of argument , we take 90% to be a typical spot contrast . we now look whether a spot areal coverage of 55% is consistent with empirical evidence . [ [ surface - coverage ] ] surface coverage + + + + + + + + + + + + + + + + to assess what constitutes typical spot surface coverages , we look to results from di and molecular bandhead fitting . using di , spot coverages for three m - dwarfs have been estimated . di maps of , , and @xcite reveal varying levels of surface coverage from 2% ( v374 peg ) up to 40% ( hk aqr ) . similarly , surface coverages derived by modeling the shape of tio bandheads of active k stars again ii peg and ek vir are typically around 30 40% , consistent with values found in di investigations @xcite . there are no measurements of spot coverages on m - dwarfs using this technique , which limits the applicability of these results , but we may cautiously extrapolate and assume that m - dwarfs are able to possess at least similar surface coverages . for one of the k stars studied using the molecular bandhead technique , , information on spot coverage from di is also available . @xcite mapped the spot coverage of ii peg and find seasonal variation in spot coverage between 5 20% , a factor of 8 and 2 lower than the coverage inferred from tio bandheads , respectively . seasonal variation is expected if magnetic activity cycles are present in other stars , so it s not surprising that one observes this phenomenon . however , the difference between the maximum surface coverage revealed by di ( 20% ) and that obtained using molecular bandhead features ( 40% ) highlights a known limitation with di studies : they are only sensitive to spot features larger than the grid resolution in the di analysis . in other words , small scale spot features will be missed by the di reconstruction . this is especially relevant since comparison of spectro - polarization measurements of m - dwarf magnetic fields using stokes @xmath192 and @xmath193 polarization indicate a majority of the magnetic energy is contained in small - scale features @xcite . therefore , spot coverages inferred from di maps must be taken as lower limits to the true spot coverage . one may then plausibly expect spot surface coverages at least as high as 40% for m - dwarfs . evidence from both di and molecular bandhead fitting support the existence of spot coverages around 40% for active late k - dwarfs and m - dwarfs . this is below the 55% needed to satisfy the use of a @xmath191 spot parameter . whether 55% is _ unrealistically _ high is unclear . if , for instance , the spot temperature contrast were around 85% , as opposed to 90% , then the required spot surface coverage would be 40% , and thus consistent with observations . one could just as well argue the opposite , that a larger temperature contrast of 94% would require a surface coverage of 87% and push the requirement further from the current empirical evidence . one further scenario is that stars are able to achieve spot coverages near 100% . observations suggest that the filling factor of magnetic fields at the surface of fully convective stars ( both main - sequence and pre - main - sequence ) approaches unity @xcite . we must then try to understand the connection between the presence of magnetic fields and star spots . this includes determining how strong of a magnetic field is necessary at the surface of an m - dwarf to generate spots of appreciable temperature contrast . it may be that pervasive 1 kg magnetic fields covering the whole surface of fully convective stars may be too weak to cause any noticeable effects on convection . such an even distribution of small spots would also likely not cause any noticeable light curve modulation , if filling factors approach unity . however , pockets of strong 5 8 kg magnetic fields @xcite may produce the noticeable light curve modulation , bias stellar radii measurements , and suppress surface convection . [ [ spatial - distribution ] ] spatial distribution + + + + + + + + + + + + + + + + + + + + arguably , whether spots cluster near the poles of active m - dwarfs is of greatest consequence for the present assessment . evidence for spots clustered at the poles is less definitive than temperature contrasts and surface coverages . about half of all stars studied using di show signs of polar caps @xcite . exceptions are very active stars and m - dwarfs , which tend to have spots distributed across all latitudes . of the three di spot maps generated for m - dwarfs , none display evidence for polar caps @xcite . a di study of cm dra to search for possible signs of significant polar coverage of spots is needed . ideally , this would be performed with existing data that was used to derive the stellar properties , but a look at the signal - to - noise of the observations reveals that the data is likely unsuitable for such an investigation ( see * ? ? ? in the meantime , we must be cautious as even polar spots would require large areal coverages ( half or more of the stellar surface ) to provide agreement between models and observations . additional evidence for non - polar - cap distributions comes from studies of the @xmath138150 myr open cluster . the statistical distribution of light curve modulations assumed to be caused by star spots in the low - mass star population can only be reproduced if the stars have randomized spot latitudinal distributions and rotational inclination angles @xcite . we do note that these latter studies also require spot temperature contrasts of 75% , which we have just discussed may be too dramatic for k- and m - dwarfs . how using a more realistic spot contrast would alter their results is unclear as there are strong degeneracies when modeling spot modulation in light curves . a first estimate would be that typical surface coverages would increase from the @xmath194% that was required in their study . there is as yet no definitive empirical evidence for polar cap spots among fully convective main sequence stars . however , there is a theoretical expectation that spots should be located near the poles in rapidly rotating stars ( e.g. , * ? ? ? * ; * ? ? ? * ) . whether this expectation applies to fully convective stars is questionable , as spots distributed more evenly at lower latitudes may be more reasonable for more distributed dynamos @xcite . based on only three data points , the apparent lack of polar spots among fully convective main sequence stars is not robust and requires further confirmation . is it then possible to abandon the idea of a largely polar cap distribution and still rely on spots to address the radius discrepancies ? in figure [ fig : spot_corr ] , we see that abandoning the idea of polar spots requires @xmath190 if model inflation is to alone correct for the observed radii . in this case , a spot temperature contrast of 89% with areal coverage of 100% would be required . temperature contrasts greater than 89% lead to areal coverages greater than 100% and are thus unphysical . even as one approaches 100% surface coverage ( for temperature contrasts below 89% ) , it becomes increasingly difficult to produce spot modulation in light curves as there are fewer regions of unblemished photosphere left to produce the asymmetries needed , as discussed in the previous section . considering patches of varying spot temperature contrast may provide the necessary asymmetries , but has not yet been investigated for consistency with the empirical data . the case for magnetic fields inflating the radii of fully convective stars appears to be tenuous . stabilization of convection in stellar models can produce the necessary radius inflation with reasonable surface magnetic field strengths . but , the models require super - mg magnetic fields in the interior . we are hesitant to suggest that real stars contain such strong interior magnetic fields for several reasons . first , turbulent dynamos can not generate magnetic fields with strengths greater than about 50 kg @xcite . how magnetic fields with strengths greater than 1 mg would accumulate in the interior of fully convective stars would be unknown . if super - mg magnetic fields are of primordial origin , they would probably not survive to the present day . nor would magnetic field strengths of that magnitude be generated within the star and survive on evolutionary timescales . to avoid this , one can assume a scaling relation between magnetic field strengths and rotation , as suggested by @xcite to explain the super - mg magnetic fields . however , this breaks down for most stars in debs . observations indicate a saturation of magnetic flux for fully convective stars that have rotation periods below @xmath195days @xcite . also , koi-126 ( b , c ) agree with stellar evolution models @xcite . these stars exist in a regime where magnetic flux saturation should occur , but are seemingly uninfluenced by the presence of any magnetic phenomena . if super - mg magnetic fields are causing dramatic inflation in cm dra , one would expect to observe it in koi-126 ( b , c ) as well . finally , suppression of convective efficiency is unable to provide a suitable solution on its own . instead , star spots may be invoked for fully convective stars , which introduces additional concerns . the presence of star spots on the stellar surfaces may provide an adequate solution , mostly through biasing the analysis of light curve data . from observations , star spots on fully convective stars can be characterized as : having temperatures about 10% cooler than the surrounding photosphere @xcite , having surface coverages ranging from a couple percent @xcite up to around 40% @xcite , and having spot distributions that appear to be more random and not clustered at the poles . at this point , there is not sufficient empirical data regarding spot properties of m - dwarfs to draw firm conclusions . still , the lack of noticeable spots at the poles is a concern . results regarding how spots bias radius measurements rely on concentrated regions of spot coverage at the poles that are darker than their surroundings . at present , this feature is not observationally supported , but more work needs to be performed on this matter . this paper addressed the question of whether magnetic fields can realistically inflate the radii of fully convective stars . we approached the problem by modeling two individual debs that contain fully convective stars , kepler-16 and cm dra . our models showed that the observed radii could be reproduced provided that a strong 50 mg magnetic field resides deep within these fully convective stars . we proceeded to show that multiple mechanisms advection due to convective flows , macroscopic diffusion , and magnetic buoyancy instability can lead to rapid destruction of such strong magnetic fields . these estimates would benefit from detailed numerical simulations to probe further whether such magnetic fields can stably reside deep within m - dwarfs . an assessment of the hypothesis that star spots are responsible for the observed inflation was also given . here , the hypothesis largely rests on the assumption that large conglomerations of spots lie near the poles ; a prediction that has not yet been observed among main sequence fully convective stars . after weighing the evidence , we find it difficult to favor the hypothesis that magnetic fields are actively inflating the radii of fully convective , main sequence stars . there are numerous uncertainties that must be addressed on both the theoretical and observational side of the hypothesis . more sophisticated magneto - convection models are required to further investigate this issue , using both simplified 1d approximations and more detailed 3d mhd models . most critical , though , will be observational studies of star spots and magnetic fields of low - mass stars . these properties must be further constrained by observations if we are to rule out any of the physical pictures presented by the current generation of 1d stellar evolution models . in the meantime , it may not be too early to start seeking other solutions . we thank bengt gustafsson for carefully reading and commenting on the manuscript , and the anonymous referee for their constructive remarks and suggestions . this research was made possible by the william h. neukom 1964 institute for computational science at dartmouth college and the national science foundation ( nsf ) grant ast-0908345 . this work made use of nasa s astrophysics data system , the simbad database , operated at cds , strasbourg , france , and the _ rosat _ data archive tools hosted by the high energy astrophysics science archive research center ( heasarc ) at nasa s goddard space flight center .

wave scattering in surface - corrugated guiding systems remains a quite interesting problem , both from the theoretical viewpoint and due to many applications ranging from the sound and electromagnetic propagation , to the conduction of electrons in mesoscopic wires . in spite of many important results obtained to date ( see , e.g. , refs . @xcite ) , there is a number of open questions to be resolved . one of such problems is to understand the interplay between different mechanisms of the scattering that are due to principally different terms in an effective scattering potential constructed with the use of various perturbative approaches . in the recent papers @xcite it was shown that apart from the well known terms in the potentials appearing in the first order perturbation theory , one should keep some specific terms that were neglected due to their seeming smallness . it was demonstrated that these terms , although formally belonging to the second order terms , can play a decisive role in the scattering in some region of parameters . since these terms are proportional to the square of the derivative of surface profiles , they can be associated with the _ square - gradient scattering _ ( sgs ) mechanism . in contrast with ref.@xcite where a particular case of one rough surface was considered , below we develop a general approach by concentrating our attention on quasi - one dimensional waveguides with two corrugated surfaces that are either statistically uncorrelated or correlated . our main interest is in the expression for the _ attenuation length _ @xmath0 ( also known as the scattering length , or the total mean free path ) , corresponding to specific @xmath1-th propagating mode . we assume that the total number @xmath2 of these modes ( or conducting channels ) is large , @xmath3 , and that the lower and upper surface profiles ( boundaries ) of such quasi-1d waveguides are described by two random functions @xmath4 and @xmath5 , respectively . in what follows , we consider three configurations of a particular interest : 1 . the waveguide with the _ uncorrelated _ boundaries , @xmath4 and @xmath5 . 2 . the waveguide with the _ antisymmetric _ boundaries , @xmath6 3 . the waveguide with the _ symmetric _ boundaries , @xmath7 in our approach we assume that due to the multiple scattering of a travelling wave from the rough boundaries , the longitudinal wave number of an @xmath1-th propagating mode can be written as @xmath8 , where @xmath9 is its unperturbed value , @xmath10 the real part @xmath11 is responsible for a roughness - induced correction to the phase velocity of a given mode . as is known , the shift @xmath11 does not change the transport properties of a disordered system . therefore , our further analysis shall be focused on the attenuation length @xmath0 only . in order to properly take into account the surface irregularities we use the approach developed in ref . specifically , by means of the canonical transformation we formally reduce the problem of the _ surface _ scattering to the _ bulk _ one described by the hamiltonian @xmath12 , @xmath13 with the effective potential @xmath14 . it should be stressed that such a transformation does not mean an equivalence between the surface and bulk scattering . the very point is that the obtained `` bulk '' hamiltonian @xmath12 is of very specific form that does not allow to use standard random matrix approaches based on a complete randomness in effective potentials . the paper is organized as follows . in sec . [ sec - problem ] we start with the general expressions describing the waveguide with any form of rough boundaries . in sec . [ sec - dysoneq ] we write the dyson - type equation and derive approximate expressions for the surface scattering potential in the form , most useful for our perturbative approach . in sec . [ sec - xi - tilde ] we obtain the fluctuating part of the potential for three configurations of a waveguide . in sec . [ sec - averagegf ] we present a brief description of the formalism for obtaining the average green s function developed in ref . @xcite . here we also specify the conditions of applicability of our method . in sec . [ sec - ln - an ] we perform a detailed analysis of the attenuation length for three specific configurations of the waveguide . in sec . [ sec - num - analysis ] we report the data that illustrates our analytical results and make clear the role of various terms in the expression for the scattering potential . in this section we also discuss the so - called `` repulsion '' phenomenon discovered in ref . , in view of its relation to the square - gradient scattering mechanism . a brief comparison of the three configurations , as well as our concluding remarks , are presented in sec . [ sec - conclusion ] . in this section we derive the general expressions describing quasi-1d waveguides with any type of the boundaries . then , we present them in the form , most suitable for our analysis of three particular waveguide configurations . specifically , we consider the standard model of an open waveguide of the average width @xmath15 , stretched along the @xmath16 axis . the lower and upper surfaces of the waveguide are assumed to be described , respectively , by the profiles @xmath17 , and @xmath18 . here @xmath19 is the root - mean - square roughness height that is assumed to be identical for both boundaries . in other words , the waveguide occupies the region @xmath20 of the @xmath21-plane . the fluctuating width @xmath22 is defined by @xmath23 , \qquad \langle w(x)\rangle = d\,,\ ] ] where the random functions @xmath24 ( with @xmath25 ) describe the roughness of the corresponding boundary . in what follows we consider the model for which both of the boundaries are statistically identical . they are assumed to be described by a statistically homogeneous and isotropic gaussian random processes with the zero mean and unit variance , [ gen - xi ] @xmath26 here the angular brackets stand for the statistical averaging over different realizations of the surface profiles @xmath27 . we also assume that the binary correlator @xmath28 decreases on the scale @xmath29 , with the normalization , @xmath30 . for three configurations we analyze below , the cross - correlations are as follows . in the first case of the uncorrelated boundaries we obviously have , @xmath31 the waveguide with the antisymmetric boundaries implies : @xmath32 as for the symmetric boundaries case , the opposite boundaries hold the correlation equality : @xmath33 the _ roughness - height power _ ( rhp ) spectrum @xmath34 is defined by : @xmath35 since @xmath28 is an even function of @xmath16 , its fourier transform ( [ ft - w ] ) is an even , real and non - negative function of @xmath36 . the rhp spectrum has the maximum at @xmath37 with @xmath38 , and decreases on the scale @xmath39 . in order to analyze the surface scattering problem we shall employ the method of retarded green s function @xmath40 . specifically , we start with the dirichlet boundary - value problem [ gen - gfp1 ] @xmath41 & & { \cal g}(x , x';z=\sigma \xi_{\downarrow}(x),z')=0 \nonumber \\ & & { \cal g}(x , x';z = d+\sigma \xi_{\uparrow}(x),z')=0 . \label{gen - bc1}\end{aligned}\ ] ] here @xmath42 and @xmath43 are the dirac delta functions . the wave number @xmath44 is equal to @xmath45 for an electromagnetic wave of the frequency @xmath46 and te polarization , propagating through a waveguide with perfectly conducting walls . for an electron quantum wire , @xmath44 is the fermi wave number within the isotropic fermi - liquid model . the equation ( [ gen - geq1 ] ) does not contain any scattering potential , therefore , the perturbation is hidden in the boundary conditions ( [ gen - bc1 ] ) . in this case it is convenient to perform the canonical transformation to new coordinates , [ gen - coortr ] @xmath47 z_{new}&=&\frac{d\ , [ z_{old}-\sigma\xi_{\downarrow}(x_{old } ) ] } { w(x_{old})}\notag \\ & = & \frac{d\ , [ z_{old}-\sigma\xi_{\downarrow}(x_{old } ) ] } { d+\sigma[\xi_{\uparrow}(x_{old})-\xi_{\downarrow}(x_{old})]},\end{aligned}\ ] ] in which both boundaries are flat . in accordance with eqs . ( [ gen - coortr ] ) , the transformations for the first and second derivatives with respect to the old coordinates are given by ( below for convenience we drop the subscript @xmath48 for @xmath16 and @xmath49 ) : [ gen - d1tr ] @xmath50 \frac{\partial}{\partial z}\ , , \nonumber\\[6pt ] \\ \frac{\partial}{\partial z_{old}}&=&\frac{\partial x}{\partial z_{old}}\frac{\partial}{\partial x}+\frac{\partial z}{\partial z_{old}}\frac{\partial}{\partial z } \nonumber \\ & = & \frac{d}{w(x)}\frac{\partial}{\partial z}\ , , \end{aligned}\ ] ] [ gen - d2tr ] @xmath51 \frac{\partial}{\partial z } \right\ } \nonumber \\ & \times & \left\ { \frac{\partial}{\partial x}- \left[\frac{\sigma\xi'_{\uparrow}(x)}{w(x)}z + \frac{\sigma\xi'_{\downarrow}(x)}{w(x)}(d - z)\right ] \frac{\partial}{\partial z}\right\ } \nonumber \\[6pt ] \\ \label{gen - d2ztr } \frac{\partial^2}{\partial z_{old}^2}&=&\frac{d^2}{w^2(x ) } \frac{\partial^2}{\partial z^2}\,.\end{aligned}\ ] ] correspondingly , we introduce the canonically conjugate green s function , @xmath52 here the prefactor before @xmath53 is due to the jacobian of the transformation to new variables . as a result , we arrive at the equivalent boundary - value problem governed by the equation , [ gen - gfp4 ] @xmath54 & & -\widehat{u}(x , z)\,{\cal g}(x , x';z , z ' ) = \delta(x - x')\delta(z - z'),\qquad\label{gen - geq4}\\[6pt ] & & { \cal g}(x , x';z=0,z')=0,\nonumber\\ & & { \cal g}(x , x';z = d , z')=0 . \label{gen - bc4}\end{aligned}\ ] ] here we also omitted the subscript @xmath48 for @xmath55 . as a result of the transformation , we have formally reduced the problem of surface scattering to that of the bulk one specified by the effective potential @xmath56 , @xmath57 \frac{\partial^2}{\partial z^2 } \nonumber \cr \\ & & + \frac{\sigma}{w(x ) } \bigg\{\bigg[\xi_{\uparrow}''(x ) + 2\xi_{\uparrow}'(x)\frac{\partial}{\partial x}\bigg ] \bigg[\frac{1}{2}+z\frac{\partial}{\partial z}\bigg ] \nonumber \cr \\ & & -\bigg[\xi_{\downarrow}''(x)+ 2\xi_{\downarrow}'(x)\frac{\partial}{\partial x}\bigg ] \bigg[\frac{1}{2}+(z - d ) \frac{\partial}{\partial z}\bigg ] \bigg\}\nonumber \cr \\ & & -\frac{\sigma^2}{w^2(x)}\bigg\{{\xi_{\uparrow}'}^2(x ) \bigg[\frac{3}{4}+3z\frac{\partial}{\partial z } + z^2\frac{\partial^2}{{\partial z}^2}\bigg]\nonumber \cr \\ & & + { \xi_{\downarrow}'}^2(x ) \bigg[\frac{3}{4}+3(z - d)\frac{\partial}{\partial z } + ( z - d)^2\frac{\partial^2}{{\partial z}^2}\bigg ] \nonumber \cr \\ & & -\xi_{\downarrow}'(x)\,\xi_{\uparrow}'(x)\bigg[\frac{3}{2}+3(2z - d ) \frac{\partial}{\partial z}+2z(z - d)\frac{\partial^2}{{\partial z}^2 } \bigg ] \bigg\}.\nonumber \cr \\\end{aligned}\ ] ] note that the prime over the function @xmath27 denotes a derivative with respect to @xmath16 . it is important to stress that eqs . ( [ gen - gfp4 ] ) and ( [ gen - u ] ) are _ exact _ ones , and valid for any form of the surface profiles @xmath4 and @xmath5 . as is expected , the general expression ( [ gen - u ] ) reduces to that obtained previously in ref.@xcite for a particular case of one rough boundary , if one substitutes @xmath58 and @xmath59 . here we start with the non - local dyson equation in @xmath36-representation , @xmath60 with @xmath61 being the unperturbed pole - factor . in this representation the effective surface scattering potential @xmath62 is given by the expression , @xmath63 here @xmath36 and @xmath64 are the lengthwise and transverse wave numbers , @xmath65 their unperturbed eigenvalues are @xmath9 , @xmath66 and @xmath67 , respectively . the total number @xmath68 of propagating waveguide modes ( conducting channels that have real values of @xmath9 ) is determined by the integer part @xmath69 $ ] of the ratio @xmath70 , @xmath71.\ ] ] after the substitution of the expression ( [ gen - u ] ) for @xmath72 in eq . , we realize that @xmath62 consists of three groups of terms of a different nature . the first one has the factor @xmath73 $ ] , the second and third groups have the factors @xmath74 and @xmath75 , respectively . one needs to note that while itself the kernel @xmath62 is hermitian , the latter two parts are non - hermitian ones . in order to present each part of @xmath62 in an hermitian form , we perform the integration by parts for the term containing @xmath76 and @xmath77 in the second group . as a result , we arrive at the final _ exact _ form for the perturbation potential , @xmath78 \frac{\partial^2}{\partial z_1 ^ 2 } + \frac{\sigma}{w(x_1)}\ , [ i(k_x+k_x ' ) ] \nonumber \cr \\ & & \times \bigg\{\xi_{\uparrow}'(x_1)\bigg[\frac{1}{2 } + z_1\frac{\partial}{\partial z_1}\bigg ] -\xi_{\downarrow}'(x_1)\bigg[\frac{1}{2}+(z_1-d ) \frac{\partial}{\partial z_1}\bigg]\bigg\}\nonumber \cr \\ & & - \frac{\sigma^2}{w^2(x_1)}\bigg\{{\xi_{\uparrow}'}^2(x_1 ) \bigg[\frac{1}{4}+2z_1\frac{\partial}{\partial z_1 } + z_1 ^ 2\frac{\partial^2}{{\partial z_1}^2}\bigg]\nonumber \cr \\ & & + { \xi_{\downarrow}'}^2(x_1 ) \bigg[\frac{1}{4}+2(z_1-d)\frac{\partial}{\partial z_1 } + ( z_1-d)^2\frac{\partial^2}{{\partial z_1}^2}\bigg ] \nonumber\cr\\ & & - \xi_{\downarrow}'(x_1)\,\xi_{\uparrow}'(x_1)\bigg[\frac{1}{2 } + 2(2z_1-d)\frac{\partial}{\partial z_1 } \nonumber\cr\\ & & + 2z_1(z_1-d)\frac{\partial^2}{{\partial z_1}^2 } \bigg ] \bigg\ } \bigg ) \exp\left(ik'_xx_1\right)\frac{\sin(k'_zz_1)}{k'_z } \label{gen - xi - exactf}\end{aligned}\ ] ] this equation has a peculiar structure , very useful for a further analysis . the kernel written in this form consists also of three groups of terms , however , now they can be associated with different scattering mechanisms . since we are interested in the averaged green s function , we have to calculate the binary correlator of @xmath62 . therefore , in order to avoid very cumbersome calculations , it is reasonable to make a simplification of @xmath62 that does not destroy the hermitian structure of each group of terms . taking also into account that our main interest is in the case of small surface corrugations @xmath79 , we can do the following : expand the factor @xmath80/d$ ] in the first term of eq . ( [ gen - xi - exactf ] ) and to put @xmath81 in all the others . in such a way , we derive the following _ approximate _ expression for the surface scattering potential , @xmath82\frac{\partial^2}{\partial z_1 ^ 2 } + \frac{\sigma}{d}\ , [ i(k_x+k_x ' ) ] \nonumber \cr \\ & & \times \bigg\{\xi_{\uparrow}'(x_1)\bigg[\frac{1}{2 } + z_1\frac{\partial}{\partial z_1}\bigg ] -\xi_{\downarrow}'(x_1)\bigg[\frac{1}{2}+(z_1-d ) \frac{\partial}{\partial z_1}\bigg]\bigg\}\nonumber \cr \\ & & - \frac{\sigma^2}{d^2}\bigg\{{\xi_{\uparrow}'}^2(x_1 ) \bigg[\frac{1}{4}+2z_1\frac{\partial}{\partial z_1 } + z_1 ^ 2\frac{\partial^2}{{\partial z_1}^2}\bigg]\nonumber \cr \\ & & + { \xi_{\downarrow}'}^2(x_1 ) \bigg[\frac{1}{4}+2(z_1-d)\frac{\partial}{\partial z_1 } + ( z_1-d)^2\frac{\partial^2}{{\partial z_1}^2}\bigg ] \nonumber\cr\\ & & - \xi_{\downarrow}'(x_1)\,\xi_{\uparrow}'(x_1)\bigg[\frac{1}{2 } + 2(2z_1-d)\frac{\partial}{\partial z_1 } \nonumber\cr\\ & & + 2z_1(z_1-d)\frac{\partial^2}{{\partial z_1}^2 } \bigg ] \bigg\ } \bigg ) \exp\left(ik'_xx_1\right)\frac{\sin(k'_zz_1)}{k'_z}\end{aligned}\ ] ] the expression ( [ gen - xi - approx ] ) contains three types of scattering terms of a different physical meaning : the terms depending on the roughness height @xmath83 , derivative of the roughness height @xmath84 , and square of the derivative of the roughness height @xmath85 . in this connection one can speak about the _ amplitude _ , _ gradient _ and _ square - gradient scattering _ ( sgs ) mechanisms . note that these terms are separately associated with each of two surfaces . apart from this , there is a specific group of terms that are due to the inter - correlations between two surfaces . in what follows we refer to these terms depending on the product @xmath86 , as to the _ gradient inter - product scattering _ ( gis ) mechanism . it should be stressed that the gis - terms , due to their proportionality to @xmath87 , were neglected in previous studies of the transport properties of surface - disordered waveguides . a similar situation occurs for the sgs - terms that have been analyzed only recently in ref . as we show below , all these terms can give the main contribution in the scattering , although formally they may be treated as negligible ones . we start our analysis with the presentation of the kernel as a sum of its average and fluctuating parts , i.e. , @xmath88 it can be tested that the average , @xmath89 , contributes only to the real part @xmath11 of the complex renormalization @xmath90 of the lengthwise wave number @xmath9 ( see eq . ( [ k+dk ] ) ) , and therefore , does not change static transport properties of the surface disordered waveguide . thus , we will omit it because our interest is in the attenuation length @xmath0 , not in @xmath11 . therefore , it is necessary to identify the terms that have a non - zero mean - value in the general expression . one can see that the mean value of the kernel in eq . is defined by the sgs - terms that are proportional to @xmath91 . in order to construct the fluctuating part of @xmath62 , we introduce the _ square - gradient functions _ @xmath92 for the lower ( @xmath93 ) and upper ( @xmath94 ) boundary , by subtracting the mean values , @xmath95 , from @xmath96 , @xmath97 the functions @xmath98 play an important role in our further consideration . in accordance with the gaussian nature of the surface - profile functions @xmath24 , there are no correlations between @xmath98 and both @xmath27 and @xmath99 , @xmath100 as well as between the product @xmath101 and @xmath102 , @xmath27 or @xmath99 . the binary correlators for @xmath98 are given by @xmath103 correspondingly , the fourier transform of @xmath104 reads , @xmath105 due to the standard relation , @xmath106 here we have introduced the _ roughness - square - gradient power _ ( rsgp ) spectrum @xmath107 , @xmath108 we also introduce the fourier transform of the product @xmath109 and its binary correlator , @xmath110 \label{corr - ft - xi1-xi2 } & & \langle v_{\updownarrow}(k_x)v_{\updownarrow}(k_x')\rangle= 2\pi\delta(k_x+k_x')\,t(k_x),\end{aligned}\ ] ] as well as the fourier transform of @xmath27 , @xmath111 it should be pointed out that by the integration by parts the power spectrum of the roughness gradients @xmath99 can be reduced to the rhp spectrum @xmath34 . however it is not possible to do for the rsgp spectrum @xmath107 . this very fact reflects a highly non - trivial role of the sgs- and gis - terms , since they can compete with the other terms , although they are proportional to @xmath87 . we can write now the fluctuating part of the total scattering potential in the form , @xmath112 here the first term is associated with the terms depending in eq . on @xmath113 and @xmath114 , therefore , on the amplitude and gradient scattering mechanisms , respectively , [ ub - xi1u ] @xmath115\right\}\exp(ik_x'x_1)\,\frac{\sin(k_z'z_1)}{k_z ' } \qquad \label{ub - xi1u - def}\\[6pt ] & & \xi_{1\uparrow}(k_n , k_{n'})= \nonumber \\ & & -\sigma\{\delta_{nn'}+\cos[\pi(n - n')](1-\delta_{nn'})\ } \tilde\xi_{\uparrow}(k_n - k_{n ' } ) . \label{ub - xi1u - exp}\end{aligned}\ ] ] the second term in eq . corresponding to the lower boundary , is of the same form , [ ub - xi1d ] @xmath116\right\ } \exp(ik_x'x_1)\,\frac{\sin(k_z'z_1)}{k_z ' } \qquad\quad \label{ub - xi1d - def}\\[6pt ] & & \xi_{1\downarrow}(k_n , k_{n'})=\sigma\ , \tilde\xi_{\downarrow}(k_n - k_{n ' } ) . \label{ub - xi1d - exp}\end{aligned}\ ] ] next two terms in eq . are the sgs - terms , with @xmath117 being replaced by @xmath118 , [ ub - xi2u ] @xmath119 \nonumber\cr\\ & & \times\exp(ik_x'x_1)\,\frac{\sin(k_z'z_1)}{k_z ' } \label{ub - xi2u - def}\\[6pt ] & & \xi_{2\uparrow}(k_n , k_{n'})= \nonumber \\ & & \frac{\sigma^2d}{2 } \bigg\{\left[\frac{1}{3}+ \frac{1}{(2\pi n)^2}\right]\delta_{nn ' } + \frac{4}{\pi^2}\,\frac{n^2+n'^2}{(n^2-n'^2)^2 } \nonumber\cr \\ & & \times \cos[\pi(n - n')]\ , ( 1-\delta_{nn ' } ) \bigg\}\ , v_{\uparrow}(k_n - k_{n ' } ) . \qquad \label{ub - xi2u - exp}\end{aligned}\ ] ] [ ub - xi2d ] @xmath120 \nonumber\cr\\ & & \times\exp(ik_x'x_1)\,\frac{\sin(k_z'z_1)}{k_z ' } \label{ub - xi2d - def}\\[6pt ] & & \xi_{2\downarrow}(k_n , k_{n'})= \nonumber \\ & & \frac{\sigma^2d}{2}\bigg\{\left[\frac{1}{3}+ \frac{1}{(2\pi n)^2}\right]\delta_{nn ' } \nonumber\cr\\ & & + \frac{4}{\pi^2}\,\frac{n^2+n'^2}{(n^2-n'^2)^2}\ , ( 1-\delta_{nn ' } ) \bigg\}\,v_{\downarrow}(k_n - k_{n'}).\qquad\quad \label{ub - xi2d - exp}\end{aligned}\ ] ] finally , the last term in eq . is due to the gis terms in eq . , [ ub - xi2lu ] @xmath121 \nonumber\cr\\ & & \times\exp(ik_x'x_1)\,\frac{\sin(k_z'z_1)}{k_z ' } \label{ub - xi2lu - def}\\[6pt ] & & \xi_{2\updownarrow}(k_n , k_{n'})= \nonumber \\ & & \frac{\sigma^2d}{2}\bigg\{\left[\frac{1}{3}- \frac{1}{2(\pi n)^2}\right]\delta_{nn'}-\frac{8}{\pi^2}\ , \frac{n^2+n'^2}{(n^2-n'^2)^2 } \nonumber\cr\\ & & \times\cos^2[\pi(n - n')/2]\,(1-\delta_{nn ' } ) \bigg\}\ , v_{\updownarrow}(k_n - k_{n ' } ) . \label{ub - xi2lu - exp}\end{aligned}\ ] ] the kernel @xmath122 written as the sum of specially designed terms , has a very convenient form . first , all the terms are chosen to have zero average . second , since the functions @xmath27 , @xmath123 and @xmath124 do not correlate with each other , their fourier transforms are also non - correlated : [ ft - corr ] @xmath125 therefore , all the terms in eq . determining the potential , are also non - correlated . their autocorrelators are , [ ub - xi1-corr - q1 ] @xmath126 \label{ub - q1 } & q_1(k_n , k_{n'})=\sigma^2w(k_n - k_{n'}),\end{aligned}\ ] ] [ ub - xi2-corr - q2 ] @xmath127 \label{ub - q2 } & q_2(k_n , k_{n'})=\frac{\sigma^4d^2}{2}t(k_n - k_{n ' } ) \bigg\{\left[\frac{1}{3}+\frac{1}{(2\pi n)^2}\right]^2\delta_{nn ' } \notag \\ & + \frac{16}{\pi^4}\,\frac{(n^2+n'^2)^2}{(n^2-n'^2)^4}\ , ( 1-\delta_{nn'})\bigg\},\end{aligned}\ ] ] [ ub - xi2lu - corr - q2lu ] @xmath128 \label{ub - q2lu } & q_{2\updownarrow}(k_n , k_{n'})=\frac{\sigma^4d^2}{4}t(k_n - k_{n ' } ) \left\{\left[\frac{1}{3}- \frac{1}{2(\pi n)^2}\right]^2\delta_{nn'}\right . \nonumber \\ & \left.+\frac{64}{\pi^4}\,\frac{(n^2+n'^2)^2}{(n^2-n'^2)^4}\ , \cos^4[\pi(n - n')/2]\,(1-\delta_{nn'})\right\}. \raisetag{-5pt}\end{aligned}\ ] ] note that in spite of the absence of correlations between the upper and lower surface profiles , there are the correlations due to their derivatives since @xmath129 ( see eq . ) . as a result , the correlator of the fluctuating scattering potential in the @xmath36-representation reads , [ ub - tilde - xi - corr ] @xmath130 \label{ub - q } & q(k_n , k_{n'})=2q_1(k_n , k_{n'})+2q_2(k_n , k_{n ' } ) + q_{2\updownarrow}(k_n , k_{n'})\ , . \raisetag{-4pt}\end{aligned}\ ] ] the factor @xmath131 in the first and second terms arises due to independent contributions , @xmath132 and @xmath133 , from the lower and upper boundaries . in this configuration both boundaries of the waveguide are identical , see eq . , therefore , antisymmetric with respect to the central line @xmath134 . for this specific case the width of the waveguide ( the distance between upper and lower boundaries ) remains constant along the @xmath135direction . instead of eq . we have to start now with the exact expression ( ) by substituting the relation ( ) . as a result , one gets the relatively simple expression for the scattering potential , @xmath136\ , \xi'(x_1)\,\frac{\partial}{\partial z_1 } \\ & -\sigma^2\,{\xi'}^2(x_1)\frac{\partial^2}{{\partial z_1}^2}\bigg ) \exp\left(ik'_xx_1\right)\frac{\sin(k'_zz_1)}{k'_z } \end{split}\ ] ] one can realize that all terms proportional to @xmath87 can be written as one sgs - term with a non - zero mean value . by making use of the functions defined by eqs . , and with @xmath94 , in what follows we omit the subscript @xmath137 . now the fluctuating part of the scattering potential is , @xmath138 the first summand related to the gradient scattering terms in eq . has the form , [ ccb - xi1 ] @xmath139\,\xi'(x_1)\,\frac{\partial}{\partial z_1 } \exp(ik_x'x_1)\,\frac{\sin(k_z'z_1)}{k_z ' } \label{ccb - xi1-def}\\[6pt ] & & \xi_1(k_n , k_{n'})=2\sigma\sin^2[\pi(n - n')/2]\ , \tilde\xi(k_n - k_{n ' } ) . \label{ccb - xi1-exp}\end{aligned}\ ] ] the second summand is due to the sgs - term , however , with the function @xmath140 instead of @xmath141 . with the use of the fourier transform ( [ ft - v ] ) for the operator @xmath142 , one gets , [ ccb - xi2 ] @xmath143 & & \xi_2(k_n , k_{n'})=\frac{\sigma^2d}{2}\,\delta_{nn'}\ , v(k_n - k_{n ' } ) . \label{ccb - xi2-exp}\end{aligned}\ ] ] note that the expressions ( [ ccb - xi1 ] ) and ( [ ccb - xi2 ] ) for the scattering potential are still _ exact_. in contrast with other configurations discussed below , the expansion in @xmath144-smallness is not required because of the constant width of the waveguide . the correlation properties of the scattering potential are described by the following formulas . due to the condition ( [ ft - corr - xi - v ] ) , the terms @xmath145 and @xmath146 are non - correlated , @xmath147 on the other hand , the autocorrelators are non - zero , [ ccb - xi1-corr - q1 ] @xmath148 . \nonumber\\\label{ccb - q1}\end{aligned}\ ] ] [ ccb - xi2-corr - q2 ] @xmath149 finally , the correlator of the fluctuating part of the total scattering potential @xmath150 has the following form , [ ccb - xi - tilde - corr - q ] @xmath151 let us now consider the waveguide with symmetric boundaries , see eq . ( [ sbcase ] ) , therefore , with the width defined by @xmath152 by substituting the symmetric profile into eq . , we obtain for the surface scattering potential , @xmath153 \nonumber\cr\\ & & -\left.\frac{\sigma^2{\xi'}^2(x_1)}{d^2}\left[1 + 4\left(2z_1-d\right)\ , \frac{\partial}{\partial z_1}+ \left(2z_1-d\right)^2\frac{\partial^2}{\partial z_1 ^ 2}\right]\right\ } \nonumber\cr\\ & & \times\exp\left(ik'_xx_1\right)\frac{\sin(k'_zz_1)}{k'_z}\,.\end{aligned}\ ] ] as for the previous configuration , see eq . , the sgs - terms corresponding to the lower boundary and the gis - terms are written in terms of @xmath154 . the potential in eq . ( [ sb - xi - approx ] ) can be written as the sum of two terms related to substantially different scattering mechanisms , @xmath155 here the first summand is related to the terms proportional to @xmath19 in eq . ( the amplitude and gradient scattering terms ) , [ sb - xi1 ] @xmath156\right\ } \exp(ik_x'x_1)\,\frac{\sin(k_z'z_1)}{k_z ' } \label{sb - xi1-def}\\[6pt ] & & \xi_1(k_n , k_{n'})=-2\sigma\{\delta_{nn'}+\cos^2[\pi(n - n')/2 ] \nonumber\\ & & \times(1-\delta_{nn'})\}\,\tilde\xi(k_n - k_{n'}).\nonumber\\ \label{sb - xi1-exp}\end{aligned}\ ] ] the second summand is associated with the fluctuating part of the sgs - terms in eq . ( [ sb - xi - approx ] ) , and consequently , with the function @xmath157 being replaced by @xmath158 , [ sb - xi2 ] @xmath159 \nonumber\cr\\ & & \times\exp(ik_x'x_1)\,\frac{\sin(k_z'z_1)}{k_z ' } \label{sb - xi2-def}\\[6pt ] & & \xi_2(k_n , k_{n'})=\frac{\sigma^2d}{2 } \bigg\{\left[\frac{1}{3}+ \frac{1}{(\pi n)^2}\right]\delta_{nn'}+\frac{16}{\pi^2}\,\nonumber\cr\\ & & \times\frac{n^2+n'^2}{(n^2-n'^2)^2 } \cos^2[\pi(n - n')/2](1-\delta_{nn'})\bigg\}v(k_n - k_{n'}).\nonumber\\ \label{sb - xi2-exp}\end{aligned}\ ] ] the correlation properties of the introduced surface scattering potentials @xmath160 and @xmath161 are described by the correlator , [ sb - xi1-corr - q1 ] @xmath162(1-\delta_{nn'})\}. \label{sb - q1}\end{aligned}\ ] ] [ sb - xi2-corr - q2 ] @xmath163 ^ 2 \delta_{nn'}\nonumber\\ & & + \frac{256}{\pi^4}\,\frac{(n^2+n'^2)^2}{(n^2-n'^2)^4 } \cos^4[\pi(n - n')/2](1-\delta_{nn'})\bigg\}.\nonumber\\ \label{sb - q2}\end{aligned}\ ] ] thus , we arrive at the expression , [ sb - xi - tilde - corr - q ] @xmath164 now we replace the problem for the random green s function @xmath165 with the problem for the green s function @xmath166 averaged over the surface disorder . to perform the averaging of eq . ( [ g - de ] ) with @xmath167 , given for the uncorrelated , antisymmetric and symmetric boundary configurations by eq . , and , respectively , and obtain @xmath168 , we can apply one of the standard and well known perturbative methods . for example , it can be the diagrammatic approach developed for surface disordered systems @xcite , as well as the technique developed in ref . . both of the methods take adequately into account the multiple scattering from the corrugated boundaries , and allow one to develop the consistent perturbative approach with respect to the scattering potential . after quite cumbersome calculations ( see ref . @xcite for details ) , we obtain the average green s function in the normal - mode representation , @xmath169 the quantity @xmath0 is the wave attenuation length or electron total mean - free - path of the @xmath1th conducting mode . this quantity is determined by the imaginary part of the self - energy and describes the scattering from the @xmath1th mode into other propagating modes , @xmath170 & & \times\left[q(k_n ,- k_{n'})+q(k_n,+k_{n'})\right ] . \label{ln - q}\end{aligned}\ ] ] here the explicit expressions for @xmath171 and @xmath172 are given by eqs . , and for the waveguide with uncorrelated , antisymmetric and symmetric boundaries . in deriving eqs . ( [ < gf > ] ) and ( [ ln - q ] ) we have made the following simplifications . first , the self - energy in the dyson - type equation for the average green s function was obtained within the second - order approximation in the perturbation potential . in terms of the diagrammatic technique this is similar to the simple vortex " or , the same , bourret approximation @xcite which contains the binary correlator @xmath173 of the surface scattering potential and the unperturbed pole factor @xmath174 . second , in order to extract the inverse attenuation length from the self - energy , we substituted the lengthwise wave number @xmath36 by its unperturbed value @xmath9 . the justification of this substitution , as well as the conditions of applicability resulting from the above simplifications , were discussed in ref . specifically , it was shown that the domain of applicability is restricted by two independent criteria of the _ weak surface scattering _ , @xmath175 here @xmath176 is the distance between two successive reflections of the @xmath1th mode from one boundary to the opposite one . the conditions in imply that the wave is weakly attenuated on both the correlation length @xmath29 and the cycle length @xmath176 . they implicitly include the requirement that @xmath79 for the surface corrugations be small in height , used in the sec . [ sec - xi - tilde ] when deriving the explicit form ( [ gen - xi - approx ] ) for the surface scattering potential @xmath167 . in this section we discuss the attenuation length @xmath0 for three configurations of the waveguide . in the first part of the subsections we give explicit expressions for the @xmath177 for any form of the binary correlator @xmath28 . in the second part , we assume that the random surface profile @xmath178 has the gaussian binary correlator , @xmath179 then the rhp spectrum ( [ ft - w ] ) is given by @xmath180 the rsgp spectrum @xmath107 defined by eq . ( [ t - def ] ) ) , can be written as @xmath181 \nonumber\\[6pt ] & \times&\exp\left[-(k_xr_c)^2/4\right]\ , . \label{t - gaus - expl}\end{aligned}\ ] ] in what follows , we express the inverse attenuation length in the form of the dimensionless quantity @xmath182 . since @xmath0 depends on as many as four dimensionless parameters @xmath183 , @xmath184 , @xmath70 , and @xmath1 , a complete analysis appears to be quite complicated . for this reason , below we restrict ourselves to an analysis of the interplay between different scattering mechanisms , as a function of the dimensionless correlation length @xmath184 for different values of @xmath183 . we concentrate our attention on a multimode waveguide with a large number of propagating modes , @xmath185 . our main interest is in two situations , namely , in a small - scale roughness of the `` white noise '' kind for which @xmath186 , and in large - scale random corrugations with @xmath187 . in view of eq . ( [ ub - q ] ) , the general expression ( [ ln - q ] ) for the inverse attenuation length shows that for a waveguide with uncorrelated boundaries this quantity consists of three terms , @xmath188 these terms are originated from different mechanisms of the surface scattering . the first attenuation length @xmath189 is related to the rhp spectrum @xmath34 through the expression for @xmath190 . in accordance with eq . ( [ ub - q1 ] ) it is given by @xmath191 & & \times\left [ w(k_n+k_{n'})+w(k_n - k_{n'})\right].\qquad\qquad\end{aligned}\ ] ] here the term @xmath192 corresponding to @xmath193 , is related to the amplitude scattering while the terms @xmath194 result from the gradient scattering . these two mechanisms of surface scattering are due to the corresponding terms in the expression for @xmath195 , see eqs . and . the first one is related to the term depending on the amplitude of the roughness profile @xmath196 , and the second one is due to the terms depending on the roughness gradient @xmath197 . the expression ( [ ub - ln1-def ] ) exactly coincides with that previously obtained by various methods ( see , e.g. , ref . ) . the second attenuation length @xmath198 related to the rsgp spectrum through @xmath199 , is associated solely with the sgs mechanism due to the operator @xmath98 , see eqs . and . in accordance with eq . ( [ ub - q2 ] ) , it is described by @xmath200 here the terms in the sum are @xmath201 ^ 2 \left[t(2k_n)+t(0)\right]\ ] ] and @xmath202 & \times\left[t(k_n+k_{n'})+t(k_n - k_{n'})\right ] . \end{split}\ ] ] the third term @xmath203 is also related to the rsgp spectrum through @xmath204 due to the product @xmath101 , see eq . ( [ ub - xi2lu - def ] ) . in accordance with eqs . ( [ ub - q2lu ] ) , it is described by @xmath205 where @xmath206 ^ 2 \\[6pt ] & \times\left[t(2k_n)+t(0)\right ] . \end{split}\ ] ] and @xmath207 & \times\left[t(k_n+k_{n'})+t(k_n - k_{n'})\right ] \cos^4[\pi(n - n')/2 ] . \end{split}\ ] ] analogously , the terms in eq . can be classified as partial terms describing the intramode ( @xmath192 , @xmath208 and @xmath209 ) , and intermode ( @xmath194 , @xmath210 and @xmath211 ) scattering . according to the discussion of eq . , the amplitude and gradient terms result in the intramode and intermode scattering , respectively . as for the square - gradient terms , they lead to both types of scattering . to continue our analysis , let us start with the widely used case of _ a small - scale boundary perturbation _ , @xmath186 . in this case the surface roughness can be regarded as a delta - correlated random process with the correlator @xmath212 and constant power spectrum @xmath213 . taking into account the evident relationship @xmath214 , one can get the following inequalities specifying this case , @xmath215 it is necessary to underline that in the regime of small - scale roughness ( [ ssr - def ] ) the second of the weak - scattering conditions in eq.([ws - ln ] ) is not so restrictive as the first one , and directly stems from it , @xmath216 . in eqs . ( [ ub - ln1-def ] ) , ( [ ub - lnn2 ] ) , ( [ ub - lnn2 ] ) , ( [ ub - lnn2lu ] ) and ( [ ub - lnn2lu ] ) for the attenuation lengths the argument of the correlators @xmath34 and @xmath107 turns out to be much less than the scale of their decrease @xmath39 under the conditions ( [ ssr - def ] ) . thus , for any term in the summation over @xmath217 the argument can be taken as zero . therefore , the first attenuation length is determined as follows , [ ub - ln1-ssr ] @xmath218 & & \approx ( k\sigma)^2\frac{n}{kd/\pi}\,\frac{kw(0)}{2 } \label{ub - ln1-ssr - w}\end{aligned}\ ] ] the approximation is made here for a large number of the conducting modes , with the change of the summation over @xmath217 by integration . for a further estimate , one can take into account the formula @xmath219 that directly follows from the definition ( [ ft - w ] ) for the fourier transform @xmath34 of the binary correlator @xmath28 . the function @xmath220 is the dimensionless correlator of the dimensionless variable @xmath221 , with the scale of decrease of the order of @xmath222 . as a result , the function @xmath220 does not depend on @xmath29 . therefore , the integral over @xmath221 entering eq . ( [ w(0 ) ] ) is a positive constant of the order of unity . for example , in the case of the gaussian correlations ( see eq . ( [ wft - gaus ] ) ) , we have , @xmath223 , therefore , the integral is @xmath224 . for the second length in eq . one gets , [ ub - ln2-ssr ] @xmath225 & & \times\bigg\{\left[\frac{1}{3}+\frac{1}{(2\pi n)^2}\right]^2 + \frac{16k_nd}{\pi^4(\pi n / d)^2 } \nonumber\\[6pt ] & & \times\left(\sum_{n'=1}^{n-1}+\sum_{n'=n+1}^{n_d}\right ) \frac{(\pi n'/d)^2}{k_{n'}d}\frac{(n^2+n'^2)^2}{(n^2-n'^2)^4}\bigg\ } \qquad\label{ub - ln2-ssr - sum}\\[6pt ] & & \approx\frac{\pi^2}{4}\frac{(k\sigma)^4}{k_nd } \frac{n^3}{(kd/\pi)}\frac{t(0)}{k^3}. \label{ub - ln2-ssr - nt}\end{aligned}\ ] ] in eq . ( [ ub - ln2-ssr - sum ] ) every term in the sum rapidly decreases with an increase of the absolute value of @xmath226 . this can be seen by making use of the following estimate , @xmath227 therefore , the sum in eq . ( [ ub - ln2-ssr - sum ] ) can be correctly estimated by three terms with @xmath228 . for simplicity , in eq . ( [ ub - ln2-ssr - nt ] ) we assume @xmath229 , and replace the curly braces by factor @xmath230 . the explicit form for @xmath231 directly follows from the definition ( [ t - def ] ) for the correlator @xmath107 , @xmath232 & = & r_c^{-3}\int_{-\infty}^\infty d\rho \left[\frac{d^2{\cal w}(r_c\rho)}{d\rho^2}\right]^2.\end{aligned}\ ] ] if the roughness correlations are of the gaussian form , then according to eq . ( [ t - gaus - expl ] ) , we have @xmath233 and the integral over @xmath221 entering eq . ( [ t(0 ) ] ) is equal to @xmath234 . the second term in eq . is bigger than the third one . therefore , a good aproximation for eq . is @xmath235 one can make sure of this if we consider in eq . the product of the fast decreasing factor and @xmath236 $ ] ( the latter cancels the terms with @xmath237 ) . thus , we arrive at the following estimate , @xmath238 by comparing eq . with eq . one can confirm eq . . the expression works well in the whole domain of applicability of our theory , not only in the region of small - scale roughness . to support this , in fig . [ fig : l2_vs_l2lu ] we compare the plot of @xmath239 with the plot of @xmath240 ; they are practically the same . plot of @xmath239 ( solid line ) and @xmath240 ( dashed line ) versus @xmath184 , for @xmath241 , @xmath242 and @xmath243 . ] therefore , in what follows for this case of uncorrelated boundaries we will neglect the third term in eq . . due to , the relationship between the first and second attenuation lengths is expressed by @xmath244 according to a substantially different behavior of the quantities @xmath245 and @xmath246 with respect to @xmath184 , it becomes clear that they have to intersect at the crossing point @xmath247 . if the crossing point falls onto the region of a small - scale roughness ( @xmath248 ) , its dependence on the model parameters is obtained by equating the expression ( [ ub - l1/l2-ssr ] ) to unity , @xmath249 to the left from this point @xmath247 the sgs length prevails , @xmath250 . to its right the main contribution is due to the first attenuation length , @xmath251 . the expression ( [ ub - cp - ssr ] ) shows that the crossing point is smaller for smaller values of @xmath252 , as well as for smaller mode indices @xmath1 , or for larger values of the mode parameter @xmath70 . the intermediate situation arises when the correlation length @xmath29 becomes much larger than the wave length @xmath253 , but still remains much less than the cycle length @xmath176 , @xmath254 as before , the first of the weak - scattering conditions ( [ ws - ln ] ) is the most restrictive , @xmath216 . since the distance @xmath176 is larger than the correlation length @xmath29 , the successive reflections of the waves from the rough surface are weakly correlated . note that the distance between neighboring wave numbers @xmath9 and @xmath255 is much smaller than the variation scale @xmath39 of the correlators @xmath34 and @xmath107 , @xmath256 this implies that the correlators @xmath34 and @xmath107 are smooth functions of the summation index @xmath217 . therefore , the sum in the expression ( [ ub - ln1-def ] ) for the first attenuation length @xmath189 can be substituted by the integral , [ ub - ln1-wc ] @xmath257 & & \times\frac{1}{k}\left[w(k_n+k_{n'})+w(k_n - k_{n'})\right ] \\[6pt ] \label{ub - ln1-wc - fin } & & = \frac{(k\sigma)^2}{\pi}\frac{n}{kd/\pi}\,\int_{-k}^{k } dk_x\sqrt{k^2-k_x^2}\,\frac{w(k_n - k_x)}{k}.\qquad\quad\end{aligned}\ ] ] eq . ( [ ub - ln1-wc ] ) shows that the first attenuation length is contributed by the scattering of a given @xmath1-th mode into other propagating modes . in order to obtain this asymptotic result , we have used only the condition of week correlations , @xmath258 . therefore , eq . ( [ ub - ln1-wc ] ) provides a reduction to eq . ( [ ub - ln1-ssr ] ) for small - scale corrugations , @xmath186 . in the case of a large - scale roughness , @xmath259 , the formula ( [ ub - ln1-wc ] ) obtained for @xmath189 allows one for further simplifications as was done in ref . . in contrast with @xmath189 , due to a rapidly decaying factor ( [ factordeltan ] ) the sgs length @xmath198 can be still described by keeping tree terms only , @xmath260 , in the sum in eq . ( [ ub - ln2 ] ) . taking into account the estimate ( [ kn - wc ] ) , for the case @xmath229 one can obtain , [ ub - ln2-wc ] @xmath261 & & \times\bigg\{\left[\frac{1}{3}+\frac{1}{(2\pi n)^2}\right]^2 + \frac{8}{\pi^4}\bigg\ } \label{ub - ln2-wc - nt}\\[6pt ] & & \approx\frac{\pi^2}{8}\frac{(k\sigma)^4}{k_nd } \frac{n^3}{(kd/\pi)}\frac{t(0)+t(2k_n)}{k^3}. \qquad\quad\label{ub - ln2-wc - fin}\end{aligned}\ ] ] to get the final expression ( [ ub - ln2-wc - fin ] ) , we replaced the curly braces in eq . ( [ ub - ln2-wc - nt ] ) by the factor @xmath230 . naturally , at small - scale corrugations , @xmath186 , the obtained result ( [ ub - ln2-wc ] ) reduces to eq . ( [ ub - ln2-ssr ] ) . for a large - scale roughness , @xmath259 , one should use eq . ( [ ub - ln2-wc ] ) because of an arbitrary value of the parameter @xmath262 . we do not consider here this case in detail due to its intermediate character . another limit case refers to the correlation length @xmath29 to be very large both in comparison with the wave length @xmath253 and with the cycle length @xmath176 , @xmath263 in this case the number of wave reflections on the scale of the correlation length @xmath29 is large . therefore , the successive reflections are strongly correlated to each other . under the relations ( [ lsr - sc - def ] ) the second of the weak - scattering conditions ( [ ws - ln ] ) is the most restrictive , therefore , the condition of the applicability reads as @xmath264 the latter requirement determines the upper limit for the value of the correlation length @xmath29 . due to eq . ( [ lsr - sc - def ] ) , the distance between neighboring wave numbers @xmath9 and @xmath255 turns out to be much larger than the variation scale @xmath39 of the correlators @xmath34 and @xmath107 , @xmath265 this indicates that the probability of the intermode ( @xmath266 ) transitions is exponentially small and the attenuation lengths are mainly formed by the incoherent intramode ( @xmath193 ) scattering . formally , for strong correlations ( [ lsr - sc - def ] ) the correlators @xmath34 and @xmath107 are sharpest functions of the summation index @xmath217 . in the sums of eqs . ( [ ub - ln1-def ] ) and ( [ ub - ln2 ] ) for the attenuation lengths the main contribution is due to the diagonal terms with @xmath193 , for which @xmath267 and @xmath268 can be neglected in comparison with @xmath269 and @xmath231 . thus , the first attenuation length reads @xmath270 correspondingly , for the sgs length one gets , @xmath271 ^ 2.\ ] ] the ratio of the first attenuation length to the second one can be presented as @xmath272 according to the first inequality in eq . ( [ ws - ln ] ) , applied to the condition ( [ lsr - sc - def ] ) of the strong - correlations and to the evident relationship @xmath273 , one can see that the amplitude scattering length always prevails over the sgs length within the interval of strong correlations . for this reason in the condition of applicability ( [ domain - sc ] ) one should substitute @xmath0 by @xmath189 . this allows one to arrive at the inequalities in the explicit form , @xmath274 before we pass to the next configuration of the waveguide , it is interesting to compare the case of two uncorrelated boundaries with the waveguide with one rough boundary , studied in ref . one can found that under the approximation , the expression for @xmath177 turns out to be as twice as bigger than that one for the one - rough - boundary configuration . this fact is quite natural since in the case of uncorrelated boundaries the two rough surfaces are of the same statistical nature ( see the factor @xmath131 in the first two terms of eq . ) . the remarkable similarity between the behavior of the inverse attenuation length for both configurations can be observed in fig . [ fig : q1d-2rb_l1l2_n_krc ] . note that the normalization factor @xmath176 for one boundary is twice as much of the quantity for two boundaries , and both curves are practically superposed . plot of @xmath182 versus @xmath184 for @xmath241 , @xmath242 and different values of @xmath183 . different sets of curves are shown for three waveguide configurations , as well as for the waveguide with one rough boundary analyzed in ref . @xcite ( dashed line ) . curves without symbols depict eq . for the waveguide with uncorrelated boundaries ; curves labelled by triangles and diamonds stand for eqs . and for waveguides with antisymmetric and symmetric boundaries , respectively ] . according to eqs . ( [ ccb - q ] ) and ( [ ln - q ] ) , the inverse attenuation length can be presented as a sum of the partial inverse lengths @xmath275 that are responsible for the scattering from @xmath1th into @xmath217th modes , @xmath276 however , it is noteworthy that in this case the partial terms describing the intramode @xmath277 and intermode @xmath278 scattering are rather different since they are associated with different scattering mechanisms . indeed , the intermode scattering results from the rhp spectrum only , through the expression in eq . ( [ ccb - q1 ] ) for the kernel @xmath279 . these partial lengths are formed by the gradient scattering mechanism , due to the term containing in the scattering potential the derivative @xmath280 , see eq . . they read as @xmath281 \nonumber\cr\\ & & \times[w(k_n+k_{n'})+w(k_n - k_{n'})].\end{aligned}\ ] ] from the above expression one can draw an interesting conclusion . as one can see , in the case of symmetric boundaries the transitions are forbidden between the modes for which the difference of the indexes @xmath282 is an even number . there are transitions only between odd and even channels . this fact can be explained by an existence of an additional integral of motion , namely , the parity . specifically , if one solves the stationary problem for the effective hamiltonian , there are two sets of eigenstates of different parity . correspondingly , there are two independent sets of eigenvalues , see refs . @xcite . the intramode scattering , related to the rsgp spectrum through the kernel @xmath283 in the eq . , is related to the sgs mechanism via the function @xmath158 in eq . . therefore , we have @xmath284.\ ] ] note that in a single - mode waveguide with @xmath285 the sum over @xmath217 in eq . ( [ ccb - ln ] ) contains only one term with @xmath286 . in this case the backscattering length @xmath287 entering in eq . ( [ ccb - lnn ] ) is in accordance with that obtained in ref . @xcite . bearing in mind very different dependencies of the @xmath34 and @xmath107 power spectra on the roughness correlation length @xmath29 , one can write the inverse attenuation length as the sum of two terms . one is related to @xmath34 , @xmath288 and the other is just the inverse of the intramode attenuation length determined by eq . . therefore , @xmath289 , and one can write , @xmath290 in fig . [ fig : q1d-2rb_l1l2_n_krc ] we depict the length defined by eq . as a function of @xmath184 for different values of @xmath252 ; the curve is labelled with triangles . as above , it is convenient to consider separately three regions of the correlation length . we start with the standard case of a small - scale boundary perturbation specified by the conditions . in this case for any term in the summation over @xmath217 the argument of the correlators @xmath34 and @xmath107 is much less than the scale of their decrease @xmath39 , therefore , it can be taken as zero . thus , for the attenuation length describing the intermode scattering one gets , [ ccb - ln1-ssr ] @xmath291 & & \times\sum_{n'=1}^{n_d}\frac{(\pi n'/d)^2}{k_{n'}d } \sin^4[\pi(n - n')/2]\qquad \label{ccb - ln1-ssr - sum1}\\[6pt ] & \approx & ( k\sigma)^2\frac{n}{kd/\pi}\,\frac{kw(0)}{2 } \label{ccb - ln1-ssr - w}\end{aligned}\ ] ] the factor @xmath292 $ ] gives alternating values of one and zero for @xmath293 and @xmath294 , respectively , therefore , for a large number of conducting modes @xmath295 it can be replaced by @xmath296 . the estimate ( [ ccb - ln1-ssr - w ] ) is obtained with the change of the summation by integration . for a further analysis one has to take into account the formula in eq . for @xmath269 . the second term in eq reads as @xmath297 where the explicit form for @xmath231 can be taken as in eq . . it is easy to realize that the relationship between lengths defined by eqs . and is of the same order as in the case of uncorrelated boundaries , see eq . . in this way one obtains the expression for the crossing point @xmath247 at which the curves of @xmath245 and @xmath246 intercept each other . this intermediate situation is specified by eq . . we replace @xmath292 $ ] by @xmath296 , and change the sum by the integral since the correlator @xmath34 is a smooth function of @xmath217 ( see the discussion before eq . ) , @xmath298 note that eq . coincides with eq . for the case of two uncorrelated boundaries . for the length @xmath198 in the large - scale - roughness region , @xmath259 , one should use the general expression ( [ ccb - lnn ] ) because of arbitrary value of the parameter @xmath262 . its normalized version reads @xmath299 in this limit case specified by eq . , the distance between wave numbers @xmath9 and @xmath255 is much larger than the variation scale @xmath39 of the correlator @xmath34 , see eq . . this means that the probability of the intermode ( @xmath266 ) transitions is exponentially small and the total attenuation length is mainly formed by the incoherent intramode ( @xmath193 ) scattering , @xmath300 for strong correlations the correlator @xmath107 is a sharp function of its argument @xmath36 . thus , the term with @xmath268 can be neglected in comparison with @xmath231 , and one gets , @xmath301 in this region of @xmath184 the sgs mechanism is recovered , determining the total attenuation length . the second inequality in eq . is automatically satisfied for any value of @xmath184 as above , for this configuration it is convenient to write down the inverse attenuation lengths as a sum of two terms , @xmath302 here each term is associated with the corresponding term in eq . , resulting from different mechanisms of the surface scattering . the first term in eq . is related to the rhp spectrum @xmath34 through the expression for @xmath303 . according to eq . ( [ sb - q1 ] ) , we have @xmath304\nonumber\\[6pt ] & & \times\left [ w(k_n+k_{n'})+w(k_n - k_{n'})\right].\qquad\qquad\end{aligned}\ ] ] the term @xmath192 corresponding to @xmath193 , is related to the amplitude scattering while the terms @xmath194 result from the gradient scattering . these two mechanisms of surface scattering are due to the corresponding terms in the expression for @xmath305 , see eq . . namely , the former follows from the term depending on the amplitude of the roughness profile @xmath196 , and the latter from the terms depending on the roughness gradient @xmath197 . the second term in eq . related to the rsgp spectrum through @xmath306 , is associated solely with the sgs mechanism due to the square - gradient function @xmath307 , see eq . ( [ sb - xi2 ] ) . in accordance with eq . ( [ sb - q2 ] ) , it is described by @xmath308 here @xmath309 ^ 2 \left[t(2k_n)+t(0)\right]\ ] ] and @xmath310 & \times\cos^4[\pi(n - n')/2]\\[6pt ] & \times\left[t(k_n+k_{n'})+t(k_n - k_{n'})\right ] . \end{split}\ ] ] note that in a single - mode waveguide with @xmath285 ( therefore with @xmath286 in eqs . and ) the backscattering length @xmath311 takes the expression derived in ref . @xcite . the diagonal part of the inverse attenuation length , @xmath312 , characterizes the intramode scattering . the off - diagonal partial attenuation length @xmath313 describes the intermode scattering . to compare with the above case of the antisymmetric boundaries , here there are no transitions between the modes with odd values for the difference @xmath282 . therefore , only the transitions between odd or even modes are allowed only . this fact is related to the underlying symmetry , and can be associated with specific integrals of motion of the total hamiltonian . in the case of a small - scale roughness the first attenuation length reads [ sb - ln1-ssr ] @xmath314 & & \approx ( k\sigma)^2\frac{n}{kd/\pi}\,\frac{kw(0)}{2}. \label{sb - ln1-ssr - w}\end{aligned}\ ] ] for the second length we have , [ sb - ln2-ssr ] @xmath315 & & \times\bigg\{\left[\frac{1}{3}+\frac{1}{(\pi n)^2}\right]^2 + \frac{256k_nd}{\pi^4(\pi n / d)^2 } \nonumber\\[6pt ] & & \times\left(\sum_{n'=1}^{n-1}+\sum_{n'=n+1}^{n_d}\right ) \frac{(\pi n'/d)^2}{k_{n'}d}\frac{(n^2+n'^2)^2}{(n^2-n'^2)^4 } \nonumber\\[6pt ] & & \times\cos^4[\pi(n - n')/2]\bigg\ } \qquad\label{sb - ln2-ssr - sum}\\[6pt ] & & \approx\frac{\pi^2}{8}\frac{(k\sigma)^4}{k_nd } \frac{n^3}{(kd/\pi)}\frac{t(0)}{k^3}.\label{sb - ln2-ssr - fin}\end{aligned}\ ] ] as well as in the configuration with uncorrelated boundaries , due to the estimate every term in the sum of eq . ( [ sb - ln2-ssr - sum ] ) rapidly decreases with an increase of the absolute value of @xmath316 . in addition , the factor @xmath317 is zero for @xmath293 . therefore , the sum in eq . ( [ sb - ln2-ssr - sum ] ) can be evaluated by three terms with @xmath318 , and one can replace the quantity in the curly braces by @xmath230 . here we also assume , @xmath229 . it is evident that in the small - roughness case the crossing point is determined by eq . . in the intermediate situation the first attenuation length @xmath189 coincides with that obtained for the previous configurations , @xmath319 as for the second length @xmath198 , for @xmath229 it can be written , @xmath320 in this region of @xmath184 the first attenuation length reads , @xmath321 for the sgs length one gets , @xmath322 ^ 2.\ ] ] the ratio of the first attenuation length to the second one can be obtained due to eq . . one can see that the amplitude scattering length always prevails over the sgs length within the interval of strong correlations , @xmath323 . for this reason , the condition of applicability coincides with eq . . in fig . [ fig : q1d-2rb_l1l2_n_krc ] by diamonds we show the behavior of the total dimensionless inverse scattering length as a function of the correlation parameter @xmath184 . in this section we present numerical data for periodic waveguides ( billiards ) with the potential @xmath56 . our main attention is payed to the structure of eigenstates for the discussed types of boundaries . although this information is not directly related to the scattering properties of open waveguides , it is instructive to reveal a role of the sgs - terms in the potential , by comparing the structure of eigenstates for different boundaries . the numerical method is described in details in ref.@xcite ; below we discuss the most interesting results and relate them to the previous studies ( see , ref.@xcite and references therein ) . our particular interest is in the `` repulsion '' effect discovered in ref . @xcite for quasi-1d billiards with rough surfaces . this phenomena can be observed for a subset of eigenfunctions with the smallest values of the transversal quantum number of the unperturbed billiard with perfectly flat boundaries . for these eigenstates it was found that the maximum of the eigenfunction intensity is strongly shifted towards from the rough boundary . this repulsion " was shown to increase with a decrease of the roughness correlation length @xmath29 . based on our numerical data , one can make a conclusion that this effect of repulsion " is directly related to the sgs - terms discussed above . to start with , we show in fig . [ fig : psi(x , z)-sgs - nointersgs ] a typical eigenstate of this type for the billiard with one rough surface ( upper panel ) . to compare with , the low panel presents the corresponding eigenstate for the same potential , however , without sgs - terms . as one can see , by removing these terms , the repulsion disappears . it is also instructive to compare the structure of such eigenstates with those computed for the case of both symmetric and antisymmetric rough boundaries , see fig . [ fig : psi(x , z)-antisym - sym ] . in both cases the potentials have the sgs - terms . however , the repulsion phenomena is absent in the antisymmetric billiard . a close inspection reveals that the difference between these two cases is entirely due to the fact that in the antisymmetric billiard there is no square - gradient scattering between different channels , in contrast to the symmetric billiard . thus one can make a conclusion that the main contribution to the repulsion is due to the intermode sgs - terms . this conclusion supports an observation found in ref . @xcite according to which the effect of the repulsion can be explained due to a strong localization in the channel space . this localization occurs due to a relatively strong interaction between different conducting channels ( waveguide modes ) . ( color online ) upper panel : intensity of a typical eigenstate manifesting a repulsion " from the rough surface . low panel : the corresponding eigenstate for the potential without the sgs - terms.,title="fig : " ] ( color online ) upper panel : intensity of a typical eigenstate manifesting a repulsion " from the rough surface . low panel : the corresponding eigenstate for the potential without the sgs - terms.,title="fig : " ] ( color online ) intensity of an eigenfunction revealing strong repulsion from two symmetric rough boundaries ( low panel ) . the same for the antisymmetric billiard for which there is no interaction between different channels ( upper panel).,title="fig : " ] ( color online ) intensity of an eigenfunction revealing strong repulsion from two symmetric rough boundaries ( low panel ) . the same for the antisymmetric billiard for which there is no interaction between different channels ( upper panel).,title="fig : " ] we have developed a perturbative approach to the surface scattering in multimode quasi-1d waveguides with rough surfaces . the approach is based on the construction of an effective hamiltonian that formally corresponds to the bulk " scattering described by the potential of a very complicated structure . the detailed analysis of this potential has led to the conclusion that there are specific square - gradient scattering ( sgs ) terms that were not taken into account in the previous studies . we show that in spite of their seeming smallness in comparison with the standard terms , they can give a major contribution to the value of the scattering length in the case of a sufficiently strong roughness characterized by a corresponding value of the derivative of boundary profiles . in order to demonstrate a strong influence of this sgs - mechanism , we have applied the approach to the waveguides with the boundaries of three different types . specifically , we considered the waveguides with two uncorrelated boundaries , and compared the results with those obtained for other two cases of symmetric and antisymmetric rough boundaries . in this way one can see the influence of various terms very clearly . according to our analysis , all terms in the expression for the effective potential can be regrouped in three parts . the first one is due to the roughness mean amplitude @xmath19 . another group is characterized by the derivative of profiles , reflecting the fact that the scattering depends on two characteristic parameters , namely , on the amplitude @xmath19 ( or , the variance @xmath87 ) , and on the correlation length @xmath29 inversely proportional to the mean derivative of the profiles . in the expression for the scattering potential the above two types of terms have linear dependence on the root - mean - square roughness height @xmath19 . there is one more group of terms that can not be reduced to the two previous ones , and that gives rise to the square dependence on the profile derivative @xmath324 . for this reason , in the previous studies it was assumed that these sgs - terms can be safely neglected . however , we show that apart from the @xmath19-dependence due to the amplitude and gradient terms , as well as the @xmath87-dependence due to the sgs - terms , there is also dependence on the correlation parameter @xmath29 . as was found , the @xmath29-dependence for the sgs - terms turns out to be very different from those corresponding to the amplitude and gradient terms . our analysis reveals that there is a region of parameters where the sgs - contribution to the inverse scattering length is compared ( or even larger ) with the standard contributions due to the amplitude and gradient scattering . in accordance with the general theory of surface scattering @xcite the wave propagation through any @xmath1-th conducting channel is determined by its total ( or outgoing ) attenuation length @xmath0 . the inverse attenuation length @xmath177 can be presented as the sum of partial inverse lengths , see eqs . , and . the first term @xmath325 is associated with both the amplitude and gradient scattering , it depends on the roughness - height power spectrum @xmath34 . to the contrary , the partial length @xmath326 being connected with the sgs - terms , depends on the power spectrum @xmath107 that is principally different from @xmath34 . for the convenience of comparison between three different types of boundaries , in eq . we have separately written the term @xmath327 that appears in the specific case of two uncorrelated boundaries . this term also depends on @xmath107 . such a representation allows us to analyze the total dependence of the inverse attenuation length @xmath177 on two control parameters , the roughness hight @xmath19 and the correlation length @xmath29 . our analysis shows that one can speak about three characteristic regions of @xmath29 , namely , the small - scale , large - scale weak - correlation and large - scale strong - correlation roughness . as we noticed above , the @xmath34- and @xmath107-power spectra have very different dependencies on the roughness correlation length @xmath29 . specifically , the inverse value of the @xmath328-length decreases with an increase of the correlation parameter @xmath184 ; this occurs in the whole interval of the roughness values . as for the inverse value of the @xmath329-length , it shows a different behavior in each of the three regions of roughness and strongly depends on the waveguide configuration . in the region of a small - scale roughness the inverse length @xmath330 increases with an increase of @xmath184 . therefore , the curves corresponding to the terms @xmath325 and @xmath331 ( @xmath332 for the case of uncorrelated boundaries ) intersect each other , see fig if the crossing point @xmath247 falls into this region of small - scale roughness , it obeys the law given by eq . ( [ ub - cp - ssr ] ) . to the left from the crossing point @xmath247 , the sgs - length @xmath328 prevails over the standard length , @xmath250 . contrary , to the right from the point @xmath247 , the @xmath329-length gives the main contribution to the scattering process , @xmath251 . the detailed analysis of three types of boundaries shows that for the small - scale roughness region one can specify that the crossover ( with an increase of @xmath184 ) occurs due to the following interplays : * uncorrelated boundaries : the two sgs - terms that depend on the @xmath107-power spectrum , versus the amplitude and gradient terms depending on the @xmath34-power spectrum ; * antisymmetric boundaries : the sgs - term versus the gradient term ; * symmetric boundaries : the sgs - term versus the amplitude and gradient terms . thus , we show that in the region of small - scale roughness and for any fixed value of the roughness height @xmath19 , one can indicate the region of small values of the correlation length @xmath29 where the sgs - length @xmath333 predominates over the the length @xmath334 . this predominance occurs in spite of the fact that @xmath335 is proportional to @xmath87 while @xmath336 is proportional to @xmath337 . it is interesting to note that in the region to the left from the point @xmath247 we have observed that the length @xmath338 for symmetric boundaries is twice as much as the length @xmath339 for the uncorrelated boundaries . on the other hand , the latter length is twice as much as the length @xmath340 for the antisymmetric boundaries , i.e. , @xmath341 . there is another border that is characterized by the transition to the large - scale roughness with strong correlations , where a very different behavior of the total inverse length @xmath177 was found for the antisymmetric boundaries , in comparison with the uncorrelated and symmetric ones . for the two last cases , the major mechanism of the scattering is due to the amplitude terms in the potential , and a similar behavior with respect to @xmath184 ( @xmath342 ) occurs for both of them . however , the attenuation length for the uncorrelated boundaries turns out to be twice as much as the length for the symmetric boundaries , i.e. , @xmath343 ( note that this relation is inverted in comparison with the region of small - scale roughness ) . as for the waveguide with the antisymmetric boundaries for which the amplitude terms are absent in the potential , the sgs - mechanism prevails in this region , leading to the dependence @xmath344 . note , that here there is the interplay between the sgs - term and gradient one . the transition point was found to be @xmath345 , with the crossover specified by the following interplays : finally , there is an intermediate region of a large - scale roughness and weak correlations where the first attenuation length mainly contributes , @xmath346 . here one can observe a strong similarity of the inverse attenuation length behavior for all configurations considered . it should be noted that this region is quite difficult for obtaining simple analytical estimates due to a strong dependence of the number of terms in the corresponding sums on the parameter @xmath184 . numerical data manifest that the attenuation length roughly remains the same inside this intermediate region .

string theory is a strong candidate for a unified theory including quantum gravity . however , it still does not have a constructive , nonperturbative definition . the main reason is the lack of our understanding on the real fundamental dynamical variables of string theory . in fact , since the advent of d - branes and the discovery of string dualities , the idea has widely spread that the fundamental dynamical variables need not be strings and can be other types of extended objects . m - theory @xcite is a description of string theory , where membranes are believed to play an important role . the worldvolume theory of membranes is equivalent to a three - dimensional gravity theory , where the target space coordinates of an embedded membrane are expressed as scalar fields in three - dimensional worldvolume ( see @xcite for a review ) . however , the analytic understanding of three - dimensional quantum gravity is still not sufficient , as compared to that of two - dimensional quantum gravity @xcite . here , the roles played by matrix models in string theory should be suggestive , where the feynman diagrams of matrix models are interpreted as triangular ( or polygonal ) discretization of string worldsheets ( see @xcite for reviews ) . furthermore , by introducing the degrees of freedom corresponding to matters on worldsheets or by considering a matrix field theory , one can define various kinds of string theory in terms of matrix models @xcite . the @xmath2 expansion of matrix models , where @xmath1 is the size of matrix , corresponds to the genus expansion of string worldsheets as in @xcite . moreover , the double scaling limit enables us to study the nonperturbative aspects of string theory @xcite as well as their integrable structure @xcite . tensor models @xcite or group field theory @xcite are natural generalizations of matrix models to three ( and higher ) dimensions . for three - dimensional models , the perturbative expansion generates random tetrahedral decompositions of three - dimensional objects . unlike the two - dimensional case , however , these objects are not always manifolds or not even pseudomanifolds . recently the situation was drastically improved by the colored tensor models ( see , e.g. , @xcite for a review ) . it is shown that the colored tensor models admit a large @xmath1 expansion and the leading contributions represent higher dimensional sphere @xcite . moreover , it is claimed that one can take a double scaling limit in the tensor models @xcite . thus , the colored tensor models give a fascinating formulation of higher dimensional quantum gravity . nevertheless , the analytic treatment of tensor models is still not so easy as that of matrix models . for example , tensors can not be diagonalized as matrices can , and an analogue of saddle point method has not been found yet . in the present paper , we propose a new class of models which generate three - dimensional random volumes , by regarding each random diagram as a collection of triangles glued together along multiple hinges as in @xcite . our models have real symmetric matrices as the dynamical variables and are characterized by semisimple associative algebras @xmath0 . although most of the diagrams represent configurations which are not manifolds , . see , e.g. , @xcite for the rigorous definition . ] we show that the set of possible diagrams can be drastically reduced such that only ( and all of the ) three - dimensional manifolds with tetrahedral decompositions appear , by introducing a color structure and taking an appropriate limit of parameters existing in the models . since our models are written with matrices , there should be a chance that various techniques in matrix models can be applied and the dynamics of random volumes can be understood more analytically . we show that our models have a novel strong - weak duality which interchanges the roles of triangles and hinges when @xmath0 is a matrix ring . this duality may suggest the analytic solvability of the models . this paper is organized as follows . in section [ def_of_model ] , we first define our models and show that the models are characterized by semisimple associative algebras @xmath0 . we then give a few examples of the feynman diagrams , and show that some diagrams are not manifolds . from the examples , we deduce a strategy to restrict the models so that only ( and all of the ) three - dimensional manifolds are generated . this strategy is implemented in section [ hm2mr ] , where matrix rings are taken as the defining associative algebras . we explicitly construct models that generate only manifolds as feynman diagrams , by introducing a color structure to the models and letting the associative algebras have centers whose dimensions play the role of free parameters . in section [ hm2gr ] we investigate the models where @xmath0 is set to be a group ring @xmath3 $ ] , and demonstrate how the models depend on details of the group structure of @xmath4 . section [ conclusion ] is devoted to conclusion and discussions . we list some of the future directions for further study of the models , and give a brief comment on the relationship of our models with the colored tensor models . in this section we define a class of models which have matrices as the dynamical variables and generate feynman diagrams consisting of triangles glued together along multiple hinges . we show that the models can be defined by semisimple associative algebras . we then give a few examples of the feynman diagrams , and show that some diagrams are not manifolds . we will conclude the section by giving a strategy to restrict the models so that only three - dimensional manifolds are generated . this strategy will be implemented in the next section , where matrix rings are taken as the defining associative algebras . we first explain the diagrams we are concerned with and give the rule to assign a boltzmann weight to each diagram . we then write down the action which generates such diagrams as feynman diagrams . we consider a set of diagrams , @xmath5 , consisting of triangles glued together along multiple hinges as in @xcite . in order to assign a boltzmann weight to diagram @xmath6 , we first decompose @xmath6 to a set of triangles and a set of multiple hinges ( see fig . [ fig : decomposition ] ) . for each edge of a triangle , we draw an arrow and assign an index from a finite set @xmath7 . we repeat the same procedure for the hinges . we then assign the real numbers @xmath8 and @xmath9 to the indexed triangles and hinges , respectively , as in fig . [ fig : tri - hinge ] . we require that @xmath8 and @xmath9 be cyclically symmetric . then , we glue the triangles and hinges to reconstruct the original diagram in such a way that the identified edges have the same index . in doing this , there may appear the case where the arrows of a triangle and a hinge have opposite directions . to treat such cases , we introduce a tensor @xmath10 which reverses the direction of an arrow ( see fig . [ fig : t - tensor ] ) . note that the tensor @xmath11 should be involutory because the direction of an arrow comes back to the original one after @xmath10 is applied twice : @xmath12 furthermore , the following relations should hold since a hinge ( or a triangle ) whose arrows are all flipped is equivalent to a hinge ( or a triangle ) with indices in reverse order : @xmath13 we define the boltzmann weight @xmath14 of diagram @xmath6 to be the product of @xmath8 and @xmath9 followed by the summation over the indices on the edges : @xmath15 here , @xmath16 are the indices on the edges , and @xmath17 is the symmetry factor of the diagram . the indices in @xmath8 and @xmath18 are contracted when the corresponding edges are identified ( with @xmath10 inserted appropriately if necessary ) . the above diagrams with the prescribed boltzmann weights can be generated as feynman diagrams from the action , the introduction of color structure excludes all such diagrams except for a tetrahedron , which is then interpreted as representing three - sphere @xmath19 obtained by gluing two tetrahedra face to face . ] @xmath20 = \frac{1}{2 } a_i b^i -\frac{\lambda}{6 } c^{ijk } a_i a_j a_k - \sum_{k \geq 2}^{\infty } \frac{\mu_k}{2k } b^{i_1 } \cdots b^{i_k } y_{i_1 \ldots i_k } , \label{hm1_action}\end{aligned}\ ] ] where the dynamical variables @xmath21 and @xmath22 satisfy the relations @xmath23 we have included the coupling constants @xmath24 and @xmath25 @xmath26 to count the numbers of triangles and @xmath27-hinges , respectively . in order to specify the directions of arrows , the index line will be a double line by setting index @xmath28 to be double index @xmath29 . it may be already clear , but we here explain how the action generates the diagrams we are concerned with . there are two kinds of interaction terms , one corresponding to triangles @xmath8 and the other to @xmath27-hinges @xmath9 @xmath30 . the kinetic term @xmath31 yields a propagator that glues an edge of a triangle and that of a hinge . note that two triangles can not be glued to each other without an intermediate hinge , and two hinges can not be glued to each other without an intermediate triangle . in order to handle the case where the tensor @xmath10 needs to be inserted , we should multiply every leg of interaction terms by the factor @xmath32 . however , this is equivalent to inserting the projector @xmath33 in every propagator , which in turn is equivalent to requiring that the dynamical variables be invariant under the action of @xmath10 . in summary , our model is characterized by the data @xmath34 that satisfy the constraints . in the next subsection we show that most of the constraints can be solved by considering semisimple associative algebras . in this subsection , we give an algebraic construction of the model data @xmath34 ( see @xcite and also @xcite for a related idea ) . let @xmath35 be a real semisimple associative algebra . that is , @xmath35 is a linear space over @xmath36 with multiplication ( denoted by @xmath37 ) that satisfies the associativity , @xmath38 . if one introduces a basis @xmath39 as @xmath40 , the multiplication is expressed in the form @xmath41 , where the structure constants @xmath42 satisfy the relations @xmath43 due to associativity . the @xmath27-hinge tensor @xmath9 can then be constructed from @xmath42 as @xmath44 it is easy to see that @xmath9 are cyclically symmetric .- hinge tensor can also be expressed as @xmath45 , where @xmath46 are the representation matrix of the basis @xmath39 in the regular representation of @xmath35 ; @xmath47 @xcite . ] the two - hinge tensor @xmath48 is called the _ metric _ of @xmath35 and will often be denoted by @xmath49 ; @xmath50 . it is known @xcite that the associative algebra @xmath35 is semisimple ( i.e. , isomorphic to a direct sum of matrix rings ) if and only if @xmath51 has its inverse @xmath52 . the constraints and can be solved if there exists an involutory _ anti_automorphism @xmath53 . in fact , the coefficients @xmath10 in @xmath54 satisfies when @xmath55 is involutory . furthermore , when @xmath55 is an antiautomorphism : @xmath56 , we have the relations @xmath57 , which ensure to hold . such an antiautomorphism can be naturally constructed when we set the index @xmath28 to be a double index @xmath29 @xmath58 in order to assign arrows to the edges of triangles and hinges . to see this , we let @xmath35 take the form @xmath59 where @xmath0 and @xmath60 are linear spaces of the same dimension @xmath1 . we fix an isomorphism from @xmath0 to @xmath60 and denote it by @xmath61 . can be taken arbitrarily because it can be absorbed into an automorphism of @xmath0 or @xmath62 . ] we assume that @xmath0 is a semisimple associative algebra with multiplication @xmath37 . we introduce a multiplication ( also denoted by @xmath37 ) to @xmath60 such that @xmath63 is an algebra _ anti_automorphism , @xmath64 @xmath65 . ] then @xmath66 naturally becomes an associative algebra as the tensor product of two associative algebras . the antiautomorphism @xmath53 now can be defined such as to map an element @xmath67 to @xmath68 one can easily show that @xmath55 is certainly an antiautomorphism . we thus find that the constraints and can be solved by giving an associative algebra @xmath0 . note that @xmath35 can also be thought of as the set of algebra endomorphisms of @xmath0 by regarding @xmath60 as the dual linear space of @xmath0 : @xmath69 one then can also define an antiautomorphism @xmath70 for the dual of @xmath35 , @xmath71 , such that the following relation holds : @xmath72 where @xmath73 is the paring between @xmath74 ( the dual of @xmath35 ) and @xmath35 . we rephrase the above construction in terms of the bases @xmath75 and @xmath76 of @xmath0 and @xmath60 : @xmath77 we first represent the isomorphism @xmath61 as @xmath78 and write the structure constants of the multiplication on @xmath0 as @xmath79 . then those of @xmath60 ( appearing in @xmath80 ) are determined from the requirement of antihomomorphism , @xmath81 , to be @xmath82 if we take the basis of @xmath35 to be @xmath83 , then the structure constants @xmath84 are given by @xmath85 from which the @xmath27-hinge tensor is given by @xmath86 with @xmath87 in particular , the metric of @xmath35 takes the form @xmath88 where @xmath89 and @xmath90 are the metrics of @xmath0 and @xmath60 , respectively ; @xmath91 , @xmath92 . can also be written as @xmath93 ] we easily see that @xmath35 is semisimple if @xmath0 is , because @xmath94 has its inverse when @xmath89 does ( and so does @xmath90 ) . the antiautomorphism @xmath95 is now expressed as @xmath96 , that is , @xmath97 for the dual algebra @xmath98 , regarding @xmath76 as the dual basis of @xmath75 , we set a basis of @xmath74 to be @xmath99 , which leads to the pairing @xmath100 . then the antiautomorphism @xmath70 on @xmath74 is expressed as @xmath101 . the dynamical variables @xmath21 and @xmath22 in can be regarded as elements of @xmath74 and @xmath35 , respectively : @xmath102 the condition is then expressed as @xmath103 , @xmath104 . with these double indices , the action is written as @xmath105 where the tensor @xmath106 is arbitrary as long as it satisfies the condition . it is often convenient to use @xmath107 , @xmath108 , and @xmath109 . then the conditions and can be rewritten to the form where @xmath61 does not appear : @xmath110 the action then becomes @xmath111 thus , a set of models can be defined by giving semisimple associative algebras @xmath0 and the tensors @xmath112 satisfying . the isomorphism @xmath113 can be taken arbitrarily and is regarded as a sort of gauge freedom in choosing the basis of @xmath0 or @xmath62 . as stated in the last subsection , the tensor @xmath112 in can be chosen arbitrarily as long as it satisfies the condition . in this paper , we set it to be @xmath114 which one can easily show to satisfy . then the feynman rules of the action become @xmath115{fig2 - 05.eps } \end{array } + \begin{array}{l } \includegraphics[height = 1.2cm]{fig2 - 06.eps } \end{array } \sim \langle a_{ij}b^{kl } \rangle = \delta_i^{\phantom{i}k } \delta_j^{\phantom{j}l } + \delta_i^{\phantom{i}l } \delta_j^{\phantom{j}k } , \label{hm2_propagator } \\ { \rm triangle}&:\ \begin{array}{l } \includegraphics[width = 2.0cm]{fig2 - 07.eps } \end{array } \sim \lambda \ , g^{j_1i_2}\,g^{j_2i_3}\,g^{j_3i_1 } , \label{hm2_triangle } \\ \mbox{$k$-hinge}&:\ \begin{array}{l } \includegraphics[width = 2.0cm]{fig2 - 08.eps } \end{array } \sim \mu_k \ , y_{i_1 \ldots i_k } \ , y_{j_k \ldots j_1 } . \label{hm2_hinge}\end{aligned}\ ] ] recall that the arrows are now expressed with double lines , and thus , when the first ( or second ) term of the propagator is used the edges are glued in the same ( or opposite ) direction . the free energy of this model takes the form @xmath116 here , the sum @xmath117 is taken over all possible connected feynman diagrams @xmath5 , and @xmath17 is the symmetry factor of diagram @xmath6 . @xmath118 is the number of triangles , and @xmath119 the number of @xmath27-hinges . @xmath120 denotes the product of @xmath121 and @xmath122 with the indices contracted according to a given wick contraction , and we call @xmath120 the _ index function of diagram _ @xmath6 . we here regard two diagrams as being the same if the indices are contracted in the same manner . the numerical coefficients in the action are chosen such that independent diagrams give only the symmetry factors to the free energy , by taking into account the symmetry ( rotation and flip ) of triangles and hinges . we stress that the free energy will have a different form from if we make a different choice for @xmath112 other than . we now show that the index function @xmath120 can be expressed as the product of the contributions from _ two - dimensional _ surfaces , each surface enclosing a vertex of diagram @xmath6 . to see this , we first note from the feynman rules that even a connected feynman diagram generally gives disconnected networks of index lines . this is because each hinge has a pair of junction points as for index lines and two index lines out of the same edge of a hinge can enter two different hinges after passing through an adjacent triangle ( see fig . [ fig : part_of_index_network ] ) . we further note that the index lines on two different hinges can be connected ( through an intermediate triangle ) if and only if the hinges share the same vertex of @xmath6 . this means that the _ connected _ index networks have a one - to - one correspondence to the vertices of @xmath6 . we thus find that the index function @xmath120 of diagram @xmath6 is the product of the contributions from connected index networks ( each assigned to a vertex of @xmath6 ) and has the form @xmath123 we also call @xmath124 the index function ( more precisely , the _ index function of vertex _ @xmath125 ) . note that every connected index network takes the form of a polygonal decomposition of a closed surface ( not necessarily a sphere and may include monogons or digons ) , where a @xmath27-valent junction ( or @xmath27-junction ) corresponds to a @xmath27-hinge where @xmath27 index lines meet ( see fig . [ fig : index_network ] ) . we here make a few comments . the first comment is on the uniqueness in interpreting an index network as a polygonal decomposition of a closed surface . in fact , if one regards an index network simply as a wire frame ( i.e. , as a collection of segments ) , then it is not a unique procedure to assign polygonal faces in the frame such that the resulting configuration forms a closed surface .- simplex ( @xmath126 ) which can be constructed from triangles and multiple hinges . [ fn : higher - dimensional - simplex ] ] however , our index network is not simply a wire frame , and has the information on how the indices are contracted . we thus can uniquely assign faces to the holes of the index network by carefully following the contraction of indices . we will see in section [ hm2mr ] that the assignment is straightforward when models are given by matrix rings as the defining associative algebras . the second comment is on the manifoldness of a diagram @xmath6 . since there is a two - dimensional surface around each vertex of @xmath6 , we can say that there is a three - dimensional cone at each vertex , the base and apex of a cone being the connected index network around a vertex and the vertex itself , respectively . for example , if an index network has the topology of two - sphere @xmath127 , then the corresponding cone is a 3-dimensional ball @xmath128 . these cones characterize the neighborhoods of the vertices of the diagram @xmath6 . ) may be better regarded as being a higher dimensional object . in this case , the above three - dimensional cone will be treated as a part of the neighborhood of a vertex in the higher dimensional object . ] note that @xmath6 represents a three - dimensional ( combinatorial ) manifold if @xmath6 gives a tetrahedral decomposition and the neighborhood of every vertex is homeomorphic to @xmath128 . in section [ hm2mr ] , by taking @xmath0 to be a matrix ring and introducing a color structure to the models , we show that the set of possible feynman diagrams can be drastically reduced such that only ( and all of the ) manifolds are generated . the index function @xmath120 can be easily evaluated by deforming each connected index network with the use of the associativity of @xmath129 @xcite : @xmath130{fig2 - 11.eps } \end{array } = y_{ij}^{\phantom{ij}l}y_{lkm } = y_{ilm } y_{jk}^{\phantom{jk}l } = \begin{array}{l } \includegraphics[height = 2cm]{fig2 - 12.eps}\end{array } . \label{2 - 2}\end{aligned}\ ] ] in deformation there may appear two kinds of index loops : @xmath131{fig2 - 13.eps } \end{array } & = y_{ik}^{\phantom{ik}l}\,y_{jl}^{\phantom{jl}k } ~~\bigl(\ , = g_{ij } = \begin{array}{l } \includegraphics[width = 1.75cm]{fig2 - 14.eps } \end{array } \bigr ) , \label{2d - bubble } \\ \begin{array}{l } \includegraphics[width = 4cm]{fig2 - 15.eps } \end{array } & = y_{ik}^{\phantom{ik}l}\,y_{lj}^{\phantom{lj}k } ~~\bigl(\ , \equiv p_{ij } \bigr ) . \label{proj_p}\end{aligned}\ ] ] the former index loop diagram can be replaced by a single solid line , while the loop in the latter index diagram can not be removed . actually @xmath132 ( or more precisely , @xmath133 ) is a projector to the center of algebra @xmath0 , @xmath134 , as can be checked easily @xcite . if a given connected index network does not produce a projector @xmath132 in the process of deformation , the index network can always be deformed to a single circle after repeatedly using and and gives the value @xmath135 , where @xmath1 is the dimension of @xmath0 . on the other hand , if a given connected index network admits the appearance of a projector @xmath132 , the value of the index network is generally less than @xmath1 . gives the linear dimension of @xmath134 . ] note that the two deformations and are actually the local moves of two - dimensional surfaces . therefore , the index function @xmath124 of vertex @xmath125 gives a two - dimensional topological invariant defined by the associative algebra @xmath0 @xcite , and thus has the form @xmath136 , where @xmath137 is the genus of the network . , @xmath138 . ] thus the index function of diagram @xmath6 is expressed as @xmath139 in this subsection we give a few examples of the diagrams generated in our models . if our aim is to apply the models to three - dimensional gravity , we should be able to assign three - dimensional volume to each diagram , and thus it is preferable that the diagrams can be regarded as collections only of tetrahedra . however , as we see in the examples below , there arise a lot of undesired diagrams . we will show in the next section that such undesired diagrams can be automatically excluded by taking specific associative algebras and modifying the form , with an appropriate limit of parameters . first we consider a diagram which represents a tetrahedral decomposition of three - dimensional sphere @xmath19 ( see fig . [ 5-cells ] ) . this is the boundary of the so - called 5-cell or a 4-simplex and can be constructed from five tetrahedra . note that the diagram has ten triangles , ten 3-hinges and five vertices . all the index networks around vertices have the same topology and give triangular decompositions of @xmath127 as in fig . [ fig : s^2 ] . the neighborhood of each vertex is homeomorphic to @xmath128 . since every index network can be deformed to a single circle , the index function of each vertex takes the value @xmath1 ; @xmath140 . thus , the contribution from this diagram to the free energy is given by @xmath141 where @xmath142 is the symmetry factor of the diagram . the next example is a diagram which represents a tetrahedral decomposition of three - dimensional torus @xmath143 ( see fig . [ t^3 ] ) . the diagram has twelve triangles , four 4-hinges and three 6-hinges . it has only a single vertex due to the identification in the diagram . the index network around the vertex also represents @xmath127 as in the previous example for @xmath19 . thus , the contribution from this diagram is given by @xmath144 we can easily generalize the above results to such diagrams that represent tetrahedral decompositions of three - dimensional closed manifolds . since the neighborhood of every vertex is homeomorphic to @xmath128 , the contribution from such a diagram to the free energy is given by @xmath145 where @xmath146 , @xmath147 and @xmath148 represent , respectively , the number of triangles , @xmath27-hinges and vertices of the diagram . note that since the topology of three - dimensional manifolds can not be distinguished by @xmath146 , @xmath147 and @xmath148 alone , , since the euler characteristic of three - dimensional closed manifold is zero , @xmath149 . ] it can happen that topologically different manifolds give contributions of the same form . however , we in principle can distinguish the topology by carefully looking at the way of tetrahedral decompositions , although this is usually a tedious task . another way to examine the topology of diagrams is to evaluate a set of topological invariants of each diagram as in @xcite . this prescription will be further studied in our future paper @xcite . our models also generate diagrams that have vertices whose neighborhoods are not three - dimensional ball @xmath128 . one of such diagrams is depicted in fig . [ t^2cone ] , which consists of four tetrahedra , eight triangles , five edges and three vertices . the neighborhood of vertex 3 is homeomorphic to @xmath128 , but that of vertex 1 ( and also that of vertex 2 ) has the topology of cone over @xmath150 . in fact , the index network around vertex 1 gives a polygonal decomposition of two - dimensional torus @xmath150 . one can check that the euler characteristic of the diagram is not zero . thus , this diagram should not give a manifold ( but still gives a pseudomanifold ) . the contribution from this diagram to the free energy can be evaluated to be @xmath151 where @xmath152 comes from the index network around vertex 3 and equals @xmath135 . by contrast , two of @xmath153 come from vertices 1 and 2 , and have the value @xmath154 , which is the linear dimension of the center of @xmath0 . there also arise diagrams which do not give tetrahedral decompositions . a few simple diagrams are depicted in fig . [ 3_triangles ] . although they have the topology of @xmath19 , it is not suitable to assign three - dimensional volume . we close this section by giving a strategy to choose the parameters in our models such that only tetrahedral decompositions of three - dimensional manifolds are generated as feynman diagrams . as we will show in the proof of the theorem in subsection [ tetrahedron ] , one can ensure a diagram to be a tetrahedral decomposition if the index network around every vertex is a _ triangular _ decomposition of two - dimensional surface . this condition can be realized by introducing a color structure to the models , as we will carry out in subsection [ intro_color ] . furthermore , the manifoldness of the resulting diagrams can be ensured by appropriately choosing the defining associative algebra @xmath0 such that the following two conditions are realized : ( i ) the number of vertices can be fixed by using free parameters in @xmath0 , and ( ii ) @xmath155 for @xmath156 . in fact , due to the expression @xmath157 [ see ] , the dominant contributions come from the diagrams whose index networks all have the topology of two - sphere ( and thus the neighborhood of every vertex has the topology of three - ball ) , namely , from the diagrams that represent ( combinatorial ) manifolds . if @xmath0 does not have free parameters to fix the number of vertices , we extend @xmath0 as needed . this extension will be carried out for matrix rings in subsection [ center ] . note that our models also generate nonorientable diagrams . however , such diagrams always have an index network _ not _ homeomorphic to @xmath127 and thus are also decoupled in the program described in the previous paragraph . in this section , we consider matrix rings as the defining associative algebras of the models . we show that such models can be constructed that generate only manifolds as feynman diagrams , by introducing a color structure to the models and letting the associative algebras have centers whose dimensions play the role of free parameters ( to count the number of vertices ) . matrix ring @xmath158 is the set of real - valued matrices of size @xmath159 . this is an associative algebra with the same rules of addition , scalar product and multiplication as those of matrices . note that @xmath0 has the linear dimension @xmath160 . matrix ring is one of the simplest semisimple associative algebras because any semisimple associative algebra is isomorphic to a direct sum of matrix rings . in this section we analyze a model where @xmath0 is set to be a matrix ring @xmath158 . we take its basis to be @xmath161 ( @xmath162 ) , where @xmath163 is a matrix unit whose @xmath164 element is given by @xmath165 . the structure constants can be read from the multiplication rule of matrices : @xmath166 we stress that the double index @xmath167 corresponds to the single index @xmath168 @xmath169 in the previous section . in subsection [ subsec : general ] thus becomes a quadruple index as @xmath170 . ] one can compute @xmath171 and @xmath172 as @xmath173 by setting the tensor @xmath112 as in : @xmath174 the action has the form @xmath175 where @xmath176 and @xmath177 satisfy the following relations because of the symmetry property : @xmath178 the feynman rules for the action can be expressed with quadruple lines as follows : @xmath179{fig3 - 01.eps } \end{array } + \begin{array}{l } \includegraphics[height = 1.5cm]{fig3 - 02.eps } \end{array } \nonumber \\ & \qquad\qquad \sim \langle a_{abcd}b^{efgh } \rangle = \delta_a^{\phantom{a}e } \delta_{b}^{\phantom{b}f } \delta_c^{\phantom{c}g } \delta_d^{\phantom{d}h } + \delta_a^{\phantom{a}g } \delta_{b}^{\phantom{b}h } \delta_c^{\phantom{c}e } \delta_d^{\phantom{d}f } , \label{hm2mr_propagator } \\ { \rm triangle}&:\ \begin{array}{l } \includegraphics[width = 3.3cm]{fig3 - 03.eps } \end{array } \sim \frac{\lambda}{n^3 } \,\delta^{d_1a_2 } \delta^{c_1b_2 } \ldots \delta^{d_3a_1 } \delta^{c_3b_1 } , \label{hm2mr_triangle } \\ \mbox{$k$-hinge}&:\ \begin{array}{l } \includegraphics[width = 2.6cm]{fig3 - 04.eps } \end{array } \sim n^2 \mu_k \ , \delta_{b_1 a_2 } \delta_{c_1d_2 } \ldots \delta_{b_ka_1 } \delta_{c_kd_1 } . \label{hm2mr_hinge}\end{aligned}\ ] ] note that each of the index lines in becomes a double line . moreover , the index line dose not have branch points in this case due to the index structure of hinges [ see ] . thus , as depicted in fig . [ fig : index_network_matrix ] , the identification of the index network with a polygonal decomposition of two - dimensional surface can be done automatically ( and uniquely ) . ] [ although this identification can also be carried out uniquely even when @xmath0 is not a matrix ring , as argued in the first comment following ] . the contribution from each index network to the free energy can be calculated just as in the standard matrix model . to see this , we first note that each polygon gives a factor of @xmath159 because each index loop ( i.e. the index contraction with respect to one of the double index ) gives @xmath180 . we also see from the coefficients in and that each segment in the polygonal decomposition gives @xmath181 ( one - third contribution from a triangle ) and each junction gives @xmath159 ( one - half contribution from a hinge ) . in total , the contribution from the index network around vertex @xmath125 in the original diagram is given by @xmath182 , where @xmath137 is the genus of the index network around @xmath125 . one can easily see that an insertion of the projector @xmath132 , , into the diagram corresponds to attaching a handle to the index network ( as in @xcite ) and decreases the power of @xmath159 by two . in subsection [ ex_not_tetra ] we argued that undesired diagrams appear in our models . in this subsection we show that they can be excluded by introducing a `` color structure '' to our models . let the size @xmath159 of matrices be a multiple of three , @xmath183 . we then modify the tensor to @xmath184 where @xmath185 is a permutation matrix of the form @xmath186 this modification , we can also introduce the color structure to other algebras by taking the tensor product of the form @xmath187 note that @xmath188 . then , the variables @xmath176 and @xmath177 are expressed as @xmath189 and @xmath190 , and the action has the form @xmath191 ] corresponds to inserting @xmath185 and @xmath192 in a pair into two index lines on every segment in each index network ( see fig . [ fig : triangle_color ] ) . note that only @xmath185 ( not @xmath193 ) are accumulated when following the arrows in each index line . thus , the value of a closed index loop forming @xmath194-gon changes from @xmath195 to @xmath196 we thus see that the index function of a diagram gives a nonvanishing value only when the index network around every vertex has a polygonal decomposition where the number of segments of each polygon is a multiple of three . note that such polygonal decompositions with nonvanishing index functions have the following dependence on the coupling constants . suppose that the index network around vertex @xmath125 has @xmath197 polygons , @xmath198 segments and @xmath199 junctions . here , @xmath200 with @xmath201 the number of @xmath194-gons , and @xmath202 with @xmath203 the number of @xmath27-junctions . it is easy to see that the function @xmath204 can be expressed as @xmath205 . thus @xmath206 is nonnegative for the @xmath207 s in because monogons and digons are excluded due to the color structure [ i.e. , @xmath208 . recalling that the contribution from each diagram is given by @xmath209 and noting that the identification rule of the polygonal decompositions gives the relations @xmath210 we find another expression of : @xmath211 \bigl ( \frac{n}{\lambda } \bigr)^{2 - 2g(v ) } \bigl ( \frac{1}{\lambda } \bigr)^{\frac{1}{3}d(v ) } \biggr ] . \label{hm2mr_colored_diag_value } \end{aligned}\ ] ] therefore , if we expand the free energy around @xmath212 with @xmath213 and @xmath214 being fixed , the leading contribution comes from such diagrams that satisfy @xmath215 for every vertex @xmath125 , namely , from the diagrams where every index network forms a _ triangular _ decomposition . one may think from that it would be possible by taking a limit @xmath216 to single out the diagrams where the index networks are all homeomorphic to two - sphere @xmath127 . however , this is not the case . for example , let us consider a diagram which includes an index network forming a two - torus @xmath150 . since the index network gives the contribution of @xmath217 , we can not distinguish a diagram whose vertices all give index networks homeomorphic to @xmath127 from a diagram which has the same number of such vertices whose index networks are homeomorphic to @xmath127 but also has extra vertices whose index networks are homeomorphic to @xmath150 , because the contributions from the two diagrams to the free energy have the same form . this problem comes from the fact that we can not control the number of vertices only with the coupling constants existing in the model with @xmath218 . however , this can be remedied by setting the algebra @xmath0 to be the direct sum of @xmath219 copies of matrix ring @xmath220 , . ] @xmath221 in fact , the index function of a diagram with @xmath148 vertices becomes proportional to @xmath222 since the index network around each vertex gives a factor of @xmath219 independently , and thus changes to equals the linear dimension of @xmath134 . ] @xmath223 \bigl ( \frac{n}{\lambda } \bigr)^{2 - 2g(v ) } \bigl ( \frac{1}{\lambda } \bigr)^{\frac{1}{3}d(v ) } \biggr].\end{aligned}\ ] ] therefore , we can single out the diagrams where every index network is homeomorphic to @xmath127 , by picking out only the diagrams whose index function gives the values with the same power of @xmath219 as that of @xmath224 . combining the results in subsections [ intro_color ] and [ center ] , we can reduce the set of possible diagrams to those whose index networks all give triangular decompositions of two - sphere @xmath127 . we then can apply the following theorem to conclude that these diagrams represent tetrahedral decompositions of three - dimensional manifolds : * theorem . * _ assume that the index network around every vertex in diagram @xmath6 gives a triangular decomposition of two - sphere . then , @xmath6 represents a tetrahedral decomposition of a three - dimensional manifold . _ we label the vertices , triangles and hinges of diagram @xmath6 as @xmath125 @xmath225 , @xmath226 @xmath227 and @xmath228 @xmath229 , respectively . let @xmath230 denote the index network around vertex @xmath125 , which we assume to have a form of triangular decomposition of two - sphere . note that every corner of a triangle in @xmath6 corresponds to a segment of the index network around some vertex ( see fig . [ proof_fig ] ) . we denote by @xmath231 the segment which is lying on triangle @xmath226 and is placed in the corner at vertex @xmath125 . we choose a vertex ( say @xmath232 ) and focus on an `` index triangle '' formed by three segments @xmath233 in @xmath234 . here , @xmath235 are the triangles on which the three segments live . since all the edges of each triangle are attached to hinges , there are hinges @xmath236 , @xmath237 , @xmath238 , @xmath239 , @xmath240 , @xmath241 as in fig . [ proof_fig ] . may represent the same vertex because the index triangles around them may belong to the same connected component of an index network . ] as is depicted there , the three index triangles @xmath242 , @xmath243 and @xmath244 ensure the existence of the corresponding triangles @xmath245 , @xmath246 and @xmath247 , respectively . we are now going to give a detailed description of these triangles and show that they all coincide , @xmath248 . we first take a look at hinge @xmath236 . we assume that @xmath249 is the positive direction of hinge @xmath176 and label the triangles such that triangle @xmath27 is to the immediate left of @xmath168 when seen from vertex 1 ( see fig . [ proof_fig ] ) . this means that triangle @xmath27 is to the immediate right of @xmath168 when seen from vertex 2 , so that @xmath250 and @xmath251 are two segments of an index triangle around vertex 2 , which will be complemented by the third segment @xmath252 as in fig . [ proof_fig ] . the triangle @xmath245 on which the segment @xmath252 lives is glued to triangle @xmath168 along hinge @xmath241 , and must be to the immediate left of @xmath168 when seen from vertex 2 in the direction of @xmath253 . we repeat the same argument for hinge @xmath237 . there , triangle @xmath254 is to the immediate right of @xmath168 when seen from vertex 1 . this means that triangle @xmath254 is to the immediate left of @xmath168 when seen from vertex 3 , so that @xmath255 and @xmath256 are two segments of an index triangle around vertex 3 , which will be complemented by the third segment @xmath257 as in fig . [ proof_fig ] . the triangle @xmath246 on which the segment @xmath257 lives is glued to triangle @xmath168 along hinge @xmath241 , and must be to the immediate right of @xmath168 when seen from vertex 3 in the direction of @xmath253 . however , this means that @xmath246 is to the immediate left of @xmath168 when seen from vertex 2 , and thus two triangles @xmath245 and @xmath246 must be the same . the same argument can also be made for hinge @xmath238 , and we obtain @xmath248 , from which we see that there exists a tetrahedron surrounded by four triangles @xmath168 , @xmath254 , @xmath27 , @xmath245 . by repeating the same arguments for all the index triangles around every vertex , we conclude that diagram @xmath6 gives a tetrahedral decomposition . furthermore , since the index network around every vertex represents a triangular decomposition of @xmath127 , the neighborhood of every vertex is homeomorphic to @xmath128 . therefore , the diagram @xmath6 gives a tetrahedral decomposition of a three - dimensional manifold . we have shown that a class of our models allow us to single out the diagrams which represent tetrahedral decompositions of three - dimensional manifolds . such models can be used to define discretized three - dimensional euclidean gravity . in fact , we only need to follow the arguments given in @xcite . the action of three - dimensional euclidean gravity is given by @xmath258 where @xmath259 corresponds to the bare gravitational coupling and @xmath260 to the bare cosmological constant . this can be discretized by using regular tetrahedra with fixed spacing @xmath261 as @xmath262s_3.\end{aligned}\ ] ] here , @xmath148 and @xmath263 denote the number of vertices and tetrahedra , respectively , and @xmath264 is the angle between two neighboring triangles in a regular tetrahedron . the free energy of this action is then given by @xmath265 where @xmath142 is the symmetry factor . in our models , on the other hand , each diagram representing a tetrahedral decomposition contributes to the free energy as @xmath266 here we have set @xmath267 @xmath268 . since the relations @xmath269 and @xmath270 hold for tetrahedral decompositions of a three - dimensional manifold , the contribution takes the form @xmath271 comparing and , we obtain the relations between the coupling constants of the two models , @xmath272 we conclude this section by commenting that there exists a novel strong - weak duality which interchanges the roles of triangles and hinges when @xmath0 is a matrix ring . we expect this duality to play an important role when we further study the analytic properties of the models in the future . we first recall that one has two choices when introducing a structure of associative algebra to the tensor product of linear spaces , @xmath273 [ see ] . the first is the algebra structure as the tensor product of two associative algebras @xmath0 and @xmath62 . this is the structure we have used exclusively so far , and gives the multiplication ( denoted by @xmath37 ) , which can also be written as @xmath274 the second is the algebra structure as the set of endomorphisms of @xmath0 ; @xmath275 . the multiplication is defined as the composition of two linear operators acting on @xmath0 and will be denoted by dot `` @xmath276 '' : @xmath277 which can also be written as @xmath278 we will show that there is a duality between the two algebra structures when @xmath0 is a matrix ring . we first set @xmath279 . this is possible because @xmath61 can be chosen in an arbitrary way [ see a comment following ] . then , when @xmath280 , the multiplications are represented as @xmath281 we now introduce the dual variables @xmath282 to @xmath177 as @xmath283 which satisfy the symmetry property @xmath284 due to . then one can easily show from and that the two multiplications are interchanged for the dual variables : @xmath285 we further introduce the variables @xmath286 dual to @xmath176 as @xmath287 then the action can be rewritten in terms of the dual variables @xmath286 and @xmath282 to the form @xmath288 note that the way to contract the indices of @xmath286 ( or @xmath282 ) in the dual action is the same as that of @xmath177 ( or @xmath176 ) in the original action . this means that a triangle for the original variables , , now plays the role of a 3-hinge for the dual variables , and a @xmath27-hinge for the original variables , , plays the role of a @xmath27-gon for the dual variables . we thus find that the action for the dual variables generates the dual diagrams to the original ones , consisting of @xmath289-hinges ( dual to original triangles ) and polygons ( dual to original hinges).-gons to appear in the original action for all @xmath290 . ] note that the large @xmath1 limit in ( @xmath291 with @xmath292 and @xmath214 being fixed ) gives @xmath293 and @xmath294 . since @xmath24 and @xmath295 are interchanged in the duality transformation , one sees that this duality is actually a strong - weak duality . in this section , we investigate the models where @xmath0 is set to be a group ring @xmath3 $ ] , and demonstrate how the models depend on details of the group structure of @xmath4 . we assume that @xmath4 is a finite group with order @xmath296 in order to avoid introducing regularizations , although most of the relations below can be applied to continuous compact groups . the action can be rewritten to a form similar to that of matrix ring , by expressing everything in terms of the irreducible representations of @xmath4 . to show this , we first write the delta function as @xmath308 where the sum is taken over all the irreducible representations @xmath309 of @xmath4 with the representation matrix @xmath310 @xmath311 , and @xmath312 is the dimension of representation @xmath309 , @xmath313 . then , the action can be rewritten to the form @xmath314 where @xmath315 this action gives the following feynman rules : @xmath316{fig4 - 01.eps } \end{array } + \begin{array}{l } \includegraphics[height = 1.5cm]{fig4 - 02.eps } \end{array } \nonumber \\ & \quad \sim \langle a^{rs}_{abcd}b^{r^\prime s^\prime}_{efgh } \rangle = \frac{1}{d_r d_s}(\delta_{ae } \delta_{bf } \delta_{cg } \delta_{dh } \delta^{rr^\prime } \delta^{ss^\prime } + \delta_{ag } \delta_{bh } \delta_{ce } \delta_{df } \delta^{rs^\prime } \delta^{sr^\prime } ) , \label{hm2gr_propagator } \\ { \rm triangle}&:\ \begin{array}{l } \includegraphics[width = 4cm]{fig4 - 03.eps } \end{array } \sim \lambda\ , d_{r_1 } d_{r_2 } d_{r_3 } \delta_{a_1d_2 } \delta_{c_1b_2 } \ldots \delta_{a_kd_1 } \delta_{c_kd_1 } , \label{hm2gr_triangle } \\ \mbox{$k$-hinge}&:\ \begin{array}{l } \includegraphics[width = 3cm]{fig4 - 04.eps } \end{array } \sim \mu_k\ , d_r\ , d_s\ , \delta_{a_1 b_2 } \delta_{d_1 c_2 } \ldots \delta_{a_k b_1 } \delta_{d_k c_1 } . \label{hm2gr_hinge}\end{aligned}\ ] ] we thus see that the index network around every vertex is again expressed as a closed surface with double index lines , and its index function is determined only by the euler characteristics of the polygonal decomposition : is regarded as the real sector of the index function for the complexified algebra @xmath317 $ ] , because the group ring @xmath318 $ ] can be expressed as the direct sum of @xmath319 over @xmath309 , @xmath318 = \bigoplus_r m_{d_r}(\mathbb{c})$ ] . ] @xmath320 .\end{aligned}\ ] ] here , elementary group theory shows that @xmath321 , and @xmath322 gives the number of irreducible representations which equals that of conjugate classes . for example , when @xmath4 is the symmetric group @xmath323 , we have @xmath324 where @xmath325 denotes the number of partitions of @xmath159 . therefore , if @xmath4 admits the relations @xmath326 @xmath327 , the index networks of spherical topology have a large value of index function compared to those of higher genera . we also can introduce a color structure as in subsection [ intro_color ] and can control the number of vertices by considering the direct sum of @xmath219 copies of group ring as in subsection [ center ] . therefore , we can again single out the diagrams which give tetrahedral decompositions of three - dimensional manifolds . in this paper we construct a class of models that generate random diagrams consisting of triangles and multiple hinges . the models are completely characterized by semisimple associative algebras @xmath0 and tensors @xmath112 . when @xmath112 are chosen as in or , each feynman diagram can be expressed as a collection of index networks around vertices . the contribution @xmath120 from each diagram @xmath6 to the free energy is expressed as the product of the index functions @xmath124 of vertices @xmath125 of @xmath6 , and @xmath124 depends only on the topology of the index network around @xmath125 besides the structure of the defining associative algebra . although most of the feynman diagrams do not represent three - dimensional manifolds , we give a general prescription to automatically reduce the set of possible diagrams such that only ( and all of the ) manifolds are generated . we implement the strategy for the models with @xmath0 set to matrix rings , by introducing a color structure and taking the direct sum of @xmath219 copies of matrix ring . we show that every diagram actually gives a tetrahedral decomposition where each vertex has a neighborhood homeomorphic to @xmath128 ( ensured by the statement that the index network around each vertex has the topology of @xmath127 ) . we further demonstrate that there is a novel strong - weak duality in the models which interchanges the roles of triangles and hinges . we also investigate the models where the defining associative algebras are group rings , and show that most of their analytic properties can be understood as a straightforward generalization of those for matrix rings . we now list some of the future directions for further study of the models . the first is about the topology summation . our models actually give a summation over all topologies of three - dimensional manifolds . it seems that we can not distinguish the topology of the feynman diagrams if the tensor @xmath112 has the form or as we took in this paper , because topologically different diagrams can give contributions of the same form to the free energy . one can optimistically think that this represents membrane instability ( see , e.g. , @xcite ) . however , it may also happen that configurations of some specific topology entropically dominate in a critical region , although we have not fully evaluated the numerical coefficients in the free energy and their dependence on topology . another way to investigate the topologies of diagrams is to change the tensor @xmath112 to other forms . in fact , this change significantly modifies the dependence of the index function on the associative algebra @xmath0 . the change of @xmath112 and its effect on topological invariants will be studied in our future paper @xcite . the second direction for further study is about the continuum limit . although we have not fully studied the continuum limit yet , and @xmath328 @xmath30 , since the dependence of @xmath148 and @xmath263 disappear from [ or ] . these values of coupling constants may correspond to a certain ( possibly uninteresting ) critical point because the models are not only diffeomorphism invariant but also weyl invariant . ] there may be a chance to analytically solve the models in the large @xmath1 limit and to determine the critical behaviors , because the dynamical variables of our models are given by symmetric matrices @xmath329 and @xmath330 . the third direction is about the introduction of matters to our models . we expect that extra degrees of freedom representing matter fields can be introduced by making copies of the variables @xmath329 and @xmath330 as in matrix models @xcite . it should be particularly interesting to introduce matter fields corresponding to the target space coordinates of embedded membranes and to study the critical behaviors . it would then be important to investigate if there is an analogue of the so - called `` @xmath331 barrier '' in the models and how the situation is modified when supersymmetry is introduced . we close this section with a brief comment on the relationship of our models with the colored tensor models . it is worth noting that our models ( with a color structure and an appropriate limit of parameters as in subsection [ tetrahedron ] ) generate _ all _ of the possible tetrahedral decompositions of three - dimensional manifolds , and thus should have more configurations than those of the colored tensor models . for example , the colored tensor models do not generate such tetrahedral decompositions where odd number of triangles are glued together along a hinge . since the colored tensor models introduce a pair of tensors as in two - matrix models , they may correspond to three - dimensional gravity with specific matters . in this sense , our models with minimum fine tunings may give a continuum theory ( if exists ) closer to pure gravity . it should be interesting to investigate whether colored tensor models can be obtained by adding some ( not necessarily unitary ) matters to our models . in a remarkable paper @xcite , it is shown that the free energy of three - dimensional colored tensor models depend on the size of tensor , @xmath332 , as @xmath333 where we have suppressed other coupling constants . @xmath334 denotes a colored graph which is dual to a tetrahedral decomposition of three - dimensional pseudomanifold , and @xmath335 is the degree of @xmath334 ( see @xcite for details ) . thus , in the large @xmath332 limit , the leading contribution comes from colored graphs with @xmath336 . it is also shown that if @xmath336 then @xmath334 is dual to a three - sphere @xcite . therefore , the leading order graphs are homeomorphic to @xmath19 . we can say the same thing for our models if we confine our attention to only the diagrams that have tetrahedral decompositions dual to colored graphs . the degree @xmath185 of such a diagram can be evaluated as in @xcite and becomes @xmath337 with which the contribution to the free energy can be rewritten to the form @xmath338 thus , if we take a limit @xmath339 with @xmath340 kept finite , the leading contribution comes from configurations with @xmath341 , that is , tetrahedral decompositions of @xmath19 . it is interesting to study the meaning of the degree for general tetrahedral decompositions which are not dual to colored graphs . the authors thank hikaru kawai for stimulating discussions . mf is supported by mext ( grant no.23540304 ) . ss is supported by the jsps fellowship . = 0.83 s. h. shenker , `` the strength of nonperturbative effects in string theory , '' in `` cargse 1990 , proceedings , random surfaces and quantum gravity '' 191200 . m. r. douglas , `` strings in less than one dimension and the generalized kdv hierarchies , '' phys . lett . b * 238 * ( 1990 ) 176 .

this paper deals with chain graphs under the classic lauritzen - wermuth - frydenberg interpretation . we prove that the regular gaussian distributions that factorize with respect to a chain graph @xmath0 with @xmath1 parameters have positive lebesgue measure with respect to @xmath2 , whereas those that factorize with respect to @xmath0 but are not faithful to it have zero lebesgue measure with respect to @xmath2 . this means that , in the measure - theoretic sense described , almost all the regular gaussian distributions that factorize with respect to @xmath0 are faithful to it . previously , it has been proven that for any undirected graph there exists a regular gaussian distribution that is faithful to it ( * ? ? ? * corollary 3 ) . a stronger result has been proven for acyclic directed graphs : in certain measure - theoretic sense , almost all the regular gaussian distributions that factorize with respect to an acyclic directed graph are faithful to it ( * ? ? ? * theorem 3.2 ) . therefore , this paper extends the latter result to chain graphs . it is worth mentioning that we have recently proved in @xcite a result analogous to the one in this paper but for strictly positive discrete probability distributions with arbitrary prescribed sample space . it is also worth noticing that a result analogous to the one in this paper has been proven in ( * ? ? ? * theorem 6.1 ) under the alternative andersson - madigan - perlman interpretation of chain graphs . there are two important implications of the result proven in this paper : * the use of chain graphs to represent independence models in artificial intelligence and statistics has increased over the years , particularly in the case of undirected graphs and acyclic directed graphs . however , there are independence models that can be represented exactly by chain graphs but that can not be represented exactly by undirected graphs or acyclic directed graphs . as a matter of fact , the experimental results in @xcite suggest that this may be the case for the vast majority of independence models that can be represented exactly by chain graphs . in other words , for most chain graphs , every undirected graph and acyclic directed graph either represents some separation statement that is not represented by the chain graph or does not represent some separation statement that is represented by the chain graph . as ( * section 1.1 ) points out , something that would confirm that this is an advantage of chain graphs for modeling regular gaussian distributions would be proving that any independence model represented by a chain graph can be represented by a regular gaussian distribution . the result in this paper confirms this point . * in the literature , there exist two graphical criteria for identifying independencies holding in a probability distribution @xmath3 that factorizes with respect to a chain graph @xmath0 : the moralization criterion @xcite and the c - separation criterion @xcite . both criteria are known to be equivalent ( * ? ? ? * lemma 5.1 ) . furthermore , both criteria are known to be sound , i.e. they only identify independencies in @xmath3 ( * ? ? ? * theorems 3.34 and 3.36 ) . the result in this paper implies that both criteria are also complete for regular gaussian distributions : if @xmath3 is a regular gaussian distribution , then both criteria identify all the independencies in @xmath3 that can be identified on the sole basis of @xmath0 , because there exists a regular gaussian distribution that is faithful to @xmath0 . the rest of the paper is organized as follows . we start by reviewing some concepts in section [ sec : preliminaries ] . in section [ sec : cg ] , we describe how we parameterize the regular gaussian distributions that factorize with respect to a chain graph . we present our results on faithfulness in section [ sec : faithfulness ] . in section [ sec : equivalence ] , we present some results about chain graph equivalence that follow from the results in section [ sec : faithfulness ] . finally , we close with some discussion in section [ sec : conclusions ] . in this section , we define some concepts used later in this paper . we first recall some definitions from probabilistic graphical models . see , for instance , @xcite and @xcite for further information . let @xmath4 be a finite set of size @xmath5 . the elements of @xmath6 are not distinguished from singletons and the union of the sets @xmath7 is written as the juxtaposition @xmath8 . we denote by @xmath9 the size or cardinality of a set @xmath10 , e.g. @xmath11 . we assume throughout the paper that the union of sets precedes the set difference when evaluating an expression . unless otherwise stated , all the graphs in this paper are defined over @xmath6 . if a graph @xmath0 contains an undirected ( resp . directed ) edge between two nodes @xmath12 and @xmath13 , then we write that @xmath14 ( resp . @xmath15 ) is in @xmath0 . if @xmath15 is in @xmath0 then @xmath12 is called a parent of @xmath13 . let @xmath16 denote the set of parents in @xmath0 of the nodes in @xmath10 . when @xmath0 is evident from the context , we drop the @xmath0 from @xmath16 and use @xmath17 instead . a route from a node @xmath12 to a node @xmath18 in a graph @xmath0 is a sequence of nodes @xmath19 such that there exists an edge in @xmath0 between @xmath20 and @xmath21 for all @xmath22 . the length of a route is the number of ( not necessarily distinct ) edges in the route , e.g. the length of the route @xmath19 is @xmath23 . we treat all singletons as routes of length zero . a path is a route in which the nodes @xmath19 are distinct . a route is called undirected if @xmath24 is in @xmath0 for all @xmath22 . a route is called descending if @xmath24 or @xmath25 is in @xmath0 for all @xmath22 . if there is a descending route from @xmath12 to @xmath18 in @xmath0 , then @xmath12 is called an ancestor of @xmath18 and @xmath18 is called a descendant of @xmath12 . let @xmath26 denote the set of ancestors in @xmath0 of the nodes in @xmath10 . a descending route @xmath19 is called a directed pseudocycle if @xmath25 is in @xmath0 for some @xmath22 , and @xmath27 . a chain graph ( cg ) is a graph ( possibly ) containing both undirected and directed edges and no directed pseudocycles . an undirected graph ( ug ) is a cg containing only undirected edges . the underlying ug of a cg is the ug resulting from replacing the directed edges in the cg by undirected edges . a set of nodes of a cg is connected if there exists an undirected route in the cg between every pair of nodes in the set . a connectivity component of a cg is a connected set that is maximal with respect to set inclusion . hereinafter , we assume that the connectivity components @xmath28 of a cg @xmath0 are well - ordered , i.e. if @xmath15 is in @xmath0 then @xmath29 and @xmath30 for some @xmath31 . the moral graph of a cg @xmath0 , denoted @xmath32 , is the undirected graph where two nodes are adjacent iff they are adjacent in @xmath0 or they are both in @xmath33 for some connectivity component @xmath34 of @xmath0 . the subgraph of @xmath0 induced by @xmath10 , denoted @xmath35 , is the graph over @xmath36 where two nodes are connected by a ( un)directed edge if that edge is in @xmath0 . a path @xmath19 in @xmath0 is called a complex if the subgraph of @xmath0 induced by the set of nodes in the path looks like @xmath37 . the path @xmath38 is called the region of the complex . a section of a route @xmath39 in a cg is a maximal subroute of @xmath39 that only contains undirected edges . a section @xmath40 of @xmath39 is a collider section of @xmath39 if @xmath37 is a subroute of @xmath39 . furthermore , a route @xmath39 in a cg is said to be superactive with respect to @xmath41 when * every collider section of @xmath39 has some node in @xmath42 , and * every other section of @xmath39 has no node in @xmath42 . a set @xmath10 is complete in an ug @xmath0 if there is an undirected edge in @xmath0 between every pair of distinct nodes in @xmath36 . we denote the set of complete sets in @xmath0 by @xmath43 . we treat all singletons as complete sets and , thus , they are included in @xmath43 . let @xmath44 denote a column vector of random variables and @xmath45 @xmath46 its subvector @xmath47 . we use upper - case letters to denote random variables and the same letters in lower - case to denote their states . unless otherwise stated , all the probability distributions in this paper are defined on ( state space ) @xmath48 . let @xmath36 , @xmath49 and @xmath42 denote three disjoint subsets of @xmath6 . we denote by @xmath50 that @xmath45 is independent of @xmath51 given @xmath52 in a probability distribution @xmath3 . likewise , we denote by @xmath53 that @xmath36 is separated from @xmath49 given @xmath42 in a cg @xmath0 . specifically , @xmath53 holds when there is no route in @xmath0 from a node in @xmath36 to a node in @xmath49 that is superactive with respect to @xmath42 . this is equivalent to say that @xmath53 holds when every path in @xmath54 from a node in @xmath36 to a node in @xmath49 has some node in @xmath42 . the independence model represented by a cg @xmath0 is the set of separation statements @xmath53 . we say that a probability distribution @xmath3 is markovian with respect to a cg @xmath0 when @xmath50 if @xmath53 for all @xmath36 , @xmath49 and @xmath42 disjoint subsets of @xmath6 . we say that @xmath3 is faithful to @xmath0 when @xmath50 iff @xmath53 for all @xmath36 , @xmath49 and @xmath42 disjoint subsets of @xmath6 . we denote by @xmath55 and @xmath56 that @xmath50 and @xmath53 do not hold , respectively . we now recall some results from matrix theory . see , for instance , @xcite for more information . let @xmath57 denote a square matrix . let @xmath58 with @xmath59 denote its submatrix @xmath60 . the determinant of @xmath61 can recursively be computed , for fixed @xmath62 , as @xmath63 , where @xmath64 denotes the matrix produced by removing the row @xmath65 and column @xmath66 from @xmath61 . if @xmath67 then the inverse of @xmath61 can be computed as @xmath68 for all @xmath69 . we say that @xmath61 is strictly diagonally dominant if @xmath70 for all @xmath62 , where @xmath71 denotes absolute value . a matrix @xmath61 is hermitian if it is equal to the matrix resulting from , first , transposing @xmath61 and , then , replacing each entry by its complex conjugate . clearly , a real symmetric matrix is hermitian . a real symmetric @xmath72 matrix @xmath61 is positive definite if @xmath73 for all non - zero @xmath74 . [ rem : inverse ] note that @xmath75 is a real polynomial in the entries of @xmath61 , and that @xmath76 is then the restriction of a fraction of two real polynomials in the entries of @xmath61 to the area where @xmath75 is non - zero . finally , we recall some results about gaussian distributions . we represent a gaussian distribution as @xmath77 where @xmath78 is its mean vector and @xmath79 its covariance matrix . we say that a gaussian distribution @xmath77 is regular if @xmath79 is positive definite or , equivalently , invertible . in this paper , we often find more convenient to work with the inverse of the covariance matrix @xmath80 , which is also known as the concentration matrix or precision matrix . since @xmath81 , we thus often write @xmath82 instead of @xmath77 . let @xmath36 , @xmath49 , @xmath42 and @xmath83 denote four disjoint subsets of @xmath6 . any regular gaussian distribution @xmath3 satisfies , among others , the following properties : * symmetry @xmath84 . * decomposition @xmath85 . * intersection @xmath86 . * weak transitivity @xmath87 with @xmath88 . the following results have been proven in ( * ? ? ? * sections 2.3.1 , 2.3.3 ) . for the sake of completeness , appendix a adapts the proofs to the notation used in this paper . let @xmath36 and @xmath49 denote two disjoint subsets of @xmath6 . let @xmath89 where @xmath90 is positive definite . then , as shown in ( * ? ? ? * section 2.3.1 ) and in appendix a , @xmath91 where @xmath92 , @xmath93 and @xmath94 are the following real matrices of dimensions , respectively , @xmath95 , @xmath96 and @xmath97 : @xmath98 @xmath99 and @xmath100 let @xmath101 and @xmath102 where @xmath92 , @xmath93 and @xmath94 are real matrices of dimensions , respectively , @xmath95 , @xmath96 and @xmath97 , and @xmath103 and @xmath94 are positive definite . then , as shown in ( * ? ? ? * section 2.3.3 ) and in appendix a , @xmath104 is a gaussian distribution @xmath105 over @xmath106 where @xmath107 and @xmath108 moreover , @xmath104 is regular because @xmath109 in this section , we describe how we parameterize the regular gaussian distributions that factorize with respect to a cg . this is a key issue because our results about faithfulness are not only relative to the cg at hand and the measure considered , the lebesgue measure , but also to the number of parameters of the regular gaussian distributions that factorize with respect to the cg at hand . we say that a regular gaussian distribution @xmath3 factorizes with respect to a cg @xmath0 with connectivity components @xmath28 if the following two conditions are met ( * ? ? ? * proposition 3.30 ) : * @xmath110 where * @xmath111 where each @xmath112 is a non - negative real function . let @xmath113 denote the set of regular gaussian distributions that factorize with respect to @xmath0 . we parameterize each probability distribution @xmath114 with the following parameters : * the mean vector @xmath78 of @xmath3 . * the submatrices @xmath115 and @xmath116 of the precision matrix @xmath117 of @xmath118 for all @xmath119 . we warn the reader that if @xmath90 denotes the precision matrix of @xmath3 , then @xmath117 is not @xmath120 but @xmath121 . it is worth mentioning that an alternative parameterization of the probability distributions in @xmath113 is presented in @xcite . the main difference between our parameterization and the alternative one is that we parameterize certain concentration matrices whereas they parameterize certain partial concentration matrices . however , both parameterizations are equivalent . we omit the details of the equivalence because they are irrelevant for our purpose . we stick to our parameterization simply because it is more convenient for the calculations performed later in this paper . note that the values of some of the parameters in the parameterization introduced above are determined by the values of the rest of the parameters . specifically , for all @xmath119 , the following constraints apply : * @xmath122 for all @xmath123 , because @xmath124 since @xmath117 is symmetric . * @xmath125 for all @xmath123 such that @xmath66 and @xmath126 are not adjacent in @xmath0 . to see it , note that @xmath66 and @xmath126 are not adjacent in @xmath127 . consequently , any path between @xmath66 and @xmath126 in @xmath127 must pass through some node in @xmath128 or @xmath33 . then , @xmath129 , which implies @xmath130 because @xmath118 is markovian with respect to @xmath127 due to the condition f2 above ( * ? ? ? * proposition 3.30 , theorems 3.34 and 3.36 ) . the latter independence statement implies @xmath131 and , thus , @xmath125 ( * ? ? ? * proposition 5.2 ) . * @xmath132 for all @xmath133 and @xmath134 such that @xmath66 and @xmath126 are not adjacent in @xmath0 , by a reasoning analogous to the one above . hereinafter , the parameters whose values are not determined by the constraints above are called non - determined ( nd ) parameters . however , the values the nd parameters can take are constrained by the fact that these values must correspond to some probability distribution in @xmath113 . we prove in lemma [ lem:121cg ] that this is equivalent to requiring that the nd parameters can only take real values such that @xmath115 is positive definite for all @xmath119 . that is why the set of nd parameter values satisfying this requirement are hereinafter called the nd parameter space for @xmath113 . we do not work out the inequalities defining the nd parameter space because these are irrelevant for our purpose . the number of nd parameters is what we call the dimension of @xmath0 , and we denote it as @xmath1 . specifically , @xmath135 where @xmath136 is the number of edges in @xmath0 : * @xmath137 due to @xmath78 . * @xmath137 due to @xmath138 for all @xmath119 and @xmath133 . * @xmath136 due to the entries below the diagonal of @xmath115 that are not identically zero and the entries of @xmath116 that are not identically zero for all @xmath119 . to see this , recall from the constraints c1-c3 above that there is one entry below the diagonal in some @xmath115 that is not identically zero for each undirected edge in @xmath0 , and one entry in some @xmath116 that is not identically zero for each directed edge in @xmath0 . [ lem:121cg ] let @xmath0 be a cg . there is a one - to - one correspondence between the probability distributions in @xmath113 and the elements of the nd parameter space for @xmath113 . we first prove that the mapping of probability distributions into nd parameter values is injective . obviously , any probability distribution in @xmath114 is mapped into some real values of the nd parameters @xmath78 , @xmath115 and @xmath116 for all @xmath119 . in particular , @xmath115 takes value @xmath139 where @xmath90 is the precision matrix of @xmath3 . then , that @xmath115 is positive definite follows from the fact that @xmath90 is positive definite @xcite . thus , @xmath3 is mapped into some element of the nd parameter space for @xmath113 . moreover , different probability distributions are mapped into different elements . to see it , assume to the contrary that there exist two distinct probability distributions @xmath140 that are mapped into the same element . note that this element uniquely identifies @xmath141 by equations [ eq : bishop1a]-[eq : bishop1c ] for all @xmath119 , where @xmath142 and @xmath143 . likewise , it uniquely identifies @xmath144 for all @xmath119 . then , @xmath145 for all @xmath119 . however , this contradicts the assumption that @xmath3 and @xmath146 are distinct by the condition f1 above . we now prove in three steps that the mapping of nd parameter values into probability distributions is injective . * step 1 * we first show that any element of the nd parameter space for @xmath113 is mapped into some regular gaussian distribution @xmath147 . note that any element of the nd parameter space for @xmath113 uniquely identifies a gaussian distribution @xmath148 for all @xmath119 by equations [ eq : bishop1a]-[eq : bishop1c ] , where @xmath142 and @xmath143 . specifically , @xmath149 where @xmath150 @xmath151 and @xmath152 in the equations above , we have assumed that the values of all the entries of @xmath115 and @xmath116 have previously been determined from the element of the nd parameter space at hand and the constraints c1-c3 above . furthermore , note that @xmath148 is regular because , by definition , @xmath115 is positive definite . clearly , @xmath148 can be rewritten as a regular gaussian distribution @xmath153 : it suffices to take @xmath154 where @xmath155 is a matrix of zeroes of dimension @xmath156 . then , @xmath157 is a regular gaussian distribution by equations [ eq : bishop2a]-[eq : bishop2c ] . likewise , @xmath158 is a regular gaussian distribution . continuing with this process for the rest of connectivity components proves that @xmath159 is mapped into some regular gaussian distribution @xmath147 . * step 2 * we now show that @xmath160 . note that for all @xmath161 and any fixed value of @xmath162 @xmath163 @xmath164 @xmath165 \ldots ] d{x_{b_{i+2}}}]d{x_{b_{i+1 } } } = 1.\ ] ] thus , for all @xmath119 , it follows from the equation above that @xmath166 @xmath167 [ \int \prod_{l = i+1}^n q^l(x_{b_l } | x_{pa(b_l ) } ) d{x_{b_{i+1 } \ldots b_n } } ] d{x_{b_1 \ldots b_{i-1 } \setminus pa(b_i)}}\ ] ] @xmath168 @xmath169 moreover , for all @xmath119 , it follows from the equation above that @xmath170 @xmath171 d{x_{b_i}}\ ] ] @xmath172 d{x_{b_1 \ldots b_{i-1 } \setminus pa(b_i)}}\ ] ] @xmath173 d{x_{b_1 \ldots b_{i-1 } \setminus pa(b_i)}}\ ] ] @xmath174 note the use of fubini s theorem to change the order of integration and produce equation [ eq : fub1 ] . this implies that the inner integral in equation [ eq : fub2 ] becomes 1 . consequently , for all @xmath119 @xmath175 due to equations [ eq : sum1 ] and [ eq : sum2 ] . therefore , @xmath176 and , thus , @xmath147 satisfies the condition f1 above . moreover , @xmath177 satisfies the condition f2 for all @xmath119 . we show this by induction on @xmath65 . let @xmath178 denote the precision matrix of @xmath177 , and note that @xmath179 by equation [ eq : equalconditionals ] . so , @xmath178 can be calculated from @xmath148 and @xmath180 via equation [ eq : bishop2b ] . specifically , it follows from equations [ eq : bishop2b ] and [ eq : epsilon ] , respectively [ eq : delta ] , that @xmath181 and @xmath182=\omega^i_{b_i , pa(b_i)}.\ ] ] consequently , due to the constraints c2 and c3 above , @xmath183 for all @xmath184 such that @xmath66 and @xmath126 are not adjacent in @xmath127 . moreover , @xmath183 is equivalent to @xmath185 ( * ? ? ? * proposition 5.2 ) . this implies that @xmath177 factorizes with respect to @xmath127 and , thus , that it satisfies the condition f2 above ( * ? ? ? * proposition 3.30 , theorems 3.34 and 3.36 ) . consequently , @xmath160 . * step 3 * we finally show that different elements of the nd parameter space for @xmath113 are mapped into different probability distributions in @xmath113 . assume to the contrary that two distinct elements of the nd parameter space for @xmath113 are mapped into the same probability distribution @xmath160 . assume that the two elements differ in the value for @xmath186 , @xmath115 or @xmath116 but that they coincide in the values for @xmath187 , @xmath188 and @xmath189 for all @xmath190 . there are two scenarios to consider : * if the two elements differ in the value for @xmath115 or @xmath116 , then they are mapped into two different @xmath177 by equations [ eq : lambda1 ] and [ eq : lambda2 ] , because two regular gaussian distributions with different precision matrices are different . however , this contradicts the assumption that the two elements are mapped into the same @xmath147 . * if the two elements differ in the value for @xmath186 but they do not differ in the values for @xmath115 and @xmath116 , then the two elements do not differ in the value for @xmath191 either , because @xmath192 and we assumed above that the two elements coincide in the values for @xmath187 for all @xmath190 . then , the two elements are mapped into the same @xmath193 but different @xmath194 in equations [ eq : delta ] and [ eq : gamma ] . that is , the two elements are mapped into two different @xmath148 and , thus , to two different @xmath195 by equation [ eq : equalconditionals ] . however , this contradicts the assumption that the two elements are mapped into the same @xmath147 . [ rem : polynomial ] note the following three observations : * for all @xmath119 , according to the constraints c1-c3 above , every entry of @xmath115 and @xmath116 is equal either to zero or to some nd parameter in the parameterization of the probability distributions in @xmath113 . * for all @xmath119 , by remark [ rem : inverse ] , every entry of @xmath196 is a fraction of real polynomials in the entries of @xmath115 and , thus , a fraction of real polynomials in the nd parameters in the parameterization of the probability distributions in @xmath113 . thus , every entry of the matrices @xmath193 and @xmath197 in equations [ eq : delta ] and [ eq : epsilon ] is also a fraction of real polynomials in the referred nd parameters . * every entry of the precision matrix of @xmath157 in the proof above is , by equation [ eq : bishop2b ] , a real polynomial in the entries of @xmath198 , @xmath199 and the precision matrix of @xmath200 , i.e. @xmath201 . likewise , every entry of the precision matrix of @xmath158 in the proof above is a real polynomial in the entries of @xmath202 , @xmath203 and the precision matrix of @xmath157 , that is , a real polynomial in the entries of @xmath202 , @xmath203 , @xmath198 , @xmath199 and @xmath201 . continuing with this process for the rest of connectivity components shows that every entry of the precision matrix of @xmath204 in the proof above is a real polynomial in the entries of the matrices @xmath201 , and @xmath193 and @xmath197 for all @xmath205 . it follows from the observations above that every entry of the precision matrix of @xmath147 in the proof above is a fraction of real polynomials in the nd parameters in the parameterization of the probability distributions in @xmath113 . consequently , by remark [ rem : inverse ] , every entry of the covariance matrix of @xmath147 is a fraction of real polynomials in the nd parameters in the parameterization of the probability distributions in @xmath113 . moreover , note the following two observations on the latter fractions : * each of these fractions is defined on the whole nd parameter space for @xmath113 : the polynomial in the denominator of the fraction is non - vanishing in the nd parameter space for @xmath113 because , as we have proven in step 1 in the theorem above , @xmath147 is a gaussian distribution . * within the nd parameter space for @xmath113 , each of these fractions vanishes only in the points where the polynomial in the numerator of the fraction vanishes because , as we have just seen , the denominator of the fraction is non - vanishing in the nd parameter space for @xmath113 . we now prove another result that will be crucial in the coming section . [ lem : spacecg ] let @xmath0 be a cg of dimension @xmath1 . the nd parameter space for @xmath113 has positive lebesgue measure with respect to @xmath2 . since we do not know a closed - form expression of the nd parameter space for @xmath113 , we take an indirect approach to prove the lemma . recall that , by definition , the nd parameter space for @xmath113 is the set of real values such that , after the extension determined by the constraints c1 and c2 , @xmath115 is positive definite for all @xmath119 . therefore , all the nd parameters except those in @xmath115 for all @xmath119 can take values independently of the rest of the nd parameters . the nd parameters in @xmath115 can not take values independently one of another because , otherwise , @xmath115 may not be positive definite . however , if the entries in the diagonal of @xmath115 take values in @xmath206 and the rest of the nd parameters in @xmath115 take values in @xmath207 $ ] , then the nd parameters in @xmath115 can take values independently one of another . to see it , note that in this case @xmath115 will always be hermitian , strictly diagonally dominant , and with strictly positive diagonal entries , which implies that @xmath115 will always be positive definite ( * ? ? ? * corollary 7.2.3 ) . the subset of the nd parameter space of @xmath113 described in the paragraph above has positive volume in @xmath2 and , thus , it has positive lebesgue measure with respect to @xmath2 . then , the nd parameter space of @xmath113 has positive lebesgue measure with respect to @xmath2 . the two theorems below are the main contribution of this manuscript . they prove that for any cg @xmath0 , in the measure - theoretic sense described below , almost all the probability distributions in @xmath113 are faithful to @xmath0 . [ the : f1 ] let @xmath0 be a cg of dimension @xmath1 . @xmath113 has positive lebesgue measure with respect to @xmath2 . the one - to - one correspondence proved in lemma [ lem:121cg ] enables us to compute the lebesgue measure with respect to @xmath2 of @xmath113 as the lebesgue measure with respect to @xmath2 of the nd parameter space for @xmath113 . moreover , the latter is positive by lemma [ lem : spacecg ] . before proving the second theorem , some auxiliary lemmas are proven . [ lem : subset ] let @xmath0 and @xmath208 be two cgs such that the undirected ( resp . directed ) edges in @xmath208 are a subset of the undirected ( resp . directed ) edges in @xmath0 . then , @xmath209 . note that a regular gaussian distribution factorizes with respect to a cg iff it is markovian with respect to the cg ( * ? ? ? * proposition 3.30 , theorems 3.34 and 3.36 ) . then , @xmath209 because the independence model represented by @xmath208 is a superset of that represented by @xmath0 . [ lem : chunk1 ] let @xmath0 be a cg such that 1 . @xmath0 has a route between the nodes @xmath65 and @xmath66 that has no collider section , and 2 . the route has no node in @xmath210 . then , there exists a probability distribution @xmath114 such that @xmath211 . the route in the lemma can be converted into a path @xmath39 between @xmath65 and @xmath66 in @xmath0 as follows : iteratively , remove from the route any subroute between a node and itself . note that none of these removals produces a collider section : it suffices to note that if the route after the removal has a collider section , then the route before the removal must have a collider section , which is a contradiction . consequently , @xmath39 is a path between @xmath65 and @xmath66 in @xmath0 that has no collider section . therefore , @xmath39 is superactive with respect to @xmath212 : since the route in the lemma has no node in @xmath212 , @xmath39 has no node in @xmath212 either . now , remove from @xmath0 all the edges that are not in @xmath39 , and call the resulting cg @xmath208 . note that @xmath208 has no complex since @xmath39 has no collider section . drop the direction of every edge in @xmath208 and call the resulting ug @xmath83 . now , note that there exists a regular gaussian distribution @xmath3 that is faithful to @xmath83 ( * ? ? ? * corollary 3 ) and , thus , @xmath211 because @xmath213 . note also that the fact that @xmath3 is faithful to @xmath83 implies that @xmath3 is markovian with respect to @xmath83 which , in turn , implies that @xmath3 is also markovian with respect to @xmath208 , because @xmath208 and @xmath83 have the same underlying ug and complexes ( * ? ? ? * theorem 5.6 ) . consequently , @xmath214 ( * ? ? ? * proposition 3.30 , theorems 3.34 and 3.36 ) and , thus , @xmath114 because @xmath209 by lemma [ lem : subset ] . [ lem : polynomial ] let @xmath0 be a cg . for every @xmath69 and @xmath210 , there exists a real polynomial @xmath215 in the nd parameters in the parameterization of the probability distributions in @xmath113 such that , for every @xmath114 , @xmath216 iff @xmath215 vanishes for the nd parameter values coding @xmath3 . let @xmath79 denote the covariance matrix of @xmath3 . note that @xmath216 iff @xmath217 ( * ? ? ? * proposition 5.2 ) . recall that @xmath218 with @xmath219 . note that @xmath220 because @xmath221 is positive definite @xcite . then , @xmath216 iff @xmath222 . thus , @xmath216 iff a real polynomial @xmath223 in the entries of @xmath79 vanishes due to remark [ rem : inverse ] . however , note that it follows from lemma [ lem:121cg ] and remark [ rem : polynomial ] that each entry of @xmath79 is a fraction of real polynomials in the nd parameters in the parameterization of the probability distributions in @xmath113 . recall also from remark [ rem : polynomial ] that the polynomial in the denominator of each of these fractions is non - vanishing in the nd parameter space for @xmath113 . therefore , by simple algebraic manipulation , the polynomial @xmath223 can be expressed as a fraction @xmath224 of real polynomials in the nd parameters where @xmath225 is non - vanishing in the nd parameter space for @xmath113 . consequently , @xmath216 iff the real polynomial @xmath215 in the nd parameters vanishes for the values coding @xmath3 . we interpret the polynomial in the lemma above as a real function on a real euclidean space that includes the nd parameter space for @xmath113 . we say that the polynomial in the lemma above is non - trivial if not all the values of the nd parameters are solutions to the polynomial . this is equivalent to the requirement that the polynomial is not identically zero , because the nd parameter space for @xmath113 contains a @xmath1-dimensional interval in @xmath2 , where @xmath1 is the dimension of @xmath0 ( recall the proof of lemma [ lem : spacecg ] ) . let @xmath226 denote an undirected route @xmath40 in a cg . hereinafter , we denote by @xmath227 the route @xmath228 . [ lem : chunk2 ] let @xmath0 be a cg such that 1 . @xmath0 has a route @xmath229 where @xmath69 and @xmath226 is an undirected route , and 2 . some node in @xmath226 is in @xmath212 or has a descendant in @xmath212 , where @xmath210 . then , there exists a probability distribution @xmath114 such that @xmath211 . the route @xmath226 can be converted into a path @xmath230 in @xmath0 as follows : iteratively , remove from @xmath226 any subroute between a node and itself . note that @xmath226 does not contain either @xmath65 or @xmath66 because , otherwise , @xmath0 would have a directed pseudocycle between @xmath65 and itself or between @xmath66 and itself , which is a contradiction . therefore , @xmath230 does not contain either @xmath65 or @xmath66 and , thus , @xmath231 is a path in @xmath0 . note that the subroutes removed from @xmath226 contain only undirected edges . therefore , every node that is in @xmath226 but not in @xmath230 is a descendant of some node in @xmath230 . consequently , some node in @xmath230 is in @xmath212 or has a descendant in @xmath212 , due to the assumptions in the lemma . we first prove the lemma for the case where some node in @xmath230 is in @xmath212 . remove from @xmath0 all the edges that are not in @xmath231 , and call the resulting cg @xmath208 . note that @xmath231 is a complex in @xmath208 and , thus , that @xmath232 . let @xmath126 denote the closest node to @xmath65 that it is in @xmath230 and in @xmath212 . we prove in this paragraph that there exists a probability distribution @xmath214 such that @xmath233 . by lemma [ lem : polynomial ] , there exists a real polynomial @xmath234 in the nd parameters in the parameterization of the probability distributions in @xmath235 such that , for every @xmath236 , @xmath237 iff @xmath234 vanishes for the nd parameter values coding @xmath147 . furthermore , @xmath234 is non - trivial . to see this , remove from @xmath208 all the edges outside the path between @xmath65 and @xmath126 , and call the resulting cg @xmath83 . note that @xmath238 by lemma [ lem : subset ] . now note that , by lemma [ lem : chunk1 ] , there exists a probability distribution @xmath239 such that @xmath240 . by an analogous reasoning , we can conclude that there exists a non - trivial real polynomial @xmath241 in the nd parameters in the parameterization of the probability distributions in @xmath235 such that , for every @xmath236 , @xmath242 iff @xmath241 vanishes for the nd parameter values coding @xmath147 . let @xmath243 and @xmath244 denote the sets of solutions to the polynomials @xmath234 and @xmath241 , respectively . let @xmath1 denote the dimension of @xmath208 . then , @xmath243 and @xmath244 have both zero lebesgue measure with respect to @xmath2 because they consist of the solutions to non - trivial real polynomials in real variables ( the nd parameters ) @xcite . then , @xmath245 also has zero lebesgue measure with respect to @xmath2 , because the finite union of sets of zero lebesgue measure has zero lebesgue measure too . consequently , the probability distributions @xmath236 such that @xmath237 or @xmath242 correspond to a set of elements of the nd parameter space for @xmath235 that has zero lebesgue measure with respect to @xmath2 because it is contained in @xmath246 . since this correspondence is one - to - one by lemma [ lem:121cg ] , the probability distributions @xmath236 such that @xmath237 or @xmath242 also have zero lebesgue measure with respect to @xmath2 . this result together with theorem [ the : f1 ] imply that there exists a probability distribution @xmath214 such that @xmath247 and @xmath248 . furthermore , as shown above @xmath232 and , thus , @xmath249 because @xmath3 is markovian with respect to @xmath208 , since @xmath214 ( * ? ? ? * proposition 3.30 , theorems 3.34 and 3.36 ) . then , @xmath250 by symmetry and weak transitivity and , thus , @xmath233 by decomposition . finally , recall that @xmath126 is the closest node to @xmath65 that it is in @xmath230 and in @xmath212 , then @xmath251 and thus @xmath252 because @xmath3 is markovian with respect to @xmath208 . then , @xmath211 by intersection on @xmath233 and @xmath252 . consequently , we have proven that there exists a probability distribution @xmath214 such that @xmath211 . moreover , @xmath114 because @xmath209 by lemma [ lem : subset ] . we now prove the lemma for the case where no node in @xmath230 is in @xmath212 but some node in @xmath230 has a descendant in @xmath212 . consider the shortest descending path between a node in @xmath230 and a node in @xmath212 . let @xmath253 and @xmath126 denote the initial and final nodes of the path , i.e. @xmath254 . remove from @xmath0 all the edges that are not in @xmath231 or in the path between @xmath253 and @xmath126 , and call the resulting cg @xmath208 . note that @xmath231 is a complex in @xmath208 and , thus , that @xmath232 . therefore , we can follow the same steps as above to prove that there exists a probability distribution @xmath214 such that @xmath233 . finally , recall that there is no path between @xmath65 and any node in @xmath255 in @xmath208 , then @xmath251 and thus @xmath252 because @xmath3 is markovian with respect to @xmath208 . then , @xmath211 by intersection on @xmath233 and @xmath252 . consequently , we have proven that there exists a probability distribution @xmath214 such that @xmath211 . moreover , @xmath114 because @xmath209 by lemma [ lem : subset ] . [ lem : non - trivial ] let @xmath0 be a cg such that @xmath256 , where @xmath69 and @xmath210 . then , there exists a probability distribution @xmath114 such that @xmath211 . we prove the lemma in two steps . in the first step , we introduce some notation that we use in the second step , the actual proof of the lemma . * step 1 * given a route @xmath39 in a cg @xmath208 , we define @xmath257 as the cg resulting from removing from @xmath208 all the edges that are not in @xmath39 . we define the level of a node in @xmath208 as the index of the connectivity component the node belongs to . we define the dlength of a route as the number of distinct edges in the route . note the difference between the dlength and the length of a route : the former counts edges without repetition and the latter with repetition ( recall section [ sec : preliminaries ] ) . we say that a route is dshorter than another route if the former has smaller dlength than the latter . likewise , we say that a route is dshortest if no other route is dshorter than it . let @xmath258 denote any total order of the nodes in the cg @xmath208 . let @xmath259 denote any total order of all the routes between two nodes in @xmath208 . finally , if @xmath260 where @xmath261 and @xmath262 , then we define @xmath263 as follows : * if there is a route in @xmath208 like that in lemma [ lem : chunk1 ] or [ lem : chunk2 ] for @xmath264 , @xmath265 and @xmath266 , then we define @xmath267 . * otherwise , we define recursively @xmath268 , where @xmath39 and @xmath126 are selected as follows . let @xmath269 denote the set of routes between @xmath270 and @xmath271 in @xmath208 that are superactive with respect to @xmath272 . let @xmath273 denote the dshortest routes in @xmath269 . let @xmath274 denote the shortest routes in @xmath273 . let @xmath39 denote the route in @xmath274 that comes first in @xmath259 . we call @xmath39 the splitting route . furthermore , let @xmath42 denote the set of nodes in @xmath39 but not in @xmath275 that have minimal level in @xmath257 . let @xmath126 denote the node in @xmath42 that comes first in @xmath258 . note that the only point with @xmath259 and @xmath258 is to select @xmath39 and @xmath126 unambiguously . note that we have implicitly assumed in the definition s2 that @xmath42 is non - empty . we now prove that this is always true . assume to the contrary that @xmath42 is empty . this means that all the nodes in @xmath39 are in @xmath275 . since the definition s1 did not apply , @xmath39 must have some collider section @xmath226 . moreover , @xmath276 is a subroute of @xmath39 : if @xmath277 ( resp . @xmath278 ) then @xmath279 ( resp . @xmath280 ) must be outside @xmath272 for @xmath39 to be superactive with respect to @xmath272 , which contradicts the assumption that all the nodes in @xmath39 are in @xmath275 . moreover , some node in @xmath226 must be in @xmath272 for @xmath39 to be superactive with respect to @xmath272 . however , this implies that @xmath281 is a route that satisfies the requirements of the definition s1 , which is a contradiction . finally , we prove that @xmath282 and @xmath283 in the definition s2 are well - defined . let @xmath284 denote the subroute of @xmath39 between the first occurrences of @xmath270 and @xmath126 in @xmath39 when going from @xmath270 to @xmath271 . note that if @xmath39 contains @xmath126 only in non - collider sections , then none of the other nodes in those sections can be in @xmath272 for @xmath39 to be superactive with respect to @xmath272 and , thus , @xmath284 is a route between @xmath270 and @xmath126 in @xmath257 that is superactive with respect to @xmath272 and , thus , @xmath285 and , thus , @xmath282 is defined . we now prove that @xmath39 contains @xmath126 only in non - collider sections . assume the contrary and let @xmath226 denote any collider section of @xmath39 that contains @xmath126 . note that @xmath276 is a subroute of @xmath39 , because if @xmath277 or @xmath278 then there exists a node in @xmath39 but not in @xmath275 with smaller level than @xmath126 in @xmath257 , which is a contradiction . moreover , some node in @xmath226 must be in @xmath272 for @xmath39 to be superactive with respect to @xmath272 . however , this implies that @xmath281 is a route that satisfies the requirements of the definition s1 , which is a contradiction . now , let @xmath286 denote the subroute of @xmath39 between the first occurrences of @xmath271 and @xmath126 in @xmath39 when going from @xmath271 to @xmath270 . by repeating the reasoning above with @xmath286 instead of @xmath284 , we can conclude that @xmath287 and , thus , that @xmath283 is defined too . moreover , note that @xmath284 and @xmath286 have dlength equal or smaller than @xmath39 and length strictly smaller than @xmath39 . therefore , the splitting routes for @xmath282 and @xmath283 are each either dshorter or shorter than @xmath39 . this guarantees that the recursive definition s2 eventually reaches the trivial case s1 . * step 2 * we prove the lemma by induction over the value of @xmath288 . if @xmath289 , then there exists a route in @xmath0 like that in lemma [ lem : chunk1 ] or [ lem : chunk2 ] . therefore , there exists a probability distribution @xmath114 such that @xmath211 by lemma [ lem : chunk1 ] or [ lem : chunk2 ] . assume as induction hypothesis that the lemma holds for any value of @xmath288 smaller than @xmath290 ( @xmath291 ) . we now prove it for value @xmath290 . recall that @xmath292 where @xmath39 is a dshortest route among all the routes between @xmath65 and @xmath66 in @xmath0 that are superactive with respect to @xmath212 , and @xmath126 is a node in @xmath39 but not in @xmath293 that has minimal level in @xmath294 . then , as shown in step 1 , @xmath295 and @xmath296 . moreover , @xmath297 and @xmath298 are both smaller than @xmath290 . then , by the induction hypothesis , there exist two probability distributions @xmath299 such that @xmath300 and @xmath301 . we prove below that there exists a probability distribution @xmath302 such that @xmath211 . note that @xmath114 because @xmath303 by lemma [ lem : subset ] . by lemma [ lem : polynomial ] , there exists a real polynomial @xmath304 in the nd parameters in the parameterization of the probability distributions in @xmath305 such that , for every @xmath306 , @xmath307 iff @xmath304 vanishes for the nd parameter values coding @xmath147 . furthermore , @xmath304 is non - trivial due to the probability distribution @xmath308 above . similarly , there exists a real polynomial @xmath309 in the nd parameters in the parameterization of the probability distributions in @xmath305 such that , for every @xmath306 , @xmath310 iff @xmath309 vanishes for the nd parameter values coding @xmath147 . furthermore , @xmath309 is also non - trivial due to the probability distribution @xmath311 above . let @xmath312 and @xmath313 denote the sets of solutions to the polynomials @xmath304 and @xmath309 , respectively . let @xmath1 denote the dimension of @xmath294 . then , @xmath312 and @xmath313 have both zero lebesgue measure with respect to @xmath2 because they consist of the solutions to non - trivial real polynomials in real variables ( the nd parameters ) @xcite . then , @xmath314 also has zero lebesgue measure with respect to @xmath2 , because the finite union of sets of zero lebesgue measure has zero lebesgue measure too . consequently , the probability distributions @xmath306 such that @xmath307 or @xmath310 correspond to a set of elements of the nd parameter space for @xmath305 that has zero lebesgue measure with respect to @xmath2 because it is contained in @xmath246 . since this correspondence is one - to - one by lemma [ lem:121cg ] , the probability distributions @xmath306 such that @xmath307 or @xmath310 also have zero lebesgue measure with respect to @xmath2 . this result together with theorem [ the : f1 ] imply that there exists a probability distribution @xmath302 such that @xmath315 and @xmath316 . note that these two independence statements together with @xmath317 would imply the desired result by symmetry and weak transitivity . we prove below @xmath318 which , in turn , implies @xmath317 because @xmath3 is markovian with respect to @xmath294 , since @xmath302 ( * ? ? ? * proposition 3.30 , theorems 3.34 and 3.36 ) . assume to the contrary @xmath319 . let @xmath284 denote any route between @xmath65 and @xmath66 in @xmath294 that is superactive with respect to @xmath320 . note that @xmath284 must contain @xmath126 because , otherwise , @xmath284 would be a route between @xmath65 and @xmath66 in @xmath0 that is superactive with respect to @xmath212 and that is dshorter than @xmath39 , which is a contradiction . furthermore , @xmath284 must contain @xmath126 only in collider sections because , otherwise , @xmath284 would not be superactive with respect to @xmath320 . let @xmath226 denote any collider section of @xmath284 that contains @xmath126 . note that @xmath321 is a subroute of @xmath284 , because if @xmath322 or @xmath323 then there exists a node in @xmath284 but not in @xmath293 with smaller level than @xmath126 in @xmath294 . since @xmath284 is a route in @xmath294 , this implies that there exists a node in @xmath39 but not in @xmath293 with smaller level than @xmath126 in @xmath294 , which is a contradiction . note also that no descendant of @xmath126 in @xmath0 can be in @xmath212 because , otherwise , @xmath229 would be a route that satisfies the requirements of the definition s1 , which is a contradiction . however , if no descendant of @xmath126 in @xmath0 is in @xmath212 , then @xmath39 must contain @xmath126 only in non - collider sections because , otherwise , @xmath39 would not be superactive with respect to @xmath212 . the last two observations imply that @xmath65 or @xmath66 is a descendant of @xmath126 in @xmath0 which , together with @xmath229 , implies that @xmath0 has a directed pseudocycle between @xmath65 and itself or between @xmath66 and itself , because @xmath226 contains @xmath126 . this is a contradiction . [ the : f2 ] let @xmath0 be a cg of dimension @xmath1 . the set of probability distributions in @xmath113 that are not faithful to @xmath0 has zero lebesgue measure with respect to @xmath2 . note that the probability distributions in @xmath113 are markovian with respect to @xmath0 ( * ? ? ? * proposition 3.30 , theorems 3.34 and 3.36 ) . then , for any probability distribution @xmath114 not to be faithful to @xmath0 , @xmath3 must satisfy some independence that is not entailed by @xmath0 . that is , there must exist three disjoint subsets of @xmath6 , here denoted as @xmath36 , @xmath49 and @xmath212 , such that @xmath324 but @xmath325 . however , if @xmath324 then @xmath256 for some @xmath326 and @xmath327 . furthermore , if @xmath325 then @xmath216 by symmetry and decomposition . by lemma [ lem : polynomial ] , there exists a real polynomial @xmath215 in the nd parameters in the parameterization of the probability distributions in @xmath113 such that , for every @xmath160 , @xmath328 iff @xmath215 vanishes for the nd parameter values coding @xmath147 . furthermore , @xmath215 is non - trivial by lemma [ lem : non - trivial ] . let @xmath329 denote the set of solutions to the polynomial @xmath215 . then , @xmath329 has zero lebesgue measure with respect to @xmath2 because it consists of the solutions to a non - trivial real polynomial in real variables ( the nd parameters ) @xcite . then , @xmath330 has zero lebesgue measure with respect to @xmath2 , because the finite union of sets of zero lebesgue measure has zero lebesgue measure too . consequently , the probability distributions in @xmath113 that are not faithful to @xmath0 correspond to a set of elements of the nd parameter space for @xmath113 that has zero lebesgue measure with respect to @xmath2 because it is contained in @xmath246 . since this correspondence is one - to - one by lemma [ lem:121cg ] , the probability distributions in @xmath113 that are not faithful to @xmath0 also have zero lebesgue measure with respect to @xmath2 . the following corollary , which follows trivially from theorems [ the : f1 ] and [ the : f2 ] , summarizes the results in this section . [ co : f3 ] let @xmath0 be a cg of dimension @xmath1 . the set of probability distributions in @xmath113 that are faithful to @xmath0 has positive lebesgue measure with respect to @xmath2 . the space of cgs can be divided in classes of equivalent cgs according to criteria such as markov independence equivalence , markovian distribution equivalence or factorization equivalence . as we prove below with the help of the theorems above , these criteria actually coincide . this result is important because the classes of markovian distribution equivalent cgs have a simple graphical characterization and a natural representative , the so - called largest cg , which now also apply to the classes of equivalence induced by the other two criteria mentioned . we also prove below that all equivalent cgs have the same dimension with respect to the parameterization introduced in section [ sec : faithfulness ] . before proving our results , we formally define the equivalence criteria discussed in the paragraph above . recall that , unless otherwise stated , all the probability distributions in this paper are defined on ( state space ) @xmath48 , where @xmath11 . we say that two cgs are markov independence equivalent if they represent the same independence model . we say that two cgs are markovian distribution equivalent if every regular gaussian distribution is markovian with respect to both cgs or with respect to neither of them . we say that two cgs @xmath0 and @xmath208 are factorization equivalent if @xmath331 . the corollary below proves that these definitions coincide . [ co : me ] let @xmath0 and @xmath208 denote two cgs . the following statements are equivalent in the frame of regular gaussian distributions : 1 . @xmath0 and @xmath208 are factorization equivalent . 2 . @xmath0 and @xmath208 are markovian distribution equivalent . 3 . @xmath0 and @xmath208 are markov independence equivalent . the equivalence of statements 1 and 2 follows from ( * ? ? ? * proposition 3.30 , theorems 3.34 and 3.36 ) . we now prove that statements 2 and 3 are equivalent . by definition , markov independence equivalence implies markovian distribution equivalence . to see the opposite implication , note that if @xmath0 and @xmath208 are not markov independence equivalent , then one of them , say @xmath0 , must represent a separation statement @xmath53 that is not represented by @xmath208 . consider a probability distribution @xmath214 faithful to @xmath208 . such a probability distribution exists due to corollary [ co : f3 ] , and it is markovian with respect to @xmath208 . however , @xmath3 can not be markovian with respect to @xmath0 , because @xmath332 implies @xmath55 . * theorem 5.6 ) gives a straightforward graphical characterization of markovian distribution equivalence : two cgs are markovian distribution equivalent iff they have the same underlying ug and the same complexes . due to the corollary above , that is also a graphical characterization of the other two types of equivalence discussed there . hereinafter , we do not distinguish anymore between the different types of equivalence discussed in the corollary above because they coincide and , thus , we simply refer to them as equivalence . * proposition 5.7 ) shows that every class of equivalent cgs contains a unique cg that has more undirected edges than any other cg in the class . such a cg is called the largest cg ( lcg ) in the class , and it is usually considered a natural representative of the class . * section 4.2 ) conjectures that , for discrete probability distributions , the lcg in a class of equivalent cgs has fewer nd parameters than any other cg in the class . this would imply that the most space efficient way of storing the discrete probability distributions that factorize with respect to a class of equivalent cgs is by factorizing them with respect to the lcg in the class rather than with respect to any other cg in the class . the corollary below proves that an analogous conjecture for regular gaussian distributions and the parameterization of them proposed in section [ sec : faithfulness ] would be false . [ co : lcg ] all equivalent cgs have the same dimension with respect to the parameterization proposed in section [ sec : faithfulness ] . let @xmath0 denote the lcg in a class of equivalent cgs . let @xmath208 denote any other cg in the class . recall that the dimensions of @xmath0 and @xmath208 with respect to the parameterization proposed in section [ sec : faithfulness ] are , respectively , @xmath333 and @xmath334 . note that @xmath208 can be obtained from @xmath0 by orienting some of the undirected edges in @xmath0 ( * ? ? ? * theorem 3.9 ) . then , @xmath335 and , thus , @xmath336 . in this paper , we have proven that , in certain measure - theoretic sense , almost all the regular gaussian distributions that factorize with respect to a chain graph are faithful to it . this result extends previous results such as * ( * ? ? ? * theorem 3.2 ) where it is proven that , in certain measure - theoretic sense , almost all the regular gaussian distributions that factorize with respect to an acyclic directed graph are faithful to it , and * ( * ? ? ? * corollary 3 ) where it is proven that for any undirected graph there exists a regular gaussian distribution that is faithful to it . there are a number of consequences that follow from the result proven in this paper : * there are independence models that can be represented exactly by chain graphs but that can not be represented exactly by undirected graphs or acyclic directed graphs . as a matter of fact , the experimental results in @xcite suggest that this may be the case for the vast majority of independence models that can be represented exactly by chain graphs . this is an advantage of chain graphs when dealing with regular gaussian distributions , because there exists a regular gaussian distribution that is faithful to each of these independence models . * the moralization and c - separation criteria for reading independencies holding in the regular gaussian distributions that factorize with respect to a chain graph are complete ( i.e. they identify all the independencies that can be identified on the sole basis of the chain graph ) , because there exists a regular gaussian distribution that is faithful to the chain graph . * some definitions of equivalence in chain graphs coincide , which implies that the graphical characterization of markovian distribution equivalence in ( * ? ? ? * theorem 5.6 ) also applies to other definitions of equivalence . * for the parameterization introduced in this paper , all the chain graphs in a class of equivalence have the same dimension and , thus , their factorizations are equally space efficient for storing the regular gaussian distribution that factorize with respect to the chain graphs in the class . this work is funded by the swedish research council ( ref . vr-621 - 2005 - 4202 ) and ceniit at linkping university ( ref . we are thankful to prof . jose a. lozano for discussions on this work . in this appendix , we derive equations [ eq : bishop1a]-[eq : bishop2c ] . our derivations are adaptations of those in ( * ? ? ? * sections 2.3.1 , 2.3.3 ) to the notation used in this paper . a gaussian distribution for @xmath337 can be written as @xmath338 where @xmath78 is a @xmath137-dimensional mean vector , @xmath339 a @xmath340-dimensional covariance matrix , and @xmath126 a normalization constant . note that @xmath341\ ] ] @xmath342 where @xmath343 is a constant , i.e. it is independent of @xmath344 . in the last equality above we have used the fact that @xmath345 because @xmath90 is symmetric . then , a gaussian distribution for @xmath337 can be written as @xmath346 where @xmath347 is a normalization constant . let @xmath36 and @xmath49 denote two disjoint subsets of @xmath6 . let @xmath89 where @xmath90 is positive definite . if we regard @xmath348 as a constant , then @xmath349 @xmath350\ ] ] @xmath351 @xmath352 + k ' '' = - \frac{1}{2 } x_j^t \omega_{j , j } x_j + x_j^t [ \omega_{j , j } \mu_j - \omega_{j , i } ( x_i-\mu_i ) ] + k'''\ ] ] where @xmath353 is a constant , i.e. it is independent of @xmath354 . in the last equality above we have used a reasoning analogous to that in equation [ eq : ap1 ] . then , @xmath355}}{k''''}\ ] ] where @xmath356 is a normalization constant , because @xmath348 can be regarded as a constant in @xmath357 since it is the value of the conditioning set . consequently , @xmath357 is a gaussian distribution since it can be written in the form given in equation [ eq : ap2 ] . by equating the term that is quadratic in @xmath337 in equation [ eq : ap2 ] with the term that is quadratic in @xmath51 in equation [ eq : ap3 ] , we conclude that the covariance matrix of @xmath357 is @xmath358 . by equating the term that is linear in @xmath337 in equation [ eq : ap2 ] with the term that is linear in @xmath51 in equation [ eq : ap3 ] , we conclude that the mean vector of @xmath357 is @xmath359 = \mu_j - ( \omega_{j , j})^{-1 } \omega_{j , i } ( x_i-\mu_i)\ ] ] @xmath360 therefore , @xmath91 where @xmath92 , @xmath93 and @xmath94 are the following real matrices of dimensions , respectively , @xmath95 , @xmath96 and @xmath97 : @xmath361 @xmath362 and @xmath363 now , let @xmath101 and @xmath102 where @xmath92 , @xmath93 and @xmath94 are real matrices of dimensions , respectively , @xmath95 , @xmath96 and @xmath97 , and @xmath103 and @xmath94 are positive definite . then , @xmath364}}{k}\ ] ] where @xmath126 is a normalization constant . note that @xmath365\ ] ] @xmath366 @xmath367 + k'\ ] ] @xmath368\ ] ] @xmath369 where @xmath343 is a constant , i.e. it is independent of @xmath370 . in the last equality above we have used a reasoning analogous to that in equation [ eq : ap1 ] . by using this reasoning further and reorganizing some terms we can rewrite the expression above as @xmath371\ ] ] @xmath372 @xmath373 then , @xmath374 can be expressed as @xmath375 where @xmath347 is a normalization constant . consequently , @xmath374 is a gaussian distribution over @xmath106 since it can be expressed in the form given in equation [ eq : ap2 ] . as we did above , the precision matrix ( resp . the mean vector ) of @xmath374 can easily be found by equating the term that is quadratic ( resp . linear ) in @xmath337 in equation [ eq : ap2 ] with the term that is quadratic ( resp . linear ) in @xmath106 in equation [ eq : ap4 ] . specifically , @xmath376 where @xmath377 and @xmath378 @xmath379 we have omitted the details of the derivation of @xmath380 from @xmath381 , but it can easily be checked that @xmath382 equals the identity matrix . note that @xmath381 is invertible because @xmath103 and @xmath94 are invertible and , thus , that @xmath374 is regular . jose m. pea . faithfulness in chain graphs : the discrete case . _ international journal of approximate reasoning _ , 50:1306 - 1313 , 2009 . peter spirtes , clark glymour , richard scheines . _ causation , prediction , and search_. springer - verlag , 1993 .

starburst galaxies are defined as galaxies that are currently undergoing a period of intense star - formation , mainly within ( but not exclusive to ) the nuclear region , at a rate that can not be maintained over their lifetime @xcite . the radio emission from such galaxies is not dominated by an active supermassive black hole , but by thermal and non - thermal emission that traces the star - forming activity ( e.g. @xcite ) . the thermal emission is the result of free - free ( or thermal bremsstrahlung ) radiation from gas ionised by hot , young stars , while the non - thermal emission is from synchrotron radiation due to the acceleration of electrons to relativistic speeds by the core - collapse of massive stars to form type ii or type ibc supernovae @xcite . interaction of the supernova with the local circumstellar medium generates a hot , shocked region that produces a prompt emission typically observed as a radio supernova ( rsn ) @xcite . as the rsn expand into the surrounding interstellar medium , they evolve into supernova remnants ( snrs ) , forming shells that emit non - thermal synchrotron radiation that can be visible over many years @xcite . studies of snrs within nearby starbust galaxies can provide important information on the astrophysical processes occurring within these galaxies . obscuration by dust and gas limit observations at shorter wavelengths , while radio wavelengths are largely unaffected . in addition , high - resolution , wide - field radio interferometry offers an opportunity to detect and resolve individual snrs ( e.g. @xcite ) . such observations over multiple epochs allow monitoring of the evolution of existing snrs and the detection of new rsne and/or snrs , enabling the supernova and star formation history of the galaxy to be reconstructed . such studies have been successfully applied to nearby starburst galaxies such as : m82 @xcite ; arp 220 @xcite ; arp 299 @xcite ; ngc 4945 @xcite ; and ngc 253 @xcite . these studies have provided a wealth of information concerning rsne and the evolution of snrs , the star - formation and supernova rates of the host galaxies , and the interstellar medium of the starburst region . for example , almost three decades of observations have identified @xmath5100 compact sources in m82 @xcite , where most have been resolved into parsec - scale shell - like structures @xcite . in addition , these multi - epoch observations have been instrumental in measuring the expansion speeds of the resolved snrs ( 2,000 - 11,000 kms@xmath2 ) , and with giving a direct estimate of the supernova rate ( 0.09 yr@xmath2 ) in m82 . the discovery of new sources such as sn2008iz @xcite and the radio transient 43.78 + 59.3 @xcite would not have been possible without regular radio monitoring of m82 . in addition , by combining multi - wavelength , high resolution radio observations of 46 compact sources in m82 , @xcite were able to identify low - frequency spectral turnovers , due to free - free absorption by the surrounding ionized medium . as one of the nearest star - forming galaxies , ngc 253 has been extensively studied from gamma - rays to radio wavelengths . recently , as part of the advanced camera for surveys ( acs ) nearby galaxy survey treasury ( angst ) , @xcite estimated the distance to ngc 253 as 3.47 @xmath6 0.24 mpc , 3.46 @xmath6 0.07 mpc and 3.40 @xmath6 0.09 mpc . the weighted average of the @xcite distance estimates , 3.44 @xmath6 0.13 mpc , is used as the distance to ngc 253 in this paper . the first high - resolution , wide - field radio interferometric observations of ngc 253 were made by ( * ? ? ? * hereafter th85 ) with the very large array ( vla ) at 15 ghz ( 2 cm ) . with a resolution of @xmath7 , th85 discovered nine compact sources ( designated th1 to th9 ) within the central nuclear region . following the results of th85 , ulvestad and antonucci conducted almost a decade of multi - frequency radio ( 1.8 , 5 , 8.3 , 15 and 23 ghz ) observations of the nuclear star - forming region of ngc 253 with the vla , that culminated in a series of papers @xcite ; they identified 64 individual compact sources , including the original nine from th85 . spectral index measurements were obtained for the 17 brightest sources between frequency pairs of 5/15 and 8.3/23 ghz . almost half of the 17 sources were identified with thermal hii regions while the remaining sources were taken to be associated with snrs . over the course of their survey , no new rsne or snrs appeared and the radio flux densities of the detected snrs were found to be stable ( * ? ? ? * hereafter ua97 ) . this led to an estimate for the upper limit on the supernova rate of 0.3 yr@xmath2 by ua97 . in an attempt to resolve the low frequency radio emission in the inner 300 pc , @xcite conducted the first wide - field vlbi observations of ngc 253 . the observations were carried out with the australian long baseline array ( lba ) at 1.4 ghz and matched the angular resolution of the ua97 23 ghz vla observations . while the observations detected only two sources ( th7 and th9 ) , @xcite showed that the radio emission at low frequencies is absorbed by ionized gas with a free - free optical depth range at 1 ghz of @xmath8 2.5 to @xmath9 8 . this result was consistent with observations between 0.33 - 1.5 ghz with the vla @xcite and radio - recombination line ( rrl ) modelling of the nuclear region of ngc 253 @xcite . motivated by these indicators of free - free absorption in ngc 253 , ( * ? ? ? * hereafter lt06 ) began a program to observe the nuclear region of this galaxy with the lba . their observations , conducted in 2004 at 2.3 ghz , were higher in sensitivity and resolution than @xcite and covered the region observed by ua97 . they identified six compact sources , which were also seen in the higher frequency observations with the vla ( ua97 ) . one of the sources , 5.48 - 43.3 was also resolved into a shell - like structure approximately 90 mas ( 1.7 pc ) in diameter . combining the lt06 data with the multi - wavelength radio data from ua97 , the spectra of 20 compact sources in the nuclear region of ngc 253 were found to be consistent with the free - free absorption interpretation of @xcite . the results indicated that while the free - free optical depth is highest towards the supposed nucleus , it varies significantly ( @xmath10 1 to @xmath9 20 ) throughout the nuclear region , implying variations in the gas density . of the 20 sources , eight were found to have flat spectral indices , indicative of thermal h ii regions , while the remaining sources were taken to be associated with snrs due to their steep spectral indices at high frequencies . with no new sources detected in ngc 253 over almost two decades , lt06 developed a monte - carlo method based on the work of @xcite , to estimate the upper limit on the supernova rate . their model took into consideration improved distance measurements , a median free - free opacity , and the sensitivity limits of six observations over a 17 yr period . a value of 2.4 yr@xmath2 was derived for the upper limit on the supernova rate . this high value suggests that there may be a large number of undetected rsne , with observations over that period only detecting the rare , bright events . detecting weaker or short - lived snrs would provide tighter constraints on the supernova rate . however , this can only be done through frequent , high sensitivity , high resolution observations of ngc 253 . in this paper , we present the results of multi - epoch , wide - field vlbi observations conducted at 2.3 ghz with the lba , of the nuclear region of ngc 253 . [ sec : obsall ] describes the observations , data analysis methods and the sources detected , including cross identifications with previous observations and investigations of possible flux density variations . free - free absorption modelling of the spectra is described in [ sec : cs ] . the morphology of the resolved snr , 5.48 - 43.3 , is examined in detail in [ sec:548 ] . an improvement to the monte - carlo method of lt06 is presented with new estimates of the upper limits on supernova and star - formation rates within the inner 300 pc region of ngc 253 in [ sec : sr ] @xmath11 [ sec : sfr ] . the results are summarised in [ sec : sum ] . [ cols="<,^,^,^",options="header " , ] ngc 253 was observed at 2.3 ghz with the australian long baseline array ( lba ) , at three epochs as described in [ tab : table1 ] . the observations were carried out with : the 64 m parkes ( pa ) antenna of the australia telescope national facility ( atnf ) ; the atnf australia telescope compact array ( atca ) 22 m antennas were used for the 2006 @xmath11 2008 epochs , while 4 @xmath12 22 m antennas were used for the 2007 epoch . ] , used as a phased array ; the atnf mopra ( mp ) 22 m antenna ; the university of tasmania s 26 m antenna near hobart ( ho ) ; and the university of tasmania s 30 m antenna near ceduna ( cd ) . in addition , the 70 m nasa deep space network antenna at tidbinbilla ( tid ) was used for the 2008 epoch . the data for each observation were recorded from 4 @xmath12 16 mhz bands ( digitally filtered 2-bit samples ) in the frequency ranges given in [ tab : table1 ] . all bands had dual circular polarization at parkes , atca , and mopra , with right circular polarization only at the remaining antennas . [ tab : table1 ] lists the parameters associated with the lba observations . during each observation , 3 minute scans of ngc 253 ( centred on : @xmath13 = 00@xmath1447@xmath15 33.178@xmath16 ; @xmath17 = @xmath025@xmath1817@xmath19 17@xmath20.060 [ j2000.0 ] ) were scheduled , alternating with 3 minute scans of a nearby phase reference calibration source , pks j0038 - 2459 ( @xmath13 = 00@xmath14 38@xmath15 14.735@xmath16 ; @xmath17 = @xmath024@xmath18 59@xmath19 02@xmath20.235 [ j2000.0 ] ) , located 2.13 degrees from the target . the recorded data for all epochs were correlated using the difx software correlator @xcite , with an integration time of 2 seconds and 64 frequency channels across each 16 mhz band ( channel widths of 0.25 mhz ) . the @xmath21 coverage for the 2007 and 2008 epochs are shown in [ fig : uvcov ] . the initial data reduction and calibration were performed using the data reduction package aips . prior to calibration , flagging of data during times at which each of the antennas were known to be slewing and time ranges that contained known bad data , were carried out via application of flag files and information provided by the observing logs . data from the first 30 seconds of each scan from baselines involving atca or parkes were flagged , in order to eliminate known corruption of the data at the start of each scan at these two telescopes . during correlation , nominal ( constant ) system temperatures ( in jansky ) for each antenna were applied to the correlation coefficients . the nominal calibration was refined by application of antenna system temperatures ( in kelvin ) measured during the observation , along with the gain ( in janskys per kelvin ) for each antenna . further refinements to the amplitude calibration were derived from simultaneously recorded atca observations of pks j0038 - 2459 , which is a bright , compact radio source , unresolved at this frequency on the lba baselines . since the lba is nt resolving out any extended structure , the atca and lba baselines will measure the same flux density . thus , the flux density measured at the atca can be used to check and refine the amplitude calibration for the lba data . suitable simultaneous recorded atca data were obtained for all but the 2008 observation . the nrao s very long baseline array ( vlba ) calibrator survey routinely observes pks j0038@xmath02459 at 2.3 ghz , but did not in 2008 . [ fig : j0038lightcurve ] plots the 2.3 ghz light curve of pks j0038@xmath02459 for eight epochs from 1997 to 2010 . the error bars are @xmath610@xmath22 of the measured flux densities . a weighted non - linear least - squares fit was used to interpolate to the flux density of pks j0038@xmath02459 at the time of the 2008 lba observation . the fit gave a flux density of 337 mjy , with a 1@xmath23 error of 19 mjy at the time of the 2008 observation . this flux density was used to refine the amplitude calibration for the 2008 lba data , by defining a model with this flux density in difmap @xcite . the difmap task gscale was used to determine corrections to the uncalibrated amplitudes of pks j0038@xmath02459 from the model . the square point plots the final imaged flux density following amplitude refinement for the 2008 epoch ( square ) . the arrows indicate the dates of the lba observations . ] following amplitude calibration refinement , global fringe - fitting solutions were determined for pks j0038@xmath02459 ( aips task fring ) with a three minute solution interval , finding independent solutions for each of the 16 mhz bands . the delay and phase solutions were examined and , following editing of bad solutions , applied to pks j0038@xmath02459 . the pks j0038@xmath02459 data were exported to difmap , where the data were vector - averaged over 30 s , flagged of bad data and imaged using standard imaging techniques ( deconvolution and self - calibration of both phase and amplitude ) . the resulting images of pks j0038 - 2459 for the 2006 , 2007 and 2008 epochs show a highly compact source , with no significant structure on these baselines at this frequency to a dynamic range of 400 . the final calibration solutions ( phase and amplitude ) of pks j0038@xmath02459 were exported via the difmap task , cordump @xcite to an aips - compatible solutions table . the solutions table was then transferred to aips and applied to all sources in the dataset . the pks j0038 - 2459 data were also used to derive a bandpass calibration via the aips task bpass which was applied to the ngc 253 data . the edge channels of each band were flagged from the data set ( 2 channels from both the lower and upper edge of each 16 channel band ) . the final calibration solutions were applied to both pks j0038@xmath02459 and ngc 253 and the visibility data exported as fits files . the loss of amplitude due to bandwidth smearing at the field s edge ( radius of @xmath24 from the phase centre ) is @xmath25 on the longest baseline ( hobart - ceduna ) . thus , to allow imaging of the inner 300 pc at a resolution of @xmath26 milli - arcseconds ( mas ) , the ngc 253 dataset was not averaged in time or frequency . to facilitate imaging in difmap , the frequency channels were converted into intermediate frequencies ( ifs ) . this conversion allowed difmap to treat the frequency channels independently in the @xmath21-plane rather than averaging them together , thus avoiding any further bandwidth smearing effects during the imaging process . ngc 253 imaging was initially performed with a reduced resolution by excluding data from the hobart and ceduna antennas ( the longest baselines ) . the imaging parameters were chosen to closely match the imaging parameters of lt06 , with cellsize of 11 mas and application of natural weighting to minimise noise at the expense of resolution . [ fig : wf ] displays the resulting contour images of ngc 253 for the three epochs . the 1@xmath23 noise measurement for the three images are listed in [ tab : table1 ] . the 1@xmath23 noise levels at the 2006 and 2007 epochs are 2 - 3 times higher than at the 2008 epoch , which can be attributed to the absence of the 70 m tidbinbilla antenna in 2006/07 . however , the 1@xmath23 noise of the 2008 image is 28@xmath22 lower than that of the lt06 ( 2004 epoch ) image due to increased bandwidth . [ fig:5.48 ] shows the images of the snr 5.48 - 43.3 for the three epochs . the images were made with the full array ( i.e. including data from both hobart and ceduna ) , with natural weighting and a cell size of 4 mas . * 15c & & & & & & & & & & & & & + @xmath13 ( j2000.0 ) & @xmath17 ( j2000.0 ) & & & & & + ( 00@xmath1447@xmath27 & ( @xmath02517 ) & ua97 & th85 & lt06@xmath28 & & @xmath29 & @xmath30 & @xmath31 & @xmath32 & & @xmath33 & @xmath34 & @xmath35 & @xmath36 + [ 1pt ] 32.876 & 21.383 & 5.48@xmath043.3 & th9 & y & & 8.6 @xmath6 0.9 & 13.5@xmath6 1.6 & 16.0 @xmath6 1.9 & 7.8 @xmath6 01.2 & & 32.0 @xmath6 3.2 & 22.8 @xmath6 2.4 & 30.2 @xmath6 3.4 & 25.1 @xmath6 9.9 + [ 1pt ] 33.012 & 19.425 & 5.62@xmath041.3 & th7 & y & & 1.7 @xmath6 0.2 & 2.7 @xmath6 0.6 & 2.4@xmath6 0.5 & 1.3 @xmath6 0.3 & & 8.8 @xmath6 0.9 & 6.3@xmath6 0.8 & 5.7 @xmath6 0.8 & 6.8 @xmath6 2.7 + [ 1pt ] 33.115 & 18.039 & 5.72@xmath040.1 & th6 & y & & 1.5 @xmath6 0.2 & @xmath32.0 & @xmath32.2 & 1.0 @xmath60.2 & & 3.8 @xmath6 0.4 & @xmath32.0 & @xmath32.2 & 3.0 @xmath6 1.2 + [ 1pt ] 33.185 & 19.084 & 5.78@xmath041.1 & .... & n & & @xmath31.2 & @xmath32.0 & @xmath32.2 & 0.9 @xmath60.2 & & @xmath31.2 & @xmath32.0 & @xmath32.2 & 4.6 @xmath6 1.8 + 33.177 & 17.830 & 5.79@xmath039.7 & th3 & y & & 1.8 @xmath6 0.2 & @xmath32.0 & @xmath32.2 & 1.0 @xmath60.2 & & 6.2 @xmath6 0.6 & @xmath32.0 & @xmath32.2 & 3.2 @xmath6 1.3 + [ 1pt ] 33.181 & 17.148 & 5.79@xmath039.0 & th2 & y & & 2.4 @xmath6 0.2 & 4.1 @xmath6 0.6 & 4.9 @xmath6 0.7 & 2.6 @xmath60.4 & & 5.7 @xmath6 0.6 & 7.8 @xmath6 1.0 & 7.3 @xmath6 0.9 & 5.9 @xmath6 2.3 + [ 1pt ] + + + + + a number of detected sources are clearly visible in [ fig : wf ] . we used the source extraction software , blobcat @xcite to identify and measure the flux densities of compact sources in the three images . seven sources were detected above 5@xmath23 by blobcat in the 2008 epoch , while only the three brightest sources ( 5.48 - 43.3 , 5.62 - 41.3 and 5.79 - 39.0 ) were found via blobcat for the less sensitive 2006 and 2007 epochs ( see [ tab : lbafluxes ] ) . included in [ tab : lbafluxes ] are the flux densities for the compact sources detected in the 2004 epoch by lt06 . only three of these sources were detected in all four 2.3 ghz lba epochs ( i.e. 2004 , 2006 , 2007 and 2008 ) , while three were detected in only the 2004 and 2008 epochs , with one possible additional source detected solely in the 2008 image . the flux density of the sources detected by lt06 were also estimated via blobcat and found to agree with lt06 s values within measurement errors . for comparison , the published source flux densities of lt06 are included in [ tab : lbafluxes ] . while stacking of the 2004 and 2008 images improved image noise by a factor of @xmath37 , the resulting image did not reveal any additional sources . errors in the flux - density estimates have two different origins . firstly , there is the overall uncertainty in the absolute flux density scale for the lba , estimated as @xmath38 @xcite . secondly , there are the errors in determining the flux density by blobcat , which ranges from 5 - 10@xmath22 . both terms are added in quadrature for the 2006 and 2007 observations . for the 2008 observation , we determine a third contribution to the error , which is described in [ sec : fluxvar ] . table [ tab : lbafluxes ] lists cross identifications with sources detected at 15 ghz and 23 ghz by ua97 and at 2.3 ghz with the lba by lt06 . the mean position difference between our 2008 and lt06 s 2004 j2000.0 source positions was found to be 8 mas , with a standard deviation of 5 mas , which is @xmath39 of the beam - width . the mean position difference between our 2008 j2000.0 source position and the b1950.0 source positions of the 23 ghz sources from ua97 is 87 mas , with a standard deviation of 51 mas . ) are plotted on the y - axis , while the observation date is listed on the x - axis . 5@xmath23 upper limits are represented by arrows for epochs without detections . the error - bars are the absolute flux density scaling and measurement errors ( see [ sec : flux ] ) added in quadrature . ] lower integrated flux densities are recorded for all sources except for 5.79@xmath039.0 in the 2008 epoch compared to the 2004 epoch ( see [ tab : lbafluxes ] and [ fig : lcurve ] ) . since it was necessary to interpolate the flux density of a variable calibrator to obtain a value for the 2008 epoch ( see [ sec : calib ] ) it would not be surprising if there were an error on the flux density scale . given the covariance of the flux densities between 2006 and 2008 ( and possibly for the other epochs too ) this would appear to be the case . adjusting empirically for the difference in flux density between 2006 and 2008 by taking the mean of each , we find that we must increase the absolute flux density calibration of the 2008 epoch by 20@xmath22 . after this adjustment , no source appears significantly variable with the exception of 5.805 - 38.92 , whose decrease in flux density would still be a 9@xmath23 error ( assuming of course that the distribution of the errors is gaussian ) . to investigate whether the flux density decrease is genuine we used the aips task uvmod to simulate datasets with the same @xmath21-coverage and noise level as the 2004 and 2008 lba datasets . in uvmod the parameter factor was set to zero to allow the injection of the fake sources at the same positions as the real ones . the fake sources were injected with the properties ( i.e. size , position angle , integrated flux density and position ) of the sources detected by lt06 . the resulting datasets were exported and imaged in difmap following the procedure described in [ sec : calib ] . the integrated flux densities of the injected sources were measured with blobcat and plotted in [ fig : uvmod ] ( left plot ) against the model flux densities . if the flux density decrease in the observed dataset were due to only source variability , the measured ( or recovered ) flux densities would be equal to the injected flux densities , represented by the dashed lines . the deviation of the uvmod flux densities from the dashed lines follows a similar trend to the observed flux densities ( right plot of [ fig : uvmod ] ) . thus it is possible the flux density decrease results mainly from systematic effects such as deconvolution errors ( e.g. clean bias @xcite ) and high side - lobes . the effect of clean bias is known to produce a systematic underestimate of flux densities @xcite by redistributing flux density from sources to noise peaks in the image during deconvolution . the magnitude of the flux density redistributed is generally independent of flux density , and thus the fractional error is largest for weak sources . this effect may have a stronger effect in the 2008 image ( as compared to the 2004 image ) due to the presence of high side - lobes . the high side - lobes may have resulted from the combination of poor @xmath21-coverage and an increased sensitivity due to a higher bandwidth . the rms ] difference between the 2008 recovered flux densities and the model flux densities is 37@xmath22 , which is added in quadrature with the absolute flux density scaling and measurement errors for the 2008 epoch ( see [ sec : flux ] ) and listed in [ tab : lbafluxes ] . the resulting flux density decrease of 5.805@xmath038.92 is then reduced to a significance of 4.5@xmath23 . while this variability may be due to unaccounted for calibration errors , we can not rule out contributions from intrinsic variability . l|cccc|cccc source & @xmath40 & @xmath41 & @xmath42 & type@xmath43 & @xmath44 & @xmath45 & @xmath46 & type + 4.81 - 43.6 & 19.92@xmath612.81 & @xmath01.46@xmath60.62 & 9.16@xmath65.16 & s & 24.1@xmath616.4 & @xmath01.53@xmath60.73 & 11.59@xmath66.84 & s + 5.48 - 43.3 & 105.6@xmath60.3 & @xmath00.77@xmath60.01 & 3.30@xmath60.02 & s & 76.4@xmath60.13 & @xmath00.63@xmath60.01 & 2.89@xmath60.01 & s + 5.49 - 42.2 & 2.31@xmath60.53 & @xmath00.33@xmath60.15 & 2.19@xmath61.65 & t & 2.47@xmath60.84 & @xmath00.35@xmath60.24 & 4.12@xmath63.11 & t + 5.54 - 42.2 & 1.54@xmath60.18 & 0.33@xmath60.06 & 2.54@xmath61.66 & t & 1.68@xmath60.30 & 0.3@xmath60.09 & 4.21@xmath62.60 & t + 5.59 - 41.6 & 0.72@xmath60.04 & 0.67@xmath60.03 & 0.29@xmath61.12 & t & 0.77@xmath60.11 & 0.65@xmath60.07 & 2.23@xmath62.30 & t + 5.62 - 41.3 & 16.11@xmath60.04 & @xmath00.31@xmath60.01 & 1.82@xmath60.02 & t & 12.25@xmath60.07 & @xmath00.21@xmath60.01 & 1.49@xmath60.02 & t + 5.65 - 40.7 & 2.71@xmath60.73 & @xmath00.25@xmath60.17 & 3.46@xmath62.28 & t & 2.89@xmath61.02 & @xmath00.27@xmath60.24 & 5.40@xmath63.66 & t + 5.72 - 40.1 & 8.25@xmath60.37 & @xmath00.02@xmath60.02 & 4.12@xmath60.53 & t & 9.03@xmath60.94 & @xmath00.05@xmath60.04 & 5.70@xmath61.23 & t + 5.73 - 39.5 & 2.07@xmath60.47 & 0.09@xmath60.12 & 3.59@xmath62.33 & t & 2.21@xmath60.65 & 0.07@xmath60.17 & 5.52@xmath63.58 & t + 5.75 - 41.8 & 76.97@xmath65.12 & @xmath01.16@xmath60.04 & 17.46@xmath60.64 & s & 86.16@xmath65.80 & @xmath01.20@xmath60.04 & 18.98@xmath60.66 & s + 5.78 - 39.4 & 121.18@xmath62.49 & @xmath00.81@xmath60.01 & 22.67@xmath60.51 & s & 129.41@xmath63.35 & @xmath00.83@xmath60.01 & 24.61@xmath60.67 & s + 5.79 - 39.0 & 150.79@xmath60.59 & @xmath00.46@xmath60.01 & 17.58@xmath60.08 & s & 149.60@xmath62.32 & @xmath00.46@xmath60.01 & 17.45@xmath60.26 & s + 5.79 - 39.7 & 4899.0@xmath62737.6 & @xmath02.58@xmath60.93 & 25.15@xmath610.52 & s & 5244.7@xmath62959.9 & @xmath02.6@xmath61.14 & 30.09@xmath613.9 & s + 5.78 - 41.1 & 75.97@xmath65.53 & @xmath01.38@xmath60.04 & 17.09@xmath60.60 & s & 33.03@xmath64.71 & @xmath01.07@xmath60.08 & 6.20@xmath60.88 & s + 5.805 - 38.92 & 87.74@xmath625.72 & @xmath01.23@xmath60.15 & 8.26@xmath61.68 & s & 97.7@xmath649.61 & @xmath01.27@xmath60.55 & 15.24@xmath67.14 & s + 5.87 - 40.1 & 70.27@xmath636.82 & @xmath01.50@xmath60.51 & 16.00@xmath66.63 & s & 84.0@xmath616.12 & @xmath01.57@xmath60.11 & 18.17@xmath61.33 & s + 5.90 - 37.4 & 2.30@xmath60.35 & 0.35@xmath60.07 & 4.72@xmath62.30 & t & 2.48@xmath60.48 & 0.32@xmath60.09 & 6.20@xmath63.05 & t + 5.95 - 37.7 & 5.08@xmath61.94 & @xmath00.80@xmath60.30 & 4.45@xmath62.67 & s & 6.15@xmath62.89 & @xmath00.87@xmath60.42 & 6.89@xmath64.37 & s + 6.00 - 37.0 & 82.41@xmath68.22 & @xmath01.32@xmath60.05 & 17.28@xmath60.80 & s & 94.1@xmath69.4 & @xmath01.37@xmath60.05 & 18.98@xmath60.82 & s + 6.40 - 37.1 & 25.72@xmath612.40 & @xmath01.10@xmath60.47 & 12.35@xmath65.91 & s & 31.17@xmath615.94 & @xmath01.17@xmath60.55 & 14.78@xmath67.40 & s + + + + a downturn in the spectra of the compact sources in ngc 253 at low frequencies was observed previously by @xcite and lt06 . different mechanisms were explored by lt06 to explain the observed effect including ; a simple power law , a power law spectrum with free - free absorption by a screen of ionized gas , and a self - absorbed bremsstrahlung spectrum . lt06 demonstrated that the spectra of the sources were consistent with a free - free absorbed power - law spectrum , given by the following equation : @xmath47 where @xmath48 and @xmath13 is the optically thin intrinsic spectral index , @xmath49 is the free - free optical depth at 1 ghz , and @xmath50 is the intrinsic flux density of the source at 1 ghz . similar spectra have been obtained for compact sources in other nearby starburst galaxies , such as m82 @xcite , arp 220 @xcite , and ngc 4945 @xcite . we therefore adopt this model and the method described by lt06 . to model the free - free absorption towards the compact sources within ngc 253 we compile multi - wavelength radio flux density measurements for these sources from the literature ( listed in [ tab : allfluxes ] ) , including : 1.4 ghz lba data @xcite ; 2.3 ghz lba data ( lt06 and this paper ) ; 23 ghz , 15 ghz , 8.3 ghz and 5 ghz vla data ( ua97 ) . @xcite detected 10 ua97 sources at 23 ghz with the vla ( a configuration ) in 2004 . they reported higher flux densities for all sources compared to ua97 , at the same frequency . only the source 5.79@xmath039.0 ( the assumed core th2 ) was similar in both epochs within a 10@xmath22 error . the authors attributed the higher flux densities in their data to a larger beam ( @xmath51 ) . to keep the spectral modelling consistent with lt06 , and maintain a comparable beam size with the lba 2.3 ghz images , the flux densities from @xcite were not used . [ fig : ffspectra ] presents multi - epoch monitoring of the free - free spectra of 20 compact sources in ngc 253 . the free - free spectra obtained by lt06 are plotted as dashed lines , while the solid lines are the spectra derived using the 2008 2.3 ghz lba data . for sources not detected in the 2008 observation , an upper limit of 0.9 ( 5@xmath23 ) was adopted . table [ tab : ffparams ] lists the fitted free parameters @xmath13 , @xmath49 , and @xmath50 for each compact source . included for reference are the values of the parameters for the 2004 data . differences in the free - free spectra and derived parameters between 2004 and 2008 , while potentially large , are still within our errors , suggesting no evidence for changes in the free - free absorbing medium between epochs . these results further confirm the free - free absorption interpretation of @xcite and lt06 . from the radio spectra , lt06 deduced that eight of the 20 sources were thermally dominated h ii regions ( indicated with a flat intrinsic spectrum , @xmath520.4 ) and the remaining 12 as consistent with steep intrinsic power - law spectra typical of rsne or snrs . we find no difference in the source classification in our study . the flux density for the sources with spectra consistent with rsne or snrs , ranges from 2 - 30 mjy at 5 ghz . the relationship between the rise time and peak luminosity for core - collapse supernovae by @xcite predicts a rise time @xmath53110 days . a typical rsne follows a power law decay that begins after this fast rise , and would be detectable over many years . however , no such decay has been observed for these compact sources over two decades , suggesting that they are probably very old snrs . the brightest source in ngc 253 at 2.3 ghz , 5.48 - 43.3 , is the only fully resolved snr in the starburst galaxy with the full lba array ( i.e. including the long baselines of cd and ho ) . in [ fig:548v2 ] we present the two high sensitivity vlbi images of 5.48 - 43.3 ( the 2004 and 2008 epochs ) , restored with a beam size of 13@xmath1215 mas and beam position angle of 58 . the contours were chosen to represent identical surface brightness levels in both images . as first noted by lt06 , 5.48@xmath043.3 appears to be a shell - type snr , with a diameter of @xmath570 - 90 mas ( 1.4 - 1.8 pc ) . at both epochs , the structure of 5.48@xmath043.3 is dominated by the eastern lobe , which has a higher flux density than the western lobe by a factor of 3 . this may be the result of interactions with a denser interstellar medium in the direction of the eastern lobe . comparing the epochs , we notice that there are several differences in the lobes . such apparent changes in the small scale structures are possibly due to ambiguities caused by the combination of structural evolution and image fidelity limitations , resulting from incomplete sampling of the @xmath21-plane . the effects of @xmath21-plane sampling on the appearance and evolution of complex small - scale structures , within spherically symmetric shell - like sources , are well documented by @xcite ; they demonstrated that sparse @xmath21-plane sampling , and the non - uniqueness of deconvolution , can add complex azimuthal structure to a radio brightness distribution that is in reality morphologically simple . one striking feature of the @xcite results is the departure from a spherically symmetric shell to a two - lobed structure ( similar to sn 1987a @xcite and 5.48@xmath043.3 ) when the @xmath21-plane coverage lacks intermediate baselines , similar to the @xmath21 coverage of both the 2004 ( see 2 of lt06 ) and the 2008 ( [ fig : uvcov ] ) observations . thus , it is possible that 5.48 - 43.3 possesses a spherically symmetric ( or slightly elliptical ) morphology that is not recovered due to low sensitivity associated with sparse sampling of the @xmath21-plane . this effect is noticeable in the epochs without the intermediate baselines to tidbinbilla ( 2006 and 2007 ) , with the non - detection of the weaker , western lobe . with multiple high resolution observations at the same frequency , it may be possible to determine the expansion speed of 5.48 - 43.3 . in order to carry out such measurements , we use the aips task iring to measure radial profiles ( averaged in azimuth ) of the source at each epoch . this method has been successfully implemented to determine the expansion speed of resolved snrs in m82 @xcite . owing to differences in the small - scale structure between the two epochs ( as discussed in the previous section ) , determination of the geometrical centre of 5.48 - 43.3 is difficult . to account for any positional offset between the epochs , the position of the peak surface brightness in the 2008 image was aligned to the position of the 2004 peak surface brightness , using the aips task ogeom . a common geometrical centre for both images was estimated by visual inspection , and indicated by the crosses in [ fig:548v2 ] . using this position as the reference point , the radial profiles were obtained by measuring the integrated flux density within a series of 4 mas thick annuli , and are plotted in [ fig : iring ] . no discernible expansion between the two epochs can be seen in [ fig : iring ] . at the smaller radii , higher integrated flux densities were recovered for the 2004 image . this is possibly due to differences in the short spacings of the @xmath21 coverages and/or deeper cleaning in the 2008 image . if we consider an upper limit of the expansion speed on @xmath54 km s@xmath2 @xcite for snrs in ngc 253 , an expansion of @xmath52.2 mas , @xmath55 of the 2.3 ghz lba beam , is expected for 5.48@xmath043.3 over 4.3 years between the 2004 and 2008 lba observations . to measure the expansion of 5.48@xmath043.3 , multiple observations separated by @xmath910 years with the current lba baselines , or observations with longer baselines ( @xmath93000 km ) separated by @xmath95 years would be needed . a lower limit on the supernova rate can be estimated based on the number of detected snrs , their size , and an assumed expansion rate ( lt06 ) . based upon the method by @xcite and an assumed radial expansion rate of @xmath1 km s@xmath2 , lt06 estimated the lower limit of the supernova rate to be @xmath56 km s@xmath57 . our results from [ sec:548 ] are consistent with this as the minimum supernova rate . no new sources have been detected in ngc 253 after two decades of multi - wavelength , high resolution radio observations . the consequences of the non - detections for estimates of the supernova rate were first investigated by ( * ? ? ? * hereafter ua91 ) . they assumed a hypothetical population of rsne whose flux densities at 5 ghz peak 100 days after the optical maxima and decay as the @xmath00.7 power of time @xcite . the rsne population was assumed to have peak luminosities uniformly distributed between 5 and 20 times that of cas a , and an ngc 253 distance of 2.5 mpc ( th85 ) was also assumed . ua91 found that @xmath52/3 of the rsne that occurred during the 18 month period between two 5 ghz epochs should be detectable above the second epoch s sensitivity limit . by assuming the events are poisson distributed in time , ua91 determined with 95@xmath22 confidence an upper limit to the supernova rate of 3.0 yr@xmath2 in ngc253 . this model was used to estimate the supernova rate upper limit for two subsequent 5 ghz epochs , @xmath51.4 yr@xmath2 for a third epoch 2.5 years later @xcite and @xmath50.3 yr@xmath2 for a fourth epoch , 4.0 years later ( ua97 ) . with new distance measurements for ngc 253 ( 3.94@xmath60.5 mpc , @xcite ) and two additional high resolution observations , lt06 developed a new model based on the principles of ua91 and using new data : a 5 ghz vla observation by @xcite ; and a 2.3 ghz lba observation by lt06 . the rsne at 2.3 ghz were assumed to peak 200 days after the optical maxima at the same flux density as at 5 ghz ( see @xcite ) . lt06 also considered the effects of free - free absorption by adjusting the flux densities of the rsne with , assuming a median value of @xmath49 = 6 . the number of rsne that occur between epochs was determined via a poisson - distributed random number , given a specified supernova rate . the radio luminosities of the rsne were allowed to evolve using the parametric equation of @xcite into supernovae remnants ( snrs ) and then , given the time and sensitivity of the observation , a test was made to determine if each snr could be detected . a monte carlo simulation was used to drive the model ( @xmath5 10000 iterations ) , where the proportion of snrs detected , @xmath58 at the end of each epoch was obtained . the simulation was seeded with an initial supernova rate of 0.1 yr@xmath2 producing a confidence level for the @xmath58 at each epoch . using linear interpolation , the supernova rate required to drive the simulation to a 95@xmath22 confidence limit was determined and the simulation was repeated until the confidence limit was achieved . the resulting supernova rate was then used to seed the next epoch and the process was continued until all epochs were processed . using this model , lt06 found that the increased distance , combined with the effects of free - free absorption , decreased @xmath58 which resulted in an upper limit of the supernova rate at the end of the final epoch of 2.4 yr@xmath2 . a summary of the model parameters used by @xcite and lt06 to derive upper limits on the supernova rate are listed in [ tab : snrmodelparams ] . with additional high resolution radio observations of ngc 253 since lt06 , improved distance measurements to ngc 253 , and availability of a more realistic rsne luminosity function @xcite , it is worth revisiting the estimates of the upper limit on the supernova rate . the additional ngc 253 observations are the three new 2.3 ghz lba observations in this paper , plus 23 ghz vla observations by @xcite and 15 ghz vla observation by @xcite . [ fig : timeline ] gives a time - line of all observations of ngc 253 used to determine the supernova rate . to further constrain the upper limit on the supernova rate with these observations , we have developed a new model based on the principles of @xcite and lt06 , as well as incorporating suggested improvements by lt06 . 15 ghz ( mga05 ) , vla 23 ghz ( bct09 ) and three lba 2.3 ghz ( this paper ) observations . note : mga05 - @xcite ; bct09 - @xcite ] llccccccc name & galaxy & type & distance@xmath43 & detected?@xmath59 & s@xmath60 @xmath28 & l@xmath61 @xmath62 & l@xmath63 @xmath62 & ref . + & & & ( mpc ) & & ( mjy ) & & & + sn1983n & m83 & ibc & 4.5 & d & 40.10 & 9.72 & 132.45 & 1 + sn1984l & ngc 4991 & ibc & 21.7 & d & 4.59 & 25.86 & 352.54 & 2 + sn1990b & ngc 4568 & ibc & 18.2 & d & 1.26 & 4.99 & 68.08 & 3 + sn1994i & m51 & ibc & 8.0 & d & 16.20 & 12.50 & 170.38 & 4 + sn1998bw & eso184-g082 & ibc & 37.3 & d & 37.40 & 622.59 & 8487.24 & 5 + sn2000c & ngc 2415 & ibc & 59.0 & e & 0.19 & 7.79 & 106.18 & 6 + sn2001b & ic391 & ibc & 24.0 & d & 2.56 & 17.64 & 240.51 & 6 + sn2001ci & ngc 3079 & ibc & 17.0 & d & 1.06 & 3.67 & 50.06 & 6 + sn2002ap & ngc 628 & ibc & 7.3 & d & 0.45 & 0.29 & 3.94 & 7 + sn2002cj & eso582-g5 & ibc & 106.0 & d & 0.22 & 29.58 & 403.19 & 6 + sn2002dg & anon . & ibc & 215.0 & e & 0.09 & 50.88 & 693.65 & 6 + sn2003l & ngc 3506 & ibc & 92.0 & d & 2.56 & 258.75 & 3527.31 & 8 + sn2003bg & mcg-05 - 10 - 15 & ibc & 19.6 & d & 52.81 & 242.74 & 3309.07 & 9 + sn2004cc & ngc 4568 & ibc & 18.0 & e & 3.26 & 12.64 & 172.28 & 10 + sn2004dk & ngc 6118 & ibc & 23.0 & d & 1.80 & 11.39 & 155.31 & 10 + sn2004gq & ngc 1832 & ibc & 26.0 & d & 5.75 & 46.51 & 634.00 & 10 + sn2007bg & anon . & ibc & 152.0 & d & 1.10 & 304.91 & 4156.62 & 11 + sn2007gr & ngc 1058 & ibc & 7.7 & d & 1.00 & 0.71 & 9.67 & 12 + sn2008d & ngc 2770 & ibc & 27.0 & e & 3.00 & 26.17 & 356.72 & 13 + sn2009bb & ngc 3278 & ibc & 40.0 & d & 12.00 & 229.73 & 3131.69 & 14 + sn2010as & ngc 6000 & ibc & 27.3 & d & 1.40 & 12.48 & 170.19 & 15 + ptf11qcj & sdss@xmath64 & ib & 124.0 & d & 5.43 & 998.98 & 13618.22 & 16 + sn1970 g & ngc 5457 & ii & 7.2 & d & 2.50 & 1.55 & 21.14 & 17 + sn1974e & ngc 4038 & ii ? & 20.9 & d & 1.24 & 6.48 & 88.35 & 18 + sn1978k & ngc 1313 & ii & 4.1 & d & 518.00 & 104.19 & 1420.28 & 18 + sn1979c & ngc 4321 & iil & 22.0 & d & 12.40 & 71.81 & 978.91 & 18 + sn1980k & ngc 6946 & iil & 7.0 & d & 2.45 & 1.44 & 19.58 & 18 + sn1981k & ngc 4258 & ii & 6.6 & d & 5.15 & 2.68 & 36.59 & 18 + sn1982aa & ngc 6052 & ii & 66.0 & d & 19.10 & 995.48 & 13570.58 & 18,19 + sn1985l & ngc 5033 & ii & 12.9 & d & 1.56 & 3.11 & 42.34 & 20 + sn1986e & ngc 4302 & ii & 16.8 & d & 0.33 & 1.11 & 15.19 & 21 + sn1986j & ngc 891 & iin & 12.0 & d & 135.00 & 232.60 & 3170.83 & 18 + sn1988z & mcg+03 - 28 - 022 & iip & 94.5 & d & 1.85 & 197.67 & 2694.71 & 22 + sn1993j & m81 & iib & 3.6 & d & 96.90 & 15.28 & 208.26 & 23 + sn1995n & mcg-02 - 38 - 17 & iin & 24.0 & e & 9.00 & 62.03 & 845.56 & 24 + sn1997eg & ngc 5012 & ii & 40.0 & e & 10.00 & 191.44 & 2609.74 & 25 + sn1998s & ngc 3877 & ii & 17.0 & e & 1.04 & 3.60 & 49.02 & 26 + sn1999em & ngc 1637 & ii & 7.8 & e & 0.30 & 0.22 & 2.98 & 26 + sn2000 ft & ngc 7469 & ii ? & 70.0 & d & 1.76 & 103.30 & 1408.25 & 27 + sn2001ig & ngc 7424 & iib & 11.5 & d & 21.90 & 34.65 & 472.41 & 28 + sn2001gd & ngc 5033 & iib & 20.0 & d & 7.96 & 38.10 & 519.34 & 29 + sn2004dj & ngc 2403 & iip & 3.6 & e & 1.90 & 0.29 & 4.02 & 30 + sn2004et & ngc 6946 & ii & 5.5 & e & 2.50 & 0.90 & 12.34 & 31 + sn2006jd & ugc 4179 & iib & 76.8 & d & 2.22 & 156.39 & 2131.91 & 32 + sn2008ax & ngc 4490 & iip & 8.1 & d & 4.00 & 3.13 & 42.70 & 33 + sn2008iz & m82 & ii & 3.2 & e & 180.00 & 22.05 & 300.64 & 34 + sn2011cb & pgc 69707 & iib & 29.2 & d & 0.73 & 7.46 & 101.73 & 35 + sn2011dh & m51 & iip & 8.0 & d & 7.80 & 6.02 & 82.04 & 36 + sn2011ei & ngc 6925 & iib & 28.5 & d & 0.56 & 5.48 & 74.72 & 37 + a25 & arp220 & iip & 44.8 & e & 0.30 & 7.20 & 98.21 & 38 + a27 & arp220 & iip ? & 44.8 & e & 0.75 & 18.01 & 245.52 & 38 + & & & & & & & & + + + + + + + + + + + + + + + + + + llccccccc name & galaxy & type & distance & s@xmath60 & l@xmath60 & l@xmath63 & p - value@xmath43 & ref . + & & & ( mpc ) & ( mjy ) & & & & + sn1980i & ngc 4374 & ibc & 22.0 & @xmath30.50 & @xmath32.90 & @xmath339.47 & 0.27 & 1,2 + sn1980n & ngc 1316 & ibc & 33.0 & @xmath30.60 & @xmath37.82 & @xmath3106.58 & 0.39 & 1 + sn1981b & ngc 4536 & ibc & 33.0 & @xmath30.30 & @xmath33.91 & @xmath353.29 & 0.30 & 1 + sn1999ex & ic 5179 & ibc & 54.0 & @xmath30.07 & @xmath32.48 & @xmath333.77 & 0.25 & 3 + sn2000cr & ngc 5395 & ibc & 54.0 & @xmath30.04 & @xmath31.47 & @xmath319.98 & 0.20 & 3 + sn2000ew & ngc 3810 & ibc & 15.0 & @xmath30.07 & @xmath30.18 & @xmath32.46 & 0.06 & 3 + sn2000fn & ngc 2526 & ibc & 72.0 & @xmath30.05 & @xmath32.79 & @xmath338.05 & 0.26 & 3 + sn2001 m & ngc 3240 & ibc & 56.0 & @xmath30.03 & @xmath31.05 & @xmath314.32 & 0.17 & 3 + sn2001ai & ngc 5278 & ibc & 118.0 & @xmath30.05 & @xmath38.16 & @xmath3111.29 & 0.39 & 3 + sn2001bb & ic 4319 & ibc & 72.0 & @xmath30.06 & @xmath33.72 & @xmath350.73 & 0.29 & 3 + sn2001ef & ic 381 & ibc & 38.0 & @xmath30.12 & @xmath31.99 & @xmath327.09 & 0.23 & 3 + sn2001ej & ugc 3829 & ibc & 63.0 & @xmath30.04 & @xmath32.09 & @xmath328.48 & 0.23 & 3 + sn2001is & ngc 1961 & ibc & 60.0 & @xmath30.07 & @xmath33.02 & @xmath341.10 & 0.27 & 3 + sn2002j & ngc 3464 & ibc & 58.0 & @xmath30.09 & @xmath33.42 & @xmath346.64 & 0.28 & 3 + sn2002bl & ugc 5499 & ibc & 74.0 & @xmath30.05 & @xmath32.95 & @xmath340.19 & 0.27 & 3 + sn2002bm & mcg013219 & ibc & 85.0 & @xmath30.07 & @xmath35.71 & @xmath377.78 & 0.35 & 3 + sn2002cg & ugc 10415 & ibc & 150.0 & @xmath30.07 & @xmath319.92 & @xmath3271.58 & 0.51 & 3 + sn2002cp & ngc 3074 & ibc & 80.0 & @xmath30.04 & @xmath32.68 & @xmath336.54 & 0.26 & 3 + sn2002dn & ic 5145 & ibc & 115.0 & @xmath30.04 & @xmath36.80 & @xmath392.76 & 0.37 & 3 + sn2002ge & ngc 7400 & ibc & 47.0 & @xmath30.13 & @xmath33.44 & @xmath346.84 & 0.29 & 3 + sn2002gy & ugc 2701 & ibc & 114.0 & @xmath30.04 & @xmath36.06 & @xmath382.67 & 0.35 & 3 + sn2002hf & mcg05320 & ibc & 88.0 & @xmath30.04 & @xmath33.71 & @xmath350.52 & 0.29 & 3 + sn2002hn & ngc 2532 & ibc & 82.0 & @xmath30.04 & @xmath33.54 & @xmath348.26 & 0.29 & 3 + sn2002ho & ngc 4210 & ibc & 42.0 & @xmath30.05 & @xmath31.14 & @xmath315.54 & 0.17 & 3 + sn2002hy & ngc 3464 & ibc & 58.0 & @xmath30.07 & @xmath32.98 & @xmath340.60 & 0.27 & 3 + sn2002hz & ugc 12044 & ibc & 85.0 & @xmath30.03 & @xmath32.51 & @xmath334.18 & 0.25 & 3 + sn2002ji & ngc 3655 & ibc & 23.0 & @xmath30.04 & @xmath30.27 & @xmath33.71 & 0.08 & 3 + sn2002jj & ic 340 & ibc & 66.0 & @xmath30.03 & @xmath31.72 & @xmath323.45 & 0.21 & 2 + sn2002jp & ngc 3313 & ibc & 58.0 & @xmath30.04 & @xmath31.53 & @xmath320.85 & 0.20 & 3 + sn2002jz & ugc 2984 & ibc & 24.0 & @xmath30.03 & @xmath30.17 & @xmath32.35 & 0.06 & 3 + sn2003jd & mcg0159021 & ibc & 81.0 & @xmath30.05 & @xmath34.08 & @xmath355.65 & 0.31 & 4 + sn2011hp & ngc 4219 & ibc & 23.7 & @xmath30.20 & @xmath31.34 & @xmath318.32 & 0.19 & 5 + sn2005cs & m51 & ii & 8.0 & @xmath30.37 & @xmath30.29 & @xmath33.89 & 0.08 & 6 + sn2009mk & pgc 474 & iib & 18.3 & @xmath30.15 & @xmath30.60 & @xmath38.19 & 0.12 & 7 + & & & & & & & & + + + + + + + at 5 ghz and 2.3 ghz our hypothetical rsn is the same as that used by ua91 and lt06 ( see previous section and [ tab : snrmodelparams ] ) . at 8.3 ghz , 15 ghz and 23 ghz we assume that the rsn luminosity peaks 100 , 70 and 50 days after the optical maximum at the same luminosity as at 5 ghz . this assumption closely agrees with the light curves of sn 1993j @xcite . in the models of ua91 and lt06 , the distribution of the 5 ghz peak luminosities of rsne was assumed to be uniformly distributed between 5 and 20 times the luminosity of cas a. this assumption follows @xcite , who came to this conclusion after comparing the peak flux densities at 5 ghz of 16 type ii supernovae ( three detections and 13 upper limits ) . since the study by @xcite there have been many more new detections of type ii as well as type ibc supernovae in the radio . thus it is worth re - investigating the 5 ghz peak luminosity distribution of these objects . ) . the upside down triangles are supernovae whose peak luminosities have not been detected ( 2 type ii and 32 type ibc , see [ tab : snrupperlims ] ) . the dashed lines is the average 3@xmath23 detection limit of the vla from @xcite . ] table [ tab : snrflux ] lists 51 core - collapse ( 29 type ii @xmath11 22 type ibc ) supernovae whose 5 ghz peak flux density have either been detected or estimated via spectral index or extrapolation using the models by @xcite or @xcite . this table is represented graphically in [ fig : snedist ] , where the type ibc ( grey symbols ) and type ii ( black symbols ) luminosity ratio to cas a s at 5 ghz is plotted on the y - axis and their distances on the x - axis . included in the plots are the 5 ghz peak luminosity upper limits for 34 core - collapse supernovae not detected in the radio ( [ tab : snrupperlims ] ) , that follow the average 3@xmath23 detection limit of the vla ( the dashed lines , from @xcite ) . using the detected core - collapse supernovae listed in [ tab : snrflux ] , we derived a more accurate luminosity distribution at 5 ghz shown in [ fig : snehist ] . the data are divided into bins of size @xmath65log@xmath66(l@xmath63 is the rsne 5 ghz peak luminosity ratio to cas a s ] ) = 1 . the dashed line presents the best fit gaussian with mean luminosity ratio of 245.5 , standard deviation of 1.27 and @xmath67 = 0.018 . this distribution is similar to that presented by @xcite . however , this analysis ignores the upper limits ( [ tab : snrupperlims ] ) entirely , and so we should consider whether these offer any significant additional information . while their true luminosities are unknown , we need to determine if , given the upper limits , these supernovae are distributed differently to [ fig : snehist ] . to examine this we calculate the probability ( or p - value ) , using the cumulative distribution function derived from the gaussian distribution in [ fig : snehist ] , that a supernova will have a luminosity less than the upper limits listed in [ tab : snrupperlims ] . the p - values ( [ tab : snrupperlims ] ) , although on average are lower than 0.5 , are probably consistent with the chosen luminosity function given that the upper limits may not have been observed at a time corresponding to when the peak flux density actually occurs . however , detailed work on testing whether the upper limits in [ tab : snrupperlims ] are consistent with the chosen model are beyond the scope of this paper . ( l@xmath63 ) . the x - axis mirrors the y - axis of [ fig : snedist ] and is given in [ tab : snrflux ] . the data are binned to @xmath65log@xmath66(l@xmath63 ) = 1 . the dashed curve shows a non - linear least - squares fitted gaussian to the underlying data ( black x s ) . the resulting fit : mean log@xmath66(l@xmath63)= 2.39 , @xmath23 = 1.27 and @xmath67 = 0.018 . ] @xcite showed that the spectral index between 6 cm and 20 cm evolves with time , for a number of rsne . this has also been shown to be true for sn 1993j for wavelength pairs of 1.2/2 cm , 2/3.6 cm and 3.6/6 cm @xcite . thus , it is expected that 2.3/5 ghz , 5/8.3 ghz , 5/15 ghz and 5/23 ghz spectral indices will display similar variations with time . to account for this , we have included light curves that best describe the spectral indices between 6 @xmath11 20 cm from sn 1980k @xcite and between 6 @xmath11 3.6 cm , 6 @xmath11 2 cm , 6 @xmath11 1.2 cm from sn 1993j @xcite . [ tab : snrate ] lists the results of applying the improved model parameters described above and listed in column 4 of [ tab : snrmodelparams ] , to the monte - carlo simulation of lt06 ( see 3.3.2 ) . we find that the proportion of snrs detected , @xmath58 at the end of each epoch are relatively consistent , with minor variations between 0.66 and 0.84 . the major changes occur during epochs where the sensitivity varies by almost an order of magnitude . the consistently high @xmath58 resulted in an upper limit on the supernova rate of @xmath68 = 0.2 yr@xmath2 at the end of the final epoch . the results of the simulation indicate that the radio observations are detecting 60@xmath22 - 80@xmath22 of all snrs in ngc 253 , thus pointing to the existence of a small population of undetected snrs . moreover , the fact that no rsne or snrs has been observed in ngc 253 during 21 years of radio observation suggests a low rate of rsne production in ngc 253 . cccccc epoch & time@xmath43 & @xmath69 & sensitivity@xmath28 & @xmath58 @xmath62 & @xmath68 @xmath70 + & ( yr ) & ( ghz ) & ( mjy ) & - & ( yr@xmath2 ) + 1 & ... & 5.0 & ... & ... & + 2 & 1.5 & 5.0 & 3.0 & 0.746 & @xmath32.71 + 3 & 2.5 & 5.0 & 3.0 & 0.713 & @xmath31.04 + 4 & 4.0 & 5.0 & 3.0 & 0.683 & @xmath30.53 + 5 & 4.0 & 8.3 & 0.3 & 0.803 & @xmath30.34 + 6 & 0.25 & 15 & 0.6 & 0.838 & @xmath30.33 + 7 & 4.5 & 2.3 & 1.2 & 0.697 & @xmath30.25 + 8 & 0.5 & 23 & 2.0 & 0.661 & @xmath30.24 + 9 & 1.7 & 2.3 & 2.0 & 0.676 & @xmath30.22 + 10 & 0.8 & 2.3 & 2.2 & 0.689 & @xmath30.21 + 11 & 1.0 & 2.3 & 0.9 & 0.791 & @xmath30.20 + + + + + + + + the final result for @xmath68 agrees with estimates determined from near infrared ( nir ) observations of the [ feii ] line ( 0.24 yr@xmath2 ) by @xcite . observations of the [ feii ] line have been shown to be a strong tracer of shocks associated with snrs @xcite . dust grains in the interstellar medium containing fe atoms are destroyed via the shocks . the process releases the fe atoms , which are then ionized by the interstellar radiation field . in the post shock region , fe@xmath71 is excited by electron collisions , causing it to emit at nir wavelengths . however , there are possible contributions to the [ feii ] line from shocks due to other processes such as mergers @xcite . thus , it is possible that the nir [ feii ] observations are overestimating the supernova rate upper limit . the major uncertainty in our model arises from the rsne peak luminosity distribution . first , the detection of rsne has primarily been through follow - up observations of optical supernovae , resulting in a small and incomplete sample , whose luminosity depends on the sensitivity of the survey . second , there is the possibility that many rsne are not visible in the optical due to extinction and we may be looking at a slightly different sub - population further uncertainty comes from the assumed rise time of the rsne at the different frequencies . in this model , for simplicity , constant rise times ( the time to reach maximum radio luminosities ) were assumed for the hypothetical rsne at the different frequencies . however , this has been observed to be variable for different rsne ( e.g. @xcite ) . additionally , the relation between the rise time and peak luminosity has also been observed to be variable @xcite and does not necessarily follow a one - to - one relationship as assumed in the model presented . further refinement to the supernova rate of ngc 253 could be obtained from either a large unbiased sample of the peak flux densities of rsne , or direct observations of the rsne population in ngc 253 . both possibilities can be provided by the _ _ square kilometre array _ _ , ( ska ) and new generation telescopes , such as the karl g. jansky vla ( jvla ) . the sensitivity of both instruments may provide the opportunity to detect a population of weak snrs that are faint due to either free - free absorption or intrinsic properties . the ska is expected to have a type ii supernova detection rate of @xmath5620 yr@xmath2 deg@xmath72 out to a redshift @xmath73 5 @xcite . the jvla and precursors to the ska may provide a detection rate of @xmath5160 yr@xmath2 deg@xmath72 out to redshift @xmath73 3 @xcite . with new upper limits on the supernova rate , we can provide a new estimate for the star formation rate ( sfr ) in ngc 253 . following @xcite , the sfr can be determined directly from the supernova rate ( and vice versa ) . by assuming that all stars with mass @xmath74 eventually form supernovae , the relation between sfr and the supernova rate is given by : @xmath75 = 24.4\left[\dfrac{\nu_{sn}}{\mathrm{yr}^{-1}}\right]\ ] ] using the new supernova rate upper limits of @xmath76 yr@xmath2 ( [ sec : sr ] ) , we find @xmath4 yr@xmath2 . we can compare this upper limit to estimates from independent methods . if one assumes that the main contribution to the observed radio luminosity in a starburst are from rsne , snrs and thermal h ii regions , then the sfr of that region is directly proportional to its radio luminosity at the observed radio wavelength and can be calculated using the relation given by @xcite and @xcite . @xcite observed a total flux density of 0.59 jy for ngc 253 at 24 ghz . using the relation by @xcite and the data from @xcite we obtain @xmath77 = 9.3@xmath78 yr@xmath2 . additional observations of ngc 253 at 23 ghz ( total flux density of 0.56 jy ) by @xcite give a similar @xmath77 = 8.7@xmath78 yr@xmath2 for the @xmath5300 pc nuclear region , using the same method . the @xmath77 can also be determined from far infrared ( fir ) emission using equation 26 from @xcite . from fir luminosity measurements with _ _ iras _ _ @xcite and scaling for the new distance 3.44 mpc , the sfr is 1.3 - 2.0 @xmath78 yr@xmath2 for the inner @xmath5300 pc nuclear region and 2.6 - 3.2 @xmath78 yr@xmath2 for the entire galaxy . fir luminosity measurements of the inner @xmath5350 pc nuclear region by _ _ _ @xcite lead to an estimate of 2.7@xmath78 yr@xmath2 . it is encouraging that the @xmath77 calculated using different estimators are in broad agreement with each other . however , the differences point to the existence of systematic or intrinsic differences between the different estimators . the supernova rate derived from the fir luminosity ( via 20 from @xcite ) is 0.05 - 0.13 yr@xmath2and is similar to the lower limit of the supernova rate deduced from expansion rates ( lt06 ) . we have presented the results of multi - epoch observations of the southern starburst galaxy , ngc 253 with the lba at 2.3 ghz . the results presented here are complementary to previous 2.3 ghz lba observations by @xcite . we find the following results : 1 . seven compact sources were detected in the highest sensitivity observation ( the 2008 epoch ) . all sources were identified with higher - frequency vla observations ( ua97 ) , while six were identified with a 2.3 ghz lba detections by lt06 . the three brightest sources were also detected in the lower sensitivity observations ( 2006 and 2007 ) . the shell - like snr , 5.48 - 43.3 , was successfully imaged with the highest resolution in all three epochs . the structure in the 2008 image shows a double - lobed morphology which is dominated by the eastern lobe . the weaker , western lobe was not detected in the 2006 and 2007 observations , which is attributed to lower sensitivity and fidelity , due the lack of the 70 m tidbinbilla antenna . 3 . observed differences in the small - scale structure of 5.48@xmath043.3 between the 2004 ( lt06 ) and the 2008 ( this paper ) observations can be explained by deconvolution errors associated with sparse @xmath21-plane sampling . no discernible expansion has been observed for 5.48@xmath043.3 between the 2004 and 2008 epochs . the spectra of the 20 compact sources in ngc 253 from lt06 are compared with the spectra obtained with the new 2.3 ghz lba flux density measurements ( 2008 epoch ) . as with lt06 , we find that the spectra fit a free - free absorption model , with little difference in the spectra between the epochs . our results show no change in the classification of the compact sources by lt06 . 12 of the 20 sources have steep spectra associated with snrs , while the remaining eight have flat intrinsic power - law spectra ( @xmath79 ) , indicative of h ii regions . we derive an improved rsne peak luminosity distribution at 5 ghz using data from the literature for 51 core - collapse supernovae ( 29 type ii @xmath11 22 type ibc ) . 7 . we estimate a value for the upper limit of the supernova rate of 0.2 yr@xmath2 for the inner 300 pc of ngc 253 , using an improved model that is based upon the principles of ua91 and lt06 . this result was found to be consistent with estimates from near infrared ( nir ) observations . the results of our model indicate that the radio observations are detecting 60@xmath22 - 80@xmath22 of all snrs in ngc 253 , thus pointing to the existence of a small population of undetected snrs . moreover , no rsne or snrs has been observed in ngc 253 during 21 years of radio observation suggests a low rate of rsne production in ngc 253 . a new upper limit to the star formation rate of @xmath4 yr@xmath2 is estimated directly from the supernova rate limits for the inner 300 pc region of the galaxy , which is consistent with independent estimates . the international centre for radio astronomy research is a joint venture between curtin university and the university of western australia , funded by the state government of western australia and the joint venture partners . the long baseline array is part of the australia telescope which is funded by the commonwealth of australia for operation as a national facility managed by csiro . sjt is a western australian premier s research fellow , funded by the state government of western australia . the centre for all - sky astrophysics ( caastro ) is an australian research council centre of excellence , funded by grant ce110001020 . we wish to thank paola castangia @xmath11 andreas brunthaler for use of their vla data , cormac reynolds for assisting with the calibration of the lba data and cath trott for the helpful discussions on statistics . this research has made use of material from the bordeaux vlbi image database ( bvid ) and the the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration . we thank the anonymous referee for comments that improved this paper .

today string theory is a promising candidate for the quantum theory of gravity . since it requires more than four dimensions for its internal consistency , there has been growing interest in higher dimensional solutions and , in particular , in higher dimensional black holes ( see e.g. @xcite ) . the higher - dimensional generalizations of the stationary axisymmetric kerr black holes were found by myers and perry @xcite , who anticipated already the existence of higher - dimensional black rings . in 2001 emparan and reall @xcite then found such black rings in five dimensions . they rotate in the direction of the ring and possess a horizon topology of @xmath0 . the phase diagram of these black rings together with the myers - perry black holes showed , that the uniqueness of 4-dimensional vacuum black holes does not generalize to higher dimensions . in myers - perry black hole spacetimes the geodesic equations are separable @xcite . however , the geodesic equations of five dimensional black rings do not seem to be separable , in general . nevertheless , it is possible to separate the equations of motion on the rotational axis ( which is actually a plane ) , in the equatorial plane and in the case @xmath1 ( which is only possible in the ergosphere ) @xcite . hoskisson @xcite studied the geodesic motion of a singly spinning black ring and discussed the separability of the hamilton - jacobi equation , and provided numerical solutions . he analyzed numerically the motion on the rotational axis and the equatorial plane in detail . some aspects of nullgeodesics in the equatorial plane were studied by elvang , emparan and virmani @xcite . the pomerasky - senkov doubly spinning black ring @xcite was studied by durkee @xcite . he showed that it is possible to separate the hamilton - jacobi equation of the doubly spinning black ring in the case @xmath1 and analyzed the effective potential on the two axes ( planes ) of rotation and in the case @xmath1 . the zero energy nullgeodesics of the singly spinning dipole black ring were analyzed by armas @xcite . in @xcite igata , ishihara and takamori concentrated on stable bound orbits in the singly spinning black ring spacetime , which they found numerically on and near the rotational axis . so far the equations of motion for test particles in black ring spacetimes were only solved numerically , and no analytic solutions have been given . the first to solve the geodesic equations in a black hole spacetime analytically was hagihara @xcite . he presented the solution of the geodesic equation for test particles in the schwarzschild spacetime in terms of the elliptic weierstra @xmath2 function . when adding the cosmological constant to the schwarzschild metric one encounters hyperelliptic curves in the geodesic equations . the equations of motion in schwarzschild-(anti ) de sitter spacetimes in four dimensions were solved in @xcite . also analytical solutions of the geodesic equations in higher dimensional myers - perry spacetime @xcite as well as in higher dimensional schwarzschild , schwarzschild-(anti ) de sitter , reissner - nordstrm and reissner - nordstrm-(anti ) de sitter @xcite were found . the mathematical method is based on the jacobi inversion problem . the solution can be found if the problem is restricted to the theta - divisor , the set of zeros of the theta function . enolski , pronine and richter developed this method in 2003 to solve the problem of the double pendulum @xcite . in this paper we present analytical solutions of the equations of motion of a singly spinning black ring . in the case @xmath1 and in the equatorial plane the equations are of elliptic type , however , on the rotational axis the geodesic equations are of hyperelliptic type . the singly spinning black ring solution can be written in the form @xcite @xmath3 .\ ] ] the metric is given in toroidal coordinates ( see [ pic : ringcoord ] ) where @xmath4 , @xmath5 and @xmath6 . @xmath7 and @xmath8 are @xmath9-periodic . the metric functions are @xmath10 and @xmath8 . solid circles correspond to @xmath11 and dashed circles corredspond to @xmath12 . @xcite , width=453 ] the parameters @xmath13 and @xmath14 describe the shape , mass and angular momentum of the ring . @xmath13 lies in the range @xmath15 to ensure the black ring is balanced . a spacelike curvature singularity is located at @xmath16 . the metric has a coordinate singularity at @xmath17 , so the event horizon lies at @xmath18 . the ergosphere of the singly spinning black ring is determined by @xmath19 , which is at @xmath20 . since we have @xmath21 , it is clear that an ergoregion does exist . the topology of the event horizon and the ergosphere is @xmath22 . + the inverse metric is @xmath23 \ , .\ ] ] the singly spinning black ring metric and its hamiltonian @xmath24 do not depend on the coordinates @xmath25 , @xmath7 and @xmath8 , so we have three conserved momenta @xmath26 with the associated killing vector fields @xmath27 , @xmath28 and @xmath29 . a dot denotes the derivative with respect to an affine parameter @xmath30 . @xmath31 @xmath32 is the energy , @xmath33 and @xmath34 are the angular momenta in @xmath7- and @xmath8-direction . the conjugate momenta in @xmath35- and @xmath36-direction are : @xmath37 to obtain the equations of motion for a particle in the singly spinning black ring spacetime we need the hamilton - jacobi equation : @xmath38 we already have three constants of motion ( @xmath32 , @xmath33 and @xmath34 ) and the mass shell condition @xmath39 gives us a fourth , so we can make the ansatz @xmath40 inserting this ansatz into ( [ eqn : hamilton - jacobi ] ) gives @xmath41 \ , . \label{eqn : hjd - sring}\ ] ] the hamilton - jacobi equation does not seem to be separable in general . however , it is possible to separate the equation in the special case @xmath1 . these zero energy null geodesics are only realisable in the ergoregion . we can also obtain equations of motion for geodesics on the @xmath7- and @xmath8-axis by setting @xmath42 ( @xmath7-axis ) or @xmath43 ( @xmath8-axis ) . the plane @xmath42 which is called the @xmath7-axis , is the equatorial plane of the black ring . the plane @xmath43 which is called the @xmath8-axis corresponds to the rotational axis of the singly spinning black ring . + in the next sections we will study these special cases and solve the corresponding equations of motion analytically . for @xmath1 it is possible to separate the hamilton - jacobi equation : @xmath44 with a separation constant @xmath45 , the equation ( [ eqn : ham - jac - sing ] ) splits into two : @xmath46 so that @xmath47 using @xmath48 and ( [ eqn : t - impuls])-([eqn : y - impuls ] ) the separated hamilton - jacobi equation gives the equations of motion : @xmath49 where we have introduced the mino - time @xcite @xmath50 . equation ( [ eqn : sing - x - gleichung ] ) and ( [ eqn : sing - y - gleichung ] ) can be written as @xmath51 @xmath52 and @xmath53 can be regarded as effective potentials ( see @xcite ) . to get real solutions for the @xmath35- and @xmath36-equation the effective potentials have to be negative . so @xmath54 is required , because @xmath55 for @xmath15 and @xmath56 . the zeros of the effective potentials and hence @xmath57 and @xmath58 mark the turning points of the motion of light or a test particle ( in this case we only have light since @xmath59 ) . a good way to determine the number of zeros are parametric diagrams . figure [ pic : sing - parameter ] shows a parametric @xmath33-@xmath13-@xmath45-diagram for @xmath52 . it turns out that @xmath60 has two real zeros in the allowed range of @xmath35 or none . one zero is always negative while the other can be positive or negative . if @xmath60 has two zeros the @xmath35-motion takes place between these two values , if @xmath60 has a single zero at @xmath61 the @xmath35-motion stays constant . @xmath62 and accordingly @xmath53 determine the type of the orbit . if @xmath63 and @xmath64 , @xmath62 has no real zeros . otherwise @xmath62 has always one real zero in the allowed range of @xmath36 . that means the only possible orbit is a terminating orbit ( to ) , where light crosses the horizon and falls into the singularity . see figure [ pic : sing - potential ] for examples of the effective potentials . -@xmath13-@xmath33-diagram for the singly spinning black ring in the case @xmath1 . inside the structure @xmath60 has two real zeros , outside the structure it has no real zeros in the allowed range of @xmath35.,width=302 ] equation ( [ eqn : sing - x - gleichung ] ) can be written as @xmath65 where @xmath57 is a polynomial of third order with the coefficients @xmath66 the substitution @xmath67 transforms the polynomial into the standard weierstra form @xmath68 where @xmath69 equation ( [ eqn : sing - weierstrass - form ] ) is of elliptic type and is solved by the weierstra elliptic function @xcite @xmath70 where @xmath71 and @xmath72 . then the solution of ( [ eqn : sing - x - gleichung ] ) takes the form @xmath73 .\ ] ] equation ( [ eqn : sing - y - gleichung ] ) can be written as @xmath74 where @xmath58 is a polynomial of third order with the coefficients @xmath75 the problem can be solved analogously to the @xmath35-equation . here the solution is @xmath76 , \ ] ] where @xmath77 and @xmath78 . using ( [ eqn : sing - x - gleichung ] ) the equation ( [ eqn : sing - phi - gleichung ] ) becomes @xmath79 @xmath80 we substitute @xmath81 to transform @xmath60 into the weierstra form @xmath82 ( see ( [ eqn : sing - weierstrass - form ] ) ) : @xmath83 now @xmath84 has the zeros @xmath85 and @xmath86 . we next apply a partial fractions decomposition upon equation ( [ eqn : sing - ix ] ) : @xmath87 @xmath88 are constants which arise from the partial fractions decomposition and depend on the parameters of the metric and the test particle . then we substitute @xmath89 with @xmath90 . equation ( [ eqn : sing - ix - partial ] ) now simplifies to @xmath91 with @xmath92 and @xmath93 . after solving the integrals of the third kind ( see e.g. @xcite ) , the final solution reads @xmath94 with @xmath95 . the index @xmath35 refers to the weierstra - functions with respect to the parameters @xmath96 and @xmath97 . using ( [ eqn : sing - y - gleichung ] ) equation ( [ eqn : sing - psi - gleichung ] ) becomes @xmath98 @xmath99 the @xmath8-equation can be solved analogously to the @xmath7-equation . with @xmath100 , @xmath93 and @xmath95 the solution is @xmath101 @xmath102 are constants which arise from the partial fractions decomposition and depend on the parameters of the metric and the test particle . the index @xmath36 refers to the weierstra - functions with respect to the parameters @xmath103 and @xmath104 . using ( [ eqn : sing - y - gleichung ] ) we can write ( [ eqn : sing - t - gleichung ] ) as @xmath105 @xmath106 the @xmath25-equation can be solved in an analogous way to the @xmath7- and @xmath8-equation . we substitute @xmath107 . the integral ( [ eqn : sing - em0-tint ] ) is of the third kind and has the poles @xmath108 and @xmath109 ( with respect to @xmath110 ) . then we apply a partial fractions decomposition , where the constants @xmath111 arise and we substitute again @xmath112 . after the solution of the occurring elliptic integrals of the third kind , the solution of ( [ eqn : sing - t - gleichung ] ) yields @xmath113 in the ergosphere of a singly spinning black ring only tos with @xmath1 are possible . figure [ pic : sing - to ] shows a to plotted in the @xmath35-@xmath36-plane in cartesian coordinates ( @xmath114 , @xmath115 ) . to change from the ring coordinates to the polar coordinates ( @xmath116 , @xmath117 ) the transformation @xmath118 is used . then conventional cartesian coordinates take the form @xmath119 ( see @xcite or @xcite ) . the singularity of the black ring is at @xmath120 , @xmath121 . , @xmath122 , @xmath123 and @xmath124 @xmath114-@xmath115-plot of a to for the singly spinning black ring in the case @xmath1 . the black dashed circles are the event horizon and the red dotted circles denote the ergosphere.,width=377 ] the surface @xmath43 is the axis of rotation of the singly spinning black ring . here the hamilton - jacobi equation depends on the coordinate @xmath35 only . we set @xmath43 , @xmath125 and @xmath126 in the hamilton - jacobi equation ( [ eqn : hjd - sring ] ) : @xmath127 \ , .\ ] ] this can be rearranged to @xmath128 - \frac{h(x)}{g^2(x)}\phi^2 : = x_s .\ ] ] then we have @xmath129 now we set the partial derivatives of @xmath130 with respect to the constants @xmath131 , @xmath32 and @xmath33 to zero in order to obtain the equations of motion . with the mino - time @xcite @xmath132 the equations of motion take the form @xmath133 where @xmath60 is a polynomial of fifth order . from ( [ eqn : sing - psi - x - gleichung ] ) we can read off the effective potential consisting of two parts @xmath134 and @xmath135 ( to be consistent with the effective potential on the equatorial plane later on ) : @xmath136 since @xmath60 can be written as @xmath137 the effective potential takes the form @xmath138 where @xmath139 and @xmath140 . figure [ pic : s - psi - orbits1 ] shows the effective potential for the motion on the @xmath8 axis . @xmath141 is plotted in red ( solid ) while @xmath142 is plotted in blue ( dotted ) . the grey area between the two parts of the potential is a forbidden zone where no motion is possible because @xmath60 becomes negative there . @xmath141 and @xmath142 are symmetric ( @xmath143 ) and meet at the horizon . @xmath144 is always a zero of @xmath60 , but since the point @xmath144 , @xmath43 corresponds to infinity in cartesian coordinates it is not a real turning point of the test particle . if @xmath145 then @xmath60 has the zeros @xmath144 and @xmath146 and possibly a third zero between -1 and + 1 . the coordinate range of @xmath35 ( @xmath147 ) only covers the space from infinity ( -1 ) to the center of the black ring ( + 1 ) . since there is no potential barrier at @xmath146 for @xmath145 , light and test particles with the right amount of energy cross the center of the black ring and continue their orbit at the other side of the black ring . if @xmath145 then none or one turning point exists . if @xmath148 there is a potential barrier which prevents the geodesics from reaching @xmath146 and one or two turning points exist . for larger @xmath13 and @xmath149 the potential has local extrema which lead to three turning points . possible orbits are bound orbits ( bo ) , where light or test particles circle the black ring , and escape orbits ( eo ) , where light or test particles approach the black ring , turn around at a certain point and escape the gravitational field . there are five different types of orbits ( see table [ tab : s - psi - typen - orbits1 ] ) . * type a : + @xmath60 has no zero in the range @xmath150 . eos without a turning point exist . the orbit crosses the equatorial plane ( @xmath146 ) and reaches infinity ( @xmath144 and @xmath43 ) . * type b : + @xmath60 has one zero in the range @xmath150 . bos with a turning point on each side of the ring exist , so that the orbit crosses the equatorial plane ( @xmath146 ) . * type c : + @xmath60 has one zero in the range @xmath150 . eos with a turning point exist . * type d : + @xmath60 has two zeros in the range @xmath150 . bos which do not cross the equatorial plane exist . * type e : + @xmath60 has three zeros in the range @xmath150 . bos which do not cross the equatorial plane and eos exist . a & 0 & ( -2.5,-0.2)(3,0.2)(-2.5,0)(3,0 ) ( -2.5,0)(3,0 ) & eo + b & 1 & ( -2.5,-0.2)(3,0.2)(-2.5,0)(3,0 ) ( -1.5,0)(3,0 ) & bo + c & 1 & ( -2.5,-0.2)(3,0.2)(-2.5,0)(3,0 ) ( -2.5,0)(2,0 ) & eo + d & 2 & ( -2.5,-0.2)(3,0.2)(-2.5,0)(3,0 ) ( -1,0)(2,0 ) & bo + e & 3 & ( -2.5,-0.2)(3,0.2)(-2.5,0)(3,0 ) ( -0,0)(2,0 ) ( -2.5,0)(-1.0,0 ) & eo , bo + equation ( [ eqn : sing - psi - x - gleichung ] ) can be written as @xmath151 where @xmath60 is a polynomial of fifth order with the coefficients @xmath1522\lambda e^2}{(1-\lambda)^2 } - 2r^2\lambda^2 \left ( \frac{(1+\lambda^2)}{(1-\lambda)^2}e^2 -m^2 \right ) \nonumber\\ a_{x,3 } & = & 2\phi^2\lambda + \frac{[2r^2\lambda^2-r^2(1+\lambda^2)]2\lambda e^2}{(1-\lambda)^2 } + ( -2r^2\lambda - r^2(1+\lambda^2)\lambda ) \left ( \frac{(1+\lambda^2)}{(1-\lambda)^2}e^2 -m^2 \right ) \nonumber\\ a_{x,2 } & = & -4 \phi^2 \lambda-\phi^2 ( 1+\lambda^2 ) + \frac{[2r^2\lambda^2+r^2(1+\lambda^2)]2\lambda e^2}{(1-\lambda)^2 } + ( 2r^2\lambda - r^2(1+\lambda^2 ) ) \left ( \frac{(1+\lambda^2)}{(1-\lambda)^2}e^2 -m^2 \right ) \nonumber\\ a_{x,1 } & = & -2\phi^2\lambda-2\phi^2(1+\lambda^2 ) + \frac{2r^2(1+\lambda^2)\lambda e^2}{(1-\lambda)^2 } + ( 2r^2\lambda + r^2(1+\lambda^2)\lambda ) \left ( \frac{(1+\lambda^2)}{(1-\lambda)^2}e^2 -m^2 \right ) \nonumber\\ a_{x,0 } & = & -\phi^2(1+\lambda^2 ) + r^2(1+\lambda^2)\left ( \frac{(1+\lambda^2)}{(1-\lambda)^2}e^2 -m^2 \right ) . \label{eqn : sing - xkoeff}\end{aligned}\ ] ] a separation of variables gives the hyperelliptic integral @xmath153 a canonical basis of holomorphic ( @xmath154 ) and meromorphic ( @xmath155 ) differentials associated with the hyperelliptic curve @xmath156 is given by ( see @xcite or @xcite ) @xmath157 @xmath158 furthermore we introduce the holomorphic and meromorphic period matrices @xmath159 and @xmath160 : @xmath161 with @xmath162 , where @xmath163 is the canonical basis of closed paths . the normalized holomorphic differentials are @xmath164 the solution of equation ( [ eqn : sing - psi - x - quadrgleichung ] ) is extensively discussed in @xcite , and is given by the derivatives @xmath165 of the kleinian sigma function @xmath166 : @xmath167 where @xmath168 the constant @xmath169 depends on @xmath170 and @xmath171 only . @xmath172 is defined by the vanishing condition of the kleinian sigma function @xmath173 so that @xmath174 is an element of the theta divisor @xmath175 ( the set of zeros of the theta function ) where @xmath176 is the vector of riemann constants and @xmath30 is the riemann period matrix defined as @xmath177 . with ( [ eqn : sing - psi - x - gleichung ] ) equation ( [ eqn : sing - psi - phi - gleichung ] ) yields @xmath178 or @xmath179 the integral ( [ eqn : psi - phiint ] ) has poles at @xmath180 and @xmath181 . now we apply a partial fractions decomposition upon ( [ eqn : psi - phiint ] ) : @xmath182 where @xmath183 are constants which arise from the partial fractions decomposition . the differentials in the equation above are of the third kind and can be solved with the help of the following equation ( see @xcite ) . @xmath184\\ & + \ln \frac{\sigma\left(\int_{\infty}^p \mathrm{d}\boldsymbol{u}- \int_{(e_2,0)}^{(z , w ) } \mathrm{d}\boldsymbol{u } - \boldsymbol{k}_\infty \right)}{\sigma\left(\int_{\infty}^p \mathrm{d}\boldsymbol{u}+ \int_{(e_2,0)}^{(z , w ) } \mathrm{d}\boldsymbol{u } - \boldsymbol{k}_\infty \right ) } - \mathrm{ln } \frac{\sigma\left(\int_{\infty}^{p ' } \mathrm{d}\boldsymbol{u } - \int_{(e_2,0)}^{(z , w ) } \mathrm{d}\boldsymbol{u } - \boldsymbol{k}_\infty \right)}{\sigma\left(\int_{\infty}^{p ' } \mathrm{d}\boldsymbol{u } + \int_{(e_2,0)}^{(z , w ) } \mathrm{d}\boldsymbol{u } - \boldsymbol{k}_\infty \right)}.\end{split } \label{main1 - 2}\end{aligned}\ ] ] @xmath185 and @xmath186 are points on the hyperelliptic curve , @xmath187 is a pole , @xmath188 and @xmath156 . the zeros @xmath189 of @xmath190 are the branch points of the curve @xmath191 . @xmath192 is the vector of the holomorphic differentials of the first kind @xmath193 with @xmath194 . @xmath195 and @xmath196 are kleinian functions and @xmath197 is the vector of riemann constants . the vector @xmath198 identified with each branch point @xmath189 is defined as @xcite @xmath199 with the vectors @xmath200 and @xmath201 whose entries @xmath202 , @xmath203 are @xmath204 or @xmath205 for every @xmath206 , @xmath207 . the matrix @xmath208 = \left [ \begin{array}{c } \boldsymbol{\varepsilon}_i'\\ \boldsymbol{\varepsilon}_i \end{array } \right]\ ] ] is called the characteristic of a branch point @xmath189 . + the @xmath209th component ( in this case genus @xmath210 ) of the vector @xmath211 is @xmath212 and for @xmath213 we have @xmath214 the @xmath215 are elementary symmetric functions of order @xmath216 built on @xmath217 branch points @xmath218 : @xmath219 , @xmath220 , etc . + then the solution of the @xmath7-equation reads @xmath221 \end{split}\ ] ] where @xmath222 and @xmath223 . with ( [ eqn : sing - psi - x - gleichung ] ) equation ( [ eqn : sing - psi - t - gleichung ] ) yields @xmath224 or @xmath225 next we apply a partial fractions decomposition upon ( [ eqn : psi - tint ] ) where the constants @xmath226 arise . @xmath227 first we will solve the holomorphic integrals @xmath228 . we introduce a variable @xmath229 so that @xmath230 . the inversion of this integral yields @xmath231 ( see section [ sec : sing - x - solution ] or @xcite ) , where @xmath232 the function @xmath233 can be found from the condition @xmath234 . equation ( [ eqn : sing - t - pbz ] ) now reads @xmath235 \ , .\ ] ] the remaining differential is of the third kind . its solution is presented in ( [ main1 - 2 ] ) . then the solution of the @xmath25-equation ( [ eqn : sing - psi - t - gleichung ] ) is @xmath236 + k_2(v - v_0 ) + k_3f_1(v - v_0 ) \right\rbrace \\ \end{split}\ ] ] where @xmath237 . on the rotational axis of the singly spinning black ring bound orbits and escape orbits are possible . the orbits either move around the ring or move directly through the center of the ring . figure [ pic : sing - psi - bo ] and [ pic : sing - psi - bo2 ] show bound orbits and figure [ pic : sing - psi - eo ] and [ pic : sing - psi - eo2 ] show escape orbits in @xmath114-@xmath115-coordinates ( see section [ sec : em0-orbit ] ) and the corresponding solution @xmath238 . since the orbits on the rotational axis are presented by lines in the @xmath35-@xmath36-plane , we will also show them in the @xmath35-@xmath7-plane where we will use the coordinates @xmath239 and @xmath7 . one can think of ring coordinates as two pairs of polar coordinates @xmath240 where @xmath241 ( see @xcite ) . if @xmath8 is constant , the horizon of the black ring consists of two @xmath242 spheres . if we look at the rotational axis where @xmath43 , the coordinates @xmath243 and @xmath244 describe the plane between these two spheres , so the horizon can not be seen in this plane . if @xmath7 is constant , the horizon has @xmath245 topology . so if @xmath246 the coordinates @xmath247 and @xmath248 describe the equatorial plane `` as seen from above '' . + the bound orbits of type b and the escape orbits of type a , which move through the center of the black ring , are lines in every plane since here both angles @xmath7 and @xmath8 are constant . so in that case we only show the @xmath114-@xmath115-plot . the surface @xmath42 is the equatorial plane of the black ring , which is divided into two parts . the first part @xmath146 is the plane enclosed by the ring ( or more precisely : enclosed by the singularity ) , which we will refer to as `` inside '' the ring . the second part @xmath144 describes the equatorial plane around the black ring ( or more precisely : around the singularity ) , which we will refer to as `` outside '' the ring . if we set @xmath42 , @xmath249 and @xmath250 in the hamilton - jacobi equation ( [ eqn : hjd - sring ] ) , it depends on the coordinate @xmath36 only : @xmath251 \ , .\ ] ] this can be rearranged to @xmath252 - \frac{h(y)}{g^2(y)}(\psi + \omega_\psi e)^2 : = y_s \ , .\ ] ] then we have @xmath253 now we set the derivatives of @xmath130 with respect to the constants @xmath131 , @xmath32 and @xmath34 to zero in order to obtain the equations of motion . with the mino - time @xcite @xmath254 and the relation @xmath255 the equations of motion take the form @xmath256 - \frac{(\pm 1-y)^2h(y)}{(1\pm\lambda)^4 } [ h(y)\psi + \omega_\psi]^2 \right\rbrace ^{1/2 } \nonumber\\ & : = & \sqrt{y(y ) } \ , , \label{eqn : sing - phi - y - gleichung}\\ \frac{\mathrm{d}\psi}{\mathrm{d}\gamma } & = & -\frac{(\psi + \omega_\psi e)(\pm 1-y)h(y)}{(1\pm\lambda)^2g(y ) } \ , , \label{eqn : sing - phi - psi - gleichung}\\ \frac{\mathrm{d}t}{\mathrm{d}\gamma } & = & \frac{r^2e}{(\pm 1-y)h(y ) } + \frac{(\omega_\psi\psi+\omega_\psi^2e)(\pm 1-y)h(y)}{(1\pm\lambda)^2g(y ) } \label{eqn : sing - phi - t - gleichung } \ , .\end{aligned}\ ] ] it might not be obvious at first glance , but @xmath62 is a polynomial of third order in @xmath36 and therefore the equations of motion are of elliptic type . from ( [ eqn : sing - phi - y - gleichung ] ) we can read off an effective potential consisting of the two parts @xmath257 and @xmath258 : @xmath259 since @xmath62 can be written as @xmath260 the effective potential takes the form @xmath261 @xmath262 the two cases @xmath146 ( geodesics inside the ring ) and @xmath144 ( geodesics outside the ring ) have to be discussed separately . let us first take a look at the motion on the surface outside the black ring . here we have @xmath144 . figure [ pic : sing - phi - orbits1 ] shows the effective potential @xmath53 for different values of the parameters . @xmath263 and @xmath264 meet at the horizon . mainly the angular momentum @xmath34 defines the shape of the effective potential . for @xmath125 the potential is symmetric and @xmath62 has none or one zero . if @xmath265 the potential is no longer symmetric and if @xmath266 is large enough up to two zeros of @xmath62 are possible . possible orbits are terminating orbits ( to ) with or without a turning point , where light or test particles cross the horizon and fall into the singularity , and escape orbits ( eo ) , where light or test particles aproach the black ring , turn around at a certain point and escape the gravitational field . the zero of @xmath62 of a to can lie directly on the event horizon . there are three different types of orbits ( see table [ tab : sing - phi - typen - orbits1 ] ) . * type a : + @xmath62 has no zeros and only tos exist . * type b : + @xmath62 has one zero and only tos exist . in a special case the zero of @xmath62 lies on the horizon . * type c : + @xmath62 has two zeros . tos and eos exist . in a special case the zero of @xmath62 lies on the horizon . a & 0 & ( -2.5,-0.2)(3,0.2)(-2.5,0)(3,0 ) ( 0.5,-0.2)(0.5,0.2 ) ( -2.5,0)(3,0 ) & to + b & 1 & ( -2.5,-0.2)(3,0.2)(-2.5,0)(3,0 ) ( 0.5,-0.2)(0.5,0.2 ) ( -2.5,0)(1,0 ) & to + b@xmath267 & 1 & ( -2.5,-0.2)(3,0.2)(-2.5,0)(3,0 ) ( 0.5,-0.2)(0.5,0.2 ) ( -2.5,0)(0.5,0 ) & to + c & 2 & ( -2.5,-0.2)(3,0.2)(-2.5,0)(3,0 ) ( 0.5,-0.2)(0.5,0.2 ) ( -2.5,0)(1,0 ) ( 2.0,0)(3,0 ) & to , eo + c@xmath267 & 2 & ( -2.5,-0.2)(3,0.2)(-2.5,0)(3,0 ) ( 0.5,-0.2)(0.5,0.2 ) ( -2.5,0)(0.5,0 ) ( 2.0,0)(3,0 ) & to , eo + the effective potential for geodesics on the surface enclosed by the black ring ( @xmath146 ) is shown in figure [ pic : sing - phi - orbits2 ] . again , if we have @xmath125 the potential is symmetric and @xmath62 has none or one zeros . in the case @xmath265 a potential barrier appears which prevents test particles and light from reaching @xmath43 . then @xmath62 has always a single zero in the allowed range of @xmath36 . the higher @xmath266 the higher the energy where @xmath263 and @xmath264 meet . note that @xmath146 , @xmath43 is the location of the center of the black ring . possible orbits are terminating orbits where @xmath62 has one zero ( type b ) or none zero ( type a ) . for the type of orbits see previous section . in a special case the zero of @xmath62 lies on the horizon . ( type b@xmath267 ) . , @xmath268 , @xmath122 und @xmath145 effective potentials @xmath257 ( red , solid ) and @xmath258 ( blue , dotted ) on the equatorial plane inside the ring . the grey area is a forbidden zone , where no motion is possible . the horizon is marked by a vertical dashed line . green dashed lines represent energies and green points mark the turning points . possible orbits are of type a , b and b@xmath267.,width=226 ] equation ( [ eqn : sing - phi - y - gleichung ] ) can be solved analogously to ( [ eqn : sing - x - gleichung ] ) . ( [ eqn : sing - phi - y - gleichung ] ) can be written as @xmath269 where @xmath270 the solution is ( see section [ sec : ergo - xsol ] ) @xmath271 , \ ] ] where @xmath272 and @xmath273 . the coefficients @xmath274 and @xmath275 of the polynomial in the weierstra form are @xmath276 with ( [ eqn : sing - phi - y - gleichung ] ) equation ( [ eqn : sing - phi - psi - gleichung ] ) yields @xmath277 or @xmath278 this can be rewritten as @xmath279 this equation can be solved analogously to the @xmath7- and @xmath8-equation for nullgeodesics in the ergosphere ( see section [ sec : ergo - phisol ] and [ sec : ergo - psisol ] ) . with @xmath280 , @xmath93 and @xmath95 the solution is @xmath281 @xmath102 are constants which arise from the partial fractions decomposition and depend on the parameters of the metric and the test particle . with ( [ eqn : sing - phi - y - gleichung ] ) equation ( [ eqn : sing - phi - t - gleichung ] ) yields @xmath282}{(1\pm\lambda)^2(1-y)(1+\lambda y)h(y ) } \right ) \frac{\mathrm{d}y}{\sqrt{y(y)}}\ ] ] or @xmath283}{(1\pm\lambda)^2(1-y')(1+\lambda y')h(y ' ) } \right)\ , \frac{\mathrm{d}y'}{\sqrt{y(y ' ) } } \ , . \label{eqn : phi - tint}\ ] ] this equation can be solved analogously to the @xmath7- and @xmath8-equation for nullgeodesics in the ergosphere ( see section [ sec : ergo - phisol ] and [ sec : ergo - psisol ] ) . with @xmath280 , @xmath93 and @xmath284 the solution is @xmath285 @xmath111 are constants which arise from the partial fractions decomposition and depend on the parameters of the metric and the test particle . on the equatorial plane around the singly spinning black ring terminating orbits and escape orbits are possible . figure [ pic : sing - phiout - eo ] and [ pic : sing - phiout - to ] show some orbits in @xmath114-@xmath115-coordinates ( see section [ sec : em0-orbit ] ) and the corresponding solution @xmath286 . also the @xmath36-@xmath8-plane is shown in the coordinates @xmath247 and @xmath248 ( see section [ sec : psi - axis - orbits ] ) . an escape orbit is depicted in figure [ pic : sing - phiout - eo ] . figure [ pic : sing - phiout - to ] shows a terminating orbit which starts at its turning point and then falls into the singularity . the frame dragging effect can be seen in figure [ pic : framedragg ] . once the particle enters the ergosphere it is dragged along by the rotation of the black ring . if the angular momentum of the particle and the black ring have opposite signs , the particle changes its direction when approaching the ergosphere . + on the equatorial plane enclosed by the black ring only terminating orbits are possible . figure [ pic : sing - phiin - to ] shows a terminating orbit which starts at the center of the black ring and then falls into the singularity . in this paper we presented the analytical solutions of the geodesic equations of the singly spinning black ring for special cases . since the hamilton - jacobi equation seems not to be separable in general , we had to concentrate on the nullgeodesics in the ergosphere ( @xmath1 ) , geodesics on the rotational axis ( @xmath43 ) and geodesics on the equatorial plane ( @xmath246 ) . + we discussed the general structure of the orbits and gave a complete classification of their types . in the ergosphere there is just one possible orbit , where light crosses the event horizon and falls inevitably into the singularity ( terminating orbit ) . the @xmath35-motion bounces back and forth between two values or stays constant at @xmath61 , while @xmath36 ranges from a turning point to @xmath287 . on the rotational axis @xmath36 is constant , so here the @xmath35-motion determines the type of orbit . we found escape orbits and bound orbits , the latter were also shown numerically by igata et al . @xcite . on the equatorial plane we found terminating orbits and escape orbits . + the separability of the hamilton - jacobi equation is a coordinate related phenomenon , so one might think of a coordinate system in which it would be possible to separate the hamilton - jacobi equation in general . but recently igata , ishihara and takamori found evidence of chaotic motion in the singly spinning black ring spacetime using the poincar map @xcite . from that one could conclude that it is not possible to separate the hamilton - jacobi equation in any coordinate system . + besides the singly spinning black ring , one could consider black rings with two angular momenta ( doubly spinning black ring @xcite ) or add charge to the black ring @xcite . also a supersymmetric black ring solution was found @xcite , @xcite . the methods shown in this paper can be applied to ( charged ) doubly spinning black rings as well as to supersymmetric black rings . this will be done in future work . we would like to thank victor enolski , norman grlebeck and volker perlick for helpful discussions . we gratefully acknowledge support by the dfg , in particular , also within the dfg research training group 1620 `` models of gravity '' . r. emparan and h. s. reall , living rev . rel . * 11 * , 6 ( 2008 ) [ arxiv:0801.3471 [ hep - th ] ] . r. c. myers and m. j. perry , annals phys . * 172 * , 304 ( 1986 ) . r. emparan and h. s. reall , phys . lett . * 88 * , 101101 ( 2002 ) [ hep - th/0110260 ] . d. kubiznak and v. p. frolov , class . * 24 * , f1 ( 2007 ) [ gr - qc/0610144 ] . d. n. page , d. kubiznak , m. vasudevan and p. krtous , phys . lett . * 98 * , 061102 ( 2007 ) [ hep - th/0611083 ] . v. p. frolov , p. krtous and d. kubiznak , jhep * 0702 * , 005 ( 2007 ) [ hep - th/0611245 ] . j. hoskisson , phys . d * 78 * , 064039 ( 2008 ) [ arxiv:0705.0117 [ hep - th ] ] . m. durkee , class . * 26 * , 085016 ( 2009 ) [ arxiv:0812.0235 [ gr - qc ] ] . h. elvang , r. emparan and a. virmani , jhep * 0612 * , 074 ( 2006 ) [ hep - th/0608076 ] . a. a. pomeransky and r. a. senkov , hep - th/0612005 . j. armas , class . * 28 * , 235014 ( 2011 ) [ arxiv:1011.5618 [ hep - th ] ] . t. igata , h. ishihara and y. takamori , phys . d * 82 * , 101501 ( 2010 ) [ arxiv:1006.3129 [ hep - th ] ] . y. hagihara , em jpn . j. astron . * 8 * , 67 ( 1931 ) e. hackmann and c. lmmerzahl , phys . d * 78 * , 024035 ( 2008 ) . v. z. enolski , e. hackmann , v. kagramanova , j. kunz and c. lmmerzahl , j. geom . phys . * 61 * , 899 ( 2011 ) [ arxiv:1011.6459 [ gr - qc ] ] . e. hackmann , v. kagramanova , j. kunz and c. lmmerzahl , phys . d * 78 * , 124018 ( 2008 ) [ erratum - ibid . * 79 * , 029901 ( 2009 ) ] [ arxiv:0812.2428 [ gr - qc ] ] . v. z. enolski , m. pronine , p.h . richter , _ j. nonlinear sci . * 13 * , pp . 157 - 174 ( 2003 ) r. emparan and h. s. reall , class . * 23 * , r169 ( 2006 ) [ hep - th/0608012 ] . y. mino , phys . rev . d * 67 * , 084027 ( 2003 ) [ gr - qc/0302075 ] . a. i. markushevich , _ theory of functions of a complex variable _ ( prentice - hall , englewood cliffs , nj , 1967 ) , vol . v. enolski , b. hartmann , v. kagramanova , j. kunz , c. lmmerzahl and p. sirimachan , journal of mathematical physics * 53 * , 012504 ( 2012 ) [ arxiv:1106.2408 [ gr - qc ] ] . y. k. lim , e. teo , _ geodesics through the black ring _ , report , university of singapore , http://staff.science.nus.edu.sg/ phylyk / downloads / reports / blackrings_report.pdf ( 2008 ) t. igata , h. ishihara and y. takamori , phys . d * 83 * , 047501 ( 2011 ) [ arxiv:1012.5725 [ hep - th ] ] . h. elvang , phys . d * 68 * , 124016 ( 2003 ) [ hep - th/0305247 ] . j. hoskisson , phys . d * 79 * , 104022 ( 2009 ) [ arxiv:0808.3000 [ hep - th ] ] . d. v. galtsov and n. g. scherbluk , phys . d * 81 * , 044028 ( 2010 ) [ arxiv:0912.2771 [ hep - th ] ] . h. elvang , r. emparan , d. mateos and h. s. reall , phys . lett . * 93 * , 211302 ( 2004 ) [ hep - th/0407065 ] . h. elvang , r. emparan , d. mateos and h. s. reall , phys . d * 71 * , 024033 ( 2005 ) [ hep - th/0408120 ] .

quantum chromodynamics , the theory describing the interactions of quarks and gluons , has been shown to provide excellent descriptions of a wide variety of phenomena ranging from hadron spectroscopy via lattice field theory to hard scattering phenomena via perturbation theory . yet qcd is a theory whose degrees of freedom are the quarks and gluons quanta that can not be observed directly due to their confinement in hadrons . on the other hand , experiments deal with hadrons and leptons . in order to calculate observables measured in high energy scattering processes a technique is needed that allows conversion from a description in terms of hadrons to one in terms of quarks and gluons ( collectively referred to as partons ) . for hadrons in the initial state this conversion is provided by parton distribution functions ( pdfs ) , while for final state hadrons a similar role is played by parton fragmentation functions . the pdfs , denoted by @xmath0 , for each type of quark and antiquark and for the gluon ( @xmath1 ) , allow one to essentially describe a beam of hadrons as an effective beam of quarks and gluons . in the infinite momentum frame , the pdfs can be interpreted as probability densities describing how the parent hadron s momentum is shared amongst the different types of partons , as a function of the hadron s momentum fraction @xmath2 . with knowledge of the pdfs and the feynman rules for qcd , one can calculate hard scattering cross sections and differential distributions that can then be compared to data . pdfs are traditionally determined by simultaneously fitting a wide variety of data for large momentum transfer processes . the parameters of the fits describe the pdfs at some initial momentum transfer scale , while evolution equations are then used to calculate the pdfs at all other scales needed for the calculations . the data sets used for such fits often include deep inelastic scattering ( dis ) of charged leptons on proton and deuterium targets , or neutrinos on heavy nuclear targets , lepton pair production on proton and deuterium targets , and the production of photons , vector bosons , or jets at large values of transverse momentum . while the pdfs provide us with a detailed description of the partonic substructure of hadrons , they contain only partial information . partons ( both quarks and gluons ) have nonzero spin , so that the fundamental distributions in nature are the pdfs for a specific helicity ( spin projection along the direction of motion ) , @xmath3 and @xmath4 , corresponding to parton spins aligned and antialigned with that of the hadron , respectively . experiments with unpolarized beams and targets are therefore sensitive only to the sums of the helicity pdfs , @xmath5 , while information on the differences , @xmath6 , can be obtained from measurements involving polarized beams and/or targets . it is the purpose of this topical review to give a current `` snapshot '' of the status of our knowledge of pdfs , both unpolarized and polarized . to this end , brief reviews of both the theory and the types of data used in the fits will be presented , and theoretical issues that need to be addressed in the perturbative calculations will be outlined . throughout this review we shall attempt to answer the question `` what do we know and how do we know it ? '' recently there have been several excellent reviews of pdfs @xcite , updating the early treatises of pdfs in refs . @xcite , which have focused on various aspects of unpolarized pdfs , such as the impact on the phenomenology of the large hadron collider ( lhc ) . the spin structure of the nucleon , including spin - dependent pdfs , was also reviewed in refs . the present pedagogical review should be viewed as complementary to these efforts , emphasizing the different aspects of hadron structure that can be revealed through the study of unpolarized and polarized pdfs , at both small and large parton momentum fractions @xmath2 . there is an old saying in experimental particle physics that `` today s discovery is tomorrow s calibration . '' a similar effect occurs for theoretical calculations . what was once touted as a test of qcd , now often plays the role of providing a description of backgrounds for searches for new types of particles . of course , one wants these calculations to be as accurate as possible and this , in turn , requires precise knowledge of the pdfs that enter into the calculation . the error estimates on pdfs will be discussed at some length for this reason . on the other hand , there are many aspects of hadron structure that are currently not understood well , or at all , either theoretically or experimentally , and clearly identifying and delineating the limits of our knowledge in the context of global pdf fits serves a valuable purpose . the plan of the review is as follows . sec . 2 contains an overview of the necessary qcd theory required to understand the results of global fits . sec . 3 contains the review of unpolarized pdfs , while sec . 4 summarizes the status of polarized pdfs . finally , sec . 5 provides some speculations as to how our knowledge of pdfs may be improved through anticipated results of ongoing and future experiments . as noted in the introduction , pdfs are necessary ingredients for obtaining predictions for hard scattering hadron hadron and lepton hadron processes . the cross section for a typical hadron hadron process involving collisions of hadrons @xmath7 and @xmath8 , with momenta @xmath9 and @xmath10 , respectively , producing a state @xmath11 in addition to other hadrons ( collectively denoted by @xmath12 ) , can be written in the form @xmath13 where @xmath14 denotes an @xmath15-th order parton parton `` cross section '' that produces the desired final state , and the functions @xmath16 and @xmath17 are the pdfs of flavor @xmath18 in hadron @xmath7 and flavor @xmath19 in hadron @xmath8 , respectively . the state @xmath11 might denote a high-@xmath20 jet , a lepton pair , a photon or weak vector boson , for example , and @xmath21 is a scale that characterizes the hard scattering . for high-@xmath20 jet or photon production the scale @xmath22 . the form ( [ eq : xsec ] ) can also be applied to the case when the hadrons @xmath7 and @xmath8 are polarized ; in the remainder of this section , however , we shall for illustration focus primarily on the unpolarized case . the parton - level cross sections will generally possess ultraviolet singularities that must be renormalized . doing so leads to the introduction of the running coupling @xmath23 which depends on a renormalization scale @xmath24 . there can also be infrared singularities associated with loop graphs ; these will cancel corresponding singularities from the emission of real soft gluons , provided that the observable is suitably defined , _ i.e. , _ that it is `` infrared safe . '' finally , there will be collinear singularities associated with , in this case , the initial partons emitting additional partons at zero angle . these collinear configurations correspond to internal propagators going on - shell and , as such , correspond to long - distance physics . such singular terms can be _ factorized _ and absorbed into the pdfs . this process introduces a factorization scale , @xmath25 , that separates the long - distance and short - distance hard scattering physics . as shown in eq . ( [ eq : xsec ] ) , the partonic hard scattering cross section can be expanded in powers of the running coupling . the dependence on @xmath24 that enters via the running coupling cancels that which appears in the partonic cross sections . similarly , the dependence on @xmath25 in the pdfs cancels against the @xmath25 dependence in the partonic cross sections . these cancellations represent the fact that the physical cross section , if calculated to all orders in perturbation theory , should not depend on the two scales introduced to control the ultraviolet and collinear singularities . however , at any fixed order in perturbation theory these cancellations will only be approximate . indeed , at order @xmath26 one has the relations @xmath27 these results suggest that the dependence on the renormalization and factorization scale choices will decrease as one carries out the perturbative calculations to higher order . this is indeed the case for the processes typically used in global fits for pdfs . ( if a new kinematic configuration opens up at a higher order , then the corrections at that order can be large . this happens when going from lo to nlo for heavy quark production , for example , in which case the scale dependence can actually increase at that order . the reduction then occurs at the next order . ) these processes are all known to at least nlo and some to nnlo , as will be discussed below . in principle , the two scales @xmath24 and @xmath25 can be chosen independently . however , in processes where there is one large scale characterizing the hard scattering it is often the case that the choice @xmath28 is used . this simplifying choice will be utilized in the results discussed in the sections to follow . the dependence on the renormalization scale of the running coupling is governed by the qcd @xmath29 function via the renormalization group equation , @xmath30 the qcd @xmath29 function is calculable in perturbation theory and can be written as @xmath31 where the coefficients for @xmath32 flavors are @xmath33 the strong coupling @xmath34 calculated using the expression for @xmath29 in eq . ( [ eq : alphas ] ) with terms up to @xmath35 is referred to as the @xmath36loop running coupling . lo calculations retain only @xmath37 , while nlo calculations retain @xmath38 , and nnlo calculations retain also @xmath39 . note that the @xmath29 function is negative , resulting in a decrease of @xmath34 with increasing scale , a phenomenon known as asymptotic freedom . this behavior is essential for the applicability of perturbation theory for hard scattering processes . in eq . ( [ eq : xsec ] ) both the pdfs and the partonic cross sections depend on the factorization scale @xmath25 . as seen in eq . ( [ eq : scale ] ) , this scale dependence cancels up to the order of perturbation theory used . this fact allows one to calculate the @xmath41 dependence of the pdfs . indeed , the singularities associated with collinear parton emission are universal and factorize from the hard scattering subprocesses . the factorization scale dependence of the parton distributions is then also universal , reflecting the independence of physical quantities on the scale @xmath25 . this leads to the @xmath40 evolution , or renormalization group equations ( rge ) , @xmath42 where the bjorken @xmath2 variable is ( related to ) the longitudinal momentum fraction of the partons . the evolution kernels @xmath43 or `` splitting functions '' stem from the collinear divergences absorbed in the distributions and represent the ( collinear ) resolution of a parton @xmath44 in a parton @xmath45 . they are calculated perturbatively as a series expansion in @xmath46 , @xmath47 at present these have been computed up to 3 loops ( @xmath48 ) , which is the necessary accuracy for a nnlo analysis . the evolution of spin - dependent pdfs @xmath49 follows in a similar manner , with a corresponding set of spin - dependent splitting functions @xmath50 . in this case the splitting functions have been computed to 2 loops , enabling analyses to be performed at nlo accuracy . traditionally , information on the pdfs of the nucleon has come from the process of deep - inelastic scattering ( dis ) of leptons from protons or nuclei , beginning with the pioneering experiments at slac in the late 1960s . in the one - boson exchange approximation , the differential dis cross section can be written as a product of lepton and hadron tensors , @xmath51 where @xmath52 is the fine structure constant , @xmath53 is the laboratory solid angle of the scattered lepton , and @xmath54 is the incoming ( outgoing ) lepton energy . for neutral - currents , the summation is over @xmath55 and the interference @xmath56 , while for charged - currents only @xmath57 exchange contributes . the leptonic tensor @xmath58 depends on the charge @xmath59 and helicity @xmath60 of the lepton , @xmath61 where @xmath62 and @xmath63 are the initial and final electron momenta , and @xmath64 and @xmath65 are the electron vector and axial - vector charges , respectively . the factors @xmath66 in eq . ( [ eq : sigtens ] ) denote the ratios of the corresponding propagators and couplings to the photon propagator and coupling squared @xcite @xmath67 where @xmath68 is the fermi weak interaction coupling constant , and @xmath69 is the @xmath70 boson mass . the hadronic tensor @xmath71 contains all of the information about the structure of the hadron target . using constraints from lorentz and gauge invariance , together with parity conservation , it can be decomposed into spin - independent and spin - dependent contributions , @xmath72 , \label{eq : wmunu}\end{aligned}\ ] ] where @xmath73 and @xmath74 are the nucleon and exchanged boson four - momenta , @xmath75 , and @xmath76 . the nucleon polarization four - vector @xmath77 satisfies @xmath78 and @xmath79 . for spin - averaged scattering , the nucleon structure is parametrized in terms of the vector @xmath80 and @xmath81 structure functions , and the vector - axial vector interference @xmath82 structure function , which requires weak currents . these are generally functions of two variables ( such as @xmath2 and @xmath40 ) , but become functions of @xmath2 only in the bjorken limit , in which both @xmath40 and @xmath83 , but @xmath2 is fixed . in this limit the @xmath80 and @xmath81 structure functions become proportional , according to the callan - gross relation , @xmath84 , and in the parton model are given in terms of quark @xmath85 and antiquark @xmath86 distribution functions , @xmath87,\ ] ] where @xmath88 is interpreted as the probability to find a quark of flavor @xmath85 and charge @xmath89 in the nucleon with light - cone momentum fraction @xmath2 . at finite energies , the logarithmic @xmath40 dependence from the evolution equations described in sec . [ ssec : q2evol ] , as well as residual @xmath40 dependence associated with power corrections ( see sec . [ ssec : power ] below ) , give corrections to the simple parton model expectations . the spin - dependent structure functions @xmath90 and @xmath91 can be extracted from measurements where longitudinally polarized leptons are scattered from a target that is polarized either longitudinally or transversely relative to the electron beam . for longitudinal beam and target polarization , the difference between the cross sections for spins aligned and antialigned is dominated at high energy by the @xmath90 structure function . the @xmath91 structure function can be determined with additional measurement of cross sections for a nucleon polarized in a direction transverse to the beam polarization . in practice one often measures the polarization asymmetry @xmath92 , which is given as a ratio of spin - dependent and spin - averaged structure functions , @xmath93.\end{aligned}\ ] ] at small values of @xmath94 , one then has @xmath95 . if the @xmath40 dependence of the polarized and unpolarized structure functions is similar , the polarization asymmetry @xmath92 will be weakly dependent on @xmath40 . in analogy with the unpolarized @xmath80 structure function , in the parton modelthe structure function @xmath90 can be expressed at lo in terms of differences between quark distributions with spins aligned ( @xmath96 ) and antialigned ( @xmath97 ) relative to that of the nucleon , @xmath98 , @xmath99 . \label{eq : g1quark}\ ] ] the @xmath91 structure function , on the other hand , does not have a simple parton model interpretation . however , its measurement provides important information on the subleading , higher twist contributions which parametrize long - range nonperturbative parton - parton correlations in the nucleon ( see sec . [ ssec : power ] ) . as discussed in sec . [ sec : qcd ] , the pdfs are universal , the collinear singularities associated with initial parton collinear emission having been absorbed into the pdfs . therefore , the parton distributions appearing , for example , in the expressions for dis structure functions , are the same that describe the structure of the incoming hadrons in hadronic production . repeating the general form given in eq . ( [ eq : xsec ] ) ( with @xmath100 ) , one has , for spin - unpolarized scattering , @xmath101 and an analogous expression for the spin - dependent cross section , involving the difference of cross sections with hadrons @xmath7 and @xmath8 polarized in the same and opposite directions , @xmath102 here @xmath103 and @xmath104 are the spin - dependent pdfs for flavor @xmath18 in a hadron @xmath7 and flavor @xmath19 in a hadron @xmath8 , respectively , and @xmath105 is the corresponding spin - dependent partonic cross section . as indicated in eqs . ( [ eq : xsec2 ] ) and ( [ eq : xsec_pol ] ) , the partonic cross sections are calculable in fixed order perturbation theory as series expansions in @xmath106 , which starts with different powers depending on the process . it will be useful for subsequent sections to describe the parton kinematics for several examples of hadron hadron processes . although all processes used in global fits are known at least to nlo , the lo parton kinematics nevertheless serves as a useful guide to what region of parton momentum fraction and which combinations of parton flavors will be constrained by a given set of data . consider high-@xmath20 dijet production in lowest order where two jets are produced with approximately balancing transverse momenta @xmath20 and rapidities denoted by @xmath107 . two - body phase constrains the momentum fractions to be @xmath108 where the dimensionless ratio @xmath109 . for centrally produced dijets one sees that the parton momentum fractions are both approximately @xmath110 . however , if one or both of the jets is produced at far forward or backward rapidity values , then the @xmath2 range can be considerably expanded towards large or small values thereby allowing a `` tuning '' of the probed @xmath2 range . for the case of inclusive production of a jet with transverse momentum @xmath20 and rapidity @xmath111 there is , at lo , an integration over one of the parton momentum fractions , say @xmath112 , given by @xmath113 again , one sees that for @xmath111 near 0 the ranges of the momentum fractions are centered near @xmath110 . another important class of experiments involves the production of a system of mass @xmath114 , rapidity @xmath111 , and transverse momentum @xmath20 . examples of such systems include @xmath115 , @xmath116 , and charged lepton pairs @xmath117 . the four momentum of the produced system is conveniently given as @xmath118 where @xmath119 , and @xmath120 is the azimuthal angle . the rapidity @xmath111 and , equivalently , the feynman@xmath2 variable @xmath121 are defined by @xmath122 which lead to the relations @xcite @xmath123 where @xmath124 . depending on the detected final state it is possible to define different cross sections for hadron hadron collisions . in the drell - yan mechanism @xcite a quark from one hadron and an antiquark from the other one annihilate into an intermediate vector boson ( @xmath125 , @xmath116 or @xmath57 ) which subsequently decays into a lepton pair . for lepton pairs of invariant mass @xmath126 , the process is dominated by virtual photon exchange . the experiments usually consist of proton(beam)nucleon ( proton or some other nucleus ) collisions and the cross sections are extracted from the detection of muon pairs ( dimuon production ) produced in the decay of the virtual photons . the double differential cross section for lepton pair production at lo is given by @xmath127 , \label{eq : dy}\end{aligned}\ ] ] where the two parton momentum fractions are given by @xmath128 from eq . ( [ eq : dy ] ) one can see that for @xmath129 collisions this process is sensitive separately to sea and valence distributions , in contrast to the neutral current dis cross sections , where they enter as @xmath130 . inclusion of the drell yan data in global qcd analyses of pdfs is in fact instrumental in fixing ratio @xmath131 for @xmath132 at lo ( see eq . ( [ eq : sig_pd_pp ] ) below ) . another alternative is the use of charged current neutrino dis structure functions , although this process entails the use of model dependent corrections for the heavy nuclear targets typically employed in such measurements . nuclear corrections can be avoided if one instead uses the time reversed processes such as @xmath133 . a process closely related to drell yan dilepton production is the hadronic production of electroweak bosons @xcite . for @xmath134 scattering , the inclusive production cross section for @xmath57 bosons , for example , is given at lo by @xmath135 where @xmath136 are the ckm matrix elements , and the sum runs over all light quark and antiquark flavors in both hadrons . the production of @xmath137 bosons is then sensitive primarily to the products @xmath138 , while @xmath139 production is sensitive to @xmath140 . at large @xmath70 boson rapidity , or equivalently @xmath132 , @xmath57 production is sensitive to the @xmath141 and @xmath142 quark pdfs in the proton , respectively . provided the antiquark distributions at @xmath143 are known to sufficiently accuracy , the @xmath144 cross section ratio would then be a clean probe of the @xmath145 pdf ratio , which is presently not well determined at large @xmath112 . conversely , if the quark pdfs are known in particular regions of @xmath146 , then the @xmath147 ratios can also be used to probe the antiquark distributions . similar considerations apply when one or both of the protons are polarized , as for the case of the spin program at rhic . for collisions of longitudinally polarized protons from unpolarized protons , @xmath148 , the generalization of eq . ( [ eq : wprod ] ) simply involves replacing one of the quark or antiquark distributions by the corresponding polarized quark ( antiquark ) pdf , @xmath149 . single - spin asymmetries can then constrain ratios of polarized to unpolarized pdfs , @xmath150 , for various flavors , over specific regions of @xmath146 ( see sec . [ ssec : pol_sea ] ) . double - spin asymmetries , formed by combinations of inclusive hadron or jet production cross sections in the scattering of polarized protons from polarized protons , @xmath151 , have also been studied in order to constrain the polarized gluon distribution @xmath152 ( see sec . [ ssec : pol_gluon ] ) . the treatment of heavy quarks in high - energy processes generally requires more care than for light quarks . the production mechanism of heavy quarks in dis is predominantly the photon gluon fusion ( pgf ) process , which makes it particularly sensitive to the gluon distribution in the nucleon . although in principle there could exist an `` intrinsic '' ( initial state ) heavy quark content of the nucleon , existing measurements are well described through `` extrinsic '' ( generated in the hard scattering ) heavy quark production only ( _ cf . _ @xcite and references therein ) , so that any intrinsic heavy quark content of the nucleon is small and restricted to large values of bjorken @xmath2 ( see sec . [ ssec : hqresults ] below ) . the natural framework for describing the pgf mechanism is known as the `` fixed flavor number scheme '' ( ffns ) . in this approach the light - quark flavors ( @xmath153 ) and the gluon are considered as massless partons within the nucleon . this scheme allows one to compute the contributions of heavy quarks ( @xmath154 ) perturbatively as final - state quantum fluctuations , taking into account the full dependence of the production cross section on the mass @xmath155 of the heavy quark . as is common in the @xmath156 renormalization scheme , the evaluation of the strong running coupling @xmath106 is nevertheless based on the usual variable flavor number scheme for the @xmath29function governing its scale dependence . this procedure automatically resums the contributions to the strong coupling from heavy quarks , and consequently improves the stability of the perturbative expansion @xcite . complete calculations of ( heavy quark ) dis structure functions are presently known at lo @xcite and at nlo @xcite . beyond this , approximations based on threshold resummation have been made @xcite ( nnlo * in sec . [ ssec : hqresults ] ) , but thus far the nnlo coefficient functions are only partially known @xcite , which constitutes a major drawback of any precision qcd analysis at nnlo accuracy . another aspect of these calculations is the mass definition used ; while the original calculations of the amplitudes employ the pole mass definition , it has been reported @xcite that the use of the running mass definition improves the stability of the perturbative series . in many situations calculations within the ( fully massive ) ffns scheme become unduly complicated , and in many cases even impossible due to the unknown massive ( @xmath157 ) matrix elements at nnlo or even nlo . for this reason it is common to generate parton distributions in the so - called `` variable flavor number scheme '' ( vfns ) , where the heavy quarks are also considered to be massless constituents of the nucleon . here , the required nlo and nnlo cross sections have been computed for a variety of important production processes , in particular at hadron colliders . in the vfns scheme the effective heavy quark distributions are generated perturbatively from the nonperturbative distributions of light quarks and gluons using the boundary conditions of ref . @xcite , typically at the unphysical `` thresholds '' @xmath158 . from there one proceeds to their renormalization group evolution with an increased number of flavors @xmath159 and ( eventually ) 6 . for situations where the ( threshold ) invariant mass of the produced system far exceeds the mass of the interacting heavy flavor in the ffns , the vfns predictions have been shown @xcite to deviate from the ffns typically by about 10% , although the exact amount depends on the particular process and energy scale . this is usually within the margins of the renormalization and factorization scale uncertainties and other theoretical ambiguities related to pdfs . eventually , one nevertheless has to _ assume _ that these massless `` heavy '' quark distributions are relevant asymptotically , in that they can correctly describe the relevant cross sections at scales much larger than the heavy quark masses involved . unfortunately this is not the case for dis structure functions , for which the strictly massless approach , known in this context as the zero mass vfns , or zm - vfns , is well known to be experimentally inadequate , particularly near the heavy quark production thresholds . to remedy this the heavy - quark mass effects are re inserted in what are known as general - mass vfnss ( gm - vfns ) , for which several prescriptions are available in the literature @xcite ( see also ref . these schemes interpolate ( in a model - dependent way ) between the ffns results near production threshold and the asymptotic results of the zm - vfns . however , since the interpolating schemes are based on the same massive matrix elements as the ffns , they do not have complete information on the @xmath160 heavy flavor wilson coefficients ( even though they are sometimes referred to as `` nnlo gm - vfnss '' ) . the importance for global pdf analysis of finite heavy quark mass effects in the calculation of dis structure functions had been previously emphasized in refs . @xcite , but was only universally recognized after the analysis of @xcite . currently the abm @xcite and jr @xcite collaborations use the ffns for the calculation of dis structure functions in their global analyses , while the mstw @xcite , ct @xcite , herapdf @xcite and nnpdf @xcite groups use variants of gm - vfns . the calculations of hadron collider cross sections are carried out in the vfns in all cases . the elegant machinery that has been developed within the framework of perturbative qcd to analyze leading twist pdfs is , strictly speaking , valid only at high values of @xmath40 and @xmath70 where all hadron mass scales are suppressed , @xmath161 . in real experiments performed at a finite beam energy @xmath162 , however , the maximum values of @xmath40 and @xmath70 are limited , which inevitably restricts the available coverage in bjorken @xmath2 . this is especially relevant at large @xmath2 , where in dis the invariant mass squared of the produced hadronic system is given by @xmath163 with @xmath164 the mass of the target nucleon . for fixed @xmath40 , as @xmath165 the final state hadron mass @xmath70 decreases as one descends into the region dominated by nucleon resonances at @xmath166 gev . the resonance region may be treated using the concept of quark - hadron duality @xcite , although this goes beyond the scope of the usual pqcd analysis . in the region of low @xmath40 , power corrections to the bjorken limit results that scale as powers of @xmath167 become increasingly important . in the operator product expansion , these are associated with higher twist corrections , which arise from multi - parton correlations and characterize the long - range nonperturbative interactions between quarks and gluons . of tremendous interest in their own right as providing glimpses into the dynamics of quark confinement , the power corrections are viewed as troublesome backgrounds to efforts aimed solely at extracting leading twist pdfs . to avoid the complications from the higher twist corrections , the usual strategy in global pdf analyses is to apply cuts specifying minimum values of @xmath40 and @xmath168 . in many analyses of unpolarized scattering data , the cuts are of the order @xmath169 gev@xmath170 and @xmath171 gev@xmath170 @xcite , which in practice restrict the range of @xmath2 that can be accessed to @xmath172 . for spin - dependent pdfs , the scarcity of high - energy data forces most global analyses to use less restrictive cuts , of the order gev@xmath170 and @xmath174 gev@xmath170 . on the other hand , there are a number of important reasons for needing to know the large-@xmath2 behavior of pdfs , and several recent unpolarized pdf analyses @xcite have been performed with relaxed cuts of @xmath175 and @xmath176 gev@xmath170 which have allowed an expanded reach into large @xmath2 . the improvement in the large-@xmath2 kinematic coverage amounts to about 1300 more data points for proton and deuteron targets , representing some @xmath177 increase in the total number of dis data points compared with the more restrictive cuts . most importantly , the resulting leading twist pdfs fits have proved to be very stable even when such low cuts have been applied @xcite . with the inclusion of the kinematic regions in which subleading @xmath178 effects play a non - negligible role , it is obviously crucial to account for the power corrections that might otherwise obfuscate the leading twist pdfs . among the different categories of @xmath178 effects , the simplest are the target mass corrections ( tmcs ) , which are formally associated with matrix elements of leading twist operators @xcite and are of kinematical origin . others include genuine higher twist corrections , which arise from dynamical , multi - parton correlations , as well as higher order perturbative qcd corrections , which can also resemble power suppressed contributions at low @xmath40 . the standard method to compute tmcs in dis is based on the operator product expansion , and was first formulated by georgi and politzer @xcite , and expressions for all unpolarized and polarized structure functions now exist in both @xmath2 and mellin space @xcite . an alternative method based on collinear factorization ( cf ) in momentum space was developed by ellis , furmanski and petronzio @xcite , and extended by various authors @xcite ( see also the recent reviews of tmcs in refs . the advantage of the ope method is that tmcs can be calculated to all orders in @xmath178 in dis , whereas tmcs in the cf approach have only been computed to @xmath179 . on the other hand , the ope is limited to inclusive dis , while the cf framework can be applied to computing tmcs also in other processes @xcite . since typically the non - dis data are taken at very high @xmath40 where tmc effects are very small , the ope method is usually adopted . all of the tmc methods also suffer to some extent from the threshold problem , whereby the target mass corrected structure function remains nonzero as @xmath165 @xcite . for the purposes of global fits , however , the region where the threshold effects become problematic is @xmath180 gev @xcite , which is mostly outside of where even the most liberal cuts in @xmath70 and @xmath40 are made @xcite . from the seminal work of de rjula _ et al . _ @xcite , the appearance of the threshold problem in the analysis of tmcs is attributed to the neglect of dynamical higher twist corrections , both of which scale as powers in @xmath178 . ( in fact , higher order perturbative qcd corrections can also resemble power suppressed contributions at low @xmath40 . ) regardless of their origin , the various power suppressed corrections that are not included in a leading twist calculation can be absorbed into phenomenological functions ; for example , for an unpolarized structure function @xmath181 , @xmath182 where @xmath183 denotes the leading twist contribution including tmcs . the higher twist corrections are sometimes assumed to be multiplicative , with the functions @xmath184 proportional to the leading twist contribution , @xmath185 . possible additional @xmath40 dependence of the higher twist contributions , from radiative @xmath186 corrections , is usually neglected . the leading twist pdfs were found in ref . @xcite to be essentially independent of the tmc prescription adopted , with the ht parameters able to compensate for the variations due to the different tmc formulations . the isospin dependence of the ht corrections was also studied in refs . existing data do not allow for an accurate determination of the @xmath187 corrections @xmath188 , and attempts to include them in global fits produce anomalously small values of @xmath189 and @xmath190 in conjuction with large compensating twist4 and twist6 contributions @xcite , possibly due to overfitting and blurring of the scaling violations . in the polarized case , the greater scarcity of data means that typically one can not afford the luxury of @xmath40 and @xmath70 cuts as stringent as those applied in some of the unpolarized pdf analyses . most global analyses of spin - dependent pdfs therefore include structure function measurements down to @xmath191 gev@xmath170 , where higher twist corrections are believed to be important . the higher twist contributions to @xmath90 and @xmath91 may be treated in an analogous way to the unpolarized @xmath181 structure functions , with the important difference that the twist3 contributions to @xmath91 are not @xmath40-supressed . phenomenological pdf analyses exist which include ht contributions to @xmath90 @xcite , @xmath91 @xcite , and to both functions simultaneously @xcite . while the focus in the present work is on the leading twist pdfs , the higher twist contributions to @xmath90 and @xmath91 are also of intrinsic interest in themselves , containing information , for example , about the correlations of color electric and magnetic fields in the nucleon with the nucleon s spin @xcite . since nucleons bound in a nucleus are not free , the parton distributions @xmath192 in a nucleus @xmath7 deviate from a simple sum of pdfs in the free proton and neutron , @xmath193 , where @xmath194 is the number of protons . this phenomenon is especially relevant at small values of @xmath2 , where nuclear shadowing ( or screening ) effects suppress the nuclear to free isoscalar nucleon ( @xmath195 ) ratio , @xmath196 , and at large @xmath2 , where the effects of fermi motion , nuclear binding , and nucleon off - shellness give rise to the `` nuclear emc effect '' @xcite . in addition , for spin - dependent pdfs , the different polarizations of the bound nucleons and nuclei need to be taken into account . in the nuclear impulse approximation , where scattering is assumed to take place incoherently from partons inside individual nucleons , the pdf in a nucleus can be expressed as a convolution of the pdf in a bound nucleon and a momentum distribution function @xmath197 of nucleons in the nucleus @xcite . coherent rescattering effects involving partons in two or more nucleons give rise to nuclear shadowing corrections to the impulse approximation . in general , at large @xmath40 the pdfs in a nucleus and in a nucleon are related by @xmath198 where the additive term @xmath199 represents nucleon off - shell or relativistic corrections , and @xmath200 parametrizes the shadowing corrections . a similar expression can be written for spin - dependent pdfs . the momentum distribution , or `` smearing function '' , @xmath197 , can be computed from nuclear wave functions , incorporating nuclear binding and fermi motion effects . at @xmath201 the smearing function has a simple probabilistic interpretation in terms of the light - cone momentum fraction of the nucleus carried by the struck nucleon , where @xmath202 and @xmath203 are the four - momenta of the nucleon and nucleus , respectively , and @xmath204 is the nuclear mass . in this case the smearing function is normalized to unity , @xmath205 . at finite @xmath40 , however , the smearing function depends in addition on the parameter @xmath206 , where @xmath207 and @xmath208 are the energy and three - momentum transfer , respectively , which characterizes the deviation from the bjorken limit @xcite . typically , the function @xmath197 is steeply peaked around @xmath209 , becoming broader with increasing mass number @xmath7 as the effects of binding and fermi motion become more important . in the limit of zero binding , @xmath210 , and one recovers the free - nucleon case . this is the usual assumption made in most global pdf analyses . recently , several analyses @xcite have accounted for the nuclear effects by explicitly calculating the corrections from microscopic nuclear models , or attempted to constrain them phenomenologically @xcite . the most straightforward calculation is for the simplest nucleus the deuteron , for which both nonrelativistic and relativistic wave functions are available @xcite , constrained by high - precision nucleon - nucleon scattering data . experiments with deuterium targets play a vital role , in fact , in determining the flavor decomposition of the proton pdfs . traditionally , the standard method for disentangling the @xmath141 and @xmath142 pdfs has been through charged lepton dis , which for a proton target is sensitive at large @xmath2 to the combination @xmath211 . dis from a neutron would constrain @xmath212 ; however , the absence of free neutron targets has necessitated the use of deuterium as effective neutron targets . while the nuclear corrections in the deuteron are typically a few percent , in some regions of kinematics , most notably at large @xmath2 , they can give rise to large uncertainties in the extracted @xmath142 quark pdf in particular . an alternative , purely phenomenological approach @xcite has attempted to constrain the nuclear corrections in deuterium directly from the data . however , without an independent , high - precision measurement of the @xmath142 quark pdf from processes other than inclusive dis , it is difficult to unambiguously separate the effects of the nuclear corrections from uncertainties in the @xmath142 quark pdf , especially at high values @xmath2 . until future experiments ( see ref . @xcite for a discussion ) are able to determine the @xmath142 quark pdf independent of uncertainties in nuclear models , a more practical approach adopted by the cj collaboration @xcite has been to produce a set of global pdfs for a range of nuclear models , corresponding to mild ( cj12min ) , medium ( cj12mid ) , and strong ( cj12max ) nuclear corrections in the deuteron . the current uncertainties in pdfs from nuclear corrections can have important consequences far beyond the fixed - target experiments where they are encountered most directly . in fact , the effects of the nuclear smearing corrections are not suppressed at large @xmath40 , and must be considered at all scales wherever data at @xmath213 are used @xcite . through @xmath40 evolution , pdfs at large @xmath2 and small @xmath40 evolve to lower @xmath2 and higher @xmath40 , so that large-@xmath2 uncertainties in fixed - target experiments can have significant consequences for collider measurements @xcite , examples of which were discussed for selected observables at the tevatron and lhc in ref . @xcite . for neutrino scattering , in order to increase the relatively low rates and obtain sufficient statistics , experiments have often resorted to using heavier nuclear targets , such as iron or lead . this is particularly relevant for determinations of the strange quark pdf , which is typically extracted from opposite - sign dimuon events in @xmath207 and @xmath214 charm production , @xmath215 or @xmath216 @xcite . such extractions are complicated by the presence of nuclear corrections in neutrino structure functions @xcite , as well as effects of the nuclear medium on the charm quark propagation in the final state @xcite . uncertainties in the strange quark pdf can have significant impact on @xmath70 and @xmath194 boson measurements at the lhc , for example , so that understanding of the nuclear effects will have impact far beyond lepton nucleus dis @xcite . in order to systematically study the nuclear dependence of pdfs , without assuming specific relations between pdfs in nuclei and nucleons , several groups have parametrized nuclear pdfs directly @xcite . one approach has been to parametrize the @xmath7 dependence of the initial proton pdf parameters , and then perform a global fit of the available hard scattering data on nuclear targets . this implicitly assumes a smooth @xmath7 dependence for the pdfs , which is reasonable for large @xmath7 , but may break down for light nuclei such as deuterium . an alternative method is to fit to one type of data set and examine how the results differ from those on proton targets . this can be useful for testing whether the nuclear corrections are consistent with those obtained from various data sets . such comparisons have revealed , for instance , a controversy about whether or not the @xmath7 dependence of the neutrino - nucleus data from the nutev collaboration is compatible with that observed in charged lepton dis @xcite . differences between neutrino and electromagnetic nuclear interactions can arise , for instance , from the presence of the parity - odd @xmath82 structure function in @xmath217 scattering , which does not contribute to charged - lepton scattering @xcite . for spin - dependent scattering , the scarcity of data and their larger uncertainties at very small @xmath2 and at high @xmath2 , where nuclear corrections are most prominent , has meant that almost all global analyses have thus far relied exclusively on the effective polarization _ ansatz _ , in which the polarized pdf in the nucleus @xmath218 is related to the polarized pdfs in the proton and neutron as @xmath219 , with @xmath220 the average polarizations of the proton ( neutron ) in the nucleus . in practice , only polarized deuterium and @xmath221he nuclei have been used in dis experiments , in addition to protons . the new global analysis of helicity pdfs by the jam collaboration @xcite is the first systematic attempt to incorporate the effects of nuclear smearing in dis structure functions of the deuteron and @xmath221he . this will be important as future spin - dependent scattering experiments probe the nucleon spin structure at increasingly large values of @xmath2 @xcite . the concept of global fitting for pdfs relies on the formalism outlined in the previous sections . partonic cross sections for various hard scattering processes are convoluted with scale - dependent pdfs to generate results for physical observables which can then be compared to data . the pdfs are parametrized at some convenient input scale @xmath222 and then evolved using the appropriate evolution equations to the scales needed for each calculation . the values of the input parameters are estimated using a @xmath189 minimization technique . all of the global pdf fitting groups use some variation of this technique . a typical parametrization at the input scale @xmath222 for a generic polarized or unpolarized parton ( quark , antiquark or gluon ) distribution function @xmath223 is @xmath224 where @xmath225 represents some smoothly varying function usually chosen as a polynomial in @xmath2 or @xmath226 , although exponential functions are used as well . in addition , there are parametrizations inspired by a statistical model of the nucleon @xcite , as well as those based on neural networks @xcite and self - organizing maps @xcite . some of the parameters in the input distributions can be determined from physical constraints . for example , in the unpolarized case valence quark number is conserved , @xmath227 while the momentum sum rule requires @xmath228 = 1 , \label{eq : mom_sum_rule}\ ] ] where we use the notation @xmath229 , and the number of flavors at the input scale @xmath230 is usually taken to be @xmath231 ( see sec . [ ssec : hq ] ) . in the polarized case the first moments of the charge - conjugation even ( or @xmath11-even ) distributions can be related to octet baryon weak decay constants . for the isovector combination , corresponding to the bjorken sum rule , @xmath232 where @xmath233 is the nucleon axial charge , while for the su(3 ) octet one has @xmath234 where the octet axial charge @xmath235 is extracted from hyperon @xmath29-decays assuming su(3 ) flavor symmetry @xcite . note that the sum rules ( [ eq : sum_rules])([eq : hyperon ] ) are preserved under @xmath40 evolution . while most of the pdf groups use similar procedures and data , one can nevertheless obtain pdf results that can be rather different . some of the reasons for this include the following : * differences in the specific parametrizations and input scale @xmath230 . * differences in data selection , choices of data sets , kinematic cuts , and the specific treatment of the correlated errors of the data . this typically limits the amount of data available at large values of @xmath2 . * differences in the theoretical framework used , including the particular form of the solutions of the rge employed , the treatment of heavy quarks , inclusion of higher twist contributions , target mass effects , and nuclear corrections . these differences should be born in mind when comparing the results from different pdf groups . pdfs are an essential ingredient for producing predictions for processes in high energy experiments . the uncertainties in these predictions depend , in part , on how well determined the pdfs are themselves . it is important to bear in mind that the predictions that are compared to data are convolutions of pdfs with partonic hard scattering cross sections . there are thus three main sources of pdf uncertainties : the fitted data , the partonic cross sections , and the parametrizations used to describe the pdfs . the following list describes the main sources of uncertainty in the determination of pdfs . * the experimental errors on the fitted data can be directly propagated to the fitted pdfs . standard techniques include the hessian @xcite , lagrange @xcite , and monte carlo @xcite methods . * uncertainties due to the use of perturbation theory : these can be estimated to some extent by doing lo , nlo , and nnlo fits , although not all processes are known to nnlo accuracy . * scale dependence : the perturbative predictions depend to some extent on the choices made for the renormalization and factorization scales for each process . these choices will change the results for each process and the fitted pdfs must compensate these changes ( see sec . iii of ref . @xcite for an example of the effects of choosing different scales ) . * choice of the value of the running coupling @xmath236 : some pdf determinations fit @xmath236 while others use the global average value . this is mostly a question of philosophy . on the one hand , since the strong coupling is a parameter of qcd , then there is no freedom to choose a different value for each process . thus , information from data types not included in the global fits ( such as data from @xmath237 processes ) provides valid constraints on the value of the coupling and are included in determining the global average . on the other hand , fitting the value of the strong coupling and comparing it to the global average provides an interesting consistency check on the description of the data provided by qcd . * choice of data sets and kinematic cuts : these choices can affect the fitted pdfs and the user should be aware of these differences . * treatment of heavy quarks : the various schemes that are currently used differ at higher orders and such differences can affect the fitted pdfs . * parametrization dependence : fitted pdfs can differ simply through the choice of the initial parametrizations . extensive efforts are made to choose flexible parametrizations which are well - constrained by data , but there is no control over the behavior outside the kinematic region covered by the data . this can lead to different extrapolations at very large or very small values of @xmath2 . a method for estimating this remaining uncertainty has been suggested in ref . @xcite . of all these sources of error , the easiest to treat is the first the propagation of the experimental errors . the hessian method is described in detail in ref . the elements of the hessian matrix are given by @xmath238 where @xmath239 denotes the @xmath240 pdf parameter . the hessian matrix is generated during the actual minimization procedure and its inverse is the error matrix . the eigenvectors of the error matrix can then be used to define eigenvector parameter sets which , in turn , can be used to calculate error bands for the pdfs or for specific processes . one particular subtlety is that the error bands generally depend on a @xmath189 tolerance . mathematically , one expects the @xmath241 parameter errors to correspond to an increase of @xmath189 by one unit from the minimum value . however , it has been suggested @xcite that inconsistencies between different data sets may require a larger value to be used . this `` @xmath189 tolerance '' varies between groups and allowance must be made for this when comparing the resulting error bands . the lagrange multiplier @xcite method is useful when one wants to determine the pdf error on a specific observable @xmath12 such as the cross section for @xmath70 or @xmath194 production . let @xmath242 denote the @xmath189 for the global data set . then one minimizes the function @xmath243 where @xmath18 denotes the set of pdf parameters and minimization is done for a range of values of @xmath244 . the end result is a relation between @xmath242 and the value @xmath12 of the chosen observable . once one specifies the @xmath189 tolerance , _ i.e. _ , the range of allowable @xmath189 values , this maps out a range of values @xmath245 where @xmath246 is the value of @xmath12 at the global minimum . an alternative to the usual linear propagation of errors ( hessian method ) which can be useful for minima that are not well behaved or defined , is the so called monte carlo method . in order to propagate the experimental errors a number of ` replica ' data sets are generated by using random numbers and the original errors to generate new data points @xcite . these replica data sets are then fitted and the resulting replica pdf sets are treated using standard statistics ; the central values are given by the average over replicas , and the uncertainties by the envelope of predictions . the main motivation for pursuing global fitting as a technique for determining pdfs is that the use of a wide range of data types with different kinematic coverage places many constraints simultaneously on the pdfs . each type of observable depends on a particular linear combination of pdfs or products of pdfs . obtaining the best fit to all of the observables simultaneously has proven to be an efficient method for extracting pdfs . nevertheless , it remains true that specific observables may be sensitive to a specific pdf or combination of pdfs . in this section some examples using simplified lo kinematics will be presented since these can be useful in understanding how pdf errors can be reduced by future measurements . * deep - inelastic scattering * experiments provide direct information on pdfs since both the structure functions and the cross sections depend linearly on the pdfs . the following list summarizes the main dependences . 1 . charged lepton neutral current measurements on a proton target constrain the combination @xmath247 where we have assumed that only single photon exchange contributes ( @xmath194 boson exchange involves a different linear combination of pdfs ) . at large values of @xmath2 the antiquark pdfs become negligible leaving the combination @xmath248 . if one had a neutron target , then the linear combination would be @xmath249 which becomes @xmath250 at large values of @xmath2 . of course , neutron targets are not available , so deuterium is often used for this purpose requiring that nuclear corrections be made in order to extract the neutron target information as discussed in sec . [ ssec : nuclear ] . 2 . charged current neutrino interactions constrains the combinations @xmath251 where the plus ( minus ) sign corresponds to @xmath252 ( @xmath253 ) . the actual linear combinations depend on the type of target used ; high statistics neutrino experiments employ different types of heavy targets for which model - dependent nuclear corrections must be made . note that one can , in principle , isolate specific combinations of @xmath85 or @xmath86 pdfs , depending on the type of target used . a special case is provided by charm production in neutrino and antineutrino scattering as these are proportional to the @xmath254 and @xmath255 pdfs , respectively . the experimental signal is the production of opposite sign muon pairs requiring one to take into account the charm fragmentation and decay as well as nuclear corrections , depending on the type of target employed . the gluon pdf enters in dis at @xmath256 . thus , it is mainly constrained through the @xmath40 dependence of the structure functions and rosenbluth ( or longitudinal / transverse ) separated @xmath257 data ( see sec . [ ssec : gluon ] ) . * vector boson production * ( lepton pairs , @xmath115 , and @xmath116 ) in lo proceed through @xmath258 annihilation . a few key examples are : 1 . lepton pair production in proton proton and proton neutron collisions depend on the combinations @xmath259 where @xmath260 , and the @xmath261indicate contributions from the @xmath254 , @xmath262 and @xmath19 quarks . of course , the @xmath263 cross section is obtained from data from a deuterium target . these cross sections can be used to constrain the @xmath264 ratio , for example . 2 . @xmath115 production constrains products of the form @xmath265 with specific weights given by the appropriate ckm matrix elements , whereas @xmath116 production constrains @xmath258 . for @xmath266 collisions at the tevatron at large values of rapidity , one has approximately , @xmath267 where @xmath268 denotes the proton ( antiproton ) , and the @xmath261 represents contributions from heavier quarks . note that these results are written in terms of proton pdfs . if the rapidity is large and positive , then @xmath269 and one can neglect the antiquark terms so that these cross sections directly constrain the @xmath270 pdfs . due to the missing neutrino from the @xmath70 decays , one can not directly reconstruct the rapidity distributions . what is usually presented is the charged lepton rapidity asymmetry for @xmath115 production . in this case the decay process means that the constraints on the pdfs are less direct , but such measurements still provide useful constraints on the @xmath145 ratio at moderate values of @xmath2 . of particular interest is the model dependent determination of the @xmath70 rapidity asymmetry by the cdf collaboration @xcite , which directly constrains the @xmath145 ratio at large values of @xmath2 since the asymmetry is for the @xmath70 and not the charged lepton from the decay . recently the atlas collaboration has presented an analysis @xcite using @xmath70 decay lepton and @xmath116 rapidity distribution data which shows potential for constraining the strange quark pdf . this analysis makes use of the fact that the @xmath116 cross section receives a significant contribution from @xmath271 annihilation while the @xmath70 cross sections help constrain the @xmath272 and @xmath273 pdfs . the other major class of observables includes inclusive * jet or photon , dijet * , and * photon + jet production*. each of these constrain products of pdfs summed over all flavors . nevertheless , there are certain pdfs that can be constrained by these observables . this follows since the @xmath141 and @xmath142 pdfs are well constrained by the dis data and the vector boson and lepton pair production data provide strong constraints on the @xmath272 and @xmath273 pdfs . hence , these data have the greatest impact on the gluon distribution . 1 . direct photon production in lo has two subprocesses : @xmath274 and @xmath275 . hence , this process was though to be a good candidate for constraining the gluon , especially at fixed target energies where the available @xmath110 range extended out to about 0.6 . however , photons can also be created by bremsstrahlung from the charged quarks often referred to as the fragmentation process . indeed , one can define photon fragmentation functions to help describe this production component . it turns out that at fixed target energies this component receives very large soft gluon corrections , requiring threshold resummation techniques @xcite . this , combined with some apparent disagreements between experimental data sets @xcite meant that photon production has not fulfilled the original promise of constraining the gluon pdf . recently , however , an analysis @xcite suggests that the use of _ isolated _ photon collider data may well help constrain the gluon pdf . the use of isolation cuts reduces the fragmentation contribution and the use of higher energy collider reduces the need for threshold resummation . high-@xmath20 jet production has played a significant role in constraining the gluon pdf . even though the quark and antiquark pdfs are well constrained by other types of data , there are still significant contributions from @xmath276 and @xmath277 subprocesses @xcite . dijet production triple differential cross sections yield more information than single jet cross sections because the rapidity of the second jet is also constrained , thereby helping to constrain the momentum fractions of the pdfs . by tuning the rapidity ranges for the two jets one can explore different regions for these momentum fractions . an example is given in ref . photon + jet production offers similar constraints , but now the subprocesses are weighted by the squared charge of the parton to which the photon couples . as with unpolarized measurements , historically most constraints on spin - dependent pdfs have come from * polarized charged - lepton dis * experiments . 1 . in the one - photon exchange approximation , the difference of inclusive dis cross sections for leptons polarized longitudinally with spin parallel and antiparallel to the target hadron polarization measures @xmath11-even quark combinations @xmath278 . for proton targets , one has the combination @xmath279 while for the neutron the combination would be @xmath280 with contributions from heavy quark polarization expected to be negligible . in practice , polarized @xmath221he targets are usually used as effective sources of polarized neutron , since the neutron carries almost 90% of the spin of @xmath221he , while polarized deuterons , to which the proton and neutron spins contribute equally , are used to provide the isoscalar combination @xmath281 because of the greater sensitivity of the proton @xmath90 structure function measurements to the polarized @xmath141 quark pdf , and uncertainties associated with nuclear corrections when extracting the neutron from @xmath221he or deuterium data , the uncertainty on the @xmath282 pdf is significantly larger at high @xmath2 than on @xmath283 . 2 . at nlo , the polarized gluon distribution @xmath152 also enters in the @xmath90 structure function . the @xmath40 evolution of the flavor singlet contribution to @xmath90 can then be used to constrain @xmath152 , as for the unpolarized gluon pdf discussed above , although the constraints are weaker in the polarized because @xmath284 at small @xmath2 . * semi - inclusive dis * provides additional independent combinations of spin - dependent pdfs that can be used to reconstruct individual quark and antiquark flavors . 1 . semi - inclusive production of hadrons @xmath285 in the current fragmentation region , primarily pions or kaons , is proportional to products of spin - dependent pdfs and quark @xmath286 hadron fragmentation functions , @xmath287 where @xmath288 is the fraction of the quark s energy carried by the hadron @xmath285 . the fragmentation functions @xmath289 can be determined from other reactions , such as inclusive hadron production in @xmath290 annihilation . one can weight particular quark or antiquark flavors by selecting favored ( such as @xmath291 or @xmath292 ) or unfavored ( @xmath293 or @xmath294 ) fragmentation functions for specific hadrons ( in this case @xmath295 ) . information on the polarized strange quark pdf @xmath296 in particular can be obtained from data on @xmath297 production . 2 . the polarized gluon distribution @xmath152 can be constrained by semi - inclusive dis data on charmed or high-@xmath20 hadron production through the photon gluon fusion process . recently the analysis of this process has been performed at nlo , extending earlier lo extractions of @xmath152 . inclusive particle production in * polarized proton proton collisions * provides an additional method of determining spin - dependent sea quark and gluon distributions . 1 . @xmath57 production cross sections for scattering longitudinally polarized protons from unpolarized protons , @xmath148 , depend on products of spin - dependent and spin - averaged pdfs , for example , @xmath298 at large positive or negative rapidities , @xmath132 or @xmath299 , the cross sections ( or asymmetries ) are dominated by a single flavor , while at mid - rapidities both @xmath141 and @xmath142 flavors contribute . inclusive jet or @xmath300 production in double - polarized proton proton scattering , @xmath301 , is sensitive to the polarized gluon pdf . the first evidence for a small , but nonzero @xmath152 was recently observed by the star collaboration at rhic in jet data at @xmath302 gev . direct photon production in polarized @xmath134 scattering has also been suggested @xcite as a means of probing @xmath152 . using the technology outlined in sec . [ sec : qcd ] , a number of global qcd analyses of the world s high - energy scattering data have produced sets of proton pdfs , up to next - to - leading order ( nlo ) or next - to - next - to - leading order ( nnlo ) accuracy . the efforts have indeed been global , with groups in europe and the us in the forefront of the data analyses . the pdfs sets include parametrizations from the mstw @xcite and abm @xcite groups , both of which use standard global fitting methodology ; the nnpdf @xcite collaboration , which uses a newer approach based on neural networks ; the jr @xcite pdfs , which are dynamically generated through @xmath40 evolution from a low @xmath40 input scale ; and the herapdf @xcite group , which includes only data from the h1 and zeus experiments at hera . the us - based efforts have been centered around the cteq collaboration @xcite , which at present involves two derivative analyses of nucleon pdfs , by the ct @xcite and cj ( cteq - jefferson lab ) @xcite groups , as well as the ncteq @xcite analysis of nuclear pdfs . the cj collaboration in particular has focused on developing the methodologies needed for describing data over a broad energy range including the low-@xmath40 and @xmath70 domain @xcite . in this section we summarize the results for unpolarized pdfs from the various pdf fitting groups , discussing their similarities and contrasting their differences . we will focus mostly on the physics issues , rather than on technical aspects of pdf fitting . of course with new data arriving or soon anticipated at the lhc , jefferson lab , and other facilities , the global fitting efforts are constantly evolving , so that the information presented here can only be viewed as a snapshot of the field at the present time . ( red ) , @xmath303 ( blue ) , @xmath304 ( green ) , @xmath305 ( violet ) , @xmath306 ( orange ) and @xmath307 ( black ) pdfs for the cj12mid fit @xcite at @xmath308 gev@xmath170 , shown on logarithmic _ ( left ) _ and linear _ ( right ) _ scales in @xmath2 . note that in the left panel the gluon is scaled by a factor 1/10.,title="fig:",width=302 ] ( red ) , @xmath303 ( blue ) , @xmath304 ( green ) , @xmath305 ( violet ) , @xmath306 ( orange ) and @xmath307 ( black ) pdfs for the cj12mid fit @xcite at @xmath308 gev@xmath170 , shown on logarithmic _ ( left ) _ and linear _ ( right ) _ scales in @xmath2 . note that in the left panel the gluon is scaled by a factor 1/10.,title="fig:",width=272 ] as a typical example of modern pdfs and their uncertainties , fig . [ fig : allpdfs ] shows the @xmath309 , @xmath303 , @xmath304 , @xmath305 , @xmath306 and @xmath307 pdfs ( with the gluon scaled by a factor 1/10 ) for the cj12 fit @xcite ( for the case of moderate nuclear corrections , cj12mid ) at a scale of @xmath308 gev@xmath170 . the general behavior of the pdfs is similar for all the parametrizations @xcite , particularly where sufficient data exist to constrain the distributions in specific regions of @xmath2 and @xmath40 . in regions where data are scarce , such as at very low @xmath2 ( @xmath310 ) or high @xmath2 ( @xmath311 ) for certain observables , or where data on specific pdfs ( such as the strange quark ) are subject to large experimental or theoretical uncertainties , the details of the pdfs can depend strongly on the assumptions adopted in the extrapolations . and @xmath312 , and the @xmath305 and @xmath313 sea distributions , for the available 3-flavor nnlo pdf sets from jr09 @xcite ( red solid line ) , mstw08 @xcite ( blue dashed ) and abm11 @xcite ( green dotted ) , at @xmath308 gev@xmath170.,width=453 ] , but for the singlet - sector distributions @xmath314 , @xmath315 and @xmath316 and the gluon @xmath307 pdf.,width=453 ] some of these differences are evident in figs . [ fig : nspdf ] and [ fig : spdf ] , which illustrate the flavor nonsinglet combinations @xmath317 , and the singlet - sector distributions @xmath318 , respectively , for the nnlo mstw08 @xcite , abm11 @xcite , nnpdf @xcite and jr09 @xcite pdf sets . while the valence @xmath319 and @xmath320 pdfs are fairly well constrained in the intermediate-@xmath2 region , at large @xmath2 there are significant uncertainties on the @xmath320 distribution in particular . the su(2 ) nonsinglet distribution @xmath321 is well determined by data from the drell - yan reaction over the range @xmath322 , but is not constrained at higher @xmath2 . the strange nonsinglet distribution is the most difficult to determine , requiring a combination of neutrino and antineutrino scattering data , which are subject to large experimental and nuclear uncertainties . for the @xmath11-even distributions in fig . [ fig : spdf ] , the @xmath323 and @xmath324 are the dominant quark distributions at large @xmath2 , but the strange quark pdf becomes increasing more important at small @xmath2 . numerically , the gluon distribution dominates all other pdfs at small @xmath2 , but has sizable uncertainties at large @xmath2 values . in the following sections we discuss features of each of the pdfs in more detail . valence quarks give the global properties of the nucleon , such as its charge and baryon number . knowledge of their momentum distributions is important for many reasons , especially at high values of @xmath2 where a single quark carries most of the nucleon s momentum . the large-@xmath2 region is in fact a unique laboratory for studying the nonperturbative flavor and spin dynamics of quarks @xcite , as well as testing predictions from perturbative qcd for the behavior of pdfs in the limit @xmath165 @xcite . reliable determination of pdfs at large @xmath2 is also important for searches for new physics beyond the standard model in collider experiments at the lhc . through perturbative qcd evolution , uncertainties in pdfs from fixed - target experiments at high @xmath2 and low @xmath40 can propagate to larger @xmath40 to affect cross sections at smaller @xmath2 values @xcite . this is especially true for the forward production of particles of mass @xmath155 at large rapidities @xmath111 , whose cross sections are given by products of pdfs with one evaluated at small @xmath325 and the other at large @xmath326 . the production of heavy @xmath327 and @xmath328 bosons , for example , is sensitive to @xmath142 quark pdf uncertainties at high rapidities , exceeding 100% in the @xmath329 channel , which places limits on the accuracy of cross section measurements for masses near the kinematic thresholds @xcite . furthermore , understanding pdfs at large @xmath2 is vital for the analysis of neutrino oscillation experiments @xcite , where one of the most significant uncertainties comes from neutrino nucleus cross sections at the interface of the dis and resonance regions . the growing need to better understand large-@xmath2 pdfs and their uncertainties has been reflected in the greater attention being paid recently to the physics of the large-@xmath2 region , with dedicated experiments planned at jefferson lab following its 12 gev upgrade @xcite , as well as at proposed new facilities such as the lhec @xcite and the electron - ion collider @xcite . it has also been a catalyst for the recent concerted theoretical efforts by the cj collaboration @xcite in their global qcd analyses of pdfs extending into the lower @xmath40 and @xmath168 regions , with the aim of providing better constraints on pdfs at large values of @xmath2 @xcite . and @xmath142 quark distributions from the cj12mid pdfs @xcite , with pdf errors only ( solid , red shaded ) with combined pdf and nuclear uncertainties ( dashed , blue shaded ) , and with a fit excluding all deuterium data ( dot - dashed , green shaded ) , relative to the reference cj12mid set at an arbitrary scale @xmath330 gev@xmath170 . note the different vertical scales for the @xmath141 and @xmath142 quark pdfs.,title="fig:",width=283 ] and @xmath142 quark distributions from the cj12mid pdfs @xcite , with pdf errors only ( solid , red shaded ) with combined pdf and nuclear uncertainties ( dashed , blue shaded ) , and with a fit excluding all deuterium data ( dot - dashed , green shaded ) , relative to the reference cj12mid set at an arbitrary scale @xmath330 gev@xmath170 . note the different vertical scales for the @xmath141 and @xmath142 quark pdfs.,title="fig:",width=283 ] the current constraints on the @xmath141 and @xmath142 quark distributions are illustrated in fig . [ fig : dunod ] for the cj12mid pdf set @xcite , where the uncertainties arising from nuclear corrections and pdf ( experimental ) errors are indicated separately . since the @xmath141 quark pdf is relatively well constrained by the proton dis data at large @xmath2 , the effect of the nuclear uncertainties is minor , increasing the total uncertainty at intermediate @xmath2 by a few percent . for the @xmath142 quark distribution , on the other hand , whose determination at present requires both proton and deuterium dis data , the nuclear correction uncertainties in the deuteron significantly increase the overall error for @xmath331 . however , even though the use of deuterium data introduces the need to confront the problem of nuclear effects and their uncertainties , without these data the error bands on the @xmath142 quark pdf would be even larger over most of the @xmath2 range , as fig . [ fig : dunod ] demonstrates . in addition to accounting for the finite-@xmath40 ( tmc and higher twist ) and nuclear effects which become increasingly important as @xmath2 tends to 1 at fixed @xmath40 , the behavior of pdfs in the @xmath165 limit also depends critically on the functional form used for the @xmath2 dependence . the conventional parametrizations utilized in most global pdf analyses assume a @xmath332 dependence for both the @xmath141 and @xmath142 quark pdfs , as in eq . ( [ eq : param ] ) above , so that in the limit as @xmath165 , the @xmath145 ratio tends either to zero or infinity . @xcite allowed for a more flexible parametrization of the valence @xmath142 quark pdf , in which the @xmath320 distribution receives a small admixture from the @xmath319 pdf , @xmath333 with @xmath19 and @xmath262 as free parameters . provided the @xmath142 quark has a softer momentum dependence ( larger exponent @xmath334 ) than the @xmath141 quark , which agrees with phenomenology , the ratio of the distributions will approach a constant value , @xmath335 , as @xmath165 . a nonzero , finite value in this limit is in fact expected in several different nonperturbative models of nucleon structure @xcite . ratio for the _ cj12min ( dot - dashed ) , cj12mid ( solid ) and cj12max ( dashed ) pdfs @xcite , and _ ( right ) _ cj12mid ( solid ) compared with the ct10 @xcite ( dot - dashed ) , mstw08 @xcite ( dashed ) and abkm09 @xcite ( dot - dot - dashed ) parametrizations at @xmath330 gev@xmath170.,title="fig:",width=283 ] ratio for the _ ( left ) _ cj12min ( dot - dashed ) , cj12mid ( solid ) and cj12max ( dashed ) pdfs @xcite , and _ ( right ) _ cj12mid ( solid ) compared with the ct10 @xcite ( dot - dashed ) , mstw08 @xcite ( dashed ) and abkm09 @xcite ( dot - dot - dashed ) parametrizations at @xmath330 gev@xmath170.,title="fig:",width=283 ] the ratio of the @xmath142 to @xmath141 quark distributions is illustrated in fig . [ fig : dupdf ] for several modern nlo pdf sets , including the cj12 @xcite , ct10 @xcite , mstw08 @xcite ( dashed ) and abkm09 @xcite distributions . as expected , the total uncertainty on the @xmath145 ratio at large @xmath2 increases when different nuclear correction models are considered , with the central values extrapolated to @xmath336 increasing from @xmath337 for the cj12min pdfs , with the minimum nuclear correction model , to @xmath338 for the cj12max set with the maximum nuclear correction model from ref . @xcite . combining all uncertainties , the cj12 analysis found @xmath339 } \pm 0.10 \,\rm{[nucl]}$ ] as @xmath165 , where the first error is from the pdf fits and the second is from the nuclear correction models . in contrast , using the more restrictive , conventional parametrizations , the @xmath145 ratios for the mstw08 and abkm09 distributions tend to zero in the @xmath165 limit , while that for ct10 tends to infinity . this behavior distorts somewhat the error bands , making the mstw08 ratio appear to have an anomalously small uncertainty , while the ct10 result has a much larger one . the relative errors , however , are similar for the mstw08 and ct10 pdfs , while the cj12 and abkm09 uncertainties are reduced at larger @xmath2 because of the additional high-@xmath2 data used in those analyses . unfortunately , the extrapolated values of @xmath145 at @xmath336 span most of the range of available predictions @xcite , so that discrimination between the physical mechanisms that lead to the different @xmath165 behaviors will only be possible with constraints from new experiments that do not involve deuterium data and the nuclear model uncertainties @xcite . inclusive electromagnetic dis probes @xmath11-even combinations of pdfs , @xmath340 , weighted by their squared charges , or in terms of valence and sea quark ( or antiquark ) distributions , @xmath341 . the shapes of the quark valence and sea distributions differ markedly as a function of @xmath2 . as we have seen in the preceding section , the valence quark pdfs dominate at intermediate and large values of @xmath2 , but vanish as @xmath342 for @xmath343 . the sea , in contrast , dominates at small @xmath2 , but is strongly suppressed as @xmath165 . for the light antiquarks @xmath272 and @xmath273 , structure function data with proton targets are sensitive at small @xmath2 to the combination @xmath344 , while deuterium targets ( nuclear corrections aside ) probe the isosinglet combination @xmath345 . under @xmath40 evolution , the singlet quark distribution mixes with the gluon , which leads to the rapid growth in the antiquark pdfs with decreasing @xmath2 . with increasing @xmath40 , the uncertainties on @xmath345 also decrease . inclusive dis measurements can therefore provide important constraints on the behavior of both the @xmath346 and @xmath86 distributions in the regions where their contributions are dominant . however , in the intermediate-@xmath2 region , @xmath347 , where the magnitude of valence and sea quark pdfs is comparable , additional information is required to uniquely determine the distributions individually . here data from neutrino scattering can allow access to the parity - violating @xmath82 structure function , which measures @xmath11-odd combinations of pdfs , @xmath348 . combined with the @xmath11-even pdfs from the parity - conserving @xmath80 and @xmath81 structure functions , one can then uniquely determine the @xmath85 and @xmath86 distributions individually . unfortunately , neutrino scattering cross sections are usually statistics limited compared with electromagnetic data , and with few exceptions , have necessitated the use of nuclear targets such as iron or lead as a means of increasing the rates . this subsequently brings with it the inherent complications associated with accounting for nuclear corrections in relating the nuclear structure functions with those in a free nucleon @xcite . more direct constraints on @xmath86 distributions can be obtained from lepton pair production in inclusive hadron hadron scattering , which involves products of pdfs evaluated at different values of beam ( @xmath112 ) and target ( @xmath143 ) momentum fractions . as discussed in sec . [ sssec : h - h ] , the drell - yan process or @xmath57-boson production in @xmath134 scattering at nonzero rapidity , for example , allows the antiquark distribution at small @xmath112 to be extracted with knowledge of the corresponding quark distribution at @xmath143 . in particular , the drell - yan reaction @xcite has been used by the na51 collaboration at cern @xcite and the e866/nusea collaboration at fermilab @xcite to determine the ratio of @xmath273 to @xmath272 distributions in the proton . naive expectations from perturbative qcd based on @xmath349 splitting suggest that since the masses of the @xmath141 and @xmath142 quarks are very similar , the production of @xmath350 and @xmath351 pairs should be nearly identical , even with the inclusion of higher - order radiative corrections @xcite . the first indication of a significant asymmetry in the proton sea was obtained by the new muon collaboration ( nmc ) @xcite at cern , who performed an accurate measurement of the ratio of dis cross sections for hydrogen and deuterium , from which the gottfried sum @xmath352 was determined @xcite , @xmath353 where charge symmetry is assumed ( see sec . [ ssec : csv ] below ) . for a flavor symmetric su(2 ) proton sea , the prediction for the gottfried sum would be @xmath354 . in contrast , the nmc measurement found @xmath355 @xcite , indicating a strong violation of flavor symmetry in the light antiquark sea , @xmath356 following the suggestion by ellis and stirling @xcite that the light antiquark sea could be more directly probed in the drell - yan process , the na51 experiment used the ratio of dimuon cross sections for @xmath357 and @xmath134 scattering to extract the ratio @xmath358 at an average @xmath359 at a feynman-@xmath2 of @xmath360 . subsequently , the e866/nusea experiment at fermilab measured the @xmath361 ratio at large @xmath121 , where at leading order @xmath362,\ \ \ \ \ x_a \gg x_b . \label{eq : sig_pd_pp}\end{aligned}\ ] ] the results , illustrated in fig . [ fig : dbar_ubar ] , confirmed a significant deviation of the cross section ratio from the naive perturbative qcd expectation , indicating a large asymmetry between @xmath273 and @xmath272 . and @xmath134 scattering as a function of the fractional momentum @xmath143 of the target parton . the data points from the fermilab e866/nusea collaboration @xcite are compared with the cross section ratio calculated using the jr09 @xcite ( red solid line ) , mstw08 @xcite ( blue dashed ) , and abm11 @xcite ( green dotted ) , and nnpdf23 @xcite ( yellow dot - dashed ) pdf parametrizations.,width=453 ] the earliest efforts to explain an enhancement of @xmath273 quarks over @xmath272 in the proton invoked the effects of pauli blocking , which arise from the presence of different numbers of valence @xmath141 and @xmath142 quarks in the proton @xcite . estimates of this effect are very difficult to quantify reliably @xcite , with some calculations @xcite even suggesting that it produces a small excess of @xmath272 over @xmath273 . alternative explanations have focused on the role of chiral symmetry breaking and the pion cloud of the nucleon @xcite . because of the heisenberg uncertainty principle , a component of the proton s wave function involves its dissociation into a nucleon and a virtual pion state . the preferential coupling of a proton to a @xmath363 , with the @xmath364 containing a valence @xmath365 pair , implies a natural mechanism for generating an excess of @xmath273 over @xmath272 . while many models have been proposed to compute this effect quantitatively ( see refs . @xcite for reviews ) , the underlying physics principles are model independent and simply the chiral symmetry properties of qcd @xcite . while most of the modern pdf parametrizations are able to accommodate an enhanced @xmath273 sea over the range @xmath366 , there is some uncertainty in the trend of the @xmath264 ratio at larger @xmath2 . the new drell - yan experiment e906/seaquest at fermilab @xcite will extend the @xmath2 coverage up to @xmath367 , and a proposal at the j - parc facility in japan @xcite would extend the range even further . most pdf analyses have been performed under the assumption of charge symmetry , or independence of interactions under rotations in isospin space . charge symmetry implies that the @xmath141 quark distribution in the proton is identical to the @xmath142 quark distribution in the neutron , and _ vice versa _ , and is believed to be accurate in nature at the @xmath368 level . this symmetry is broken explicitly by light quark mass differences , @xmath369 , as well as by electromagnetic corrections . such effects lead to nonzero values of the `` majority '' and `` minority '' charge symmetry violating asymmetries in the nucleon , defined as @xmath370 respectively , and similarly for the antiquarks @xmath371 . although there is currently no direct evidence for charge symmetry violation ( csv ) in pdfs , it has been predicted in several nonperturbative models for both valence and sea quark pdfs @xcite . in addition , electromagnetic radiative effects can be included in the @xmath40 evolution equations ( sec . [ ssec : q2evol ] ) , in analogy with usual gluon radiation , by adding a term that accounts for the emission of photons by quarks @xcite , @xmath372 here @xmath373 are the new qed splitting functions ( expanded in terms of the electromagnetic coupling @xmath52 ) , and the sum over partons includes also the photon pdf @xmath374 . even with charge symmetric pdfs at the input scale , the different electric couplings of the photon to @xmath141 and @xmath142 quarks in the modified evolution equations will induce nonzero csv distributions ( [ eq : csv_u ] ) and ( [ eq : csv_d ] ) at a higher scale . the valence quark asymmetries resulting from the qed - modified @xmath40 evolution ( [ eq : qeddglap ] ) are illustrated in fig . [ fig : csv_ssbar ] at a scale @xmath375 gev@xmath170 . the perturbatively generated csv distributions from glck _ et al . _ @xcite are calculated with an average isoscalar quark mass @xmath376 mev , with the qed radiative corrections computed after fixing the qcd evolution effects . the majority valence @xmath377 distribution is negative at intermediate @xmath2 values , and is of opposite sign to the minority valence distribution @xmath378 , which is smaller in magnitude . qualitatively , these features are similar to the nonperturbative csv valence asymmetries computed in the bag model @xcite , which arise from differences between the spectator @xmath379 and @xmath380 diquark masses , as well as the proton neutron mass difference . the mrstqed @xcite analysis of csv was similar to that in ref . @xcite , but involved unequal photon pdfs in the proton and neutron resulting from qed radiation with quark masses @xmath381 mev . since the photon now also contributes to the momentum sum rule , eq . ( [ eq : mom_sum_rule ] ) , the csv photon distributions induce charge symmetry violation in the valence quark pdfs , assuming the sea and gluon are charge symmetric . despite the different assumptions , the resulting distributions turn out to be similar to those in fig . [ fig : csv_ssbar ] . and `` minority '' @xmath378 valence quark distributions for the radiative qed fit of ref . @xcite at @xmath375 gev@xmath170 , compared with the bag model estimates have been taken from ref . _ ( right ) _ strange antistrange asymmetry at @xmath382 gev@xmath170 ( appropriate for the nutev experiment ) for the asymmetric gjr08 @xcite ( red solid ) and the mstw2008 @xcite ( blue dashed ) pdfs.,title="fig:",width=253 ] and `` minority '' @xmath378 valence quark distributions for the radiative qed fit of ref . @xcite at @xmath375 gev@xmath170 , compared with the bag model estimates have been taken from ref . _ ( right ) _ strange antistrange asymmetry at @xmath382 gev@xmath170 ( appropriate for the nutev experiment ) for the asymmetric gjr08 @xcite ( red solid ) and the mstw2008 @xcite ( blue dashed ) pdfs.,title="fig:",width=364 ] while the overall magnitude of the possible csv effects is rather small , they can nevertheless play an important role in understanding the discrepancy between the nutev measurement of @xmath383 @xcite and the standard model value @xcite . the nutev collaboration extracted the paschos - wolfenstein ( pw ) ratio @xcite of total neutral to charged current cross sections , involving the difference between @xmath207 and @xmath384 cross sections . in addition to assuming small uncertainties on the nuclear corrections necessary for translating the measured iron target data to an isoscalar nucleon target , the analysis also neglected any csv effects . including csv corrections , the experimental pw ratio would receive a contribution proportional to @xmath385 @xcite . depending on the sign of the csv distributions , the asymmetry can therefore either remove some of the discrepancy with the nutev result , or exacerbate it . the signs and magnitude of the predicted perturbative qed and nonperturbative quark model asymmetries in fig . [ fig : csv_ssbar ] would in fact remove approximately 2/3 of the discrepancy . an additional correction to the pw ratio can arise from differences between strange and antistrange distributions in the nucleon , @xmath386 , which we discuss in the next section . traditionally the strange quark distribution has been more difficult to determine experimentally than the non - strange sea . in practice one has often assumed that the @xmath254 and @xmath255 pdfs ( which are often assumed to be identical ) are proportional to the @xmath345 distributions , @xmath387 with @xmath388 ranging between @xmath389 and 0.5 . the @xmath254 and @xmath255 distributions can be directly determined , however , through the measurement of dimuon pairs in neutrino dis , which are produced through the charged current by scattering from strange quarks with subsequent generation of charm . in particular , while neutrino scattering produces charmed hadrons ( @xmath390 ) , antineutrino scattering results in anticharmed hadrons ( @xmath391 ) . the semileptonic decay of the charmed hadrons ( such as @xmath392 and @xmath393 mesons ) then yields @xmath394 pairs along with the associated hadronic debris . because of the low rates involved in neutrino scattering experiments , in order to increase the statistics these have typically made use of heavy nuclear targets , such as iron . since the nuclear effects for neutrino and charged lepton scattering are in general different @xcite , the nuclear corrections introduce an additional source of uncertainty in extractions of the strange quark pdf . a number of nonperturbative models have in fact predicted an asymmetry between the @xmath254 and @xmath255 distributions in the nucleon , which can be probed by taking the difference between the @xmath207 and @xmath384 cross sections . for example , in analogy with the pion cloud models invoked to explain the @xmath273 excess over @xmath272 in the proton in sec . [ ssec : lightsea ] , the dissociation of a nucleon to a hyperon and a virtual kaon naturally produces unequal @xmath254 and @xmath255 distributions , although the magnitude and even the sign is difficult to determine unambiguously @xcite . in addition , an asymmetry can also be generated through perturbative qcd evolution at nnlo because of unequal valence @xmath141 and @xmath142 distributions @xcite . the ccfr @xcite and nutev @xcite collaborations at fermilab have collected the highest statistics data on inclusive charm production in @xmath207 and @xmath384 scattering on an iron target , enabling the most direct constraints to be placed on the strange quark pdfs as a function of @xmath2 . the latter in particular was the first analysis using a complete nlo description of charm production . the cross section for ( anti)neutrino charm production was computed to nlo in refs . @xcite within the ffns , in which , apart from the gluon , only the light quark flavors ( @xmath395 ) are included as massless partons ( see secs . [ ssec : hq ] and [ ssec : hqresults ] ) . the fully differential nlo calculation , including acceptance corrections , was performed in ref . ( note , however , that the use of the expressions in @xcite together with vfns distributions , as implemented in the nutev analysis @xcite , is strictly not consistent . ) comparison of the @xmath207 and @xmath384 events found a preferred fit for the @xmath396 distribution that peaked at @xmath397 , with the compensating negative contribution required by the sum rule @xmath398 to be restricted to the unmeasured region at @xmath399 @xcite . for the first moment , the value obtained at @xmath382 gev@xmath170 was @xmath400 , where the errors are statistical , systematic , and external . the latter includes a dominant contribution from the charm semileptonic branching ratio uncertainty , as well as from nuclear corrections @xcite . the strangeness asymmetry in the nucleon is shown in fig . [ fig : csv_ssbar ] at @xmath382 gev@xmath170 , as relevant for the nutev experiment , for several different pdf parametrizations . the mstw08 @xcite strange asymmetry is fitted to the ccfr and nutev data assuming the form @xmath401 where the factor @xmath402 ensures zero net strangeness in the nucleon . the strange asymmetry in ref . @xcite is generated radiatively from the boundary condition @xmath403 at @xmath404 gev@xmath170 , but allowing @xmath405 to have the form in eq . ( [ eq : sminus ] ) . the resulting @xmath405 asymmetry at the experimental scale is similar to the mstw08 fit , and to the phenomenological extraction in ref . @xcite , within the currently sizable uncertainties . the values for the momentum weighted asymmetry are the order @xmath406 for the mswt08 fit @xcite and @xmath407 for the dynamically generated pdfs @xcite at scales @xmath408 gev@xmath170 . the cteq collaboration found a strange asymmetry in the range @xmath409 @xcite . these values have the correct sign and similar magnitude to that required to reconcile the nutev @xmath383 measurement with the standard model value @xcite , once the charge symmetry violating effects in sec . [ ssec : csv ] are also included @xcite . finally , a recent measurement by the atlas collaboration @xcite of inclusive @xmath57 and @xmath194 boson production in @xmath134 collisions at the lhc found a significantly larger value of the strange to non - strange sea ratio in eq . ( [ eq : kappa ] ) , @xmath410 at @xmath411 for @xmath412 gev@xmath170 , than in previous analyses . a reanalysis of the atlas and hera data by the nnpdf group @xcite found , however , that the uncertainty on the extracted @xmath388 was significantly underestimated , and that a more complete global analysis gives @xmath388 values consistent with the traditional values . heavy quark production contributes considerably to inclusive dis ; for example , the charm ( bottom ) contributions to the dominant @xmath81 structure function constitute up to 30% ( 3% ) of the total in the small@xmath2 region . besides these contributions , heavy quark electroproduction is directly accessed experimentally in semi inclusive dis , typically by detecting charm ( bottom ) mesons in the final state . because the main production mechanism is photon gluon fusion , data on this process provide valuable constraints on the gluon pdf in the nucleon . furthermore , they provide an appropriate context for testing the different approaches to these calculations that have been adopted in various pdf analyses ( see sec . [ ssec : hq ] ) . since it is by far the most relevant , we focus the discussion on the charm flavor , although similar considerations often also apply to bottom ( top contributions to this process , on the other hand , are negligible at accessible kinematics ) . the most relevant data on heavy quark electroproduction are the ( h1 + zeus ) combination of hera measurements @xcite . a comparison of these data with ffns calculations at lo , nlo and nnlo * ( see sec . [ ssec : hq ] ) in fig . [ fig : charm ] ( left ) shows excellent agreement between the ffns calculations . in fact , the experimental analysis of the hera data is based primarily on the ffns fully differential ( exclusive ) calculations of ref . @xcite . despite this fact , the constructs in the gm - vfnss are often motivated by arguing that pseudo - mass - divergences ( non - collinear logarithms of the form @xmath413 ) which appear in the wilson coefficients of the massive calculations need to be resummed . however , the ffns results are notably stable even for very large values of @xmath414 @xcite , in particular at the largest @xmath40 values accessible at hera , suggesting little need to resum these supposedly `` large logarithms '' in the context of global pdf analyses . finally , in addition to the perturbative generation of charm , it is also possible for charm to be produced through nonperturbative mechanisms , such as those discussed in refs . @xcite ( and references therein ) . first postulated to account for some early charm production data in hadronic reactions , these `` intrinsic '' charm contributions share several characteristic features , even though their details depend somewhat on the models . in particular , they are typically valence - like , with significantly harder @xmath2 distributions than those generated perturbatively , and can produce large asymmetries between the @xmath262 and @xmath415 distributions . measurements of the charm structure function @xmath416 by emc @xcite , especially the two data points at the highest @xmath40 in fig . [ fig : charm ] ( right ) , have been interpreted as indicating a nonzero intrinsic charm contribution at large @xmath2 . the evidence is not conclusive , however , in view of the disagreement evident in fig . [ fig : charm ] ( right ) between the emc data and theoretical predictions at smaller @xmath2 , where the predictions are in agreement with the more recent hera results in fig . [ fig : charm ] ( left ) . the current limits on the normalization of the nonperturbative charm quark distribution range from @xmath417 to 2% , depending on which data sets are fitted and which models of intrinsic charm considered @xcite . although current global pdf analyses do not require an intrinsic charm component , the existence of nonperturbative charm remains an intriguing issue which can be addressed by high - precision measurements of charm production at future facilities such as j - parc @xcite , gsi - fair @xcite , jefferson lab at 12 gev @xcite , after@cern @xcite , or an electron - ion collider @xcite . in addition to the heavy quark production discussed above , there are three primary sources of information on the gluon distribution : the @xmath40 dependence of the dis structure function @xmath81 at low values of @xmath2 ; the longitudinal structure function @xmath257 at all @xmath2 values ; and jet ( both at hera and at hadron colliders ) and photon production cross sections at moderate to high values of @xmath2 . furthermore , the momentum sum rule ( [ eq : mom_sum_rule ] ) provides an additional constraint on the gluon pdf . since the normalizations of the valence pdfs are fixed by the flavor sum rules in eq . ( [ eq : sum_rules ] ) and the contributions of the light antiquarks are relatively well determined by dis and lepton pair production data , the momentum sum rule effectively constrains the gluon contribution . the effect is an anticorrelation between the low- and high-@xmath2 behavior of the gluon pdf : a decrease ( increase ) of the gluon distribution at low @xmath2 leads to an increase ( decrease ) at high values of @xmath2 . interestingly , this can result in more constrained gluon distributions in the medium-@xmath2 region ( see fig . [ fig : gluon ] ) . gev@xmath170.,width=377 ] to understand the relationship between the scale dependence of @xmath81 and the gluon pdf , consider eq . ( [ eq : rge ] ) for the scale dependence of the quark pdf . multiply both sides by @xmath2 and by the squared parton charge and sum over the active flavors . in this way one derives ( using the lo expression for @xmath81 ) , @xmath418 \label{eq : f2q2}\ ] ] where @xmath419 and @xmath420 are the quark quark and quark gluon splitting functions . at small values of @xmath2 the gluon term dominates the integrand so that the scale dependence of @xmath81 places a constraint on the gluon pdf . a second constraint on the gluon from dis comes from the longitudinal structure function @xmath257 , which is defined as @xmath421 in lowest order , where only quarks contribute to @xmath80 and @xmath81 , one has @xmath422 so that @xmath423 . the first nonzero contribution to @xmath257 starts at @xmath256 where , at small @xmath2 , there is a significant contribution from the subprocess @xmath424 , which again imposes strong constraints on the gluon pdf at small @xmath2 . at larger @xmath2 values the valence quark contributions increase , the gluon pdf decreases , and the subprocess @xmath425 dominates . however , since the same combinations of quark distributions are well determined from @xmath81 , where they enter with one power of @xmath34 less and the gluon does not contribute , precision measurements of the longitudinal structure function provide valuable information on the gluon pdf at large @xmath2 @xcite . note also that the gluon pdf is accompanied by the same powers of @xmath34 in all dis structure functions , starting with @xmath426 . this results in a correlation between the value of @xmath34 obtained in global pdf analysis and the size and shape of the gluon distribution , with larger @xmath34 leading to a smaller gluon pdf in the small-@xmath2 region and ( via the momentum sum rule correlation ) a larger gluon pdf at large @xmath2 . this can be seen in fig . [ fig : gluon ] , where the jr09 @xcite and abm11 @xcite analyses obtain @xmath34 values and small-@xmath2 gluon pdfs of similar size , while the mstw08 fit @xcite finds a larger @xmath34 value and a smaller gluon pdf in this region ( which even turns negative ) . both possibilities describe well the inclusive cross sections measurements at hera . this ambiguity can be reduced by using processes with a different leading @xmath34 power , typically jet production at the tevatron , which , as noted in sec . [ sssec : unpol ] , can provide constraints on the gluon pdf . to understand this , recall that the calculation of the jet cross section involves a sum over @xmath427 , @xmath276 and @xmath277 induced subprocesses where `` @xmath85 '' stands for any flavor of quark or antiquark . at low values of jet @xmath110 , the @xmath277 subprocesses dominate , at intermediate values all three contribute , and eventually as @xmath110 approaches one only the @xmath427 terms survive see fig . 3 in ref . @xcite , for example . jet data at the tevatron extend to @xmath110 values of approximately 0.5 , so these data provide some constraints on the gluon pdf , at least in the mid-@xmath2 range where the @xmath276 subprocesses make substantial contributions . as discussed in ref . @xcite , isolated photon production at collider energies has the potential to tighten the error bands in the @xmath2 range below about 0.2 . it remains a challenge to find processes that will constrain the gluon beyond @xmath428 . the fundamental problem is that the valence pdfs here are larger than the gluon pdf . most hard processes that receive contributions from @xmath276 subprocesses also receive contributions from @xmath427 processes , which reduce the sensitivity to the gluon pdf . one counterexample is provided by direct photon production in @xmath134 collisions . the two dominant subprocesses are initiated by @xmath258 and @xmath276 initial states . at large values of @xmath110 the fragmentation process is suppressed ; it can be further reduced by photon isolation cuts . the single photon cross section involves an integration over the awayside jet rapidity which tends to smear out the @xmath2 values of the contributing pdfs . there are regions where one @xmath2 is small and one is large one might think that this could constrain the large-@xmath2 gluon . however , the favored configuration will be where the gluon or @xmath86 is at small values of @xmath2 while the valence quark is at the larger values . instead , consider the photon + jet cross section with equal and opposite photon and jet rapidities , in which case both momentum fractions will be equal to @xmath429 using lo kinematics . going to large values of @xmath110 and @xmath430 would then yield information on the gluon at large @xmath2 , since the @xmath86 contribution is much smaller there than that of the gluon . this would also be the case for lepton pair @xcite and vector boson production in @xmath134 collisions , provided that one considers @xmath431 in order for the parton momentum fraction to be large . the analysis in ref . @xcite suggests that photon + jet data with increased statistics from the lhc may help constrain the gluon and light quark pdfs . another source of information on the gluon pdf is top quark pair production , for which the lowest order subprocess is @xmath432 . in ref . @xcite the constraints on the gluon pdf provided by @xmath433 data from the tevatron and lhc were examined , with the results indicating that these data place strong constraints on the large-@xmath2 gluon pdf . also of note is the analysis of ratios and double ratios of cross sections measured at different lhc energies @xcite . such ratios have reduced systematic errors and have the potential to provide strong constraints on large-@xmath2 pdfs . in addition to the issues discussed above , an obstacle for the determination of the gluon distribution , and the highly correlated @xmath34 values from pdf analysis , is that the calculations for the relevant processes are not available beyond nlo , with the important exceptions of the ( light quark contributions to the ) inclusive dis structure functions and drell - yan cross sections . for example , the relative uncertainties of the jr09 gluon pdf in fig . [ fig : gluon ] are below about 10% for @xmath434 , and increase as one proceeds to higher values of @xmath2 , where the constraints of the @xmath81 structure function from fixed target experiments used in that fit are weaker . the uncertainty at large @xmath2 could in principle be reduced by including jet data . however , this would also affect the central values obtained for the gluon pdf , changing them in the relevant region by an amount proportional to the missing nnlo corrections , which in the case of jet production are expected to be large @xcite . global nnlo fits which include jet data , such as the mstw08 @xcite parametrization in fig . [ fig : gluon ] , generally find larger gluon pdfs in the high-@xmath2 region and correspondingly also larger values of @xmath34 . yet another factor which can alter the pdfs and the value of @xmath34 obtained in a particular analysis are the kinematic cuts applied ; of particular relevance are the cuts on the dis structure function data and the related treatment of contributions beyond leading twist @xcite . although data selection ( data sets and cuts ) bring about certain arbitrariness in global pdf analysis , it can not account for all the existing differences between the various pdf sets , which are also affected by theoretical issues such as heavy quark schemes ( sec . [ ssec : hq ] ) and the particular solutions of the rge equations used . even within a given framework there are uncertainties due to the inability of the estimation procedure to find the optimal solution , for example , shortcomings of the parametrization to reproduce the optimal shape of the distributions , and the statistical estimation procedure . these effects , referred in general as ` procedural bias ' , induce a dependence of the results on the choice of the scale at which the input distributions are parametrized , which in principle should not depend on these choices . for particularly sensitive quantities such as the gluon distribution and @xmath34 , this uncertainty can be comparable to the experimental uncertainty @xcite , as can be seen in fig . [ fig : alphas ] . values obtained for different parametrizations in ref . the band indicates the uncertainties due to propagation of the experimental errors for a particular parametrization ( nnlo20).,width=377 ] another example of this is the differences between `` standard '' distributions , generated from input pdfs at values for the input scale @xmath435 gev@xmath170 , and the so called dynamical distributions @xcite , for which input scale is chosen to be @xmath436 gev@xmath170 . at these low scales the input distributions naturally tend to valence like ( positive definite ) shapes @xcite that vanish in the small@xmath2 limit , not only for valence but also for the sea quark and gluon input densities . it was shown in refs . @xcite , where also `` standard '' distributions were generated , that this implies that the dynamical distributions at small @xmath2 are more restricted and have smaller uncertainties than their standard counterparts . furthermore , this procedure typically results in smaller @xmath34 values and steeper gluons in the small@xmath2 region , although both approaches provide comparable descriptions of data , in particular , of the inclusive dis cross sections from hera . an alternative approach to fitting @xmath34 as part of the global fitting is to use the world average value for @xmath236 , thereby avoiding the correlations during the fitting process . since @xmath34 is a parameter of the qcd lagrangian , it should be the same for all processes , not just those in the global fit . of course , interesting information could be obtained by determining the pdfs both with a fixed value of @xmath236 and with a fitted value . comparing the results could indicate which processes in the global fit are chiefly responsible for any differences . the moments of leading twist pdfs can be evaluated from first principles using lattice qcd . within the operator product expansion , the moments of pdfs are related to matrix elements of local operators between hadron states , which can be computed numerically on the lattice . the lattice pdf moments can be related to the quark and antiquark distributions in the nucleon as @xmath437 , \label{eq : mom_unp } \\ \langle x^n \rangle_{\delta q } & = & \int_0 ^ 1 dx\ , x^n\ [ \delta q(x)+(-1)^n \delta \bar q(x ) ] , \label{eq : mom:_pol}\end{aligned}\ ] ] for the unpolarized and helicity distributions ( see sec . [ sec : spin ] ) , respectively . the lattice moments alternate between the total ( or @xmath11-even ) @xmath438 and valence ( @xmath11-odd ) @xmath348 distributions , depending on whether @xmath15 is even or odd . while reconstructing the @xmath2 dependence of the pdfs from a finite number of calculated moments is challenging @xcite , comparison of the lattice moments with the phenomenological pdf moments can provide important tests of lattice techniques and provide constraints on specific pdfs which may be difficult to measure experimentally . because of the discretization of space - time on the lattice , the rotational symmetry of the continuum is broken to the hyper - cubic subgroup , which introduces mixing with lower dimensional operators under renormalization . as a result , only the lowest few moments , corresponding to @xmath439 , can at present be reliably computed . as well as the usual approximations of working at a finite lattice spacing @xmath18 , in a finite lattice volume @xmath440 , and at unphysically large values of the quark mass , simulations usually include contributions only from `` connected '' diagrams , representing operator insertions on quark lines connected to the source . meaningful comparisons of lattice data can therefore only be made with nonsinglet combinations , such as @xmath441 , where the `` disconnected '' contributions cancel . significant progress has been made in recent years in the computation of matrix elements of twist - two operators on the lattice , using dynamical ( unquenched ) configurations and quark ( or pion , by the gell - mann oaks renner relation ) masses approaching the physical limit . unfortunately , the results for the lowest nontrivial ( @xmath442 ) nonsinglet moment of the unpolarized distributions , @xmath443 , lie systematically in the range @xmath444 at a scale @xmath445 gev@xmath170 , which is some 40% above the phenomenological pdf moments of @xmath446 @xcite . a smaller , but persistent 10% overestimate of the lowest ( @xmath447 ) moment of the nonsinglet spin - dependent distribution , @xmath448 , has posed a serious challenge to lattice simulations @xcite . it has been suggested @xcite that the discrepancy between the calculated and experimental pdf moments could be resolved by the nontrivial chiral behavior of the moments , when extrapolating the results from the lowest @xmath449 values at which lattice data exist to the physical mass . verification of this behavior will require lattice simulations at very small quark masses , as well as at large enough volumes to fully include the physics of the pion cloud @xcite ( see also sec . [ ssec : lightsea ] ) . while it may take some time before the lattice parameters can become sufficiently close to those in the physical realm , such comparisons will continue to reflect the important synergy between lattice qcd and pdf phenomenology . considerable progress has been made in understanding the spin structure of the nucleon since the first precision polarized dis experiments at cern in the late 1980s @xcite indicated an anomalously small fraction of the proton spin carried by quarks . a rich program of spin - dependent inclusive and semi - inclusive dis , as well as polarized proton proton scattering experiments has followed , vastly improving our knowledge of spin - dependent pdfs of the nucleon over the last two decades ( for a recent review , see ref . @xcite and references therein ) . while the spin - dependent data have not been as abundant as those available for constraining spin - averaged pdfs , the existing empirical information from polarized lepton and hadron facilities at cern @xcite , slac @xcite , desy @xcite , jefferson lab @xcite and rhic @xcite has enabled several dedicated global qcd analyses of spin - dependent parton distributions to be performed . these include the nlo analyses from the more established dssv @xcite group , the european - based lss @xcite and bb @xcite groups , as well as the japanese asymmetry analysis collaboration ( aac ) @xcite , using the standard global fitting methodology outlined in sec . [ sec : qcd ] . more recent efforts have been made by the nnpdf collaboration @xcite , extending the neural network approach to the polarized sector , and the jefferson lab angular momentum ( jam ) collaboration @xcite , which has focused on utilizing data over a large kinematic range that includes the large-@xmath2 and low-@xmath40 regions , with a simultaneous fit of unpolarized pdfs . and @xmath450 , the antiquark @xmath451 , @xmath452 and @xmath453 , and polarized gluon @xmath454 distributions at @xmath191 gev@xmath170 , for the dssv09 @xcite , lss10 @xcite , bb10 @xcite , aac09 @xcite , and jam13 @xcite pdf sets.,width=548 ] representative examples of the spin - dependent pdfs are shown in fig . [ fig : spinpdfs ] , illustrating the total ( or @xmath11-even ) @xmath455 and @xmath456 pdfs , as well as the sea quark @xmath457 , @xmath458 and @xmath459 and polarized gluon @xmath152 distributions , for the dssv09 @xcite , lss10 @xcite , bb10 @xcite , aac09 @xcite , and jam13 @xcite pdf sets at @xmath191 gev@xmath170 . the bb10 @xcite and jam13 @xcite analyses are based on inclusive dis data only , while the lss10 @xcite fit includes also semi - inclusive dis data . the dssv09 pdfs are constrained in addition by polarized @xmath134 scattering data . the latest aac analysis @xcite is based on inclusive dis data and a @xmath297-factor approximation for the nlo corrections for the @xmath134 data . as for the unpolarized case in fig . [ fig : nspdf ] , the @xmath455 distribution is the best constrained polarized pdf , mostly by measurements of the proton @xmath90 structure function over a relatively broad range of @xmath2 . the corresponding @xmath456 distribution , which has the opposite sign , is smaller in magnitude compared with the @xmath455 with somewhat larger uncertainties , especially at larger @xmath2 values . the uncertainty band for the dssv09 parametrization of @xmath455 and @xmath456 is smaller than the variation between the different pdf sets , which reflects the fact that the systematic uncertainties associated with the choice of data sets and parametrization assumptions , as well as other theoretical inputs , are currently larger than the experimental errors . the polarization of the sea is considerably smaller , and more strongly dependent upon assumptions about the flavor dependence in the analysis of semi - inclusive dis data . at present there is no conclusive experimental evidence of a nonzero light quark sea , @xmath457 and @xmath458 , although there is a slight trend towards a more negatively polarized @xmath273 distribution than @xmath272 , with a positive @xmath460 . the polarization of the strange sea is also very small , in contrast to suggestions from the early spin - dependent dis data analyses of a large negative @xmath459 . the polarized gluon distribution is essentially unconstrained by existing inclusive and semi - inclusive dis data , and almost all information on @xmath152 comes from measurements of @xmath461 production in semi - inclusive dis @xcite , and inclusive pion and jet production in polarized @xmath134 scattering @xcite . the data are consistent with a small value of @xmath462 , consistent with zero , although new measurements from rhic have the promise of resolving a small nonzero distribution . at large values of @xmath2 , the spin - dependent pdfs are even more sensitive to the quark - gluon dynamics responsible for spin - flavor symmetry breaking than the spin - averaged valence quark pdfs discussed in sec . [ ssec : valence ] . however , while considerable progress has been made in determining spin - dependent pdfs over the last two decades , especially in the small-@xmath2 region , relatively little attention has been paid to structure function measurements at large @xmath2 . the dearth of data in the valence region is especially striking for the neutron , where , with the exception of several data points from jefferson lab hall a @xcite for the polarization asymmetry @xmath463 extending to @xmath464 , there are essentially no constraints for @xmath465 . on the other hand , there are a number of predictions for the behavior of polarized pdfs and ratios of polarized to unpolarized pdfs in the limit as @xmath165 that differ even in sign . at large @xmath2 there are dramatically different predictions for the valence @xmath283 and @xmath282 distributions from nonperturbative and perturbative calculations . in the simple su(6 ) spin - flavor symmetric quark model , the equal probabilities of the spin-0 and spin-1 spectator diquark configurations lead to the simple relation @xmath466 , and corresponding polarized to unpolarized ratios @xmath467 and @xmath468 . in this limit the polarization asymmetry of the proton , which at lo can be written @xmath469 is predicted to be @xmath470 , while that of the neutron , @xmath471 would vanish , @xmath472 . while spin - flavor symmetry is of course broken in nature ( in some cases badly ) , with the unpolarized @xmath142 quark distribution significantly softer than the @xmath141 , for example , the details of this breaking can affect the spin - dependent pdfs in strikingly different ways . the mechanism of spin-1 diquark suppression , which leads to the vanishing of the @xmath145 ratio as @xmath165 , also predicts that @xmath473 in this limit , while @xmath474 remains unchanged from the su(6 ) expectation . arguments based on helicity conservation in perturbative qcd , on the other hand , predict that for valence quarks in the ground state of the nucleon the scattering at large @xmath2 is predominantly from configurations in which the quark is coupled to a diquark of zero helicity @xcite . in this case the pdfs associated with a quark polarized in the same direction as the nucleon are favored over the helicity - antialigned configurations , @xmath475 as @xmath165 . this model leads to the expectation that @xmath473 and @xmath476 in the @xmath165 limit , and consequently a rapid approach to unity of both the proton and neutron @xmath92 asymmetries . an additional feature of the helicity conservation model is that it correlates the @xmath165 behavior of the pdfs with the large-@xmath40 behavior of elastic form factors @xcite , which can be measured in elastic lepton nucleon scattering experiments at jefferson lab and elsewhere @xcite . this is also closely related to the notion of quark - hadron duality , which connects averages over structure functions in the resonance region at low @xmath70 with those extrapolated from the deep - inelastic continuum at higher @xmath70 ( lower @xmath2 ) @xcite . since the existing dis data are for the most part consistent with a vanishingly small neutron asymmetry at @xmath477 , a very dramatic change in @xmath463 would be expected at larger @xmath2 . an even more rapid transition at high @xmath2 may occur if quark orbital angular momentum plays an important role in nucleon structure @xcite . and @xmath478 distributions and their uncertainties for the jam13 pdf set @xcite at @xmath191 gev@xmath170 , illustrating the effects of the nuclear smearing ( green dotted ) , target mass ( blue dashed ) and higher twist ( red solid ) corrections , relative to the reference fit ( yellow dot - dashed ) . note that the yellow and green bands overlap for @xmath479 and the green and blue bands overlap for @xmath478 . _ ( right ) _ corresponding ratios of polarized to unpolarized @xmath480 and @xmath481 distributions.,title="fig:",width=302 ] and @xmath478 distributions and their uncertainties for the jam13 pdf set @xcite at @xmath191 gev@xmath170 , illustrating the effects of the nuclear smearing ( green dotted ) , target mass ( blue dashed ) and higher twist ( red solid ) corrections , relative to the reference fit ( yellow dot - dashed ) . note that the yellow and green bands overlap for @xmath479 and the green and blue bands overlap for @xmath478 . _ ( right ) _ corresponding ratios of polarized to unpolarized @xmath480 and @xmath481 distributions.,title="fig:",width=302 ] experimentally , the extraction of leading twist spin - dependent pdfs at high @xmath2 faces similar challenges to those discussed in sec . [ ssec : valence ] for the unpolarized distributions . in particular , information on the structure of the neutron is typically obtained from experiments involving polarized deuterium or @xmath221he nuclei , requiring the nuclear corrections to be understood . analysis of high-@xmath2 data also necessitates careful treatment of finite-@xmath40 corrections such as target mass and higher twist contributions . the effects of these corrections are illustrated in fig . [ fig : pol_val ] , which shows the total @xmath455 and @xmath456 pdfs from the jam13 analysis @xcite for various fits with or without the finite-@xmath40 and nuclear corrections . at small values of @xmath2 their effects are negligible ; however , at @xmath482 they can give up @xmath483 corrections for @xmath455 and more @xmath177 corrections for @xmath456 ( the relative correction can be even larger at @xmath484 , although the pdfs are not constrained in this region ) . the same effects are more clearly illustrated through the polarization ratios @xmath480 and @xmath481 in fig . [ fig : pol_val ] , where the unpolarized distributions are fitted simultaneously within the same analysis as the polarized @xcite . this in principle eliminates any bias arising from the use of spin - averaged pdfs taken from analyses performed under different sets of assumptions . in the intermediate-@xmath2 region the ratios for both @xmath141 and @xmath142 quarks are generally consistent with the symmetric quark model expectations , with the @xmath480 increasing towards unity at larger @xmath2 . the @xmath481 ratio , on the other hand , remains negative for all @xmath2 where it is constrained , and shows no indication of the upturn predicted by the helicity conservation models . the nuclear and finite-@xmath40 effects can significantly impact the limiting behavior as @xmath165 , and clearly additional data are needed in order to constrain their @xmath2 dependence at high @xmath2 . a dedicated program of large-@xmath2 structure function measurements is planned at the 12 gev energy upgraded jefferson lab facility @xcite . more complete information on spin - dependent pdfs will also be available once the @xmath91 structure function of the nucleon is measured more accurately @xcite . this will provide additional information on possible higher twist corrections that enter in the extraction of leading twist pdfs from dis polarization asymmetries . as for the unpolarized inclusive charged lepton dis measurements , inclusive @xmath90 structure function experiments measure @xmath11-even combinations of pdfs , @xmath278 . the individual quark and antiquark distributions can be separated with the help of semi - inclusive charged lepton scattering data , with coincident measurement of fast pions or kaons in the final state . typically such experiments measure the semi - inclusive polarization asymmetry , which at lo can be written @xmath485 where @xmath486 is the fragmentation function for the scattered quark or antiquark to produce a hadron @xmath285 ( @xmath487 ) in the current fragmentation region with a fraction @xmath488 . for large @xmath288 , the produced hadron has a high probability of containing the scattered parton , hence providing a tag on the initial state distribution of quarks and antiquarks . the fragmentation functions @xmath289 are extracted from independent fits to single hadron production cross sections in @xmath290 , @xmath134 and other reactions . , from the compass semi - inclusive dis measurement at @xmath489 gev@xmath170 @xcite , compared with several parametrizations and model calculations , and the unpolarized @xmath305 asymmetry ( solid curve ) . ( figure from ref . @xcite . ) _ ( right ) _ single polarization asymmetry @xmath490 for inclusive @xmath57 production in @xmath491 scattering from the star experiment at rhic , at central electron pseudorapidity @xmath492 . ( figure from ref . copyright ( 2011 ) by the american physical society.),title="fig:",width=340 ] , from the compass semi - inclusive dis measurement at @xmath489 gev@xmath170 @xcite , compared with several parametrizations and model calculations , and the unpolarized @xmath305 asymmetry ( solid curve ) . ( figure from ref . _ ( right ) _ single polarization asymmetry @xmath490 for inclusive @xmath57 production in @xmath491 scattering from the star experiment at rhic , at central electron pseudorapidity @xmath492 . ( figure from ref . copyright ( 2011 ) by the american physical society.),title="fig:",width=226 ] the program of semi - inclusive dis measurements from proton and deuteron targets at hermes @xcite , compass @xcite , and earlier smc @xcite , have produced intriguing glimpses into the flavor content of the polarized sea . the extracted sea quark and antiquark asymmetries are generally rather small , and consistent with zero within the current uncertainties . for the light quark polarized sea , there is a slight trend towards a positive @xmath457 ( with an @xmath2-integrated value of @xmath493 from the latest compass data @xcite ) and a negative @xmath458 ( with an @xmath2-integrated value of @xmath494 ) , so that the difference @xmath460 is slightly positive , at the @xmath495 level @xcite , as fig . [ fig : pol_sea ] indicates . this is in contrast to the large negative values of the unpolarized @xmath496 asymmetry ( see sec . [ ssec : lightsea ] ) , which would disfavor models such as the chiral quark soliton model in the large-@xmath497 limit @xcite that predict large @xmath498 . very small antiquark spin asymmetries would be expected in models based on the pion cloud of the nucleon , such as those discussed in sec . [ ssec : lightsea ] . one should caution , however , that the experimental distributions in fig . [ fig : pol_sea ] were extracted assuming lo expressions for the semi - inclusive asymmetries , which , strictly speaking , can not be directly compared with nlo pdfs @xcite . the comparison should therefore be viewed as a qualitative one , although the general indication of a small polarized non - strange sea is unlikely to be qualitatively modified in an nlo treatment . the semi - inclusive dis measurements are also consistent with a very small strange quark polarization , @xmath296 . the hermes data @xcite give a slightly positive value when integrated between @xmath499 and 0.6 , while the compass data @xcite are consistent with zero . this is in contrast with the analysis of inclusive dis data on protons , deuteron and @xmath221he , which suggests a small negative value . the extractions of @xmath296 from the inclusive and semi - inlcusive data are somewhat dependent on the assumptions of su(3 ) flavor symmetry for the axial charges of hyperons , and on the specific fragmentation functions used in the semi - inclusive analysis . future measurements of kaon fragmentation functions @xmath500 should improve the accuracy of the flavor decomposition of the proton s helicity distributions . a complementary method , discussed by bourrely and soffer @xcite , of constraining the @xmath457 and @xmath458 distributions , which does not depend on knowledge of fragmentation functions , is through longitudinal polarization asymmetries in inclusive @xmath57 boson production from @xmath134 scattering with one proton polarized and the other unpolarized , @xmath501 . the asymmetry for @xmath139 production , for example , can then be written at lo as @xcite @xmath502 where @xmath503 is the scattering angle of the electron in the partonic center - of - mass frame , while the corresponding @xmath137 asymmetry at lo is @xmath504 at large negative rapidity , where @xmath299 , the @xmath139 asymmetry then directly probes the @xmath457 pdf , @xmath505 , while at large positive rapidity one has @xmath132 , and the asymmetry is sensitive to the @xmath282 pdf at large @xmath2 , @xmath506 . for @xmath137 production , the term proportional to @xmath507 will be similarly enhanced at small @xmath299 ; however , the angular factor will produce a suppression of this contribution at backward angles relative to the @xmath508 term , so that both terms will give competing contributions . the first data from the star experiment ( run 9 ) at rhic @xcite are shown in fig . [ fig : pol_sea ] for zero electron rapidity , and are generally in agreement with nlo calculations using existing pdf parametrizations . data from run 12 at rhic have subsequently been taken at both forward and backward rapidities @xcite , and will provide important constraints on the antiquark polarization . information on the polarized gluon distribution @xmath152 can be obtained in several complementary ways . firstly , from the @xmath40 evolution equations , the evolution of the quark singlet contribution to the @xmath90 structure function mixes with the gluon contribution , so that in principle the study of scaling violations in @xmath90 can constrain @xmath152 . in practice , however , the @xmath40 range and precision of the available inclusive dis data are not sufficient to provide meaningful constraints , and dis - only fits of spin - dependent pdfs effectively leave @xmath152 undetermined . values extracted at lo from charm and high-@xmath20 hadron production data from compass , smc and hermes . ( figure from ref . copyright ( 2013 ) by the american physical society . ) _ ( right ) _ longitudinal double - spin asymmetry @xmath509 for inclusive jet production , as a function of the jet transverse momentum , in polarized @xmath510 scattering from the star experiment at rhic @xcite , compared with calculations based on several pdf parametrizations . ( figure from ref . @xcite.),title="fig:",width=321 ] values extracted at lo from charm and high-@xmath20 hadron production data from compass , smc and hermes . ( figure from ref . copyright ( 2013 ) by the american physical society . ) _ ( right ) _ longitudinal double - spin asymmetry @xmath509 for inclusive jet production , as a function of the jet transverse momentum , in polarized @xmath510 scattering from the star experiment at rhic @xcite , compared with calculations based on several pdf parametrizations . ( figure from ref . @xcite.),title="fig:",width=321 ] more direct information on gluon polarization has come from the production of charm mesons and high-@xmath20 hadrons in semi - inclusive dis , through the process of photon gluon fusion ( see sec . [ ssec : hqresults ] ) . data on hadron production from hermes , smc and compass , and open charm data from compass have been analyzed over a range of @xmath2 values , averaging from @xmath511 to @xmath512 , and @xmath40 between @xmath513 gev@xmath170 and 13 gev@xmath170 . most of the analyses have been performed at lo , finding values of @xmath462 consistent with zero , albeit with large uncertainties , as illustrated in fig . [ fig : pol_g ] . the recent compass experiment @xcite was the first to be analyzed at nlo , yielding @xmath514 over the range @xmath515 with an average @xmath516 . an alternative reaction that can be used to constrain @xmath152 is the inclusive production of jets or neutral pions in polarized proton proton scattering . the longitudinal double - spin asymmetry for this process , @xmath517 where @xmath518 is the differential cross section for longitudinally polarized protons with equal ( opposite ) helicities , is then sensitive to the polarized quark gluon scattering cross section at the parton level . following an early measurement of @xmath509 for multi-@xmath519 pairs from two - jet type events by the fermilab e581/704 collaboration @xcite , which found a small @xmath462 at @xmath520 , most of the experimental effort at measuring @xmath152 in @xmath134 scattering has been by the phenix and star collaborations at rhic . the star experiment measures inclusive jet production in @xmath521 at an invariant center of mass energy @xmath302 gev . the @xmath509 data from the 2006 and 2009 runs @xcite shown in fig . [ fig : pol_g ] provide the clearest evidence to date for a small but nonzero @xmath152 . a nonzero asymmetry is also observed by star in preliminary data on dijet events at mid - rapidity @xcite , which may provide even better constraints on the shape of @xmath152 . the phenix data on inclusive neutral pion production , @xmath522 , from 2005 to 2009 suggest a very small asymmetry up to @xmath523 gev @xcite . taken together , the inclusive rhic data from several different channels suggest a small but non - vanishing gluon polarization in the proton . though not as large as had been proposed in some early explanations of the small contribution of quarks to the proton spin , the precise role played by @xmath152 in the proton spin budget will be intriguing to understand from future measurements . in this topical review we have presented a summary of the current knowledge of spin - averaged and ( longitudinal ) spin - dependent pdfs . while great progress has been made in recent years , there are still regions of bjorken-@xmath2 where the pdf uncertainty bands are uncomfortably large . this is primarily due to either lack of data in kinematic regions that would otherwise provide constraints on the pdfs , or to the particular pdf being smaller in that region than those of other flavors , or both . one might naively conclude that if existing data do not constrain particular pdfs in certain domains , then it may not be too important to seek ways to constrain them there . however , this viewpoint overlooks two important points : ( i ) there is intrinsic interest in the behavior of the different pdfs and relations to the properties of the parent hadrons , and ( ii ) it is possible that new phenomena may occur in extended kinematic regions which would require precise knowledge of the pdfs there . as an example , suppose one were searching for a massive state produced at large rapidity in hadron hadron collisions . this would typically require one parton to have a large value of @xmath2 and the other a small value . the errors at large @xmath2 may well dominate the uncertainty in the resulting prediction . the key here is to measure standard model processes in expanded kinematic regions and include these processes in the global fits . the resulting reduction in the errors can then be used to improve searches for new phenomena in these regions . moreover , if one reduces the errors at large values of @xmath2 , this will necessarily lead to reductions in the errors at intermediate @xmath2 . since deuterium targets in dis are a prime source of information on the @xmath142 quark pdf at large values of @xmath2 , nuclear corrections are a significant source of uncertainty in its determination . several planned experiments at the 12 gev energy upgraded jefferson lab @xcite will use techniques that reduce the effects of nuclear corrections on the extraction of @xmath145 at large @xmath2 . constraints on the @xmath145 ratio in the proton can also be obtained , free of nuclear effects , through the measurement of @xmath57 charge asymmetries in @xmath134 collisions at very high rapidity @xcite , although this would require reconstruction of the @xmath57 distributions themselves from the decay lepton distributions and , unlike in @xmath524 collisions at the tevatron , would require precise knowledge of the @xmath272 and @xmath273 distributions @xcite . another important area of investigation is the flavor separation of the pdfs . for example , currently our knowledge of the @xmath254 and @xmath255 pdfs comes primarily from one neutrino experiment which used an iron target @xcite , thus requiring the use of model - dependent nuclear corrections . other sources , for example weak vector boson production at the lhc @xcite , can provide additional constraints , especially as the statistics of the measurements improve . in a complementary effort , the hermes collaboration has measured @xcite @xmath525 production in semi - inclusive dis and used these data to place constraints on the strange quark pdf . while an interesting analysis , it does depend on knowledge of the @xmath525 fragmentation functions , and a recent , more complete analysis @xcite of the hermes data suggests a somewhat smaller @xmath254 pdf than that extracted in ref . better knowledge of the fragmentation functions would clearly help in the interpretation of the semi - inclusive dis measurements . ideally , a high statistics neutrino experiment on _ hydrogen _ could further provide knowledge of both the @xmath254 and @xmath255 pdfs through the measurement of charm production ( using the @xmath526 and @xmath527 subprocesses ) without having to make model - dependent nuclear corrections . note that such an experiment could also yield information on the @xmath145 ratio at high values of @xmath2 . information on the su(2 ) flavor asymmetry of the proton sea will be provided by experiment e906/seaquest @xcite at fermilab , which is currently taking data on muon pair production using a 120 gev proton beam on a variety of targets . it is designed to improve our knowledge of @xmath528 over the @xmath2 range out to @xmath529 , whereas the currently available data run out of statistics at @xmath530 . the experiment will further enable a study of the effects of nuclear binding on the sea quark pdfs . there is also the possibility of follow - on measurements of lepton pair production at the nuclear and particle physics facility at j - parc @xcite in japan . the question of charge symmetry violation in pdfs can be addressed in future through study of drell - yan lepton pair production in @xmath531 scattering from the deuteron , which could be measured at the fermilab main injector or with the compass experiment at cern @xcite , or via measurement of @xmath532 electroproduction ratios in semi - inclusive dis from isoscalar targets @xcite . another avenue to explore will be parity - violating dis on a deuterium target @xcite , for which the parity - violating asymmetry arising from @xmath533 interference is expected to be independent of hadronic structure at large @xmath2 in the absence of csv effects @xcite . beyond searches for csv effects , a program of parity - violating dis studies at the 12 gev jefferson lab @xcite could open up a new window on the vector @xmath534 interference structure functions , as well as provide glimpses into the parity - odd @xmath535 structure function . for heavier quarks , the issue of intrinsic charm will require measuring the charm structure function at large values of @xmath2 , which may be possible at a future electron - ion collider facility @xcite . the semi - inclusive production of prompt photons and charmed jets in @xmath134 collisions at the lhc , @xmath536 , may produce an enhancement at large transverse momenta from intrinsic charm @xcite . measurements of jet , dijet , isolated @xmath519 , and @xmath519 + jet production at the lhc will all play a role in helping to further constrain the gluon pdf . of particular interest would be @xmath519 + jet production where both the @xmath519 and the jet have equal and opposite large rapidities . the dominant @xmath274 subprocess would thus require both a high-@xmath2 gluon and a high-@xmath2 quark . furthermore , continued progress on resummation , especially for the fragmentation contribution at fixed target energies may provide additional information on the gluon at high values of @xmath2 . in the polarized sector , several experiments planned at jefferson lab will measure the longitudinal polarization asymmetries for hydrogen , deuterium and @xmath221he to unprecedented large values of @xmath2 @xcite . these will constrain the spin - dependent valence quark pdfs , especially the @xmath282 distribution , to values of @xmath537 . this should help answer long - standing questions about the behavior of the @xmath538 ratios as @xmath165 , and reveal insights into the role of quark orbital angular momentum in the nucleon . the flavor asymmetry of the polarized light quark sea , @xmath460 , will be further probed via @xmath70 boson production in polarized @xmath134 scattering at rhic at @xmath539 gev . both the star and phenix experiments anticipate significantly improved statistics and kinematic coverage in their 2012 and 2013 runs @xcite . the strange quark polarization , despite appearing to be smaller than initially surmised from early polarized dis experiments , is neverthess important to pin down , as it allows a portal to nonperturbative qcd effects in the nucleon . improved constraints on kaon fragmentation functions should lead to more reliable determinations of @xmath296 in semi - inclusive dis . finally , the polarized gluon distribution is being actively investigated in semi - inclusive deep - inelastic hadron production at compass , and through double - spin asymmetries for pion and jet production in polarized @xmath134 collisions at rhic . preliminary results from star on jet and dijet cross sections hint at nonzero values for @xmath152 , which will need to be confirmed by further measurements at small @xmath2 . access to @xmath152 below @xmath540 will be possible with future running at @xmath541 gev and at large forward rapidity @xcite , although reaching values down to @xmath542 will only be possible with an electron - ion collider @xcite . on the theory front , continued progress on nnlo calculations will eventually result in global fits where all the processes are treated consistently at nnlo , for both unpolarized and polarized observables . the increased precision and decreased scale dependence will help to reduce the uncertainty bands on the pdfs . the parametrization dependence of global fits will continue to be explored , allowing for greater flexibility and reducing bias introduced with the use of specific forms , such as for the @xmath145 ratio at large @xmath2 . improvements in electroweak radiative corrections are being sought , incorporating , for example , pdfs of a photon with @xmath543 @xmath40 evolution . progress on computing moments of pdfs in lattice qcd is also anticipated , with more reliable simulations performed on larger lattice volumes and at quark masses near the physical limit . a new approach @xcite to calculating the @xmath2 dependence of the pdfs directly in the infinite momentum frame may provide complementary constraints on pdfs that are difficult to access experimentally . we thank a. accardi and k. griffioen for helpful comments and a careful reading of the manuscript . the work of p.j .- d . and was supported by doe contract no . de - ac05 - 06or23177 , under which jefferson science associates , llc operates jefferson lab . the work of j.f.o . was supported by doe contract no . de - fg02 - 13er41942 . 10 a. accardi , _ large - x connections of nuclear and high - energy physics _ , to appear in mod . phys . a ; pos ( confinement x ) 227 , + http://pos.sissa.it/archive/conferences/171/227/confinement%20x_227.pdf . jefferson lab experiment e12 - 06 - 109 , _ the longitudinal spin structure of the nucleon _ , d. crabb _ et al . _ , spokespersons . jefferson lab experiment e12 - 06 - 110 , _ measurement of the neutron spin asymmetry @xmath463 in the valence quark region using an 11 gev beam in hall c _ , j .- p . chen _ et al . _ , spokespersons . jefferson lab experiment e12 - 06 - 122 , _ measurement of the neutron asymmetry @xmath463 in the valence quark region using 8.8 and 6.6 gev beam energies and the bigbite spectrometer in hall a _ , b. wojtsekhowski _ et al . _ , spokespersons . a. accardi _ et al . _ , _ electron ion collider : the next qcd frontier - understanding the glue that binds us all _ , bnl-98815 - 2012-ja , jlab - phy-12 - 1652 , arxiv:1212.1701 [ nucl - ex ] . jefferson lab experiment e12 - 10 - 102 [ bonus12 ] , _ the structure of the free neutron at large @xmath2-bjorken _ , s. bltmann , m. e. christy , h. fenker , k. griffioen , c. e. keppel , s. kuhn and w. melnitchouk , spokespersons . jefferson lab experiment e12 - 10 - 103 [ marathon ] , _ measurement of the @xmath544 , @xmath145 ratios and @xmath545 emc effect in dis off the tritium and helium mirror nuclei _ , g. g. petratos , j. gomez , r. j. holt and r. d. ransome , spokespersons . fermilab e906 experiment , _ drell - yan measurements of nucleon and nuclear structure with the fermilab main injector _ , d. f. geesaman and p. e. reimer , spokespersons ; http://www.phy.anl.gov/mep/seaquest/index.html . a. airapetian _ et al . _ [ hermes collaboration ] , in print ; h. jackson , talk given at the _ xxi international workshop on deep inelastic scattering and related subjects ( dis2013 ) _ , marseilles , france , april 22 - 26 , 2013 . jefferson lab experiment e12 - 09 - 002 , _ precise measurement of @xmath532 ratios semi - inclusive dis : charge symmetry violating quark distributions _ , k. hafidi _ et al . _ , spokespersons . c. y. prescott _ et al . _ , b * 77 * , 347 ( 1978 ) .

recent years have seen a crescendo of observational and theoretical interest in the population of nearby m dwarfs . these stars , which make up the majority of the stars in the solar neighborhood ( e.g. * ? ? ? * ) , have historically been difficult to observe due to their optical faintness . during the past two decades , great strides have been made in the refinement of techniques for identifying nearby m dwarfs . multiple groups have used parallaxes ( e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) , photometry , and proper motions to identify thousands of m dwarfs in the solar neighborhood @xcite and throughout the galaxy @xcite . in parallel , tools for observationally characterizing these stars have improved dramatically @xcite . even so , the precise measurement of their stellar parameters remains challenging . precise stellar radii and masses have only been measured for a small set of targets through interferometry and the study of eclipsing binaries ( e.g. * ? ? ? * ; * ? ? ? the measurement of m dwarf stellar compositions has also presented a formidable challenge , as the molecular species in their atmospheres mask the continuum , typically rendering the direct spectroscopic modeling techniques intractable ( although progress is being made , e.g. @xcite ) . instead of direct modeling , the detailed analysis of atomic line and molecular absorption in m dwarf spectra has also been demonstrated @xcite . rather than depending solely on the m dwarf spectra to derive abundance information , several groups have instead relied on abundance measurements of associated stars . these calibrator systems typically contain an fgk dwarf in a common - proper - motion system with an m dwarf , separated by several arcseconds or more . the composition of the higher - mass star is known or readily determined using established spectroscopic techniques ( as deployed for m subdwarfs in * ? ? ? the m dwarf is then assumed to have the same composition , and their photometric and spectroscopic properties are used to construct empirical models that can then be applied to single stars . this empirical calibration strategy was first implemented for optical photometry , taking advantage of the strong and weak [ fe / h ] dependence for m dwarf @xmath4 and @xmath5-band absorption , respectively . multiple groups have developed similar photometric calibrations based on absolute stellar magnitudes and colors @xcite , and @xcite have further leveraged these photometric [ fe / h ] calibrations into more precise optical spectroscopic calibrations . this empirical calibration technique has also been implemented for near - infrared ( nir ) spectra , taking advantage of the relative brightness of m dwarfs in the nir . the first of these was the @xmath5-band calibration developed by @xcite , which leveraged the @xmath5-band doublet and triplet , along with h@xmath6o absorption indices to account for t@xmath3 effects . this technique was expanded by ( * ? ? ? * , t12 ) , refined in @xcite and @xcite ( hereafter n14 ) , and further generalized in @xcite ( hereafter m13a ) . presently , the nir calibration with the largest number and diversity of calibrators is that of m13a for early - mid m dwarfs , and @xcite ( hereafter m14 ) for late - type m dwarfs ; these calibrations provide [ fe / h ] and [ m / h ] , are well - calibrated across a wide range of [ fe / h ] ( @xmath7 [ fe / h ] @xmath8 ) and spectral type ( late k to m8 ) , and have good precision ( @xmath9 dex in [ fe / h ] and [ m / h ] ) . in addition to the empirical techniques for measuring stellar composition , various groups have developed similar techniques for the measurement of spectral type , effective temperature , stellar mass , and stellar radius , based on spectral type standards , stars with interferometrically - determined parameters , or stars with parameters measured through other means . notably , techniques applicable to high resolution optical spectra from harps @xcite and keck @xcite have achieved good precision for the m dwarfs observed in these programs . techniques based on lower - resolution ( @xmath2 ) nir spectra , using t@xmath3-sensitive indices @xcite ( hereafter m13b ) and atomic absorption features @xcite have been also been developed . the use of broadband photometry has also enabled precise measurements of the temperatures and radii of these stars @xcite . the resulting suite of techniques has been used extensively in the characterization of _ kepler _ low - mass planet hosts @xcite . the tools and data described above , combined with advances in structural @xcite and atmospheric models , provide a fertile ground for a variety of topics related to m dwarfs , including galactic evolution ( e.g. * ? ? ? * ; * ? ? ? * ) , stellar structure ( e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) , and exoplanets . nearby m dwarfs are appealing targets for doppler radial velocity ( rv ) and transit exoplanet searches because their low masses and luminosities make planets orbiting in their habitable zones ( hz * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) more easily detectable than those around higher - mass stars . moreover , given the prevalence of m dwarfs , their planets may represent the most common exoplanetary environment in the galaxy . upcoming nir ground - based rv surveys such as the habitable - zone planet finder on the hobby - eberly telescope ( hpf , * ? ? ? * ) , carmenes @xcite at calar alto , spirou on the cfht @xcite , ird on subaru @xcite , and ilocater on lbt as well as photometric surveys like tess @xcite , plato @xcite , and mearth @xcite will specifically target planets around nearby m dwarfs . these surveys will provide unprecedented sensitivity to planets orbiting the nearby m dwarfs , whose optical faintness has made them difficult targets for many existing rv and transit surveys . despite being optimized for optically bright fgk stars , results from _ kepler _ @xcite clearly indicate that m dwarfs host a large population of small planets in the hz @xcite . the characteristics of planets around m dwarfs will be useful for testing theories of planet formation ( e.g. * ? ? ? * ) in the regime of low stellar and protoplanetary disk masses . in order to construct a well - characterized target list for hpf , we have carried out an extensive spectroscopic survey of 886 nearby m dwarfs , using the spex spectrograph at the nasa infrared telescope facility ( irtf ) . we take advantage of the nir spectroscopic techniques described above to measure the metallicities of these stars , in order to form the largest catalog yet of nearby m dwarf compositions based on spectroscopy . we also provide measurements of the t@xmath3-sensitive indices , which can be used to estimate stellar mass , radius , or luminosity . this work echoes the spirit of previous large - scale efforts to characterize our most common stellar neighbors ( e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ? * n14 ) . in section [ obs ] we describe our observations and data reduction , including target selection and cross - identification with other catalogs . in section [ specmeas ] we describe our spectral measurements and error estimates . in section [ gendisc ] we provide a brief summary and discussion of our measurements and comparisons to other catalogs . this dataset has already enabled a variety of studies of the nearby m dwarfs . t12 developed a new @xmath10-band [ fe / h ] calibration for m dwarfs , extending established techniques and providing a valuable comparison for the development of improved calibrations ( m13a ) and a starting point for work with higher resolution spectra ( e.g. sdss - iii apogee or igrins , * ? ? ? * ; * ? ? ? @xcite employed the t12 [ fe / h ] measurement technique to characterize the low - mass eclipsing binary cm draconis ( cm dra * ? ? ? * ) , helping to remove the [ fe / h]-related degeneracy and to clarify the disagreement between models and observations of this system . @xcite again applied the t12 [ fe / h ] measurement technique to confirm the first low - mass members of the nearby coma berenices cluster @xcite . finally , @xcite leveraged m dwarf activity catalogs ( e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) and improvements in parallax measurements of the nearby m dwarfs ( e.g. * ? ? ? * and the research consortium on nearby stars , recons ) to develop new spectroscopic indices for m dwarf luminosity , surface gravity , and possibly @xmath11-enrichment . over a period of two years , beginning in 2011 may and concluding in 2013 august , we observed 886 m dwarfs with the spex nir spectrograph on the nasa - irtf @xcite . this set of observations includes 21 stars observed multiple times . we drew m dwarf targets primarily from the ( * ? ? ? * lspm - n ) and ( * ? ? ? * lg11 ) proper - motion based catalogs of nearby m dwarfs , and from published lists of m dwarf planet hosts and known wide binaries ( drawn from * ? ? ? * ) that serve as abundance calibration stars . the majority of target stars ( 829 ) were taken from lg11 , the remainder being the aforementioned binary calibrator stars , planet hosts , or that were targeted in a contemporaneous high - resolution survey @xcite . for stars from lg11 and lspm - n , we prioritized targets with redder optical - to - infrared color ( @xmath12 ) , a good indication of mid - late m spectral subtype ( m4 or later ) . the resulting compilation of targets included both high - priority targets ( calibrator stars , planet hosts , late - type m dwarfs ) and low - priority targets ( early type m dwarfs ) . from this super - set of potential targets we constructed our nightly observing plans , and during each night we actively added and removed targets to account for changes in observing conditions . our primary goal was to ensure very high signal - to - noise ( snr @xmath13 per pixel ) for high - priority targets , and our secondary goal was to observe as many m dwarfs as possible with reasonable snr ( @xmath14 ) . the typical resulting observing pattern was a series of groups ( in position and time ) which contained a few faint late - type ( m4 or later ) m dwarfs and several brighter early ( earlier than m4 ) m dwarfs . each group was sufficiently localized that a single nearby standard star ( within one hour of time and 0.1 @xmath15 of target star ) could be used for telluric correction , and the same arc lamp exposure could be used for wavelength calibration . this strategy enabled a large number of m dwarfs to be observed , with minimal time losses from large slews , wavelength calibrations , and standard star observations . we obtained uniform data for all observations , operating spex in the short cross - dispersed ( sxd ) mode with the @xmath16 slit , which produces @xmath17 spectra from 0.8 - 2.4 @xmath18 . as described in t12 , we extracted these spectra with the facility - provided spextool package @xcite . we telluric - corrected using the ` xtellcor ` program @xcite and observations of an a0v star ( or similar ) , at an airmass and observation time within 0.1 ( @xmath15 ) and 1 hour of the target , respectively . the ` xtellcor ` program also flux - calibrates the spectrum , using the optical magnitudes ( from simbad ) of the standard star to estimate its flux as a function of wavelength . @xcite estimate that the flux calibration using this method agrees to within a few percent of fluxes derived using broadband magnitudes . a typical example spectrum is shown in figure [ specex ] . all observations presented here took place prior to the spex detector upgrade , which occurred in the spring of 2014 . we cross - identified our targets with several published catalogs , in order to provide supplementary data for our target stars and to facilitate the use of our catalog , each target has a 2mass @xcite identifier , and we use the 2mass j2000 coordinates in our final catalog . the 2mass i d also provides the most convenient uniform identifier for all our targets . we include in our catalog the 2mass @xmath19 photometry and the respective quality flags for each star . we cross - identified our targets with the ucac4 @xcite catalog as well . ucac4 provides optical magnitudes based on the aavso photometric all - sky survey ( apass ) . optical magnitudes are useful for m dwarfs as their optical - to - infrared colors are excellent indicators of spectral type and t@xmath3 ; we include ucac4 aperture ( @xmath20)-magnitudes ( for 874 stars ) and @xmath4-magnitudes ( for 732 stars ) as indications of their optical brightness . for the majority of our targets , we include proper motions as measured by lg11 or lspm - n . for targets without proper motions in these catalogs , we include measurements from ucac4 or the ppmxl catalog @xcite . we also provide parallax measurements for a nearly half of our targets ( 427 ) . many of these parallax measurements are drawn from the recons database @xcite and the newly - published mearth parallax catalog @xcite . we also found parallaxes in several other catalogs @xcite . we describe here the various spectral measurements we performed on the m dwarf spectra . the nir spectra of m dwarfs are dominated by the broad absorption of h@xmath6o , the strength of which is known to correlate strongly with t@xmath3 and spectral type ( e.g. * ? ? ? . using @xmath2 nir spectra like ours , several groups have developed h@xmath21o - based and other indices that are well - correlated with spectral type and t@xmath3 ( * ? ? ? * ; * ? ? ? * ; * ? ? ? * t12,m13a , m13b , n14 , m14 ) . calibrations of these indices provide an efficient and simple method for measuring spectral type and temperature for the m dwarfs in our catalog . the relatively large widths of the wavelength ranges used in these measurements makes them insensitive to small errors in the rv measurement . for the m dwarfs in our catalog , we measured the t@xmath3-sensitive indices used in m13b and n14 @xcite ( see table [ h2oparams ] ) . lccc + @xcite @xmath22 index & @xmath23 & @xmath24 & @xmath25 + @xcite @xmath10 index & @xmath26 & @xmath27 & @xmath28 + @xcite @xmath29 index & @xmath30 & @xmath31 & @xmath32 + @xcite h@xmath6o-@xmath22 & @xmath33 & @xmath34 & @xmath35 + @xcite h@xmath6o-@xmath10 & @xmath36 & @xmath37 & @xmath38 + @xcite h@xmath6o-@xmath29 & @xmath39 & @xmath40 & @xmath41 to measure spectral type for the m dwarfs in our catalog , we applied the nir calibration developed in n14 , which is based on the h@xmath21o - k2 index developed in @xcite . this calibration yields nir m sub - types " on a uniform scale for our stars . the precise relationship between this nir m sub - type and the established palomar / michigan state university ( pmsu ) scale @xcite is considered in @xcite and n14 , and is known to be sensitive to [ fe / h ] for at least the early m dwarfs . we also include in our catalog the measured pmsu spectral types from the n14 pmsu spectral type calibration over the range in which it is valid ( m1-m4 ) . to check the consistency of our nir spectral type measurements , we considered the overlap between our sample and that of n14 , which includes 152 m dwarfs . we found no significant offset between the respective measures of spectral type , and a scatter of half a subtype , consistent with the scatter expected from n14 . this nir spectral type measurement was necessarily the first in our chain of measurements , as it defined the template for our measurement of rv . the t@xmath3-sensitive features are sufficiently broad that they are insensitive to rv shifts in our @xmath2 spectra . nonetheless , after measuring the rv we re - measured the spectral type in the rest frame of the star , and this is the value we report in our catalog . we measured the rv of each order of each spectrum by fitting a gaussian to the cross - correlation ( in log-@xmath42 ) of the observed spectrum with a spectral template of similar spectral type . we used three spectral templates from the irtf cool stars library @xcite : for stars m4 and earlier we used the m1.5v star hd 36395 , for stars m5 to m6 we used the m5v star gl 51 , and for stars later than m6 we used the m9v star lhs 2065 . for each observation , we report the weighted ( by snr ) average rv from all orders , and this is value we used to shift all spectra to a common frame for further analysis . to verify our rv measurements , we compared our ( barycentric - corrected ) rv measurements to those measured in @xcite and n14 . for the 33 stars in common with @xcite , we find a median offset of @xmath43 km s@xmath44 with a standard deviation of @xmath45 km s@xmath44 . for 152 stars in common with n14 , we find a median offset of @xmath46 km s@xmath44 with a standard deviation of 13.2 km s@xmath44 . figure [ rv_validation ] shows the differences in rv between this work and the literature sources . among our common targets with the n14 study , there were seven with differences greater than 40 km s@xmath44 . two of these ( 2mass j07362513 + 0704431/gj 3454 and 2mass j17074083 + 0722066/gj 1210 ) are known visual binaries with sub - arcsecond separation @xcite . given the consistency of the rv measurements of the majority of the common targets , we suspect that these highly discrepant targets are either rv - variable or binary stars . we also measure a large rv for one star ( 2mass j06052936 + 6049231/lhs 1817 , @xmath47 km s@xmath44 ) with similar magnitude but the opposite sign as n14 and @xcite . that the systematic offset from n14 and @xcite would be similar is unsurprising , as n14 implement a @xmath48 km s@xmath44 correction to match the rv system of @xcite . the origin of the @xmath49 km s@xmath44 offset for our data is unclear ; since we rely solely on the arc lamp wavelength calibrations , we are subject to different and possibly worse instrumental systematics than n14 , who use telluric spectra to refine their wavelength calibrations . a systematic issue with the wavelength calibration of approximately 8 km s@xmath44 is also supported by the magnitudes of the inter - order and rv template differences , as discussed in section [ uncertainties ] . in order to obtain barycentric rvs in the same system as @xcite and n14 , the median offset above ( @xmath507.6 km s@xmath44 ) should be added to the rvs in our catalog . . [ rv_validation ] ] due to the difficulty of modeling m dwarf atmospheres and spectra , model - dependent measurements of t@xmath3 can systematically differ by 100 k or more ( * ? ? ? * and references therein ) . interferometric t@xmath3 measurements have therefore been crucial in defining the m dwarf t@xmath3 scale , and in parsing the relations between the fundamental parameters of m dwarfs ( * ? ? ? * and references therein ) . however , there are few targets with spectral type m5 or later that are bright enough or have large enough angular diameters to support precise interferometric radius measurements . to enable stellar radius measurements without interferometry , multiple groups have constructed empirical calibrations based on photometry @xcite , spectral indices ( m13b ) , or individual absorption feature strengths @xcite . we measured t@xmath3 for the m dwarfs in our catalog using the h@xmath21o indices as discussed in section [ section : bands ] . we used the @xmath51 t@xmath3 calibrations developed in m13b over their well - calibrated range ( e.g. @xmath52 ) , which corresponds to @xmath53 k. m13b provide empirical relations for stellar radius , luminosity , and mass as a function of t@xmath3 , calibrated on a set of stars with interferometrically measured parameters . we favor the use of the @xmath5-band index and calibrations due to the robustness of this index to small changes in the telluric correction ( as discussed in section [ section : bands ] ) . we also measured t@xmath3 , stellar radius , and stellar luminosity using the @xmath10-band atomic feature strengths studied in @xcite . we measure the pseudo - equivalent widths ( ews ) of these features using the continuum and feature definitions in their table 1 . they construct relations using the strengths and ratios of mg and al lines in m dwarfs , empirically calibrated on stars with interferometrically measured parameters . these relations use a similar calibration set to that of m13b , so the range of validity is similar : @xmath54 k. we measured [ fe / h ] and [ m / h ] using the empirical spectroscopic calibrations from t12 , m13a , m14 , and n14 . these calibrations were all developed using data from the same spectrograph ( irtf - spex ) in the same settings ( sxd ) as ours , so their recipes can be applied directly to our data . m13a present separate calibrations for features in the @xmath51 bands that yield both [ fe / h ] and [ m / h ] for early - mid m dwarfs ( m1-m5 ) . they derived these calibrations from a blind search of the most [ fe / h ] and [ m / h]-sensitive regions of the spectra in 112 stars with known [ fe / h ] and [ m / h ] , and are based on the ews of a small number ( four or fewer ) of spectral regions in each case . the calibrations are valid for k5-m5 dwarfs and over approximately @xmath55[fe / h ] @xmath56 . we report the results of these calibrations separately for each target in our catalog . for a similar range of targets in spectral type and [ fe / h ] , we also report the @xmath5-band [ fe / h ] calibration developed in n14 , which depends only on the ew of the @xmath5-band doublet . n14 calibrated this relation using 36 m dwarfs with known [ fe / h ] , and it is similar in form to the relations developed in @xcite and @xcite . it is valid for m1-m5 dwarfs and for a range of @xmath57[fe / h]@xmath58 . for the late m dwarfs ( m5 or later ) , the only applicable published [ fe / h ] calibration is that of ( m14 ) . the m14 calibration is based on two strong features in the @xmath5-band , the triplet and the doublet , which were also used in the calibration of @xcite and @xcite for early - type m dwarfs . this calibration is based on 44 m dwarfs with known metallicities . it is an extension of the binary calibration technique described above , which extends the calibration set by including some early+late m dwarf pairs where the earlier - type m dwarf [ fe / h ] has itself been measured using empirical calibrations . this late - type calibration is valid for m4.5-m9.5 dwarfs and for a range of @xmath59[fe / h]@xmath60 . the summary of the parameters of this [ fe / h ] calibration and those above is shown in table [ calparams ] . for general comparison with other catalogs , and for cases below in which it is useful to have a single [ fe / h ] measurement for a given star , we elect to use the @xmath5-band calibrations of m13a for m1-m5 ( see section [ prefmeas ] ) , along with the output of m14 for m5 and later . we also report the [ m / h ] values where available . for the purposes of presenting and vetting our catalog , we primarily consider [ fe / h ] as this is the specific quantity addressed by the bulk of work on the topic of m dwarf compositions . lccc + @xcite [ fe / h]@xmath61 & k5.0@xmath62m5.0 & @xmath63}<+0.56 $ ] & 0.11 dex + @xcite [ m / h]@xmath61 & k5.0@xmath62m5.0 & @xmath64}<+0.50 $ ] & 0.10 dex + @xcite [ fe / h]@xmath61 & m4.5@xmath62m9.5 & @xmath65}<+0.56 $ ] & 0.07 dex + @xcite [ fe / h ] & m1.0@xmath62m5.0 & @xmath66}<+0.35 $ ] & 0.12 dex + @xcite nir spt & m1.0@xmath62m9.0 & & 0.5 subtype our spectra extend to approximately 0.8 @xmath1 m on the blue end , and so contain the 8200 doublet ( hereafter ) and the 8600 triplet ( hereafter ) , strong features that have shown promise as indicators of activity ( e.g. * ? ? ? * ; * ? ? ? * ) and surface gravity @xcite . our measurements of these features for a subset of the observed m dwarfs showed a strong relationship with the t@xmath3 and [ fe / h ] measures in these stars , and can be used to derive information about their radii , luminosities , and possibly @xmath11-enrichment . these measurements are discussed in @xcite ; we repeat the salient details here . we measured the ews of the and features ( ew@xmath67 , ew@xmath68 ) for each of our targets . for , we chose line and pseudo - continuum regions similar to those used in @xcite , @xcite , and @xcite . for , we defined feature regions for each of the three lines , as well as pseudo - continuum regions blueward and redward of each line . in each case , we fit a line to the pseudo - continuum regions in order to define the pseudo - continuum , and calculated the feature ew relative to this line . for ew@xmath67 and ew@xmath68 we follow the convention that positive ew corresponds to line absorption . we caution that the and features are embedded in regions of strong tio and other absorption ( as can be seen in e.g. @xcite ) , so the ews we measure are likely strongly sensitive to effects of the line species themselves as well as to the species that define the local pseudo - continuum . moreover the snr in the relevant orders of the spex spectra ( orders seven and eight ) is typically only a third of that in the @xmath51 bands . accordingly , these feature strengths should be carefully vetted before they are used in any application . the regions we defined for this analysis are listed in table [ canaparams ] and are shown on example spectra in figure [ emission ] . figure [ emission ] also shows the three ( possibly young ) m dwarfs in our sample for which the lines are in emission . lcccc + & 8192 & 34.0 & 8149 - 8169 & 8230 - 8270 + @xmath69 & 8500 & 12 & 8456 - 8488 & 8512 - 8536 + @xmath6 & 8544 & 12 & 8512 - 8536 & 8553 - 8573 + @xmath70 & 8664 & 12 & 8635 - 8652 & 8672 - 8688 in order to estimate our measurement uncertainties ( for all quantities other than rv ) , we rely on a set of 21 m dwarfs with multiple observations . we calculate the rms of the epoch - to - epoch differences in each measurement to obtain an observational estimate of our internal uncertainty . these values are shown for [ fe / h ] , [ m / h ] , m spectral sub - type , , and in figure [ errcomp ] , for the t@xmath3-sensitive indices in [ h2oerrcomp ] , and for the @xmath10-band features of @xcite in figure [ nferrcomp ] . for the and features , we exclude three pairs with very low snr ( @xmath71 ) . the observed rms differences naturally include effects that arise from variable telluric conditions and by - hand choices in the telluric correction pipeline . the resultant variable quality of the telluric correction can induce correlated noise which dominates our errors at the high snrs we achieve ( n14 ) . we note that the t@xmath3-sensitive indices may be affected by the broad characteristics of the telluric correction , including the flux calibration ( which uses simbad @xmath72 magnitudes , section [ datared ] ) and the construction of the telluric spectrum . by repeating the telluric correction with varying @xmath72 magnitudes for a small set of stars , we found that the @xmath22-band indices were generally much more sensitive to differences in @xmath72 magnitude and small differences in the by - hand components of the telluric correction . we favor the use of the @xmath5-band indices ( h@xmath6o - k2 or the m13b @xmath5-index ) , which are stable to a few percent or better with @xmath72 variations of 0.5 mag . the t@xmath3-sensitive indices can be used to estimate stellar t@xmath3 ( also mass and bolometric luminosity ) as described in m13b , who present uncertainties of @xmath73 k for these relations . our adopted measurement uncertainties for their t@xmath3-sensitive indices ( @xmath74 ) correspond to median uncertainties of 110 k ( @xmath22 ) , 170 k ( @xmath10 ) , and 78 k ( @xmath5 ) when propagated through the t@xmath3-index relations from m13b . the @xmath10-band atomic feature strengths presented in @xcite can similarly be used in empirical calibrations to derive stellar t@xmath3 ( residual scatter = 73 k ) , luminosity ( @xmath75 ) , and radius ( @xmath76 ) . our repeated measurements of these @xmath10-band features ( figure [ nferrcomp ] ) suggests errors of @xmath77 , which corresponds to measurement errors of approximately 90 k , @xmath78 , and @xmath79 . to estimate the rv measurement uncertainty , we report the standard deviation of the rvs measured from each spectral order . this empirical measure of uncertainty is conveniently available for all targets , and avoids potential issues of rv variability which may occur using the epoch - to - epoch rms differences . the median rv spread is @xmath80 km s@xmath44 . we note that a significant component of this rv spread is contributed by systematic offsets of @xmath81 km s@xmath82 between spectral orders . we found these shifts to be variable depending on the choice of rv template spectrum , suggesting issues with the wavelength calibration , line shapes , or psf changes . we also found overall rv differences of a similar magnitude ( @xmath83 km s@xmath44 ) for each star depending on the template used , suggesting that our rv precision is dominated by these systematic issues . this also suggests that the @xmath84 km s@xmath44 offset between our rvs and those in the literature could be an artifact of the same systematic wavelength calibration issue . alternatively , it is possible to calculate measurement uncertainties based on the errors reported by the spextool pipeline , which reports a combined noise that includes photon noise , read noise , and residual background noise components . measurement uncertainties can then be estimated using monte carlo realizations of the noise . this process does not naturally include correlated noise effects like those arising from the telluric correction , although an ad - hoc technique is presented in n14 that mimics the effects of correlated noise . briefly , this procedure involves convolving the noise spectrum " with a 1.5 pixel gaussian , which is a good approximation for the actual autocorrelation function for observed spectra . this procedure effectively spreads the noise for each pixel over the neighboring pixels , mimicking the effect of the systematic correlated errors . applying this technique to our measurements yields error estimates that are consistently smaller than the observed epoch - to - epoch rms differences , so we elect to report the more conservative rms numbers as our estimated measurement uncertainties . we note that the ew@xmath67 and ew@xmath68 are thought to be sensitive to chromospheric activity @xcite , and may be variable on the timescales ( days - months ) of our repeated observations . the epoch - to - epoch rms differences may therefore be an overestimate of our measurement error for these features . we also estimate a lower limit of @xmath85 on the measurement uncertainties for these features using a 100-trial monte carlo technique based on the errors reported by the spextool pipeline . following n14 , we include an ad hoc correction to account for the effects of correlated noise . the true measurement errors for and likely lie in the range bracketed by these estimates , @xmath86 . we provide here a brief description of the spectra of the m dwarfs observed in this survey , as well as the set of measured and compiled values for each m dwarf . the spectra for each of the 886 m dwarfs are available as a ` gzipped ` archive of standard fits tables hosted at the penn state scholarsphere data hosting service . the file for each spectrum is identified by the 2mass i d of the target star . the fits headers contain the standard information output by the spextool pipeline , including observation meta - data ( instrument configuration , airmass , observing time , etc . ) and the spextool - generated file history . we append to this the measured rv shifts and barycentric correction for each spectrum , which can be used to velocity shift the spectra into a common reference frame . the spectra themselves are in the standard format described fully in the spextool documentation @xcite . briefly , each spectrum is a three - dimensional array , with the following axes : 1 . order number ( @xmath87 ) . data type ( wavelength , flux , or variance ) . wavelength is in vacuum , and is in the observer s reference frame . pixel number . table [ tablecols ] describes the columns included in our catalog . there is one line per star , including astrometry and photometry compiled from the literature , measurements of spectral features , and derived stellar parameters . the full machine - readable table is available in the online edition of the astrophysical journal supplement . llc + i d & 2mass identifier & + radeg & right ascension in decimal degrees ( j2000 ) & deg + dedeg & declination in decimal degrees ( j2000 ) & deg + jmag & 2mass @xmath22 magnitude & mag + hmag & 2mass @xmath10 magnitude & mag + ksmag & 2mass @xmath5 magnitude & mag + e_jmag & error in 2mass @xmath22 magnitude & mag + e_hmag & error in 2mass @xmath10 magnitude & mag + e_ksmag & error in 2mass @xmath5 magnitude & mag + 2mqf & 2mass quality flag & + amag & ucac4 @xmath20 magnitude & mag + e_amag & error in ucac4 @xmath20 magnitude & mag + pmra & proper motion in right ascension & mas / yr + pmdec & proper motion in declincation & mas / yr + r_pm & reference for proper motion & + plx & parallax & mas + e_plx & error in parallax & mas + r_plx & reference for parallax & + vmag & @xmath4 magnitude from ucac4 & mag + rv & radial velocity & m / s + e_rv & error in radial velocity ( scatter among spectral orders ) & m / s + bc & barycentric correction & m / s + mh2oj & h@xmath6o - j index defined by mann et al . 2013a & + ch2oh & h@xmath6o - h index defined by covey et al . 2010 & + rh2ok2 & h2o - k2 index defined by rojas - ayala et al . + mji & j index defined by mann et al . 2013b & + mhi & h index defined by mann et al . 2013b & + mki & k index defined by mann et al . 2013b & + nspt & m spectral subtype defined in newton et al . 2014 & + nspt_pmsu & pmsu spectral type defined in newton et al . + mteffj & effective temperature from j index defined in mann et al . 2013 & k + mteffh & effective temperature from h index defined in mann et al . 2013 & k + mteffk & effective temperature from k index defined in mann et al . 2013 & k + mfehj & [ fe / h ] from @xmath22-band defined in mann et al . + mfehh & [ fe / h ] from @xmath10-band defined in mann et al . + mfehk & [ fe / h ] from @xmath29-band defined in mann et al . + mmhj & [ m / h ] from @xmath22-band defined in mann et al . + mmhh & [ m / h ] from @xmath10-band defined in mann et al . + mmhk & [ m / h ] from @xmath29-band defined in mann et al . + mfehl & [ fe / h ] for late - type m dwarfs defined in mann et al . + nfeh & [ fe / h ] defined in newton et al . + t12fehh & [ fe / h ] from @xmath10-band defined in terrien et al . + ewna & equivalent width of 820 nm doublet & 0.1 nm + ewca & equivlanet width of 860 nm triplet & 0.1 nm + nmg148 & equivalent width of 1.48 @xmath1 m mg feature defined in newton et al . 2015 & 0.1 nm + nmg150 & equivalent width of 1.50 @xmath1 m mg feature defined in newton et al . 2015 & 0.1 nm + nk151 & equivalent width of 1.51 @xmath1 m k feature feature defined in newton et al . 2015 & 0.1 nm + nmg157 & equivalent width of 1.57 @xmath1 m mg feature feature defined in newton et al . 2015 & 0.1 nm + nsi158 & equivalent width of 1.58 @xmath1 m si feature feature defined in newton et al . 2015 & 0.1 nm + nco161 & equivalent width of 1.61 @xmath1 m co feature feature defined in newton et al . 2015 & 0.1 nm + nco162 & equivalent width of 1.62 @xmath1 m co feature feature defined in newton et al . 2015 & 0.1 nm + nal167a & equivalent width of 1.67 @xmath1 m al - a feature feature defined in newton et al . 2015 & 0.1 nm + nal167b & equivalent width of 1.67 @xmath1 m al - b feature feature defined in newton et al . 2015 & 0.1 nm + nmg171 & equivalent width of 1.71 @xmath1 m mg feature feature defined in newton et al . 2015 & 0.1 nm + n15teff & effective temperature from newton et al . 2015 calibration & k + n15rad & stellar radius from newton et al . 2015 calibration & solrad + n15logl & logarithm of stellar luminosity from newton et al . 2015 calibration & [ sollum ] + notes & notes & the overall distribution of our sample in nir spectral type and [ fe / h ] is shown in figure [ feh_spt ] . the [ fe / h ] distribution of our sample is centered around solar , with a median [ fe / h ] of @xmath88 dex and a scatter of @xmath89 dex . this sample is neither volume nor magnitude - limited , but is consistent with the quoted [ fe / h ] measurement uncertainty ( @xmath90 dex ) and volume - limited measurements of the fgk stars @xcite and m dwarfs @xcite in the solar neighborhood , which find a median [ fe / h ] of @xmath91 dex and a scatter of @xmath92 dex . the distribution of nir spectral types , which peaks around m4 and has a long tail toward earlier spectral types , is mostly a byproduct of our observational strategy , although the mass function of the solar neighborhood also peaks around this spectral type @xcite . distributions for all m dwarfs observed in our survey . our [ fe / h ] values are the average of the @xmath51-band calibrations from m13a for targets with spectral type m5 and earlier , and the [ fe / h ] derived from @xmath5 band features using the technique from m14 for targets with spectral type later than m5 . the hatched region shows [ fe / h ] values that exceed the calibrated range . the projection plots show a kernel density estimation for each parameter , clarifying the peaks , widths , and tails of each distribution . [ feh_spt ] ] the majority of m dwarfs in our catalog have spectral types of m5 or earlier , and so can have their [ fe / h ] measured with multiple techniques . we focus on the most recent of these calibrations ( from m13a and n14 ) , along with the @xmath10-band calibration of t12 for comparison . our measures of [ fe / h ] are compared to each other in figure [ fig_internal ] , which clearly shows systematic offsets among the various calibrations of up to @xmath93 dex . in general , the @xmath10 and @xmath22-band techniques are in agreement , while the @xmath5-band techniques show good agreement up to the saturation of the n14 calibration around 0.3 dex . to explore whether issues with the wavelength calibration could be responsible for these systematic differences , we tested the sensitivity of the various [ fe / h ] calibrations to rv shifts . we measured [ fe / h ] for each target 50 times with a normal distribution of rvs centered around the original measured rv and with a spread equivalent to our estimated systematic and random measurement errors ( typically 10 - 20 km s@xmath44 ) . among the m13a calibrations , we found that the mean scatter in the @xmath5-band calibration was best ( at @xmath94 dex ) , compared to the @xmath10-band ( @xmath95 dex ) and @xmath22-band ( @xmath96 dex ) . this suggests that the features chosen for the m13a @xmath22 and @xmath10-band calibrations are marginally more sensitive than the @xmath5-band features to small wavelength shifts , and the magnitude of the expected rv errors is not sufficient to explain the observed offsets . we found that an rv shift of @xmath14 km s@xmath44 can remove the offset between the @xmath5 and @xmath10-band measurements , although this causes the @xmath22-band measurements to diverge further . in any case , such a large shift is not justified given the agreement among our per - order rv measurements , those for different rv templates , and between our overall rv measurements and literature values . the differences among our implementations of the m13a techniques may be due to undiagnosed issues with the wavelength calibration or other unidentified differences between our reduction procedure and those of m13a ( e.g. the telluric correction ) . we present measurements from each band in our catalog , and justify below the choice of the @xmath5-band calibration of m13a as our preferred abundance calibration for m1-m5 dwarfs ( section [ prefmeas ] ) . to check our results and to probe the relationships between different measures of metallicity , we compared our values ( using the calibrations of m13a , m14 , and n14 ) to those presented by a number of recent works ( * ? ? ? * ; * ? ? ? * n14 ) for common targets . @xcite rely on the strengths of the @xmath5-band and features along with the h@xmath6o - k2 index to account of t@xmath3 ; n14 rely on the strength of the @xmath5-band feature alone ; @xcite used the @xmath4-band spectroscopic calibration presented by m13a . the comparison of our values to those from other works is shown in figure [ fehcomp ] , along with the median and scatter in each case . our application of the m13a @xmath5-band [ fe / h ] calibration shows good agreement with these other works , although we find systematically higher [ fe / h ] for those stars with the highest [ fe / h ] values compared to values reported in @xcite and ( n14 ) . this behavior may be a byproduct of the features used in each case , whose sensitivity to [ fe / h ] can saturate ( as noted in n14 ) . particularly , the na and ca @xmath5-band features used in @xcite , n14 , and m14 are known to saturate as [ fe / h ] approaches 0.5 dex . the @xmath10 and @xmath22-band calibrations are known to have larger scatter ( m13a ) , and this is borne out in our comparisons of these measurements to other works . in both cases , the [ fe / h ] we measure for stars with solar [ fe / h ] and higher is systematically @xmath77 dex lower than that reported in other works . this offset , which is not present for the @xmath5-band [ fe / h ] values , may result from structure remaining in the residuals of these calibrations as shown in e.g. figure 5 of m13a . there are several m dwarfs for which we measure later spectal type than m5 , which also have [ fe / h ] reported in other works . for these stars , the n14 and @xcite calibrations appear to perform well compared to the m14 calibration , likely due to their common reliance on the @xmath5-band feature . our application of the n14 [ fe / h ] calibration is consistent with the published [ fe / h ] measurements in their work , verifying our implementation . for targets in common with @xcite , our [ fe / h ] measurements are systematically higher , consistent with the offset for later spectral types observed in figure 18 of n14 . values we measure . each plot lists the median offset ( @xmath97 ) and standard deviation ( @xmath98 ) of the differences , although these do not capture higher - order systematic differences . [ fehcomp ] ] a similar catalog to the one presented here is that being developed for the carmenes m dwarf exoplanet survey @xcite , for which an extensive optical spectroscopic survey has been carried out @xcite . there are 276 targets in the carmenes catalog with coordinate matches in our catalog to within 5 . we plot the respective measures of composition and spectral type in figure [ ccomp ] . as an indication of abundance , @xcite measure the [ fe / h]-sensitive @xmath99 parameter ( based on tio and cah absorption in the optical , * ? ? ? * ; * ? ? ? we find that @xmath99 is weakly correlated with our measured [ fe / h ] ( kendall s @xmath100 , two - sided @xmath101 ) . this is unsurprising since @xmath99 is a coarse indicator of abundance , which has been primarily used for discriminating between dwarfs and subdwarfs . for all targets in the overlap sample , our [ fe / h ] measurements and the @xmath99 measurements of @xcite are both consistent with dwarf compositions . to measure spectral type , @xcite use a suite of spectral standard template matching and spectral index based techniques , and report spectral types to 0.5 subtypes . our spectral type estimates broadly agree with those of @xcite : our nir m subtype measurements have a median offset of 0.37 with a residual standard deviation of 0.52 , while our pmsu spectral types have a median offset of 0.12 subtypes and a residual standard deviation of 0.41 . @xcite use pmsu standards for much of their reference set , so it is expected that the pmsu measurements should show better agreement . ] in light of the systematic offsets shown in sections [ intcomp ] and [ extcomp ] , it is useful to choose and justify a preferred measure of [ fe / h ] for m1-m5 dwarfs . we choose the @xmath5-band calibration of m13a for the following reasons : * the breadth and number of calibrators exceeds those of other similar calibrations ( t12 , * ? ? ? * ; * ? ? ? * ) . * the better agreement between the @xmath5-band measurements and other literature measurements of the star ( section [ extcomp ] ) , which are themselves based on @xmath5-band features ( * ? ? * , n14 ) and visible features @xcite . * the stability against small rv shifts of the @xmath5-band measurements compared to those of the @xmath22 and @xmath10-band of m13a . the stellar population of the galaxy comprises multiple sub - populations with distinct kinematic , spatial , and compositional characteristics . these populations are often grouped into three components the thin disk , the thick disk , and the galactic halo . large scale surveys of fgk stars and luminous evolved stars have found rich information about the history and evolution of the galaxy within the detailed abundances of these different populations ( e.g. * ? ? ? * ; * ? ? ? similar large - scale studies of m dwarfs ( e.g. * ? ? ? * ; * ? ? ? * ) have also been carried out , and measurements of metallicity - sensitive band indices have enabled the identification and study of the oldest metal - poor m subdwarf populations ( e.g. * ? ? ? refined metallicity measurement techniques are under development , and have demonstrated great potential for application with these large m dwarf catalogs @xcite . precise metallicity measurements of the m dwarfs in the solar neighborhood reveal correlations between composition and stellar kinematic and activity characteristics that are consistent with the increasing age and decreasing level of activity from the thin disk to the thick disk @xcite . these results suggest the promise of precise abundance measurements in these long - lived stars . we explored whether the [ fe / h ] values in were consistent with the expected trends , as a check on our [ fe / h ] values and to probe for outliers and systematics . we calculated galactocentric uvw motions for all of our targets with parallax measurements . combining these parallaxes with proper motion values from lg11 and lspm - n and rv measurements from the literature where available ( or our rv measurements where literature rvs were unavailable ) , we have 283 targets with relatively precise uvw velocities ( parallax error @xmath102 , literature rv or measured rv error @xmath103 km s@xmath44 , no indications of binarity ) . these velocities are shown in the toomre diagram in the left panel of figure [ uvw ] , along with rough demarcations of the thin disk , thick disk , and halo @xcite . the toomre diagram is effectively a projection of the rotational kinetic energy of each star relative to the local standard of rest , with stars of higher energies at increasing radii @xcite . consistent with expectations from galactic kinematics , those targets with the highest [ fe / h ] values are kinematically within the thin disk region , and targets with lower [ fe / h ] values are spread across the entire range of velocities . the right panel of figure [ uvw ] shows the total galactocentric velocity plotted against h@xmath11 ew , a common activity indicator for m dwarfs ( e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ? the concentration of chromospherically active m dwarfs at low galactocentric velocities , compared with the absence of active m dwarfs at higher velocities , suggests that we are indeed probing both the thick and thin disks , consistent with the chemical and kinematic properties of the m dwarfs in our catalog . the boundary between the thick and thin disk populations is not discrete @xcite , and there are indications that the compositional gradient between the thick and thin disks should be less than 1 dex @xcite , consistent with figure [ uvw ] . the most significant outlier in these plots is cm dra ab , a well - studied short - period eclipsing binary m dwarf system ( e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ? constraints from the cooling age of a nearby co - moving white dwarf @xcite support the old age of this system , consistent with its high velocity , low [ fe / h ] , and possible @xmath11-enrichment @xcite . that this system is chromospherically active at its old age is not surprising due to its short period ( @xmath104 d ) and the tidal interactions that have presumably taken place . the infrared colors of m dwarfs have long been known to show sensitivity to stellar metallicity . this is in large part due to two primary sources of continuum opacity : h@xmath105 absorption and h@xmath6o absorption . lower - metallicity stars have higher h@xmath105 and h@xmath6o absorption , leading to a loss of flux especially in the @xmath10-band . the effects of metallicity on infrared colors have been exploited by many authors to construct photometric metallicity calibrations ( * ? ? ? * m13a , n14 ) , which are complementary ( observationally cheaper but less precise ) to the spectroscopic [ fe / h ] provided in this work . as another check of the integrity of our catalog of metallicities , we show in figure [ colfeh ] the @xmath106 colors along with the [ fe / h ] values for each of our targets . we confirmed that our data show a tight relation between position in the @xmath107-space and [ fe / h ] , a connection established through examination of kinematically - defined populations in @xcite . also shown are the kinematic age boundaries derived in @xcite , transformed to the 2mass photometric system using the updated 2mass color transformations @xcite . space is well - correlated with metallicity , as noted in @xcite . [ colfeh ] ] stellar composition is of fundamental importance in studies of exoplanets . the properties of observational relationships relating to composition , including the giant planet - metallicity @xcite and the planet mass - metallicity @xcite correlations , provide unique insight into the environments and mechanisms of planet formation . moreover , reliable planet host star compositions are necessary in order to apply empirical relationships to derive the stellar luminosities and temperatures @xcite , and hence the location of the habitable zone ( e.g. * ? ? ? * m13b ) . this breadth of impact motivates the empirical and theoretical efforts to define an m dwarf metallicity scale . multiple groups have already shown that the giant planet - metallicity correlation extends to the m dwarf regime . a well - calibrated and precise metallicity scale is crucial for these and future studies of m dwarf planet populations . the metallicity values in this catalog are derived from the most precise and well - calibrated scales available for m dwarfs , and it is valuable to compare our results with those of previous studies for planet - hosting m dwarfs . our catalog includes 16 m dwarfs that are known to be planet hosts , and in table [ ptable ] we show our measured [ fe / h ] , [ m / h ] , and h@xmath6o - k2 based spectral types . generally , we measure metallicities consistent with those of previous nir spectroscopic studies . in many cases our [ fe / h ] measurements are substantially different from previous photometric estimates , which is expected as those calibrations are inherently less precise and use fewer calibrators than the relations we employed . lccccccccccccc gj 1214 & & @xmath108 & & @xmath109 & @xmath110 & @xmath111 & @xmath108 & & @xmath88 & & @xmath112 & @xmath113 & m4.5 + hip 57050 & @xmath114 & @xmath115 & @xmath116 & & @xmath117 & @xmath118 & & @xmath88 & & & @xmath118 & @xmath119 & m4.0 + gj 876 & @xmath108 & @xmath120 & @xmath121 & @xmath122 & @xmath123 & @xmath124 & @xmath125 & @xmath126 & @xmath125 & & @xmath127 & @xmath124 & m3.5 + gj 317 & & & & & & & @xmath128 & @xmath127 & @xmath128 & & @xmath123 & @xmath129 & m3.5 + gj 581 & @xmath130 & @xmath131 & @xmath131 & @xmath132 & @xmath114 & @xmath131 & @xmath133 & @xmath134 & @xmath135 & & @xmath114 & @xmath119 & m3.0 + gj 179 & & @xmath129 & @xmath121 & @xmath111 & & @xmath122 & @xmath125 & @xmath136 & @xmath117 & & @xmath137 & @xmath117 & m3.0 + hip 79431 & @xmath138 & @xmath139 & & @xmath140 & @xmath141 & @xmath142 & & @xmath143 & & & @xmath144 & @xmath145 & m2.5 + gj 436 & @xmath119 & @xmath137 & @xmath146 & @xmath147 & @xmath148 & @xmath88 & @xmath149 & @xmath121 & @xmath119 & & @xmath148 & @xmath119 & m2.5 + gj 849 & @xmath125 & @xmath150 & @xmath111 & @xmath151 & @xmath152 & @xmath127 & @xmath153 & @xmath128 & @xmath128 & & @xmath143 & @xmath154 & m2.5 + gj 3470 & & & & & & & & & & @xmath111 & @xmath113 & @xmath155 & m2.0 + gj 15a & & & & & & & & & & @xmath156 & @xmath157 & @xmath158 & m2.0 + gj 649 & @xmath133 & @xmath118 & @xmath159 & @xmath119 & @xmath125 & @xmath160 & & @xmath146 & & & @xmath118 & @xmath161 & m1.5 + gj 433 & & & & & & & @xmath159 & & @xmath158 & & @xmath119 & @xmath162 & m1.5 + kepler-138 & & & & & & & & & & @xmath133 & @xmath163 & @xmath164 & m1.0 + wasp-80 & & & & & & & & & & @xmath165 & @xmath166 & @xmath116 & m0.5 + wasp-43 & & & & & & & & & & @xmath149 & @xmath112 & @xmath109 & m0.5 we have presented nir spectra and spectroscopic parameters for 886 nearby m dwarfs . this catalog represents the largest such compilation of m dwarf abundances , t@xmath3-sensitive indices , and nir spectra yet published , and includes : * [ fe / h ] and [ m / h ] measurements ( uncertainty @xmath85 dex ) for m1-m5 dwarfs using @xmath22 , @xmath10 , and @xmath5-band features , derived from multiple published calibrations . * [ fe / h ] with similar uncertainties derived from @xmath5-band features for m5-m8 dwarfs . * measurements of @xmath51-band h@xmath6o - based and other indices which are known to track t@xmath3 . * measurements of strengths of the 820 nm doublet and the 860 nm triplet , which are sensitive to activity and surface gravity . * stellar photometry and astrometry from the literature . * rvs ( uncertainty @xmath167 km s@xmath44 ) . we performed cross - checks among the different [ fe / h ] calibrations within our catalog and in the literature . we uncovered some systematic issues with the implementations of the @xmath22 and @xmath10-band calibrations , which led us to select the @xmath5-band calibration of m13a as the preferred measure of [ fe / h ] for m1-m5 dwarfs where multiple calibrations exist . using this [ fe / h ] , we confirmed the expected trends with galactocentric velocity activity levels , and stellar colors . we also provide updated abundances measurements for many of the known m dwarf exoplanet hosts . the original motivation for this survey was to support upcoming planet surveys targeting m dwarfs , including ground - based rv surveys @xcite and space - based transit surveys @xcite , to enable both survey target planning and rapid interpretation of results ( e.g. whether a planet is in the habitable zone of its host star ) . this catalog will also be useful for other studies of the nearby m dwarfs , and has already proven its value in this context . data from this catalog enabled insight into the tension between observations and model predictions for the eclipsing binary cm dra @xcite , supported the first confirmation of the low - mass members of the coma berenices cluster @xcite , and were used to demonstrate the use of new gravity and activity - sensitive spectral indices @xcite . this collection of m dwarf spectra provide a rich dataset on which to build even better calibrations , or to discover new useful features . it is our hope that this dataset , complementing other similar recent works ( e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) , will facilitate the work of other groups working to understand the nearby m dwarfs , and to further our understanding of these most common stars and their planets . this work was partially supported by funding from the center for exoplanets and habitable worlds . the center for exoplanets and habitable worlds is supported by the pennsylvania state university , the eberly college of science , and the pennsylvania space grant consortium . this work was also partially supported by the penn state astrobiology research center and the national aeronautics and space administration ( nasa ) astrobiology institute . we acknowledge support from nsf grants ast 1006676 , ast 1126413 , and ast 1310885 in our pursuit of precision radial velocities in the nir . this research has made use of the simbad database , operated at cds , strasbourg , france . this publication makes use of data products from the two micron all sky survey , which is a joint project of the university of massachusetts and the infrared processing and analysis center / california institute of technology , funded by the national aeronautics and space administration and the national science foundation . the authors wish to recognize and acknowledge the very significant cultural role and reverence that the summit of mauna kea has always had within the indigenous hawaiian community . we are most fortunate to have the opportunity to conduct observations from this mountain .

over the last 25 years , there has been significant effort in understanding the morphological evolution of galaxies and the dynamical evolution of the clusters of galaxies . the early work in this area proved very fruitful . it provided strong evidence for galaxy morphological segregation in clusters ( oemler 1974 and melnick @xmath3 sargent 1977 ) , _ i.e. _ early type galaxies preferentially occupy denser environments and/or are preferentially found near the cluster centers . in recent years , there have been many refinements to these initial observations e.g. dressler ( 1980 ) , kent @xmath3 sargent ( 1983 ) , whitmore @xmath3 gilmore ( 1991 ) , biviano et al . ( 1997 ) , stein ( 1997 ) , adami et al . ( 1998 ) , metevier et al . ( 2000 ) . these works have , however , not fully determined whether this morphological segregation is simply a function of redshift and/or the cluster environment _ e.g. _ does the galactic content of clusters evolve with cosmic epoch or as a function of the cluster s virial mass ? to try to understand these issues , there has been considerable effort over the last two decades dedicated to studying the dynamical and morphological evolution of clusters via multi object spectroscopy and photometric imaging ; for example , dressler ( 1980 ) ; butcher & oemler ( 1984 ) ; gunn et al . ( 1986 ) ; couch et al . ( 1991 ) ; bower et al . ( 1997 ) ; carlberg et al . ( 1997 ) ; dressler et al . ( 1999 ) . over the last few years , large surveys of distant clusters of galaxies _ e.g. _ the canadian network of observational cosmology ( cnoc ) ( see carlberg et al . 1997 ) have begun to provide interresting answers . for example , the red galaxies in the cnoc cluster sample ( assumed to be early type galaxies ) have a velocity dispersion nearly @xmath4 times lower than the blue galaxies ( assumed to be later type galaxies ) . ccccccc @xmath5 & mean z & vel . corrected vel . disp . & sampling & center ( j2000 ) + pdcs16 & 0.398 & 944@xmath6300 km.s@xmath7 & 802 km.s@xmath7 & 12 & 2h28m27.7s 0deg3158 `` + pdcs30/45 & 0.250 & 607@xmath6470 km.s@xmath7 & 516 km.s@xmath7 & 14 & 9h54m38.3s 47deg4540 '' + pdcs30/45 & 0.331 & 588@xmath6480 km.s@xmath7 & 500 km.s@xmath7 & 9 & 9h54m34.7s 47deg1210 `` + pdcs38 & 0.469 & 1819@xmath6720 km.s@xmath7 & 1546 km.s@xmath7 & 13 & 9h51m35.3s 47deg4540 '' + pdcs57 & 0.459 & 1282 km.s@xmath7 & 1090 km.s@xmath7 & 7 & 13h23m35.3s 30deg610 `` + pdcs62 & 0.462 & 1757@xmath6630 km.s@xmath7 & 1493 km.s@xmath7 & 19 & 13h23m33.7s 30deg2248 '' + [ ] we have also initiated a new study , the cfh optical pdcs ( cop ) survey , described in detail in adami et al . ( 2000a ) and holden et al . the cop survey contains approximately 650 galaxy redshift measurements ( from 6 nights of cfht time ) towards more than 10 pdcs clusters at a mean redshift of 0.4 . we have also obtained r band photometry at the cfht for these targets . we have combined our r - band data with the published pdcs photometry in v and i. this combined data set is allowing us to study the dynamical and morphological status of optically selected distant clusters of galaxies . we can then directly compare with the enacs ( eso nearby abell cluster survey : e.g. katgert et al . 1996 ) and cnoc samples ( carlberg et al . 1997 ) . due to the relatively low number of galaxies , the statistics of an individual cluster are too low to compare individually to nearby clusters . therefore , we combined the data from all the clusters in our cop study ( at z@xmath00.4 ) whose velocity dispersions were greater than 500 km.s@xmath7 into one composite cluster . we also applied the same technique to the available cnoc data from which we built two composite clusters , one at z@xmath00.32 and one at z@xmath00.48 . our results and conclusions , then , are based upon the comparison of the composite clusters constructed in the same way from the enacs , cop and cnoc surveys . in section 2 , we summarize the data and describe the methods used in analyzing them . we then discuss evolutionary effects , since our clusters extent over a relatively large redshift range and have been all analyzed in the same manner . in section 3 , we describe our measurements of galaxy segregation . in section 4 we discuss the galaxy evolution in our cluster sample as a function of redshift and the implications that our data pose on the formation epoch of these clusters . we use q@xmath8=0 and @xmath9=0 . distances are given in units of h=@xmath10/100 . in this paper we use the cop data described in adami et al . ( 2000a ) and holden et al . ( 2000 ) as well as the enacs and cnoc data . the cop survey is an imaging and spectroscopic follow - up of 10 pdcs ( postman et al . 1996 ) lines of sight which contain 15 pdcs cluster candidates at z@xmath110.4 ( adami et al . 2000a and holden et al . 2000 ) . here , we compare the dynamical characteristics of the galaxies belonging to those pdcs clusters to their enacs ( adami et al 1998 ) and cnoc counterparts . the low redshift data are based on a compilation from the enacs database ( katgert et al . 1996 , mazure et al . 1996 ) and the literature ( adami et al . 1998 ) . all the galaxies in the nearby sample have a redshift and a magnitude that were derived in the same manner as the current data . they have also been classified into morphological types , but no color data comparable to our cop data are available . we also use 5 cnoc clusters available in the literature ( ms0016 + 16 , ms0302 + 16 , ms1008 - 12 , ms1224 + 20 and ms0451 - 03 ) . ms1358 + 62 and ms1621 + 26 were excluded from the sample because of their distorded apparent shape . the cnoc galaxies have r and g magnitudes and are classified into spectral types according to their color and spectral distribution ( yee et al . we compute an absolute magnitude according to the redshift and the spectral type of each galaxy using the k - corrections given in frei @xmath3 gunn ( 1994 ) ( see below for the definition of the cnoc spectral types ) . we used the most massive ( @xmath12 ) clusters in the pdcs sample in our comparison with the enacs and cnoc surveys in order to match the same class of clusters . we compute the velocity dispersions of the massive systems detected in the cop survey ( adami et al . 2000a ) using the bi weight robust estimators of beers et al . ( 1990 ) and applying a `` 3-@xmath13 clipping '' . this is exactly the same methodology used by adami et al . ( 1998 ) for the enacs systems and this ensure us to have robust estimators despite the relatively low numbers of galaxies used . our final velocity dispersion measurements are within 5@xmath14 of those presented in holden et al . ( 2000 ) for all the clusters in common except pdcs30/45 . this last group has a complicated structure in redshift space which probably produced deviations from holden et al . in the velocity dispersion determination . however , if we excluded that cluster from our sample , the results described in the following sections do not change by more than 5@xmath14 . we decided , therefore , to not exclude those galaxies when building our composite cluster ( see next section ) . due to the small number of redshifts per cluster in our sample ( @xmath15 ) , we were unable to distinguish between inter - loper galaxies those which are passing - through the cluster but are not physically bound to the cluster and real cluster members . katgert et al . ( 1996 ) showed that not excluding inter - lopers results in an increase of the velocity dispersion of the clusters by between 10@xmath14 and 25@xmath14 . to account for this effect in our sample , we have , therefore , systematically decreased our measured velocity dispersions by 15@xmath14 . for this paper , we selected only the clusters with a redshift greater than 0.25 , a velocity dispersion greater than 500 km.s@xmath7 and with more than 7 redshifts ( see table 1 ) . in total , we have 6 clusters at a mean redshift of 0.4 and 74 cluster galaxies with measured redshift and v@xmath16 and i@xmath16 magnitudes ( 56 have also measured r magnitude ) . only one of the clusters in our sample has fewer than 9 redshifts . this is not a problem because the velocity dispersion can be reliably estimated using bi - weight estimators with 9 or more redshifts ( see lax 1985 ) . we analyzed the cnoc clusters in the same way as the pdcs and enacs clusters . first of all , we defined groups along the lines of sight by using the same gap technique as in adami et al . ( 1998 ) or adami et al . ( 2000a ) . this technique searches for gaps of more than 1000 km.s@xmath7 in the distribution of the galaxies ordered by increasing redshift . the choice of 1000 km.s@xmath7 is well adapted to the data we have ( see also katgert et al . we checked , however , that values between 600 and 1500 km.s@xmath7 do not significantly change the results . when more than 5 galaxies were between two velocity gaps , we assumed that this was a structure . with this technique we detected all the groups defined in carlberg et al . ( 1997 ) with very similar mean redshifts ( see table 2 ) plus several small and probably diffuse groups not listed by carlberg et al . in order to use a uniform set of selection criteria , we chose only groups with a velocity dispersion larger than 500 km.s@xmath7 . when the detected groups were found to be listed in carlberg et al . ( 1997 ) , we used the velocity dispersion quoted in that paper . for groups not detected by carlberg et al , we computed the velocity dispersion by the same method described in the previous section . we kept only groups with velocity dispersions larger than 500 km.s@xmath7 . in total , only one structure was added to the 5 main structures in the cnoc sample ( carlberg et al . this group , in the ms0016 + 16 field , is at a redshift of 0.328 and has a velocity dispersion of 610 km.s@xmath7 ( denoted ms0016 + 16@xmath17 in the following ) . excluding this group did not change the results significantly . finally , we only used the cnoc galaxies within a radius of 1 h@xmath7 mpc from the center in order to match approximately the pdcs and enacs criteria . for ms0016 + 16@xmath17 , we used the position of the brightest galaxy as the center . ccccccc @xmath5 & mean z & vel . disp . & sampling + ms0016 + 16 & 0.548 & 1243 km.s@xmath7 & 42 + ms0016 + 16@xmath17 & 0.326 & 610 km.s@xmath7 & 8 + ms0302 + 16 & 0.425 & 656 km.s@xmath7 & 33 + ms0451 - 03 & 0.539 & 1354 km.s@xmath7 & 43 + ms1008 - 12 & 0.307 & 1059 km.s@xmath7 & 75 + ms1224 + 20 & 0.326 & 798 km.s@xmath7 & 24 + [ ] the end result was 6 cnoc structures ( see table 2 ) : three of them at z@xmath00.32 ( ms0016 + 16@xmath17 , ms1008 - 12 and ms1224 + 20 ) ; the other three in the redshift interval [ 0.4;0.55 ] ( ms0016 + 16 , ms0302 + 16 and ms0451 - 03 , mean redshift of z@xmath00.48 ) . we built as described in section 2.3 two cnoc composite clusters with these structures . the z@xmath00.32 composite cluster has 107 galaxies and the z@xmath00.48 composite cluster has 118 galaxies . these numbers are similar to the number of galaxies in the cop composite cluster . following adami et al . ( 1998 ) , we built `` composite clusters '' made by combining the data from all the individual systems . we compute galaxy velocities relative to the mean velocity and normalized to the velocity dispersion of each cluster . the normalized velocity of galaxy i in cluster j , v@xmath18 , is defined as v@xmath18 = ( v@xmath19-v@xmath20)/@xmath21 , with v@xmath20 being the mean velocity of cluster @xmath22 , @xmath21 being its velocity dispersion and v@xmath19 the original velocity of galaxy `` i '' . we merged these normalized velocities into the composite clusters . the uncertainties could be introduced by systematic errors on the @xmath21 ; however , the use of robust estimators prevents us from severe biases in the estimation of the @xmath21 . we stress that this technique of using composite clusters has been already used successfully , for example in adami et al . ( 1998 ) . we have also investigated possible bias due to velocity dispersion inhomogeneity between the 3 composite clusters . we are not very likely to have such selection effects because at z@xmath00.32 , 0.40 and 0.48 , the mean velocity dispersions of the individual clusters used to make the composite clusters are 966 km.s@xmath7 , 1047 km.s@xmath7 and 1119 km.s@xmath7 and they are overlapping if we consider error bars . ccccc color ( z ) & e & sbc & scd & i m + v@xmath23i ( z=0.25 ) & 1.69 & 1.21 & 1.00 & 0.78 + v@xmath23r ( z=0.25 ) & 0.74 & 0.46 & 0.42 & 0.31 + r@xmath23i ( z=0.25 ) & 0.95 & 0.75 & 0.58 & 0.47 + v@xmath23i ( z=0.30 ) & 2.00 & 1.28 & 1.08 & 0.81 + v@xmath23r ( z=0.30 ) & 0.87 & 0.53 & 0.48 & 0.35 + r@xmath23i ( z=0.30 ) & 1.13 & 0.775 & 0.60 & 0.46 + v@xmath23i ( z=0.40 ) & 2.22 & 1.43 & 1.24 & 0.87 + v@xmath23r ( z=0.40 ) & 1.13 & 0.66 & 0.60 & 0.43 + r@xmath23i ( z=0.40 ) & 1.09 & 0.77 & 0.64 & 0.44 + v@xmath23i ( z=0.45 ) & 2.33 & 1.55 & 1.34 & 0.97 + v@xmath23r ( z=0.45 ) & 1.12 & 0.73 & 0.64 & 0.49 + r@xmath23i ( z=0.45 ) & 1.21 & 0.82 & 0.70 & 0.47 + [ ] before we begin to discuss the morphological content of the pdcs clusters , we must first establish a common language by which we can compare our results to past and present work in the field _ e.g. _ the use of the early and late morphological types . our imaging data did not have the resolution necessary to allow us to assign galaxies a morphological type . we could , nevertheless , classify our galaxies in terms of their spectral energy distribution using our photometry and spectroscopy . our first classification scheme separates galaxies into emission line and non - emission line objects . this classification relates well to the early / late morphological type classification ( e.g. biviano et al . 1997 ) . however , in some cases early type galaxies , like sos , can exhibit emission lines features in their spectra . in any case , biviano et al . ( 1997 ) have shown that less than 4@xmath14 of elliptical or so cluster galaxies have emission lines . this implies that our spectroscopically determined `` late type galaxies '' ( see fig . 4 ) are probably free from significant contamination by early type objects . in other cases , some late type galaxy spectra do not have strong emission lines , however . moreover , we used blocking filters to obtain the spectra of the cop galaxies ( see adami et al . 2000a ) which significantly reduced the available spectral range . it is possible , therefore , that we missed emission line features due to this reduced spectral coverage . our `` early type galaxies '' bin may , therefore , be contaminated by late type objects . the other classification scheme we used was based on the three - color information ( v , r and i ) we had for most of the galaxies in the sample ( 56 out of 74 ) . once the redshift was known , the spectral energy distribution was constrained and a rough classification between different spectral types was assigned . we used the frei @xmath3 gunn ( 1994 ) colors to classify our galaxies into 4 spectral types . table 3 gives the v@xmath23i , v@xmath23r and r@xmath23i colors from frei @xmath3 gunn ( 1994 ) for the 4 different morphological types we used and for the 4 redshifts of our composite clusters . we matched the 3 colors of each object with this table . first , we transformed the v@xmath24i@xmath16 , v@xmath24r@xmath25 and r@xmath26i@xmath16 into the standard johnson system used in frei @xmath3 gunn ( 1994 ) . as in postman et al . ( 1996 ) and adami et al ( 2000a ) we applied the following relations for the color transformations : v@xmath23i = 0.41 + 0.855(v@xmath24i@xmath16)+ 0.012(v@xmath24i@xmath16)@xmath17 , v@xmath23r = @xmath230.02+(v@xmath24r@xmath25)@xmath270.056(v@xmath24i@xmath16)+ 0.012(v@xmath24i@xmath16)@xmath17 , and r@xmath23i = 0.43+(r@xmath26i@xmath16)@xmath230.089(v@xmath24i@xmath16 ) . each color will , in principle , correspond to a single color - type , but this did not always happen . we therefore developed a method of using the colors to uniquely determine the galaxy color - type . first , if all 3 independent color - type determinations derived by the colors did yield the same result , we selected that color - type . this occurred for 24@xmath14 of the sample . second , if 2 color - types among the 3 were identical , and if the third color - type was close to these two ( for example , i m is close to scd ) , we selected the two identical color - types ( 50@xmath14 of the sample ) . third , if 2 color - types among the 3 were identical , but if the third color - type was not close to these two ones ( for example , i m is not close to sbc ) we again selected the two identical color - types ( 5@xmath14 of the sample ) . fourth , if the 3 color - types were different and close ( for example , e , sbc and scd ) , we selected the mean color - type ( 14@xmath14 of the sample ) . finally , if the 3 color - types were different but not close ( for example , e , sbc and i m ) , we selected the color - type providing the lowest difference with the values of table 3 ( 7@xmath14 of the sample ) . we have assumed that the @xmath28 bin of frei @xmath3 gunn ( 1994 ) was a mix of elliptical and so galaxies , the @xmath29 bin was a mix of sa , sb and sc galaxies , the @xmath30 bin was a mix of sd and sm galaxies and the @xmath31 bin contained only irregular galaxies . this classification technique is similar to the one used in the cnoc survey ( carlberg et al . 1997 ) , except that we used three colors instead of two , and we assigned a morphological type ( called color - type hereafter ) using the frei @xmath3 gunn ( 1994 ) templates . the classification estimates with the largest uncertainties are for the last two cases and they constitute only 21@xmath14 of the sample . the main uncertainties in the color classification scheme came from the uncertainties in the magnitudes . possible dust obscuration found , for example , in cl0039 + 4713 ( z=0.41 : smail et al . 1999 ) can lead to misclassifications , but we ignored this effect . we also compare the predicted types of our two classification schemes . overall , emission line galaxies should be mostly late type galaxies and absorption line galaxies should be mostly early type galaxies . the direct comparison shows that , on one hand , 100@xmath14 of the emission line galaxies of our sample have a color - type later than sbc ( sbc : 33@xmath14 , scd:42@xmath14 , i m : 25@xmath14 ) . on the other hand , 14@xmath14 of the non - emission line galaxies have a color - type later than scd ( e : 50@xmath14 , sbc : 36@xmath14 , scd:0@xmath14 , i m : 14@xmath14 ) . these percentages are well within our expectations ( see above ) and these results give strength to our classification schemes . however , we must consider this color classification scheme ( red galaxies : early types , blue galaxies : late types ) only as an @xmath32 of the true morphological type . the spectral assignment is clearly not as reliable as the @xmath33 morphological types obtained , for example , via hst observations by dressler et al . ( 1999 ) or lubin et al . the morphology and the stellar content of a galaxy can be affected differently by various astrophysical processes and consequently the two tentative classifications by colors or spectral features described in this section must be considered only as an approximation of the morphological types . however , in order to compare our distant cluster with their local counterparts ( enacs ) we needed to establish such a relation between spectral and morphological classifications . in order to relate the cnoc composite clusters with the rest of our sample ( significantly increasing the number of galaxies ) , we decided to split the cnoc spectral classification into two bins that could be related to our previous classification . hereafter , we refer to these subsamples as early and late type cnoc galaxies we assigned the cnoc spectral types 1 , 2 ( elliptical galaxies ) and 3 ( e+a galaxies ) to the early type class . these are mainly elliptical galaxies because the cnoc type 3 galaxies represent less than 2.5@xmath14 of the sample . the cnoc types 4 and 5 are assumed to be late type galaxies . in order to compare with the enacs data , we associated this late type bin to the bin sc+sd+sm+irr of adami et al . as shown in adami et al . ( 1998 ) , there exists both a magnitude and a morphological segregation of galaxies in nearby clusters that agrees with simple dynamical models of cluster formation and evolution . the bright galaxies have a lower velocity dispersion compared to the fainter ones . this variation is expected from energy equipartition in clusters . the elliptical galaxies have a lower velocity dispersion compared to the later morphological types in nearby clusters . the ratio of these dispersions between the elliptical and the late type spiral galaxies is consistent with a model where the late type spirals are currently falling into the cluster . we now compare the results of adami et al . ( 1998 ) for nearby clusters with the results from the 3 distant composite clusters . we plot in figures 1 , 2 and 3 the variation of the normalized velocity dispersions versus the absolute magnitude for the 3 distant composite clusters . we made 4 bins of 17 distant galaxies for the pdcs composite cluster at z@xmath00.4 ( fig . 2 ) , 5 bins of 17 galaxies for the cnoc composite cluster at z@xmath00.32 ( fig . 1 ) and 6 bins of 17 galaxies for the cnoc composite cluster at z@xmath00.48 ( fig . we transformed the v@xmath16 magnitudes into r magnitudes ( v@xmath24r@xmath111 ) to over plot the results for the pdcs composite cluster on the results for the nearby clusters ( which used r magnitudes ) . we have not modified the values of the cnoc r magnitudes because according to frei @xmath3 gunn ( 1994 ) , the correction between r and r magnitudes is less than 0.2 magnitudes . this uncertainty is negligible compared , for example , to the uncertainty in the v@xmath24r value . we have excluded galaxies whose velocities were @xmath343-@xmath13 away from the mean in each bin for the z@xmath00.4 and z@xmath00.32 composite clusters . the figures show qualitatively the same results of adami et al . ( 1998 ) superposed to each composite cluster . for the z@xmath00.32 and z@xmath00.4 composite clusters , the agreement is apparently good except for the faintest bin of the z@xmath00.4 composite cluster and the r@xmath0 - 21.5 point of the z@xmath00.32 composite cluster ( see the discussion section for an explanation of this peculiar value ) . for the z@xmath00.48 composite cluster , the r@xmath0 - 22.3 and r@xmath0 - 22 points seem to be higher than the corresponding z@xmath00.07 points as well as the energy equipartition prediction ( thick solid line ) . the energy equipartition law ( e.g. adami et al . 1998 : @xmath13=10@xmath35 with m the magnitude ) has been normalized to r = @xmath2321.5 ( + 5@xmath36log@xmath37(h ) ) . the agreement with the low redshift data and the energy equipartition for this last composite cluster seems to be worse compared to the other two composite clusters at z@xmath00.32 and z@xmath00.4 . however , taking into account the error bars , the differences are not very significant and we need to estimate quantitatively the level of agreement . in order to find a way to quantify more efficiently the possible differences between the z@xmath00.07 results and the higher redshift distributions , we used a two dimensional kolmogorov - smirnov test ( e.g. press et al . 1992 ) in the normalized velocity dispersion - r magnitude phase space . we used all the individual values of each galaxy ( without binning ) in the magnitude interval r=[-23.;-19.5 ] . each kolmogorov - smirnov test was run 1000 times ( bootstrap resampling ) to estimate the significance level . at the 10@xmath14 level , we found no significant differences between the normalized velocity dispersion versus r magnitude for the z@xmath00.07 , z@xmath00.32 and z@xmath00.4 composite clusters . however , the distribution of the z=0.48 cluster was found to be significantly different ( even at 1@xmath14 level ) from the three lower redshift composite clusters . the results show , therefore , a significant evolution of the magnitude segregation in the most distant composite cluster , but none for the z@xmath00.07 , z@xmath00.32 and z@xmath00.4 composite clusters . in this section , we also compared the variation of the @xmath38 versus the @xmath39 . aa a first step , we do not consider the spatial distribution of the galaxies ( see next section ) . the types 1 , 2 , 3 and 4 of fig . 4 were respectively assigned to the closest galaxy types : elliptical , so , sa+sb and sc+sd+sm+irr . we stress that the error bars are too large to draw firm conclusions regarding the morphological segregations shown in fig . 4 . however , the variation of the mean values is still suggestive . we now comment on this . for the z@xmath00.4 galaxy sample , we used both the emission line ( large filled squares ) and color ( large empty squares ) classifications to assign morphological types . we used the previously defined early and late spectral types for the z@xmath00.32 and z@xmath00.48 cnoc samples . the results are shown in figure 4 , together with the z@xmath00.07 results ( small empty circles ) of adami et al . ( 1998 ) . to plot these results on the same figure , we assumed that the z@xmath00.4 pdcs galaxies with : only absorption lines were an equal mix of elliptical and so galaxies only emission lines were an equal mix of spiral and irregular galaxies a color - type equal to the @xmath28 type of frei @xmath3 gunn ( 1994 ) corresponded to type 1 of adami et al . ( 1998 ) a color - type equal to the @xmath29 type of frei @xmath3 gunn ( 1994 ) corresponded to type 3 of adami et al . ( 1998 ) a color - type equal to the @xmath30 + @xmath31 type of frei @xmath3 gunn ( 1994 ) corresponded to type 4 of adami et al . ( 1998 ) further , we assumed that the cnoc galaxies which are classified as early types were elliptical galaxies and that those classified as late types were type 4 of adami et al . ( 1998 ) the results were found to be qualitatively similar for the z@xmath00.07 , z@xmath00.32 and z@xmath00.4 cluster galaxies . the trend shown is in agreement with the theoretical predictions . for example the ratio between the two z@xmath00.32 bins is close to the @xmath4 value predicted for a relaxed and an infalling galaxy population ( see e.g. adami et al . the ratio is also close to the value given in carlberg et al . ( 1997 ) . the results for the z@xmath00.48 are , however , slightly different compared to the other redshifts . the ratio of @xmath4 is still observed . there seems to be a systematic shift towards higher values of the normalized velocity dispersions , however . these distant clusters could be dynamically younger compared to the other not so distant clusters . we now consider the spatial distribution of the galaxies with respect to their magnitude and spectral type . ccccccccccc z bin / r & -22.75 & -22.5 & -22.25 & -22 & -21.75 & -21.5 & -21.25 & -21 & -20.75 & -20.5 + 0.32 & - & 0.69 & 0.65 & 0.83 & 0.98 & 0.98 & 0.95 & 0.98 & 0.98 & 0.99 + 0.40 & - & - & - & - & 0.25 & 0.65 & 0.95 & 0.95 & 0.95 & 0.91 + 0.48 & 0.79 & 0.94 & 0.99 & 0.99 & 0.99 & 0.99 & 0.99 & 0.99 & 0.99 & 0.99 + [ ] we used a two dimensional kolmogorov - smirnov test based on the distance from the center _ versus _ normalized velocity dispersion phase space ( d-@xmath13 phase space hereafter ) to investigate the segregation as a function of spectral type and magnitude . in the z@xmath00.32 , z@xmath00.4 and z@xmath00.48 composite clusters , we showed that the distribution of the early type is significantly different , at the 10@xmath14 level , compared to the late type galaxies . this agrees with the results of adami et al . ( 1998 ) : early type galaxies in nearby clusters are more concentrated at the cluster center than late type galaxies . we also investigated the spatial distribution of the different spectral types as a function of magnitude using a kolmogorov - smirnov test ( with 1000 bootstrap resampling for each case ) . the results are given in table 4 . for the z@xmath00.32 cluster , we show that the distribution in the d-@xmath13 phase space of early type galaxies brighter than r@xmath11 - 21.75@xmath60.25 is significantly different at the 10@xmath14 level compared to the distribution of late type galaxies . in the same time , there is no significant difference between late type and early type galaxies fainter than r@xmath11 - 21.75@xmath60.25 . this means that only the bright early type galaxies are more concentrated at the center than the late type galaxies . biviano et al . ( 2000 ) found a similar result in nearby clusters . for the z@xmath00.4 cluster , we found the same trend if we used the emission line / absorption line classification . this time , the segregation appears at magnitudes brighter than r@xmath11 - 21.25@xmath60.25 . for the z@xmath00.48 cluster , however , the segregation starts to be significant at r@xmath11 - 22.5@xmath60.25 , brighter than for the z@xmath00.32 and z@xmath00.4 clusters . once again , the most distant composite cluster seems to be different form the less distant composite clusters . only the very bright early type galaxies show a more concentrated distribution than the late type galaxies . the cop clusters are @xmath40 while the cnoc clusters are @xmath41 selected . we now consider whether these different selection methods are responsible for the differences seen between the z@xmath00.48 composite cluster ( x - ray selected ) and the other composite clusters at lower redshift . for the magnitude and the spatial distribution segregations , these different selection methods did not seem to play a significant role because the z@xmath00.32 ( x - ray selected ) composite cluster exhibits the same characteristics as the z@xmath00.07 and z@xmath00.4 composite clusters ( optically selected ) and it was selected in the same way as the z@xmath00.48 composite cluster . in the case of the morphological segregations , this effect is not statistically significant . we , therefore , assume that our results do not depend on whether the clusters were x - ray or optically selected . we stress that our conclusions are strictly valid only for the clusters of our sample . however , these clusters are a fair sample of the general massive cluster population , and the results presented here can probably be generalized to other clusters . based on our previous analyses we found no evidence for redshift evolution of the dynamical state in our cluster sample from z@xmath42 to z@xmath43 . we detected , however , significant evolution between the z@xmath43 and z@xmath44 composite clusters using a kolmogorov - smirnov test . we also find that the cluster galaxy content changes from z@xmath00.07 , z@xmath00.32 , z@xmath00.4 to z@xmath00.48 . in the nearby sample of cluster galaxies of adami et al . ( 1998 : z@xmath00.07 ) , the type 1 galaxies ( ellipticals ) are 24@xmath14 of the sample , the type 2 ( so ) are 49@xmath14 , the type 3 ( sa+sb ) are 20@xmath14 and the type 4 ( sc+sd+sm+irr ) are only 7@xmath14 . in the sample at z@xmath00.32 , the early type galaxies are 78@xmath14 , the late type galaxies are 22@xmath14 ( type 4 ) . in the sample at z@xmath00.4 , the redder galaxies ( assumed to be 1+2 : ellipticals + so ) are 44@xmath14 of the sample ( versus 24@xmath14 + 49@xmath14 for the nearby sample ) , the intermediate color galaxies ( assumed to be type 3 ) are 38@xmath14 and the bluer galaxies ( type 4 ) are 18@xmath14 . and finally , in the sample at z@xmath00.48 , the early type galaxies are 70@xmath14 , the late type galaxies are 30@xmath14 ( type 4 ) . the percentages of late type galaxies are higher in the distant cluster samples . the estimate of the uncertainties in these percentages is not straightforward but we assumed the section 2.4 maximum contamination of 21@xmath14 ( @xmath03-@xmath13 ) in assigning spectral types when using colors in the z@xmath00.4 sample . to be conservative , we applied this to the relative percentages of red , intermediate and blue galaxies of our sample . the uncertainties on the numbers given in this paragraph are , therefore , based on a 7@xmath14 uncertainty . in other words , the distant cluster galaxies are statistically bluer ( or present later types ) than the nearby cluster galaxies . this is the `` classical '' butcher @xmath3 oemler ( 1984 ) effect that already has been extensively reported in the literature . at even higher redshifts , van dokkum et al . ( 1999 ) found the same effect in the cluster ms1054 - 03 at z@xmath00.83 . they showed clear evidences of further evolution of late type galaxies into early types ( so @xmath45 e ) . in order to compare results free from possible bias in our classification for the z@xmath00.4 sample , we also compare the percentage of emission line galaxies in this sample ( 19@xmath63@xmath14 ) to the enacs sample of nearby clusters ( biviano et al . 1997 : 10@xmath61@xmath14 ) , without using any color classification . we see that the emission line galaxy fraction increases by a factor of @xmath02 . in principle , our ability to measure redshifts for emission line galaxies is higher than that for absorption line galaxies , but this effect is canceled out by the short spectral ranges we used ( see adami et al . 2000a ) which occasionally can lead to missed emission line features . the region ( @xmath46 mpc ) covered by the cop and enacs surveys are similar ( see adami et al . 2000a and adami et al . 1998 ) , which is important as we discuss below . the evolution in the galaxy content of the clusters in our sample between z@xmath00 and z@xmath00.48 seems to be real . it does not depend on the sample used to test it , whether it is selected according to their spectral features or to their colors . these results are consistent with the recent studies of dressler et al . ( 1999 ) of z @xmath47 clusters . they find evolution of the morphological types of cluster galaxies . the metric aperture used , however , plays a key role in determining the blue fraction : for example , the implied evolution of the blue fraction with redshift was not confirmed by lubin et al . ( 1998 ) and postman et al . they showed that the massive cluster 1604 + 4304 at z=0.84 has a morphological content similar to that of nearby clusters . but , these studies used images of the inner core of the cluster ( @xmath48 ) which is significantly smaller than the area sampled in adami et al . ( 1998 ) , our comparison study . if we assume that the percentage of late type galaxies increases from the inner part to the outer parts of the clusters ( e.g. adami et al . 1998 ) , then this is an explanation of the discrepancy between our results and the results of lubin et al . ( 1998 ) and postman et al . ( 1998 ) . in order to check this point , we have computed the percentage of red , intermediate and blue galaxies in our z@xmath00.4 sample but in the same physical area as in lubin et al . ( 1998 ) : the core of the clusters . we found in this region , 65@xmath14 red galaxies , 18@xmath14 intermediate galaxies and 17@xmath14 very blue galaxies . these percentages are in agreement with the work of lubin et al . the choice of metric aperture is therefore important , and it could be an explanation for why the longer look back time studies of lubin and postman did not yield evidence for evolution . it may be that at longer look back times ( z@xmath00.8 ) other effects could lead to a relatively low blue galaxy fraction ( see section 5 ) . the entire evolutionary process is complicated , however , and it probably also relates to the dynamical state of the clusters as well as its redshift . for example , wang et al . ( 1997 ) showed a correlation between cluster ellipticity and blue galaxy fraction , and metevier et al . ( 2000 ) showed that there exist nearby clusters that have bi - modal galaxy or x - ray distributions which have relatively high blue fractions and are yet at low ( @xmath49 ) redshifts . when we compared the dynamical and the cluster content evolution of our composite clusters between z@xmath00.4 and z@xmath00.48 , the situation is consistent : we both found `` younger '' clusters and a `` younger '' population at higher redshift . however , the composite clusters from z@xmath00.07 to z@xmath00.4 give the following puzzle : how did the cluster contents evolve in morphological type , while the clusters have not apparently evolved in terms of their dynamical state ? we now examine this point . the answer is , it is possible to have the bright galaxies virialize before changing morphological type . this would explain why our composite nearby and distant clusters have the same distribution of galaxies versus magnitude , but they do not have the same ratio of late to early type galaxies . we can see this by assuming that galaxies virialize via dynamical friction , which takes 0.1 gyrs for the brightest galaxies for example in our @xmath50 sample and 1 to 1.5 gyrs for the faintest galaxies ( sarazin , 1986 ) . yet , if galaxy - galaxy interactions cause the galaxies to change morphological type , the galaxies take about 1 gyr to change ( crossing time ) , and clearly the more massive ( brighter ) galaxies can reach a `` relaxed state '' before changing morphological type . without additional assumptions , however , this scenario runs into a difficulty ; namely , the time between @xmath51 and @xmath50 is about 4 gyrs , and the nearby clusters should have virialized such that the dynamical state of the nearby and distant clusters should not be the same . a way to deal with this is to assume that clusters are ( not surprisingly ) continually evolving ( e.g. ulmer et al 1992 and references therein ) . this scenario is clearly not new and has been investigated in details by authors such as oemler ( 1974 ) , melnick @xmath3 sargent ( 1977 ) , dressler ( 1980 ) or whitmore @xmath3 gilmore ( 1991 ) . however , here , we have a unique case to check this scenario with homogeneously analyzed clusters from z@xmath00 to z@xmath00.5 . clusters are evolving being fed continuously by the in - fall of field galaxies and groups ( or other clusters ) . the majority of the infalling field galaxies are late - type . the most recent episode of in - fall could have been such that the relative low mass late type galaxies in the outer regions have been able to evolve into early type galaxies ( @xmath01 gyr ) while yet not virializing ( @xmath01.5 gyr ) . this leads to the nearby clusters having higher velocity dispersions in their outer portions while , at the same time , having relatively low fractions of late - type galaxies . this hypothesis is consistent with finding a correlation with dynamical age and blue fraction ( wang et al . it is not completely clear how this hypothesis relates to finding nearby clusters which have bi - modal distributions and appear to have relatively large blue fraction ( metevier et al . 2000 ) . in a high matter density universe , however , the cluster formation occurs at late epochs ( close to the present time ) and a high density universe could then leads to the production of dynamically young structures until relatively recent epochs . if we assume the general picture of cluster evolution and formation , we can extrapolate this scenario back in time to place a lower limit on when clusters formed and also suggest what clusters should look like in terms of morphology and dynamical state if we are able to extend studies to redshifts up to @xmath52 . this is discussed in the next subsection . an interesting question to ask is : when did all these clusters form ? this epoch depends strongly of the cosmological model ( e.g. oukbir @xmath3 blanchard 1992 and 1997 ) . in a high @xmath53 universe , the clusters form later than in a low @xmath53 universe , but they evolve faster . to adress this question , we can use the composite clusters of our sample from z@xmath00.07 to z@xmath00.4 . these are already , at least partially , relaxed . we can then assume that the `` formation '' process has occurred at an epoch earlier than z@xmath00.4 . the clusters of the z@xmath00.48 sample have younger dynamical characteristics and we can assume that they are close to the @xmath54 of their `` first formation '' process , which probably occured around z@xmath00.5 . now we have to estimate the time needed to form such a cluster to estimate the epoch of the @xmath55 of the formation process . the surface covered by the clusters of our samples is a circle of about @xmath01h@xmath7 mpc radius . the minimal time to form such a cluster is then less than 1 gyear in a violent relaxation process ( 1 crossing time , see sarazin 1986 ) . adding 1 gyear to z@xmath00.5 gives the epoch of the beginning of the cluster formation , i.e. in a q@xmath8=0 universe @xmath07.2 gyears ( not significantly different from when we use low values of q@xmath8 but different from 0 ) . we then conclude that the cluster formation process for the clusters of our sample began at z@xmath340.8 . consistent with this hypothesis are the recent detections of clusters of galaxies at z@xmath00.8 ( in addition to the two quoted in section 4.1 : ms1054 - 03 and 1604 + 4304 ) . nichol et al . ( 1999 ) and ebeling et al . ( 1999 ) report the discovery of an x - ray luminous cluster ( rxj0152 ) at z=0.83 , gioia et al . ( 1999 ) has also discovered a young x - ray luminous cluster at z=0.81 and donahue et al . ( 1998 ) reported one at z=0.82 . there is also the detection of an x - ray cluster ( rosati et al . 1999 ) at z=1.26 with a low x - ray luminosity . nearly all these clusters at z@xmath340.8 have a disturbed x - ray shape ( elongated and clumpy ) and their nature ( partially relaxed clusters or proto - clusters ) is not clear yet . the formation epoch of these massive clusters of galaxies , though , was probably only slightly greater than z=0.8 , which is consistent with our estimates of the minimum z of cluster formation . selection effects may also have played a role in finding only distorded clusters ( cf . adami et al . 2000b ) , however , and the cluster formation may occur at even higher redshifts ( cf . rosati et al . 1999 ) for ( at least ) the low mass clusters . we have produced different composite clusters based on combining the results of 6 rich optically selected clusters with redshifts near 0.4 , and 6 x - ray selected clusters with redshifts near 0.3 and 0.5 . we compared these to a similar composite cluster at redshift 0.07 . we have shown that in terms of brightness , type and radius versus velocity dispersion , the @xmath51 , @xmath56 and @xmath57 composite clusters are very similar ( at least within the statistical limitations of the data ) while the @xmath58 composite cluster is significantly different . we conclude from this work that the clusters of the sample have at least partially virialized by a redshift of 0.5 and that massive clusters ( @xmath010@xmath59 solar masses ) have formed at the very latest at z = 0.8 . our derived lower limit to the time of cluster formation epoch is consistent with work on finding massive clusters at z = 0.8 or higher . we have confirmed that the more distant clusters of our sample contain more late - type galaxies that nearby clusters , as found many times before . at the same time , however , we have shown that the dynamical state of these clusters does not seem to evolve for the redshifts lower than 0.4 . in order to explain this , to the extent that there are some bright blue galaxies in the core of the clusters , we hypothesize that these massive blue galaxies have virialized before changing morphological classification . this is not a large effect , however , as the cores of the clusters are mainly red ( cf . section 4.1 ) . we must , therefore , conclude ( at the level of statistics we have here ) that it is primarily the outer regions of the clusters that contribute the relatively large blue fractions of distant clusters . although the nearby clusters should have virialized completely , the outer portions of these clusters are not virialized and yet are dominated by red galaxies . continuous in - fall of relatively numerous , low mass galaxies such that the galaxies have evolved into early type galaxies is just consistent with the time scales involved ( 1gy for relaxation for low mass or faint galaxies and 1.5 gy for virialization ) . it is not clear what our proposed scenario predicts at greater look back times . one possibility is that early type galaxies formed first and formed the initial clusters ( this hypothesis is , however , not supported by van dokkum et al . 1999 ) . then the later type galaxies , having formed later _ in the field _ are just beginning to fall into the cluster and even the bright , massive ones have not yet fallen in . in this scenario , the cluster core will have a low blue fraction / low fraction of late type galaxies even at the bright end . this is approximately consistent with the finding of wirth et al . ( 1994 ) on cl 0016 + 16 ( albeit an x - ray luminous cluster ) or of butcher @xmath3 oemler ( 1984 ) or as suggested by lubin et al ( 1998 ) and postman et al ( 1998 ) ( see , however , section 4.1 regarding the lubin et al . 1998 and postman et al . 1998 results ) the detailed scenario we suggested to explain the results of our analysis and comparisons with a nearby optically selected sample should be considered only that , a suggestion . more elaborate theoretical work , that is beyond the scope of this paper , is necessary to develop the full range of possibilities . the statistics of the results we found here also clearly need to be improved by analyzing the data from more clusters with larger telescopes . what our discussion has showed is the power of the analysis we performed here , and how more analysis of @xmath60 systems coupled with an extension to even higher redshift systems will do much to clarify picture of how galaxy populations evolve within clusters . adami , c. , ulmer , m.p . , romer , a.k . 2000b , apj , submitted adami , c. , holden , b.p . , castander , f. , et al . 2000a , aj , submitted adami , c. , biviano , a. , @xmath3 mazure , a. 1998 , a@xmath3a , 331 , 439 beers , t.c . , flynn , k. , @xmath3 gebhardt , k. 1990 , aj , 100 , 32 biviano , a. , et al . 2000 , a@xmath3a , in preparation biviano , a. , katgert , p. , mazure , a. , et al . 1997 , a@xmath3a , 321 , 84 bower , r.g . , castander , f.j . , ellis , r.s . , couch , w.j . , @xmath3 boehringer , h. 1997 , mnras , 291 , 353 butcher , h. , @xmath3 oemler , a. 1984 , apj , 285 , 426 carlberg , r.g . , yee , h.k.c . , ellingson , e. , et al . 1997 , apj , 476 , 7 couch , w.j . , ellis , r.s . , maclaren , i. , malin , d.f . 1991 , mnras , 249 , 606 van dokkum , p.g . , franx , m. , fabricant , d. , kelson , d.d . , @xmath3 illingworth , g.d . 1999 , apj , 520 , l95 donahue , m. , voit , g.m . , gioia , i.m . , et al . 1998 , apj , 502 , 550 dressler , a. , smail , i. , et al . 1999 , apjs , 110 , 213 dressler , a. 1980 , apj , 236 , 351 ebeling , h. , jones , l.r . , perlman , e. , et al . 1999 , apj , submitted astroph : 9905321 frei , z. , @xmath3 gunn , j.e . 1994 , aj , 108 , 4 gioia , i.m . , henry , j.p . , mullis , c.r . , et al . 1999 , aj , 117 , 2608 gunn , j.e . , hoessel , j.g . , oke , j.b . 1986 , apj , 306 , 30 holden , b.p . , 2000 , aj , submitted katgert , p. , mazure , a. , perea , j. , et al . 1996 , a@xmath3a , 310 , 8 kent , s.m . , @xmath3 sargent , w.l.w . 1983 , aj , 88 , 697 lax , d. 1985 , j. am . assoc . , 80 , 736 lubin , l.m . , postman , m. , oke , j.b . , et al . 1998 , aj , 116 , 584 mazure , a. , katgert , p. , den hartog , r. , et al . 1996 , a@xmath3a , 310 , 31 melnick , j. , @xmath3 sargent , w.l.w . 1977 , apj , 215 , 401 metevier , a. , romer , a.k . , ulmer , m.p . 2000 , aj , in press nichol , r.c . , romer , a.k . , holden , b.p . , et al . 1999 , apj , 521 , 21 oemler , a.jr . 1974 , apj , 194 , 1 oukbir , j. , @xmath3 blanchard , a. 1997 , a@xmath3a , 317 , 1 oukbir , j. , @xmath3 blanchard , a. 1992 , a@xmath3a , 262 , l21 postman , m. , lubin , l.m . , oke , j.b . 1998 , aj , 116 , 560 postman , m. , lubin , l.m . , gunn , j. , et al . 1996 , aj , 11 , 615 press , w.h . , teukolsky , s.a . , vetterling , w.t . , flannery , b.p . 1992 , numerical recipes : second edition rosati , p. , stanford , s.a . , eisenhardt , p.r . , 1999 , aj , 118 , 76 sarazin , c.l . 1986 , rev . phys . , 58 , 1 smail , i. , morrison , g. , gray , m.e . , et al . 1999 , apj , 525 , 609 stein , p. 1997 , a@xmath3a , 317 , 670 ulmer , m.p . , wirth , g.d . , kowalski , m.p . 1992 , apj , 397 , 430 wang , q.d . , ulmer , m.p . , lavery , r.j . 1997 , mnras , 288 , 702 whitmore , b.c . , @xmath3 gilmore , d.m . 1991 , apj , 367 , 94 wirth , g.d . , koo , d.c . , @xmath3 kron , r.g . , 1994 , apj , 435 , l105 yee , h.k.c . , ellingson , e. , abraham , r. , et al . 1996 , apjs , 102 , 289

the problem of interacting cooper pairs and/or bosons , together with the possibility of a quantum control of long range phase coherence in low dimensional systems has received considerable interest since the experiments@xcite on homogeneous lead and bismuth films , which exhibit a transition from a superconducting to an insulating phase as a function of thickness . it is known , since the work of abrahams et al . @xcite , that no true metallic behavior can be expected to be observed for 2d non - interacting electrons at t=0 , because all the states will be localized by an arbitrarily small amount of disorder . when one includes coulomb interaction , the situation is less clear , but the common belief is that a metallic phase should still not appear at t=0 in the presence of disorder - though no rigorous proof is available@xcite . yet , in the presence of attractive interactions , one expects a superconducting state both at t=0 and finite temperature , even in presence of a finite amount of disorder , due to its relative ineffectiveness for a transition of kosterlitz - thouless type ( the degree of relevance being given by the harris criterion@xcite ) . the existence of a superconducting state at t=0 in 2d systems is then considered to be directly linked to that of an insulating state , with no intermediate metallic phase present . one hence should be able to observe a direct superconductor - insulator transition ( sit ) at zero temperature in 2d as a function of disorder , interaction strength , magnetic field or any other external parameter which can drive the system away from the superconducting phase . there are different theoretical scenarios which are commonly discussed in connection with the salient features of such quantum phase transitions : i ) dissipative models , considering a network of josephson coupled superconducting grains shunted by resistors@xcite , where the competition between the josephson intergrain coupling and the charging energy yields an increase of the phase fluctuations of the superconducting order parameter and hence leads to a phase disordered state . dissipation thereby plays the role of suppressing quantum phase fluctuations and thus competes with the charging mechanism . ii ) bose - hubbard models , where the superconducting phase is due to the bose - einstein condensation of charge-2e bosons and the insulating phase due to a proliferation of vortices and localization of pairs@xcite . while the sit was long thought to be a paradigm of the above theories , recent experiments have thrown some doubt about that . they revealed , what seems to be a low - temperature metallic state which is intertwined in such a sit and thus requires a more appropriate theoretical description . in magnetic field tuned experiments on mo - ge samples@xcite , in granular superconductors ( ga films@xcite and pb films@xcite ) , and in josephson junction arrays@xcite a metallic phase has been observed . moreover , the recent experiments of kapitulnik et al.@xcite , in which a metallic phase has been observed sandwiched between the superconductor and the insulating phase , suggest that perhaps two phase transitions accompany the loss of phase coherence in 2d superconductors : ( i ) a superconductor to a `` bose metal '' and ( ii ) a `` bose metal '' to an insulator . a `` bose metal '' in this context , is thought of just a gapless non - superfluid liquid with metallic like transport@xcite . these experimental results have led to reconsider the whole issue of the sit . one recent proposal to handle such a new viewpoint of the sit has been to reexamine the standard on - site charging model . by including nearest neighbor charging terms it was shown that the resulting uniform bose metal state lacks any trace of either phase or charge order@xcite , due to a competition between the order parameters which describe the onset of charge order and that of phase coherence . a different point of view@xcite is , that in the quantum disordered regime a cancellation happens between the exponentially long quasi - particle scattering time and the exponentially small quasi - particle population , which ultimately leads to a finite dc conductivity . finally , intrinsic as well as extrinsic sources of dissipation have been suggested as potentially relevant for the occurrence of a non - superfluid metallic phase in proximity to the superconducting phase@xcite . in this context , the dissipation is a relevant perturbation@xcite , which , depending on the strength of the coupling to the dissipative source , can drive the system from a _ superconductor - to - insulator transition _ to a _ superconductor - to - metal transition_@xcite . several other ideas supported a scenario where the system could break into superconducting and insulating `` islands '' weakly linked via percolating paths before going into a metallic phase , dominated by vortex dissipation@xcite . this `` puddle '' scenario matches with the view to describe the 2d superconductor - metal transition via superconducting islands embedded in thin metal films@xcite . detailed analysis of the puddle - like model considered strongly fluctuating superconducting grains embedded in a metallic matrix@xcite , predicting a _ metal - to - superconductor transition _ with a metallic phase just above the transition dominated by andreev reflections between the superconducting grains . finally , recent investigations directed the attention to new phases with bose metallic features , where the dissipation is dynamically self - generated , as in the quantum phase glass model , where disorder in the distribution of the tunnelling amplitudes and quantum fluctuations destroy phase coherence@xcite . on similar lines as those dealt with in a great variety of such different approaches discussed in the literature , our aim here is to investigate a system where the breakdown of superconductivity situates itself in between the case of a `` bosonic '' and a `` fermionic '' mechanism for superconductivity suppression . we focus on determining the possible ground states for interacting cooper pairs , in close relation to the classical notions of superfluidity and localization of bosons and , on a more general basis , how the phase coherence can be tuned to a phase disordered state whose transport properties may be unconventional . in particular , we consider within the framework of a boson - fermion model ( bfm ) , a system where localized bound electron pairs ( hard - core bosons ) form an array of `` andreev''-like scattering centers coupled to a fermionic subsystem of itinerant uncorrelated electrons ( quantum pair - exchange ) . this scenario goes beyond that of pure phase models widely discussed in the literature , since here one is dealing with bosonic degrees of freedom ( localized bound pairs ) as well as fermionic ones ( itinerant electrons ) for which , due to the emergence of pair correlations , one is dealing with both amplitude as well as phase modes . the possibility to tune from short to long - range phase coherence arises from the following competing effects : \(i ) on the one hand , the short range interaction between bosons and fermion pairs ( holes ) induces a local phase locking in a configuration with a quantum superposition between bosons and electron pairs , leaving the common phase undetermined . \(ii ) on the other hand , the itinerancy of the electrons tends to lock and rigidly extend these initially arbitrary local phases . as a result , phase coherence develops over longer distances by suppressing the quantum fluctuations of the local phase , which will involve the dynamics of amplitude fluctuations . out of this competition one can recover either a superfluid state in the regime of a small scattering rate or a phase disordered state in the limit where the pair exchange dominates and the local quantum phase fluctuations do not allow for long - range phase coherence . as we shall show , the physics described by this scenario strongly depends on the concentration of fermions and bosons , on the coupling strength , as well as on temperature and magnetic fields . the purpose of this study is to analyze how a transition from a superconductor to an insulator may occur and discuss the features which can give rise to unconventional dynamics and , eventually , unconventional transport properties in the proximity of such a transition . the outline of the paper is as follows . in sect . ii we will introduce the bfm and describe its main qualitative features and the phenomena which can be described by it . in sect . iii , we shall develop a path integral representation of this model and derive an effective coarse grained action , which , after integrating out the fermionic degrees of freedom , is able to describe the low energy dynamics of the phase and amplitude modes . in sect . iv we will discuss the phase diagram based on the simplest approximation of such a coarse grained effective action in terms of a phase - only action and explore the transition from a superconductor to a phase disordered state as a function of the coupling and the density of bosons . in sect . v , we will study the physical features which arise from the intrinsic berry phase term present in our effective action and which arises from the hard core nature of the bosons , represented by quantum pseudo spin-@xmath0 variables . in section vi we compare the salient features of this bfm scenario with similar scenarios , such as the negative - u hubbard model , the bose - hubbard model and josephson junction arrays . in the discussion , section vii , we review the main results obtained in this paper and indicate further developments planed for the near future . the boson - fermion model ( bfm ) in recent years has attracted considerable attention as a model capable of capturing basic physical properties in many body systems with strong interaction , giving rise to the formation of resonant pair states of bosonic nature inside a reservoir of fermions . such a scenario was initially proposed by one of us ( jr ) as an alternative to the scenario of the hypothetical and yet to be experimentally verified _ bipolaronic superconductivity_. it was meant to describe the intermediary coupling regime between the adiabatic and anti - adiabatic limits in polaronic systems , where an exchange between localized bipolarons and pairs of uncorrelated electrons can be assumed to take place ( for a recent intuitive justification of such a scenario see for instance ref.@xcite ) . as it has turned out , this boson - fermion scenario has a much wider range of applicability than that for which it was initially proposed and seems to apply to very different physical situation such as : hole pairing in semiconductors@xcite , isospin singlet pairing in nuclear matter@xcite , d - wave hole and antiferromagnetic triplet pairing in the positive - u hubbard model@xcite ( and possibly also in the t - j model ) , entangled atoms in squeezed states in molecular bose einstein condensation in traps@xcite and superfluidity in ultracold fermi gases induced by a feshbach resonance@xcite . the bfm is reminiscent of an anisotropic kondo lattice model in terms of a pseudospin-@xmath0 , but characterizing localized electron pairs instead of localized impurity spins as in the kondo analogue . its hamiltonian is given by @xmath1 the pseudo - spin operators @xmath2 $ ] denote the local bound electron pairs ( bosons ) and @xmath3 $ ] the itinerant pairs of uncorrelated electrons . @xmath4 $ ] stand for the creation and annihilation operators of the itinerant electrons ( fermions ) and @xmath5 is the strength of the @xmath6 exchange interaction . the hopping integral for the itinerant electrons , which is assumed to be different from zero only for nearest neighbor sites , is given by @xmath7 with a band half width equal to @xmath8 , @xmath9 denoting the coordination number of the underlying lattice . the energy level of the bound electron pairs is denoted by @xmath10 . the number of the ensemble of bosons and fermions being conserved , @xmath11 , implies a common chemical potential @xmath12 for both subsystems . @xmath13 , @xmath14 indicate the occupation number per site of the hard core - bosons and of the electrons with up and down spin states . the exchange coupling between the bosons and the fermion pairs can be considered as an effective andreev - like scattering leading to local states which are quantum superpositions of the form @xmath15\times \nonumber \\ & & [ \cos(\phi_i ) + \tau_{i}^{+}\sin(\theta_i/2)\sin(\phi_i ) ] |0\rangle .\end{aligned}\ ] ] such states evolve gradually out of the system of localized dephased bosons and essentially uncorrelated free fermions , which characterize the high temperature phase of this model , when the temperature is decreased below a certain @xmath16 where resonant pairing ( not bound pairs ! ) starts to be induced in the fermionic subsystem . these pair states have already features built in which are reminiscent of those which characterize cooper pairing of fermions as well as superfluidity of bosons . the phases of the two coherent states , corresponding to the two subsystems , are the same and hence locked together , but are averaged over all angles as a consequence of the conserved particle number on any given site , @xmath17 . roughly speaking , the ground state of the system is then given by a product state @xmath18 over a finite length and time scale . in this way the localized bosons and fermion pairs acquire itinerancy@xcite which eventually leads to a superfluid state in both subsystems@xcite , provided that the effect of the local correlations between the bosons and the fermion pairs can be sufficiently diminished , but remaining still sufficiently strong to guarantee the formation of pairing in the fermionic subsystem . achieving or not this situation will depend on the relative importance of the local exchange coupling versus the fermion hopping rate ( given by the ratio @xmath19 ) and as well as on the concentration of the bosons , as shown by exact diagonalization study@xcite on finite clusters of this bfm . what we shall attempt in the present study is to describe this physics in terms of an effective action for the phase and amplitude fields of the bosonic fields . in order to achieve this we shall put the discussion on a level which is more familiar , namely that one of josephson junction arrays and bose - hubbard models . for that purpose let us briefly sketch the analogy which exists ( up to a certain point ) between the bfm and those systems which have been widely discussed in the literature . a physically possible realization of this bfm scenario can be imagined in form of a network of superconducting grains embedded in a metallic environment and where the only mechanism of interaction between the grains and the fermionic background is that of andreev reflections . via such a mechanism an electron ( hole ) is reflected on the grain as a hole ( electron ) leaving behind a surplus of two holes ( electrons ) in the fermionic subsystem and of two electrons ( holes ) in the grain . if the grains are such that they have a large charging energy , the fluctuations of the number of pairs on them are energetically unfavorable and hence are largely suppressed . we then have a situation where the state of the grains switches essentially between zero to double occupancy with respect to the average occupation , any time an electron ( hole ) is reflected at the interface of the grain . thus , the quantum dynamics of the single grain can be directly represented by a pseudospin@xmath0 , in order to account for the doublets which represent the two possible states of the grains . for such a possible experimental setup , the effective sites in the bfm have to be considered as defining a regular array of grains and having the same periodicity as the underlying lattice on which the fermions move with a hopping amplitude @xmath7 . moreover the size of the grains has to be such that it is much smaller than the distance between them . a pictorial view of such an experimental set up in given in fig . [ fig2-clustconf ] . the analogy between the bfm and the array of superconducting grains which scatter pairs of fermions in a metallic matrix via andreev like reflections , may ultimately serve as an experimental device on which to test and analyze the theoretical issues which will be discussed in this paper . let us now construct an effective action which describes the bfm , with the aim to extract the dynamical properties of the low energy degrees of freedom of the phase and amplitude modes for the bosons and fermion - pairs . we start by expressing the partition function in terms of a coherent - state path integral representation@xcite , where the fermionic part is formulated by means of the usual grassmann variables and the bosonic part is described by a pseudospin - coherent state representation@xcite : @xmath20}\end{aligned}\ ] ] where @xmath21=\int d\tau \sum_i [ i s ( 1-\cos\theta_i)~\partial_{\tau } \phi_i + \nonumber\\ ( \delta_b -2 \mu ) \cos\theta_i]+ \sum_{\langle ij \rangle } \bar{\psi}_j(\tau ) g_{ij}^{-1 } { \psi}_i(\tau).\end{aligned}\ ] ] @xmath22 denotes the imaginary matsubara time variable and a nambu spinor representation for the grassmann variables , related to that of the original fermionic operators by @xmath23 the pseudospin is described by a bosonic field which in spherical coordinates is given by @xmath24 ( see fig . [ fig1-sphere ] ) . @xmath25 describes the polar angle of the vector @xmath26 with respect to the north pole of the @xmath9 axis , while @xmath27 is the azimuthal angle which defines the angular position of the basal plane projection of this vector . the first term of the action @xmath28 is the wess - zumino term@xcite , ensuring the correct quantization of the quantum pseudospin variable . for any path , parameterized by @xmath29 and @xmath30 , the contribution of this term is equal to @xmath31 times the surface area of the sphere between this path and the north pole . for closed paths this has exactly the form of the berry phase @xcite . the second term is linked to the density of the bosons @xmath32 through the @xmath33 dependence of the pseudospin . finally , the last contribution of the action contains the coupling between the fermionic and bosonic subsystem through the green s function @xmath34 , determined by : @xmath35 where @xmath36 , @xmath37 and @xmath38 and @xmath39 being the conjugate field of @xmath40 . integrating out the fermionic part , one obtains the action in terms of exclusively the bosonic fields : @xmath41\end{aligned}\ ] ] where the trace has to be carried out over all internal as well as space - time indices . up to this point no approximation has been made in the derivation of the action which describes the coupling between the bosonic and fermionic degrees of freedom . it is important to notice that the variation of the bosonic variable @xmath30 describes both , the density fluctuations of the bosonic subsystem ( via the projection of the pseudospin vector on the @xmath9 axis , i.e. @xmath42 , being the longitudinal component ) and the amplitude fluctuations of the fermionic pair field ( via the transverse part as projection of the pseudospin vector onto the basal plane , i.e. @xmath43 ) ( see fig . [ fig1-sphere ] ) ) . the variable @xmath29 determines the rotational degrees of freedom of the pseudospin vector , expressing its phase dynamics . for extracting the relevant terms which control the low energy dynamics of the coupled phase and amplitude modes and performing an expansion , which is meaningful in terms of the phase variable , it is judicious to make the following steps : i ) gauge away the phase dependence from the term @xmath40 which permits to separate the trace into a part which does not depend on the phase of the bosonic field and another part that contains only spatial and time variations of @xmath44@xcite , ii ) rewrite the term @xmath40 after the gauge transformation , as a sum of two pieces , one not dependent on time ( which is linked to the average density of bosons ) and another term containing the fluctuations with respect to its mean value . the first operation is performed by applying to the operators under the trace the rotation @xmath45 , where @xmath46 denote the pauli matrices . hence , @xmath47 where @xmath48 \sigma_3+g~\sin\theta_i(\tau ) ~\sigma_1\right]\delta_{ij } + \\ & & t_{ij } e^{i\,[\phi_i(\tau)-\phi_j(\tau ) ] \sigma_3/2}.\end{aligned}\ ] ] since the bosonic density is fixed in average , we now separate the part which depends on the polar angle in a time independent contribution and its time dependent correction . that is , the term @xmath49 is decomposed into its average value @xmath50 ( which is determined by fixing the density of bosons due to the spherical constraint ) plus a time dependent contribution @xmath51 $ ] which contains the fluctuation around its average value ( fig . [ fig2-sphere ] ) . thus this local field , due to the constraint , will describe both : the time dependent variation of the density as well as of the pairing amplitude . next , let us write the green s function in the usual form as @xmath52 with @xmath53\delta_{ij}$ ] and @xmath54 , where @xmath55 \sigma_1\\ t_{ij}&=&t_{ij } e^ { -i ( \phi_i(\tau)-\phi_j(\tau ) ) \frac{\sigma_3}{2}}\\ \tilde{g}&=&g \langle\sin[\theta]\rangle.\end{aligned}\ ] ] from this point onward we shall assume the average density of bosons to be homogeneously distributed and thus given by @xmath56 . this implies that the bare coupling @xmath5 is renormalized to @xmath57 as a consequence of the spherical constraint of the pseudospin variable . moreover , since @xmath58 and the density of bosons is fixed in average via a suitable choice of the chemical potential and of the on - site bosonic energy , one can treat the exchange coupling as an external boson - density tunable parameter . the variation in the @xmath59 variable is then simply related to the variation of the bosonic density such that if @xmath59 varies in the range @xmath60 $ ] then @xmath13 varies in the interval @xmath61 $ ] . before expanding the trace , let us write down explicitly the expression of the zero order green s function , as it will be frequently used in the following steps : @xmath62\end{aligned}\ ] ] with @xmath63 and where we introduced @xmath64 . we now evaluate the contribution of the self energy @xmath65 to the effective action . this is done in the usual way by making a loop expansion in the trace@xcite . we shall construct that expansion up to second order in the time and space derivatives of the phase variable and in the terms which contain both , the fluctuations of the density and the amplitude . for that purpose we use the standard identity : @xmath66\\ & = & tr \ln { g_0}^{-1}+tr \ln \left [ 1 + { g_0 } \sigma\right]\end{aligned}\ ] ] and then expand the second term of this expression up to second order in @xmath67 , such as to keep all the contributions up to quadratic order in the gradient of the phase . this gives @xmath68 -\frac{1}{2 } tr [ { g_0 } \sigma]^2 \,.\end{aligned}\ ] ] the first term of this expression is just a constant and does not contribute to the dynamics . in the second term @xmath69=tr [ { g_0 } t]+tr [ { g_0 } d]+tr [ { g_0 } k]\end{aligned}\ ] ] the only parts different from zero are @xmath70 $ ] and @xmath71 $ ] . @xmath72 $ ] gives no contribution once one makes the trace over the site indices . @xmath71 $ ] introduces a contribution which is proportional to the chemical potential multiplied by the time derivative of the phase @xmath73 . we will see below that this contribution describes an effective _ off - set charge _ ( in terms of the terminology of a similar josephson junction array scenario ) and which arises from the total fermionic and bosonic static density distribution via their dependence on the chemical potential . @xmath74 $ ] describes the lowest order fluctuations of the bosonic ( fermion pair ) density due to the presence of the spontaneous pair / hole creation out of the condensate . its direct evaluation gives a contribution @xmath75\end{aligned}\ ] ] where @xmath76 represents the trace over the internal pseudospin index . with @xmath77 and integrating over the matsubara times gives : @xmath78 \\ \nonumber & & = 2 g { g_0}_i^{12}(0 ) \int d\tau l[\theta(\tau)]\\&&= -\frac{g \tanh[\beta \omega_0]}{\omega_0 } \int d\tau l[\theta(\tau)],\end{aligned}\ ] ] or in a compact form @xmath79 \nonumber \\ e_1&&=-\frac{g \tanh[\beta \omega_0]}{\omega_0}.\end{aligned}\ ] ] let us next come to the evaluation of the terms which contribute to the quadratic order in this loop expansion of the trace . the parts which are non zero in these terms are the following : @xmath80 \nonumber \\ a_3&=&tr~[{g_0}_i~k_i~{g_0}_i~k_i ] \nonumber \\ a_4&=&tr~[{g_0}_i~d_i~{g_0}_i~d_i ] \nonumber \\ a_5&=&tr~[{g_0}_i~t_{ij}~{g_0}_j~t_{ji}]\end{aligned}\ ] ] the term @xmath81 contains processes where the fermionic background locally couples at different times to the fluctuations of the bosonic density and to the phase fluctuations : @xmath82 = \qquad \\ & & \int d\tau_1d\tau_2 tr_{ps}[g_{0\,i}(\tau_1-\tau_2 ) k_i(\tau_2 ) g_{0\,i}(\tau_2-\tau_1)d_i(\tau_1 ) ] . \nonumber \\\end{aligned}\ ] ] since @xmath83 depends exclusively on the time differences , we introduce the new variables @xmath84 and @xmath85 , after which the integral over the matsubara time variables becomes : @xmath86.\ ] ] as we are interested in the gradient expansion in the bosonic phase and density , related to @xmath29 and @xmath30 , we keep only the lowest order in their time derivatives . this is done by considering the expansion of the product @xmath87 around @xmath88 in order to separate the parts which depend exclusively on the local variable of the green function and those which are linked to the phase and the density fluctuations . the expansion then reads as follows , @xmath89 due to the symmetry of the green s functions ( even for @xmath90 ) , the linear contribution of the series expansion of @xmath87 cancels in the effective action after integrating over the time . hence , to lowest order one obtains , @xmath91 \nonumber \\ & = & \frac{g}{2 } \int d\tau \left[{g_0}_i^{12}(\tau ) { g_0}_i^{11}(-\tau ) + { g_0}_i^{11}(\tau ) { g_0}_i^{12}(-\tau ) \right.\nonumber \\ & - & \left . { g_0}_i^{22}(\tau ) { g_0}_i^{12}(-\tau ) - { g_0}_i^{12}(\tau ) { g_0}_i^{22}(-\tau)\right]\times \nonumber \\ & & \int d\eta \frac{i}{2 } \frac{\partial \phi(\eta)}{\partial \eta } l[\theta(\eta)]\end{aligned}\ ] ] or in a compact form , @xmath92\ ] ] where the coefficient @xmath93 , after integrating over the matsubara time , is given by : @xmath94 } ) } { \omega_0 ^ 3(1+\cosh{[\beta~\omega_0])}}.\ ] ] let us next consider the term @xmath95 , which expresses the coupling of the fermionic field to the fluctuations of bosonic density at different times . in order to extract the lowest order gradient contributions , we follow the same procedure as that just used above , giving us : @xmath96\ ] ] performing again the expansion in time up to quadratic order in the time derivatives of @xmath97 , we have : @xmath98 and hence can rewrite @xmath95 in the following way : @xmath99 + \nonumber \\ & & \frac{1}{2 } \int d\tau~d\eta~ tr_{ps}[{g_0}_i(\tau ) k^{'}_i(\eta ) { g_0}_i(-\tau ) k_i^{'}(\eta)].\end{aligned}\ ] ] carrying out the trace over the internal indices gives : @xmath100=[{g_0}_i^{22}(\tau ) { g_0}_i^{11}(-\tau ) \nonumber \\ + 2~{g_0}_i^{12}(\tau ) { g_0}_i^{12}(-\tau)+{g_0}_i^{11}(\tau ) { g_0}_i^{22}(-\tau)]~(\tilde{g}^2 l[\theta(\eta)]^2 ) \nonumber \\\end{aligned}\ ] ] so that @xmath101 ^ 2- e_{3b } \int ~d\eta \frac{\partial ~l[\theta(\eta)]}{\partial \eta}^2.\ ] ] by evaluating the integrals over the matsubara times we obtain : @xmath102}{\omega_0 ^ 3 ( 1+\cosh[\beta~\omega_0 ] ) } \nonumber \\ e_{3b}&&=\frac{\tilde{g}^2}{2 } [ \frac{sech[(\beta \omega_0/2)]^2 ( -6 \beta \mu^2 \omega_0+\beta^3 ( -\mu^2 \omega_0 ^ 3 + \omega_0 ^ 5)}{24 \omega_0 ^ 5 } \nonumber \\ & & + \frac{6\mu^2 \sinh[\beta \omega_0])}{24 \omega_0 ^ 5}].\end{aligned}\ ] ] applying the same procedure for the evaluation of @xmath103 ( which has the same functional form as @xmath95 but involving the coupling between the fermionic degrees of freedom and the phase velocity at different time ) one gets an analogous expression , providing we discard terms of higher order in the derivatives than @xmath104 ^ 2 $ ] : @xmath105 \nonumber \\ & & = \frac{1}{8 } c_0 \int ~d\eta ( \frac{\partial \phi(\eta)}{\partial \eta})^2.\end{aligned}\ ] ] here @xmath106 \}~,\end{aligned}\ ] ] which after evaluating the integration over the matsubara times gives , @xmath107 ^ 2 ( \beta \mu^2 \omega_0+\tilde{g}^2 \sinh[\beta \omega_0])}{2 \omega_0 ^ 3}. \label{selfcapac}\ ] ] finally , we come to the evaluation of the last term @xmath108 which involves the coupling between phase fluctuations on different sites and the process of single particle hopping . this contribution will yield terms which are quadratic in the time derivative of the phase ( _ charging like _ , in the terminology of a similar josepson junction array scenario ) and moreover will generate an effective hopping induced inter - site phase coupling which turns out to be similar to the josepshon coupling in arrays of superconducting grains . considering again the lowest order gradient expansion contributions we have : @xmath109\\\nonumber & & \simeq \frac{1}{2 } \int d\tau~d\eta tr_{ps } \left\ { { g_0}_i(\tau ) t_{ij}(\eta ) { g_0}_j(-\tau ) t_{ji}(\eta)-\right.\\ \nonumber & & \left . \tau^2 \left[{g_0}_i(\tau ) \frac{\partial t_{ij}(\eta)}{\partial \eta } { g_0}_j(-\tau ) \frac{\partial t_{ij}(\eta)}{\partial \eta}\right]\right\}\\ & & = p_1+p_2\end{aligned}\ ] ] the terms @xmath110 are given by : @xmath111~\int d\eta \cos[\phi_i(\eta)-\phi_j(\eta ) ] \nonumber\\ & p_2&=-\frac{t^2_{ij}}{8 } \int d\tau~ \tau^2~[g_{0\,i}^{11}(\tau ) g_{0\,j}^{11}(-\tau)+g_{0\,i}^{22}(\tau ) g_{0\,j}^{22}(-\tau)]~\nonumber\\ & & \int d\eta [ \frac{\partial \phi_i(\eta)}{\partial \eta}-\frac{\partial \phi_j(\eta)}{\partial \eta}]^2.\end{aligned}\ ] ] by carrying out the integration over the matsubara times one obtains for the hopping induced inter - site amplitude and phase coupling : @xmath112 \nonumber \\ & e_{j}&=\frac{t^2_{ij}~\tilde{g}^2 \left ( -\beta \omega_0 sech[\beta \omega_0/2]^2 + 2 \tanh[\beta \omega_0/2 ] \right)}{8 \omega_0 ^ 3}.\nonumber \\ \label{joseph}\end{aligned}\ ] ] similarly , by calculating the coefficient of the term @xmath113 , one obtains the strength of a _ mutual capacitance _ term ( in the terminology of a similar josephson junction array scenario ) for neighboring effective sites : @xmath114 ^ 2 \\ \nonumber c_{1}&&=-t^2_{ij } \int_{-\beta/2}^{\beta/2 } d\tau~ \tau^2~[g_i^{11}(\tau ) g_i^{11}(-\tau)+g_i^{22}(\tau ) g_i^{22}(-\tau)]\\ \nonumber & & = -{t^2_{ij } } \left [ \frac{\exp[\beta \omega_0 ] ( -6 \beta \tilde{g}^2 \omega_0 + \beta^3 \omega_0 ^ 3 ( \mu^2+\omega_0 ^ 2)}{12 ( 1+\exp[\beta \omega_0])^2 \omega_0 ^ 5 } \right.\\ & & + \left . \frac{6 \tilde{g}^2 \sinh[\beta \omega_0])}{12 ( 1+\exp[\beta \omega_0])^2 \omega_0 ^ 5 } \right ] . \label{mutualcap}\end{aligned}\ ] ] the final effective action is then given by the sum of all the terms evaluated above which are grouped together in form of three different contributions : @xmath115 . \ , \label{action}\end{aligned}\ ] ] @xmath116 , @xmath117 , @xmath118 are the contributions arising from exclusively ( i ) the phase dynamics , ( ii ) the fluctuations of the bosonic density and of the amplitude @xmath119 $ ] and ( iii ) the coupling between them . they are given by : @xmath120\right ] \nonumber \\ & + & i \sum_{i}\mu \frac{\partial \phi_(\tau)}{\partial \tau}\nonumber \\ s_{\theta}&=&\sum_i \left[e_1~l[\theta_i(\tau ) ] + e_{3a } l[\theta_i(\tau)]^2-e_{3b } [ \frac{\partial ~l[\theta_i(\tau)]}{\partial \tau}]^2\right ] \nonumber \\ \nonumber s_{\phi-\theta}&=&\sum_i\left [ i~s\,\frac{\partial \phi_{i}(\tau)}{\partial \tau}~(1-\cos[\theta_i(\tau ) ] ) \right . \\ & + & \left.i~e_2\frac{\partial \phi_i(\tau)}{\partial \tau}~ l[\theta_i(\tau ) ] \right ] . \label{action1}\end{aligned}\ ] ] we expand the berry phase contribution ( the first term in @xmath118 ) up to second order in @xmath51 $ ] and subsequently redefine this field as @xmath121 = a + b \bar{l}[\theta(\tau ) ] $ ] , with the time independent constants @xmath122 chosen in such a way as to eliminate any terms linear in @xmath123 in the the action @xmath124 . we then find two contributions to the action which are linear in @xmath125 : one which is time independent and which can be absorbed into the chemical potential and another one which is quadratic in @xmath126 $ ] . with this , @xmath118 can be rewritten as : @xmath127 where @xmath128 is quadratic in @xmath126 $ ] . the effective action thus constructed for the bfm ( eqs . ( [ action],[action1 ] ) ) is similar to that of josephson junction arrays with nearest neighbor , as well as , self _ capacitance _ , _ josephson _ coupling and _ off - set charge _ terms . the action for the bfm goes however beyond that for such josephson junction arrays in the following respects . we have extra terms which control the dynamics of the amplitude modes , given by @xmath117 and an intrinsic berry phase term which gives rise to a direct phase - amplitude coupling ( eq . [ action2 ] ) , where the dynamical amplitude fluctuations would correspond to a time dependent _ offset charge _ term in an analogous josephson junction array picture . finally , the berry phase term , being an intrinsic topological term , may give rise to a magnus force on a vortex , as it will be discussed in section v. in a first approach to analyze the stability region of long - range phase superconducting coherence , we examine the effective action ( eqs . [ action],[action1 ] ) , restricting ourselves to the phase only part of it . this means a study of the sit driven by a competition between the phase coherence induced by pair hopping and the disrupting effect of local boson density ( or equivalently pair field amplitude ) fluctuations in the presence of a source term for the bosons which controls the global boson density via an effective chemical potential . within such an approximation our study is equivalent to that of josephson junction arrays , that the effective coupling constants entering in such an action depend in a highly non trivial way on the parameters which characterize the original bfm hamiltonian . this , as we shall see , will lead to novel features concerning the phase diagram with a sit for the bfm when we examine it in terms of the boson - fermion exchange coupling @xmath19 and the boson concentration @xmath13 . we shall determine the phase diagram by means of the so - called coarse graining approximation which has been successfully applied for this kind of problem and which permits to capture the relevant qualitative and quantitative features of such a josephson junction array like action@xcite . it is known that , as a consequence of the uncertainty relation between the phase @xmath27 and the pair number operator @xmath129 , the system can switch from a phase ordered to a disordered state . an essential part of this study will concern how the relevant parameters of the bfm hamiltonian influence the equivalent amplitudes of the josephson coupling , the _ capacitance _ and the _ off - set charge _ terms . thus , at zero temperature , by fixing the bosonic density distribution , we find that the variation of the ratio between the josephson coupling energy @xmath130 and the _ charging _ energy @xmath131 , which controls the phase - density interplay , increases from zero , goes through a maximum and then decreases to zero with increasing @xmath19 . or else , if one fixes the coupling @xmath19 , then by varying the density of the bosonic distribution one is able to control the effective coupling which appears in @xmath130 and @xmath132 bt varying the average bosonic density @xmath133 ) . it is thus immediately evident that in the bfm scenario there is a non - trivial interplay between the renormalization of the josephson coupling and the charging effect . if one goes to the limit of empty ( @xmath134 ) and full bosonic occupation ( @xmath135 ) , the effective coupling @xmath136 , and the critical temperature consequently reduces to zero . the general form of the action we then have to examine is given by : @xmath137 + % \\ & & % \sum_{i , j}\frac{1}{8}\frac{\partial \phi_i(\tau)}{\partial \tau } c_{ij } \frac{\partial \phi_j(\tau)}{\partial \tau}.\end{aligned}\ ] ] the first term describes the effect of a static _ off - set charge _ ( @xmath138 ) the second term contains the physics of the pair hopping processes , an analog to the josephson tunnelling processes , and has a coupling strength @xmath130 with @xmath139 if @xmath140 are nearest neighbors and zero otherwise . finally , the third term describes the _ charging _ term arising from the local exchange of boson and fermion pairs and from quasi - particle hopping between nearest neighbor sites . the strength of this _ charging _ type interaction is given by @xmath141 which represents an effective general _ capacitance _ matrix , with the vector @xmath142 running over the nearest neighbors . @xmath143 denotes the _ self - capacitance _ and @xmath144 the mutual one ( @xmath9 being the coordination number ) . as mentioned before , to extract the phase diagram we make use of the coarse graining approximation@xcite . the main idea of this approach is to introduce a hubbard - stratonovich auxiliary field which is conjugate to the average of @xmath145 and which plays the role of an order parameter for the transition from a superconducting to an insulating state . since the phase transition has a continuous character , one can expand the action in powers of the auxiliary field and determine the occurrence of phase coherence by looking at the coefficients of the quadratic - term in the limit of long - wavelengths and zero frequency . we briefly sketch the main steps of such an approximation and adapt it to the present scenario of the bfm . the partition function for @xmath146 is given by the sum of all the possible paths of the phase variables in the imaginary time and in the real space : @xmath147\right.\nonumber\\ % capacitance \nonumber & & -\left.\frac{1}{8 } \sum_{ij } \frac{\partial \phi_i(\tau)}{\partial \tau } c_{ij } \frac{\partial \phi_j(\tau)}{\partial \tau } \right\}\end{aligned}\ ] ] to perform the hubbard - stratonovich transformation , one rewrites the _ josephson coupling _ term as @xmath148 \alpha_{ij } \exp[i \phi_j]$ ] and then , by using the usual gaussian identity , introduces an auxiliary field @xmath149 . the partition function @xmath150 then becomes : @xmath151\right\}.\end{aligned}\ ] ] hence , starting from the effective action for the auxiliary field , one can perform an expansion up to second order in @xmath152 and thus derive the usual ginzburg landau type free energy functional which permits to determine the boundary line between the superconducting and the insulating state . the corresponding effective action for these auxiliary fields @xmath153 @xmath154\end{aligned}\ ] ] is then expanded to second order , giving : @xmath155 where @xmath156 } \rangle_0\ ] ] denotes the two - time phase correlator , which is equal to the second derivative with respect to the auxiliary field @xmath152 and its conjugate at different time and space positions , evaluated around their zero values . by performing the functional derivation , one gets the following expression for it : @xmath157 } \exp[-s_0]}{\int \prod_i d\phi_i \exp[-s_0]}\end{aligned}\ ] ] with @xmath158 being the part of the action which contains exclusively the _ charging _ contributions , i.e. : @xmath159.\end{aligned}\ ] ] according to the scheme outlined above , we now develop the partition function @xmath150 up to second order in the auxiliary fields , thus putting it into a familiar form : @xmath160 with @xmath161 \psi_i(\tau^{'}).\ ] ] the determination of the conditions for the boundary line between the superconducting and the insulating phase then reduces to the explicit evaluation of @xmath162 . the result for that has been first obtained in ref . [ 48 ] , and we sketch below the main steps of this derivation . in determining @xmath162 it has turned out to be essential to treat the phase variable @xmath27 in a compact form . in order to separate the imaginary time evolution of the phase into a periodic part @xmath163 and an non - periodic part , one introduces the following parameterization : @xmath164 with @xmath165 and @xmath166 being an integer which counts how many times the phase winds over an angle which is a multiple of @xmath167 . the use of such a relation allows to express the sum over all @xmath27 as an integration over @xmath168 plus a sum over all the possible integer values of the winding number @xmath166 . after performing a number of suitable algebraic operations@xcite , one ends up with the following expression for the two - time phase correlator : @xmath169 } { \sum_{\{n_i\ } } \exp\left [ -\sum_{ij } 2\beta c_{ij}^{-1 } n_i n_j \right]}\times \nonumber \\ & & \sum_{\{n_i\ } } \exp \left [ -\sum_{ij } 2\beta c_{ij}^{-1 } n_i n_j - \sum_{k } 4 c_{ik}^{-1 } n_k \tau \right],\end{aligned}\ ] ] where @xmath170 . this two - time phase correlator is local in space and its time dependence follows an exponential behavior , if one assumes that the static _ offset charge _ @xmath138 has a distribution which is homogeneous in space . in the general case , of a dynamic _ offset charge _ @xmath171 , this time dependence will be modified and can lead to qualitative changes in the nature of the sit . having obtained the expression for the local two - time phase correlator in the time and space representation , we can now express the effective ginzburg - landau free energy functional in the fourier space in the following way : @xmath172 \psi_{k}(\omega_n ) \right]\ , .\end{aligned}\ ] ] expanding the inverse matrix @xmath173 in the form @xmath174 , this free - energy functional finally is written as : @xmath175 the transition line is then given by the condition that the coefficient of the quadratic term vanishes in the limit of vanishing @xmath176 and @xmath177 , that is : @xmath178 for a quantitative analysis , one has to determine the explicit zero frequency limit of the two - time phase correlator . as mentioned above , the inclusion of the time dependent _ offset - charge _ coming from the fluctuations of the bosonic density can modify the low frequency behavior of the local phase correlations . in order to get a first insight into the underlying physics at play here we evaluate the two time phase correlator in the limit of a purely local _ capacitance _ ( the so called _ self - charging _ limit ) , by keeping only the on - site part in the original structure of the _ capacitance _ matrix . under those conditions the evaluation of the two time phase correlator at finite frequency gives : @xmath179 \left ( \frac{1}{c_{ii}^{-1}}-\left [ i \omega_n -4 c_{ii}^{-1 } n_i \right ] \right)\nonumber \\\end{aligned}\ ] ] with @xmath180=\exp[-\sum_i 2 \beta c_{ii}^{-1 } n_i^2]$ ] and @xmath181 $ ] . with these expressions we finally can cast eq . [ boundary ] in the form @xmath182}{1 - 4 n^2}}{\sum_{m } \exp[-4 e_c \beta m^2]}=0 \,,\end{aligned}\ ] ] expressed in terms of the _ josephson coupling _ @xmath130 and the _ charging energy _ @xmath183 , as given by the eqs.([selfcapac],[joseph],[mutualcap ] ) . given this defining equation for the boundary between the superconducting and the insulating phase , we want to see now how the intrinsic dependence of the josephson and charging energy on the exchange coupling and the density of bosons , manifests itself in the competition between phase and boson density degrees of freedom . in the loop expansion given in the previous section for the bfm , we have obtained the amplitude of the intersite phase coupling and the _ charging _ effect as a function of temperature , the ratio @xmath19 and the bosonic density . it is the non trivial dependence of those coupling constants on the parameters of the original microscopic hamiltonian of the bfm which leads to the interesting features in the competition between the phase and boson density fluctuations . thus , upon increasing the pair exchange coupling it is expected that the _ josephson _ coupling is reduced while the _ charging _ energy is increased due to the local transfer between fermions pairs and bosons as well as due to the single particle hopping of fermions between nearest neighbor sites . we determine the critical line for two different cases in order to highlight the role played by , on the one hand , the coupling and , on the other hand , by the effective boson density . in fig . [ fig4-phasediagpi1 ] , we illustrate the transition line @xmath184 , separating a phase coherent state from a phase disordered one , as a function of the coupling strength and the effective boson density @xmath13 . the evolution of the transition line is non monotonic as a function of @xmath19 and goes through a maximum at @xmath185 . the quantum critical point , where the sit occurs , is given by @xmath186 . the critical behavior close to the transition is that of an xy model in d+1 dimensions . more interesting still is the behavior of @xmath184 as a function of the @xmath187 . the variation of @xmath184 is qualitatively different for the different parameter regimes : a ) the weak coupling case for @xmath188 , b ) the intermediate one with @xmath189 and c ) the strong coupling limit for @xmath190 ( see fig . [ fig5-phasediagpi1 ] ) . going from the limit @xmath191 to @xmath192 we find that with increasing @xmath13 the critical line decreases to zero , goes through a maximum starting at a finite @xmath193 and finally shows a sit at a critical density . in the preceding section we have analyzed the basic physics resulting from the effective action by neglecting , ( i ) the influence of any feedback between the density and amplitude fluctuations ( included in the field @xmath194 $ ] ) on the phase dynamics , and ( ii ) the contribution from the berry phase term , responsible for the correct quantization of the pseudospin variable , which is given by the integral over all the possible paths of @xmath195 ) \partial_{\tau } \phi_i$ ] . in this section , we discuss the consequences of the presence of such a berry phase term in the case where the phase action has a vortex solution . the existence of a vortex solution is assured for 2d systems where the phase correlations are described by a xy type dynamics . we shall show here that the berry phase term will produce an intrinsic magnus force on the vortex which is analogous to the lorentz force for a charged particle , whose effective magnetic field depends on the spatial distribution of the @xmath59 field . using the correspondence with the bosonic pseudospin variable , this implies a relation with the spatial density distribution of the bosons . the magnus force has been widely discussed in the context of normal bcs type superconductors where it has been shown that it arises from the berry phase caused by the adiabatic motion of a vortex along a closed loop coming back to its starting position@xcite . the adiabatic vortex motion on a loop in the superconducting state turns out to be affected by an effective magnetic field generated by the supercurrent arising from the gradient of the phase which encircles such a vortex . in the bfm scenario discussed here , we find that on top of the usual contribution due to the superfluid electrons , there is an intrinsic berry phase term which will generate such a magnus force and which has an intensity proportional to @xmath196 , where @xmath13 is the bosonic density . this magnus force , arising solely from the quantum nature of the pseudospin variable , is clearly independent on any superconducting state , and does not require a coherent superfluid current induced by the presence of the vortex itself . moreover , since @xmath13 can be controlled externally , one has the possibility to tune the strength of the effective magnetic field acting on the vortex , and hence to alter its dynamical properties . this will result in a possible measurable effect on transport coefficients , such as the hall coefficient , resistance , etc . to be more explicit and following the procedures used in different approaches treating with the magnus force problem , let us assume that one has a vortex centered at the position @xmath197 . let us furthermore consider that we are in the continuum and at zero temperature so that the berry phase term in the lagrangian now reads @xmath198 \partial_{t } \phi(r , t)$ ] . the contributions linear in @xmath199 and having time independent coefficients do not contribute to the dynamics . moreover , since we consider that the phase @xmath200 and the variable @xmath59 are linked to a vortex solution centering at @xmath197 , it is judicious to introduce a new relative variable @xmath201 , which defines the intrinsic position dependence of the variables with respect to the vortex center . starting from the above given expression for the lagrangian @xmath202 and making a simple change in the time derivative we can rewrite it in the following form : @xmath203 \partial_{t } \phi(r - r , t)\\ & & = \frac{d { \bf r}}{dt}\int d^2 r cos[\theta(r - r , t ) ] { \bf \nabla}_r \phi(r - r , t).\end{aligned}\ ] ] at this point , one recognizes that the term @xmath204 { \bf \nabla}_r \phi(r - r , t)$ ] plays the role of an effective vector potential @xmath205 and @xmath206 represents the velocity of the vortex . this lagrangian is thus equivalent to that of a charged particle in presence of an effective magnetic field in the z - direction given by @xmath207 . @xmath208 hence creates a magnus force @xmath209 $ ] which is transverse and proportional to the vortex velocity and whose strength is linked to the magnitude of this effective magnetic field @xmath210 $ ] . its magnitude is given by the asymptotic value of the background bosonic density at large distance @xmath211 $ ] for @xmath212 . in case of a 2d superconductor described by the bfm scenario , the usual effective magnetic field @xmath213 arising from the supercurrent of the electrons circling the vortex has thus to be supplemented by the above mentioned contribution @xmath214 ( intrinsic to such a scenario ) and permits in principle to modulate the total magnus force upon changing the density of the bosons . clearly , a full description of such an eventuality , which is beyond the scope of the present analysis , has to be studied in more detail and has to include the derivation of the effective action in presence of magnetic field such as to determine the superfluid density of the electrons @xmath215 . at this stage we simply want to point out that the sign of the magnetic field arising from the berry phase term in the action of the bfm changes if one tunes the bosonic density between zero and unity . in terms of the variable @xmath216 $ ] this implies a variation in the range @xmath217 $ ] . in other words , the sign of the intrinsic berry phase induced magnus force will change at @xmath218 when going from the limit of small density of bosons to high density . this extra contribution to the magnus force has to be added of course to the conventional one known for standard superconductors . a frequently asked question concerns the qualitative differences which exist between the bfm and similar scenarios such as the negative - u hubbard model and pure bosonic systems such as the bose - hubbard model and josephson junction arrays . let us start with a comparison of the bfm and the negative - u hubbard model which has been studied in a great variety of different approaches and discussed especially in connection with the bcs - bec crossover . the main issue of such a comparison is the occurrence or not of a quantum critical point in the negative - u hubbard scenario at a finite value of the coupling @xmath219 which introduces the pairing among the electrons . as we have shown in this paper , in the bfm there exists a critical value for the exchange coupling @xmath19 responsible of the pairing in the fermionic subsystem above which the system undergoes a sit . we have seen that within a path integral formalism , one can extract an effective action describing such a transition as being primarily due to the interplay between the quantum pair exchange ( between a boson and a fermion pair ) and the inter - site pair hopping leading to a phase dynamics originating from the itinerancy of the fermions . due to the non - trivial dependence of the strength of these two competing mechanisms on the microscopic parameters of the bfm , there occurs a sit for a finite density as well as finite exchange coupling @xmath19 . for the negative - u hubbard scenario this is not the case and the sit does not occur at any finite value of the ratio @xmath220@xcite . this model merely describes a continuous cross - over between a bcs superconductor and a bec of tightly bound electron pairs as u is increased from 0 to @xmath221 . in order to better understand this difference between the bfm and the negative - u hubbard model , let us consider two scenarios in equivalent situations , i.e. , the half filled band case for the negative - u hubbard model and the fully symmetric case for the bfm ( with the bosonic level lying in the middle of the fermionic band such that both , the fermionic band as well as the bosonic level are half occupied ) . we then address the question how , in the strong coupling limit , the pair hopping is generated out of the basic configurations of states and processes which contribute in this regime . for the case of the negative - u hubbard model with fermions interacting via a local attractive potential , the ground state wave function in the limit of @xmath222 and at half - filling is highly degenerate and is composed of all the possible configurations comprising equal distribution of zero and doubly occupied site . in this limit , one can perform a mapping of this model on the hard - core boson model described by the hamiltonian : @xmath223 @xmath224 and @xmath225 stand for the creation ( destruction ) operators of hard - core bosonic particles ( tightly bound electron pairs ) and for their density operator , respectively . due to the presence of a coupling , which is isotropic both in the boson hopping and in the charge interacting channel , one has a superconducting state for any finite value of the ratio @xmath220 , and possibly a supersolid phase due to the symmetry of the charging interactions characterized by a coexistence of diagonal and off - diagonal long range order . this implies that the quantum critical point is strictly pushed to @xmath222 . in the bfm the phase space in the large @xmath19 limit is completely different . in the fully symmetric case of this model ( corresponding to a total density equal twice the number of sites ( @xmath17 ) ) , the ground state configuration is given by a wave function where all the bosons are strongly coupled with pairs of fermions , thus resulting in a product state of local bonding states : @xmath226 this wave function can be viewed as a ferromagnetic ising - type state in the sense that it is made up of a ferro - type bonding order and is not degenerate with respect to all the other possible local configurations . moreover , by construction it contains bonding - bonding correlations on a long range ( bonding solid ) , but has zero phase correlation length . the latter can be shown by evaluating the static correlation function for the bosons or , equivalently , the fermion pairs which gives @xmath227 and @xmath228 for any distance @xmath229 . next , let us consider the low energy configurations which are mixed into such a @xmath230 ground state when the kinetic energy operator @xmath231 is switched on . applying @xmath232 to @xmath233 , there will occur states which are separated by an energy gap with respect to @xmath234 . the low energy states to be considered as the relevant quantum mixed configurations are of non - bonding nature , such as @xmath235 and @xmath236 . in order to construct those states , let us to begin with consider a local excitation on two adjacent sites @xmath237 given by @xmath238 in the background of bonding states . this configuration can be seen as a ferro - type order interrupted by two domain walls of non - bonding / bonding nature . now let us consider the dynamics of such objects and under which conditions they can be rendered itinerant in a way which leaves the number of bonding configurations unchanged . the problem is thus analogous to that of domain wall dynamics in an ising type systems . it so happens ( a detailed discussion of this is beyond the scope of the present study and will be given at a later stage elsewhere ) , that the local degeneracy of the non - bonding states will be removed by the action of the kinetic term . in this way it induces a global lowering of their excitation energy due to the dispersion of each domain wall which is of the form @xmath239\,\end{aligned}\ ] ] and where @xmath240 . only after the condensation of the domain wall like excitations in the presence of the bonding state background one can meet the conditions for setting up long range boson and fermion - pair phase correlations . this simple sketch of the nature of the excitations in the bfm implies that the onset of phase coherence can not be activated for an arbitrarily small hopping amplitude since one has to overcome the energy gap between the ground state and the manifold of non - bonding states , which is of the order of @xmath241 . as we can see from the expression for the dispersion of the domain walls , this is achieved when @xmath242 becomes zero for @xmath243 , i.e. , when @xmath244 . for z=2 this gives @xmath245 , which determines when those defects - like excitations become gapless and thus induce a proliferation within the background of the bonding states . it is worth pointing out that this value for the critical exchange coupling matches rather well the value obtained from our functional integral approach discussed in this paper ( see fig 3 ) . the picture which hence emerges for the bfm is very similar to that of a xy model in a transverse field . there , the hopping term is responsible of the xy dynamics , while the local boson - fermion pair exchange provides the role of the effective transverse field . this itself is already a strong indication that this model is of different nature to that of the negative - u hubbard scenario , for which we know that it is , on the contrary , akin to an isotropic heisenberg model . a further essential difference between the bfm and the negative - u hubbard model can be seen from their respective path - integral formulations . when the path integral representation is considered for this model@xcite , the first step is to make use of a hubbard - stratonovich transformation to rewrite the quartic interaction term in a bilinear form , where the fermionic operators are now coupled to random auxiliary fields . limiting oneself to purely superconducting order , the usual procedure for manipulating such a bilinear action is to separate the complex auxiliary field into its modulus and a pure phase part before performing an expansion around the saddle point solution for the amplitude of this auxiliary field . this way of proceeding allows then to extract an effective action for the slow phase dynamics . this is distinctly different from the functional integral representation for the bfm which we have presented above and where from the very beginning and throughout such a procedure the fermionic pair fields are coupled to physically real bosonic modes which have their own proper dynamics . as we have seen in section iii , the bosonic part of the action is treated within a coherent pseudo - spin representation where the dynamics of the pseudospin is parameterized by the time dependence of the spherical variables . the role of the dynamics within the spherical representation of the bosonic field and of the berry phase term ( which is a consequence of the quantum interference in the local pseudospin space ) is a distinct feature of this bfm and presents specific differences with respect to the negative - u model . such differences imply in particular that in the bfm case one has feedback effects between the amplitude and the phase fluctuations which are totally absent in an equivalent description of the negative - u hubbard model . such features are important because of intrinsic dissipation effects that eventually can change the nature of the transition and possibly are relevant for the emergence of a `` bosonic metal '' in proximity of the sit , a feature which seems to be outside the framework of the negative - u hubbard model . let us conclude this comparison of the bfm with similar models with a brief discussion on the bose - hubbard model and josephson junction arrays . the main aspect which emerges as a common denominator in all those models , at least as far as the phenomenon of the sit is concerned , is that the mechanism which is responsible for the degrading of the phase coherence is analogous and originates from the competition between the phase and charge degrees of freedom . in the bose - hubbard model this manifests itself as a competition between the boson hopping and their charge repulsion , while in the josephson junction array scenario it appears as an interplay between the josephson tunnelling amplitude and the charging energy . in spite of this , at first sight , similarity between the bfm and those scenarios , we would like to stress that in a system where the dynamics is described by the bfm scenario , the interplay between the phase and amplitude fluctuations is intrinsically related to the coupling between the fermionic and bosonic degrees of freedom . this introduces amplitudes of the different processes at work in form of a non - trivial dependence on the microscopic parameters of the starting hamiltonian . furthermore , as we have seen in the above discussion of the effective action of the bfm , this model contains features which go beyond those which characterize the pure josephson - type dynamics and which arise from the peculiar feedback between the fluctuations of the bosonic density ( or amplitude pair field ) and fluctuations of the phase . last not least , the appearance of an intrinsic berry phase - term in the bfm scenario together with the effect of dissipation due to the fermionic dynamics , can give rise to an unconventional phenomenology when topological phases play a role , and especially in the presence of vortices . in this paper we examined the nature of a superconductor - insulator transition in a system of localized bosons and itinerant fermions coupled together via a pair exchange term . an effective action was derived from such a microscopic model , which , after integrating out the fermionic fields , could be phrased in terms of amplitude and phase fluctuations of the bosons . in order to make the presentation more familiar we discussed the action in a terminology frequently employed in connection with the study of josephson junction arrays . we stress that our system does not necessarily imply any charged fermions and bosons . considering the phase - only part of the effective action it is fully equivalent to the quantum phase model for the josephson junction arrays , discussed in terms of : ( i ) a _ josephson coupling _ term , ( ii ) a _ charging or capacitance _ term and ( iii ) an _ offset charge _ term . equivalent to that in our scenario is : ( i ) a boson hopping term , ( ii ) a term which takes into account the reduction in hopping amplitude due to a fluctuating local boson density arising from the intrinsic on - site exchange coupling between the bosons and the fermions and ( iii ) a chemical potential term controlling the bosonic concentration . new in the present study within already the lowest ( phase - only ) approach to the boson - fermion system is the intricacy of the dependence of the effective josephson coupling , the capacitance term and the off - set charge term on the parameters of the initial hamiltonian , i.e. , the exchange coupling and the density of bosons ( or else the total particle density ) . it turns out that within the lowest ( phase - only ) contribute a superconductor to insulator transition can not only be triggered by a change in the exchange coupling but also by a variation of the boson density . the latter presents evidently a certain interest from the experimental point of view and can possibly be tested in such transition occurring in optical lattices for ultracold fermi gases with feshbach resonance pairing . apart from the phase - only part of the effective action we established the existence of an intrinsic berry phase term , which arises from the hard core nature of the bosons and gives rise to an additional magnus force when the system is such that topological ground state configurations like vortices are stabilized . this again is of potential interest for experiments since in principle one can change the sign of the magnus force upon changing the density of bosons and thus deviating the motion of a vortex from one direction into the opposite one . these very preliminary results will be dealt with in greater detail in some future studies . a further property of the berry phase term is that it gives rise to a bilinear coupling term between the phase and amplitude fluctuations , and is hence much more direct and relevant than similar terms in , for instance , scenarios based on the negative - u hubbard model where they occur only at a much higher order in a corresponding loop expansion of the trace in the effective action . this again merits to be investigated in some detail with the aim to study the dissipation introduced by such amplitude phase coupling and its effect on the nature of the transition , in view of exploring the possibility of an intermediary bosonic metallic ground state . finally , we have dealt with a frequently asked question concerning the differences between the negative - u hubbard scenario - mainly studied in connection with the bcs - bec cross - over and the presently studied boson - fermion model . far from being able to give a complete account for the major differences we found mainly two aspects which distinguish the physics of these models on a qualitative and robust level . the one is that in the negative - u hubbard model a superconductor - insulator transition can not take place at any finite coupling u , nor can such a transition be triggered by the change in particle concentration . the second point is that the ground states in the strong coupling limit of the two models are quite different : a highly degenerate ground state for the negative - u hubbard model with excitations being controlled by an isotropic heisenberg model when the hopping term is switched on . contrary to that , for the boson fermion model the ground state is non - degenerate ( corresponding to a ferro pseudo - magnetic ising type system of singlets formed by bosons and fermion pairs ) , which , after switching on the fermion hopping term , gives rise to propagating domain like structures . the topological structures appearing in the ground state might have measurable consequences in the transport properties near the superconductor - insulator transition . haviland , y. liu , and a.m. goldman , phys . lett . * 62 * , 2180 ( 1989 ) e. abrahams , p. w. anderson , d. c. licciardello and t. v. ramakrishnan , phys . rev 42 , 673 ( 1979 ) . p. a. lee and t. v. ramakrishnan , rev . 57 , 287 ( 1985 ) ; d. belitz and t.r . kirkpatrick , rev . mod . phys . * 66 * , 261 ( 1994 ) . a.b . harris j. phys . c : solid state phys . 7 1671 ( 1974 ) . k. efetov , zh . . fiz . * 78 * , 2017 ( 1980 ) . v. ambegaokar , u. eckern , and g. schon , phys . rev . lett.*48 * , 1745 ( 1982 ) . larkin and y.n . ovchinnikov , phys . b * 28 * , 6281 ( 1983 ) . u. eckern , g. schon , and v. ambegaokar , phys . b * 30*,6419 ( 1984 ) . s. chakravarty , s. kivelson , g.t . zimanyi , and b.i . halperin , phys . b * 35 * , 7256 ( 1987 ) . v.j . emery and s.a . kivelson , phys . * 74 * , 3253 ( 1995 ) . m. p. a. fisher , p. b. weichman , g. grinstein and d. s. fisher , phys . rev . b 40 , 546 ( 1989 ) . fisher , g. grinstein , and s.m . girvin , phys . * 64 * 587 ( 1990 ) . a.p . kampf and g.t . zimanyi , phys . b * 47 * , 279 ( 1993 ) sorensen et al . , phys . rev . * 69 * , 828 ( 1993 ) . d. ephron , a. yazdani , a. kapitulnik , and m.r . beasley , phys . * 76 * , 1529 ( 1996 ) . jaeger , d.b . haviland , a.m. goldman , and b.g . orr , phys . b * 34 * , 4920 ( 1986 ) . a. gerber , j. phys . : condens . matter * 2 * , 8161 ( 1990 ) . van der zant , f.c . fritschy , w.j . elion , l.j . geerligs , and j.e . mooij , phys . * 69 * , 2971 ( 1992 ) ; h.s.j . van der zant , w.j . elion , l.j . geerligs , and j.e . mooij , phys . b * 54 * 10 081 ( 1996 ) . n. mason and a. kapitulnik , phys . lett . * 82 * , 5341 ( 1999 ) ; _ ibid _ phys . b * 65 * , 220505(r ) ( 2002 ) . p. phillips and d. dalidovich , science * 302 * , 243 - 247 ( 2003 ) and references therein . d. das and s. doniach , phys . b * 60 * 1261 ( 1999 ) . d. dalidovich and p. phillips , phys . b * 64 * 052507 ( 2001);p . phillips and d. dalidovich _ ibid _ * 65 * 081101 ( 2002 ) . k. wagenblast , a. van otterlo , g. schon , and g.t . zimanyi , phys . lett . * 78 * , 1779 ( 1997 ) . rimberg et al . lett . * 78 * , 2632 ( 1997 ) . s. sachdev , phys . b * 57 * , 7157 ( 1998 ) . a. kapitulnik , n. mason , s.a . kivelson , and s. chakravarty , phys . b , * 63 * 125322 ( 2001 ) . m. feigelman and a. larkin , chem . phys . * 235 * , 107 ( 1998 ) . b. spivak , a. zyuzin , and m. hruska , phys . b * 64 * 132502 ( 2001 ) . j. ranninger and a. romano , phys . rev . b * 66 * , 94508 ( 2002 ) . a. mysyrowicz , e. benson , and e. fortin , phys . 77 , 896 ( 1996 ) . a. schnell , g. ropke and p. schuck , phys . 83,1926 ( 1999 ) . e. altman and a. auerbach , phys . b 65 , 104508 ( 2002 ) . v. a. yurovsky and a. ben - reuven , phys . a 67 , 043611 ( 2003 ) . e. timmermans , p. tommasini , m. hussein and a. kerman , phys . rep . * 315 * , 199 ( 1999 ) . j. ranninger , j.m . robin and m. eschrig , phys . * 74 * , 4027 ( 1995 ) . p. devillard and j. ranninger , phys . lett . , * 84 * , 5200 ( 2000 ) . t. domanski and j. ranninger , phys . b , * 63 * , 134505 ( 2001 ) . j. ranninger and l. tripodi , phys . rev . b , * 67 * , 174521 ( 2003 ) . m. cuoco , c. noce , j. ranninger , and a. romano , phys . b * 67 * , 224504 ( 2003 ) . j.w . negele and h. orland , _ quantum many - particle systems _ ( addison - wesley , reading ma , frontiers in physics 1987 ) . e. fradkin , _ field theory in condensed matter _ , ( addison - wesley , reading ma , frontiers in physics 1991 ) . m. stone , phys . rev . d * 33*,1191 ( 1986 ) . haldane , phys . lett . * 57 * , 1488 ( 1986 ) . schakel , mod . b * 14 * , 927 ( 1990 ) ; u. eckern , g. schon , and v. ambegaokar , phys . rev . b * 30 * , 6419 ( 1984 ) ; i.j.r . aitchison , p. ao , d.j . thouless , and x .- m . zhu , phys . b * 51 * 6531 ( 1995 ) . s. doniach , phys . b * 24*,5063 ( 1981 ) . c. bruder , r. fazio , a. kampf , a. van otterlo and g. schon , physica scripta * 42 * , 159 ( 1992 ) . r. fazio , and h.s.j . van der zant , phys . rep . * 355 * , 235 ( 2001 ) and references therein . p. ao and d.j . thouless , phys . lett . * 70 * , 2158 ( 1993 ) . t. schneider and j. m. singer , _ phase transition approach to high temperature superconductivity _ imperial college press ( london ) , ( 2000 ) . t. k. kopec , phys . b * 65 * , 054509 ( 2002 ) ; s. de palo , c. castellani , c. di castro and b. k. chakraverty , phys . b * 60 * , 564 ( 1999 ) .

quantum information theory is an area of scientific investigation that has witnessed an enormous progress in the last decade @xcite . in the framework of quantum information science , paradigmatic systems of condensed matter physics such as , e.g. , spin chains and harmonic lattices , are analyzed from the point of view of their information content and ability to process and transfer information . fundamental concepts of statistical mechanics and probability , such as the shannon - von neumann entropy @xcite , play a central role in the quantification of bipartite quantum entanglement @xcite . on the other hand , exciting advances in quantum information research are in a sense paying back the debt : the mathematical and theoretical toolkit developed for the characterization and quantification of quantum entanglement has proven useful to tackle questions and improve our understanding in the investigation of strongly correlated systems and quantum phase transitions @xcite . perhaps the most interesting development stemming from the application of the tools of entanglement theory concerns the study of correlation scaling in diversely connected systems , with its relations to conformal field theory and the proof of longstanding conjectures on entropic area laws in spin and harmonic lattice systems @xcite . these investigations and the associated results are of fundamental interest because they allow to enhance the control and establish firmer bounds and limits of applicability on many important tools of simulation and numerical analysis of complex many - body systems , a fundamental task in condensed matter , such as the density matrix renormalization group ( dmrg ) algorithms @xcite , matrix product representations @xcite , the entanglement renormalization ansatz @xcite , and weighted graph states methods @xcite . the pioneering application of quantum information concepts to condensed matter was the observation that two - body entanglement , as quantified by the concurrence , in the ground state of a cooperative system , exhibits peculiar scaling features approaching a quantum critical point @xcite . these seminal studies helped to clarify that at quantum phase transitions @xcite , the dramatic change in the ground state of a many - body system is associated to , or reflected by , a change in the way that quantum fluctuations are correlated , that is , in the way entanglement is distributed among the elementary constituents . many recent efforts have thus been aimed at understanding the behavior of different measures of entanglement at quantum phase transitions , and assessing the corresponding enhancement of properties useful for applications in quantum technology and quantum engineering @xcite . however , since most entanglement measures can be often rewritten in terms of the conventional n - point correlation functions , the presence of a divergence in the former at quantum criticality can be directly traced back to a divergence in the latter @xcite , although in some particular instances entanglement - based studies allowed to discover novel types of phase transitions @xcite . finally , a more technologically oriented product of the interaction between these two areas of research is the study of quantum spin chains as natural information carriers and distributors @xcite , able to realize tasks such as quantum state transfer @xcite and storage @xcite , possibly with maximal fidelity in chains with suitably engineered couplings and dynamics . quantum engineering and quantum technology are being indeed developed at a fast pace . quantum devices are vigorously pursued for applications ranging from nano - sciences to quantum computation and entanglement - enhanced metrology @xcite . despite a large variety of possible implementations involving different physical systems , many relevant properties of such devices can be in fact investigated in a unified setting by appropriate mappings to quantum spin models @xcite . control of separability and entanglement in ground states of quantum spin models plays an important role in entanglement - enabled quantum technology applications , and needs to be characterized and quantified in different physical regimes @xcite . from a more fundamental perspective , the determination of exact solutions endowed with precisely known properties of separability or entanglement , can be of great relevance in the study of advanced models of condensed matter and cooperative systems that are in general not exactly solvable . in this paper we exploit quantum information tools to tackle another important problem in condensed matter physics : the determination and characterization of exact ground states of cooperative systems characterized by the property of being in the form of a fully factorized , tensor product of single - particle pure states , with no quantum correlations between the individual constituents . historically , the occurrence of totally factorized ( unentangled ) ground states of quantum many - body systems was first discovered in the one - dimensional anisotropic heisenberg model with nearest - neighbor interactions by kurmann , thomas and mller @xcite , by adopting a direct method in terms of product state ansatz . this result was later re - derived and extended to two dimensions using quantum monte carlo numerical methods @xcite . the direct product - state ansatz method of kurmann , thomas and mller has been later extended to anisotropic rings with interactions of arbitrary range , by exploiting the observation that factorized ground states break the parity symmetry @xcite . complex quantum systems exhibiting cooperative behaviors , whose ground states are typically entangled @xcite , may thus admit , for some non trivial values of the hamiltonian parameters , a ground state which is completely separable . the occurrence of factorization at finite or even strong values of the couplings is thus an effect of a delicate balancing between interactions and external fields . the phenomenon of ground state factorization is particularly intriguing as it appears to be associated with the presence of an `` entanglement phase transition '' with no classical counterpart @xcite . furthermore , for the purposes of quantum engineering applications , that employ distributed entanglement in order to manipulate and transfer information @xcite , factorization points need to be exactly identified and either avoided , in order to guarantee the reliable implementation of quantum devices , or properly exploited for the dynamical creation of strongly entangled multipartite states of large assemblies of microsystems ( graph and cluster states ) . finally , and also quite importantly , for models not admitting exact general solutions , achieving knowledge of the exact ground state , even if only for the restricted nontrivial set of parameters associated to factorization , would allow ( i ) to prove the existence of an ordered phase and characterize it ; ( ii ) to build variational or perturbative approximation schemes around the exact factorized solution , that may then be used as test benchmarks for the validity and the precision of numerical algorithms and simulations . unfortunately , the direct method based on the ansatz of a factorized solution is neither sufficiently general nor mathematically tractable , apart from the notable exceptions of one - dimensional spin chains with short - range @xcite , infinite - range @xcite , and some types of long - range exchange interactions @xcite . in a recent work @xcite we introduced an analytic method that allows to determine exactly the existence of factorized ground states and to characterize their properties in quantum spin models defined on regular lattices of any size ( finite as well as at the thermodynamic limit ) , in any spatial dimension , and with spin - spin interactions of arbitrary range . in correspondence to rigorously established ground state factorizability , the method also allows to determine novel sets of exact solutions in generally non exactly solvable models . all the previous particular findings can be rigorously re - derived and extended within a unified framework inspired by concepts of quantum information science . in the present work we extend the original method introduced in ref . @xcite and we develop a general self - contained theory of ground state factorization in ( frustration - free ) quantum spin systems : with no _ a priori _ assumption on the nature of the magnetic ordering , we derive a complete and closed set of conditions that have to be satisfied by the hamiltonian interaction parameters in order for the ground state to be factorized , at precisely determined values of the external magnetic field , with well defined values of the magnetization and of the energy . besides the completely rigorous re - derivation of the few previously known results on quantum factorization points @xcite , our general analytic method allows to determine novel , exact factorization points and factorized ground states in , generally non exactly solvable , spin-@xmath0 models as well as in the corresponding hamiltonian models of any higher spin . these results hold true regardless of spatial dimensionality and interaction range , and can be easily extended even to non - translationally invariant systems . we illustrate the versatility of the method with a series of different applications , including models with nearest - neighbor interactions ( for which we provide the full factorization diagram , generalizing the results of refs . @xcite and of our recent work @xcite ) , models with long - range interactions ( notably including , among others , the fully connected infinite - range lipkin - meshkov - glick ( lmg ) model @xcite and systems with unbalanced next - to - nearest - neighbor interactions on cubic lattices ) , and models with spatial anisotropies . the paper is organized as follows . an excursus on the problem of ground state factorization is presented in sec . [ sec.history ] . in sec . [ sec.squoexe ] we recall the quantum information toolkit upon which our method is based , in particular the formalism of single - spin , or single - qubit , unitary operations and associated entanglement excitation energies , previously introduced for the characterization and quantification of bipartite entanglement from an abstract geometrical perspective @xcite . the general method for the rigorous determination of factorized ground states , that generalizes the scheme introduced in ref . @xcite , is discussed in full detail in sec . [ sec.method ] . relevant applications and examples , including models defined on lattices of different spatial dimensions and with different interaction ranges as well as extensions to models with spatial anisotropies and with dimerized interactions are discussed in sec . [ sec.short ] , sec . [ sec.long ] , sec . [ sec.lmg ] , and sec . [ sec.anisotropy ] . concluding remarks and outlooks on future perspectives are presented in sec . [ sec.concl ] . quantum fluctuations in the ground states of cooperative many - body systems are typically highly correlated , that is , those states contain in the most general instance a strong degree of entanglement distributed among the individual components . this is one of the reasons why their exact characterization is often hard to be accomplished by analytical or even numerical methods . however , there may exist particular solutions corresponding to special values of the hamiltonian parameters such that the ground state becomes `` classical - like '' , in the sense of being a full product state of single - site factors ( completely separable , fully factorized state ) . it is quite surprising that despite the inherent appeal of such exact , special solutions , and the renewed interest stemming from possible applications to quantum technology , very few analytical advances were obtained in the last quarter of a century . the first systematic work on ground state factorization in quantum spin models was completed by kurmann , thomas , and mller in 1982 @xcite . they focused on the one - dimensional nearest - neighbor anisotropic heisenberg - like antiferromagnetic spin-@xmath0 model in transverse field , with the assumption that all the spin - spin couplings take nonnegative values . et al . _ identified a precise value of the external magnetic field , the `` factorizing field '' , corresponding to which the system admits a fully factorized ground state . they employed a very direct , brute force " method , i.e. formulating a product state ansatz and verifying if and for what values of the hamiltonian parameters it satisfies the stationary schrdinger problem with lowest energy . even though this approach appears obvious and extremely easy to pursue , the positive result obtained by kurmann _ _ in the case of the 1-d heisenberg model was enabled by some special properties of the case they analyzed . in fact , in order to verify the eigenvalue equation for the product state ansatz , one needs to decompose the total hamiltonian of the system into a series of pairwise terms , which could be done easily for the model studied by kurmann _ _ since all the `` pairwise components '' of the factorizing magnetic field had the same weight in this particular case . however , this direct method of determining factorization is in fact significantly less obvious , useful , and comprehensive , when one tries to apply it to more general cases . for instance , considering models with interaction terms of growing spatial range , and/or increasing the spatial dimensionality of the lattice , and/or treating extensions to spatially anisotropic or non - translationally invariant models , results in increasingly hard `` guesses '' concerning the partitioning of the external field into pairwise components and in increasingly nontrivial verification steps . depending on the structure of the model and on the form of the product state ansatz , the direct method may either fail to detect factorization points at all , or it may detect only a subset of all physically realizable factorized solutions , or finally it may lead to intractable sets of conditions . these limitations of the direct method become evident , as we will see in the following , even in the simplest 1-d @xmath1 model with nearest - neighbor interactions analyzed by kurmann and coworkers . in their original work , due to the insufficient generality of their proposed product ansatz , they failed to identify a wide range of situations that allow the existence of factorized ground states . as we will see when discussing models with competing interactions of different spatial ranges , a further serious fault of the direct method , that becomes incurable in the presence of frustration , is its incorrect assessment of factorized ground states that are in fact excited energy eigenstates of the system . following the seminal analysis by kurmann _ _ , the direct method has been applied to the fully connected infinite - range lipkin - meshkov - glick model by dusuel and vidal @xcite , and to the anisotropic heisenberg chain with the same interactions of arbitrary range along the three spatial directions by hoeger et al . @xcite and by rossignoli _ we would like to remark that these works , that include the results of refs . @xcite as special cases , provide an interesting characterization of factorization points in finite ( one - dimensional ) translationally invariant lattice spin models with periodic boundary conditions as those points where the parity symmetry is broken . a general and rigorous approach to the problem of ground state separability , completely different from the direct method based on the product state ansatz , has been introduced by us in ref . @xcite and is fully developed in the present work . this analytic method allows _ inter alia _ to complete the factorization diagram of the anisotropic spin-@xmath0 models without restrictions on the sign of the couplings , and to go significantly beyond the limitations of the previously mentioned studies . our approach is based on an additional characterization of product states ( including product ground states ) that employs concepts and tools of quantum information and entanglement theory . the strong point of the quantum informatic approach to ground state factorization lies in its generality and its ensuing independence from the types and ranges of interactions present in the hamiltonian , the lattice size , and the spatial dimension , which makes it applicable in the most general cases and therefore especially useful in all the situations that are either intractable or incomplete if one resorts to the direct method or to numerical techniques . in fact , in the original formulation of the quantum informatic method @xcite there was still trace of a residual `` ansatz '' in the initial stage , since the magnetic ordering of the candidate factorized ground state had to be in some sense guessed , and imposed _ a priori_. moreover , in the final stage , we were not yet able to derive a completely general set of conditions in order to determine in each situation whether a candidate factorized state that turned out to belong to the spectrum of the hamiltonian was actually a ground state or belonged to the set of excited states . this had then to be decided case by case in each specific instance . in the present work we proceed further and develop the complete method to determine and characterize factorized ground states in quantum cooperative systems . we provide the systematics that allows to determine the occurrence of full ground state separability given a generic quantum spin hamiltonian . the magnetic ordering is derived as a consequence of the theory and not imposed _ a priori_. a complete set of general conditions is established for the verification of ground state factorization , and the elegance of the method does not hinder practical usefulness in the application to specific models and situations . we would like to recall that numerical advances on the problem of ground state factorization have been realized by the monte carlo study of ground state factorization in two - dimensional anisotropic nearest - neighbor spin models @xcite . unfortunately , beyond this important result the numerical approach appears to have a very restricted potential for extensions , since the required computational power increases rapidly with the spatial dimension and/or with the range of interactions . we will show that the difficulties inherent to the direct analytic approach and to the numerical method are intrinsically absent in our formalism . in the present section we provide a brief overview of some recent results on the geometrical interpretation of the entanglement measure known as `` tangle '' ( linear entropy ) @xcite and the consequences it entails for the relationship between energy and pairwise entanglement @xcite . such results embody the premises for our all - analytic characterization of ground state separability in quantum spin systems . here and in the following we consider generic systems made of @xmath2 spin-@xmath0 elementary constituents , or , in the language of quantum information , @xmath2 qubits . for such a system , we introduce the set of single - qubit unitary operations ( _ squos _ ) @xmath3 @xcite , defined as the unitary transformations that act separately as the identity on any spin of the system except the @xmath4-th one , on which they act instead as hermitian , unitary , and traceless operators : @xmath5 in eq . ( [ e.uk ] ) @xmath6 stands for the identity operator on the @xmath7-th spin of the system while the generic hermitian , unitary , and traceless operator @xmath8 can be expressed as a linear combination of the standard spin operators ( @xmath9 ) defined on the @xmath4-th spin : @xmath10 in the previous definition the parameters @xmath11 and @xmath12 take values respectively in the ranges @xmath13 $ ] and @xmath14 . physically , @xmath15 corresponds to a rotation in the spin space , and the traceless condition imposes the rotation operation to be orthogonal to the identity , so basically a combination of spin flip and phase flip . a squo transforms every pure state @xmath16 into a new state @xmath17 , that in general differs from @xmath18 . we may quantify the action of the squo in terms of the trace distance between the original and the transformed state , defined as @xmath19 . the distance @xmath20 varies in the interval @xmath21 $ ] and vanishes if and only if the two states coincide , meaning that the considered squo leaves the original state unchanged . for a given initial state and for any site @xmath4 , one can determine the _ extremal - squo _ ( e - squo ) that minimizes such a distance . one can prove that the e - squo is uniquely determined by the following conditions on the angular variables @xcite : @xmath22 where @xmath23 denotes the expectation value of the spin operator @xmath24 on the unperturbed state @xmath18 ( @xmath25 ) . the e - squo plays a crucial role in our analysis : for any pure state defined on a general system of interacting qubits one can prove @xcite that the square of the distance associated to the action of the e - squo coincides with the linear entropy @xmath26 $ ] of the @xmath4-th qubit ( where @xmath27 denotes the reduced density matrix of qubit @xmath4 ) . this quantity , also known as the _ tangle _ @xmath28 in the literature , is an entanglement monotone for qubit systems and satisfies important monogamy constraints @xcite . being a monotonic function of the von neumann entropy of entanglement @xcite , it quantifies the bipartite block entanglement present in the state @xmath18 between spin @xmath4 and the remaining @xmath29 constituents , and is thus a proper measure of single - site entanglement @xcite . one has the following relation @xcite : @xmath30 \ ; .\ ] ] this geometric - operational characterization of the tangle allows to develop a quantum informatic approach to the problem of characterizing the correlation properties of the ground state in many - body quantum systems @xcite . let us consider a collection of spin-@xmath0 systems defined on a regular lattice , with hamiltonian @xmath31 , and let @xmath18 denote from now on the ground state of @xmath31 . if the system is in the ground state , the application of a squo on a generic spin of the lattice is necessarily associated to an energy transfer that one must provide to the system in order to realize the selected squo . in other words , a squo perturbs the system , and this perturbation results in an increase of the average energy ( obviously , in general the transformed state needs not to be an eigenstate of the hamiltonian ) . such energy deviation from the ground - state value can be quantitatively defined as @xmath32 because @xmath18 is the ground state of @xmath31 , @xmath33 is a non negative defined quantity . considering the special relation between the e - squo and the tangle we may expect that the associated amount of energy difference @xmath34 , appropriately named ` entanglement excitation energy ' ( exe ) , will have as well a strong , direct connection with the tangle and in general with the single - site entanglement in the ground state . in fact , starting from the above definition of eq . ( [ e.ees ] ) and taking into account the property of the e - squo of leaving unchanged a fully disentangled state , it has been established in full generality @xcite that if the system is invariant under spatial translation and the hamiltonian of the system does not commute with any possible squo , then the vanishing of the exe is a _ necessary and sufficient _ condition to admit a fully factorized ground state @xmath18 @xcite . hence , if the system that we want to analyze satisfies all the former hypotheses , the vanishing of the exe provides an additional condition , besides the vanishing of the tangle @xmath28 , that one needs to include in establishing a general analytic approach to the problem of ground state factorization . the general method that was first partially developed in ref . @xcite for the determination of existence , location , and exact form of factorized ground states rests on the following two main observations : the ground state @xmath18 of a spin-@xmath0 hamiltonian @xmath31 is factorized if and only if the single - site entanglement ( tangle ) @xmath28 vanishes for all spins @xmath4 in the lattice ; provided that @xmath35\neq 0 $ ] for every possible squo @xmath36 of the form eq . ( [ e.uk ] ) , the ground state @xmath18 of a spin-@xmath0 hamiltonian @xmath31 is factorized if and only if the exe @xmath37 vanishes @xmath38 . in ref . @xcite , the strategy to the determination and understanding of factorization for a given hamiltonian bore on the above two facts and on the requirement to assume as working point a phase endowed with some kind of magnetic order . by imposing the vanishing of the exe and of the tangle , we showed how to determine uniquely the form of the completely disentangled state compatible with the assumed magnetic order , candidate to be the ground state of the system at a particular value of the external magnetic field ( labeled as `` factorizing field '' ) . the final steps of the procedure concern the determination of exact analytic conditions for the candidate factorized state to be an eigenstate of the hamiltonian with lowest energy . in ref . @xcite , we derived a closed set of eigenstate conditions , but establishing whether a factorized eigenstate is actually a ground state had to be left to a case by case analysis . in the present paper we generalize the method , extend its scope , and apply it to different classes of quantum spin models . firstly , we will show that there is no need for _ a priori _ assumptions on the magnetic order . in fact , we will prove that the existence of a factorized ground state is a necessary and sufficient condition for the existence of quantum phase transitions and ordered phases in translationally - invariant quantum spin models with exchange interactions and in the presence of external fields . moreover , the actual structure of the factorized ground state determines automatically the kind of ordered phase . further , we derive a complete set of rigorous conditions for the candidate disentangled state to be an eigenstate and a ground state . in particular , for systems with no frustration we prove that if the factorized state is an eigenstate , then it is always a ground state of the system . we illustrate the method in detail by considering its application to the general , translationally invariant , exchange hamiltonian @xmath31 for spin-@xmath0 systems on a @xmath20-dimensional regular lattices , with spin - spin interactions of arbitrary range and arbitrary anisotropic couplings : @xmath39 here @xmath40 ( and similarly @xmath41 ) is a @xmath20-dimensional index vector identifying a site in the lattice , @xmath42 @xmath43 stands for the spin-@xmath0 operator on site @xmath40 , @xmath44 is external field directed along the @xmath45 direction , @xmath46 is the distance between two lattice sites , and @xmath47 is the spin - spin coupling along the @xmath48 direction . translational invariance is ensured by the fact that all the couplings depend only on the distance @xmath49 between the spins . this type of hamiltonian encompasses a large variety of models and spans over several universality classes including , among others , the ising , xy , heisenberg , and xyz symmetry classes . besides their importance in quantum statistical mechanics and in the theory of quantum critical phenomena , these models play an important role in the study of various schemes of quantum information and quantum communication tasks with quantum many - body systems @xcite . it is a rather straightforward exercise to verify that the hamiltonian eq . ( [ e.hamiltonian ] ) never commutes with the squos @xmath50 . translational invariance and non - commutativity of the hamiltonian with the squos guarantee that the vanishing of the exe is a necessary and sufficient condition for the full separability of the ground state [ statement ( ii ) above ] . since the external field @xmath44 is uniform on the lattice , it forces the existence of a site - independent , nonvanishing magnetization @xmath51 along the @xmath45 axis : @xmath52 . on the other hand , at a factorization point , statement ( i ) above imposes the vanishing of the single - site entanglement on all sites of the lattice . therefore , if a factorization point exists , the magnetizations along the directions orthogonal to the external field must assume the following form : @xmath53 where @xmath54 is the modulus of the projection of the magnetization on the @xmath55 plane and the local orientations @xmath56 , and hence the type of magnetic ordering in the @xmath55 plane , remain undetermined . since @xmath51 is site - independent , then @xmath57 is site - independent as well . taking into account eq . ( [ e.minimumpoint ] ) of sec . [ sec.squoexe ] one sees that , at least at the factorization point , @xmath58 , i.e. one of the two angles that fix the orientation of the e - squo , does not depend on the site index . therefore @xmath59 @xmath60 . collecting the former results and considering that in the presence of a factorized ground state , if it exists , all the correlation functions are products of single - site expectations values , we obtain that the condition on the vanishing of the exe associated to the generic site @xmath61 reads : @xmath62 in the former equation @xmath63 is the _ factorizing field _ , i.e. the value , if it exists , of the external field for which ground state factorization occurs . since we are looking for a fully factorized ground state , eq . ( [ e.exe ] ) must be satisfied simultaneously for all @xmath61 and , henceforth , eq . ( [ e.exe ] ) must be site - independent . this fact in turn implies that the following relation must be satisfied by every angle @xmath56 : @xmath64 eq . ( [ e.exe ] ) must hold independently of the site on which the e - squo acts and the value of the constant @xmath65 must be identified by determining the expression of the energy density , i.e. the energy per site , associated to the factorized ground state . given the total ground - state energy @xmath66 and the total number of sites @xmath2 , the latter quantity can be written as @xmath67 where @xmath68 is the coordination number , that is the number of sites at distance @xmath49 from a given site . in order to determine the magnetic order in the candidate factorized ground state , one can now exploit the fact that , by definition , it must minimize the energy of the system . therefore such state will be simply characterized by the magnetic order that minimizes the per - site contribution @xmath65 , depending on the hamiltonian parameters and the geometric structure of the lattice . thus the general conditions for full separability of the ground state lead to the uncovering of the magnetic order in the @xmath55 plane . in the original formulation of the quantum - informatic method @xcite one had to assume from the beginning the type of ordered phase , and this resulted in constraints on the magnetization and limitations in the range of the coupling coefficients . the outcome of the minimization of the energy density depends on the presence or the absence of frustration in the system , either due to the lattice geometry , or to the structure of the hamiltonian , or to both causes . for hamiltonians belonging to the class of eq . ( [ e.hamiltonian ] ) , frustration can be due both to the geometry of the lattice , e.g. for systems with nearest - neighbor antiferromagnetic interactions defined on a lattice that admits closed loops with an odd number of spins , and to the competition of interactions of different spatial orders , e.g. in systems defined on linear chains with nearest - neighbor ferromagnetic and next - nearest - neighbor antiferromagnetic interactions . frustration effects in systems with interactions of different spatial ranges arise whenever it happens that the minimization of the energy associated to interactions over a certain spatial scale precludes the possibility of minimizing the energy associated to interactions over a different spatial scale . in the absence of frustration , i.e. when the lattice geometry and the interactions between the spins are such to allow all energy minimizations simultaneously , it is straightforward to prove that every possible factorized ground state must be characterized by one of only four magnetic orders out of the many possible ones . the complete proof of this statement is reported in appendix [ appendixfourcases ] . this restricted set includes ferromagnetic ordering along the @xmath69 direction ( @xmath70 ) or along the @xmath71 direction ( @xmath72 ) , and antiferromagnetic ordering along the @xmath69 direction ( @xmath73 ) or along the @xmath71 direction ( @xmath74 ) , the type of order effectively present in the ground state can be determined by comparing the values of the different `` net interactions '' along the @xmath69 and @xmath71 directions , where the net interactions are defined as follows : @xmath75 with @xmath76 . the net interaction along the @xmath45 direction is defined equivalently as @xmath77 as a function of the net interactions , the magnetic orders in the system are determined by the values of @xmath78 : @xmath79 it is interesting to observe that , in terms of the net interactions , one can immediately establish the presence or absence of frustration in the system : any frustration arising from the presence of competing interactions would imply that @xmath80 , with @xmath76 depending on which of the two axes is characterized by a non vanishing value of the magnetization . we remark that here and in the following we are not considering the particular situation in which a _ saturation _ occurs rather than a true ground state factorization : this instance will be discussed separately in subsec . [ s.sec.saturation ] . from now on , without loss of generality , we will specialize our analysis to the the case of an antiferromagnetic ordering along the @xmath69 direction . with trivial modifications , all the steps and results that will be obtained in the following hold as well in the remaining three cases of eq . ( [ fourcases ] ) . at the end of the procedure we will present a summary table collecting the main results for all the four types of magnetic orders compatible with factorization . given an antiferromagnetic ordering along the @xmath69 direction , and keeping in mind that in such case @xmath73 , eq . ( [ e.magnxy ] ) implies that @xmath81 while @xmath82 with @xmath83 . the @xmath84 sign reflects the fact that the antiferromagnetic order discriminates between two sublattices , each characterized by the opposite sign of the staggered magnetization along @xmath69 . consequently , the general condition eq . ( [ e.exe ] ) takes the form @xmath85 \ , , \end{aligned}\ ] ] that admits two solutions for @xmath86 , @xmath87 however , since the e - squo must be unique @xcite , the two solutions must coincide . factorization thus requires that @xmath88 moreover , the requirement of a vanishing tangle ( single - site entanglement ) imposes @xmath89 combining eqs.([e.mz][e.mx ] ) via eq . ( [ e.minimumpoint ] ) yields a closed expression for the phase @xmath86 as a function of the hamiltonian parameters and of the factorizing field @xmath63 : @xmath90 therefore , eq . ( [ e.theta ] ) , together with the knowledge of @xmath91 , once inserted into eq . ( [ e.ok ] ) , determines , independently of the value of the magnetizations , the form of the e - squo at factorization : @xmath92 the form of the candidate factorized state @xmath93 is then readily determined by imposing that the action of the e - squo on all lattice sites leaves the ground state invariant : @xmath94 > from eqs.([e.inalterate][e.oz ] ) we find then that the candidate factorized ground state @xmath93 has to be a tensor product of single - site states that are the exact eigenstates of the operators @xmath95 with eigenvalue @xmath0 : @xmath96 equation ( [ e.candidate ] ) expresses the general form that a factorized ground state must assume given an antiferromagnetic order along @xmath69 . this result is independent of the actual values assumed by the staggered magnetizations @xmath97 s . we remark that in order to establish the general form of the candidate ground state it is essential to impose the vanishing of the exe . in fact , without requiring it , the phase @xmath86 and the candidate factorized ground state become explicitly dependent on the magnetizations . then , imposing the conditions for eq . ( [ e.candidate ] ) to be an eigenstate of the hamiltonian , one merely obtains a relation between the factorizing field @xmath63 and the magnetizations @xmath98 s . if the analytic expression of at least one of the @xmath98 s is known , it is still possible to obtain an expression of the field @xmath63 in terms of the hamiltonian parameters . however , this actually occurs in few special , exactly solvable , cases ( for instance , the xy model @xcite ) . otherwise , in general one needs to resort to numerical evaluations of the magnetizations or introduce other approximations in the analysis . this problem is completely eliminated in the general analytic framework based on the invariance of the factorized ground states under the action of e - squos and on the vanishing of the exe at factorization . before proceeding further let us clarify a relevant point . because of the even - odd symmetry of the energy spectra of the system , as elucidated also in refs . @xcite , the assignment of one of the two different staggered magnetizations to each sublattice is arbitrary . according to this observation , one sees that , inverting the sign of the magnetization along the @xmath69 direction on each site , all the conditions for factorization are still satisfied but the candidate ground state takes the form @xmath99 obviously the two states in eq . ( [ e.candidate ] ) and in eq . ( [ e.candidate-1 ] ) are distinguishable and both @xmath93 and @xmath100 are legitimate candidate ground states of the system , as well as general linear combinations of the two ( the latter will be in general highly entangled ) . here and in the following we assume to be working in a situation of broken symmetry , i.e. after that a small perturbing external field along the @xmath69 direction has lifted the degeneracy between @xmath93 and @xmath100 , e.g. lowering the energy of the former and rising that of the latter . after a time long enough to ensure convergence to equilibrium , i.e. relaxation to the state @xmath93 , the perturbation is switched off : in the thermodynamic limit , this ensures that the system will remain indefinitely in the state @xmath93 . having determined the exact form of the candidate factorized ground state , the subsequent step concerns the determination of the conditions for its occurrence , i.e. the conditions under which a state of the form eq . ( [ e.candidate ] ) is indeed the eigenstate of @xmath31 , eq . ( [ e.hamiltonian ] ) , with the lowest energy eigenvalue . to this aim , taking the value of the external field at the factorization point , @xmath101 , it is useful to decompose the total hamiltonian as a sum of pairwise hamiltonian terms @xmath102 , so that @xmath103 , where each term @xmath102 involves only the degrees of freedom of a single pair of spins : @xmath104 here the quantity @xmath105 plays the role of the `` component '' of the factorizing field that acts on the selected pair of spins and obeys the following relation : @xmath106 the above ensures that @xmath107 satisfies eq . ( [ e.theta ] ) . proving that the candidate factorized ground state @xmath93 is an eigenstate of the total hamiltonian @xmath31 is equivalent to prove that it is a simultaneous eigenstate of all pair hamiltonians @xmath108 , or , more precisely , that the projection of @xmath93 onto the subspace of two given spins @xmath40 and @xmath109 which is still a pure state since @xmath93 is a tensor product of single - spin pure states is an eigenstate of @xmath108 for every pair @xmath110 . to proceed , we associate to each lattice site a set of orthogonal spin operators defined as follows : @xmath111 it is immediate to observe that @xmath112 . by inverting eqs . ( [ e.axayaz ] ) , we can conveniently re - express the standard spin operators as functions of the new set of operators @xmath113 : @xmath114 inserting eq . ( [ e.sxsysz ] ) in eq . ( [ e.couple ] ) we obtain the expression of the pair hamiltonian @xmath115 as a function of the sets of operators @xmath113 and @xmath116 : @xmath117 -2 h_f^r \sin \theta \left [ { a_{\underline{i}}^x \left ( a_{\underline{j}}^z -\frac{1}{2}\right ) + \left ( a_{\underline{i}}^z-\frac{1}{2}\right ) a_{\underline{j}}^x } \right ] \nonumber \\ & & + \ a_{\underline{i}}^x a_{\underline{j}}^x \left ( \sin^2 \theta j_z^r+\cos^2\theta ( -1)^r j_x^r \right)+ a_{\underline{i}}^y a_{\underline{j}}^y j_y^r ( -1)^r - h_f^r \cos \theta \left ( { a_{\underline{i}}^z + a_{\underline{j}}^z } \right ) \ , , \end{aligned}\ ] ] where we recall that @xmath49 is the distance between sites @xmath40 and @xmath109 and that , in the presence of an antiferromagnetic order along the @xmath69 direction , @xmath118 . the spin - pair projection @xmath119 of @xmath93 is an eigenstate of the pair hamiltonian @xmath120 either if all terms in eq . ( [ e.pairhij ] ) admit @xmath119 as an eigenstate or if they annihilate it . by construction , recalling that @xmath121 and that @xmath122 is the eigenstate of @xmath123 with eigenvalue equal to @xmath0 , the spin - pair projection of @xmath93 is an eigenstate of the pair hamiltonian if the following condition holds : @xmath124 for general models of the type eq . ( [ e.hamiltonian ] ) with a given , arbitrary , maximum spatial range of interaction @xmath125 , eq . ( [ e.condition ] ) admits the following solution : @xmath126 for all distances @xmath127 , i.e. for all distances associated with nonvanishing couplings ( we recall that @xmath128 and @xmath129 ) . the set of equations ( [ e.solcondition ] ) determines the conditions that must be satisfied simultaneously by the phase @xmath86 , which is unique , in order for @xmath93 to be an eigenstate of the total hamiltonian eq . ( [ e.hamiltonian ] ) . ( [ e.solcondition ] ) discriminate quite clearly between short - range models , i.e. models with only one nonvanishing coupling at @xmath130 ( models with nearest - neighbor interactions ) and all other possible models containing finite- or infinite - range interaction terms . specifically , models with only nearest - neighbor interactions are characterized by the fact that eqs . ( [ e.solcondition ] ) reduce to a single condition and hence if the value of the r.h.s . of eq . ( [ e.solcondition ] ) falls in the interval @xmath21 $ ] ( associated to the permitted values of @xmath131 in terms of the interaction parameters ) , one can immediately conclude that the models under investigation admit a factorized eigenstate . on the other hand , for models containing interaction terms of longer range so that eqs . ( [ e.solcondition ] ) include two or more conditions factorized energy eigenstates are allowed if and only if the r.h.s . of all conditions in the set ( [ e.solcondition ] ) take values in the interval @xmath21 $ ] and , moreover , they all coincide . summing term by term over the index @xmath49 all the relations in eq . ( [ e.condition ] ) , taking into account eq . ( [ e.theta ] ) , and solving for @xmath63 , one eventually obtains the exact expression of the factorizing field as a function of the net interactions : @xmath132 we remark that eq . ( [ e.factorizingfield ] ) , which was derived in ref . @xcite using some unnecessary auxiliary assumptions , is completely general and holds for lattices of arbitrary spatial dimension and for spin - spin interactions of arbitrary range . accordingly , the angle @xmath86 that determines the direction of the e - squo can be also expressed as a function of the net interactions as follows @xmath133 in the case of a maximum range of interaction @xmath134 , relation eq . ( [ e.theta2 ] ) for the net interactions is a necessary but not sufficient condition for factorization , and further use of eqs . ( [ e.condition ] ) and eqs . ( [ e.solcondition ] ) is needed , as we will show with explicit examples in sec . [ sec.long ] and sec . [ sec.anisotropy ] . we are finally left with the problem of establishing conditions for a factorized energy eigenstate @xmath93 to be indeed a ground state of the system . a very simple sufficient but not necessary condition is that the spin - pair projection of @xmath93 onto the subspace of a pair of spins be the simultaneous ground state of every pair hamiltonian @xmath108 . in fact , given the relation existing between all the pair hamiltonians @xmath108 in eq . ( [ e.couple ] ) and the total hamiltonian @xmath31 eq . ( [ e.hamiltonian ] ) , it follows that the simultaneous ground state of all pair hamiltonians @xmath108 is also ground state of @xmath31 . to proceed , let us re - express the pair hamiltonians eq . ( [ e.pairhij ] ) in matrix form in the basis spanned by the eigenstates of the operators @xmath135 . one has @xmath136 where @xmath137 and @xmath138 . making use of eq . ( [ e.theta2 ] ) and eq . ( [ e.hfij ] ) the eigenvalues of the matrix eq . ( [ e.matrix ] ) can be put in the form @xmath139 \nonumber \\ \varepsilon_1 & = & \frac{1}{8 } \left [ 3 ( -1)^r j_x^r - j_z^r - \left ( { j_z^r -(-1)^r j_x^r } \right)\cos \left ( { 2\theta } \right ) \right ] \nonumber \\ \varepsilon_2 & = & \frac{1}{8 } \left [ - 3\left ( { ( -1)^rj_x^r + j_z^r } \right ) + \left ( { j_z^r-(-1)^r j_x^r } \right)\cos \left ( { 2\theta } \right ) \right]\nonumber \\ \varepsilon_3 & = & \frac{1}{8 } \left [ { - 3(-1)^rj_x^r + 5j_z^r + \left ( { j_z^r- ( -1)^r j_x^r } \right)\cos \left ( { 2\theta } \right ) } \right ] \nonumber \\ & & \end{aligned}\ ] ] the eigenvalue @xmath140 is the one associated to the projection @xmath119 of the factorized eigenstate @xmath93 onto the four - dimensional hilbert space associated to the pair of spins @xmath141 . therefore , the spin - pair projection is the ground state of the pair hamiltonian if @xmath142 . imposing this condition yields the following inequalities : @xmath143 & \ge & 0 \ ; ; \label{e.suffcondition.1 } \\ ( -1)^r j_x^r - j_z^r & \le & 0 \ ; . \label{e.suffcondition.2}\end{aligned}\ ] ] notably , these inequalities are automatically satisfied for any interacting spin system that is not frustrated and that verifies all of the eqs . ( [ e.solcondition ] ) . namely , for every even value of the distance @xmath49 in the interaction range ( @xmath127 ) eq . ( [ e.suffcondition.2 ] ) implies that the denominator in the corresponding eq . ( [ e.solcondition ] ) is non - positive . therefore , an acceptable value of @xmath131 can be obtained only if also the numerator is non - positive and exceeds or equals the denominator . these conditions in turn imply the order relation , @xmath144 . on the other hand , comparing eq . ( [ e.suffcondition.2 ] ) and eq . ( [ e.suffcondition.1 ] ) , one has that @xmath145 . therefore , comparing the two order relations yields finally @xmath146 ( @xmath49 even ) . this condition is always verified in systems with an antiferromagnetic order along the @xmath69 direction and in absence of frustration and therefore the projection of the factorized energy eigenstate on the hilbert space of every pair of spins is indeed the ground state of every pair hamiltonian . on the other hand , in the case of odd @xmath49 , eq . ( [ e.suffcondition.2 ] ) implies that the denominator of eq . ( [ e.solcondition ] ) is non - negative and , similarly to the former case , the numerator must be non - negative and not exceeding the denominator . hence we obtain @xmath147 while from eq . ( [ e.suffcondition.1 ] ) we recover @xmath148 that , in turn , implies @xmath149 . again , such inequalities are always verified in the presence of an antiferromagnetic order and in the absence of frustration . [ cols="^,^,^,^ " , ] collecting all these results , we have proved that the following holds : * theorem*. _ for any cooperative system of spin-@xmath0 particles described by hamiltonians @xmath31 of the type [ eq . ( [ e.hamiltonian ] ) ] , characterized by an anti - ferromagnetic order along the @xmath69 axis as emerging from eq . ( [ fourcases ] ) and in the absence of frustration , the simultaneous verification of all eqs . ( [ e.condition ] ) is necessary and sufficient for the fully factorized state eq . ( [ e.candidate ] ) to be the exact ground state of @xmath31 when the external magnetic field takes the value @xmath101 determined by eq . ( [ e.factorizingfield])_. this central result holds as well , with obvious modifications , when one considers the other different ordered phases that form the set identified in eq . ( [ fourcases ] ) . in tab . [ t.summary ] we summarize and compare results for each of the four different possibilities . before moving to apply the general method to specific models and examples of conceptual and physical relevance , we discuss briefly the meaning of factorization and its marked differences with the phenomenon of saturation " ( see below ) . as we have seen , and as it will appear even clearer in the discussion of the examples in the following sections , the physical mechanism of ground state factorization in quantum spin systems is due to a delicate kind of balancing " between the coupling strengths that regulate the intensity ( and range ) of the interactions and the aligning effect of the external field , that tends to orient all spins along a given direction and to destroy all quantum correlations . the remarkable aspect of factorization is then that it occurs , according to well defined conditions and constraints , at precisely defined , finite values of the couplings and of the fields in an ordered ( or symmetry - broken ) phase . this means that factorization occurs when the system is relatively strongly interacting and the external aligning field is relatively weak . moreover , we will see that the factorization point is always a precursor from below ( from the ordered phase ) of the quantum critical point , as first observed for the @xmath1 model with nearest - neighbor interactions @xcite . saturation is the phenomenon of ground state factorization that occurs trivially when the value of the external field grows unboundedly compared to all other hamiltonian parameters . therefore its aligning effect on the spins prevails against all other effects and quantum correlations are suppressed . the main physical difference between proper factorization and saturation is then clear and is reflected at the level of the hilbert space of states . namely , suppose that the system under study admits a fully factorized state that is the true ground state of the system at @xmath101 . then , when one moves away from the factorizing field , i.e. when @xmath150 , the factorized state ceases to be an eigenstate of the hamiltonian . on the contrary , in the presence of saturation , the trivially ensuing factorized state is always an eigenstate of the hamiltonian and remains the ground state of the system when perturbing the value of the external field @xmath44 . let us formalize the above discussion . consider a system that admits a fully factorized ground state . in this case the pair hamiltonian in eq . ( [ e.pairhij ] ) admits as eigenstate a factorized spin - pair state . however , eq . ( [ e.pairhij ] ) is the expression of the pair hamiltonian only at @xmath101 . if we consider a value of the external field different from the factorizing field we must include in eq . ( [ e.pairhij ] ) a term proportional to @xmath151 . obviously , for nonvanishing values of @xmath152 the presence of such term prevents the factorized state from being an eigenstate of the hamiltonian of the system . on the contrary , if @xmath153 and hence if @xmath154 the extra hamiltonian term vanishes and the factorized state @xmath155 remains an energy eigenstate of the system for all values of the external field . this result can be understood from a different point of view . when @xmath156 , from eqs.([e.solcondition ] ) we immediately obtain that @xmath157 , which implies that the total magnetization along the @xmath45 axis is preserved , @xmath158=0 $ ] . therefore the hamiltonian of the system and the total magnetization operator admit a complete set of common eigenstates , discriminated by the eigenvalue of the total magnetization : the factorized state is the one ( and only one ) characterized by the maximum eigenvalue . we now move on to show that the factorized eigenstate , in the case of saturation , is the system s ground state in an infinite range of values of the external field . here and in the following we recall that since the external field is directed along the conventional @xmath45 direction , only saturation with spin alignment along the @xmath45-axis is allowed . at very low values of @xmath44 the ground state of the system will possess a small total magnetization along the field direction . however , as @xmath44 increases , also the total ground - state magnetization along the @xmath45 direction will increase until all spins align with the field and saturation occurs . by increasing @xmath44 further , it is not possible for the system to evolve in a state with higher values of the magnetization along the field direction . hence , the factorized state remains the ground state of the system for all values of the external field @xmath159 , where by @xmath160 we denote the value of the field for which the system reaches saturation ( `` saturating field '' ) . let us consider the matrix representation , similarly to eq . ( [ e.matrix ] ) , of the pair hamiltonian at arbitrary values of @xmath44 . taking into account the condition @xmath153 we have @xmath161 it is immediate to verify that if @xmath162 , the projection of the factorized state onto the considered pair of spins is associated to the lowest energy . re - summing over all values of the spatial range index @xmath49 and taking into account the coordination number @xmath68 of the lattice , we have that the factorized state is the ground state of the total system if @xmath163 eq . ( [ e.hfsaturation ] ) is the rigorous condition for saturation . we notice that it coincides with eq . ( [ e.factorizingfield ] ) in the limit @xmath164 , meaning that in this simple instance no ground - state factorization occurs other than saturation . in the following sections we apply the general theory of factorization to different quantum spin models . we will derive a series of exact results concerning the occurrence of factorized ground states in systems with various types of two - body interactions of different spatial ranges defined on regular lattices of arbitrary spatial dimensionality . the general method enables to determine many novel exact factorized ground state solutions as well as the existence and the nature of ordered phases in models that are , in general , non exactly solvable . moreover , it allows to recover quite straightforwardly the existing analytical and numerical results . in the process , we will come to appreciate that ground state factorization is a phenomenon more common than previously believed . determining the existence of exact factorized solutions in non exactly solvable models should in turn allow to envisage controlled approximation schemes , e.g. perturbative or variational , to investigate the physics of non exactly solvable quantum spin models in the vicinity of quantum factorization points . in this and in the following sections we will consider quantum spin models of the @xmath1 type eq . ( [ e.hamiltonian ] ) defined on regular lattices of arbitrary dimension . let us first consider models with short - range interactions including only nearest - neighbor couplings : @xmath165 as in sec . [ sec.method ] we restrict ourselves to the case in which such models are defined on regular lattices of a generic dimension , whose geometry does not induce any frustration , i.e. in which any closed loop is characterized by an even number of spins . the absence of geometrical frustration together with the fact that all the couplings beyond the nearest neighbor spins vanish , ensure that it is always possible for the system to satisfy all the conditions on the net magnetic interactions . therefore , as shown in sec . [ sec.method ] , if the r.h.s . of eq . ( [ e.solcondition ] ) or , better , if the r.h.s . of the equation relative to each possible magnetic order ( see tab . [ t.summary ] ) takes an acceptable value ( @xmath166 ) then the system admits a fully factorized ground state . in the case of antiferromagnetic ordering along the @xmath69 direction , which is obtained if @xmath167 , from eq . ( [ e.solcondition ] ) it follows that a factorized ground state exists if @xmath168 . analogous conditions are obtained for the other types of magnetic orders compatible with full ground state separability . the results are summarized in fig . [ f.factor ] , where @xmath169 is set assume positive values for reference . a very similar diagram of factorization holds in the case @xmath170 . and the coupling @xmath169 is set to be positive for reference . analogous results hold in the case @xmath170 . blank regions : no factorization allowed . rightmost ( yellow online ) shaded region : presence of a fully factorized ground state supporting an antiferromagnetic order along the @xmath69 direction . leftmost ( light blue online ) shaded region : fully factorized ground state with a ferromagnetic order along @xmath69 . bottommost ( light red online ) shaded region : fully factorized ground state with a ferromagnetic order along the @xmath71 direction . topmost ( light green online ) shaded region : fully factorized ground state with an antiferromagnetic order along @xmath71 . the solid line @xmath172 corresponds to models that exhibit saturation rather than true factorization ; the dashed line @xmath173 accommodates for models with unresolved superpositions of two fully factorized ground states each one supporting a different magnetic order . either one of the two possible ground states is then `` chosen '' by the system depending on the way the symmetry is broken , see text for further details . all plotted quantities are dimensionless.,width=321 ] as one can see from fig . [ f.factor ] , the type of magnetic order corresponding to the factorized ground state determines four different regions in the space of the hamiltonian parameters ( one for each different ordering ) associated to the existence of a factorized ground state . the four different domains of the interaction parameters are separated by two bisectrices that individuate special cases . the line @xmath172 is constituted by the set of points in which saturation occurs in place of genuine factorization ( see subsec . [ s.sec.saturation ] ) . on the other hand , the line @xmath174 , corresponds to models that , for a given value of the factorizing field @xmath63 and of the e - squo orientation @xmath86 , admit two degenerate , but distinct factorized states . the choice of the ground state is then determined by the usual mechanisms of symmetry breaking : if an infinitesimal magnetic field is applied in the @xmath55 plane , the axis of magnetic alignment in the ground state is determined by the order in which the two limits @xmath175 and @xmath176 are realized . it is important to observe that all the shaded regions of fig . [ f.factor ] correspond to exact analytical ground - state solutions of the investigated short - range models , which are , in general , non - exactly solvable . the exact , fully factorized ground states are tensor products of the form reported in the fourth column of tab . [ t.summary ] , in which the value of the orientation parameter @xmath86 of the e - squo is retrieved by solving the equation in the second column of tab . [ t.summary ] . obviously , such an exact solution holds only at the factorizing field @xmath101 . in the case of antiferromagnetic order either along the @xmath69 or the @xmath71 direction , the latter is given by ( see tab . [ t.summary ] ) @xmath177 while in the case of a ferromagnetic phase , again unrespectfully if along @xmath69 or along @xmath71 , it reads @xmath178 given that the ground state is known exactly at the factorization point , all the relevant physical observables can be evaluated straightforwardly . for instance , the ground state energy density at factorization reads @xmath179 in the presence of an antiferromagnetic ordering , while in the ferromagnetic case it is @xmath180 both the factorizing field and the ground state energy , unlike the e - squo orientation @xmath86 and the expression of the factorized ground states , are functions of the number of nearest neighbors per site ( the coordination number @xmath181 , where the index @xmath182 denotes the nearest - neighbor range of the interactions ) . hence , they depend on the geometry and the spatial dimension of the lattice . to begin with , let us consider one - dimensional lattices ( spin chains , @xmath183 ) , for which the coordination number @xmath184 . substituting this value in the previous equations , we immediately recover , among others , the value of the factorizing field and the ground state energy obtained analytically by kurmann _ et al . _ for the @xmath1 spin chain using the direct method @xcite . et al . _ had to restrict their investigation to coupling constants @xmath185 taking values in the interval @xmath21 $ ] , thus establishing ground state factorization only in a restricted set of conditions . instead , the general method allows to assess the occurrence of factorized ground states associated to different types of magnetic orderings in the entire region of the hamiltonian parameters . one of the most important advantages of the general method is that it is easily applicable irrespective of the spatial dimensionality , while the direct method becomes increasingly difficult to apply with growing spatial dimensions of the lattice . numerical techniques can overcome some of the limits of the direct method , although they fall short of the power and extension of the general analytic method . the most important numerical result has been obtained by applying quantum monte carlo techniques to the study of the @xmath1 model in @xmath186 space dimensions ( square lattice ) @xcite . applying the present method to the @xmath1 model on the square lattice allows not only to immediately reobtain analytically the previous numerical findings , but also , again , to extend the investigation to the entire space of the hamiltonian parameters by simply taking into account that for the square lattice @xmath187 . moreover , the flexibility of the general method allows to include in the analysis different planar models . for instance , considering models defined on planar hexagonal lattices , the previous analysis needs to be modified only in that the value of the coordination number , in this case is @xmath188 . let us now move to @xmath189 and investigate ground state factorization in nearest - neighbor @xmath1 models on three - dimensional regular lattices , for which there were so far no results ( apart the preliminary ones obtained in our previous work @xcite ) either analytic or numerical ( such a task would be extremely demanding if approached either via the direct method or via numerical techniques ) . resorting to the general method , we can instead fully characterize the factorized ground state and the associated physical quantities just by entering the correct value of the coordination number , e.g. @xmath190 for a simple cubic lattice . obviously , the results extend straightforwardly also for models defined on lattices of arbitrary higher dimension @xmath191 . the analytic method can be applied with minor complications to models with interactions of longer spatial range . as already mentioned in sec . [ sec.method ] , concerning ground state factorization , the main difference between models with short- and finite - range interactions is that for the latter there exist more than one equation in the set ( [ e.solcondition ] ) . each of these equations is associated to a different value of the range @xmath127 , where @xmath125 is the maximum interaction range , that is the maximum distance between two directly interacting spins : all the couplings vanish for pairs of spins with inter - spin distance greater or equal to @xmath192 . clearly , the fact that all the equations have to be solved simultaneously yields tighter conditions on factorizability compared to the case of models with nearest - neighbor interactions . namely , it is not sufficient that the r.h.s . of each eq . ( [ e.solcondition ] ) takes values in the interval @xmath21 $ ] . it is also necessary to require that all these values coincide . this constraint yields a set of further conditions that must be satisfied by all the nonvanishing couplings . for instance , let us suppose that the system is in an antiferromagnetic ordered phase along the @xmath69 direction . labeling by @xmath193 the value of @xmath131 and imposing that for every @xmath49 the r.h.s . of eqs.([e.solcondition ] ) must equate @xmath193 , we find that @xmath194 where , to ensure the absence of frustration , @xmath195 must obey the following inequality : @xmath196 . analogously , for an antiferromagnetic order along @xmath71 : @xmath197 and to exclude frustration we must also have that @xmath198 . in the presence of a ferromagnetic phase , respectively for an ordering along @xmath69 and along @xmath71 , the constraints and the no - frustration condition become @xmath199 eqs . ( [ e.exe.cantix][e.exe.cferroy ] ) generalize the instance analyzed in ref . @xcite , in which we limited the analysis to the case of vanishing @xmath169 that , according to eqs . ( [ e.exe.cantix][e.exe.cferroy ] ) , implies @xmath200 for all @xmath127 . especially in the presence of a ferromagnetic order in the lattice , eqs.([e.exe.cferrox][e.exe.cferroy ] ) allow us to simplify the expressions of both the factorizing field and the energy per site . namely , taking into account the definition of the ferromagnetic net interactions eq . ( [ e.netinteractions ] ) and the expression of the factorizing field in the presence of a ferromagnetic order ( see tab . [ t.summary ] ) , we have @xmath201 respectively for an order along the @xmath69 and the @xmath71 directions . similarly , for the energy density we obtain @xmath202 we now discuss a specific example ( out of many possible ones ) which demonstrates the power of the analytic method in conditions where the more conventional direct approach based on the factorized ansatz of refs . @xcite is completely ineffective without the supplement of a remarkable intuition in guessing the correct forms of the possible factorized ground states . let us consider a @xmath1-type spin model defined on a spatially isotropic cubic lattice of arbitrary size , possessing nearest - neighbor and next - nearest - neighbor interactions , and admitting arbitrarily different couplings along the @xmath69 , @xmath71 , and @xmath45 directions . the coordination numbers associated to the two types of interaction can be straightforwardly computed from the geometry of the problem and are @xmath190 and @xmath203 . suppose the model is deduced from an explicit experimental realization of a spin lattice and we are provided with explicit values of the various interaction strengths . for example , @xmath204 . the application of our `` on - demand '' ground - state factorization search engine is now immediate . from eq . ( [ fourcases ] ) we have that the candidate factorized state must possess antiferromagnetic order along the @xmath69 direction . therefore the general conditions for the existence of a factorization point , as determined by eq . ( [ e.solcondition ] ) , can be summarized as follows @xmath205 this condition must hold together with the frustration - free constraint . it is readily verified that @xmath206 provides a solution to the combined set of necessary and sufficient conditions for ground state factorization . therefore if the experimentalist tunes the external field @xmath44 at the factorizing value @xmath207 , obtained from eq . ( [ e.factorizingfield ] ) , the engineered cubic spin lattice system will relax in a completely factorized ground state of the form given in table [ t.summary ] ( second row , last column ) with @xmath208 . many other applications can be considered to models with arbitrary spin - spin interactions of longer range , and they can be treated with the same simplicity within the framework of the general analytic method . for instance , a very interesting example of a model with infinite - range interactions will be treated in the next section . however , before moving to the study of this case , we should comment on the important issue of the interplay between factorization and frustration in quantum spin systems with multiple spatial scales of interaction . the investigation of general models including many interaction terms of different spatial ranges needs to be carried out with particular care , because the interplay between the different interactions can lead in general to important effects of frustration that tend to suppress the occurrence of ground state factorization . the very delicate and tricky nature of the problem is at the origin of a recent incorrect prediction about the existence of factorized ground states in one - dimensional @xmath1 models on finite rings with long - range interactions , obtained exploiting the direct method @xcite . in ref . @xcite the author , resorting to the direct method , concludes that all the factorized energy eigenstates in one - dimensional long - range @xmath1 models , with interaction terms of spatial ranges that are in integer ratio with the total number of sites , are also the states of lowest energy . however , the proof of this statement obtained using only the direct method requires that there must be no frustration in the system @xcite . unfortunately , the interplay between the interactions of different spatial range can in general introduce frustration effects ( e.g. in the case of antiferromagnetic nearest - neighbor and next - to - nearest neighbor interactions , and so on ) . the existence of these effects nullifies all proofs , based on the direct method , of the existence of fully separable ground states . in fact , immediate counterexamples can be given of fully factorized energy eigenstates that are not ground states . let us consider for instance an @xmath209 model on a finite ring of six spins with nearest - neighbor and next - to - nearest - neighbor antiferromagnetic interactions , and @xmath210 . following ref . @xcite , this model should admit a factorized ground state . in fact , a factorized energy eigenstate with antiferromagnetic order does indeed exist , with an energy @xmath211 . however , solving the model by exact diagonalization yields that the lowest energy is @xmath212 and that the associated ground state is entangled . essentially the same type of counterarguments nullifies a recent claim that the factorized energy eigenstates are the factorized ground states of mixed - spin models of ferrimagnetism of the @xmath1 type @xcite . a general analysis of ground state factorization in ferrimagnetic models with mixed spin-@xmath0-spin-@xmath182 interactions will be presented elsewhere @xcite , based on a proper extension of the formalism of squos to include spin-@xmath182 systems , as sketched in the final part of ref . @xcite . in conclusion , the study of factorization in quantum spin models with many different finite- and long - range interaction terms requires a very careful analysis and the use of the general analytic method whenever frustration effects are present . a rigorous and systematic study of the crucial ( and subtle ) interplay between frustration and factorization in quantum spin systems will be the subject of a forthcoming paper @xcite . a very interesting limit of the cases analyzed in subsec . [ sec.long ] is given by models with infinite range interactions , such as the fully connected or lipkin - meshkov - glick ( lmg ) model @xcite , extensively studied in condensed matter physics @xcite , which is described by the following hamiltonian @xmath213 with @xmath214 . the pre - factor @xmath215 ( with @xmath2 denoting the total number of spins ) ensures the linear divergence of the ground state energy as @xmath2 increases . let us remark that normalizing the interaction part of the lipkin - meshkov - glick hamiltonian requires some care : being a connected interaction , it scales as @xmath216 , i.e. as the product of the total number of spins and the number of pairs of spins . therefore , to ensure that the interaction energy is extensive , i.e. scales with @xmath2 , the interaction part of the hamiltonian must be normalized by @xmath29 . usually , for calculational ease , the interaction part is normalized by @xmath2 , so that it scales as @xmath29 . using this convention is strictly speaking incorrect , but is clearly harmless for large @xmath2 and in the thermodynamic limit . for instance , concerning the separability of the ground state , it introduces a slight , spurious @xmath2-dependence of the factorization point through the multiplicative factor @xmath217 , and the error vanishes in the thermodynamic limit . however , on fundamental physical grounds , the factorization points of translationally invariant systems , if they exist , must be independent of the size of the system , and must take the same value both at finite @xmath2 and in the thermodynamic limit . applying the kurmann - thomas - mller factorized ansatz @xcite and the @xmath218 normalization , dusuel and vidal obtained an analytic expression of the factorizing field for the lmg model that is @xmath2-dependent through the multiplicative factor @xmath217 @xcite . although defined according to an anomalous convention , this result is essentially exact in the limit of large @xmath2 . we first observe that the lmg model in eq . ( [ e.exe.hlmg ] ) can be obtained as the limiting case of a generic model described by eq . ( [ e.hamiltonian ] ) assuming that the range of interactions diverges ( i.e @xmath219 ) while the individual strengths become independent of the distance ( @xmath220 ) , scale with @xmath2 ( @xmath221 ) , and vanish along the @xmath45-direction ( @xmath222 ) . therefore , assuming @xmath223 , @xmath224 and @xmath222 in eq . ( [ e.hamiltonian ] ) we recover eq . ( [ e.exe.hlmg ] ) . because of the sign of the interactions and of the range of values that can be assumed by @xmath225 , we immediately obtain that the energy per site ( [ e.energy ] ) is minimized by a state possessing ferromagnetic order along the @xmath69 direction . moreover , taking into account that the coupling strengths are proportional to the ratio @xmath218 , we have that the net interactions converge to finite values in the thermodynamic limit : @xmath226 , @xmath227 . it is then rather straightforward to apply the analytic method to the lmg model and hence prove , exactly , that the ground state is fully factorized at the factorizing field @xmath228 , which is , correctly , independent of @xmath2 . the form of the fully separable ground state is the tensor product of local states of the form @xmath229 with @xmath230 and the energy density reads @xmath231 . these results are in agreement with those obtained , using the direct method , by dusuel and vidal @xcite , modulo a proper normalization factor . the versatility and generality of the analytic method can be demonstrated further by considering extensions of the translationally - invariant models discussed in the previous sections . in fact , till now , we have always considered systems in which the coupling amplitudes depend only on the distance between the spins involved . in the present subsection we shall focus instead on models which contemplate the possibility of spatial anisotropies , i.e. given the spin @xmath232 associated to site @xmath40 , the interaction coupling with spin @xmath233 associated to site @xmath109 depends not only on the distance @xmath234 but also on the location of site @xmath109 relative to site @xmath40 . to simplify the notations , the physical analysis , and the number of possible cases that need to be considered , we limit ourselves to models with short - range interactions , i.e. with all the couplings vanishing for @xmath235 . we will comment briefly on general models with spatial anisotropies and interactions of arbitrary range at the end of this section . let us assume that for each site @xmath40 the total number of nearest neighbors is @xmath236 , where @xmath237 is the number of nearest neighbors whose coupling strength with @xmath232 takes a certain value @xmath238 , and @xmath239 is the number of nearest neighbors whose coupling with @xmath232 is given by a different value @xmath240 . obviously , we could consider situations more complicated at will , with an arbitrary number of @xmath241 different types of nearest neighbors . for the given example , the system hamiltonian reads @xmath242 , \nonumber \\\end{aligned}\ ] ] where the index @xmath243 runs over the first @xmath237 nearest neighbors , and the index @xmath244 runs over the remaining @xmath239 nearest neighbors for each spin @xmath232 . according to the different values of @xmath237 and @xmath239 this hamiltonian describes various possible models . for instance , if @xmath245 and @xmath246 eq . ( [ e.exe.hamiltonian.anis ] ) describes @xmath1 models on ladders constituted by two coupled linear chains ; the case @xmath247 and @xmath246 corresponds to models defined on two coupled square lattices . in both cases , every spin in each chain ( plane ) , besides the usual interactions within a single chain ( plane ) involving @xmath237 nearest neighbors , is also coupled to @xmath239 companion spins located on the second chain ( plane ) . the instance @xmath248 corresponds to a @xmath1-type model defined on a linear chain with nearest - neighbor interactions of alternating strengths . the factorization properties of this latter model have been studied by g. l. giorgi in the limit @xmath249 that , belonging to the @xmath209 symmetry class , can be solved exactly @xcite . as we have seen in sec . [ sec.method ] , the first step of the method is to single out the magnetic order that minimizes the energy associated to the candidate factorized ground state , at fixed magnetization along the @xmath45 direction ( the direction of the external field ) . provided there are no effects of frustration , in the present case we obtain that every possible hypothetical factorized ground state can assume one of eight different magnetic orders . these eight possibilities stem from the four possible orders of eq . ( [ fourcases ] ) , that the system can support when the effects of @xmath240 can be neglected , duplicated according to the two possible arrangements , parallel or antiparallel , induced by the interactions associated to @xmath240 . the four possible orderings along a single direction are represented in fig . [ f.disegno ] for a model of two transversally coupled one - dimensional spin chains . or @xmath71 ) characterizing the candidate factorized ground states of models with spatially anisotropic interactions . the case considered corresponds to a ladder of two coupled one - dimensional spin chains with coordination numbers @xmath245 and @xmath246 . topmost row : parallel ( a ) and antiparallel ( b ) ferromagnetic order . bottommost row : parallel ( c ) and antiparallel ( d ) antiferromagnetic order.,width=321 ] by imposing the minimization of the energy per site , eq . ( [ e.energy ] ) , the kind of order actually present in the ground state can be determined comparing the antiferromagnetic and ferromagnetic net interactions along the @xmath69 and @xmath71 axes , defined as @xmath250 where the @xmath84 sign discriminates between parallel / antiparallel ordering with respect to @xmath251 . as a function of the net interactions , the type of magnetic order present in the system is determined by the value of the minimum , i.e. of @xmath252 . accordingly , we have the following correspondences : @xmath253 obviously , also in the presence of spatial anisotropies we may observe that , in terms of the net interactions , the fact that in the system there is no frustration ( and especially no frustration arising from the spatial anisotropies ) implies that latexmath:[$\mu = - z_1^{(1 ) } according to which axis is characterized by the given magnetic order . without loss of generality , let us consider the situation in which the system realizes a parallel antiferromagnetic order along the @xmath69 direction , see fig . [ f.disegno](c ) . all other cases can be treated analogously . following once again the steps described in sec . [ sec.method ] , and in complete analogy with eq . ( [ e.condition ] ) , it is possible to derive the set of conditions that the orientation @xmath86 of the e - squo must satisfy simultaneously in order for the system to admit a factorized energy eigenstate : @xmath255 for the two conditions to be satisfied simultaneously the two r.h.s . must coincide and must assume values in the interval @xmath21 $ ] . hence , by equating them it is possible to derive an equation that plays the same role of eqs . ( [ e.exe.cantix][e.exe.cferroy ] ) . in the hypothesis that the model under investigation satisfies both eqs . ( [ e.exe.theta.anis.1][e.exe.theta.anis.2 ] ) , what is left to prove is that the associated factorized state be indeed the ground state . following the route illustrated in sec . [ sec.method ] it is not difficult to verify that the following inequalities must hold : @xmath256 it is straightforward to establish that they are both satisfied , provided there are no frustration effects . namely , for spins interacting via couplings of the first kind @xmath238 ( referring to fig . [ f.disegno ] , these are the intra - chain interactions ) , the second of eqs . ( [ e.exe.suffcondition.anis.1 ] ) yields that the denominator in the r.h.s . of eq . ( [ e.exe.theta.anis.1 ] ) is non - negative and , therefore , the corresponding numerator must be non - negative and not exceeding the denominator . hence , we obtain @xmath257 . on the other hand , from the first of eqs . ( [ e.exe.suffcondition.anis.1 ] ) it follows that @xmath258 which , in turn , implies @xmath259 . the latter relation is in agreement with the hypothesis of absence of frustration . furthermore , for pairs of neighboring spins interacting with a coupling of the second kind @xmath240 ( referring to fig . [ f.disegno ] , these are the inter - chain interactions ) , the second of eqs . ( [ e.exe.suffcondition.anis.2 ] ) yields that the denominator in the r.h.s . of eq . ( [ e.exe.theta.anis.2 ] ) must be non - positive . hence , also the corresponding numerator must be non - positive and therefore @xmath260 . simultaneously , the first of eqs . ( [ e.exe.suffcondition.anis.2 ] ) implies @xmath261 and therefore @xmath262 , again in complete agreement with the hypothesis of absence of frustration . collecting all these results we can conclude that in the absence of frustration , if a system described by the model hamiltonian eq . ( [ e.exe.hamiltonian.anis ] ) satisfies simultaneously all conditions in eqs.([e.exe.theta.anis.1][e.exe.theta.anis.2 ] ) , there exist a factorizing field @xmath63 at which the ground state is fully factorized and characterized by a parallel antiferromagnetic order along @xmath69 . the factorizing field is determined according to the procedure described in sec . [ sec.method ] , and it is not difficult to verify that it reads : @xmath263 the exact form of the factorized ground state reads @xmath264 in eq . ( [ e.antiferroparrallelolungox ] ) , the superscript `` @xmath265 '' ( `` @xmath266 '' ) denotes the upper ( lower ) chain of the ladder model considered in fig . [ f.disegno](c ) . we observe that @xmath267 does not depend on the choice of the chain , implying that the two generic @xmath4-th spins in both legs of the ladder are in the same state equipped with an antiferromagnetic order with respect to the nearest neighbor sites along the corresponding chain . analogous results hold , with the appropriate trivial modifications , when considering the other possible magnetic orders compatible with factorization ( see eq . ( [ e.exe.eightcases ] ) ) . the present work has been motivated both by the need for exact solutions to complex many - body quantum spin models , in particular by the question `` when is a mean - field solution exact ? '' , and by the necessity to acquire improved knowledge and control on the structure of quantum correlations for potential technological applications of spin systems . we have introduced a simple and rather powerful all - analytic general method to determine the conditions for the existence and the properties of fully factorized ( fully separable ) ground states in translationally invariant quantum spin models with general two - body exchange interactions and subject to external fields . the theory developed in the present paper builds on and extends the scheme originally introduced in ref . it provides a readily useful procedure to establish , in terms of the hamiltonian parameters , whether the ground state is entangled or completely factorized and , in the latter case , what is the exact form of the state , the expression of the ground - state energy and of the magnetic observables , and the type of ordered phase ( magnetic order ) compatible with factorization . the task is achieved by exploiting tools imported from quantum information theory such as the formalism of single - qubit unitary operations introduced in refs . @xcite and the corresponding entanglement excitation energies , whose property of vanishing if and only if a quantum ground state is totally factorized plays a central role in the analysis of ground - state separability . using these tools we have completed the program initiated in ref . @xcite by determining a general set of exact relations and inequalities that , for frustration - free systems , allow to establish rigorously the occurrence of all types of possible factorized ground states . we have applied the analytic method to spin-@xmath0 models with general heisenberg - like exchange interactions of arbitrary range ( short , finite , long ) , either isotropic or anisotropic , defined on regular lattices of arbitrary dimensions . the method is insensitive to the size of the system and applies rigorously in the thermodynamic limit as well as for finite lattices . the method allows to establish novel exact , fully factorized ground state solutions of generally non exactly solvable models , corresponding to nontrivial sets of values of the hamiltonian parameters . key results include , for instance , the rather straightforward determination of the existence and form of factorized ground states in @xmath1-type models with different types of short - range and long - range interactions on cubic lattices , a task that would be of formidable complexity if tackled with numerical techniques or resorting to the simple - minded direct method based on the product ansatz . the complete factorization diagram for nearest - neighbor heisenberg - like anisotropic models might be particularly useful for those technological implementations which employ spin systems as information processors exploiting their ground state entanglement @xcite . we have provided in fig . [ f.factor ] a `` minefield '' map of the unentangled working points that need to be avoided , when engineering the couplings and tuning the external fields , in order to achieve satisfactory quantum transmission performances . on the other hand , there are also alternative tasks in quantum information which instead require as initial working points factorized states , e.g. for engineering , via suitable dynamics , specific classes of long - distance entangled states useful for quantum state transfer and quantum teleportation @xcite or the instantaneous creation of strongly entangled graphs or cluster states involving a mesoscopic or macroscopic number of spins for one - way quantum computation @xcite : in these instances the minefield may turn into a treasure map . the analysis of ground - state factorization would be particularly useful in the study of quantum spin models on open - end lattices . such systems are not translationally invariant , and therefore the analytic method requires to be suitably extended to this type of instances . this generalization is under way and should be in reach in the near future . notably , according to a general theorem by kurmann , thomas , and mller on factorization @xcite , given any hamiltonian of the form eq . ( [ e.hamiltonian ] ) involving higher spins @xmath268 , the ground state of that spin-@xmath269 system is fully factorized at the same value of the external field @xmath101 eq . ( [ e.factorizingfield ] ) , at which factorization occurs in the corresponding spin-@xmath0 model . therefore , the method and the results derived in the present paper have a much wider scope of application and can be straightforwardly generalized to interacting systems with arbitrarily high values of the spin , provided they are of the same hamiltonian structure as in the spin-@xmath0 case . absence of frustration effects , either of geometric or dynamical nature , has been essential to the theoretical scheme derived in the present work . for instance , we have determined the factorization diagram for specific examples of frustration - free systems endowed with isotropic interaction terms of different spatial ranges , and for systems with nearest - neighbor anisotropic interactions . however , it is clear that there are many extremely interesting instances of more complex systems that would in general be subject to frustration effects . continuing a step - by - step strategy of applying the general method to models of increasing complexity , natural further stages of investigation would include models subject to frustration , e.g. , models on linear chains with competing isotropic interactions of different range such as a 1-d @xmath1-type model with nearest - neighbor ferromagnetic interactions and next - to - nearest neighbor antiferromagnetic interactions . increasing further the degree of complexity , one could then consider models with competing anisotropic interactions of different ranges , that according to the type of anisotropy could be subject to both geometrically- and dynamically - induced frustrations . when considering frustrated models , except for very special cases , even when the considered system admits a fully factorized state as an energy eigenstate , the set of conditions derived in the present work are not sufficient to establish that such an eigenstate is indeed the ground state of the system . in fact , a competition arises between factorization , which requires a magnetic ordering , and frustration , which tends to destroy magnetic orders , ultimately leading to a chaotic correlating behavior . the latter unavoidably tends to enhance ground - state entanglement , thus suppressing the occurrence of situations where the most favorable state , in terms of energy content , is completely uncorrelated . in such cases , a naive application of the direct method based on the factorized ansatz , for instance to models including anisotropic interaction terms of different magnetic types on many different spatial scales ( or even on all scales ) , leads unavoidably to overestimate the factorizing effect of the balancing between interactions and external fields and thus to badly incorrect predictions on the occurrence of fully factorized ground states in these classes of models , as done , unfortunately , in the second part of ref . @xcite . in a forthcoming work @xcite we will present a systematic and thorough study of some important classes of frustrated quantum spin models in order to establish the conditions for the occurrence of true ground state factorization and to characterize it in terms of compatibility thresholds with frustration . an intriguing issue that arises in connection with frustration is the relationship between factorization and the existence of ordered phases . for frustration - free hamiltonian systems that conserve the even - odd parity , spontaneous symmetry breaking and the existence of an ordered phase are necessary conditions for ground state factorization , because factorized states have no definite parity . however , in the presence of frustration there may exist factorized energy eigenstates without ground state factorization . it would then be very interesting to understand whether in these situations there is still a causal relation between factorization and the existence of phase transitions in the system . factorization might then be used as an heuristic tool to gain insight in the phase diagram of frustrated quantum systems . it would also be important to understand whether chiral interactions can enhance factorization beyond the limits imposed by the presence of frustration effects . in fact , since factorization can be seen as `` mean field becoming exact '' , it seems to require quite naturally a balancing between interactions and external fields as the only possibility for its occurrence . however , one can not exclude , in principle , that other mechanisms might lead to the same effect without requiring the presence of external driving fields . future investigations will be concerned with the application of the analytic method to the determination of the spectrum of factorization , that is the study of the occurrence of factorized excited states in different classes of quantum spin hamiltonians , and the implications of factorization diagrams on the structural and informational properties of quantum spin models for potential applications in quantum technology . we acknowledge cnr - infm research and development center `` coherentia '' , infn , and isi foundation for financial support . one of us ( f.i . ) , would like to thank ylenia dautilia for invaluable exchanges . in sec . [ sec.method ] we have determined the general conditions for ground state factorization by selecting as candidate factorized ground states only those characterized by a ferromagnetic or antiferromagnetic order along the @xmath69 or @xmath71 directions depending on the value of the minimum value @xmath270 in the set of the net interactions [ see eq . ( [ fourcases ] ) ] , and neglecting all other possible magnetic arrangements . in the following , we will prove that indeed , given any spin hamiltonian of the type eq . ( [ e.hamiltonian ] ) , if a factorized ground state exists , then it is characterized by one of the four above - mentioned magnetic orders . let us consider the most general factorized state compatible with the constraint of a site - independent magnetization @xmath51 : @xmath271 where the local spin state @xmath272 is eigenstate of @xmath273 with eigenvalue equal to @xmath0 . let us suppose that there exists an hamiltonian of the type eq . ( [ e.hamiltonian ] ) that admits , for some particular value of the external field @xmath101 , the state in eq . ( [ e.candidate.gen ] ) as an eigenstate . under these hypotheses we must have that the projection of the factorized eigenstate onto the four - dimensional hilbert space of a pair of spins in the lattice must be an eigenstate of the pair - hamiltonian term @xmath102 defined as in eq . ( [ e.couple ] ) . following the same strategy as in sec . [ sec.method ] let us introduce for each site @xmath274 a set of three auxiliary mutually orthogonal spin operators among which one , that we name @xmath123 , coincides with @xmath275 : @xmath276 by inverting eqs . ( [ e.axayaz.gen ] ) , we can conveniently re - express the conventional single - spin operators as functions of the new set of operators @xmath113 : @xmath277 substituting eq . ( [ e.sxsysz.gen ] ) into the pair hamiltonian we obtain the expression for @xmath115 at factorization as a function of the sets of operators @xmath113 and @xmath116 , which reads @xmath278 -2 h_f^r \sin \theta \left [ { a_{\underline{i}}^x \left ( a_{\underline{j}}^z -\frac{1}{2}\right ) + \left ( a_{\underline{i}}^z-\frac{1}{2}\right ) a_{\underline{j}}^x } \right ] \nonumber \\ & & + a_{\underline{i}}^x a_{\underline{j}}^x \left ( j_z^r-2 h_f^r \cos \theta\right)+ a_{\underline{i}}^y a_{\underline{j}}^y \left ( j_x^r \sin\varphi_{\underline{i } } \sin \varphi_{\underline{j } } + j_y^r \cos \varphi_{\underline{i } } \cos \varphi_{\underline{j } } \right ) - h_f^r \cos \theta \left ( { a_{\underline{i}}^z + a_{\underline{j}}^z } \right ) \nonumber \\ & & + \left ( a_{\underline{i}}^z a_{\underline{j}}^y \cos \theta + a_{\underline{i}}^x a_{\underline{j}}^y \sin \theta \right)\left(-j_x^r \cos \varphi_{\underline{i } } \sin \varphi_{\underline{j}}+ j_y^r \sin \varphi_{\underline{i } } \cos \varphi_{\underline{j } } \right ) \\ & & + \left ( a_{\underline{i}}^y a_{\underline{j}}^z \cos \theta + a_{\underline{i}}^y a_{\underline{j}}^x \sin \theta \right)\left ( -j_x^r \sin \varphi_{\underline{i } } \cos \varphi_{\underline{j}}+ j_y^r \cos \varphi_{\underline{i } } \sin \varphi_{\underline{j } } \right ) \ , , \nonumber\end{aligned}\ ] ] > from eq . ( [ e.pairhij.gen ] ) , one has that to ensure that the projection of the factorized state onto the four - dimensional hilbert space of spins @xmath232 and @xmath233 is an eigenstate of the pair hamiltonian , it is necessary to require that the terms associated to the operators @xmath281 , @xmath282 , @xmath283 and @xmath284 vanish . this condition yields that @xmath285 solving eq . ( [ e.condition2.1.gen ] ) and eq . ( [ e.condition2.2.gen ] ) we obtain that if @xmath286 , then the phases @xmath287 obey to the following relations : @xmath288 with @xmath241 integer . it is immediate to verify that these relations are consistent only with a ferromagnetic or with an antiferromagnetic order along the @xmath69 or the @xmath71 direction . this concludes the proof . g. vidal , j. i. latorre , e. rico , and a. kitaev , phys . * 90 * , 227902 ( 2003 ) ; v. e. korepin , phys . lett . * 92 * , 096402 ( 2004 ) ; g. c. levine , phys . rev . lett . * 93 * , 266402 ( 2004 ) ; g. refael and j. e. moore , phys . 93 * , 260602 ( 2004 ) ; p. calabrese and j. cardy , j. stat . mech . * p06002 * ( 2004 ) ; m. b. plenio , j. eisert , j. dreiig , and m. cramer , phys . rev . lett . * 94 * , 060503 ( 2005 ) ; m. cramer , j. eisert , and m. b. plenio , phys . rev . lett . * 98 * , 220603 ( 2007 ) ; j. eisert , m. cramer , and m. b. plenio , arxiv:0808.3773 . g. vidal , phys . * 99 * , 220405 ( 2007 ) ; m. rizzi , s. montangero , and g. vidal , phys . rev . a * 77 * , 052328 ( 2008 ) ; v. giovannetti , s. montangero , and r. fazio , phys . rev . lett . * 101 * , 180503 ( 2008 ) . s. bose , phys . lett . * 91 * , 207901 ( 2003 ) ; m. christandl , n. datta , a. ekert , and a. j. landahl , phys . lett . * 92 * , 187902 ( 2004 ) ; d. burgarth and s. bose , phys . rev . a * 71 * , 052315 ( 2005 ) ; a. kay , phys . rev . a 73 , 032306 ( 2006 ) . y. makhlin , g. schn , and a. shnirman , rev . * 73 * , 357 ( 2001 ) ; v. giovannetti , s. lloyd , and l. maccone , science * 306 * , 1330 ( 2004 ) ; f. dellanno , s. de siena , and f. illuminati , phys . rep . * 428 * , 53 ( 2006 ) . m. j. hartmann , f. g. s. l. brando , and m. b. plenio , phys . . lett . * 99 * , 160501 ( 2007 ) ; d. g. angelakis , m. f. santos , and s. bose , phys . a 76 , 031805(r ) ( 2007 ) ; a. kay and d. g. angelakis , europhys . lett . * 84 * , 20001 ( 2008 ) . a. srensen and k. mlmer , phys . 83 * , 2274 ( 1999 ) ; a. b. kuklov and b. v. svistunov , phys . lett . * 90 * , 100401 ( 2003 ) ; l .- m . duan , e. demler , and m. d. lukin , phys . * 91 * , 090402 ( 2003 ) .

recently there has been much progress in the study of synchronization of nonlinear maps @xcite and flows @xcite . this may lead to the clarification of the intelligence supposed to come from the synchronization among the neurons in the neural network , e.g.@xcite . especially the globally coupled map lattice ( gcml ) , a system of identical maps coupled via their mean field , has surprisingly rich clustering phases in the space of the nonlinearity parameter and the coupling . in the ordered two - cluster phase , gcml exhibits a posi - nega switch between clusters of synchronizing maps controlled by external inputs@xcite . furthermore even in the turbulent regime there emerges some hidden correlation which was first observed as the violation of the law of large numbers ( lln ) in the fluctuation of the mean field in the time series@xcite . gcml is a natural extension of the spin - glass models@xcite and these interesting features may be relevant to the intelligence activity . all the maps in gcml contribute to the mean field with an equal weight and hence there is no notion of distance between maps . it is conjectured that the intriguing properties of gclm , both the posi - nega switch and the hidden coherence , come from this bona fide scaling invariance of gcml@xcite . however , neither the reason why there occurs complete swapping of the maps between clusters in the posi - nega switch in the two - cluster phase nor the global feature of the turbulent phase has been elucidated so far . the purpose of this article is twofold . firstly , we revisit gcml . we show that there is a natural mechanism which guarantees the complete swapping of maps in the two - cluster posi - nega switch . the key to this mechanism is the motion of the mean field during the transient process . we show that even though whole maps are released from tightly synchronizing clusters to chaotic violent movements during the transient process , still they never mix across their mean field . we clarify the mechanism for this phenomenon . we also perform an extensive survey of the turbulent regime of gcml . we show that this regime has a prominent period three window and is full of periodicity remnants . a correlator analysis detects quasi - periodic clusters with high mixing rate among them almost everywhere in the turbulent regime of the large - size gcml . the so - called hidden coherence occurs in - between the regions of the quasi - clusters and the valleys with no violation of lln as the most modest appearance of the periodicity remnants . secondly , we construct a new extended globally coupled map lattice ( egcml ) . our egcml is a two - dimensional map lattice and the maps are under all - to - all interaction as in the gcml . but instead of the uniform interaction in gcml , the interaction in egcml obeys an inverse power law with respect to the distance between the maps on the lattice . egcml locates at an intermediate position between the nearest neighbor coupling map model and the gcml with uniform interaction . thus egcml facilitates a test whether the salient feature of gcml is due to the scale invariant interaction in it . we find that egcml indeed inherits the intriguing properties of gcml though they turn out diffuse to some extent . for instance , in the two - cluster phase of egcml , the two clusters follow the same bifurcation tree with that of gcml when the population unbalance between the clusters are increased but the synchronization of maps in each cluster is looser in egcml than in gcml . we also observe in the turbulent regime of the egcml the same periodic window as well as the quasi - periodic clusters with high mixing rate . gcml exhibits the basic posi - nega switch as well as more intricate controlled switches among coded clusters at various hierarchical levels in its simplest construction@xcite and certainly serves as a model for intelligence activity . but the very fact that it is a system of globally coupled maps with an equal coupling makes it lose the notion of the distance between maps . in a way it is a zero dimensional chaotic field theory . apart from the intelligence one is interested to know what is the ( classical ) field theory of chaotic maps on the lattice at criticality . in this aspect the synchronizing maps on the lattice may be regarded as the analog of higgs condensation in the field theory@xcite . we can use egcml to see the general trend how the spatial clustering of maps occurs under the scale invariant interaction on the lattice . we find that egcml does exhibit a self - organized formation of the spatial clusters ( condensation ) due to the synchronization between the maps during the chaotic transient process in the switch and after it . in the egcml switch the positive ( negative ) cluster becomes negative ( positive ) evolving in the same two - cluster attractor as before with only a few percent change of the population unbalance between the two clusters in most of the runs . however , there occurs the mixing of maps between the clusters . the organization of this article is as follows . in sec . ii we pay attention to two important rules in the gcml . these are the reciprocity of the couplings among the maps and an invariance rule that the maps do not change their mean field under the interaction . we then construct egcml based on these rules . in sec . iii we compare the phase diagrams of both models and show that they are almost the same each other . in sec . iv we analyze the two - cluster regime of both models . for the gcml we clarify how the basic posi - nega switch between two clusters in the two - cluster regime can be realized . we then turn to egcml and investigate the switch process in it . we show that amazing spatial clusters are formed during the chaotic transient process in the switch . in sec . v we present the results of an extensive statistical analysis over the whole turbulent regime of both gcml and egcml . we show that even in the turbulent regime the maps are under coherence with various manifestations of periodicity effect . we conclude in sec . vi . the coupled map lattice model in general is a system of one - dimensional field on a lattice @xmath1 with number of sites @xmath2 which evolves simultaneously by an iteration of the following two steps : @xmath3 in the first step all @xmath4 on the lattice ( called as map conventionally ) are simultaneously mapped by a nonlinear function @xmath5 . the function @xmath5 could be different site by site but we consider here the simplest case that @xmath5 is a simple logistic map @xmath6 common to all sites . in the second step the maps undergo interaction between themselves . the @xmath7 constant matrix @xmath8 represents the coupling between the maps at @xmath9 and @xmath10 on the lattice and the parameter @xmath11 is an overall coupling constant . thus the model in the simplest form has only two parameters , the nonlinear parameter @xmath12 and the coupling constant @xmath11 . if we rewrite ( [ generic ] ) in terms of @xmath13 , we obtain @xmath14 which is a familiar form in the neural network analysis @xcite . in gcml @xcite we consider n maps @xmath15 which evolve as @xmath16 where @xmath17 is the mean field on the lattice @xmath18 thus the coupling matrix of gcml is simply @xmath19 and clearly satisfies the reciprocity @xmath20 from ( [ global1 ] ) and ( [ global2 ] ) we can also derive a relation @xmath21 this relation guarantees that the mean field value is kept invariant when the maps undergo the interaction at the second step of the interaction . these two relations serve as a guide to construct the map lattice models in general . let us examine the implication of the invariance rule ( ii ) . the nonlinearity of @xmath5 generally magnifies the variance among the maps . therefore , the first step may be regarded as a _ defocusing lens _ with some _ random prisms _ installed in it . the larger the parameter @xmath12 is , the more strongly the defocusing lens acts on the system of maps . in the second step every @xmath22 is pulled to the mean field @xmath17 at a fixed rate @xmath23 . thus the second step acts as a _ focusing lens_. the larger the coupling @xmath11 is the more strongly the focusing lens acts . since ( [ global1 ] ) is a conformal contraction transformation the position of the _ center of mass _ @xmath17 of the system is kept invariant during the interaction . this is what the invariance condition ( ii ) implies . thanks to the invariance condition the fluctuation of the mean field of the maps is caused only at the first defocusing process . the maps in gcml evolve in time repeating this two - step process of defocusing and focusing . under the battle of the two conflicting tendencies the maps eventually form an attractor in which fluctuations among maps is maximally suppressed by the averaging effect of the interaction via the mean field . what is the condition on the coupling matrix @xmath24 for a general map model of the form ( [ generic ] ) satisfies the invariance condition ( ii ) ? by summing up ( [ generic ] ) over @xmath25 and using ( ii ) we obtain a relation @xmath26 at any step of the iteration this must hold . thus the necessary and sufficient condition on the coupling @xmath24 is @xmath27 from reciprocity ( i ) this is equivalent to @xmath28 this type of a model is known to show an interesting spatial cluster formation similar to the spin glass@xcite . on the other hand it is also known that the interesting two features of the gcml , i.e. the posi - nega switch and hidden coherence are lost . nonetheless it is instructive to check that this model satisfies ( i ) and ( ii ) like gcml . let us denote the sub - lattice of @xmath1 as @xmath29 which is a set of the nearest neighbor sites to @xmath25 including @xmath25 itself . the model is @xmath30 where @xmath17 is the mean field of the maps in @xmath29 at @xmath25 , i.e. @xmath31 the coupling matrix is then @xmath32 where @xmath33 is the number of maps in the sub - lattice @xmath34 in dimension @xmath35 . but if @xmath36 , then @xmath37 . hence @xmath38 and the condition ( i ) follows . also we have @xmath39 and ( ii ) follows . note that the interaction represents a diffusion process . for instance in @xmath40 we can rewrite ( [ ising1 ] ) and ( [ ising2 ] ) as @xmath41 and the second term is nothing but the laplacian on the one - dimensional lattice . we now wish to construct the following model . firstly it must have all to all interaction and satisfy the relations ( i ) and ( ii ) in order to have the same defocusing and focusing dynamics as in the gcml . secondly we wish to investigate the consequence of distance dependent interaction ; in particular , we wish to know to what extent the salient features of the gcml such as the posi - nega switch or the hidden coherence depend on the fact that the model is scale invariant trivially by the absence of distance scale . do these features disappear if the interaction between the maps decreases with distance ? certainly at the nearest interaction limit , we know neither the posi - nega switch nor the violation of the law of large numbers occurs . most crucial case is the interaction which decreases with distance but in a power law . in contrast to the exponential law this does not introduce the length scale and yet the interaction is all to all . thus the model we would like to construct is an extended gcml ( egcml ) in which the interaction decays with a power law and satisfies the conditions ( i ) and ( ii ) . below we show there is such a natural extension of gcml . in this article we choose @xmath42 for simplicity though our construction goes through for all dimensions . let us denote the set of sites at distance @xmath43 from @xmath25 as @xmath44 . the symbol @xmath45 is the memento of @xmath46 since in @xmath42 such a set is a loop ( in fact a square on the lattice space ) of radius @xmath43 around @xmath25 . we denote by @xmath47 the number of sites in @xmath48 . for @xmath42 , @xmath49 and @xmath50 . we denote the maximal radius of the loop realized in the lattice @xmath1 as @xmath51 . this is the perimeter of the lattice and for @xmath42 , @xmath52 $ ] . to be precise we present our result for @xmath53 odd but we have checked no change of results occurs for @xmath53 even . in order to treat all of the sites in @xmath1 democratically and to minimize the finite size effect we impose the periodic boundary conditions . for any point @xmath25 on the lattice we calculate the effect of all maps within @xmath51 taking into account the periodic boundary condition . the egcml is then compactly described as @xmath54 where @xmath55 at @xmath25 is given by @xmath56 in words the effect of all the maps on the lattice to the map at @xmath25 is determined by a double average . firstly any one of the maps at an equal distance @xmath43 from @xmath25 must affect the map at @xmath25 with the same weight . thus they are averaged with an equal weight and this gives @xmath57 . then , the effects of maps with various distances from @xmath25 must be accounted for . in the egcml this is done again by taking an equal weight average of @xmath57 over @xmath43 from zero to @xmath51 which gives @xmath55 as a natural non - local extension of the diffusion process . the first average must be an equal weight average to respect the rotational symmetry . the second average must be again an equal weight average . it is necessary for the invariance condition ( ii ) as we show below . the coupling matrix element @xmath24 is the weight factor for the @xmath58 to contribute to @xmath59 via the double average and is given by @xmath60 thus in egcml any map @xmath58 at equal distance @xmath43 from @xmath25 contributes with an equal weight to the mean field @xmath55 and the interaction decreases with an inverse power with respect to @xmath43 for @xmath42 . the reciprocity ( i ) follows just as in the nearest neighbor case since @xmath61 is equivalent to @xmath62 . also the invariance condition ( ii ) can be proved as : @xmath63 the gcml phase diagram in the @xmath64 plane given in @xcite is reproduced in fig . [ fig1]a to facilitate the comparison with the egcml phase diagram . for gcml we analyzed statistically the cluster composition of @xmath65 maps in the final attractor from 1000 random initial configurations and we verified the original phase diagram . the various phases are the outcome of the battle between chaotic motions of maps ( the defocusing effect ) and the interaction via the mean field forcing the maps to move coherently ( the focusing lens effect ) . since the gcml phase structure is elucidated in ref.@xcite we recapitulate here only the essential features necessarily in the following discussion . let us consider a line at @xmath66 for definiteness which is well - beyond the critical value @xmath67 of a single logistic map to the chaos@xcite . for a sufficiently large @xmath11 ( @xmath68 the maps in the final attractor are strongly bunched together in a cluster and evolve as a unity in the same chaotic motion as that of one single logistic map . this is the coherent phase . for @xmath11 in the range @xmath69 $ ] the interaction via the mean field can no longer exerts strong bunching and the final maps divide into two clusters . the maps in each cluster are still tightly synchronizing each other , while the two clusters mutually oscillate opposite in phase . this is as a solution of a minimum fluctuation in the mean field . we discuss this ordered two - cluster phase extensively in sec . iv . for smaller @xmath11 , the number of final clusters increases but the number of clusters at a given @xmath11 does not depend on the number of maps @xmath2 . the typical number of clusters at various @xmath11 ranges is indicated in the phase diagram . for further smaller @xmath11 ( @xmath70 ) the binding of the maps is almost resolved and the number of clusters is proportional to @xmath2 . this region is the turbulent phase and we extensively analyze this regime in sec . v. final remark is that the boundary values of @xmath11 between phases simultaneously increase ( decrease ) in order to maintain the balance between the conflicting tendencies if the coupling @xmath12 is increased ( decreased ) . we show in fig . [ fig1]b the corresponding phase diagram of the egcml . we used @xmath71 map lattice and analyzed statistically the final attractor configuration using typically 200 runs for each @xmath72 . for each map at the lattice site @xmath25 must be individually calculated using the inverse power law . for @xmath73 egcml , the necessary cpu time is about a hundred times of that for gcml of the same size . ] there is a remarkable agreement between two phase diagrams except for slight shifts in the phase boundaries . for instance , for the egcml at @xmath74 , the coherent phase , the ordered two - cluster phase , and the turbulent regime are respectively realized at @xmath75 , @xmath76 and @xmath77 to be compared with the boundaries quoted above . this agreement between two phase diagrams indicates strongly that egcml is in the same universality class@xcite with gcml . the ordered two - cluster phase of gcml is realized in a band in the @xmath79 parameter space with @xmath80 $ ] at @xmath81 and @xmath82 $ ] at @xmath83 ( see fig . [ fig1]a ) . in this phase , the maps divide into two clusters after some transient steps for most of the initial values . the clusters oscillate around the unstable fixed point @xmath84 mutually opposite in phase and the maps in each of the two clusters are in tight synchronization . in order to distinguish the two clusters we follow the convention in @xcite and name the cluster which takes a value above ( below ) @xmath84 at an even step @xmath85 from the first of the iteration as the positive ( negative ) cluster . we denote the number of maps in the positive ( negative ) cluster as @xmath86(@xmath87 ) and define the population ratio parameter @xmath88 . the values of maps in the positive ( negative ) cluster are uniformly @xmath89 if the synchronization of maps in each cluster is really tight . the maps are evolving in two clusters repeating the two step process of mapping and interaction and the dynamics of maps is reduced to that of two clusters : @xmath90 and the mean field of maps at time @xmath85 is given by @xmath91 where the second equality comes from the invariance condition ( ii ) for the two - cluster configuration . thus when a set of three numbers , @xmath12 , @xmath11 and @xmath78 is given the attractor of the two clusters is uniquely determined . for the discussion of the posi - nega switch below let us explain these by an explicit sample in fig . [ fig2 ] . in the left box we illustrate the mechanism of eqs . ( [ twocluster1 ] ) and ( [ twocluster2 ] ) and in the right box we show a sample attractor of two clusters realized at @xmath92 , @xmath93 and the corresponding motion of the mean field . the two clusters , represented by white and black circles , evolve in time under the iteration of mapping ( solid arrows ) and focusing by interaction ( dashed arrows ) . the mean field ( a short bar ) is kept invariant under the interaction as dictated by the invariance condition ( ii ) and ( [ twocluster2 ] ) . the first time @xmath94 in this run happened to be even . hence the positive cluster is the white circle moving as @xmath95 and the negative one is the black circle moving as @xmath96 . we see that they move in period four and mutually opposite in phase . the mean field is closer to the negative cluster , which means that the positive cluster is the minority cluster and @xmath78 must be less than half . numerically @xmath97 for this attractor . note that it is probabilistic whether the first time @xmath94 is even or odd . from another initial configuration the same attractor _ modulo translation in time _ may be equally realized . if the translation is odd steps in time we will have the same figure but with @xmath94 is now odd . the dominant cluster ( black circle ) must be then named as the positive cluster and @xmath98 by definition . in fig . [ fig3]a and b we show the attractors of gcml and egcml at @xmath92 and @xmath93 respectively as a function of @xmath78 . we used @xmath99 runs for gcml and @xmath100 runs for egcml . in one run for gcml , @xmath65 maps are iterated from a random initial configuration and to account for the first transient process the values of maps during the @xmath101 steps after the first @xmath102 iteration are plotted at the fraction of maps with @xmath103 . for instance the run shown in fig . [ fig2 ] with @xmath94 even contributes four points @xmath104 at @xmath97 and other four points @xmath105 at @xmath98 . then the run is repeated and the results of all runs are superposed in order to see the dependence of the attractor on the population unbalance between the clusters . note that this is equivalent to plot the maps only at the last even iteration steps at @xmath78 when the results from many initial configurations are superposed . the gcml attractor in fig . [ fig3]a agrees with that found in ref.@xcite . when @xmath78 approaches an upper threshold @xmath106 from below ( or a lower threshold @xmath107 from above ) the attractor undergoes successive period doubling and at the thresholds the maps fall into chaos ( @xmath106=@xmath108 and @xmath109 for @xmath92 ) . for egcml in fig . [ fig3]b the final attractor of @xmath110 maps is measured in a similar way but in order to shorten the large computing time necessary for egcml a routine is included in the iteration program which judges whether the maps have fallen into a periodic attractor of a certain period @xmath111 less than or equal to @xmath112 or not is calculated at every even step @xmath85 and when it stops to vary during ten successive steps the routine is activated at @xmath94 . it calculates the judgment condition @xmath113 and if this is satisfied for some @xmath114 the maps are judged to have fallen into an attractor of a period @xmath111 . if not , the routine is inactivated and the next stabilization of @xmath86 is waited for . ] . if the judgment condition is satisfied the maps are plotted at the fraction of maps with @xmath115 only during the next one cycle of the attractor and the next run is started . this not only shortens the calculation time but also enables the statistical analysis of the frequency of the occurrence of the attractors and of the fluctuation of maps in a given attractor since indefinitely duplicated plotting of the maps is avoided . if on the other hand the maps do not fall into an attractor all through the @xmath102 steps the maps are plotted for the next @xmath101 steps just as in the gcml run . in egcml we find that the maps in a cluster synchronize more loosely than those in gcml . the dispersion of maps in each cluster is approximately 0.1 for all @xmath78 to be compared with the gap size @xmath116 between the two bands of the attractor at @xmath117 . even though the maps do not mix between the clusters in the final stable attractor this fluctuation hides the details of bifurcation structure with respect to @xmath78 in a direct superposition of all runs in terms of the bitmap graphics . thus we have done a statistical analysis of the plot density and removed those points with density less than one percent of the maximum density from the diagram . this procedure is justified because we truncated each run just after one cycle of the periodic attractor so that each run contributes to the points on the attractor diagram with an equal weight . it was a remarkable instance in this work when we observed the same bifurcation diagram with gcml turned out in the central region of the egcml attractor ( @xmath118 $ ] ) as is shown in fig . [ fig3]b . in egcml the synchronization of maps in a cluster becomes looser than that of gcml but still the dynamics of final clusters is common for both gcml and egcml and controlled by @xmath78 . we have already seen above that egcml is endowed with a phase diagram remarkably similar to that of gcml and now we find that the attractors in both models in the ordered two - cluster regime are also similar . the posi - nega switch in gcml uses the ability of @xmath78 to control the attractor . but before delving into the detail discussion of the posi - nega switch let us examine to what extent the reduction of dynamics of maps is realized . if the reduction is really effective down to two degrees of freedom , all @xmath86(@xmath87 ) maps in the positive ( negative ) cluster should have a single value @xmath119 ( @xmath120 ) . then eqs . ( [ global1 ] ) and ( [ global2 ] ) should reduce to ( [ twocluster1 ] ) and ( [ twocluster2 ] ) or equivalently to a matrix coupled two - map model @xmath121 with @xmath122 and @xmath123 . from the remarkable agreement between gcml attractor ( fig . [ fig3]a ) and two - map model attractor ( fig . [ fig3]c ) we see that in the two - cluster phase of gcml the dynamics of the @xmath2 maps reduces to that of two maps for all @xmath78 between the thresholds . in egcml the coupling @xmath24 is not a single constant but it depends on the distance between @xmath25and @xmath124 . the interaction acts on a map at a site @xmath25 via the position dependent mean field @xmath55 which is a double sum given by ( [ tsm2 ] ) and ( [ tsm3 ] ) contrary to a single sum @xmath17 given by ( [ global2 ] ) in gcml . therefore the reduction of egcml to two maps we find in fig . 3b requires an explanation . what is important here is not the uniformness of the couplings but the scale invariance of them . in both gcml and egcml the interaction has no cutoff scale and hence it prevails over the lattice and the maps are forced to synchronize among themselves due to the averaging effect of the interaction via the mean field . thus the possible difference of egcml from gcml comes solely from the variation of the site dependent @xmath125 around @xmath17 . if the variation is small both models will show similar attractors . egcml is constructed in such a way to assure this small variance when the interaction forces the maps to synchronize . first let us consider a simple case where the nonlinear parameter @xmath12 is so small that the chaotic randomness in each map can not overcome the averaging effect . then after iterations all the maps on the lattice will synchronize and take approximately the same values with only a small variance . if we ignore this variance , the value @xmath22 common to all @xmath25 becomes a multiplicative factor to the sum for the mean field and @xmath126 follows from the invariance conditions ( [ unitaritygcml1 ] ) for gcml and ( [ unitarityegcml ] ) for egcml . in gcml the variance is negligible but in egcml the weaker synchronization causes a variance typically to the order @xmath127 in @xmath125 and the egcml attractor turns out reflecting this variance . but otherwise egcml behaves just like gcml . now in the two - cluster phase at higher @xmath12 the maps can not be bunched together in a single cluster and divide into two clusters , a and b. the maps in each of the clusters take approximately the same value and synchronize among themselves . hence we may write @xmath128 here @xmath129 ( @xmath130 ) is the collective coordinate for a ( b ) ( the average of maps taken over a ( b ) only ) and the @xmath131 ( @xmath132 ) represents small fluctuation around the @xmath133 ( @xmath130 ) . now let us analyze @xmath125 for egcml . an important observation is that the double sum for @xmath125 can be written as a sum of two double sums , one only over cluster a and the other only over cluster b. this decomposition allows one to apply the above argument to each of the clusters and the @xmath134 ( @xmath135 ) becomes a multiplicative factor to the double sum for cluster @xmath136 ( @xmath137 ) . as for the remaining double sum of couplings , some more consideration is necessary . in egcml a ` polarized ' distribution of maps is in principle allowed in which , for instance , the maps in the cluster @xmath136 ( @xmath137 ) accumulate around a given point @xmath25 and maps in cluster @xmath137 ( @xmath136 ) accumulate far from @xmath25 . but egcml has no interaction to induce such a polarization and the randomness in each map also acts to break the polarization . we have numerically checked the dominance of unpolarized distributions . therefore the remaining double sum for egcml again counts the number of maps in each cluster just in the same way the single sum for gcml does in ( [ twocluster2 ] ) . thus we have also in egcml @xmath138 here the approximation takes into account both the polarization in fluctuation and the small variance of maps in each cluster due to approximate synchronization . thus the collective coordinates @xmath133 and @xmath130 of the clusters is described by the matrix coupled two - map model and the maps in each of the clusters fluctuate around the collective coordinates . this explains the approximate reduction of egcml to two maps . the posi - nega switch found in gcml@xcite is based on the ability of @xmath78 as the control parameter of the gclm attractor . if maps are successively transported from one cluster to the other by input ( @xmath139 ) the population unbalance can be enhanced and the attractors undergo successive bifurcation . when @xmath78 reaches the threshold the whole maps go into a grand chaotic motion almost bunched together in a unit . see fig . [ fig4 ] for a typical sample of this transition . this state induced by an excess of the last one map is unstable as we see in fig . [ fig3]a that there is no self - organized stable attractor above @xmath106 ( or below @xmath140 ) . thus after some transient time the maps again form the allowed stable configuration , that is , tightly bound two clusters mutually oscillating opposite in phase . just as before let us again label the new cluster as positive if it takes a value above @xmath84 at even step @xmath85 ( counting from the start of iteration ) and the other as negative . an amazing finding in @xcite is that the composition of the new two clusters is limited only in two ways . ( 1 ) all the maps involved in the positive cluster before the chaotic transient are now involved in the negative cluster . ( 2 ) all the maps in the old positive cluster are again in the new positive cluster . the case ( 1 ) is the posi - nega switch . in both cases the maps behave as if they keep a _ memory _ of their former clusters . but how is it possible ? the maps pass the chaotic transient process where many two - cluster states as well as many other channels are open . what is the mechanism that protects the maps from mixing ? before answering this old puzzle let us point out some trivial aspects of the posi - nega switch . first there is no essential difference between case ( 1 ) and ( 2 ) . how long the chaotic transient process persists and in particular whether the duration time is even or odd is ( almost ) probabilistic . it is of course deterministic but highly dependent on the map values at the onset of the transient process . whether a cluster after the transition is positive or negative depend on the parity ( even or odd ) of the duration time by definition and there is no sense to distinguish ( 1 ) and ( 2 ) when the parity is indefinite . second we should note that in the two - cluster phase of gcml with fixed @xmath12 and @xmath11 the self - organized attractor is determined solely by the population unbalance between the two clusters . in case ( 1 ) @xmath141 and in case ( 2 ) @xmath142 . in either case , the population _ unbalance _ is the same before and after the transition and hence the attractor before and after must be the same . to be precise , to put the attractor before the transition upon that after the transition we need a temporal translation of odd steps in case ( 1 ) and even steps in case ( 2 ) . modulo this difference the attractor before and after the transition are the same in either case . there is no chance of appearance of a new attractor _ provided that _ the population _ unbalance _ before and after the transition is the same . therefore the real puzzle boils down to a single question . what is the mechanism which keeps the population unbalance between two clusters unchanged before and after the chaotic transient process ? we solve this puzzle below in two steps . first we discuss how the chaotic transient process starts and ends . this provides the temporal boundary conditions of the transient process . then we show there is a very simple mechanism which transfers the population unbalance rate through the transient process . the logistic map @xmath143 copies the region @xmath144 $ ] onto itself twice - fold as follows the limits are @xmath145 and the unstable fix point is @xmath146 but the argument here is independent of the particular parametrization of the map . all we need is the universal folding nature of the map . ] . i+ii+iii \nonumber \\ i & \equiv & [ -x_l , 0 ] \stackrel{f}{\longrightarrow } i+ii+iii \nonumber \\ ii&\equiv & [ 0 , x^ * ] \stackrel{f}{\longrightarrow } iii \\ iii&\equiv&[x^ * , x_l ] \stackrel{f}{\longrightarrow } ii+i ~. \nonumber \\ \end{aligned}\ ] ] the first two mappings are monotonously increasing while the last is monotonously decreasing . see fig . [ fig5 ] . below the threshold the two clusters are moving mutually opposite in phase around @xmath84 so that one must be in region iii and the other in region i or ii . when the unbalance is increased by sending the maps of the minority cluster to the majority cluster the mean field comes closer to the majority cluster and oscillates in phase with it . the interaction conformally contracts the maps to the mean field at every iteration step , and the minority cluster is pulled to the mean field more strongly at this highly unbalanced situation . in a few iterations after the unbalance exceeds the threshold by the last input , the non - dominant cluster crosses the boundary @xmath84 . this is the very time of the start of grand chaotic transient motion of maps . at the next step of iteration all maps must move into the same direction . the synchronization opposite in phase ceases and the maps start almost coherent evolution . the mean field evolves together in the middle of them . the focusing effect of the interaction now works only to bunch the maps close together around the mean field . the high nonlinearity of each map then becomes manifest and all maps move around the three regions bunched together in a pseudo - cluster with a slight variance around their mean field . in the typical sample run shown in fig . [ fig4 ] the mean field ( depicted by a short sold bar ) oscillates closely to the dominant cluster ( connected by the solid line ) for several steps . at time @xmath148 the minority cluster ( the dashed line ) is pulled down across the @xmath84 and at the next step both clusters are mapped upwards together . then the grand chaotic motion starts at @xmath149 . the pseudo - cluster is formed by the external input and is unstable . to bunch it stably the coupling must be larger ; it must have the value for the coherent region ( e.g. , @xmath150 for @xmath92 ) . the tiny variance between maps in the pseudo - cluster is soon magnified and the transient pseudo - cluster is broken to the self - organized stable two clusters again . in the following we use the fact that the maps starts and exits the transient process with small variance among them . let us define the relative coordinates @xmath151 with respect to the mean field @xmath152 since all of the maps are now close to the mean field we may expand @xmath153 as @xmath154 averaging this over @xmath155 using @xmath156 we obtain a relation @xmath157 but the interaction does not affect the mean field ( the condition ( ii ) ) . so we obtain @xmath158 therefore the evolution equation of the relative coordinates is given by @xmath159 we used ( [ averagerelation ] ) once more in order to derive the third line . the last line is crucial . the factor @xmath160 is common to all maps and in particular it acts with a common sign . thus at every step in the chaotic transient process the maps separated by the mean field never mix each other . this is the mechanism which keeps the population unbalance unchanged before and after the chaotic transient process . the sign of the factor at @xmath85 depends on the mean field value @xmath161 . hence case ( 1 ) and ( 2 ) may occur equally . we note that the invariance condition ( ii ) is used in proving the mechanism . this condition is really important to explore the model which shares the property of gcml . in the above resolution we examined the case in which the maps are confined in a pseudo - cluster and their variance around the mean field is kept small due to synchronization . numerically we can trace the motion of the pseudo - cluster . almost always the maps moves in a pseudo - cluster around the mean field with squared variance less than @xmath162 . but sometimes it intermittently splits into two or more tight clusters around the mean field . this does not invalidate above resolution . the danger of the mixing of members between clusters occurs only when they come close each other . our mechanism gives a guarantee that no mixing occurs even in such emergence . in fig . [ fig6]a we show a posi - nega switch in @xmath163 gcml ( @xmath164 ) . in the upper ( lower ) diagram the maps at the even ( odd ) iteration steps are plotted and connected by lines . a similar figure was given in the first report of the gcml switch@xcite but the evolution of maps only at the even iteration steps was presented . by such a diagram only it looks as though the attractor undergoes a transition to a new periodic attractor after the posi - nega switch . but as we noted above , the attractor before and after the switch is the same modulo temporal transport of odd steps . we can see clearly in fig . [ fig6]a the attractor on the whole , the even and odd iteration steps together , is precisely the same before and after the chaotic transition . as indicated by the arrows the switch from positive to negative is trivially a change in the count ( even or odd ) of iteration steps . what is not trivial is that the memory of the composition of two clusters is kept during the posi - nega switch . in fig . [ fig6]b we present a posi - nega switch event of @xmath165 gcml at even iteration steps only as a three - dimensional landscape plot . the peaks ( valleys ) before the chaotic transient process become the valleys ( peaks ) which is the posi - nega switch . in the transient process the maps form a surface which is chaotically oscillating but at any instance the segment of the surface looks remarkably horizontal . this shows that in the transient process the maps are in quasi - coherent motion with only a tiny variance . in fig . [ fig6]c we show a _ print circuit pattern _ of the same switch event only at even iteration steps distinguishing the maps with respect to the mean field value . those maps with ( @xmath166 ) are shown with white squares and the others by black squares . just as we proved above , the pattern clearly shows that there is no mixing of maps across the mean field all through the switch event . it can be also seen in fig . [ fig6]c that the transient process ceases when the last transferred map ( @xmath167 ) comes back to its home cluster . this is a typical way of the end of gcml switch . this completes our resolution of the posi - nega switch mechanism . the egcml has almost the same phase diagram with gcml as we showed in sec . iii and the attractor of egcml in the two - cluster phase has the same bifurcation tree structure with that of gcml . but as we showed in fig . [ fig3]b the attractor of egcml turns out with some variance . that is , the dynamics of clusters in egcml is reduced to that of matrix - coupled of two maps just as in gcml but the maps in each egcml cluster are in phase - synchronization rather than in tight synchronization in contrast to gcml . the all to all interaction with no cutoff scale in egcml is effective enough to maintain the reduction to two - cluster dynamics for any @xmath78 between @xmath168 $ ] . in the numerical simulation of egcml we find that the two clusters of egcml undergo a switch process by inputs which is similar to that in gcml . a typical sample of the egcml switch is shown in fig . [ fig7 ] . but let us first list the general differences of egcml switch from gcml switch below and add description particular to fig . [ fig7 ] afterwards . \(1 ) _ slower reaction _ : by an injection of a pulse we transfer a map in the minority cluster to the majority cluster . in gcml the reaction of the attractor to this input is quite rapid and the attractor shifts to the new movement only after a few steps of the iteration . on the other hand in egcml the attractor reacts to the input gradually in several iteration steps this has a simple reason . the synchronization of maps in an egcml cluster is looser . when a map is transferred there occurs a negotiation among maps in each cluster to reform the clusters and then the two clusters mutually move to the new attractor . this is similar to the two step phase - synchronization found in the globally coupled flows@xcite . \(2 ) _ self - organized transition _ : in egcml the effect of one pulse depends on the unbalance between clusters at the time of injection . if the unbalance is not high just a slow reaction of the attractor described above occurs . if the unbalance is high the map transportation by an input is followed by a self - movement of another map between clusters in the same direction . this occurs after a few tens of steps for the egcml of the size @xmath169 . if the unbalance is close to the threshold determined by the above statistical analysis ( fig . [ fig3]b ) the transportation of a single map triggers a cascade . the last injected pulse induces follower maps to move from a cluster to the other cluster in the same direction and in some 100 iteration steps the move grows up to a series of avalanches at various scales . the corresponding change of the spatial distribution of positive and negative clusters is similar to a percolation transition . under this process no more external input is given and the maps are themselves forming a metastable state . both the existence of a threshold and the formation of clusters at various scales suggest a remarkable similarity to the self - organized criticality , e.g. @xcite , but further investigation on a larger size lattice , in particular the measurement of the critical exponent , is required to verify the relation . \(3 ) _ the mixing _ : in egcml the two - cluster attractor after the chaotic transient process is for most cases approximately the same with that before . the change of the population unbalance is only a few percent in most of the runs . but there occurs swapping of maps between clusters and the composition of the clusters in terms of maps is often different before and after the egcml switch . the posi - nega switch in gcml is realized by the fact that no mixing of maps across the mean field occur during the chaotic transient process . in egcml the mixing does occur since each map is under the influence of position - dependent @xmath55 . thus the egcml switch is a posi - nega switch as a cluster process but not as the map process . in view of the brain dynamics this difference may be immaterial since what is important is the activity of whole neurons rather than the identity of individual neurons . \(4 ) _ self - organized spatial clusters in egcml _ : before the switch the maps in the positive cluster and those in the negative cluster are almost randomly distributed on the lattice . the transition to chaotic transient process goes like a percolation transition . when the self - organized mass movement starts , the positive maps form several large - size spatial clusters and so do the negative maps . since these spatial clusters are formed by the long - range interaction the formation of them is more remarkable in the lattice of larger size . the spatial clusters can be best traced in the chaotic process by distinguishing the maps by the mean field of whole maps . at short time scale often we observe a posi - nega switch of spatial patterns . but due to mixing the spatial clustering pattern changes at longer time scale . now we explain fig . [ fig7]a which is a typical egcml switch event sampled in @xmath170 lattice ( @xmath171 ) . _ preparation : _ the maps are iterated from random start and a few tens of pulses are injected during the first 1500 iteration steps until the two synchronizing clusters exhibit period 32 oscillation mutually opposite in phase . _ start of chaotic process : _ at @xmath172 we injected ten pulses simultaneously to randomly chosen maps on the lattice . this was necessary to show the whole process in one figure . had we injected the pulses one by one with say a hundred steps in between we had been able to see the accelerated cascade transition as described in ( 2 ) above . but still we can see in fig . [ fig7]a that the process after @xmath173 is a self - organized cascading process . _ the posi - nega switch between clusters : _ at @xmath174 the transient process ceased and a periodic attractor of two clusters appeared . as indicated by arrows we see clearly that it is the same attractor as before . the change of the unbalance of population between the two clusters is less than one percent in this switch . _ the spatial clusters : _ the snap - shot patterns of the maps at the even iteration steps are shown in the boxes below the evolution plot . in the upper boxes we distinguish the maps with respect to the mean field while in the lower boxes we used the unstable fixed point @xmath84 . in either way we can see that spatial clusters are formed by the self - interaction of maps during the chaotic transient process . we see clearly that the maps undergo a posi - nega switch among themselves between snap - shots 4 - 7 in the upper boxes . on the other hand in the lower boxes the pattern becomes sometimes totally black and sometimes totally white . hence , we see that the distinction by the mean field is a better way to sense the spatial cluster formation on the lattice . a distinction by the mean field is sensitive to the fluctuation around the collective coordinate . the clustering pattern extracted by this method is the long - range fluctuation dynamics of maps on the lattice . this long - range interaction dynamics can be seen more remarkably on a larger lattice which allows the formation of the spatial pattern of longer wavelength . in fig . [ fig7]b we show the samples of amazing spatial patterns formed during the chaotic transient process in the run of @xmath175 lattice ( @xmath176 ) . the snap - shots were sequentially taken at every four iteration steps . we observe that a small seed of fluctuation tends to grow into a circular dense pattern by the chaotic diffusion process on the lattice in accord with the isotropic interaction in egcml . it is interesting to note the similarity of the patterns observed on the lattice to the galaxies in the universe and in particular the appearance of patterns like globular nebula . the egcml may be used as a toy model for the formation of density fluctuations in a nonlinear diffusion process , though further study is required to test this speculation . the turbulent regime is generically the region of small @xmath11 and known by the so - called hidden coherence phenomenon . generally it is understood that in this region maps evolve almost randomly at small lattice size @xmath2 and that there occurs the hidden coherence at large @xmath2 . but actually the turbulent regime has a subtler feature . we are going to show that it is a regime where the grand periodic or quasi - periodic motion of the whole maps is in action almost everywhere . but first let us quote briefly previous observations in the literature in three items . \(i ) when gcml is iterated from a random configuration with a small coupling ( @xmath177 ) and a large nonlinear parameter ( @xmath178 ) it reaches a final state which is an ensemble of maps and tiny clusters , each moving chaotically due to high nonlinearity of the map . ( we show below that they are under certain coherence almost everywhere in the turbulent regime even though each element looks in an independent random motion . ) the number of elements increases proportionally to the number of whole maps in sharp contrast to the ordered regime where the dynamics of maps reduces to that of a few clusters@xcite . \(ii ) it was found in the statistical analysis of the mean field fluctuations that there emerges certain coherence between elements for large lattice size @xmath2@xcite . ( we show below some reservation is necessary to this statement . ) let us denote the mean field of the @xmath2 maps on the lattice @xmath1 at time @xmath85 as @xmath179 if @xmath180 are really @xmath2 independent random variables with a common distribution , the mean square deviation ( msd ) of the fluctuation ( @xmath181 ) should decrease proportionally to @xmath182 by the law of large numbers ( lln ) and the distribution of the @xmath183 must be a gaussian for sufficiently large @xmath2 by the central limit theorem ( clt ) . however the time series analysis of the mean field fluctuation shows that there is a threshold @xmath184 in @xmath2 ( depending on both @xmath12 and @xmath11 ) above which msd ceases to shrink even though the distributions remains in a gaussian shape . even though the elements ( maps and tiny clusters ) look evolving independently each other the violation of lln must reflect some hidden coherence between the elements . in fact an introduction of a very small independent noise term in each map restores lln@xcite . \(iii ) this violation of lln observed in the time averaged distribution of @xmath183 reflects that the probability distribution @xmath185 that a map takes a value @xmath186 explicitly depends on time . a noise intensity analysis of ensembles successfully proves lln@xcite . if lln should hold in the time average , @xmath185 would have to be the fixed point distribution of the frobenius - perron(fp ) evolution equation@xcite . it has been shown that the fixed point distribution is unstable due to the periodic windows inherent to the chaotic logistic map@xcite . the coherence manifests itself in the mutual information@xcite . on the other hand the temporal correlation function similar to the edwards - anderson order parameter for the spin glass@xcite decays to zero exponentially . thus it may not be due to freezing between two elements@xcite . all these are important observations for the turbulent regime of gcml . but this is not the whole story . especially one should not simply conclude by the above list that the turbulent regime of gcml ( and also egcml ) is a regime with complete randomness at small @xmath2 and with the hidden coherence for larger @xmath2 . the msd and distributions previously reported @xcite are only for the limited regions of the turbulent regime where the mean field distributions looks semi - gaussian and the violation of lln for large @xmath2 is remarkable . but there do exist regions in the turbulent regime where the coherence is suppressed strongly and the msd obeys lln in a good approximation . oppositely there are also large regions where the final attractors are two or three clusters which evolve stably ( until the end of @xmath187 iterations ) in period three ; a grand periodicity in the turbulent regime . furthermore even in a large part of the hidden coherence region the distribution is not really gaussian ; it is only semi - gaussian ( trapezoidal ) or has visible peaks on top of semi - gaussian distribution . we can detect the remains of the periodic coherent motion by an analysis of the decay of a temporal correlation function for these regions . the turbulent regime is actually the region which is full of periodicity remnants and the hidden coherence is realized in between the genuinely turbulent regions and manifestly periodic regions . the egcml shares the same features with gcml . in the following we show these facts by an extensive analysis over the whole turbulent regime . we measure the distribution of @xmath183 and its msd at every point on the grid on the @xmath188 plane in the turbulent regime at a fixed value of the nonlinear parameter @xmath12 . here we choose @xmath189 as a canonical value but our observations below generally hold for other values of @xmath12 larger than @xmath190 . at @xmath191 the turbulent regime is from @xmath192 to @xmath193 for gcml and from @xmath192 to @xmath194 for egcml ( see fig . [ fig1 ] a and b ) . thus we set a @xmath195 @xmath188 grid where the lattice size @xmath196 varies from @xmath197 to @xmath198 with an increment of @xmath199 two is necessary for two - dimensional egcml with the periodic boundary condition . for gcml we keep the increment the same for the sake of comparison . ] and the coupling @xmath11 varies from @xmath200 to @xmath194 with an increment @xmath201 . at every point on the @xmath188 grid with @xmath12 fixed at @xmath202 , the maps are iterated from a random start and the mean field @xmath183 is measured during the time interval between @xmath203 and @xmath204 . thus the result for one run at @xmath191 is a @xmath195 table of distributions and each distribution is a histogram of @xmath205 @xmath183 . we have repeated the run several times for the same @xmath12 and verified that the general result is independent from the change of the initial configuration of randomly produced maps . we have also checked in two ways that our @xmath188 grid is sufficiently fine to see the general feature of the turbulent regime . firstly we randomly chose more than a hundred squares of the grid and subdivided them . using this finer grid we have verified that either the variation of @xmath11 or @xmath2 within each square does not cause any significant change of the distribution ( a local check ) . secondly the compiled @xmath206 distributions are sufficient to make an animated show of the distribution . we observed by this animation that the distribution changes systematically the shape with variation of either @xmath11 or @xmath2 over the whole @xmath188 grid though the onset of the apparent periodic distribution is quite rapid ( a global check ) . we find that a special care must be taken over for the transient time when we are on the edge of the periodic window region . there the maps may fall sometimes into the final periodic attractor after , for instance , @xmath207 iterations but sometimes only after @xmath208 iterations depending on the random initial values . for this edge region we have done a separate analysis using a sequence of runs , each consists of @xmath204 to @xmath187 iterations . we comment on this region in sec . v d. this simple but tedious analysis distributions for both models . ] was the start of the work to unveil the real feature of the turbulent regime . in fig . [ fig8]a and b we show the msd of the mean field ( @xmath183 ) distributions at @xmath189 as a surface plot over the @xmath209 grid for gcml and egcml respectively . both msd and @xmath2 are plotted in the logarithmic scale so that the linear edge of the surface in the front panel at @xmath192 is of course due to lln . for a non - zero but very small @xmath11 @xmath210 , the lln still holds to a good approximation but for a larger @xmath11 we can clearly see that the surfaces have many peaks . the global features of the msd surfaces may be summarized as follows . * the valleys * : at the valleys of the surface the msd still approximately respects lln and we may call these valleys as genuinely turbulent regions . apart from the valleys lln does not hold . this does not imply the real violation of the lln in the ensemble average@xcite . what is meant by the violation of lln here is that there is a larger fluctuation in the time series of @xmath183 than expected by lln . we are interested in detecting the coherence among the evolving elements by the enhanced msd . for both gcml and egcml the peaks divide into two classes except that the peaks are more parallel each other for gcml . below we quote figures for gcml . * the first - class peaks * : we can see two very high and broad peaks at @xmath211 and @xmath212 and lln is violated extremely on them . here the excess msd is observed also for small @xmath2 . let us call them as the first - class peaks of the msd surface . * the second - class peaks * : separated by the large first - class peaks there are also several minor peaks . their locations for @xmath191 are @xmath213 let us call them as the second - class peaks . provided that one keeps eyes only on the second - class peaks the msd first decreases with increasing @xmath2 following lln and then above certain threshold @xmath214 the msd ceases to decrease and remains approximately in a plateau . this is a characteristic behavior of msd at the hidden coherence . however we show below that the @xmath183 distribution is not a gaussian but is a distorted gaussian in the bulk of the second - class peak regions . now we analyze directly the table of @xmath195 distributions of the mean field @xmath183 at @xmath191 for both gcml and egcml to understand how the distribution looks like for the enhanced msd regions . before this analysis we had thought from the literature@xcite that the distributions are all gaussian with possible msd enhancement at least for the central region of the @xmath183 histogram . distribution is limited in the range @xmath215 $ ] , it can not be a finite width gaussian distribution over @xmath216 . when we discuss whether the @xmath183 distribution is gaussian or not we concern whether the essence of the clt , that the convolution of independent random distributions peaks like gaussian in the central region , is in action or not . the ratio of the fourth to the second moment must be three for a perfect gaussian and for the @xmath183 distribution a small deviation from three is reported @xcite . but it is sensitive to the tail region and the dynamics in the central region is obscured in a ratio . for instance , the rank three distribution below is often two overlapping gaussian peaks due to the periodic clusters with high mixing but a ratio of moments can not tell it . ] but to our first surprise we find that there are a variety of distributions involved in the table of gcml and egcml . we assign each distribution a pattern classification number or a rank from zero to four by its shape . 1 . the distribution is gaussian . compared with the gaussian distribution at @xmath192 with common @xmath12 and @xmath2 , the msd agrees within 20 percent error . 2 . still gaussian but with more than 20 percent increase of the msd . at large @xmath2 we see the msd can become even more than twice . a singly peaked distribution but the shape is distorted from gaussian . typically a gaussian - like peak polarized to the right or left , or trapezoidal . either it has a few sharp peaks on top of a broad band , or it is an apparent overlapping gaussian distribution . it manifestly shows the periodic motion of elements . the distribution consists of a few sharp peaks only . at rank zero the mean field distribution obeys both lln and clt and the maps may be thought as independent random numbers with a common probability distribution . oppositely at rank four the maps are in periodic motion and so is the mean field . the ranks are organized in a way that the periodicity of the elements becomes more manifest with an increase of the rank . in fig . [ fig9]a and b we show the rank distribution on the @xmath188 grid as a density plot for gcml and egcml respectively . we find the followings . \(1 ) there is a remarkable match between the msd surface plots and the distribution density plots . the msd surface is high ( low ) wherever the rank of the distribution is high ( low ) . the rank distribution plot is almost a contour plot of the msd surface . this is highly non - trivial . that the rank plot is a contour plot of a surface means that there occurs no jump of ranks when we vary either @xmath11 or @xmath2 . for instance to go to rank four from rank zero one has to pass the intermediate rank regions . this indicates that the dynamics of maps changes continuously ( not necessarily smoothly ) when either @xmath11 or @xmath2 is varied over the turbulent regime . \(2 ) the fist class peaks of the msd surface occur just at the rank three and four distributions . the first - class peaks are due to the periodic coherent motions of elements . \(3 ) in the second - class peaks the mean field distribution is rank zero ( both clt and lln hold . ) for small @xmath2 . when @xmath2 is increased with @xmath11 fixed , it becomes rank one ( clt holds but lln violated ) above certain threshold . with a further increase of @xmath2 it almost always become rank two ( even clt is not fully operative ) . no higher rank distribution ( rank three or four ) appears in the second - class peak regions . only in small band(s ) of @xmath11 ( for instance , only in a narrow band around @xmath217 for @xmath191 ) the rank remains one for large @xmath2 . \(4 ) the threshold between rank zero and rank one distribution precisely agrees with the break point @xmath214 of the drop of msd following lln . for reference we superpose the approximate curve for the @xmath214 ( @xmath218 ) given by @xcite in fig . [ fig9 ] and indeed it runs roughly between the rank zero and rank one regions . we consider the fact in ( 3 ) that the rank one distribution changes smoothly to the rank two distribution with increasing @xmath2 is a key to the hidden coherence ; it indicates that the coherence between maps hidden in the rank one distribution region becomes manifest in the adjacent rank two distribution region . hereafter we focus our attention to gcml . in fig . [ fig10 ] we show a sample of ( 3 ) in the left three boxes . we fix @xmath12 and @xmath11 ( @xmath219 and give the @xmath183 distributions at various @xmath2 ( @xmath220 ) which are sampled in @xmath204 iterations from random starts discarding the first @xmath102 steps . we also display in each box the @xmath183 distribution with @xmath11 set to zero as a reference distribution . in the box for @xmath221 the two distributions are indistinguishable . the @xmath183 distribution is hence rank zero and both lln and clt hold . at @xmath222 the distribution with @xmath223 becomes a gaussian with enhanced msd . it is rank one and indicates the hidden coherence ; clt holds but lln does not and @xmath224 ) . but the continuous change of dynamics to higher ranks , such as from the second box to the third box was not noted . ] . now in the third box at @xmath225 the distribution turns out to be the rank two distorted gaussian distribution . even clt no longer holds . in the fourth box we also exhibit the @xmath183 distribution at @xmath226 , @xmath223 and @xmath65 . this is in our classification the rank three distribution with two overlapping gaussian peaks . in the model with @xmath191 , we can move to this dynamics region by varying @xmath11 , i.e. by traversing the turbulent regime . this point is further substantiated in the following subsections . at this point we need some explanation on the terminology of the _ hidden _ coherence . the interesting phenomenon that above some @xmath214 the lln in the time series is violated certainly occurs but now we have found that it is restricted to the second - class peak regions only . there are valleys where lln approximately holds even for large @xmath2 and also the first - class peak regions where lln is violated even for small @xmath2 . and now even in the second - class peak regions the place where the clt holds ( with lln violated ) is again restricted . for @xmath191 it is only a narrow band around @xmath217 and it moves to the larger @xmath11 for larger @xmath12 . thus if a distinction between the rank one and two distribution is made and the hidden coherence is defined strictly as the phenomenon that clt holds with no approximation ( i.e. excluding the rank two distribution ) but with broken lln , it can live only in a very restricted region among the second - class peak regions . we will further show below that a careful analysis of the decay exponent of the temporal correlator can also unveil the coherence among maps . but still we consider the first discovery@xcite of the onset of the violation of lln _ even in the region where the clt holds ( rank one ) _ is remarkable . our point in emphasizing the distinction between rank one and two distributions is that the recognition that the rank one dynamics region is adjacent in the @xmath188 space to the rank two dynamics region which in turn is adjacent to the rank three and so on enables us to track down the change of dynamics and thus leads us to a working hypothesis that the hidden coherence is the most modest manifestation of the coherent quasi - periodic motion of maps ubiquitous in the turbulent regime . now that we have identified the various regions in the turbulent regime with distinct manifestation of periodicity let us to proceed to clarify how the dynamics of maps in these regions are related each other . to collect necessary information it is best to traverse the turbulence regime at fixed @xmath227 and varying @xmath11 as a tuning parameter . in fig . [ fig11 ] we present a sample record of such an expedition at @xmath228 . the first column shows the mean field distributions as well as the single map distributions . the former is the average during the steps from @xmath203 to @xmath229 , and the latter is the average during the last @xmath230 steps of @xmath204 iterations . in the second column the rank of the @xmath183 distribution is denoted at the top in each box . the evolution of a hundred selected maps is displayed for the last seven steps . in order to chase the periodic motion of elements we devise a simple temporal correlator of the _ relative coordinate vector _ @xmath231 of maps around the mean field : @xmath232 the average over @xmath85 is taken for the last @xmath99 steps . this correlator is displayed in the third column . we show it just as defined by ( [ correlator ] ) but for the measurement of the decay exponent we quote values obtained by the signed logarithm plot ( @xmath233log@xmath234 ) of @xmath235 . for smaller lattice analysis we also averaged over data with different initial configurations to improve the statistics but this did not cause any significant change for the exponent estimation . the last column shows the return maps of @xmath183 ( the set of points ( @xmath236 ) from the last @xmath99 steps . let us start at pure randomness @xmath192 . the map average distribution has many sharp peaks with fractal hierarchy structure reflecting the unstable fixed points of a single map in the chaotic region . the @xmath183 distribution obeys of course both lln and clt and naturally sharp gaussian ( rank zero ) for @xmath65 . the evolution plot shows a simple logistic pattern and the return map is a simple tiny cloud of points . the correlator is one at @xmath237 by definition and decays almost instantly . we can foresee that periodicity three window in the first - class peaks is waiting us ( @xmath238)$ ] but let us choose @xmath239 to see if any precursor of it emerges . we find that the @xmath183 distribution has become rank one . the map average distribution has lost some sub - peaks since any one of the maps is pulled by the mean field in fluctuation and it has become more difficult for it to stay at the unstable sub - fixed points for long time . the coherence is still invisible neither in the map evolution nor in the return map . but we find that remarkably the correlator now oscillates in period three and decays exponentially ( @xmath240 steps for @xmath241 decay ) . this period three damping oscillation of the correlator starts at @xmath242 except for the valleys and the decay exponent becomes smaller for larger @xmath11 . at @xmath243 the @xmath183 distribution is rank two . we are now very close to the first - class peaks . the correlator survives longer period ( @xmath244 steps for @xmath241 decay ) in period three motion . but the return map does not sense this . from the evolution diagram we can immediately tell that this failure of the return map is due to mixing of the maps between the quasi - clusters at a high rate . this inadequacy of a return map should be noted . at @xmath245 we are on the edge of the first - class peak regions . now very careful analysis is necessary to account for the first transient behavior of the maps . the truncation of the first @xmath102 steps is not sufficient and we have to watch the evolution carefully until at least during @xmath204 steps . we show two typical samples at @xmath245 . in the first sample at @xmath245 the maps remain in a few quasi - clusters with mixing until the end of the run . the unstable quasi - clusters oscillate in period three and the correlator decays to @xmath241 only after @xmath246 due to the smaller mixing than before . now we can see in the return map three clouds of points turn out . the map average distribution has only three prominent peaks and all other sub peaks are lost . the corresponding @xmath183 distribution is rank three with both a visible peak and a shoulder on top of a broad band . the high msd of the @xmath183 distribution comes from the large separations between the dominant unstable fixed points of the maps in interaction . this distribution is an example of overlapping gaussian distributions . for the perfect periodic motion of the tight clusters , sharp delta peaks would be observed in both map and @xmath183 distributions . due to the decay of the quasi - clusters caused by the mixing the sharp delta peaks become broad and turn into overlapping gaussian distributions . we frequently observed more prominent overlapping two or three gaussian peaks in the whole analysis of the turbulent regime using both @xmath12 and @xmath11 as tuning parameters . ( see , for instance , the fourth box in fig . [ fig10 ] . ) in the second sample at @xmath245 the maps literally dropped into period three attractor of three clusters at @xmath247 . as the map evolution plot shows the map positions in each cluster exhibit some variance in evolution so that we would better describe the clusters as quasi - clusters . the broad band in this rank three @xmath183 distribution is an artifact of the transient motion before the maps fall into the final attractor . there is no mixing at all between the clusters . the @xmath183 distribution is a sharp single peak and the @xmath183 return map is confined in a very small range . this is due to the fact that the three clusters happen to be almost equally populated in this run so that the mean field fluctuation is minimized . at @xmath248 we are at the summit of the first - class peak . here the maps almost always drop into period three attractor of two clusters . by only two clusters in period three motion the mean field must also fluctuates in period three . thus the @xmath183 distribution has sharp three peaks and msd turns out high . the value of msd is solely determined by the dynamics of clusters and @xmath2 independent . also the return map is widely spread . while we traverse the summit of the first - class peaks till @xmath249 the attractor remains in period three . sometimes it bifurcates to period six . in further expedition we depart from the first - class peaks and go down into the second class peak regions . everything then occurs in reverse way in this descent till @xmath250 . from @xmath251 we traverse the second - class peaks as well as the valleys . at the second - class peaks the correlator survives above @xmath241 until @xmath252 but with no apparent low periodicity . then above @xmath253 the correlator starts again catching the precursor of the period two opposite phase motion in the two - cluster regime of the gcml . thus the indeterminacy of the period of quasi - clusters beyond second peak regions comes from the fact that the regions are just in between the period three window and two - cluster opposite - phase motion regions . after @xmath254 the correlator shows perfect period two motion and the decay is smoothly prolonged with increasing @xmath11 . at @xmath255 it only decreases by a few percent even at @xmath256 . correspondingly , above @xmath254 the middle peak in the map average distribution starts approaching to the peak at higher @xmath186 and above @xmath257 the map average distribution splits into two ( no events in the middle gap region ) . still in the evolution plot we can see the maps are in loose clusters with mixing between each other . the high msd of the rank two and three distributions above @xmath254 is clearly induced by the widely spread dominant unstable fixed points of low periodicity . we have done a similar analysis at various @xmath12 up to @xmath226 and verified that all above features at @xmath191 are unchanged . the various manifestation of the periodicity is summarized in table [ table1 ] for @xmath191 . only a minor difference for larger @xmath12 is that the regions of rank one distribution above the first rank msd peaks become larger shifted to larger @xmath11 and also the rank three distribution starts to appear there . since the rate of the mixing of maps between clusters is the key to the turbulent regime we have also measured the temporal correlator of the clustering pattern matrix similar to the anderson - edwards order parameter@xcite . it exhibits the same variation of the decay exponents with our correlator . we examine below how the dynamics in various regions are related each other . ( 1 ) let us first compare the dynamics of the periodic regions with that at @xmath258 . at @xmath258 maps evolve almost independently with long visit at all of the unstable fixed points of the logistic map . thus the map average distribution has many peaks with fractal structure peaking . since the maps behave as almost independent random variables with the same probability distribution , naturally both lln and clt hold . at the periodic window in the first - class msd peak regions , on the other hand , the @xmath23 focusing effect tunes the maps to form tightly bound two or three clusters mutually oscillating in period three attractor and this leads to the rank four distribution with sharp delta peaks . the dynamics of clusters in the periodic attractor is solely determined by @xmath12 , @xmath11 and the effective couplings between the clusters determined by the population _ ratios _ among the clusters ) does exhibit the same period three attractor at the same value of @xmath11 . ] . the msd of the @xmath183 distribution is again determined by the @xmath12 , @xmath11 and the population _ ratios _ between the clusters , and there is no room for the explicit @xmath2 dependence . similar coherent dynamics is realized also in the ordered two - cluster regime beyond the turbulent regime ( @xmath259 for @xmath191 ) . the turbulent regime has a period three window in the center of it and the two - cluster regime beyond it . we discuss below how these affect other regions . \(2 ) when @xmath11 deviates from the range necessary for the period three cluster dynamics ( or for the period two - cluster dynamics ) , the perfect synchronization of maps among each cluster is no longer realized . but what we observed in the last subsection is that even for some large deviation of @xmath11 the maps still move under certain coherence if @xmath2 is beyond the threshold @xmath214 . such coherent motion of maps can be best described by _ unstable quasi - clusters _ the decay life of which is given by the inverse of the exponent of the correlator . the maps now divide into a few masses _ at any instance _ of the evolution . in a short time scale these masses oscillate mutually in the same periodicity as before as we observed in both the map evolution plot and the correlator . but the variance of map positions among each mass is not so small and the synchronization of maps in each mass is also not perfect . hence we call these masses as quasi - clusters . due to the fail of perfect synchronization , the quasi - clusters mix each other by the exchange of maps and soon they loose their identities . so the quasi - clusters are unstable and have short life time . but still at any instance of evolution there do exist quasi - clusters . the quasi - clusters defined at a certain time decay but in the evolution of maps new quasi - clusters are kept created . we may regard these quasi - clusters as _ remnants _ of the perfect periodic motion of maps . in fact when we approach the periodic regions the decay exponent of the correlator decreases smoothly , that is , the life time of the quasi - clusters increases smoothly , signaling the phase transition from turbulence to periodicity . in pure turbulence the system is invariant under the exchange of maps , while in the system of maps divided into clusters this permutation symmetry is broken . simultaneously the system in pure turbulence is invariant under the temporal translation , while the system in periodic motion is only invariant under the temporal translation just for the period . the self - organized state of maps in periodic quasi - clusters emerges as the state of broken symmetry from pure turbulence by a weak self - coupling and it emerges only for large @xmath2 . this is interestingly reminiscent of the onset of ordered parameter at the spontaneous break down of the global symmetry in the field theory where an infinite number of dynamical degree of freedom is necessary in order to support the ordered vacuum . see , for instance , weinberg@xcite where the broken symmetry state of a chair is elucidated . it is important that the maps in the quasi - clusters approximately follow the previous cluster attractor in the life time of the quasi - clusters . this means that the @xmath183 distribution is controlled by the scale of the previous attractor approximately and again determined by the @xmath12 , @xmath11 and the population _ ratios _ and not by @xmath2 . thus there emerges a @xmath183 distribution whose msd does not decrease with @xmath2 and lln is violated . depending on the rate of mixing and the allowed population ratios at the @xmath11 , distributions of various ranks emerge . \(3 ) if the deviation of @xmath11 from the necessary range for periodic cluster attractor is not large , the failure in the map synchronization is not also large . the quasi - clusters decay by low rate mixing of members so the correlator survives for long time exhibiting clearly the periodicity of the previous periodic cluster attractor . but the would - be delta - peaks of both the average map distribution and the @xmath183 distribution at no mixing now change into overlapping gaussian distributions with visible separation between the peaks . these changes of the correlator and the @xmath183 distribution at small deviation of @xmath11 are precisely what we observe around the first - class msd peaks and near the upper boundary of the turbulent regime adjacent to the ordered two - cluster phase . with further deviation of @xmath11 the mixing rate becomes higher . the correlator decays more quickly and the @xmath183 distribution becomes from rank three to rank two - the distorted gaussian distribution . we observe this when we go down from the first - class peaks to the region of @xmath258 or when we approach the first - class peak regions from above . in the former case the correlator keeps exhibiting the period three damped oscillation even if we go down into the rank one @xmath183 distribution region . in the latter case the periodicity of the correlator becomes indefinite below @xmath254 for @xmath191 since naturally we are coming down from two - cluster regime with period two to the period three window , and we finally loose the track of a periodicity remnant ( rank two distribution ) when we go further into the narrow band of rank one @xmath183 distribution region . this is the genuine hidden coherence region . \(4 ) the place the hidden coherence can live has now become very limited since we have improved the sensitivity to the coherence effect . in the old analysis@xcite , the distinction between rank one and two distributions was ignored and neither the change of the decay exponent of the correlator with @xmath11 nor the periodicity of it was analyzed . the periodic window and the valleys were not discussed also . then the coherence seemed to live hidden everywhere in the turbulent regime if @xmath260 . now , both the periodic window with manifest coherence and the valleys with valid lln should be excluded first of all . furthermore , in the regions where the correlator has clearly caught the periodicity remnant , it is no longer legitimate to describe the coherence as hidden . thus , regarding the correlator the coherence is hidden only in a limited region in between the period - three and period - two remnant regions among the second - class msd peak regions . this is the region with @xmath251 for @xmath191 . see table [ table1 ] and eq . ( [ second ] ) . here the correlator does show longer decay time ( @xmath241 decay within @xmath261 steps ) than that at @xmath258 but it decays with large fluctuation showing no low periodicity . this is natural for the region in between two different periodicity regions . but not only the correlator but also the shape of the @xmath183 distribution is a vital quantity to keep track of the change the dynamics . almost everywhere in the possible residence of the hidden coherence regarding the correlator , the @xmath183 distribution is rank two ( rather than gaussian ) and naturally understood as the smooth continuation of the rank three overlapping gaussian distribution . the clt is apparently violated here and the distribution is still controlled by the scale of the attractor . therefore , the place where the coherence is literally hidden from both the correlator and distribution analyses is now very limited . it is the narrow band at @xmath262 for @xmath191 and around @xmath263 for @xmath226 where the @xmath183 distribution is really gaussian with enhanced msd ( rank one ) and the correlator does not show the low periodicity . here the coherence seen in the enhanced msd can be only substantiated either by the finite mutual information or by the instability of the fp fixed point distribution . but in ( 3 ) above we successfully tracked down the smooth change of distribution ( and the corresponding change of the decay exponent of the correlator ) with the deviation of @xmath11 from rank four down to rank two distribution regions which surround the literally hidden coherence region . this gives a new way to look at the hidden coherence . higher rate mixing with further smooth deviation of @xmath11 should necessarily weaken the coherence among the maps and change the distorted gaussian ( rank two ) to the gaussian distribution . if the scale of the attractor still remains to govern the map dynamics the @xmath183 distribution with @xmath2-independent msd would emerges . we propose this last statement as a working hypothesis for the origin of the hidden coherence . up to this point we have heavily depended on the numerical simulation . our concern is the statistical properties of the large - size gcml ( egcml ) and the monte - carlo simulation is naturally the basic approach to the phenomenology of such system . we have shown that a faithful extensive analysis of the data unveils amazing new features of the system . but , of course , if by some analytical method , one could relate the systematic manifestation of periodicity in the turbulent regime to the periodic behavior of the element of the system , namely the period three window of a single logistic map , it would be certainly a step forward to the real understanding of the turbulent regime . in this last subsection we discuss such an analytic approach . let us consider a typical case where the n maps are divided into equally populated three clusters @xmath264 ( @xmath265 ) and the clusters move cyclically round the three fixed positions @xmath266 . in such a maximally symmetric and period three motion of gcml maps , @xmath267 , the meanfield of all maps at time @xmath85 becomes a time - independent constant which is nothing but an equal weight average @xmath268 of @xmath269 . to see this let us first rewrite the meanfield using the invariance rule ( [ sumrule ] ) as @xmath270 then for the configuration in case the last average can be calculated as @xmath271 where @xmath272 denotes the coordinate of the cluster @xmath273 at time n. therefore , if such a configuration is produced by the interaction between the maps via the meanfield , any one of gcml maps must be evolving by a common evolution equation @xmath274 we can cast the evolution equation ( [ consthevolution ] ) with a constant meanfield @xmath268 and @xmath275 to a simple logistic map @xmath276 with a reduced coupling constant @xmath277 by a linear scale transformation @xcite @xmath278 the reduction rate @xmath279 is given by @xmath280 the period three window of the logistic map ( [ yevolution ] ) starts at @xmath281 by the tangent bifurcation and after the sequential bifurcation it closes at @xmath282 by the crisis . therefore the reduction rate @xmath279 must be for @xmath191 in the range @xmath283 and at a given rate @xmath279 within this range , ( [ rconstarint ] ) gives a curve of constraint on the @xmath284 plane . there is another constraint from self - consistency . each of the maps @xmath285 is related to the gcml maps @xmath286 by the linear transformation ( [ lineartransformation ] ) and so is the average value @xmath287 of maps @xmath285 to @xmath268 : @xmath288 since @xmath287 is simply an equal weight average of the period three stable orbits of ( @xmath289 ) , it can be estimated solely by the property of the logistic map without any recourse to the gcml evolution equation . for instance at the tangent bifurcation point ( @xmath290 ) , we have @xmath291 and from matching of the coefficients we obtain @xmath292 . numerically we find that @xmath293 slightly below the threshold @xmath294 , drops sharply ( @xmath295 ) to @xmath296 at the threshold @xmath297 , and @xmath287 gradually changes around @xmath298 until the end of the window ( @xmath299 ) . by eliminating @xmath268 from ( [ rconstarint ] ) and ( [ yconstaraint ] ) we obtain @xmath300 the estimated @xmath11 for @xmath191 from ( [ epsilonestimates ] ) is @xmath301 the estimates a , b , c and d are evaluated respectively below the threshold , at the threshold , at the first bifurcation point at the middle , and at the upper edge of the period three window in the order of increasing @xmath277 . before comparing with the simulation result we should note that a maximally symmetric configuration of equally populated three clusters in period three motion is assumed in deriving ( [ epsilonestimates ] ) . thus , the value of @xmath11 necessary for such a configuration should be deduced from the genuinely period three region , namely from the interval from b to c. we can predict that @xmath11 must be in the range ( 0.0365 , 0.0422 ) for the formation of such a maximally symmetric gcml configuration . for such a configuration the meanfield should not fluctuate in time and it will show up as a deep valley in the gcml msd surface . we have added two exterior estimates in ( [ estimates ] ) expecting that these extensions will cover the asymmetrically populated clusters in period three motion . now let us compare ( [ estimates ] ) with the range of the period three first - class peaks found in section v b , namely , @xmath302 and @xmath212 . firstly there is a remarkable agreement between the range of peaks and the range of estimates a to d in ( [ estimates ] ) . the whole period three window of a single logistic map shows up as the whole first - class peaks . a closer comparison with the observations in section v d gives further support for the link of the structure observed in the numerical simulation and the period three window of the logistic map . * we noted that in the deep valley of the msd surface between the two first - class peaks the formation of three clusters in period three motion . the valley region @xmath303 remarkably matches the above estimated range ( 0.0365,0.0422 ) for the maximally symmetric configuration . * we noted that in the upper first - class peak ( @xmath212 ) the maps almost always form only two clusters in period three motion . it is natural that the larger @xmath11 reduces the number of clusters from three to two . as we noted the extremely high msd in the first - class peak is induced by the asymmetric population configuration ( @xmath304 ) . furthermore we have verified by simulation that the first - class peak regions shift to higher @xmath11 when @xmath12 is increased precisely in accord with the prediction by ( [ epsilonestimates ] ) . it is remarkable that the maps with high nonlinearity are driven into coherent synchronizing clusters with very small interaction parameter @xmath11 once it takes the tune up value . a lot of further effort has to be devoted to fully appreciate the implication . the analytic approach here is only a step forward to this direction . a survey of the overall relation of peak - valley structure of the msd surface to the hierarchical windows of a single logistic map as well as the dynamical variation of the population configuration is now underway . in this article we have revisited the gcml and first resolved the puzzle of the posi - nega switch realized the two - cluster phase . the puzzle was that the maps posses the memory of the cluster to which they subject even though they pass through a transient chaotic process where many channels are open . we have proved that maps never mix across the mean field of them in the transient process . we have then analyzed the turbulent regime of gcml . to date the so - called hidden coherence has been the main target of interest in this region . we have presented our new finding that there exists a remarkable period three attractor regions in this regime . the turbulent regime locates in - between the genuinely turbulent region around @xmath192 and two - cluster regime of period two attractor and has this period three attractor regions in the center of it . we have shown that the remnant of the periodic coherence can be found almost everywhere in the turbulent regime by the measurement of the temporal correlator of the fluctuation of maps around the mean field and by the extensive analysis of the @xmath183 distributions over the whole turbulent regime . the dynamics of the turbulent regime can be well described by the quasi - clusters with mixing of elements between each other . the hidden coherence may be regarded the most modest manifestation of this remnant periodicity . by an analytic approach we have shown that the prominent periodicity manifestation in the turbulent regime is caused by the period three window of a single logistic map . a new extended globally coupled map lattice(egcml ) is constructed in which the interaction between maps decreases proportionally to the distance . egcml at @xmath42 has the same phase diagram with that of gcml . the bifurcation tree structure of the attractor of egcml in the two - cluster regime is also the same with that of gcml . the maps in each egcml cluster are phase synchronizing rather than synchronizing . the posi - nega switch between the two clusters is realized also in egcml . in the chaotic transient process in egcml - switch , formation of amazing spatial clusters of maps is observed . egcml may be viable for the simulation of chaotic diffusion process . we have also shown that the turbulence regime of egcml is also not immune from periodicity just like gcml turbulent regime . as a whole this work is an exploration of order in the chaos . the posi - nega switch in gcml is realized because there occurs no mixing in the chaotic transient process . in the chaotic transient process in a egcml - switch , interesting spatial clusters are formed . and the turbulent regime of both of gcml and egcml is dominated by the quasi - clusters of maps in periodic motion . it is a pleasure to note the author s debt of gratitude to k. kaneko . this work was started several years ago stimulated by the pioneering articles on gcml by him@xcite and the compact model gcml has been an unfailing source of interest since then . the author is thankful to fumio masuda who worked on this subject together with the author until the last one year of completion and the nice collaboration time with him is memorized . thanks also go to hidehiko shimada for many crucial suggestions , hayato fujigaki for his collaboration on the study of coupled flows done in parallel with this work , and kengo kikuchi for his unfailing effort on the analytic approach to gcml now underway . last but not at least the author wishes to thank wolfgang ochs at the max - planck institute for sharing his insight in this research and reading the manuscript . this work was supported by the faculty collaborative research grant from meiji university , grant - in - aids for scientific research from ministry of education , science and culture of japan , and grant for high techniques research from both organizations . k. kaneko , phys . lett . * 65 * , 1391 , ( 1990 ) ; + k. kaneko , physica ( amsterdam ) * 55d * , 368 , ( 1992 ) ; ibid . , * d86 * , 158 , ( 1995 ) . k. kaneko , physica ( amsterdam ) * 23d * , 436 ( 1986 ) ; + j. p. crutchfield and k. kaneko , _ in directions in chaos _ , edited by b. -l . hao ( world scientific , singapore , 1987 ) . see , for instance , + j.j . hopfield , proc . usa ( biophysics)*79 * , 2554 ( 1982 ) ; + w.a . little , math . * 19*,101 ( 1974 ) ; + d. j. amit , h. gutfreund and h. sompolinski , phys . rev . * a32 * , 1007 ( 1985 ) . .the various periodicity manifestation in the turbulent regime of gcml ( @xmath191 and @xmath65 ) . note that in each range of @xmath11 there are also valleys with an approximate lln along with listed peaks with the broken lln . [ table1 ] [ cols="^,^,^,^ " , ]

compact radio sources in nearby starburst galaxies often appear to be supernova remnants ( snr ) . snr in our galaxy and the large magellanic cloud ( lmc ) are sources of bright near - infrared [ feii ] line emission with luminosities ranging from @xmath5w to @xmath6w ( for the [ feii]@xmath0 m line , e.g. , oliva , moorwood , & danziger 1989 ; burton & spyromilio 1993 ) . this behavior suggests a connection between supernova activity and the [ feii ] emission in starbursts . this is supported by near - infrared spectroscopy ( moorwood & oliva 1988 ; greenhouse et al . 1991 ; vanzi & rieke 1997 ) , by high - resolution [ feii ] imaging studies ( forbes et al . 1993 ; greenhouse et al . 1997 ; alonso - herrero et al . 2000 ) , and by the tight correlation between the radio 6 cm and the [ feii ] emissions ( forbes & ward 1993 ) . further determination of the relation of the infrared [ feii ] emission with the supernova activity in nearby starforming galaxies would be helpful both for the understanding of these galaxies and for using the near - infrared [ feii ] luminosity as estimator of the supernova rate in more distant galaxies where individual radio supernovae can not be identified . there are indications of some complexities in this relation . for example , the bright [ feii ] emission may not be spatially coincident with the radio snr . lumsden & puxley ( 1995 ) conducted a spectroscopic survey for [ feii ] emission from radio snrs in m33 and found that in some cases the [ feii ] emission does not peak at the position of the radio snr , as well as that some radio snr show no [ feii ] emission . recently , morel , doyon , & st - louis ( 2002 ) have carried out an [ feii ] imaging survey of optically selected snr in m33 and found a similar result . both works have concluded that the [ feii ] emission from snr may be produced over a time scale of @xmath7 10@xmath8 years . greenhouse et al . ( 1997 ) found that none of the brightest sources of [ feii ] emission in m82 were coincident with the position of the radio snrs . m82 and ngc 253 are the nearest starburst galaxies ( distances @xmath9mpc and @xmath10mpc respectively ) . both galaxies show a large number of radio compact sources ( e.g. , @xmath11 in m82 kronberg et al . 1985 and huang et al . 1994 ; @xmath12 in ngc 253 turner & ho 1985 and ulvestad & antonucci 1991 , 1997 and references therein ) . their proximity makes these two galaxies ideal to study the relation between supernova activity and the [ feii ] emission in starburst galaxies . greenhouse et al . ( 1997 ) presented a fabry - perot [ feii]@xmath0 m map for m82 with a spatial resolution of @xmath13 . their main conclusions were that the compact [ feii ] sources contribute only @xmath14 of the total [ feii ] emission in m82 and that they trace a population of snr older than the radio snr . their work is somewhat inconclusive because of the limited angular resolution and also because strong extinction could hide [ feii ] emission from some of the radio snrs . this paper presents _ hst_/nicmos images of m82 and ngc 253 . the high spatial resolution of the nicmos [ feii]@xmath0 m images ( @xmath15 ; this corresponds to 5pc and 4pc for m82 and ngc 253 , respectively ) allows a detailed study of the location of the bright sources of [ feii ] and of the rle of supernova remnants in this line emission . _ hst_/nicmos observations were obtained on april 7 , 8 , 12 and 22 1998 ( for m82 ) and on august 13 and 20 1998 ( for ngc 253 ) using the nic2 and nic3 cameras . table 1 lists details of the observations . the observational strategy consisted of taking images in a spiral dither with a 5.5 pixel spacing with four positions for a given field . the plate scales for the nic2 and nic3 cameras are 0.076pixel@xmath16 and 0.20pixel@xmath16 , respectively . the fwhm of the point sources in the nic3 f166n continuum images is @xmath17 . lcclccc nic2 & f160w & @xmath18 & m82 & 192 & 4 & @xmath19 + & & & ngc 253 & 192 & 2 & @xmath20 + nic2 & f222 m & @xmath21 & m82 & 448 & 4 & @xmath19 + & & & ngc 253 & 512 & 2 & @xmath20 + nic2 & f187n & pa@xmath1 & m82 & 384 & 4 & @xmath19 + & & & ngc 253 & 384 & 2 & @xmath20 + nic2 & f190n & continuum & m82 & 384 & 4 & @xmath19 + & & & ngc 253 & 384 & 2 & @xmath20 + nic3 & f164n & [ feii]@xmath0 m & m82 & 4096 & 2 & @xmath22 + & & & ngc 253 & 2560 & 1 & @xmath23 + nic3 & f166n & continuum & m82 & 3840 & 2 & @xmath22 + & & & ngc 253 & 2560 & 1 & @xmath23 + nic3 & f215n & continuum & m82 & 2304 & 2 & @xmath22 + & & & ngc 253 & 2304 & 1 & @xmath23 + the reduction of the nicmos images used routines from the package nicred ( mcleod 1997 ) . a master dark image was produced by combining between 10 and 20 darks for a given sample sequence after the subtraction of the first readout . the darks used were part of other proposals observed close in time , and the flatfield images were on - orbit data . the data reduction was performed using the following steps : subtraction of the first readout , dark current subtraction on a readout basis , correction for linearity and cosmic ray rejection ( using fullfit ) , and flatfielding . the individual dithered galaxy images were registered to a common position using fractional pixel offsets and cubic spline interpolation , and combined to produce the final image of each field . prior to the flux calibration of the images , the background needs to be subtracted ( only for wavelengths longwards of @xmath24 m ) . due to the large projected size of both galaxies , the field of view of the nic2 images is not large enough to enable measurements of the background on empty corners of the images , so we used background measurements taken during the servicing mission observatory verification ( smov ) program for the filter nic2 f222 m . the f215n images cover a larger field of view , and an estimate of the thermal background was obtained from the corners of the images . the flux calibration of the broad - band , on - line and off - line images was performed using the conversion factors ( from adu / s to mjy ) based on measurements of the standard star p330-e during smov ( marcia rieke 1999 private communication ) . ideally one would like to measure the continuum on both sides of the emission line to estimate the continuum at the emission line . however , nicmos only provides continuum bands to the red of the emission line . in cases of objects such as m82 and ngc 253 , where the extinction to the nuclear regions is very high ( even at infrared wavelengths ) , a straight subtraction of the longer wavelength off - line image may result in an over correction of the continuum at the wavelength of the emission line . to estimate the continuum for the nic3 f164n images , we fitted the continuum on a pixel - by - pixel basis between @xmath25 m and @xmath26 m by linear regression of the flux as a function of the wavelength for the nic3 f166n and nic3 f215n line - free images . the extrapolated continuum was then subtracted from the nic3 f164n image to produce the final continuum - subtracted [ feii]@xmath0 m line image . this procedure does not correct for extinction , but it rather ensures the subtraction of the correct _ obscured _ continuum at each position of the galaxy . for the nic2 f187n image ( pa@xmath1 ) , since we did not have other narrow - band images taken with camera 2 , we used the nic2 f190n continuum image . fortunately , the continuum need not be subtracted as accurately for pa@xmath1 as for the [ feii ] emission line because the pa@xmath1 emission is much brighter . the fully - reduced images of each individual field were combined to produce the mosaics . the images presented in this paper were taken during different orbits . this resulted in slightly different orientations ( from @xmath27 to @xmath28 for m82 and from @xmath29 to @xmath30 for ngc 253 ) . before combining the images into mosaics , they were rotated by linear interpolation to north up , east to the left . this provided the same orientation for all the images , and more importantly allows comparisons with previously published data . to construct extinction maps to correct the [ feii ] emission , and [ feii]/pa@xmath1 maps , the nic2 f187n , f190n , f160w and f222 m images were demagnified to match the plate scale of the nic3 images and filtered with a gaussian with @xmath31 to match the point spread function ( psf ) of the nic3 images . figure 1a and 1b are mosaics ( after rotation ) of the continuum at @xmath25 m and the continuum - subtracted [ feii]@xmath0 m line emission , respectively . both images are displayed on a logarithmic scale to stress the contrast between the diffuse extended continuum and the compact sources . the [ feii]@xmath0 m line map has been smoothed with a gaussian filter with @xmath32 . the line emission image shows some residuals which are due to both the under - sampling of the nic3 camera psf and the variation in the shape of the psf with wavelength . in addition , the right hand side of the line image shows a slightly lower s / n because the f164n and f166n images at that position had different orientations . the [ feii]@xmath0 m line emission exhibits two distinct components : a number of relatively compact regions distributed along the disk of the galaxy and an extended diffuse component in the form of filaments extending above and below the plane of the galaxy . the filaments seen in the [ feii]@xmath0 m emission line show a remarkable similarity with the chimneys at the base of the superwind detected in radio ( wills et al . 1999 ) , which are interpreted by these authors as expelled material from the central region of m82 . in figures 2a and 2b we present contour maps of the @xmath25 m continuum emission and the [ feii]@xmath0 m line emission , respectively . comparison of the nicmos [ feii]@xmath0 m line map with that of greenhouse et al . ( 1997 ) reveals that some of the sources seen in the lower resolution map of greenhouse et al . ( 1997 ) break up into individual sources at higher spatial resolution . in particular , their source fe1 contains some 4 or 5 individual sources . source fe6 ( the most south - west source in greenhouse et al . 1997 map ) appears as a point source at our resolution in both the nic3 f164n and nic3 f166n images , and the signal to noise on it is compromised by the differing image orientations ( see above ) . therefore it is not surprising we do not detect line emission from it . the nicmos images detect a number of compact sources in addition to those reported in greenhouse et al . ( 1997 ) . in figures 2a and 2b we display the positions of 44 radio snrs with peak flux densities above 1mjy ( list provided by dr . karen wills , see more details in section 5.1 ) . the distribution of the radio snrs and their relation with the [ feii]@xmath0 m emission will be discussed in section 5 . in figures 3a and 3b we display the mosaics of the @xmath33 m continuum ( nic2 f190n images ) and the pa@xmath1 line emission ( continuum subtracted nic2 f187n images ) respectively . the pa@xmath1 image is dominated by a ring of ionized gas with an approximate diameter of 400pc . the resemblance with the mid - infrared [ neii ] line emission map at @xmath34 m ( achtermann & lacy 1995 ) is remarkable . the south side of this ring of ionized material appears to be incomplete ( see achtermann & lacy 1995 ) , and it also shows a `` hole '' of emission at the center of the galaxy , which approximately coincides with the peak of the @xmath35 m surface brightness . ccccccc + & ( ) & ( w m@xmath36 ) & ( w m@xmath36 ) & ( mag ) + pa@xmath1/h@xmath1 & 5.8 & @xmath37 & @xmath38 & @xmath39 & this work , oconnell & mangano ( 1978 ) + pa@xmath1/pa@xmath40 & 6 & @xmath37 & @xmath41 & @xmath42 & this work , mcleod et al . ( 1993 ) + pa@xmath1/pa@xmath43 & 3 & @xmath44 & @xmath45 & @xmath46 & this work , mcleod et al . ( 1993 ) + pa@xmath1/pa@xmath40 & 3 & @xmath44 & @xmath47 & @xmath42 & this work , mcleod et al . ( 1993 ) + br@xmath43/pa@xmath1 & 3.8 & @xmath48 & @xmath49 & @xmath50 & lester et al . ( 1991 ) , this work + br@xmath1/pa@xmath1 ( e1 ) & 6 ? & @xmath51 & @xmath52 & @xmath53 & achtermann & lacy ( 1995 ) , this work + br@xmath1/pa@xmath1 ( w1 ) & 6 ? & @xmath54 & @xmath55 & @xmath56 & achtermann & lacy ( 1995 ) , this work + br@xmath1/pa@xmath1 ( w2 ) & 6 ? & @xmath57 & @xmath58 & @xmath59 & achtermann & lacy ( 1995 ) , this work + @xmath60feii]1.64/[feii]1.25 & 3 & @xmath61 & @xmath62 & @xmath46 & this work , mcleod et al . ( 1993 ) + + & ( ) & ( mjy ) & ( mjy ) & ( mag ) + f222m / f160w & 3 & 237 & 166 & @xmath56 & this work + ccccccc + & ( ) & ( w m@xmath36 ) & ( w m@xmath36 ) & ( mag ) + pa@xmath1/br@xmath43 ( center ) & @xmath63 & @xmath64 & @xmath65 & @xmath66 & this work , harrison et al . ( 1998 ) + pa@xmath1/br@xmath43 ( center ) & @xmath67 & @xmath68 & @xmath69 & @xmath70 & this work , puxley & brand ( 1995 ) + pa@xmath1/br@xmath43 ( 3 ne ) & @xmath63 & @xmath71 & @xmath72 & @xmath73 & this work , harrison et al . ( 1998 ) + pa@xmath1/br@xmath43 ( 3.9 ne ) & @xmath67 & @xmath74 & @xmath75 & @xmath76 & this work , puxley & brand ( 1995 ) + + & ( ) & ( mjy ) & ( mjy ) & ( mag ) + f222m / f160w ( center ) & @xmath67 & 213 & 142 & @xmath56 & this work + it has been suggested that higher extinction in the south half of m82 may be causing the uneven distribution of ionized gas in the ring . however , the ring appears to be complete in [ feii]@xmath0 m line emission . if extinction were the cause , we should expect to see a similar morphology in the [ feii ] emission , which is more affected by extinction than the pa@xmath1 emission . it is therefore tempting to suggest that the south part of the ring of ionized gas is actually _ broken _ , perhaps due to sn explosions . this hypothesis would also explain the excess of [ feii ] emission in this region . the presence of the hole of gas emission at the center of m82 along with the anti - correlation between the morphology of the hydrogen recombination lines and the co index has been invoked as evidence for an outwards propagating starburst ( e.g. , satyapal et al . the diffuse component of pa@xmath1 emission is also in the form of `` fingers '' or `` filaments '' above and below the plane of the galaxy . it is now known from ground - based h@xmath1 imaging that this emission along the minor axis of the galaxy extends over 11kpc ( devine & bally 1999 ) . figure 1b and figure 3b show how there is a very close spatial correspondence between the morphologies of the extended [ feii]@xmath0 m and pa@xmath1 emissions . figures 4a and 4b are the @xmath25 m continuum and the [ feii]@xmath0 m line emission maps of ngc 253 . as seen in m82 , the [ feii]@xmath0 m emission is in the form of a number of compact regions along the disk of the galaxy superimposed on a diffuse emission component . figures 5a and 5b are contour plots of these two emissions . we also show the positions of the 6 cm radio sources with @xmath77mjy from ulvestad & antonucci ( 1991 ) , excluding those sources that are classified as hii regions in ulvestad & antonucci ( 1997 ) . ground - based [ feii]@xmath0 m and br@xmath43 maps at lower spatial resolution ( @xmath78 ) were previously observed by forbes et al . they found that the central [ feii ] emission of ngc 253 is dominated by three peaks of emission . our higher resolution nicmos image shows a larger number of [ feii ] compact sources . for instance , source b ( in the forbes et al . 1993 notation ) breaks up into three individual sources . in addition to the nucleus and sources a and b detected by forbes et al . ( 1993 ) , there are other sources along the disk of the galaxy , as well as diffuse emission perpendicular to the disk , more prominent in the south half of the galaxy . we believe that forbes et al . ( 1993 ) did not detect the diffuse component above and below the disk of the galaxy because of the lower s / n of their [ feii ] map . we also mark in these figures the position of the peak ( the filled square ) of the near - infrared emission ( also coincident with the peak of the pa@xmath1 emission ) which was identified as the nucleus of the galaxy by forbes et al . other authors ( e.g. , keto et al . 1999 ) claim that a bright non - thermal radio source located ne of the peak of the infrared continuum emission ( source 2 in turner & ho 1985 ) may be the true nucleus of ngc 253 . it is shown in these figures as a filled star symbol . the nicmos pa@xmath1 line map ( figure 6b ) shows very bright emission stemming from the peak of infrared continuum emission , together with more diffuse emission in a plume - like morphology along the plane of the galaxy , primarily to the ne of the nucleus . as for m82 , there is an extended component of the pa@xmath1 emission perpendicular to the disk , better seen in our nic3 images ( not shown here ) which are more sensitive to diffuse emission . this component is however not as strong as that in m82 . there is a remarkable resemblance of the central pa@xmath1 emission with both the [ neii]@xmath34 m emission line and the nearby continuum images presented in bker , krabbe , & storey ( 1998 ) and keto et al . both m82 and ngc 253 are known to be highly obscured ( see for instance the comparison between the optical _ hst_/wfpc2 f555w and the _ hst_/nicmos f222 m images of m82 in alonso - herrero et al . 2001 ) . to calibrate the supernova rate in terms of the [ feii]@xmath0 m emission and to obtain the total number of ionizing photons , it is necessary to correct the observed quantities for extinction . we can estimate the extinction to the gas in m82 using observations of the ratios of hydrogen recombination and near - infrared [ feii ] lines . we used published values of the line fluxes for the smallest apertures and matched those apertures on the nicmos line images when measuring the pa@xmath1 and [ feii]@xmath0 m fluxes . the intrinsic ratios of hydrogen recombination lines are from hummer & storey ( 1987 ) for an electron temperature of @xmath79k and electron density of @xmath80@xmath81 . the ratio between [ feii]@xmath82 m and [ feii]@xmath0 m does not depend upon the physical conditions of the gas . we used a polynomial fit to the near - infrared extinction law given in he et al . ( 1995 ) . since we do not have nicmos images of two hydrogen recombination lines , to measure extinction at the full nicmos angular resolution we made use of the broad - band filters nic2 f160w and nic2 f222 m to construct a map of extinction to the stars in m82 . we assumed a color for the unreddened stellar population of @xmath83 ( which applies almost independently of the age of the population ) , and , for computational convenience , a foreground dust screen model following the extinction law of he et al . the contribution of dust emission at the @xmath21-band was assumed to be negligible . figure 7a is the extinction map ( @xmath84 ) from the @xmath85 color map of m82 , with the positions of the radio sources superimposed . the values of the extinction in the @xmath21-band derived from this map are between @xmath86mag and @xmath87mag . the m82 extinction map qualitatively looks very similar to those of satyapal et al . ( 1995 ) , but quantitatively the values of the extinction are different . the estimates of the extinction to the gas for m82 are presented in the first part of table 2 along with the line fluxes . the errors in table 2 are the uncertainties associated with the aperture size , aperture centering and flux calibration . for comparison , we also estimated the extinction through a small aperture ( 3 ) using the f222m / f160w continuum flux ratio , which gives an estimate of the visual extinction to the stars . as can be seen from this table the agreement between the extinctions to the gas and to the stars is excellent at infrared wavelengths . satyapal et al . ( 1995 ) reached a similar conclusion . a much lower level of extinction is indicated by comparing the h@xmath1 and pa@xmath1 line fluxes . this behavior is known to arise from the different optical depths probed by optical and near - infrared lines . this behavior is confirmed by the much higher extinction of @xmath88mag inferred by puxley et al . ( 1989 ) from the radio h53@xmath1 to mid - infrared br@xmath1 line ratio . we also computed the extinctions to the e1 , w1 and w2 regions in achtermann & lacy ( 1995 ) notation ( see table 2 for the positions ) . these were found to be bright in [ neii]@xmath34 m and br@xmath1 . we used their br@xmath1 measurements along with the mid - infrared extinction curve in rieke & lebofsky ( 1985 ) . since achtermann & lacy ( 1995 ) did not quote aperture sizes , we assumed that their fluxes were measured over the extent of these regions ( approximately @xmath89 in diameter ) . the values of the extinction for these three regions are slightly larger than the values we would measure from the @xmath85 extinction map , perhaps indicating minor differences in the optical depth . a further test of the extinction to the gas was based on the total number of ionizing photons from the extinction corrected pa@xmath1 line emission map . the number of ionizing photons obtained from the extinction - free h53@xmath1 radio recombination line is @xmath90s@xmath16 ( puxley et al . 1989 ) , and from the @xmath91 mm continuum is @xmath92s@xmath16 ( carlstrom & kronberg 1991 ) . we measured through a large aperture the pa@xmath1 flux on the observed and the extinction - corrected images . we obtained @xmath93s@xmath16 ( observed ) and @xmath94s@xmath16 ( corrected for extinction ) , for electron temperature @xmath95k . the total number of ionizing photons corrected for extinction is in excellent agreement with a number of estimates including those of puxley et al . ( 1989 ) , carlstrom & kronberg ( 1991 ) and frster schreiber et al . ( 2001 ) . as for m82 , the extinction to the stars was computed from the @xmath85 color map of ngc 253 ( figure 7b ) . it shows values between @xmath96mag and @xmath97mag and is very similar to that of sams et al . ( 1994 ) derived from a @xmath98 color map . we find a similar value of the extinction ( @xmath99mag ) for the point with the highest value of the extinction which was identified by sams et al . ( 1994 ) as the true nucleus of ngc 253 . to calculate the extinction to the gas in ngc 253 we used pa@xmath1 and br@xmath43 fluxes ( the latter from puxley & brand 1995 and harrison et al . 1998 ) . both works centered their apertures on the approximate coordinates of the secondary near - infrared peak and therefore we will assume that the largest br@xmath43 fluxes in these works correspond to the primary near - infrared peak . if their ( 0,0 ) position were to be identified with the peak of the continuum emission , their fluxes would not be consistent with forbes et al . ( 1993 ) br@xmath43 line emission map . this map and our pa@xmath1 line emission map ( figure 6b ) show that the brightest hydrogen line emission arises from the peak of the continuum emission . in table 3 we give estimates for the extinction for the primary near - infrared peak ( center ) and secondary near - infrared peak ( positions 3 and 3.9 to the ne of the primary peak for puxley & brand 1995 and harrison et al . 1998 works respectively ) . the errors take into account 10% and 20% combined flux calibration / centering uncertainties for the nucleus and ne source , respectively . the values of the gas extinction to these two regions as derived from the pa@xmath1/br@xmath43 line ratio are consistent ( to within the errors ) with the extinction to the stars from the @xmath85 color map . despite this apparent good agreement we caution the reader of the aforementioned centering uncertainties . we also measured the observed and corrected for extinction pa@xmath1 fluxes which yielded a number of ionizing photons of @xmath100 ( observed ) and @xmath101 ( corrected for extinction ) , for the field of view shown in figure 6b . the latter value is consistent with engelbracht et al . ( 1998 ) and keto et al . ( 1999 ) estimates of the number of ionzing photons from their large aperture br@xmath43 and [ neii ] measurements , but smaller than puxley et al . ( 1997 ) estimate ( @xmath102 ) from millimeter - wavelength hydrogen recombination lines . to determine whether there is a spatial correspondence between the compact radio sources and the [ feii ] sources in both m82 and ngc 253 , very precise coordinates are needed . unfortunately , due to problems with the _ hst _ telemetry during the observations of m82 , and the uncertainties when producing and rotating the mosaics , the coordinates of the nic3 images may not be completely reliable . in addition , it is expected that there are minor discrepancies between the radio and the optical / infrared coordinate reference systems . our approach was to determine the coordinates of the [ feii ] image by trying to find the highest possible number of coincidences between the positions of the radio snrs and the [ feii ] sources . we use daofind within the daophot package in iraf to find compact sources . the term `` compact '' is not used in the sense of point sources . in fact , most of these sources seem to be slightly resolved , although due to the under - sampling of the nic3 psf , some uncertainties remain . the sources were selected to be easily distinguishable from the extended diffuse emission detected above and below the plane of the galaxy . in m82 we identified 65 `` compact '' sources in the [ feii]@xmath0 m map ( not corrected for extinction ) . for the radio sources in m82 we used the positions of snrs identified by searching for compact features within radio maps at 408mhz , 1.4ghz , 5ghz and 8.4ghz , with peak flux densities above 1mjy ( dr . karen wills , private communication 2000 ) . we excluded from this list radio sources that are classified as hii regions based on their radio spectral index ( see mcdonald et al . 2002 for details ) , resulting in 44 radio snrs . we began the comparison of positions using the coordinates in the header of the nicmos nic3 f164n images and made small offsets of the [ feii ] image , cross - correlating the positions of the [ feii ] sources with the radio snrs . based on the resolution of the nic3 images and the average angular size of the radio snrs ( @xmath103 , huang et al . 1994 ) , we counted a coincidence when the angular separation between a radio source and an [ feii ] source was @xmath104 . with a small shift , we found 21 coincidences between radio snrs and [ feii ] sources , whereas if we used the original coordinates we could only find 7 coincidences . monte carlo simulations showed that the probability of finding 21 coincidences by pure chance is @xmath105 ( see whitmore & zhang 2002 for details on the method ) , so we believe that now the infrared and the radio images are properly aligned . for ngc 253 we found discrepancies between the coordinates derived for the near - infrared peak of emission from the nic2 and nic3 images ( which were taken during two different orbits ) . although no problems were reported for these two sets of observations , some error must be affecting the coordinates given in the headers of the images . again as with m82 , we cross - correlated the positions of the [ feii]@xmath0 m ( 36 sources ) and 35 radio sources taken from ulvestad & antonucci 1991 excluding those sources that are classified as hii regions in ulvestad & antonucci ( 1997) using the coordinates of the nic3 images . we find a total of 9 coincidences for a separation between the radio sources and [ feii ] sources @xmath106 . the probability of finding 9 coincidences in by pure chance in ngc 253 is approximately @xmath107 . one important conclusion , despite the uncertainties of the image alignment , is that we can not assign an [ feii ] source to each radio snr or vice versa . we find approximately @xmath3% of radio snrs in m82 and ngc 253 have an [ feii ] . this point is illustrated in the contour images of the [ feii]@xmath0 m line emission and continuum emission at @xmath25 m in figures 2a and 2b for m82 and figures 5a and 5b for ngc 253 . in all figures we show the positions of the radio snrs as asterisks . we also tried to determine if there is a relation between the presence of a coincidence and the age of the radio snr . such a correlation would be expected if the [ feii ] emission were brighter in older supernova remnants . for m82 we used the angular sizes of the radio snr as an indication of their age ( see huang et al . in general , we did not find that older radio snrs tend to have a coincidence with an [ feii ] source . we performed aperture photometry on the [ feii ] sources identified in m82 image registered to match the field of view and position of the pa@xmath1 image , the number of detected sources ( 71 ) is slightly different from that in section 5.1 ] and ngc 253 using a fixed @xmath108 diameter aperture . this aperture corresponds to a physical size of 12pc and 10pc for m82 and ngc 253 , respectively . figures 9a and 9b show the distributions of the measured [ feii]@xmath0 m luminosities ( not corrected for extinction ) . some of the sources may be larger than the aperture used for the photometry ; on the other hand , patchy extinction will make some of the sources appear smaller . from inspection of typical source sizes , however , the luminosities would be accurate on average . recent studies of galactic snr show that the [ feii ] emission is filamentary and have sizes of the order of 11pc ( e.g , ic443 rho et al . 2001 ; see also morel et al . 2002 for snr in m33 ) . for the assumed distances to m82 and ngc 253 the observed [ feii]@xmath0 m luminosities of these sources range from @xmath109w to @xmath110w . as a consistency check for the photometry , we measured the [ feii]@xmath0 m fluxes for the fe1 , fe2 and fe3 sources in m82 identified by greenhouse et al . ( 1997 ) using the aperture diameters listed in their table 1 . the agreement with greenhouse et al . ( 1997 ) fluxes for these sources is excellent , to within @xmath111 . however , when we compared our photometry for the sources reported in forbes et al . ( 1993 ) for ngc 253 we found some discrepancies . the corrected [ feii]@xmath0 m luminosities for compact sources in m82 and ngc 253 range from @xmath112w to @xmath113w . as a comparison , galactic snrs yield [ feii]@xmath0 m luminosities of @xmath114w to @xmath115w ( oliva et al . 1989 ; keller et al . 1995 ) , snrs in the lmc yield luminosities from @xmath116w to @xmath112w ( oliva et al . 1989 ) and snrs in m33 ( assuming a distance of 720kpc ) @xmath117 to @xmath112w ( lumsden & puxley 1995 and morel et al . 2002 ) . the individual sources account for some 22% and 27% of the total [ feii]@xmath0 m emission ( extinction corrected ) measured in m82 and ngc 253 , respectively . given the estimates that the [ feii ] emitting phase of a snr has a duration of @xmath118yr ( lumsden & puxley 1995 ; morel et al . 2002 ) and the supernova rates in m82 and ngc 253 are @xmath119yr@xmath16 ( huang et al . 1994 and ulvestad & antonucci 1997 ) , the detected [ feii ] sources account for @xmath120yr worth of supernovae . presumably much of the remaining unresolved [ feii ] arises from additional supernovae that are packed so densely that they are confused in our images . also , snrs in these two galaxies may expand and merge into a general hot ism within a few thousand years losing their identity as individual sources . photometry on the [ feii ] emission arising from the disk of m82 shows that as much as 70% of the total emission ( disk plus component perpendicular to the disk ) is associated with snrs . this together with the good agreement between the radio and [ feii ] morphologies of the components perpendicular to the disk gives further indication for a snr origin of the [ feii ] emission in starburst galaxies . lumsden & puxley ( 1995 ) pointed out that one possibility is that the [ feii ] lines reach maximum luminosity when the remnant as a whole undergoes the transition from adiabatic expansion to radiative expansion . according to shull & draine ( 1986 ) , this happens when radiative cooling occurs faster than dynamical times , around @xmath121yr after the supernova explosion . therefore if the brightest [ feii ] emission occurs at approximately @xmath121yr , it is not surprising that we do not find a correspondence for all the [ feii ] sources and all the radio snrs . muxlow et al . ( 1994 ) found that half of the radio sources in m82 are unresolved indicative of a young age , but the other resolved half are probably intermediate age . this together with the fact that we found that @xmath122 of the radio snr have an [ feii ] counterpart implies that the [ feii ] sources may be @xmath123yr . one possibility for the lack of one - to - one coincidences between the radio snrs and the [ feii ] sources would be that most of the [ feii ] emission in these two galaxies is produced in hii regions ( photoionization ) . to check this , we measured the pa@xmath1 fluxes for the sources detected in the [ feii ] extinction corrected map of m82 , and assumed an [ feii]@xmath0m / pa@xmath1 ratio for hii regions as that observed in orion . we found that only @xmath124 of the total [ feii]@xmath0 m arises from hii regions . this further supports the hypothesis that most of the [ feii]@xmath0 m emission along the plane of m82 is produced by snrs . the [ feii]/br@xmath43 ratios in hii regions and snr differ by a factor of a hundred or more , and therefore can be used to indicate the excitation mechanism of the [ feii ] line within our images . in alonso - herrero et al . ( 1997 ) we showed that the behavior of the [ feii]@xmath0m / br@xmath43 line ratio in starbursts and active galaxies can be understood as a progression from pure photoionization to pure shock excitation going from pure hii regions through starbursts , transition objects , seyferts to supernova remnants . figures 8a and 8b are the observed [ feii]@xmath0m / pa@xmath1 maps for m82 and ngc 253 , respectively . the values of the [ feii]@xmath0m / pa@xmath1 ratio range from 0.02 in black ( mostly along the disk of the galaxy ) to 0.8 in white ( some of the bright [ feii ] sources ) . these ratios translate into [ feii]@xmath0m / br@xmath43 line ratios of between 0.24 and 10 ( see also the histograms in figure 10 ) . since these maps are not corrected for extinction , these numbers are actually lower limits . when the positions of radio sources are plotted in the [ feii]@xmath0m / pa@xmath1 maps we find again that there is a number of coincidences between the radio snrs and the regions of enhanced [ feii]@xmath0 m emission . this indicates that the relative contribution of the snrs is nearly as large in the diffuse background as it is in the radio snr . interestingly , we find that most of the coincidences with the radio snrs occur in regions of enhanced [ feii]@xmath0m / pa@xmath1 line ratio , or conversely we do not find a large number of coincidences in regions with high pa@xmath1 flux ( that is , hii regions ) . the supernova rate in m82 has been estimated from number counts of the radio supernova remnants and their ages to be @xmath125yr@xmath16 ( huang et al . the total [ feii]@xmath0 m luminosity in m82 is @xmath126([feii]@xmath127w . our value for the [ feii ] luminosity of m82 is approximately 1.5 and 2 times higher than the previous estimates by greenhouse et al . ( 1997 ) and vanzi & rieke ( 1997 ) , respectively . this difference is most likely due to better extinction estimates made possible by our nicmos images . if the supernova rate in m82 is @xmath128yr@xmath16 , then we can derive the following calibration for the supernova rate in terms of the [ feii]@xmath0 m luminosity : @xmath129}}{10^{34 } { \rm w}}\,\,\ , ( { \rm yr}^{-1})\ ] ] the total [ feii]@xmath0 m luminosity ( corrected for extinction ) measured in ngc 253 is @xmath130 . an upper limit on the supernova rate from radio supernova counts is @xmath131 ( ulvestad & antonucci 1997 ) . a rate of @xmath132 has been calculated assuming that the non - thermal radio flux is entirely due to synchrotron emission ( van buren & greenhouse 1994 ) ; this value is consistent with the similar number of compact radio sources in this galaxy compared with m82 . the calibration of supernova rate in terms of [ feii]@xmath0 m luminosity for this galaxy is then @xmath133}}{10^{34 } { \rm w}}\,\,\ , ( { \rm yr}^{-1})\ ] ] since m82 is the more thoroughly studied galaxy in the radio , we give higher weight to the calibration for it in averaging these results to estimate @xmath134}}{10^{34 } { \rm w}}\,\,\ , ( { \rm yr}^{-1})\ ] ] in the above calculations , we have included the entire [ feii ] luminosity , since that is the parameter usually measured in a distant galaxy . however , in these nearby examples , we can compare the [ feii ] emission produced in hii regions with that from supernovae . we assume that the ratio for a typical hii region is @xmath135([feii]@xmath0m@xmath136 ( e.g. , orion ) . we obtained that the [ feii ] luminosity produced in hii regions accounts for less than 8% the total [ feii ] luminosity in m82 . we have presented _ hst_/nicmos observations of the starburst galaxies m82 and ngc 253 . the morphology of the [ feii]@xmath0 m line emission in both galaxies displays two components : ( 1 ) a disk component with a number of superimposed compact sources and ( 2 ) a diffuse component in the form of filaments above and below the disk of the galaxy . in m82 we find a remarkable similarity between the extended component , the fingers or chimneys above and below the disk , of the radio and [ feii ] emissions . in m82 , we have detected in both pa@xmath1 and [ feii ] a nuclear ring of star formation . this ring , with an approximate diameter of 400pc , has also been observed in other near- and mid - infrared hydrogen recombination lines and in the [ neii]@xmath34 m line . the peak of the @xmath35 m continuum brightness of m82 lies very close to ( within 1 ) the dynamical center , and corresponds to the center of the ring of star formation . in ngc 253 , bright pa@xmath1 emission stems from the near - infrared continuum peak and extends in a plume - like morphology along the disk of the galaxy . there is also a diffuse pa@xmath1 component above and below the plane of galaxy , however , much fainter than that in m82 . prior to calibrating the surpernova rate in terms of the [ feii ] emission , we have derived the extinction to the central regions of ngc 253 and m82 . to assess the extinction correction , we have compared the number of ionizing photons infered from the extinction - corrected pa@xmath1 maps with other estimates based on longer wavelength indicators , and found a good agreement . since the [ feii ] emission in starburst galaxies is thought to be associated with supernova activity , we have cross - correlated the positions ( with a spatial resolution of a few tenths of arcseconds ) of the radio snrs and the identified compact [ feii ] sources in m82 and ngc 253 . we have found that approximately @xmath3% of the radio snrs have an [ feii ] counterpart . however , we can not associate an [ feii ] source to each radio snr or vice versa . this is well understood in terms of the different ages traced by the radio snrs ( a few hundred years ) and the [ feii ] stage of snrs ( @xmath4yr ) . the compact [ feii ] emitting sources in m82 and ngc 253 only contribute to some @xmath137 of the total [ feii]@xmath0 m emission . however , much of the remaining emission in the disk of the two galaxies is most likely produced by snrs that expanded and merged into a general ism a few @xmath138yr ago . photometry on the [ feii]@xmath0 m emission arising from the disk of m82 shows that as much as @xmath139 of the total [ feii]@xmath0 m emission ( corrected for extinction ) is associated with snrs . the remaining diffuse [ feii ] emission perpendicular to the disk of m82 seems to be related to the superwind ( via expulsion of material from the central starburst ) , as argued from the radio emission morphology ( wills et al . 1999 ) . all these arguments support the supernova origin for the observed near - infrared [ feii ] line emission in starburst galaxies . thus we have recalibrated the supernova rate in terms of the [ feii ] luminosity relation for starburst galaxies using our newly measured ( corrected for extinction ) [ feii]@xmath0 m luminosities of m82 and ngc 253 . we would like to thank drs . karen wills , t. muxlow and a. pedlar for kindly providing us with the radio images for comparison with the nicmos images , and the list of radio snr . we are also grateful to dr . ernesto oliva for enlightening discussions , and an anonymous referee for constructive comments that helped improve the paper . during the course of this work aa - h was supported by the national aeronautics and space administration on grant nag 5 - 3042 through the university of arizona . the work was also partially supported by the national science foundation under grant ast-95 - 29190 . satyapal , s. , watson , d. m. , pipher , j. l. , forrest , w. j. , coppenbarger , d. , raines , s. n. , libonate , s. , pich , f. , greenhouse , m. a. , smith , h. a. , thompson , k. l. , fisher , j. , woodward , c. e. , & hodge , t. 1995 , apj , 448 , 611

the description of quantum mechanical systems by means of the schrdinger equation has been developed more than 80 years ago . at that time only a few resonance states in nuclei and atoms were known which are well separated from one another . the energies of these states are well described by means of the schrdinger equation with a hermitian hamiltonian . in order to describe the finite lifetimes of these states , the r - matrix theory has been developed which is however too complicated for practical calculations . the finite lifetimes of the individual states are calculated usually perturbatively , see e.g. @xcite . later , systems at high level density were in the center of interest . under these conditions , a statistical description of the system turned out to be very efficient @xcite . the resonances are well isolated from one another , and the lifetimes of the states are very long and do not play any role . in the course of time , experimental studies have been performed for different quantum systems with a much improved accuracy . also the theoretical calculations are carried out today not only for states being well separated in energy from one another but also for resonance states in the regime of overlapping . here the single individual states can no longer be identified what results in problems of their interpretation . contradictions between experimental results and conventional hermitian quantum physics appeared in different small quantum systems . for example , heiblum et al . @xcite found experimentally a crossover from the mesoscopic to a universal phase for electron transmission in quantum dots about than 10 years ago . these results could not be explained in conventional quantum physics in spite of much effort @xcite . recently koehler et al . @xcite observed that neutron resonance data exclude random matrix theory . deviations between experimental data and random matrix theory in nuclear physics studies were observed also earlier , e.g. @xcite . these and other experimental results show that the schrdinger equation originally introduced with a hermitian hamilton operator for the description of well isolated resonances , has to be extended . above all , the lifetimes of the resonance states have to be calculated also in the regime of overlapping resonances and the justification of statistical assumptions has to be proven for small quantum systems . the lifetimes of the resonance states can be calculated when the system is explicitly considered to be open and the calculations are performed quantum mechanically for both the system and the environment of scattering wavefunctions into which the system is embedded . using the feshbach projection operator technique @xcite , first the _ energy - independent _ many - body problem of the system ( with the hermitian hamiltonian @xmath4 ) is solved in the standard manner . these solutions provide the energies @xmath5 and wavefunctions @xmath6 of the discrete states with inclusion of the so - called _ internal _ interaction . in a second step , the _ energy - dependent _ scattering wavefunctions @xmath7 of the environment are calculated and , further , the ( energy - dependent ) coupling matrix elements @xmath8 between the discrete states of the system and the environment are evaluated ( see @xcite , section 2.1 ) . the corresponding schrdinger equation contains an energy dependent ( nonlinear ) source term describing the coupling between system and environment . the hamiltonian @xmath9 of this equation is non - hermitian and provides the lifetimes of the states , but without any additional interaction of the states via the environment . the schrdinger equation with @xmath9 and source term can be rewritten into a schrdinger equation without source term but with a non - hermitian hamiltonian @xmath10 that contains the interaction of the states via the environment ( the so - called _ external _ interaction ) in the nondiagonal matrix elements @xcite . the hamiltonian @xmath11 reads @xmath12 where @xmath13 and @xmath14 stand for the coupling between system and environment and @xmath15 is the green function in the subspace of scattering states . the external interaction of the states via the continuum is complex , generally . the principal value integral is @xmath16 and the residuum reads @xmath17 the interaction of the states of the system via the environment is involved in the eigenvalues @xmath18 and eigenfunctions @xmath19 of the hamiltonian @xmath10 . it is therefore relatively easy , in this formalism , to study the influence of the environment onto the states of the system . that means , the non - hermitian quantum physics is in contrast to the widely spread meaning _ not _ a further approximation introduced in the theory . it is an expression of the fact that the system considered is really open and its properties are influenced by the coupling to the environment . this influence is unimportant at low level density where it can be described by perturbation theory . it becomes , however , decisive in the regime of overlapping resonances . meanwhile , many calculations for open quantum systems are performed with high accuracy . without using any perturbation theory or statistical assumptions , the schrdinger equation with the non - hermitian hamilton operator @xmath10 is solved and the eigenvalues @xmath20 and eigenfunctions @xmath19 are obtained . that means , not only the energies @xmath21 of the states of the system are evaluated but also their lifetimes @xmath22 . the control of the eigenvalues by a parameter allows to draw conclusions on the dynamics of open quantum systems . the limits of the applicability of the standard quantum theory with hermitian hamilton operator can be seen best at high level density , where the individual resonance states overlap and interact with one another via the continuum of scattering wavefunctions . here , the dynamics of the system is determined by singular points , the so - called _ exceptional points_. they cause level repulsion in energy as well as a bifurcation of the widths ( inverse proportional to the lifetimes ) of the states . results are obtained theoretically as well as experimentally , which are counterintuitive at first glance : due to width bifurcation , coherent short - lived states are formed together with states that are almost decoupled from the continuum of scattering states @xcite . an example is the above - mentioned crossover from the mesoscopic to a universal phase in the transmission in quantum dots @xcite and its qualitative explanation on the basis of a schrdinger equation with non - hermitian hamilton operator @xcite . this phenomenon is observed in different systems and is called usually _ dynamical phase transition _ @xcite . the short - lived states are analog to the coherent superradiant states considered by dicke @xcite in 1954 as shown by means of a toy model @xcite . moreover , it could be shown in an experiment on non - locally coupled pairs of quantum point contacts that discrete states undergo a robust interaction that is achieved by coupling them to each other through the continuum @xcite . most of these results are of high value for fundamental questions of quantum mechanics as well as for applications . it is the aim of the present paper to show some generic ( mostly numerical ) results for open quantum systems in order to receive a deeper understanding for the dynamical phase transitions occurring in the regime of overlapping resonances . a toy model is used in order to receive conclusions on the role played by exceptional points and avoided level crossings at high level density in open quantum systems . of special interest are the effects arising from the imaginary part ( [ form12 ] ) of the coupling term via the environment . the calculations are performed with respectively two , four and ten resonance states coupled mostly to one open decay channel ( corresponding to the common continuum of scattering wavefunctions ) under the condition that they avoid crossing or even cross in a certain parameter range . the coupling of @xmath23 resonance states to @xmath24 channels corresponds to the general situation of open quantum systems . the formalism used in the calculations is the same as that discussed in our earlier paper @xcite . the results show very clearly that the imaginary part of the coupling term between system and environment causes width bifurcation and , finally , the formation of different time scales in the regime of overlapping resonances . due to width bifurcation , the system splits into two parts that exist at different times . this process is irreversible . the wavefunctions of the short - lived _ aligned _ states are mixed coherently in relation to the open decay channel while those of the long - lived _ trapped _ states are mixed incoherently such that the states are almost decoupled from the open decay channel . the results confirm the characteristic features of the dynamical phase transitions appearing in open quantum systems at high level density . they show that the number of states of the system is reduced during the dynamical phase transition due to ejecting the aligned short - lived state and , furthermore , that the wavefunctions of the states before and beyond the dynamical phase transition are completely different from one another . while the states of the original system ( without interaction via the continuum ) have individual spectroscopic features , those of the system consisting of the long - lived states beyond the dynamical phase transition show chaotic features . these results prove once more that the experimental results @xcite on the crossover from the mesoscopic to a universal phase for electron transmission in quantum dots can be explained by means of the non - hermiticity of the hamiltonian . additionally it should be mentioned here that the aligned state corresponds to the dicke superradiant state and the trapped states to the subradiant states which are considered in optics , e.g. @xcite . in section [ form ] , we sketch the formalism used in the present calculations . for @xmath25 states with equal decay widths , the analytical and numerical solutions of the problem show clearly the effect of width bifurcation . numerical results for eigenvalues and eigenfunctions obtained for @xmath25 , 4 and 10 states are provided in sections [ 2 ] to [ v10 ] . the results are discussed in section [ disc ] while conclusions on the relation between exceptional points and width bifurcation are drawn in the last section [ concl ] . we consider an @xmath26 matrix @xmath27 the diagonal elements of which are the @xmath23 complex eigenvalues @xmath28 of a non - hermitian operator . the @xmath29 are the so - called selfenergies of the states arising from their coupling to the environment of scattering wavefunctions into which the system is embedded . in atomic physics , these values are known as lamb shift . our calculations are performed with coupling matrix elements @xmath29 the values of which do not depend on the parameter considered . in such a case , the @xmath29 can considered to be included into the diagonal matrix elements , which read @xmath30 . the @xmath31 and @xmath32 denote the energies and widths , respectively , of the @xmath23 states ( including their selfenergies ) without account of the interaction of the different states via the environment . the internal interaction of the two states @xmath33 and @xmath34 ( appearing in the closed system ) _ as well as _ their external interaction ( via the environment ) are contained in the @xmath35 . the internal interaction can be caused only by some part of re@xmath36 while the external interaction contains complex @xmath35 , see equations ( [ form11 ] ) and ( [ form12 ] ) . most interesting part of the external interaction is therefore im@xmath36 . it becomes important at high level density where the corresponding resonance states overlap . when the number @xmath23 of states is equal to the number @xmath37 of common channels ( i.e. equal to the number @xmath37 of different common continua of scattering wavefunctions ) all the coupling matrix elements @xmath35 are different from zero and the matrix ( [ form1 ] ) is full . in the case with only one open decay channel @xmath38 , all @xmath35 different from @xmath39 and @xmath40 are zero . an example of @xmath41 states coupled to only the fourth channel is the following matrix @xmath42 with @xmath43 for @xmath44 and @xmath45 . the eigenvalues of @xmath11 will be denoted by @xmath46 where @xmath21 and @xmath47 stand for the energy and width , respectively , of the eigenstate @xmath33 . the eigenfunctions of the non - hermitian @xmath11 are biorthogonal ( see sections 2.2 and 2.3 of @xcite ) , @xmath48 it follows @xmath49 and @xmath50 at an exceptional point , @xmath51 and @xmath52 . the @xmath53 and @xmath19 contain global features that are caused by many - body forces induced by the coupling @xmath35 of the states @xmath33 and @xmath34 via the environment , see equations ( [ form11 ] ) , ( [ form12 ] ) and the corresponding discussion in @xcite . in the case @xmath25 , the two eigenvalues of @xmath10 are @xmath54 according to this expression , two interacting discrete states ( with @xmath55 ) avoid always crossing since @xmath0 and @xmath56 are real in this case . resonance states with nonvanishing widths @xmath47 repel each other in energy according to the value of re@xmath57 while the widths bifurcate according to the value of im@xmath58 . the two states cross when @xmath59 . this crossing point is an exceptional point according to the definition of kato @xcite . at the exceptional point @xmath59 , the eigenfunctions of ( [ form1 ] ) of the two crossing states are linearly dependent from one another , @xmath60 according to analytical as well as numerical and experimental studies , see appendix of @xcite and section 2.5 of @xcite . that means , the wavefunction @xmath61 of the state @xmath62 jumps , at the exceptional point , via the wavefunction @xmath63 of a chiral state to @xmath64 @xcite . from ( [ eif5 ] ) follows : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \(i ) when two levels are distant from one another , their eigenfunctions are ( almost ) orthogonal , @xmath65 . + ( ii ) when two levels cross at the exceptional point , their eigenfunctions are linearly dependent according to ( [ eif5 ] ) and @xmath66 . + _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ these two relations show that the phases of the two eigenfunctions relative to one another change when the crossing point is approached . this can be expressed quantitatively by defining the _ phase rigidity _ @xmath67 of the eigenfunctions @xmath68 , @xmath69 it holds @xmath70 . the non - rigidity @xmath67 of the phases of the eigenfunctions of @xmath11 follows also from the fact that @xmath71 is a complex number ( in difference to the norm @xmath72 which is a real number ) such that the normalization condition ( [ int3 ] ) can be fulfilled only by the additional postulation im@xmath73 ( what corresponds to a rotation ) . when @xmath74 , an analytical expression for the eigenfunctions as a function of a certain control parameter can , generally , not be obtained . an exception is the special case @xmath75 for which @xmath76 . in this case , the condition @xmath59 can not be fulfilled if @xmath77 is real due to @xmath78 the exceptional point can be found only by analytical continuation into the continuum @xcite and the two states avoid crossing . this is analogue to the avoided level crossings of discrete states . the condition @xmath59 can be fulfilled however if @xmath79 is imaginary , @xmath80 and two exceptional points appear . it holds further @xmath81 independent of the parameter dependence @xmath82 . in the first case , the eigenvalues @xmath83 differ from the original values @xmath84 by a contribution to the energies and in the second case by a contribution to the widths . the width bifurcation starts at one of the exceptional points and becomes maximum in the middle between the two exceptional points . this happens at the crossing point @xmath85 where @xmath86 . some years ago , the case @xmath25 with @xmath87 , fixed real @xmath88 , and different fixed values of @xmath32 , including @xmath89 , is studied as a function of the parameter @xmath90 in the neighborhood of avoided and true crossings of the two levels @xcite . the results for the @xmath25 case @xcite show further that the wavefunctions of the two states @xmath61 and @xmath91 are mixed in a finite range of the parameter @xmath90 around the critical value @xmath92 at which the two states avoid crossing . this holds true not only for resonance states but also for discrete states . the eigenfunctions @xmath19 of @xmath10 can be represented in the set of basic wavefunctions @xmath93 of the unperturbed matrix ( corresponding to the case with vanishing coupling matrix elements @xmath94 ) , @xmath95 also the @xmath96 are normalized according to the biorthogonality relations of the wavefunctions @xmath97 . in our calculations , the mixing coefficients @xmath96 of the wavefunctions of the two states due to their avoided crossing are not calculated . we simulate the fact that the two wavefunctions are mixed in a finite parameter range around the critical value of their avoid crossing @xcite by assuming a gaussian distribution @xmath98 for the coupling coefficients . the results reproduce very well those discussed in @xcite for 2 levels and real coupling @xmath0 . of special interest is the situation at high level density where the ranges of avoided crossings , defined by ( [ int7 ] ) , of different levels overlap . some generic results obtained with 2 , 4 and 10 resonance states will be presented in the following sections . we start our calculations with the two - level case which is studied mostly in literature . we choose the matrix ( [ form1 ] ) with @xmath87 , different fixed values of @xmath32 and fixed @xmath0 . the functional dependence of the energies over the parameter @xmath90 is similar as in @xcite . however , the @xmath0 are real in @xcite while they are mostly complex in the present paper as well as in @xcite . according to ( [ int6 ] ) to ( [ int6d ] ) , the exceptional points appear at different values of the parameter @xmath90 when the ratio re@xmath1 to im@xmath1 is varied , see also figs . 1 to 3 in @xcite . in fig . [ fig1 ] we show the numerical results for the avoided level crossing phenomenon as a function of the parameter @xmath90 beyond , at and below the exceptional point . in all cases , the value of @xmath0 is fixed to @xmath100 while the @xmath101 are different in the different subfigures ( with @xmath102 in all cases ) . the exceptional point appears at @xmath103 , see figs . [ fig1].c , d. when @xmath104 ( figs . [ fig1].a , b ) , the eigenvalue trajectories cross in energy while the widths @xmath105 and @xmath106 are always different from one another . the situation is another one when @xmath107 . here , the eigenvalue trajectories avoid crossing in energy while the trajectories @xmath105 and @xmath106 cross at certain values of @xmath90 ( figs . [ fig1].e to h ) . the comparison of fig . [ fig1 ] with complex @xmath108 and fig . 2 in @xcite with real @xmath109 shows the influence of im@xmath1 . while the @xmath110 vary symmetrically around the critical value @xmath111 when @xmath0 is real , this is not the case when @xmath0 is complex . in the last case , the difference @xmath112 blows up in the critical region . furthermore , the @xmath47 approach their asymptotic values in a relatively small parameter range of @xmath90 when @xmath0 is real in contrast to the case with complex @xmath0 . these results will be discussed further in the following sections . and widths @xmath113 ( full lines ) of @xmath25 states beyond ( a , b ) , at ( c , d ) and below ( e to h ) the exceptional point . the parameters of the subfigures are @xmath114 ( a , b ) ; @xmath115 ( c , d ) ; @xmath116 ( e.f ) ; @xmath117 ( g , h ) . further parameters : @xmath118 . the dashed lines show @xmath82 . , width=529,height=529 ] of @xmath25 states beyond ( a ) , at ( b ) and below ( c , d ) the exceptional point . the parameters are the same as in fig . [ fig1 ] . , width=264,height=529 ] in fig . [ fig2 ] , we show the mixing coefficients @xmath119 of the eigenfunctions as a function of the parameter @xmath90 corresponding to the eigenvalue figures shown in fig . [ fig1 ] . the mixing coefficient is complex , @xmath120 . this figure can be compared with fig . 5 in @xcite calculated exactly with real @xmath109 . in both cases , the difference between the two curves @xmath119 is always 1 when @xmath104 . this result corresponds to the fact that the two states will not be exchanged when @xmath104 . the energy trajectories of the eigenstates cross freely and the widths of the two states are always different from one another , see figs . [ fig1].a , b. at the exceptional point ( where @xmath121 ) , @xmath122 in both cases . when @xmath107 , the two curves @xmath119 coincide at a certain value of @xmath90 . this happens for real ( see @xcite ) as well as for complex ( fig . [ fig2 ] ) @xmath0 , but is symmetrically only for real @xmath0 . at these points , the two states are exchanged . in both cases @xmath107 and @xmath104 , the asymptotic values 1 and 0 are reached in a smaller parameter range when @xmath0 is real than in the case with complex @xmath0 . within this parameter range the two eigenfunctions are mixed . the range of mixing shrinks to one point when the two states really cross ( at the exceptional point ) . it is especially large when two discrete states ( with @xmath123 ) avoid crossing , as can be seen from fig . 5 in @xcite ( where the calculations are performed exactly with real @xmath0 for all @xmath96 ) . the parameter range in which two states are mixed due to the existence of an exceptional point in the neighborhood plays an important role in realistic systems it will be discussed in the following sections with more than one avoided crossing for both real and complex @xmath0 . and widths @xmath113 ( full lines ) of @xmath25 states ( a , b , g , h ) and @xmath41 states ( c to f and i , j ) coupled to @xmath38 channel . the parameters of the subfigures are @xmath125 ( a to d ) ; @xmath126 ( e , f ) ; @xmath127 ( g , h ) ; @xmath128 ( i , j ) . further parameters : @xmath129 . the dashed lines show @xmath82 . , width=491,height=642 ] but @xmath130 ( a , b , e , f ) ; @xmath109 ( c , d , g , h ) ; @xmath41 ; @xmath131 ( a to d ) ; @xmath132 ( e to h ) . , width=529,height=529 ] but @xmath41 and @xmath133 ; @xmath130 ( a , b ) ; @xmath134 ( c , d ) ; @xmath109 ( e , f ) ; @xmath135 . + , width=529,height=396 ] let us first compare the calculations with @xmath25 and @xmath41 states and imaginary coupling @xmath134 between discrete states and environment of scattering states ( fig . [ fig3 ] ) . in the calculations , the eigenvalue trajectories @xmath136 and @xmath110 are traced as a function of the parameter @xmath90 which is varied in such a manner that one state crosses in energy one and three , respectively , other states . the results show two exceptional points in both cases appearing at the two different parameter values @xmath137 and @xmath138 , see figs . [ fig3].a , b for two states and figs . [ fig3].c to f for four states . in between these two values , the widths bifurcate : the differences @xmath139 blow up . the two - level case is described analytically by equations ( [ int6 ] ) to ( [ int6d ] ) . the width bifurcation is completely reproduced in the numerical results . the parameter range @xmath140 is larger in the case with four states than in the other one due to the larger region in which avoided level crossings take place . furthermore , the difference @xmath141 is larger in the case of 4 states than in the case of 2 states . in the case that all states have equal widths @xmath142 , the widths @xmath113 of only two states bifurcate also in the case with four states . the widths of the two other states remain unchanged ( figs . [ fig3].c , d ) . this result holds also when the different @xmath142 differ slightly from one another ( figs . [ fig3].e , f ) . in this latter case , the two states avoid crossing at the two critical values @xmath137 and @xmath138 and the width bifurcation is almost the same as in the case with equal widths @xmath32 . the situation changes completely when the widths @xmath32 differ stronger from one another . when @xmath143.g to j ) . the eigenvalue trajectories cross freely in energy and the widths do ( almost ) not bifurcate . the dynamics of the system is therefore completely different from that determined by the results shown in figs . [ fig3].a to f. in fig . [ fig4 ] , we show the eigenvalues @xmath21 and @xmath113 for the case with four levels and one channel with complex and real coupling coefficients ( @xmath120 and @xmath109 , respectively ) . width bifurcation can be seen when @xmath0 is complex and the widths @xmath32 are equal ( figs . [ fig4].a , b ) , while the @xmath144 are independent of @xmath90 when @xmath0 is real ( figs . [ fig4].c , d ) . the results with different @xmath32 differ from those obtained with imaginary coupling @xmath145 ( figs . [ fig3].g to j ) : the widths of two states cross at parameter values at which their energies avoid crossing ( figs . [ fig4].e to h ) . in fig . [ fig5 ] , we show the results of calculations with equal number of states and channels , @xmath41 and @xmath133 . although usually @xmath24 in realistic systems , the results allow us to receive a deeper understanding for the spectroscopic redistribution processes taking place in the critical region . the calculations are performed with the same parametric dependence of the energies as in the foregoing calculations with @xmath41 and @xmath38 . further assumptions : @xmath146 and @xmath147 , and 0.05 , respectively . the results show width bifurcation when @xmath0 is complex or imaginary ( figs . [ fig5].a to d ) which is however smaller than in the corresponding cases with @xmath38 ( figs . [ fig4].a , b and figs . [ fig3].c , d , respectively ) . instead , the widths @xmath47 of the states are changed asymptotically due to the coupling to the different channels . when @xmath0 is real ( figs . [ fig5].e , f ) , the widths @xmath113 are independent of @xmath90 and are equal to @xmath148 ( @xmath149 1 to 4 ) . of @xmath25 and @xmath41 states the eigenvalues of which are shown in fig . [ fig3 ] ; @xmath38 channel ; @xmath134 . , width=264,height=680 ] of @xmath41 states the eigenvalues of which are shown in fig . [ fig4 ] ; @xmath38 channel ; @xmath130 ( a , c ) ; @xmath109 ( b , d ) . , width=302,height=585 ] of @xmath41 states the eigenvalues of which are shown in fig . [ fig5 ] ; @xmath133 channels ; @xmath130 ( a ) ; @xmath134 ( b ) ; @xmath109 ( c ) . + , width=302,height=453 ] the coefficients @xmath96 defined in ( [ int20 ] ) determine the mixing of the eigenfunctions @xmath19 in relation to the basic wavefunctions @xmath150 . the mixing is nonvanishing in the neighborhood of the critical points of avoided level crossings as has been shown in exact numerical calculations for two crossing states and real coupling coefficients @xmath94 @xcite . in the calculations of the present paper , the mixing of the wavefunctions is simulated by assuming a gaussian distribution for the @xmath94 , equation ( [ int7 ] ) . it reproduces very well the exact results obtained in @xcite . in the following , we consider the mixing coefficients @xmath96 for all wavefunctions @xmath19 the eigenvalue trajectories of which are studied in figs . [ fig3 ] to [ fig5 ] as a function of the parameter @xmath90 . in fig . [ fig6 ] , the coefficients @xmath119 are shown for two and four states with imaginary coupling @xmath145 ( see the corresponding eigenvalue trajectories in fig . [ fig3 ] ) . when the widths @xmath32 of all states are equal ( figs . [ fig6].a , b ) , @xmath151 at the two exceptional points @xmath137 and @xmath138 . in the parameter range @xmath152 , the wavefunctions are mixed while @xmath153 and @xmath154 for @xmath155 beyond the two values @xmath156 . the states are completely ( 1:1 ) mixed in the range @xmath152 in the two - level - case ( fig . [ fig6].a ) . the picture is more complicated in the four - level - case ( fig . [ fig6].b ) . here , all states are involved in the redistribution taking place in the critical parameter range @xmath152 . the exchange of the two states the widths of which remain unchanged according to fig . [ fig3].d can be seen from the energy eigenvalue trajectories , fig . [ fig3].c , as well from the mixing coefficients , fig . [ fig6].b . the figures are similar when the widths @xmath32 of the states differ slightly from one another . instead of exceptional points , there are avoided level crossings at @xmath137 and @xmath138 , see fig . [ fig6].c for the four - level - case . the mixing coefficients show a dependence on @xmath90 which is similar to that obtained for equal widths ( fig . [ fig6].b ) . when the widths @xmath32 of the states @xmath33 differ stronger from one another than in fig . [ fig6].c , the eigenfunctions remain almost unmixed for all parameter values @xmath90 ( figs . [ fig6].d , e ) . this result corresponds to the almost constant width trajectories @xmath47 as a function of @xmath90 ( figs . [ fig3].h , j ) . an exchange of states does not take place ( figs . [ fig3].g , i ) . these results show the great influence of the exceptional points onto the dynamics of the system considered . the calculations in figs . [ fig3 ] and [ fig6 ] are performed with imaginary coupling @xmath145 . for comparison , we show in fig . [ fig7 ] the mixing coefficients @xmath119 when @xmath0 is complex and real , respectively , for the four - level - case ( corresponding to the eigenvalue trajectories in fig . [ fig4 ] ) . when all @xmath32 are equal to one another and @xmath0 is complex , the mixing coefficients @xmath119 point to the participation of all states in the redistribution process in the critical parameter range ( fig . [ fig7].a ) . this corresponds fully to the corresponding eigenvalue trajectories that show level exchange as well as width bifurcation ( figs . [ fig4].a , b ) . when the coupling is real , however , the wavefunctions of only two states are mixed while the wavefunctions of the other two states remain almost unchanged in the critical region ( fig . [ fig7].b ) . the mixing of the eigenfunctions is completely different from that discussed above when the widths @xmath32 differ more strongly from one another . a mixing of the wavefunctions appears only when the energy trajectories @xmath136 avoid crossing and the width trajectories @xmath110 cross , compare figs . [ fig4].e to h with the corresponding figs . [ fig7].c , d. a width bifurcation does not take place . the mixing coefficients @xmath119 for the case with @xmath41 states coupled to @xmath133 channels are shown in fig . [ fig8 ] ( parameters the same as in fig . [ fig5 ] ) . the differences between the cases with coupling to 4 channels to those with coupling to only 1 channel can be seen by comparing fig . [ fig8].a with fig . [ fig7].a , fig . [ fig8].b with [ fig6].b and fig . [ fig8].c with fig . [ fig7].b . the mixing of the wavefunctions by coupling the system to four channels is more complicated than that by coupling to only one channel . this statement holds true also when @xmath0 is imaginary . furthermore , an additional mixing is caused by the avoided level crossing at small @xmath90 in the four - channel - case when @xmath0 is complex . we continue our studies with the matrix ( [ int2 ] ) by choosing @xmath158 states coupled to @xmath62 channel . the eigenvalue trajectories @xmath136 and @xmath110 are shown in figs . [ fig9 ] and [ fig10 ] . in these cases , one of the states crosses in energy the remaining nine states one by one . the nine states have the same parametric energy dependence of @xmath90 and are shifted equidistantly relative to one another . and widths @xmath113 ( full lines ) of @xmath158 states coupled to @xmath38 channel . the parameters of the subfigures are @xmath125 ( a , b ) ; @xmath159 ( c to f ) ; @xmath160 ( a to d ) ; 0.07 ( e , f ) . further parameters : @xmath161 . the dashed lines show @xmath82 . , width=529,height=415 ] but @xmath134 ( a , b ) ; @xmath162 ( c , d ) ; @xmath163 ( e , f ) ; @xmath164 ( g , h ) . , width=529,height=529 ] but @xmath165 ; @xmath166 ( a , b ) ; @xmath167 ( c , d ) and @xmath168 ( e , f ) . , width=529,height=415 ] in fig . [ fig9].a , b we show the results obtained with @xmath169 and @xmath170 for all states . the results are similar to those for the 4-level case ( figs . [ fig3].c , d ) . two exceptional points appear as well as width bifurcation : the width of one of the states is much larger while that of another one is much smaller than the widths of all the other states in the whole critical parameter range between the two exceptional points . in figs . [ fig9].c to f , the widths @xmath32 differ from one another . we compare the results obtained with an imaginary and with a real coupling constant , @xmath171 and @xmath172 , respectively . when @xmath0 is real ( figs . [ fig9].e , f ) , the energy trajectories @xmath136 are very regular : avoided level crossings appear one by one with all nine states . an exceptional point can be found only by analytical continuation into the continuum ( see @xcite ) . the width trajectories @xmath110 are less regular : they cross at several parameter values @xmath90 . according to the results obtained for a smaller number of states in sections [ 2 ] to [ f4 ] , the widths do not bifurcate when @xmath0 is real . when @xmath0 is imaginary , the energy trajectories @xmath136 are also regular ( fig . [ fig9].c ) . [ fig9].d shows nicely how widths continue to bifurcate such that the broad state remains broad and the narrow state remains narrow for all @xmath90 in the critical region . mostly , the widths of two states are different from one another when the energy trajectories cross . the states can , nevertheless , be exchanged as can be seen from fig . [ fig9].c . in fig . [ fig10 ] , we show results obtained with the imaginary coupling constant @xmath145 ( being smaller than in figs . [ fig9].c , d ) as well as those obtained with the larger values @xmath173 ( figs . [ fig10].a , b and figs . [ fig10].c to h , respectively ) . we see width bifurcation similar as in fig . [ fig3].d for @xmath41 states and in fig . [ fig9].b for 10 states with equal widths @xmath32 . in any case , width bifurcation appears in the whole critical parameter range between the two exceptional points . the width bifurcation occurring in the case of 10 levels is stronger than that with a smaller number of levels ( compare section [ v4 ] ) . the 9 states crossed by the state 10 are exchanged in the critical parameter region as can be seen from the corresponding figs . [ fig10].a , c , e , g. the results shown in figs . [ fig9 ] and [ fig10 ] follow from calculations with a toy model being symmetrical in relation to the crossing states . in the critical parameter range , the states are exchanged in all cases but 8 of them do not contribute actively to the width bifurcation when @xmath0 is imaginary . this result follows from the high symmetry of the exceptional points in relation to the crossing states , i.e. from the linear dependence of the energy of the crossing state on the parameter @xmath90 , @xmath174 . the 8 states play the role of observers similarly as discussed in a toy model with 4 states @xcite and in a realistic model with 3 states @xcite . in order to exclude the high symmetry of the exceptional points in relation to nearby states we show in fig . [ fig11 ] the results of another version of the toy model . instead of @xmath174 in figs . [ fig9 ] and [ fig10 ] , we use @xmath165 in fig . [ fig11 ] . the width @xmath47 of one of the states becomes large in the critical parameter range when @xmath0 is imaginary or complex ( figs . [ fig11].b and d , respectively ) , while the widths of all the other states are much smaller . this result reflects the situation observed in many realistic cases ( see e.g. the review @xcite and also @xcite ) . a separation of a short - lived state from the other ones does not appear when @xmath0 is real ( fig . [ fig11].f ) . in sections [ 2 ] to [ v10 ] , we showed results obtained numerically for the eigenvalues @xmath175 and eigenfunctions @xmath19 by using the matrix ( [ form1 ] ) ( or ( [ int2 ] ) ) with 2 , 4 and 10 levels , respectively . only the energies @xmath31 of the states are varied as a function of a certain parameter @xmath90 . the widths @xmath32 are assumed to be constant , the interaction @xmath94 between the levels @xmath33 and @xmath176 is either zero or simulated by the gaussian distribution ( [ int7 ] ) where @xmath0 is the same for all states ( and also independent of @xmath90 ) , and the self - energy terms are considered to be included into the diagonal matrix elements . most interesting is the one - channel case for different @xmath23 , e.g. the matrix ( [ int2 ] ) for @xmath41 . this toy model describes well the generic features of open quantum systems . the @xmath94 are complex . they stand for the interaction of the states via the environment ( continuum of scattering wavefunctions ) : re(@xmath94 ) arises from the principal value integral ( [ form11 ] ) and im(@xmath94 ) is the residuum ( [ form12 ] ) . both parts play an important role for the dynamics of the open quantum system at high level density where the resonance states overlap . the influence of re(@xmath94 ) onto eigenvalues @xmath18 and eigenfunctions @xmath19 is studied numerically exact for @xmath25 levels in an earlier paper @xcite . it causes the avoided crossing of the states which is well known for discrete as well as for narrow resonance states . equations ( [ int6a ] ) to ( [ int6d ] ) show analytically the main difference between re(@xmath94 ) and im(@xmath94 ) in the 2-level case when the widths of the two states are equal , @xmath177 . according to ( [ int6a ] ) , an exceptional point can be found only by analytical continuation into the continuum when @xmath94 is real . thus , the two states avoid crossing as it is very well known for discrete as well as for narrow resonance states @xcite . in contrast , im(@xmath94 ) causes two exceptional points according to equation ( [ int6b ] ) . most interesting is the width bifurcation arising in between the two exceptional points according to ( [ int6d ] ) . the analytical results ( [ int6a ] ) to ( [ int6d ] ) are well reproduced in our numerical calculations for the eigenvalues @xmath18 , figs . [ fig3].a , b. also the eigenfunctions @xmath19 show the two exceptional points : @xmath122 in approaching them , fig . [ fig6].a . in between the two exceptional points , the wavefunctions of the two states are strongly mixed . the mixing is 1:1 in the middle between the two singular points . beyond the critical region between the two exceptional points , the @xmath119 approach 1 when @xmath178 and 0 when @xmath155 . here , the two states may be exchanged , at most . the figures 3.a , b and 6.a represent the basic process of width bifurcation according to the analytical expressions ( [ int6c ] ) and ( [ int6d ] ) . in sections [ v4 ] , [ f4 ] and [ v10 ] of the present paper , numerical results for 4 and 10 states are shown under similar conditions , i.e. with equal or nearly equal @xmath32 for all states : figs . [ fig3].c to f and figs . [ fig6].b and c for the eigenvalues and eigenfunctions , respectively , of 4 states , and figs . [ fig9].a , b for the eigenvalues of 10 states . the development of width bifurcation as a function of increasing imaginary coupling vector @xmath0 can be seen in fig . [ fig10 ] for @xmath158 states when the single widths @xmath32 are different from one another . the uniform width bifurcation in fig . [ fig9].b appears at a smaller value of im@xmath1 than in fig . [ fig10].h due to the different values of the single @xmath32 in the last case . generally , the width bifurcation is stronger when the number of states is larger . the influence of re(@xmath94 ) onto the eigenvalue trajectories and eigenfunctions is also shown for @xmath41 and 10 ( figs . [ fig4].c , d for the eigenvalues of 4 states , fig . [ fig7].b for the eigenfunctions of 4 states and figs . [ fig9].e , f for the eigenvalues of 10 states ) . under the influence of re(@xmath94 ) , the states avoid crossing in energy and the widths do not bifurcate . when the calculation is performed with complex coupling vector ( re@xmath179 _ and _ im@xmath179 ) , the widths bifurcate with some shift of the position of the maximum relative to that of the minimum ( figs . [ fig4].a , b ) and a larger parameter range of mixed wavefunctions ( fig . [ fig7].a ) . according to the results presented in sections [ v4 ] to [ v10 ] the eigenvalue trajectories are strongly influenced by the values of the external ( fixed ) parameters when the resonances overlap and exceptional points determine the dynamics of the system . however , the eigenvalue trajectories are almost independent of one another when the degree of resonance overlapping is small . examples are figs . [ fig3].g to j and [ fig4].e to h for the eigenvalues of , respectively , two and four resonance states and the corresponding figs . [ fig6].d , e and [ fig7].c , d for the eigenfunctions of these states . these results show the strong influence of external parameters onto the dynamics of the system at high level density which is known from the study of realistic cases . an example is the enhanced transmission through microwave cavities when the formation of whispering gallery modes is supported by the manner the leads are attached to the cavity @xcite . width bifurcation is directly related to the _ alignment _ of one of the @xmath23 resonance states of the open quantum system to a decay channel ( @xmath38 ) with the consequence that it becomes short - lived while other states become _ trapped _ ( long - lived ) , i.e. more or less decoupled from the environment . mathematically , the alignment of a resonance state is possible since the eigenfunctions of a non - hermitian operator are biorthogonal and their phases ( relative to those of the eigenfunctions of the other states ) are not rigid in approaching an exceptional point , see equation ( [ eif11 ] ) . width bifurcation appears therefore in our calculations when @xmath38 and @xmath41 or 10 . this corresponds to realistic situations in which usually @xmath24 . width bifurcation does , however , not appear when @xmath41 states are coupled to @xmath133 channels ( fig . [ fig5 ] ) although the wavefunctions are strongly mixed in the critical parameter region also in this case ( fig . [ fig8 ] ) . an artifact of our model calculations is the assumption of equidistant energies @xmath31 of all but one state which are crossed by one state the energy of which depends linearly on the parameter @xmath90 . thus , the critical points are symmetrically in relation to the two neighboring states , and @xmath180 states do not contribute actively to the width bifurcation taking place in the whole system . in fig . [ fig11 ] , we show results obtained when the symmetry is somewhat disturbed ( as it is usually the case in realistic systems ) . in such a case , the width of one state separates from those of _ all _ the other states . that means , all nine states remain trapped ( decoupled ) in the whole critical parameter range . this result corresponds to results obtained theoretically as well as experimentally in different realistic systems ( see section 4 in the review @xcite ) . it should be added here , that asymptotically ( beyond the critical parameter region ) the states can be exchanged at most . this is the case , indeed , in all our calculations . all redistribution processes caused by the exceptional points take place only in the critical parameter region . this holds true for real as well as for imaginary coupling coefficients @xmath94 and can be seen also in the mixing coefficients @xmath96 which approach 1 or 0 when @xmath178 and @xmath155 , respectively , beyond the critical parameter region . the two cases with real and imaginary @xmath0 differ , however , by the length of the critical region . the results of the present paper show that exceptional points cause width bifurcation because the coupling of the states via the environment ( continuum of scattering wavefunctions ) is complex . an example are the bound states in the continuum appearing at a _ finite _ value of the coupling strength between system and environment when the interplay between internal and external interaction is taken into account , see section 4.4 of the review @xcite . of particular interest is the width bifurcation appearing at high level density . here , many exceptional points are near to one another . as shown in the present paper , width bifurcation causes , under this condition , a splitting of the system into two parts one of which exists in the short - time scale while the other one appears in the long - time scale . the two parts can not be observed together . the long - lived states occur as fluctuations in the short - time scale while the short - lived states perform a smooth background in the long - time scale . this phenomenon is known in literature from theoretical as well as experimental studies on different small quantum systems and systems equivalent to them ( see examples in the review @xcite and in @xcite ) . in many - body open quantum systems , it is called mostly _ dynamical phase transition_. it is known also in optics where it is called _ superradiant phase transition _ according to dicke @xcite . in pt - symmetric systems , the phase transition is called mostly _ pt - symmetry breaking _ @xcite . common to all these studies is that the dynamical phase transition is very robust when the necessary conditions are fulfilled , i.e. when the level density is high and the dynamics of the system is determined by exceptional points . although the existence of exceptional points is decisive ( as shown in the present paper ) , the dynamical phase transition does _ not _ appear at the parameter value which gives the position of the exceptional point itself . instead , the phase transition occurs in the neighborhood of one ( or several ) exceptional points . such a result is known from different experimental studies on concrete realistic systems . it is discussed in e.g. @xcite . the results of the present paper show that the critical parameter range is determined by the distance between ( at least ) two exceptional points . the dynamical phase transition appearing in small quantum systems at high level density , causes a new understanding of time as discussed in @xcite . time which is characteristic of the system , is inverse proportional to the widths @xmath47 and related therefore directly to the non - hermitian part of the hamilton operator . the widths do not increase limitless as shown in the present paper . instead , the system is dynamically stabilized : the lifetimes of the states of the system are increased and the system as a whole is stabilized by ejecting the short - lived state from the system at the dynamical phase transition . thus , the time characteristic of the system is bounded from below in a similar manner as the energy . beyond the dynamical phase transition , all states have lost their original spectroscopic features and the number of states is reduced . this fact is very well known in nuclear physics ( although very seldom interpreted in this manner ) . the properties of compound nuclei are described well by the _ unified theory of nuclear reactions _ developed by feshbach @xcite , which contains both the short - lived direct reaction part and the long - lived compound nuclear reaction part . however , feshbach introduces statistical assumptions in order to describe the long - lived compound nuclear states . this is in contrast to the calculations in the present paper where the long - lived states are described _ without _ any statistical assumptions . instead the long - lived states are shown by us to arise from width bifurcation causing a dynamical stabilization of the system at high level density , a phenomenon that is caused by exceptional points . many of these results seem to be counterintuitive . they allow us , however , to explain some unexpected experimental results . for example , the crossover from the mesoscopic to a universal phase for transmission in quantum dots observed experimentally @xcite can be explained qualitatively by the formation of the short - lived resonance state at the dynamical phase transition . the results of the present paper support this interpretation @xcite of the experimental results . another example is the experimental observation @xcite of non - statistical effects in nuclear reactions on middle - heavy nuclei . the data show directly the formation of different time scales in the system at high level density . other examples are discussed in @xcite . the long - lived states beyond the dynamical phase transition are strongly mixed and show chaotic features . the random matrix theory is surely applicable only after averaging over different decay channels ( as usually done ) , i.e. it is not applicable to the description of small systems coupled to one channel ( and , respectively , to two channels in the case of transmission through the system ) . a theoretical analysis of the spectra ( without any statistical assumptions ) by restricting to the results obtained from one decay channel according to the recent experimental data on compound nuclei @xcite is not performed up to now . on the basis of the results of the present paper , it can be stated today only the following . the mixing of the long - lived states ( beyond the dynamical phase transition ) is caused by complex many - body forces via the continuum of scattering wavefunctions ( simulated by the @xmath94 in our calculations ) and not by two - body forces . this may provide an explanation of the fact why compound nucleus spectra ( after averaging over different channels and beyond decay thresholds ) may be described by a gaussian orthogonal ensemble ( containing many - body forces ) , but not by a two - body random ensemble . this point should be studied in future in more detail . in order to prove the role of exceptional points in an open many - level quantum system it is highly interesting to study experimentally time symmetry breaking caused by the influence of a nearby state onto an exceptional point . theoretical studies with a symmetrical influence ( as in most calculations of the present paper ) can surely not describe the properties of realistic systems with broken time symmetry . symmetry breaking influences not only width bifurcation ( as shown in fig . [ fig11 ] ) but will prove , above all , the irreversible processes caused by exceptional points in open quantum systems . these processes are decisive , among others , for the dynamical stabilization of quantum systems and the formation of quantum chaos . finally we remark that the sensitive dependence of the dynamics of an open quantum system on the value of external parameters ( as shown in the present paper ) can be used in order to construct small systems with desired properties . this point is important for applications . a. yacoby , m. heiblum , d. mahalu , and h. shtrikman , phys . lett . * 20 * , 4047 ( 1995 ) ; r. schuster , e. buks , m. heiblum , d. mahalu , v. umansky , and h. shtrikman , nature * 385 * , 417 ( 1997 ) ; m. avinun - kalish , m. heiblum , o. zarchin , d. mahalu , and v. umansky , nature * 436 * , 529 ( 2005 ) m. goldstein and r. berkovits , new j. phys . * 9 * , 118 ( 2007 ) ; d.i . golosov , and y. gefen , new j. phys . * 9 * , 120 ( 2007 ) ; y. oreg , new j. phys . * 9 * , 122 ( 2007 ) ; p.g . silvestrov and y. imry , new j. phys . * 9 * , 125 ( 2007 ) i. rotter , contribution to the special issue _ quantum physics with non - hermitian operators : theory and experiment _ , fortschritte der physik - progress of physics * 61 * , 178 ( 2013 ) ; doi : 10.1002/prop.201200054 g.a . lvarez , e.p . danieli , p.r . levstein , and h.m . pastawski , j. chem . phys . * 124 * , 194507 ( 2006 ) ; g.a . lvarez , e.p . danieli , p.r . levstein , and h.m . pastawski , phys . rev . a * 75 * , 062116 ( 2007 ) ; h.m . pastawski , physica b * 398 * , 278 ( 2007 ) y. yoon , l. mourokh , t. morimoto , n. aoki , y. ochiai , j.l . reno , and j.p . bird , phys . lett . * 99 * , 136805 ( 2007 ) ; y. yoon , m.g . kang , t. morimoto , l. mourokh , n. aoki , j.l . reno , j.p . bird , and y. ochiai , phys . b * 79 * , 121304(r ) ( 2009 ) ; y. yoon , m.g . kang , t. morimoto , m. kida , n. aoki , j.l . reno , y. ochiai , l. mourokh , j. fransson , and j.p . bird , phys . x * 2 * , 021003 ( 2012 ) h. eleuch and i. rotter , contribution to the special issue _ quantum physics with non - hermitian operators : theory and experiment _ , fortschritte der physik - progress of physics * 61 * , 194 ( 2013 ) ; doi : 10.1002/prop.201200062 in studies by other researchers , the factor @xmath33 in ( [ eif5 ] ) does not appear . this difference is discussed in detail and compared with experimental data in the appendix of @xcite and in section 2.5 of @xcite . a. guo , g.j . salamo , d. duchesne , r. morandotti , m. volatier - ravat , v. aimez , g.a . siviloglou , and d.n . christodoulides , phys . . lett . * 103 * , 093902 ( 2009 ) ; c.e . rter , k.g . makris , r. el - ganainy , d.n . christodoulides , m. segev , and d. kip , nature physics * 6 * , 192 ( 2010 ) ; t. kottos , nature physics * 6 * , 166 ( 2010 )

as their name suggests , compact groups are small systems of several galaxies in a compact configuration on the sky . the first example was found over one hundred years ago by stephan ( 1877 ) who observed it visually using the 40-cm refractor of the observatoire de marseille . stephan s quintet is a small group of five galaxies , three of which show strong tidal distortions due to gravitational interaction . a second example was found 71 years later by seyfert ( 1948 ) from a study of harvard schmidt plates . seyfert s sextet ( figure 1 ) is one of the densest groups known , having a median projected galaxy separation of only @xmath1 kpc ( the hubble constant @xmath2 km s@xmath3 mpc@xmath3 ) . the palomar observatory sky survey ( poss ) provided a new and extensive resource for the systematic investigation of small groups of galaxies . two catalogs , the _ atlas of interacting galaxies _ ( vorontsov - velyaminov 1959 , 1975 ) and the _ atlas of peculiar galaxies _ ( arp 1966 ) , contain galaxies or galaxy groups selected on the basis of visible signs of interaction or peculiar appearance . in addition to stephan s quintet and seyfert s sextet , these include many new compact groups , including a striking chain of five galaxies , vv 172 . prior to these , shakhbazian ( 1957 ) had discovered a small dense cluster of 12 faint red galaxies that appeared so compact that they were initially mistaken for stars . over the next two decades , shakhbazian and collaborators examined over 200 poss prints covering 18% of the sky and cataloged 376 additional `` compact groups of compact galaxies '' ( shakhbazian 1973 , shakhbazian & petrosian 1974 , baier et al 1974 , petrosian 1974 , 1978 , baier & tiersch 1975 - 79 ) . apart from occasional photographic or spectroscopic observations ( eg . mirzoian et al 1975 , tiersch 1976 , massey 1977 , shakhbazian & amirkhanian 1979 , vorontsov - velyaminov et al 1980 , vorontsov - velyaminov & metlov 1980 ) these systems initially received little attention . however , interest in them is growing . although the majority seem to be small clusters , they share some of the properties , and pose some of the same questions , as do compact groups . when redshifts were measured for galaxies in the first compact groups ( burbidge & burbidge 1959 , 1961a ) , surprises were found . both stephan s quintet and seyfert s sextet contain a galaxy with a discordant redshift . it seemed unlikely that a foreground or background galaxy would appear so often projected within such compact systems ( burbidge & burbidge 1961a ) . this impression was further reinforced with the discovery of yet another discordant redshift in vv 172 ( sargent 1968 ) . are these examples of physical association between objects of widely different redshifts , as has been advocated for many years by arp ( 1987 ) ? even if the discordant galaxies are ignored , the velocity dispersions of these systems are generally higher than would be expected given their visible mass ( burbidge & burbidge 1959 , 1961b , 1961c , burbidge & sargent 1971 ) . it was argued that such groups must be unbound and disrupting ( eg . ambartsumian 1961 ) , although limber and mathews ( 1960 ) showed that the virial theorem could be satisfied for stephan s quintet , given the uncertainties in the projection factors , if the individual galaxy masses were considerably larger than those of isolated galaxies . the observations can of course also be explained if the bulk of the mass is in a non - visible form . in hindsight , this was one of the earliest indications of the possible existence of dark matter in galactic systems . a new problem emerged with the realization that bound groups would be unstable to orbital decay resulting from gravitational relaxation processes ( peebles 1971 ) . a simple calculation indicated that the dynamical - friction timescale was much shorter than the hubble time ( hickson et al 1977 ) , as was soon confirmed by numerical simulations ( carnevali et al 1981 ) . at the same time , it became increasingly clear that mergers played an important role in the evolution of many , if not all , galaxies ( press & schechter 1974 , ostriker & tremaine 1975 ) . compact groups emerged as prime locations for investigations of the dynamical evolution of galaxies . motivated by the desire for a homogeneous sample that could be subject to statistical analysis , rose ( 1977 ) , and later , hickson ( 1982 ) produced the first catalogs of compact groups having specific , quantitative , selection criteria . subsequent detailed investigation , at many wavelengths , has produced a large body of observational data for the hickson catalog . as a result , it has now become possible to address some of the outstanding questions concerning the nature of compact groups and their role in galaxy evolution . not surprisingly , new questions have been raised and new controversies have appeared . however , much progress has been made in resolving both old and new issues . this review is organized as follows : in section 2 , the definition of a compact group is discussed , along with methods of identification and surveys that have been made . in sections 3 - 5 , observed properties of these systems are summarized and discussed . sections 6 and 7 focus mainly on interpretation of the observations , and on implications of these results . all work on compact groups of galaxies can not possibly be discussed in this short paper , although an attempt is made to touch upon most current topics . other recent reviews of compact groups and closely related subjects include those by white ( 1990 ) , hickson ( 1990 , 1997 ) , whitmore ( 1992 ) , kiseleva & orlov ( 1993 ) , sulentic ( 1993 ) , and mamon ( 1995 ) . by `` compact group '' , we mean a small , relatively isolated , system of typically four or five galaxies in close proximity to one another . such groups do not necessarily form a distinct class , but may instead be extreme examples of systems having a range of galaxy density and population . because of this , the properties of the groups in any particular sample may be strongly influenced by the criteria used to define the sample . the early surveys used qualitative criteria that , while successful in finding many interesting individual objects , do not easily allow one to draw broad conclusions about the groups as a whole . thus , the focus in recent years has been on samples selected using specific , quantitative , criteria . these criteria define the minimum number and magnitude range of the galaxies , and also consider the galaxy spatial distribution . the use of quantitative selection criteria was pioneered by rose ( 1977 ) who searched for groups that have three or more galaxies that are brighter than a limiting magnitude of 17.5 , and that have a projected surface density enhancement of a factor of 1000 compared to the surrounding background galaxy density . searching an area of 7.5% of the sky , he found 170 triplets , 33 quartets , and 2 quintets . unfortunately , the sample received little follow - up study . sulentic ( 1983 ) re - examined the 35 rose groups which contain four or more galaxies and found that only a third actually satisfied the selection criteria . this is testimony to the difficulty of visual searches . a more fundamental problem is that the fixed magnitude limit in the selection criteria makes the sample susceptible to strong distance - dependent biases . in an attempt to reduce such effects , hickson ( 1982 , 1993 , 1994 ) adopted a relative magnitude criterion , selecting systems of four or more galaxies whose magnitudes differ by less than 3.0 . a distance independent ( to first order ) compactness criterion was employed : @xmath4 , where @xmath5 is the mean surface brightness of the group calculated by distributing the flux of the member galaxies over the smallest circular area containing their geometric centers . to avoid including the cores of rich clusters , an isolation criterion was necessary so as to reject the group if a non - member galaxy , not more than 3 mag fainter than the brightest member , occurred within three radii of the center of the circle . ( a non - member galaxy is a galaxy which if included in the group would cause the group to fail one or more of the selection criteria . ) from a search of 67% of the sky ( all the poss prints ) , and using magnitudes estimated from the poss red prints , exactly 100 groups were found satisfying these criteria ( hereafter hcg s ) . as the hcg sample is now the most widely studied , it is important to examine the biases introduced by the criteria , and by the visual search procedure . any sample selected on the basis of surface density will suffer from geometric and kinematic biases . the former occurs because non - spherical systems will be preferentially selected if they are oriented to present a smaller cross - sectional area ( eg . prolate systems pointed towards us ) ; the latter because we will preferentially select systems that , owing to galaxy orbital motion , are momentarily in a more compact state ( transient compact configurations ) . thus , a compact group might result from a chance alignment or transient configuration within a loose group ( mamon 1986 ) . this question will be considered in more detail in section 6.2 . other biases arise from the subjective nature of the search procedure . the original catalog contains a few mis - identifications , such as compact galaxies mistaken for stars , and marginal violations of the isolation criteria . in addition , when photometry was obtained for the galaxies in the catalog ( hickson et al 1989 ) , it was found that some groups would not satisfy the selection criteria if photometric magnitudes are used . attempts to rectify these problems have been made by hickson et al ( 1989 ) and sulentic ( 1997 ) . however , it should be emphasized that changes , such as using photometric magnitudes in the selection criteria , are not corrections to the catalog , but are actually the imposition of _ additional _ a postiori selection criteria . the resulting subsample is by no means complete because the new criteria are applied only to the visually - selected hcg catalog and not to the entire sky . because of the difficulty of identifying faint groups , the hcg catalog starts to become significantly incomplete at an integrated magnitude of about 13 ( hickson et al 1989 , sulentic & rabaa 1994 ) . a more subtle effect results from the difficulty of recognizing low - surface - brightness groups . because of this , the catalog also becomes incomplete at surface brightnesses fainter than 24 ( hickson 1982 ) . yet another effect is that groups may be more noticeable if the magnitude spread of their members is small . thus , the catalog may also be incomplete for magnitude intervals greater than about 1.5 ( prandoni et al 1994 ) . these effects are of critical importance in statistical analyses of the sample . one immediate conclusion is that the actual number of groups which satisfy the selection criteria may be considerably larger than the number found by a subjective search . it has recently become feasible to find compact groups by automated techniques . mamon ( 1989 ) used a computer to search tully s ( 1987 ) catalog of nearby galaxies and identified one new compact group , satisfying hickson s criteria , in the virgo cluster . prandoni et al ( 1994 ) applied similar criteria to digital scans of @xmath6 deg@xmath7 around the southern galactic pole and detected 59 new southern compact groups ( scgs ) . observations are presently underway to obtain accurate photometry and redshifts for this sample ( iovino private communication ) . the digitized palomar sky survey ii also offers new opportunities for the identification of compact groups ( de carvalho & djorgovski 1995 ) . an alternative approach is to identify groups of galaxies from redshift information , as was first done by humason et al ( 1956 ) . with the advent of large - scale redshift surveys , it has become possible to identify a reasonably - large sample of compact groups in this way . barton et al ( 1996 ) have compiled a catalog of 89 redshift - selected compact groups ( rscgs ) found by means of a friends - of - friends algorithm applied to a complete magnitude - limited redshift survey . galaxies having projected separations of @xmath8 kpc or less and line - of - sight velocity differences of 1000 km s@xmath3 or less are connected and the sets of connected galaxies constitute the groups . the numerical values were chosen to best match the characteristics of the hcg sample , and indeed , many of those rscgs that have at least four members are also hcgs . there are some significant differences , however : because foreground and background galaxies are automatically eliminated by the velocity selection criteria , this technique is more effective at finding groups in regions of higher galaxy density , which would fail the hcg isolation criterion . this criterion requires that the distance to the nearest neighbor be at least as large as the diameter of the group . the rscg criteria , on the other hand , require only that the nearest - neighbor distance be larger than the threshold distance ( @xmath8 kpc for the rscgs ) , which may be considerably smaller . it will therefore allow the inclusion of groups that are physically less isolated ( and therefore less physically distinct ) than would the hcg criterion . one would also expect that the numbers of groups found in a given volume by the less - restrictive rscg criteria to be larger than by the hcg criteria , as seems to be the case . while the redshift - selection method compliments the hcg angular - selection technique , one should keep in mind that it also selects groups according to apparent ( projected ) density the velocity information serves only to reject interlopers . thus it will be subject to some of the biases discussed above . also , because the galaxy sample used is magnitude - limited , rather than volume - limited , there will be redshift - dependent biases in the rscgs . however , the well - defined selection criteria and the completeness of the sample , should allow a quantitative determination of the effects of this bias . the space distribution and environment of compact groups provides important clues to their nature . the median redshift of the hcgs is @xmath9 , placing most of them at distances well beyond the virgo cluster ( hickson et al 1992 ) . a cursory inspection reveals that they are fairly uniformly distributed and show no preference for rich clusters . this is at least partly due to the isolation criterion . however , galaxies in rich clusters have rather different kinematical and morphological properties than do those in compact groups , so one might justifiably argue that small clumps of galaxies within clusters are not compact groups . a natural question is whether or not compact groups are associated with loose groups . rood & struble ( 1994 ) observed that 70% of the hcgs are located within the bounds of cataloged loose groups and clusters . studies of the distribution of galaxies in redshift space ( vennik et al 1993 , ramella et al 1994 , sakai et al 1994 , garcia 1995 , barton et al 1996 ) indicate that compact sub - condensations do occur within loose groups and filaments . vennik et al ( 1993 ) and ramella et al ( 1994 ) find that most hcgs are indeed associated with loose groups . while the above studies show that compact groups trace large - scale structure , it is also clear that they prefer low - density environments . sulentic ( 1987 ) , rood & williams ( 1989 ) , kindl ( 1990 ) , and palumbo et al ( 1995 ) have examined the surface density of galaxies surrounding the groups . they generally agree that about two - thirds groups show no statistically significant excess of nearby neighbors . this is not inconsistent with the redshift - space results because most of the hcg associations identified by ramella et al ( 1996 ) contain fewer than 5 excess galaxies . thus while compact groups are associated with loose groups and filaments , these tend to be low - density and sparsely - populated systems . are the galaxies in compact groups in any way distinct from those in their immediate environments ? rood & williams ( 1989 ) and kindl ( 1990 ) both found that compact groups , including those in rich environments , contain a significantly smaller fraction of late - type ( spiral and irregular ) galaxies than do their neighborhoods . this result is of particular importance to the question of the physical nature of compact groups , discussed below in section 6 . in addition , many independent studies have found significant differences between galaxies in compact groups and those in other environments . these are examined in section 5 . the first studies of individual compact groups ( burbidge & burbidge 1959 , 1961b , 1961c , burbidge & sargent 1971 , rose & graham 1979 , kirshner & malumuth 1980 ) indicated short dynamical times and mass - to - light ratios intermediate between those of galaxies and rich clusters . however , because of the small number of galaxies , estimates of the space velocities and physical separations of galaxies in individual groups are highly uncertain . meaningful dynamical conclusions about systems containing only four or five galaxies requires statistical analysis of large homogeneous samples . by 1992 , velocities had been measured for almost all 462 galaxies in the hcg catalog ( hickson et al 1992 ) . the distribution of galaxy velocities , relative to the median of each group , is shown in figure 2 . it can be seen that the majority of velocities fall within a roughly gaussian distribution of standard deviation @xmath10 km s@xmath3 ( the sharp peak at zero velocity results from the use of the median ) . this characteristic velocity is quite similar to velocity dispersions found in loose groups , and much smaller than typical velocity dispersions in rich clusters . in addition to the gaussian core , a flat component is seen in the velocity distribution of figure 2 . this is expected as some galaxies which are not physically related to the group will appear projected on the group by chance . this component contains about 25% of the total number of galaxies . whether of not chance projection can account for such a large number of `` discordant '' galaxies is still a matter of some debate , and is discussed further in section 6.1 . for a system of characteristic linear size @xmath11 and internal velocity @xmath12 , a characteristic dynamical time is @xmath13 . a characteristic mass density is @xmath14 , from which one can estimate the total mass within the region occupied by the galaxies . after removing galaxies whose velocities differ from the group median by more that 1000 km s@xmath3 , hickson et al ( 1992 ) found that @xmath15 and obtained a mean mass - to - light ratio of @xmath16 ( solar units ) for the hcgs . similar values were found for several shakhbazian groups ( tikhonov 1986 , amirkhanian & egikian 1987 , amirkhanian 1989 , lynds et al 1990 , amirkhanian et al 1991 ) . since the mass - to - light ratios of individual hcg galaxies is on the order of @xmath17 ( rubin et al 1991 ) , the galaxies appear to contain only @xmath18% of the total mass . the spatial distribution and luminosities of the member galaxies provides further clues to the nature of compact groups . if they are primarily projections or transient configurations , the luminosity function should be the same as that of the parent systems , and the spatial distribution of the galaxies should be consistent with a random distribution . if they are bound physical systems , the luminosity function and spatial distribution might show features that reflect the origin or subsequent evolution of compact groups . the shapes of compact groups were first investigated by arp ( 1973 ) who concluded that galaxy `` chains '' were unusually predominant . however , rose ( 1977 ) determined that the ellipticities of his groups were consistent with a random distribution of galaxies . using the larger hcg sample , hickson et al ( 1984 ) and malykh & orlov ( 1986 ) reached the same conclusion as arp the groups are typically more elongated than would be a random distribution of galaxies . an immediate consequence of this result is that compact groups can not easily be explained as random projections or chance crossings , as this would largely erase any inherent ellipticity of a parent loose group . from static simulations , hickson et al ( 1984 ) concluded that the observed ellipticities are best matched by three - dimensional shapes that are intrinsically prolate . the same result was found by oleak et al ( 1995 ) in a recent study of the shapes of 95 shakhbazian compact groups . these conclusions , however , are not unique . hickson et al ( 1984 ) also found the shapes to be consistent with those seen in dynamical simulations of compact groups seen in projection as subgroups within loose groups . in addition , one must always be concerned about possible selection biases . it may be that highly elongated groups ( such as vv 172 ) are more easily noticed in visual searches . it will be interesting to see if these results are confirmed by studies of groups found by automated searches . if the intrinsic shapes of compact groups are related to their formation process , one might expect to see a relationship between the orientation angle of a group and the environment . palumbo et al ( 1993 ) examined the environments of the hcgs , and found that the orientations of the major axes of the groups were consistent with an isotropic distribution . if compact groups are not simply projection effects , they might be expected to show a centrally - concentrated surface density profile , as is seen in clusters of galaxies . although the number of galaxies in individual compact groups is small , with a large sample it is possible to estimate a mean profile . by scaling and superimposing the hcgs , hickson et al ( 1984 ) found evidence for central concentration . mendes de oliveira & girard ( 1994 ) , using a similar analysis , concluded that the mean surface density profile is consistent with a king ( 1962 ) model with typical core radius of 15@xmath19 kpc . most recently , montoya et al ( 1996 ) have analyzed the profiles of the 42 hcg quartets which have accordant redshifts . their technique uses the distribution of projected pair separations and thus avoids assumptions about the location of the group center . they find a smaller core radius ( @xmath20 kpc for a king model ) . the fact that montoya et al ( 1996 ) find a consistent density profile for all groups , without any scaling , is particularly interesting . this would not be expected if most groups are chance alignments within loose groups . it also implies that compact groups have a unique scale , which seems counter to the concept of hierarchical clustering . montoya et al ( 1996 ) suggest that this arises as a result of a minimum mass density and velocity dispersion that is required for the groups to be virialized ( mamon 1994 ) . there have been several studies of the morphological types of galaxies in compact groups ( hickson 1982 , williams & rood 1987 , sulentic 1987 , hickson et al 1988b ) . most studies agree that the fraction @xmath21 of late type galaxies is significantly less in compact groups than in the field . hickson et al ( 1988b ) find @xmath22 for the hcgs ; prandoni et al ( 1994 ) obtain @xmath23 for the scgs . both these values are substantially lower than those found for field galaxy samples ( @xmath24 , gisler 1980 , nilson 1973 ) . also well established is morphological type concordance , observed in both the hcgs and scgs ( sulentic 1987 , hickson at al 1988b , prandoni et al 1994 ) . a given compact group is more likely to contain galaxies of a similar type ( early or late ) than would be expected for a random distribution . white ( 1990 ) has pointed out that such concordance could result from a correlation of morphological type with some other property of the group . the strongest such correlation found to date is between morphological type and velocity dispersion ( hickson et al 1988b ) . as figure 3 shows , groups with higher velocity dispersions contain fewer late - type ( gas - rich ) galaxies . they also tend to be more luminous . the importance of velocity dispersion , in addition to local density , on the galaxy morphology had previously been emphasized by de souza et al ( 1982 ) . a crucial clue is that the morphology - density relation seen in clusters and loose groups ( dressler 1980 , postman & geller 1984 , whitmore & gilmore 1992 ) is not the dominant correlation in compact groups ( hickson et al 1988b ) , although some effect is present ( mamon 1986 ) . this suggests that the velocity dispersion is more fundamental , ie of greater physical relevance to the formation and evolution of galaxies in compact groups , than is apparent physical density . there is much evidence that interaction is occurring in a large fraction of galaxies in compact groups . the strongest direct support comes from kinematical studies . rubin et al ( 1991 ) found that two thirds of the 32 hcg spiral galaxies that they observed have peculiar rotation curves . these show asymmetry , irregularity and in some cases extreme distortion , characteristic of strong gravitational interaction . this result has recently been challenged by mendes de oliveira et al ( preprint ) who obtained h@xmath25 velocity maps for 26 hcg spiral galaxies and found that only 1/3 showed abnormal rotation curves . they suggest that the difference is due to the more - complete spatial sampling of their data . in their study , rubin et al ( 1991 ) observed 12 hcg elliptical galaxies and detected nuclear emission in 11 of them . this high fraction suggests that interactions and mergers may be supplying gas to these galaxies . this idea received independent support from radio observations in which neutral hydrogen emission was detected in three compact groups which contain only elliptical galaxies ( huchtmeier 1994 ) . zepf & whitmore ( 1993 ) found that elliptical galaxies in compact groups tend to have lower internal velocity dispersions than do ellipticals in other environments having the same effective radii , absolute magnitudes and colors . they therefore do not lie on the fundamental plane defined by other elliptical galaxies . this discrepancy correlates with isophote shape in that those galaxies that have `` disky '' or irregular isophotes tend to have lower velocity dispersion . both zepf & whitmore ( 1993 ) and bettoni & fasano ( 1993 , 1995 , 1996 , fasano & bettoni 1994 ) report that hcg elliptical galaxies are less likely to have `` boxy '' isophotes , and more likely to have irregular isophotes . such effects are consistent with results of simulations of tidal encounters ( balcells & quinn 1990 ) . the luminosity function of compact groups was first estimated by heiligman & turner ( 1980 ) . they examined a sample consisting of stephan s quintet , seyfert s sextet and eight more compact groups from the arp and vorontsov - velyaminov catalogs , and concluded that compact groups contain relatively fewer faint galaxies than does a comparable field galaxy sample . analysis of the relative luminosities within individual hcgs ( hickson 1982 ) , and studies of several shakhbazian groups ( kodaira et al 1991 ) , showed a similar effect , although tikhonov ( 1987 ) found a luminosity function similar to that of field and cluster galaxies . the larger hcg sample allows the question of the galaxy content of compact groups to be addressed with greater certainty . the standard technique for determination of the luminosity function weights each galaxy by @xmath26 , where @xmath12 is the volume of the smallest geocentric sphere containing the group , and @xmath27 is the volume of the largest such sphere within which the group could have been detected . using this approach sulentic and rabaa ( 1994 ) obtained a luminosity function for hcg galaxies similar to that of field galaxies . however , mendes de oliveira and hickson ( 1991 ) argued that the standard calculation does not address the selection effects of the hcg sample . for example , the luminosity range within an individual group is limited by the 3-mag range of the selection criteria . because of this , fainter galaxies within compact groups are not included in the catalog . in order to account for such biases they used a modeling technique in which galaxies were drawn from a trial luminosity function and assigned to groups . redshifts were given to each group according to the observed distribution , and groups that failed to meet the hcg selection criteria were rejected . the luminosity distribution of the resulting galaxy sample was then compared to the observations and the process repeated with different trial luminosity functions . their best - fit luminosity function is deficient in faint galaxies , although a normal field - galaxy luminosity function is not excluded . to avoid the selection problem , ribeiro et al ( 1994 ) obtained deeper photometry for a subsample of the hcgs in order to include the fainter galaxies explicitly . since redshifts are not known for these galaxies , a correction for background contamination was made statistically . the luminosity function that they obtained is similar to that of field galaxies . most recently , the luminosity function for the rscgs has been computed by barton et al ( 1996 ) . they find it to be mildly inconsistent with that of field - galaxies , in the same sense as that of mendes de oliveira & hickson ( 1991 ) for the hcgs . figure 4 summarizes these estimates of the luminosity function of compact galaxies in terms of the schechter ( 1976 ) parameters @xmath28 and @xmath29 . how do we interpret these apparently conflicting results ? prandoni et al ( 1994 ) have argued that the hcg catalog is biased toward groups with a small magnitude range @xmath30 , because the scgs have a larger fraction of high @xmath30 groups . however , it is not known what fraction of such groups are physically real , as few redshifts have yet been obtained . such a bias could affect the luminosity function of mendes de oliveira & hickson ( 1991 ) , particularly at the faint end , but it is not evident that the bias is sufficient to account for the apparent faint - galaxy deficit . on the other hand , the small sample used by ribeiro et al ( 1994 ) may not be representative of compact groups in general . hickson ( 1997 ) points out that the ribeiro et al ( 1994 ) sample has a spiral fraction of 0.60 , substantially higher than that of the whole hcg catalogue , and contains 7 of the 16 hcgs found to be in a high density environment by palumbo et al ( 1995 ) . this suggests that the their sample has more than the usual amount of field galaxy contamination . de carvalho et al ( 1994 ) , note that the faint galaxies form a more extended distribution than do the brighter galaxies . thus they may be a dynamically distinct component , or simply unrelated field galaxies . finally , the redshift - selected rscg sample also shows mild evidence for a faint galaxy deficiency . while the faint end of the lf in compact groups appears to be depleted , there is evidence that the bright end may be enhanced . limber and matthews ( 1960 ) were the first to remark that `` the members of stephan s quintet are to be classed among the brightest of galaxies '' . it is possible that this may be in part due to interaction - induced star formation , at least for the spiral galaxies . on the other hand mendes de oliveira & hickson ( 1991 ) compared their luminosity function of elliptical galaxies in compact groups with those in the virgo and coma clusters ( as reported by sandage et al 1985 and thompson & gregory 1980 ) and found that that the compact - group ellipticals have a luminosity enhancement of more than 1 mag compared to cluster ellipticals . sulentic and rabaa ( 1994 ) find a similar enhancement in their morphological - type - specific luminosity function . this suggests that compact group elliptical galaxies may have a unique formation mechanism . from the luminosity function , one can estimate the contribution of compact groups of galaxies @xmath31 to the total galaxian luminosity density @xmath32 . mendes de oliveira & hickson ( 1991 ) obtained a ratio of @xmath33 . applying the same analysis to the luminosity function of ribeiro et al ( 1994 ) gives a ratio of 3.3% . for the rscgs , the figure is comparable : for groups of four or more galaxies , barton et al ( 1996 ) obtain a compact group abundance of @xmath34 mpc@xmath3 which leads to a luminosity density ratio of approximately 3% . these are surprisingly high figures considering the short dynamical times of most compact groups . evidence is accumulating that tidal interactions play an important role in triggering starburst activity in galaxies ( eg . maccagni et al 1990 , campos - aguilar & moles 1991 , kormendy & sanders 1992 , sanders & mirabel 1996 ) . compact groups , with their high galaxy density and evident signs of galaxy interaction should be ideal systems in which to study such effects . many hcgs do in fact contain galaxies showing starbursts or harboring active galactic nuclei ( agn ) . for example , hcg 16 is found to contain a seyfert 2 galaxy , two liners , and three starburst galaxies ( ribeiro et al 1996 ) . hcg 31 contains five galaxies showing signs of recent starburst activity ( rubin et al 1990 , iglesias - pramo & vlchez preprint ) . seyfert galaxies are also found in hcg 77 , 92 , 93 and 96 . the general degree of star formation activity in compact group galaxies can be determined from infrared observations . to date , studies have been based primarily on data from the iras satellite . hickson et al ( 1989 ) found sources in 40 hcg from a search of the point source catalog . they concluded that the ratio of far - infrared - to - optical luminosity is greater by about a factor of two in compact group galaxies , compared to that of isolated galaxies . this result was disputed by sulentic and de mello rabaa ( 1993 ) who argued that the low spatial resolution of the data made the assignment of infrared flux to individual galaxies ambiguous . they concluded that redistribution of the flux could result in little or no infrared enhancement , a conclusion echoed by venugopal ( 1995 ) . however , in cases of doubt , hickson et al ( 1989 ) identified the infrared galaxy on the basis of radio emission . the well - known correlation between infrared and radio continuum emission makes it unlikely that the results are much in error . analysis of improved data ( eg . allam et al 1996 ) should soon resolve questions about the identifications and infrared fluxes . the resolution problem can be avoided by considering the infrared colors of the sources instead of the infrared / optical ratio . zepf ( 1993 ) compared the ratio of 60 @xmath35 m to 100 @xmath35 m fluxes of compact group galaxies with those of isolated galaxies and also with those of galaxies believed to be currently merging . he found that the compact group sample was significantly different from both other samples , and estimated that about 1/3 of the compact group galaxies had warm colors ( larger 60/100 @xmath35 m ratios ) similar to those of merging galaxies . another approach to interpreting the infrared results was taken by menon ( 1991 ) who emphasized that the strong correlation between radio and infrared radiation indicates that these likely originate from a common region . in compact group spirals the radio emission is primarily nuclear whereas in isolated spirals it originates in the disk . if this is also true for the infrared flux , there must be an enhancement of the infrared / optical ratio , in the nuclear region , of more than an order of magnitude . this idea is supported by recent millimeter - wavelength observations ( menon et al 1996 ) in which co emission is detected in 55 of 70 iras - selected hcg galaxies . the inferred ratio of infrared luminosity - to - h@xmath36 mass showed an enhancement which correlates with the projected nearest - neighbor distance . further clues are provided by radio continuum studies . nonthermal emission from spiral galaxies can arise from both disk and nuclear sources . disk emission is predominantly due to supernova remnants and is thus related to the star formation rate . nuclear emission can arise both from star formation and from an active nucleus . menon ( 1995 ) observed 133 spiral galaxies in 68 hcg , and found that overall they typically show less continuum emission than those in isolated environments , which is consistent with the neutral hydrogen observations . however , when considering the nuclear regions alone , the radio emission is found to be an order of magnitude higher compared to isolated spirals . the implication is that star formation and/or agn activity is substantially enhanced in the nuclear regions of many compact group spiral galaxies . this is generally consistent with a picture in which galaxy interactions remove gas from the outer regions of galaxies , while simultaneously allowing gas to flow inwards toward the nucleus , resulting in enhanced star formation in the nuclear region , and possibly fueling an active nucleus . although there is a clear example of tidal interaction stimulating disk radio emission in at least one compact group ( menon 1995a ) , statistical evidence for a link between interactions and radio emission in compact groups is only now accumulating . if interactions are stimulating nuclear radio emission , one would expect the radio luminosity to be correlated with some index describing the degree of interaction such as the projected distance to the nearest neighbor . evidence in support of this was found by vettolani & gregorini ( 1988 ) who observed that early - type galaxies which have a high ratio of radio - to - optical emission show an excess of nearby neighbors . a similar effect was observed by malumian ( 1996 ) for spiral galaxies in groups . examining compact group galaxies , menon ( 1992 ) found that elliptical and s0 galaxies detected at a wavelength of 20 cm had closer neighbors than the undetected galaxies . the effect was not found for spiral galaxies , but if one considers only the detected galaxies , there is a significant correlation between radio - to - optical luminosity and nearest neighbor distance for both early and late type galaxies ( tk menon , private communication ) . continuum radio emission has also been detected in a number of hcg elliptical galaxies . unlike those found in cluster ellipticals , the radio sources are low - luminosity and compact . where spectral indices are available , they indicate that the radio emission arises from an agn rather than from starburst activity ( tk menon , private communication ) . in the hcg sample , there is a significant preference for radio - loud elliptical galaxies to be first - ranked optically ( menon & hickson 1985 , menon 1992 ) . the probability of radio emission does not correlate with absolute luminosity , but instead correlates with _ relative _ luminosity within a group . spiral hcg galaxies do not show this effect . although the tendency of radio galaxies in rich clusters to be first - ranked has been known for many years , it is surprising to find a similar effect in small groups , where the number of galaxies and luminosity range is small , the gravitational potential well is much less clearly defined , and it is unlikely that any individual galaxy holds a central location . the effect of optical rank on radio emission had been previously noted in other small groups by tovmasian et al ( 1980 ) although these authors made no distinction between elliptical and spiral galaxies . it is difficult to imagine any explanation for this result in which the compact group is not a true physical system . it would appear that , regardless of absolute luminosity , only the first - ranked ( presumably the most massive in the group ) elliptical galaxy can develop a radio source . stars stripped from galaxies by tidal forces should accumulate in the potential well of the group . and may be detectable as diffuse light . in an early photographic study , rose ( 1979 ) found no evidence for diffuse light in his groups , and was lead to the conclusion that most of his groups must be transient configurations . however , bergvall et al ( 1981 ) were successful in detecting ionized gas and a common halo around a compact quartet of interacting early - type galaxies , and evidence for a common halo in vv 172 was reported by sulentic & lorre ( 1983 ) . diffuse light can clearly be seen in hcg 94 , and has been found in hcg 55 ( sulentic 1987 ) , but pildis et al ( 1995b ) did not detect any in seven other compact groups . analysis by mamon ( 1986 ) indicated that while the expected diffuse light should be detectable with modern techniques , it would generally be very faint . estimates of the total amount of diffuse light in the detected groups are rather uncertain as they depend sensitively on subtraction of the galactic light , and the sky background . deeper photometry and improved image processing techniques may yet reveal diffuse light in other compact groups ( sulentic 1997 ) . the mass and distribution of cool galactic and intergalactic gas , can be obtained from observations of the 21-cm line of neutral hydrogen . the first such study of a large sample of compact groups is that of williams & rood ( 1987 ) who found a median hi mass of @xmath37 . they concluded that compact groups are typically deficient in neutral hydrogen by about a factor of two compared to loose groups . this effect is consistent with similar deficit in continuum radio emission seen in the disks of compact group spiral galaxies ( menon 1995 ) , and suggests that interactions in compact groups has removed much of the gas from the galaxies . simulations suggest that in addition to an outflow of gas , inflow also occurs which may fuel nuclear star - formation , as suggested by the strongly - enhanced radio emission seen in the nuclear regions of compact group spiral galaxies ( menon 1995 ) . high - resolution studies of individual groups ( williams & van gorkom 1988 , williams et al 1991 ) showed clearly that the gas is not confined to the galaxies . in two of three groups studied , the radio emission originates from a common envelope surrounding the group and in the third group there are signs of tidal distortion . these results strongly indicate that at least these compact groups are physically dense systems and not chance alignments or transient configurations in loose groups . they also show that many groups have evolved to the point that gas contained within individual galaxies has been distributed throughout the group . in contrast to the hi results , initial co - line observations of 15 compact - group galaxies ( boselli et al 1996 ) indicated a normalized molecular gas content similar to that of isolated spiral galaxies . however , this result is based on normalizing the flux by the optical area of the galaxy , rather than by the infrared luminosity , and may be biased by the relatively small sizes of compact - group galaxies . further co studies , currently in progress , should soon settle this question . x - ray observations of hot gas in clusters of galaxies can reveal the amount , distribution , temperature and metallicity of the gas , as well as the relative amount and distribution of the total gravitating mass . temperature , metallicity ( fraction of solar abundance ) , and bolometric luminosities are estimated by fitting a spectral model , such as that of raymond & smith ( 1977 ) to the data . x - ray emission from compact `` poor clusters '' was first reported by schwartz et al ( 1980 ) , who concluded that their x - ray properties were similar to those of rich clusters . using the einstein observatory , bahcall et al ( 1984 ) first detected x - ray emission from stephan s quintet . the x - ray map revealed that the emission is diffuse and not centered on individual galaxies . however , we now know that most of this emission is associated with a shock front rather than gas trapped in the group potential well ( sulentic et al 1995 ) . although several other groups ( bahcall et al 1984 , biermann & kronberg 1984 ) were detected by the einstein observatory , further progress required the improved sensitivity of the rosat x - ray observatory . pointed rosat observations revealed massive hydrogen envelopes surrounding the ngc 2300 group , a bright elliptical - spiral pair with two fainter members ( mulchaey et al 1993 ) , and hcg 62 , a compact quartet of early - type galaxies ( ponman & bertram 1993 ) , and showed that these systems are dominated by dark matter . subsequent investigations detected x - rays from 18 additional compact groups , either from individual galaxies , or from diffuse gas ( ebeling et al 1994 , pildis et al 1995a , sarraco & ciliegi 1995 , sulentic et al 1995 ) . these studies showed that the physical properties of individual systems span a wide range , but that the ratio of gas - to - stellar mass is significantly lower than in rich clusters . moreover , the detected compact groups all contained a majority of early - type galaxies . no spiral - rich groups were detected ( although mulchaey et al 1996b pointed out that they might be found from qso absorption spectra ) . this result is consistent with the fact that x - ray - selected groups ( henry et al 1995 ) and loose groups ( mulchaey et al 1996a ) tend to be spiral poor , and led to the suggestion that spiral - rich compact groups might not be physically dense systems at all . the most extensive x - ray study of compact groups to date is that of ponman et al ( 1996 ) . these authors combined pointed and survey - mode observations of a complete sample of 85 hcgs and detected diffuse emission in 22 groups . they conclude that , when the detection limits are considered , diffuse emission is present in at least 75% of the systems . significantly , they detected diffuse emission in several spiral - rich groups . in these the surface brightness is lower and the x - ray emission has a lower characteristic temperature , as would be expected given the lower velocity dispersions of spiral - rich compact groups . the diffuse x - ray luminosity was found to correlate with temperature , velocity dispersion , and spiral fraction , but not with optical luminosity . the last result suggests that the gas is mostly primordial and not derived from the galaxies . the correlations with temperature and velocity dispersion appear to be consistent with a single relation for clusters and groups ( figure 5 ) . the total mass in compact groups typically exceeds the stellar and gas mass by an order of magnitude . pildis et al ( 1995 ) derived baryon fractions of 12 - 19% . davis et al ( 1996 ) obtains 10 - 16% for the ngc 2300 group . these are comparable to the fractions found for poor clusters ( dellantonio et al ( 1995 ) and are about half the typical values found for rich clusters . however , the derived baryon fraction depends sensitively on the radius within which it is measured and on the assumed background level ( henriksen & mamon 1994 ) . both the total mass , which is dominated by dark matter , and the gas mass continue to increase with radius . consequently , both the baryon fraction and the gas fraction are poorly determined . the contribution of compact groups to the x - ray luminosity function has been estimated by pbeb , who find that on the order of 4% of the total luminosity in the range @xmath38 erg s@xmath3 comes from hcgs . this is higher than the contribution of hcg galaxies to the local optical luminosity density estimated at 0.8% by mendes de oliveira & hickson ( 1991 ) , but it is comparable to the value found by ribeiro et al ( 1994 ) . the metallicity inferred for the x - ray - emitting gas in compact groups is relatively low . pbeb obtain a mean metallicity of 0.18 solar , compared to the value , 0.3 - 0.4 solar , found in rich clusters . this is comparable to the low value ( @xmath39 solar ) found for the ngc 2300 group ( davis et al 1996 ) . these figures suggest that the gas is largely primordial , a result supported by the absence of a correlation between x - ray and optical luminosity . however , given the limited spectral resolution of rosat , these low metalicities can not yet be considered secure . the higher spectral resolution and sensitivity of the asca satellite should provide more definitive results . recent observations of hcg 51 and the ngc 5044 group found metal abundances comparable to those of clusters ( fukazawa et al 1996 ) . the nature of the discordant redshift members of compact groups has been a subject of debate for many years ( eg . burbidge & burbidge 1961a , burbidge & sargent 1971 , nottale & moles 1978 , sulentic 1983 ) . if the frequency of discordant galaxies is inconsistent with the statistics of chance projection , it might signify the need for new physical theories ( arp 1987 ) , or for gravitational amplification of background galaxies ( hammer & nottale 1986 ) . initial estimates of the chance probability of finding discordant galaxies in groups like stephan s quintet , seyfert s sextet and vv 172 , were very small ( burbidge & sargent 1971 ) . however , such probabilities were recognized to be difficult to determine reliably , because the a priori probability of _ any _ particular configuration of galaxies is also very small ( burbidge & sargent 1971 ) . only with a well - defined sample of groups and a complete characterization of selection effects , can meaningful estimates of the probabilities be made . the explicit selection criteria of the hcg catalog in principal makes this sample suitable for a quantitative statistical investigation of the discordant - redshift question . sulentic ( 1987 ) first concluded that the number of discordant redshifts in the catalog is too large to explain by chance . on the other hand , hickson et al 1988a and mendes de oliveira ( 1995 ) , applying the selection criteria more rigorously , found no strong statistical evidence for this . their result , however , may be biased by incompleteness in the hcg catalog : there seem to be too - few low surface brightness groups in the catalog , and the `` missing '' groups may have a higher fraction of discordant redshifts ( sulentic 1997 ) . in order to address the incompleteness issue , iovino & hickson ( 1996 ) combined observational results from both the hcg and scg catalogs with monte - carlo simulations . their technique exploits the unbiased nature of the scg catalog and the complete redshift coverage of the hcg sample . they conclude that for all except the two highest - surface - brightness quintets ( stephan s quintet and seyfert s sextet ) , the number of discordant redshifts is consistent with chance projections . for these two , the chance probabilities are low . however , for both of these systems there is independent physical evidence that the discordant galaxies are at the cosmological distances that correspond to their redshifts and are therefore not group members ( kent 1981 , wu et al 1994 ) . one should not assume that the situation is now completely settled . further studies will be possible when redshifts have been obtained for the scg galaxies . there are still other questions that have not been adequately addressed , such as reported redshift quantization ( cocke & tifft 1983 ) . however , at this point it appears that the frequency of discordant galaxies does not require a new interpretation of galaxy redshifts . in fact , physical evidence suggests the opposite . the discordant galaxies all have physical properties consistent with a cosmological distance . for example those with higher redshift tend to be smaller and fainter than other members of the group , and vice versa ( mendes de oliveira 1995 ) . because we can measure only three phase - space dimensions for galaxies in compact groups ( two components of position and one of velocity ) , the groups are subject to projection effects . because of this , they may not be physically dense , or even physically related systems . the following interpretations have so far been suggested for compact groups : 1 . transient dense configurations ( rose 1977 ) 2 . isolated bound dense configurations ( sulentic 1987 , hickson & rood 1988 ) 3 . chance alignments in loose groups ( mamon 1986 , walke & mamon 1989 , mamon 1995 ) 4 . filaments seen end - on ( hernquist et al 1995 ) 5 . bound dense configurations within loose groups ( diaferio et al 1994 , governato et al 1996 ) evidence for and against physical association and high density in the hcg sample , to 1988 , was summarized by hickson & rood ( 1988 ) , and by walke and mamon ( 1989 ) respectively . since that time , several new results have emerged . from an analysis of optical images , mendes de oliveira & hickson ( 1994 ) concluded that 43% of all hcg galaxies show morphological features indicative of interaction and/or merging , and that 32% of all hcgs contain three or more interacting galaxies . these percentages are likely to rise with more - detailed studies and sophisticated image analysis ( longo et al 1994 ) . this high frequency of interactions observed in compact groups is difficult to reconcile with the chance alignment and filament hypotheses , even if the alignments contain physical binaries ( mamon 1995 ) . the high fraction of hcgs showing diffuse x - ray emission is very strong evidence that a large fraction of these systems are physically dense , and are not transient configurations or projection effects . although the exact numbers are not final , due to the faintness of the sources and the problems of contamination by sources associated with the individual galaxies , it seems evident that many groups are dense bound systems . the correlations seen between x - ray and optical properties , and the fact that the x - ray properties of compact groups are not inconsistent with those of clusters reinforces this conclusion . ostriker et al ( 1995 ) have argued that the relatively low x - ray luminosities of compact groups might not be due to a low gas fraction , but instead could be understood if the groups are filaments seen in projection ( hernquist et al 1995 ) . however , ponman et al ( 1996 ) point out that in order to explain even the fainter compact groups , gas temperatures @xmath40 kev and densities @xmath41 @xmath42 would be required . these appear to be ruled out by both observations ( briel & henry 1995 ) and simulations ( diaferio et al 1995 , pildis et al 1996 ) . even if compact groups are physically dense , they may not be as dense as they appear . as mentioned in section 2 , a sample of groups selected on the basis of high apparent density will be biased by the inclusion of looser systems which appear more compact due to geometrical or kinematic effects . is this bias large ? its magnitude can be estimated as follows : consider @xmath43 galaxies randomly located within a circle of radius @xmath11 on the sky . what is the probability @xmath44 that they will fall within some circular subarea of radius @xmath45 ? the answer can be obtained using analytic expressions derived by walke & mamon ( 1989 ) . from their equations 1 and 6 ( setting @xmath46 and @xmath47 ) we obtain f(x , n ) = n!^n _ 0^x dr _ 0 ^ 1-r d where dn = 2^n-1 n r^2n-3(n-2 ) ! 2ddr is the number of possible configurations with radius between @xmath48 and @xmath49 and distance from the center between @xmath50 and @xmath51 . here we have neglected a small edge contribution that is unimportant for small values of @xmath52 ( walke and mamon s case 3 ) . this gives f(x , n ) = n^2 x^2n-2 now , an observer would infer a galaxy space density that is higher by a factor @xmath53 , so the average apparent space density enhancement is < > & = & _ 0 ^ 1 dx + & = & _ 0 ^ 1 n^2 x^2n-5 dx + & = & 2n^2(n-1)(2n-5)(2n-3)(n-2 ) thus we expect to typically overestimate the space density by about a factor of 12.0 for triplets ( or quartets containing a physical binary ) , 3.2 for true quartets and 2.0 for quintets . a key question that remains is the position of compact groups in the clustering hierarchy . are compact groups distinct entities ( sulentic 1987 ) or an intermediate stage between loose groups and triplets , pairs and individual galaxies ( barnes 1989 , white 1990 , cavaliere et al 1991 , rampazzo & sulentic 1992 , diaferio et al 1994 ) . some compact groups are purely projection effects , others may be small clusters ( ebeling et al 1995 ) , but most appear to be real . it seems that they can arise naturally from subcondensations in looser groups , but further studies are needed to better determine both the observed space density of groups as a function of population and the timescales involved in the evolutionary process . this question is related to that of the formation mechanism of compact groups . two mechanisms have been discussed in the literature . diaferio et al ( 1994 ) conclude that compact groups form continually from bound subsystems within loose groups . this gains some support from the observation ( see section 3 ) that most hcgs are embedded in loose groups , although it is not obvious that these loose groups are sufficiently rich ( sulentic 1997 ) . governato et al ( 1996 ) , proposed a model in which merging activity in compact groups is accompanied by infall of galaxies from the environment . this naturally explains the observed mix of morphological types , and it allows compact groups to persist for longer times . where do the shakhbazian groups fit in this picture ? recent studies ( tikhonov 1986 , amirkhanian & egikian 1987 , amirkhanian et al 1988 , 1991 , amirkhanian 1989 , kodaira et al 1988 - 91 , stoll et al 1993 - 1996 ) show that these objects are typically compact clusters or groups of early - type galaxies . although the systems were selected on the basis of red colors and compact appearance of their galaxies , both of these factors result from their large distances because k - corrections and contrast effects become significant . the galaxies are in fact relatively normal , although luminous ( del olmo et al 1995 ) . however , the number of blue ( gas rich ) galaxies in these systems does seem to be very small . thus it appears the shakhbazian groups are mostly small clusters , possibly intermediate in physical properties between classical compact groups and clusters . if the groups are dynamically bound , galaxy mergers should commence within a few dynamical times ( carnevali et al 1981 , ishizawa et al 1983 , barnes 1985 , ishizawa 1986 , mamon 1987 , 1990 , zheng et al 1993 ) . both n - body and hydrodynamic simulations indicate that the dark matter halos of individual galaxies merge first , creating a massive envelope within which the visible galaxies move ( barnes 1984 , bode et al 1993 ) . kinematic studies of loose groups ( eg . puche & carignan 1991 ) indicate that the dark matter is concentrated around the individual optical galaxies . in contrast , the x - ray observations indicate that in most compact groups , the gas and dark matter is more extended and is decoupled from the galaxies . this may explain the observation that galaxies in compact groups typically have mass - to - light ratios 30% to 50% lower than more isolated galaxies ( rubin et al 1991 ) . is there any observational evidence that galaxies in compact groups are merging ? by 1982 it was evident that first - ranked galaxies in compact groups did not appear to be merger products , because the fraction of first - ranked galaxies that are type e or s0 is the same as for the general population of hcg galaxies ( hickson 1982 ) . if mergers were a dominant effect , the first - ranked galaxies would be expected to be more often elliptical . the same conclusion was reached by geller & postman ( 1983 ) who found that the luminosities of first - ranked galaxies were consistent with a single luminosity distribution for all group galaxies . of course this may just mean that in small groups the first - ranked galaxy is not necessarily the most evolved . rather , one should ask if _ any _ galaxies in compact groups show indications of merging . the relative paucity of merging galaxies in compact groups was first noted by tikhonov ( 1987 ) , from a visual inspection of optical images . zepf & whitmore ( 1991 ) realized that elliptical galaxies formed by recent mergers of gas - rich systems should have bluer colors than normal . examining the hcgs , they found only a small enhancement in the fraction of early - type galaxies having blue colors , a conclusion reinforced by an independent study by moles et al ( 1994 ) . on the other hand , caon et al ( 1994 ) argued that the large effective radii of compact group elliptical galaxies is indicative of an origin by merging or accretion of companions . zepf ( 1993 ) estimated that roughly 7% of the galaxies in compact groups are in the process of merging . this conclusion was based on roughly consistent frequencies of _ ( a ) _ optical signatures of merging , _ ( b ) _ warm far - infrared colors , and _ ( c ) _ sinusoidal rotation curves . however , few galaxies show all of these effects simultaneously . the merging fraction may thus be as high as 25% if one allows that any one of these criteria would be considered to be sufficient to indicate a merger ( hickson 1997 ) . given the small numbers of objects in these studies , it is fair to say that the fraction of merging galaxies is highly uncertain at present . it seems safe to conclude that current observations do not rule out a significant amount of merging in compact groups . detailed studies of individual compact groups can be quite revealing . many galaxies that at first appear normal are revealed to have peculiar morphology or spectra when examined more closely . many , perhaps most , compact groups clearly contain galaxies that are dynamically interacting . however , the groups likely span a range of evolutionary states . at the extreme end are high - density groups like seyfert s sextet , hcg 31 , hcg 62 , hcg 94 ( pildis 1995 ) and hcg 95 ( rodrigue et al 1995 ) in which we find strong gravitational interactions . at the other end are lower density compact groups , such as hcg 44 , which most likely are in a less - advanced stage of evolution . this picture is supported by radio observations : seyfert s sextet and hcg 31 are both embedded in extended hi clouds whereas in hcg 44 the hi is associated with individual galaxies ( williams et al 1991 ) . it seems clear that the groups as we now see them can persist for only a fraction of a hubble time . simulations indicate that merging should destroy the group on a time scale @xmath54 that is typically an order of magnitude larger than @xmath55 , depending on the distribution of dark matter ( cavaliere et al 1983 , barnes 1984 , navarro et al 1987 , kodaira et al 1990 ) and initial conditions ( governato et al 1991 ) . assuming that the groups are in fact bound dynamical systems , we can draw two conclusions : _ ( a ) _ there must be an ongoing mechanism for forming or replacing compact groups , and _ ( b ) _ there must be a significant population of relics of merged groups . what are the end - products of compact groups ? it is tempting to identify them with field elliptical galaxies , following a suggestion first made by toomre ( 1977 ) . simulations ( weil & hernquist 1994 ) indicate that multiple mergers in small groups of galaxies best reproduce the observed kinematical properties of elliptical galaxies . the resulting galaxies are predicted to possess small kinematic misalignments , which can be detected by detailed spectroscopic and photometric studies . neverthess , it remains to be demonstrated that these merger remnants can reproduce the tight correlation between size , luminosity and velocity dispersion found in present - day elliptical galaxies . if compact groups have lifetimes on the order of @xmath54 , and form continuously , then the number of relics , per observed group , is expected to be on the order of @xmath56 . thus , the number of relics could exceed that of present day groups by as much as an order of magnitude . mamon ( 1986 ) estimated that , if all hcgs are real , then the relics would account for about 25% of luminous field elliptical galaxies . as we have seen , the true space density of compact groups is uncertain by at least a factor of two , and may be underestimated because of selection biases . there is then the potential problem of producing too many relics . a second problem is the fact that the integrated luminosities of compact groups are typically a factor of three to four times greater than luminosities of isolated elliptical galaxies ( sulentic and rabaa 1994 ) . it is possible that interaction - induced star formation has boosted the luminosities of some compact - group galaxies , and that some degree of fading of the merger product is expected . however , at this point it is not clear whether or not the relics can be identified with isolated elliptical galaxies . despite these problems , a fossil compact group may have actually been found . ponman et al ( 1995 ) have detected a luminous isolated elliptical galaxy surrounded by diffuse x - ray emission which is consistent with the expected end - product of a compact group . if more objects like this are found , it may be possible to compare their space density with that expected for compact group relics . interactions are often implicated in the development of active nuclei in galaxies ( eg . freudling & almudena prieto 1996 ) . the hcg catalog includes several examples of compact groups containing both starburst galaxies and agn . several recent examples of associations between starburst galaxies or agn and what appear to be compact groups have been reported : del olmo & moles ( 1991 ) have found a broad - line agn in shakhbazian 278 ; zou et al ( 1995 ) find that the luminous infrared source iras 23532 coincides with a compact group that includes a seyfert 1 as well as a starburst galaxy . if this association extends to qsos , one would expect to find numerous compact groups at redshifts @xmath57 , where the comoving number density of qsos peaks ( eg . hartwick & schade 1984 ) . the tendency for qsos to have close companions has been known for some time ( eg . stockton 1982 , bahcall et al 1997 ) . recently , several examples of compact groups associated with luminous infrared galaxies , agn and qsos at @xmath58 have been found using hst ( pascarelle et al 1996 , francis et al 1996 , matthews et al 1994 , tsuboi & nakai 1994 , hutchings 1995 , hutchings et al 1995 ) . these observations provide support to the idea that tidally - triggered star formation is a predominant factor in the galaxy formation process ( lacey & silk 1991 , lacey et al 1993 ) . in this model disk star formation occurs relatively late , after the compact group has formed and tidal interactions are strong . this seems at least qualitatively consistent with the fragmentary nature of high - redshift galaxies observed with the hubble space telescope ( schade et al 1995 ) , although these fragments appear to be much less luminous and more irregular than most present - day compact group galaxies . the model also offers a possible explanation for the excess numbers of faint blue galaxies found in field galaxy count as dwarf galaxies undergoing star formation at a redshift of @xmath59 . compact groups may possibly play a role in the formation of other systems . we have seen that giant galaxies may be formed as the end product of compact - group evolution . at the other end of the scale , dwarf galaxies have physical properties distinct from normal galaxies , which suggests a unique formation mechanism . one possibility is that they form during gravitational interactions from tidal debris ( duc & mirabel 1994 ) . if this is the case , one would expect to find evidence for this in compact groups of galaxies . from an examination of condensations in tidal tails , hunsberger et al ( 1996 ) concluded that the fraction of dwarf galaxies produced within tidal debris in compact groups is not negligible . there is also evidence that star clusters form from tidal debris . longo et al ( 1995 ) have found an excess population of unresolved blue objects around hcg 90 which appear to be recently formed star clusters . these may be similar to the population of new star clusters recently reported in the merger remnant ngc 7252 ( whitmore et al 1993 ) . because compact groups have a high galaxy surface density , they may form effective gravitational lenses . gravitational amplification of background field galaxies was proposed by hammer and nottale ( 1986 ) as a possible explanation for the presence of the high - redshift discordant member of this group . mendes de oliveira and giraud ( 1994 ) and montoya et al ( 1996 ) find that most hcgs are too nearby to produce strong lensing effects . however , because the critical mass density required for strong lensing depends reciprocally on distance , analogous systems 5 - 10 times more distant should produce a non - negligible fraction of giant arcs . studies of small groups may provide clues to the overall structure of the universe . the baryon fractions found in clusters of galaxies appear to be inconsistent with a density parameter @xmath60 , unless the dark matter is more prevalent outside clusters ( white 1992 , babul & katz 1993 ) . compact groups provide a means to study dark matter in such regions . the baryon fractions found for compact groups do appear to be lower than those for clusters . david et al ( 1995 ) argued that the gas is the most extended component ; galaxies being the most compact and the dark matter being intermediate . they concluded that the baryon fraction approaches 30% on large enough scales , which is comparable to the values found for clusters . given the constraints of standard big - bang nucleosynthesis this would imply that the density parameter @xmath61 is at most 0.2 . on the other hand , the infall picture of compact - group evolution ( governato et al 1996 ) requires a high - density @xmath62 universe . in a low - density universe the infall rate is insufficient . as there is at present no other clear mechanism for avoiding the overproduction of relics by merging compact groups , this may be a strong argument for a high - density universe . during the last two decades we have seen a resurgence of interest in compact groups . while initially little more than a curiosity , these systems are now viewed as potentially important sites of dynamical evolution , shaping the structure of many galaxies . it now seems clear that while many compact groups are contaminated by projections , a large fraction of at least the high - surface - brightness hcgs are physically dense . they form by gravitational relaxation processes within looser associations of galaxies . the densest are generally in an advanced stage of evolution characterized by strong interactions , starburst and agn activity , stripping of stellar and dark matter halos , and merging . they contain large amounts of dark matter and primordial x - ray - emitting gas trapped within the gravitational potential well . despite this progress , many questions remain unanswered . what are the end products of compact group evolution , and do they have properties consistent with any know population of objects ? what is the space density of such relics ? where do compact groups fit in the overall clustering hierarchy ? what is their role in the evolution of galaxies both past and present ? given the current interest and research activity in this area , it is likely that many of these questions may soon be addressed . it is a pleasure to thank the observatories of brera and capodimonte for hospitality during the initial work on this review . i have benefitted from discussions with many individuals , but i would like to acknowledge particularly the contributions of a iovino , e kindl , g longo , g mamon , c mendes de oliveira , tk menon , g palumbo , h rood , and j sulentic . i thank g mamon , a sandage and j sulentic for providing helpful comments on an earlier version of the manuscript . financial support was provided by the natural sciences and engineering research council of canada and nato . [ a]llam s , assendorp r , longo g , braun m , richter g. 1996 . _ _ 117:39 - 82 [ a]mbartsumian ba 66:536 - 40 [ a]mirkhanian as . 1989 . ssr bjurak . 61:25 - 28 [ a]mirkhanian as , egikian ag . 1987 . _ astrofizika _ 27:395 - 97 [ a]mirkhanian as , egikian ag , silchenko ok . _ pisma astron . zh . _ 14:404 - 08 [ a]mirkhanian as , eghikian ag , tikhonov na , shakhbazian rk . _ astrofizika _ 35:67 - 75 [ a]rp h. 1966 . _ ap . j. suppl . _ 14:1 - 20 [ a]rp h. 1973 . _ 185:797 - 808 [ a]rp h. 1987 . _ quasars , redshifts and controversies _ , berkeley : interstellar media [ b]abul a , katz n. 1993 . _ ap . j. lett . _ 406:l51-l54 [ b]ahcall jn , kirharos s , saxe dh , schneider dp . 1997 . _ ap . j. _ 479 , in press [ b]ahcall na , harris de , rood hj . _ 284:l29-l33 [ b]aier fw , petrosian , mb , tiersch h , shakhbazian rk . _ astrofizika _ 10:327 - 35 [ b]aier fw , tiersch h. 1975 . _ astrofizika _ 11:221 - 28 [ b]aier fw , tiersch h. 1976 . _ astrofizika _ 12:7 - 12 [ b]aier fw , tiersch h. 1976 . _ astrofizika _ 12:409 - 15 [ b]aier fw , tiersch h. 1978 . _ astrofizika _ 14:279 - 82 [ b]aier fw , tiersch h. 1979 . _ astrofizika _ 15:33 - 35 [ b]alcells n , quinn pj . j. _ 361:381 - 93 [ b]arnes j. 1984 . _ mnras _ 208:873 - 85 [ b]arnes je . _ mnras _ 215:517 - 36 [ b]arnes je . 1989 . _ nature _ 338:123 - 26 [ b]arton e , geller mj , ramella m , marzke ro , da costa ln . 112:871 - 86 [ b]ergvall n , ekman a , lauberts a. 1981 . _ astron . _ 95:266 - 77 [ b]ettoni d , fasano g. 1993 . 105:1291 - 307 [ b]ettoni d , fasano g. 1995 . _ astron . j. _ 109:32 - 55 [ b]ettoni d , fasano g. 1996 . _ _ 118:429 - 39 [ b]iermann p , kronberg pp . 1984 . _ physica scr . _ t7:169 [ b]ode pw , cohn hn , lugger pm . 1993 . _ ap . j. _ 416:17 - 25 [ b]oselli a , mendes de oliveira c , balkowski c , cayatte v , casoli f. 1996 . _ 314:738 - 44 [ b]riel ug , henry jp . 1995 . astrophys . _ 302:l9-l12 [ b]urbidge em . , burbidge gr . 130:23 - 25 [ b]urbidge em . , burbidge gr 66:541 - 50 [ b]urbidge em . , burbidge gr . 1961b . _ ap . j. _ 134:244 - 47 [ b]urbidge em . , burbidge gr . 1961c . _ ap . j. _ 134:248 - 56 [ b]urbidge em , sargent wlw . 1971 . 35:351 - 78 [ c]ampos - aguilar a , moles m. 1991 . _ 241:358 - 64 [ c]aon n , capaccioli m , donofrio m , longo g. 1994 . astrophys . _ 286:l39-l42 [ c]arnevali p , cavaliere a , santangelo p. 1981 . _ ap . j. _ 249:449 - 61 [ c]avaliere a , colafrancesco s , scaramella , r. 1991 . _ ap . j. _ 380:15 - 23 [ c]avaliere a , santangelo p , tarquini g , vittorio n. 1983 . in _ clustering in the universe , d gerbal , a mazure , pp . gif - sur - yvette : ed . front . [ c]ocke wj , tifft wg . _ ap . j. _ 268:56 - 59 [ d]a costa ln , geller mj , pellegrini ps , latham dw , fairall ap , marzke ro , willmer cna , huchra jp , calderon jh , ramella m , kurtz mj , _ ap . j. lett . _ 424:l1-l4 [ d]avid lp , jones c , forman w. 1995 . _ ap . j. _ 445:578 - 90 [ d]avis ds , mulchaey js , mushotzky rf , burstein d. 1996 . _ ap . j. _ 460:601 - 11 [ d]e carvalho rr , djorgovski sg 1995 . soc . _ 187:53.02 [ d]e carvalho rr , ribeiro alb , zepf se . _ 93:47 - 63 [ d]ellantonio ip , geller mj , fabricant dg . 1995 . 110:502 - 12 [ d]el olmo a , moles m. 1991 . _ 245:27 - 30 [ d]el olmo a , moles m , perera j. 1995 . in _ groups of galaxies _ , asp conference series , 70:117 - 126 [ d]e souza re , capelato hv , arakaki l , logullo c. 1982 . _ ap . j. _ 263:557 - 63 [ d]iaferio a , geller mj , ramella m. 1994 . _ astron . j. _ 107:868 - 79 [ d]iaferio a , geller mj , ramella m. 1995 . 109:2293 - 304 [ d]ressler a. 1980 . _ ap . j. _ 236:351 - 65 [ d]uc p - a , mirabel if . 1994 . _ , 289:83 - 93 [ e]beling h , mendes de oliveira c , white da . 1995 . _ mnras _ 277:1006 - 32 [ e]beling h , voges w , boehringer h. 1994 . _ ap . j. _ 436:44 - 55 [ f]asano g , bettoni d. 1994 . _ astron . 107:1649 - 67 [ _ ap . j. _ ] 457:490 - 99 [ f]reudling w , almudena prieto m. 1996 . _ _ 306:39 - 48 [ f]ukazawa y , makishima k , matsushita k , yamasaki n , ohashi t , et al _ 48:395 - 407 [ g]arcia a. 1995 . _ 297:56 - 60 [ g]eller mj , postman m. 1983 . _ ap . j. _ 274:31 - 38 [ g]isler g. 1980 . 85:623 - 25 [ g]overnato f , chincarini g , bhatia r. 1991 . _ ap . j. _ 371:l15-l18 [ g]overnato f , tozzi , p , cavaliere a. 1996 . _ ap . j. _ 458:18 - 26 [ h]ammer f , nottale l. 1986 . _ 155:420 - 22 [ h]eiligman gm , turner el . 1980 . _ ap . j. _ 236:745 - 49 [ h]enriksen mj , mamon ga . 1994 . _ ap . j. lett . _ 421:l63-l66 [ h]enry jp , gioia i m , huchra jp , burg r , mclean b , et al . 1995 . _ ap . j. _ 449:422 - 30 [ h]ernquist l , katz n , weinberg dh . 1995 . _ ap . j. _ 442:57 - 60 [ h]ickson p. 1982 . _ ap . j. _ 255:382 - 91 [ h]ickson p. 1990 . in _ paired and interacting galaxies , _ int . astron . union colloq . huntsville , al : nasa marshall space flight center [ h]ickson p. 1993 . _ ap . lett . & commun . _ 29:1 - 207 [ h]ickson p. 1994 . _ atlas of compact groups of galaxies _ , basel : gordon & breach [ h]ickson p. 1997 . _ ap . lett . & commun . _ in press [ h]ickson p , kindl e , huchra jp . 1988a . _ ap . j. _ 329:l65-l67 [ h]ickson p , kindl e , huchra jp . 1988b . _ ap . 331:64 - 70 [ h]ickson p , kindl e , auman jr . _ 70:687 - 98 [ h]ickson p , mendes de oliveira c , huchra jp , palumbo gg 399:353 - 67 [ h]ickson p , menon tk , palumbo ggc , persic m. 1989 . _ ap . 341:679 - 84 [ h]ickson p , ninkov z , huchra jp , mamon ga . 1984 . in _ clusters and groups of galaxies _ , ed . f mardirossian , g giuricin , m mezzetti , pp . dordrecht : reidel [ h]ickson p , richstone do , turner el . 213:323 - 26 [ h]ickson p , rood hj . 1988 . _ ap . j. _ 331:l69-l72 [ h]uchtmeier w. 1994 . _ 286:389 - 94 [ h]unsberger sd , charlton jc , zaritsky d. 1996 . _ ap . j. _ 462:50 - 56 [ h]umason ml , mayall nu , sandage ar . 1956 . 61:97 [ h]utchings jb 1995 . 109:928 - 34 [ h]utchings jb , crampton d , johnson a. 1995 . 109:73 - 80 [ i]ovino a , hickson p. 1996 . _ mnras _ in press [ i]shizawa t. 1986 . _ astrophys . space sci . _ 119:221 - 25 [ i]shizawa t , matsumoto r , tajima t , kageyama h , sakai h. 1983 . _ 35:61 - 76 [ k]ent sm _ 93:554 - 57 [ k]indl e. 1990 . a photometric and morphological study of compact groups of galaxies and their environments , ph.d . thesis , university of british columbia , vancouver , canada , 168 pp . [ k]irshner rp , malumuth em . 236:366 - 72 [ k]iseleva lg , orlov vv . 1993 . _ vistas astron . _ 36:1 - 30 [ k]odaira k , doi m , ichikawa s - i , okamura s. 1990 . _ 1:283 - 95 [ k]odaira k , iye m , okamura s , stockton a. 1988 . _ 40:533 - 45 [ k]odaira k , sekiguchi m , sugai h , doi m. 1991 . _ 43:169 - 76 , [ k]ormendy j , sanders db . 390:l53 - 56 [ l]acey c , guiderdoni b , rocca - volmerange b , silk j. 1993 . _ 402:15 - 41 [ l]acey c , silk j. 1991 . _ 381:14 - 32 [ l]imber dn , mathews wg . 132:286 - 305 [ l]ongo g , busarello g , lorenz h , richter g , zaggia s. 1994 . _ astron . astrophys . _ 282:418 - 24 [ l]ongo g , grimaldi a , richter g. 1995 . astrophys . _ 299:l45-l48 [ l]ynds cr , khachikian ee , amirkhanian as . _ pisma astron . zh . _ 16:195 - 98 [ m]accagni d , gioia i m , henry jp , maccacaro t , vettolani gp 100:1461 - 67 [ m]alumian vh . 1996 . 317:101 [ m]alykh sa , orlov vv . 1986 . _ astrofizika _ 24:445 - 51 [ m]amon ga . 307:426 - 30 [ m]amon ga j. _ 321:622 - 44 , [ m]amon ga . _ 219:98 - 100 [ m]amon ga . 1990 . in _ paired and interacting galaxies , _ int . union colloq . huntsville , al : nasa marshall space flight center [ m]amon ga . 1995 . in _ groups of galaxies , _ asp conf . 70:83 - 94 [ m]arzke ro , huchra jp , geller mj . 1944 . _ ap . j. _ 428:43 - 50 [ m]assey p. 1977 . _ 89:13 - 18 [ m]atthews k , soifer bt , nelson j , boesgaard h , graham jr . , et al . _ 420:l13-l16 [ m]endes de oliveira c. 1995 . _ mnras _ 273:139 - 45 [ m]endes de oliveira c , giraud e. 1994 . _ ap . j. _ 437:l103-l106 [ m]endes de oliveira c , hickson p. 1991 . _ ap . 380:30 - 38 [ m]endes de oliveira c , hickson p. 1994 . _ ap . j. _ 427:684 - 95 [ m]enon tk . 372:419 - 23 [ m]enon tk _ mnras _ 255:41 - 47 [ m]enon tk 110:2605 - 9 [ m]enon tk _ mnras _ 274:845 - 852 [ m]enon tk , hickson , p. 1985 . _ j. _ 296 , 60 - 64 , [ m]enon tk , leon s , combes f. 1996 . _ bull . am . 188:4502 [ m]irzoian lv , miller js , osterbrock de . 1975 , _ ap . 196:687 - 8 [ m]oles m , del olmo a , perea j , masegosa j , marquez i , costa , v. 1994 . _ _ 285:404 - 14 [ m]ulchaey js , davis ds , mushotzky rf , burstein d. 1993 . _ ap . j. lett . _ 404:l9-l12 [ m]ulchaey js , davis ds , mushotzky rf , burstein d. 1996a . _ ap . 456:80 - 97 [ m]ulchaey js , mushotzky rf , burstein d , davis ds . 1996b . _ ap . j. lett . _ 456:l5-l8 [ n]avarro jf , mosconi m , garcia lambas d. 1987 . _ mnras _ 228:501 - 11 [ n]ilson pn . _ uppsala general catalogue of galaxies , _ uppsala obs . [ n]ottale l , moles m. 1978 . _ 66 , 355 - 58 [ o]leak h , stoll d , tiersch h , macgillivray h. 1995 . 109:1485 - 89 [ o]striker jp , lubin lm , hernquist l. 1995 . _ j. _ 444:l61-l64 [ o]striker jp , tremaine sd . _ 202:l113-l117 [ p]alumbo ggc , saracco p , hickson p , mendes de oliveira c , 1995 . 109:1476 - 84 [ p]alumbo ggc , saracco p , mendes de oliveira c , hickson p , tornatore v , baiesi - pillastrini gc . 1993 . _ ap . j. _ 405:413 - 18 [ p]alumbo g , saracco p , hickson p , mendes de oliveira c. 1995 . _ astron . 109:1476 - 84 [ p]ascarelle sm , windhorst ra , driver sp , ostrander ej , keel wc . 1996 . _ ap . j. lett . _ 456:l21-l24 [ p]eebles pje . _ physical cosmology . _ princeton : princeton univ . press [ p]etrosian mb . 1974 . _ astrofizika _ 10:471 - 76 [ p]etrosian mb . _ astrofizika _ 14:631 - 35 [ p]ildis ra . 455:492 - 96 [ p]ildis ra , bregman jn , evrard ae . 1995a . _ ap . j. _ 443:514 - 26 [ p]ildis ra , bregman jn , schombert jm 110:1498 - 506 [ p]ildis ra , evrard ae , bregman jn . 1996 . 112:378 - 87 [ p]onman tj , bertram d. 1993 . _ nature _ 363:51 - 54 [ p]onman tj , allan dj , jones lr , merrifield m , mchardy i m , lehto hj , luppino ga . _ nature _ 369:462 - 64 [ p]onman tj , bourner pdj , ebeling h , bohringer h. 1996 . _ mnras _ 283:690 - 708 [ p]ostman m , geller mj . 1984 . _ 281:95 - 99 [ p]randoni i , iovino a , macgillivray ht . 107:1235 - 44 [ p]ress wh , schechter p. 1974 . _ ap . j. _ 187:425 - 38 [ p]uche d , carignan c. 1991 . _ ap . j. _ 378:487 - 95 [ r]amella m , diaferio a , geller mj , huchra jp . 1994 . _ astron . 107:1623 - 28 [ r]ampazzo r , sulentic jw . 1992 . _ astron . astrophys . _ 259:43 - 60 [ r]aymond jc , smith bw _ 35:419 - 39 [ r]ibeiro alb , de carvalho rr , zepf se . 1994 . _ mnras _ 267:l13-l16 [ r]ibeiro alb , de carvalho rr , coziol r , capelato hv , zepf se . _ 463:l5-l8 [ r]odrigue m , schultz a , thompson j , colegrove t , spight ld , et al . 109:2362 - 67 [ r]ood hj , struble mf . 1994 . _ 106:413 - 16 [ r]ood hj , williams ba 339:772 - 82 [ r]ose ja . 211:311 - 18 [ r]ose ja 231:10 - 22 [ r]ose ja , graham ja . 1979 . _ ap . j. _ 231:320 - 26 [ r]ubin vc , ford wkj , hunter da . 1990 . _ ap . j. _ 365:86 - 92 [ r]ubin vc , hunter da , ford wkj . 1991 . _ ap . j. suppl . _ 76:153 - 83 [ s]akai s , giovanelli r , wegner g. 1994 . 108:33 - 43 [ s]andage a , binggeli b & tammann ga . 90:1759 - 71 [ s]anders db , mirabel if . 1996 . _ 34:749 - 92 [ s]aracco p , ciliegi p. 1995 . _ astron . _ 301:348 - 58 [ s]argent wlw _ 153:l135 - 37 [ s]chade d , lilly sj , crampton d , hammer f , le fevre o , tresse l. 1995 . _ j. _ 451:l1-l4 [ s]chechter p. 1976 . _ ap . 203:297 - 306 [ s]chwartz da , schwarz j , tucker w. 1980 . _ ap . j. _ 238:l59-l62 [ s]eyfert ck . j. _ 53:203 - 4 [ s]hakhbazian rk 1957 . _ 177:11 [ s]hakhbazian rk 1973 . _ astrofizika _ 9:495 - 501 [ s]hakhbazian rk , amirkhanian as . 1979 . _ astrofizika _ 14:455 - 66 [ s]hakhbazian rk , petrosian mb . _ astrofizika _ 10:13 - 20 [ s]tephan me . mnras _ 37:334 - 39 [ s]tockton a. 1982 . _ 257:33 - 39 [ s]toll d , tiersch h , oleak h , baier f , macgillivray ht . 1993a . _ 314:225 - 67 [ s]toll d , tiersch h , oleak h , baier f , macgillivray ht . 1993b . _ 314:317 - 60 [ s]toll d , tiersch h , oleak h , macgillivray ht _ 315:11 - 61 [ s]toll d , tiersch h , oleak h , macgillivray ht . 1994b . _ 315:97 - 150 [ s]toll d , tiersch h , braun m. 1996a . _ 317:239 - 57 [ s]toll d , tiersch h , braun m. 1996b . _ 317:315 - 32 [ s]ulentic jw . 1983 . _ 270:417 - 21 [ s]ulentic jw . 1987 . _ ap . j. _ 322:605 - 17 [ s]ulentic jw . 1993 . in _ progress in new cosmologies : beyond the big bang hc arp et al , pp . new york : plenum press [ s]ulentic jw . j. _ in press [ s]ulentic jw , de mello rabaa df . 1993 . _ ap . j. _ 410:520 - 25 [ s]ulentic jw , lorre jj . _ 120:36 - 52 [ s]ulentic jw , pietsch w , arp h. 1995 . _ 298:420 - 26 [ s]ulentic jw , rabaa cr 429:531 - 39 [ t]hompson la , & gregory sa . 242:1 - 7 [ t]iersch h. 1976 . _ 297:301 - 3 [ t]ikhonov na . _ soobsc . spets . _ 49:69 - 84 [ t]ikhonov na _ astrofizika _ 27:253 - 64 [ t]ovmasian gm , shakhbazian ets , shirbakian ms . 1980 . _ astrofizika _ 15:513 - 14 [ t]oomre a. 1977 . in _ the evolution of galaxies and stellar populations , _ ed . bm tinsley , rb larson , pp . new haven : yale univ . [ t]suboi m , nakai n. 1994 . jpn . _ 46:l179-l182 [ t]ully rb . _ nearby galaxies catalog _ , cambridge : cambridge univ . press [ v]ennik j , richter gm , longo g. 1993 . 314:393- [ v]enugopal vr . _ mnras _ 277:455 - 57 [ v]ettolani g , gregorini l. 1988 . _ astron . _ 189:39 - 41 [ v]orontsov - velyaminov ba _ atlas and catalog of interacting galaxies , _ vol 1 . moscow : sternberg inst . [ v]orontsov - velyaminov ba . _ atlas of interacting galaxies , part ii _ , _ astron . suppl . _ 28:1 [ v]orontsov - velyaminov ba , dostal va , metlov vg . _ soviet astron . _ 6:394 - 97 [ v]orontsov - velyaminov ba , metlov vg . 1980 . _ soviet astr . _ 6 , 199 - 201 [ w]alke dg , mamon ga . 1989 . _ _ 225:291 - 302 [ w]eil ml , hernquist l. 1994 . _ ap . j. lett . _ 431:l79-l82 [ w]hite sdm . 1990 . in _ dynamics and interactions of galaxies , _ ed . r weilen , pp . 380 - 88 , heidelberg : springer - verlag [ w]hite sdm . 1992 . in _ clusters and superclusters of galaxies _ , nato advanced study institute , pp . 1 - 15 . [ w]hitmore bc . 1992 . in _ physics of nearby galaxies : _ nature _ or nurture ? , _ tx thuan , c balkowski , j tran thanh van , _ 12th moriond astrophys . meet . giv - sur - yvette : ed . front . [ w]hitmore bc , gilmore dm 367:64 - 68 [ w]hitmore bc , schweizer f , leitherer c , borne k , robert c. 1993 . _ astron . j. _ 106:1354 - 70 [ w]illiams ba , mcmahon pm , van gorkom jh . 101:1957 - 68 [ w]illiams ba , rood hj . _ 63:265 - 94 [ w]illiams ba , van gorkom jh 95:352 - 55 [ w]u w , rabaa cr , sulentic jw 1994 . soc . _ 26:1494 [ z]epf se . 407:448 - 55 [ z]epf se , whitmore bc . 1991 . _ ap . j. _ 383:542 - 49 [ z]epf se , whitmore bc . 1993 . _ ap . j. _ 418:72 - 81 [ z]epf se , whitmore bc , levison hf . 1991 . _ 383:524 - 41 [ z]heng j , valtonen mj , chernin ad . 1993 . 105:2047 - 53 [ z]ou z - l , xia x - y , deng z - g , wu h. 1995 . _ astron . astrophys . _ 304:369 - 73

the investigation of the effects of the interplay between discreteness and nonlinearity is a long - standing argument of research in the study of the dynamical properties of nonlinear lattice models @xcite . a typical feature exhibited by nonlinear classical hamiltonian lattices is the existence of discrete breathers , i.e. time - periodic and space - localized solutions of the equations of motion . the study of their dynamical stability , as well as their robustness in long transient processes and thermal equilibrium , has been the subject of an intense experimental and theoretical work @xcite . the interplay between discreteness and nonlinearity is also crucial for the occurrence of modulational instabilities ( mi ) , well known in the theory of nonlinear media @xcite . mi are dynamical instabilities characterized by an exponential growth of arbitrarily small fluctuations resulting from the combined effect of dispersion and nonlinearity . the occurrence of modulational instabilities has been studied in a number of physical systems , ranging from fluid dynamics @xcite to nonlinear optics @xcite . the role and the consequences of the mi in the dynamics of discrete systems have been extensively studied : the mi was discussed in the context of the discrete nonlinear schrdinger equation ( dnlse ) @xcite , which is a paradigmatic lattice model @xcite used to study nonlinear discrete dynamics @xcite . the dnlse is commonly used to describe the effective dynamics in different physical systems of interest , including the dynamics of ultracold atoms in optical lattices @xcite and optical waveguide arrays @xcite . for ultracold bosons in optical lattices the onset of mi was analytically predicted @xcite and experimentally observed @xcite , and in nonlinear waveguide arrays the experimental observation of the mi was also reported @xcite . in this paper we study the occurrence of mi in the dnlse in the presence of non - local long - range hoppings and interactions . a motivation for such a study comes from experiments with ultracold dipolar bosonic gases @xcite which have been bose condensed recently by several groups @xcite , from the attainment of quantum degeneration for ensembles of polar molecules @xcite , and from the recent experimental investigations of strongly interacting rydberg gases @xcite . since the interaction potential in ( di)polar gases decay as a power law @xmath11 ( for rydberg gases interacting through van der waals interactions as @xmath12 ) , recent experiments with dipolar gases in optical lattices @xcite , as well as the realization of long - lived dipolar molecules in a three - dimensional periodic potential @xcite and in perspective the dynamics of rydberg atoms in optical lattices @xcite , open the possibility to study dnlse with non - local interactions . our other motivation is related to the wide interest in systems with long - range interactions @xcite . in these systems the range of interaction of the constitutive units is not bounded . a typical form of interactions , relevant for a number of systems ranging from gravitational ones to dipolar magnets and gases , is provided by the power - law decay @xmath13 ( e.g. for gravitational systems @xmath14 ) where @xmath15 is the distance among the constituents . for statistical mechanics models , like the ising or more generally the @xmath16 models , the possibility to have power - law couplings makes possible the appearance of a rich phase diagram @xcite . a first criterion to determine the long - rangedness of a system with power - law decaying interaction is the comparison with the dimension @xmath17 of the space ; as @xmath18 is smaller than or equal to @xmath17 , if the system is homogeneous and the interaction favours homogeneity we obtain a diverging energy density , thus to obtain a well defined thermodynamic limit ( if relevant ) a rescaling of the energy is in order ( the so called kac rescaling ) @xcite . in the following we will refer to this region as the non - extensive long - range region . if @xmath18 is larger than @xmath17 the energy of the system is extensive and it is normal to individuate a value of @xmath18 , which we denote by @xmath19 , such that for @xmath20 the system behaves as a short - range system . since it is @xmath21 , there is a region of values of @xmath18 , given by @xmath22 , in which the behavior of the system significantly differs from the properties of the same system with short - range interactions , although the energy is extensive . such a region is the extensive long - range region , also referred to as the weak - long - range region @xcite , where `` weak '' refers to the extensivity of the energy . the actual value of @xmath19 depends on the specific model and the dimension : for @xmath16 models in @xmath23 it is @xmath24 @xcite . both the thermodynamics and the dynamics of long - range interacting systems are extremely interesting @xcite . in particular , in the long - range region the dynamical evolution evidences that the system may stay in a quasi - stationary metastable state ( different from the thermal equilibrium one ) for a time exponentially growing with the size of the system . such a metastable state is reached after a short - time dynamics , referred to as violent relaxation @xcite . while the main bodies of the studies on the dynamics of nonlinear lattices have dealt with short - ranged systems , the extensions of these results to long - ranged systems appeared in the literature addressing the properties of discrete systems with different kinds of non - local dispersion or non - local nonlinear interaction we mentioned @xcite and focused on the existence and stability of localized excitations ( for a recent review on nonlinear waves in lattices see @xcite ) . the purpose of the present paper is to study how modulational instabilities emerge in nonlinear lattices with non - local interactions and hoppings , aiming both at unveiling if ( and in which conditions ) short time dynamical instabilities occur in nonlinear lattices and at clarifying the nature of the emerging modulational instability . we choose the dnlse as a case study not only due to its paradigmatic usefulness , but also due to its relation with @xmath25 [ i.e. , @xmath26 models : when the fluctuation of the number of particles are frozen , the kinetic term in the dnlse energy is basically the @xmath25 model ( see the discussion in sec . [ the_model ] ) . this is the reason why we choose to consider not only power - law interactions ( as it is relevant for experiments with ultracold dipolar bosons in optical lattices ) , but also power - law hoppings [ which corresponds to power - law couplings in @xmath27 models ] . using the dnlse we study the modulationally stable and unstable regions in the presence of power - law non - local interactions and hoppings , discussing also the interplay between local and non - local interactions , e.g. local attraction and non - local repulsion . the plan of the paper is as follows . in sec . [ the_model ] , we introduce the dnlse with long - range hoppings and interactions and discuss its relation with other statistical mechanics models . in sec . [ mi_analysis ] we derive the bogoliubov spectrum of elementary excitations , present the general framework for the determination of the stability regions in presence of power - law interactions and hoppings and specialize it to the analysis to the short - range limit reminding the known results of the mi analysis @xcite . our findings for power - law interactions are presented in sec . [ mi_lr_interaction ] where we also consider the case of attractive and competing local and non - local interactions . [ mi_lr_hopping ] deals with a system with non - local hopping and local interaction which presents some peculiar features with respect to the long - range interaction which are further investigated in [ mi_interaction_hopping_comparison ] . the physical applications of our results , in particular to ultracold dipolar gases in optical lattices , are discussed in sec . [ phys_app ] , while our conclusions are in sec . [ conclusions ] . the dnlse with non - local interactions and hoppings reads @xmath28 in eq . @xmath29 is the time and the indices @xmath30 denote the sites of a lattice . for simplicity we assume that the lattice is one dimensional , but the subsequent analysis can be extended to higher dimensional lattices . the indices @xmath31 then assume the values @xmath32 ( @xmath33 is the number of the sites , taken to be even ) . periodic boundary conditions will be also assumed , so that the wavefunction satisfies the condition @xmath34 . the hamiltonian corresponding to eq . reads @xmath35 we denote the diagonal interaction by @xmath36 and the next - neighbor interaction as @xmath37 the interaction coefficients @xmath38 in eq . are assumed to be power - law decaying with exponent @xmath0 , i.e. @xmath39 . since @xmath40 , implementing the periodic boundary conditions amounts to require that @xmath41 , where @xmath42 and @xmath43 . we have therefore @xmath44 the non - local hopping rates @xmath45 will be also assumed to be power - law decaying with exponent @xmath1 . we consider vanishing diagonal hopping ( @xmath46 ) and a nearest - neighbor hopping @xmath47 therefore , with periodic boundary conditions we have @xmath48 since we want a finite expression for the energy per particle , we will consider @xmath49 to treat the cases @xmath50 or @xmath51 one should do a kac rescaling , as it is usually done in statistical mechanics models with nonextensive long - range interactions @xcite : e.g. , for @xmath51 one has to perform the substitution @xmath52 a non - extensive ground - state energy is found in our case by @xmath53 and/or @xmath54 : the region @xmath55 is the weak - long - range hopping region . recall that such region is of high interest in the study of statistical mechanics models with long - range couplings . let us consider an ising model of the form @xmath56 with power - law couplings @xmath57 . as usual , @xmath58 denote the sites of a lattice with dimension @xmath17 . it is well known that for nearest - neighbor couplings ( formally corresponding to @xmath59 ) there is order at finite temperature only if @xmath60 @xcite . however , if the interactions are sufficiently long - ranged it is possible to have a phase transition at finite temperature between a ferromagnetic and a paramagnetic phase @xcite even for @xmath23 : if @xmath61 , then the critical temperature @xmath62 is vanishing ( as in any short - range ising chain @xcite ) , while for @xmath63 the energy is non - extensive . after the kac rescaling one easily sees that for @xmath63 there is a phase transition having the critical exponents of the mean - field universality class ( see e.g. @xcite ) . for @xmath18 between @xmath64 and @xmath65 ( weak - long - range region ) there is a phase transition at a finite critical temperature . for @xmath66 a kosterlitz - thouless transition takes place @xcite before moving on to the derivation of the bogoliubov spectrum of elementary excitations , we pause here to discuss the relation between the dnlse and a model widely used in the treatment of long - range systems , i.e. the hamiltonian mean field ( hmf ) model @xcite . we observe that the dnlse hamiltonian can be written as the sum of a kinetic term @xmath67 and an interaction term @xmath68 . the kinetic part reads @xmath69 : by writing @xmath70 in terms of the local density @xmath71 and phase @xmath72 , i.e. , @xmath73 the kinetic part reads then @xmath74 . one sees than when the number fluctuations are frozen , i.e. @xmath75 ( where @xmath76 is the average number of particles per site ) , the dnlse hamiltonian reduces to the potential energy ( and it has the same equilibrium properties ) of the hmf model @xcite @xmath77 ( where @xmath78 ) , which is nothing but a long - range @xmath25 model . this result is of course expected in the sense that interacting bosons on a lattice are in the @xmath25 universality class , and interacting bosons with long - range hoppings have to be in the universality class of the long - range @xmath25 model . we study in this section the bogoliubov spectrum of elementary excitations , describing the energy of small perturbations with ( quasi)momentum @xmath79 on top of a plane - wave state with ( quasi)momentum @xmath80 @xcite . the final part of the section is devoted to briefly recall the results of the short - range limit ( @xmath81 and @xmath82 , i.e. , only nearest - neighbor hopping ) . the stationary solutions of eq . are plane - waves @xmath83}:\ ] ] @xmath84 is the chemical potential given ( for @xmath85 ) by @xmath86.\ ] ] in eq . @xmath76 is the plane - wave density ( @xmath87 ) ; furthermore @xmath88 is the riemann zeta function @xmath89 and we introduce the function @xmath90 some useful properties of the function are recalled and discussed in the appendix [ appendix_a ] . the stability analysis of plane - waves stationary solutions can be carried out by perturbing the carrier waves as @xmath91 e^{i ( k j - \nu t)}.\ ] ] retaining only the terms proportional to @xmath92 and @xmath93 , one gets @xmath94 the quantities @xmath95 , @xmath96 , and @xmath97 in eq . are defined by @xmath98 + \rho \tilde{v}_q,\ ] ] @xmath99 + \rho \tilde{v}_q\ ] ] and @xmath100 in the previous expressions , @xmath101 is the fourier transform of the interaction : for finite @xmath33 it is @xmath102 for @xmath103 one obtains @xmath104 from eq . it follows that the excitation spectrum ( i.e. , the bogoliubov dispersion relation ) for the dnlse with power - law hoppings and interactions is @xmath105 where @xmath106 is a function of @xmath80 and @xmath79 ( respectively momentum of the perturbed and perturbing plane - waves ) . furthermore @xmath107 where we introduced the function @xmath108 . in the following we will use the convenient dimensionless parameters @xmath109 in terms of the parameters @xmath110 , @xmath111 , the quantity @xmath112 entering eq . reads @xmath113 the carrier wave becomes modulationally unstable when the eigenfrequencies @xmath114 in eq . acquire a finite imaginary part : the condition for stability is then @xmath115 . a momentum @xmath80 is then _ modulationally stable _ if for each @xmath79 the eigenfrequencies @xmath114 are real , otherwise if it exists a @xmath79 such that @xmath116 then @xmath80 will be modulationally _ unstable_. since the eigenvalues are unaffected by substituting @xmath80 with @xmath117 and @xmath79 with @xmath118 , we will consider @xmath80 and @xmath79 both belonging to the interval @xmath119 $ ] . notice that for @xmath120 it is @xmath121 . when a momentum @xmath80 is modulationally unstable , there will be some momentum @xmath79 for which @xmath122 and @xmath123 have an imaginary part @xmath124 . for those values of @xmath80 and @xmath79 we write @xmath125 to quantify how unstable is the perturbed plane - wave [ notice that the imaginary parts of @xmath122 and @xmath123 are by definition opposite in sign and equal in modulus ] . for @xmath80 unstable we will use the notation @xmath126 , where the @xmath127 is taken on all the @xmath79 such that @xmath128 . the value of @xmath79 for which the maximum value of @xmath129 is obtained will be denoted by @xmath130 [ i.e. , @xmath131 . in this section we review the results and the region of stability for the short - range limit , having only local interaction ( @xmath132 ) and nearest - neighbor hopping : @xmath133 and @xmath134 for @xmath135 [ formally this is the limit @xmath82 in eq . ] . it has been shown in @xcite that the onset of mi occurs at @xmath136 , i.e. , the momenta @xmath137 are modulationally stable , while for @xmath138 are unstable . the quantity @xmath139 , defined in eq . and giving the stability regions reads @xmath140 one readily sees that the momenta @xmath137 are stable . for @xmath138 all the momenta @xmath80 are unstable , irrespective of @xmath141 . however , one sees that for @xmath138 and @xmath142 each @xmath79 is unstable , while for @xmath143 there are stability regions : these stable regions are found to be bounded by the regions in which @xmath79 is between @xmath144 and @xmath145 and @xmath80 is between @xmath146 and @xmath145 , where @xmath147 and @xmath148 the resulting plots of stable and unstable regions for @xmath142 and @xmath143 are in fig . [ fig1 ] , where we plot as well as contour plot the value of @xmath129 in the unstable regions ( the larger @xmath149 , the more unstable is the dynamics ) . in fig . [ fig2 ] we also plot as a solid line the values @xmath150 for which the maximum value of @xmath149 , at the fixed @xmath80 , is reached . regions for the short - range dnlse ( @xmath132 and only nearest - neighbor hoppings ) for @xmath151 ( a ) and @xmath152 ( b ) in the @xmath80 ( carrier wave momentum ) @xmath79 ( perturbing wave momentum ) plane . the absolute value of the imaginary part of the frequencies @xmath129 is depicted [ from light to dark as @xmath129 increases ] . the continuous lines are equispaced isolines whose spacing is set to @xmath153 . the thick ( red ) line indicates the position of @xmath130.,title="fig : " ] regions for the short - range dnlse ( @xmath132 and only nearest - neighbor hoppings ) for @xmath151 ( a ) and @xmath152 ( b ) in the @xmath80 ( carrier wave momentum ) @xmath79 ( perturbing wave momentum ) plane . the absolute value of the imaginary part of the frequencies @xmath129 is depicted [ from light to dark as @xmath129 increases ] . the continuous lines are equispaced isolines whose spacing is set to @xmath153 . the thick ( red ) line indicates the position of @xmath130.,title="fig : " ] in this section we consider the case of power - law interactions in the presence of ( local ) nearest - neighbor hoppings ( @xmath133 and @xmath134 for @xmath135 ) . one has then @xmath154 . \label{pr}\ ] ] we consider only the cases where @xmath155 and the local interaction @xmath156 can be either or positive or negative . the cases where @xmath157 is negative can be easily derived from the cases where @xmath155 by noticing that under the following transformation @xmath158 one obtains the same dependence on @xmath79 and @xmath0 of the stability conditions that one has when @xmath159 is positive . for example , if for a fixed value @xmath160 and @xmath161 one has that the momentum @xmath162 is stable , then momentum @xmath163 will also be stable for @xmath164 and @xmath165 . -@xmath79 plane ( see caption of fig . [ fig1 ] ) for power - law interactions in presence of local nearest - neighbor hoppings with @xmath168 and @xmath151 . we plot @xmath129 for the four values of @xmath169 , @xmath170 , @xmath65 , and @xmath171 in panels @xmath172 , @xmath173 , @xmath174 , and @xmath17 , respectively . , title="fig : " ] -@xmath79 plane ( see caption of fig . [ fig1 ] ) for power - law interactions in presence of local nearest - neighbor hoppings with @xmath168 and @xmath151 . we plot @xmath129 for the four values of @xmath169 , @xmath170 , @xmath65 , and @xmath171 in panels @xmath172 , @xmath173 , @xmath174 , and @xmath17 , respectively . , title="fig : " ] -@xmath79 plane ( see caption of fig . [ fig1 ] ) for power - law interactions in presence of local nearest - neighbor hoppings with @xmath168 and @xmath151 . we plot @xmath129 for the four values of @xmath169 , @xmath170 , @xmath65 , and @xmath171 in panels @xmath172 , @xmath173 , @xmath174 , and @xmath17 , respectively . , title="fig : " ] -@xmath79 plane ( see caption of fig . [ fig1 ] ) for power - law interactions in presence of local nearest - neighbor hoppings with @xmath168 and @xmath151 . we plot @xmath129 for the four values of @xmath169 , @xmath170 , @xmath65 , and @xmath171 in panels @xmath172 , @xmath173 , @xmath174 , and @xmath17 , respectively . , title="fig : " ] we consider here @xmath141 and @xmath159 to be positive . since @xmath175 for @xmath3 and @xmath176 $ ] , one can show the following : * for @xmath177 the momenta @xmath137 are stable . * for @xmath178 the momenta @xmath137 are unstable . it follows that the critical value @xmath146 as a function of @xmath111 is given by @xmath179 i.e. , for @xmath180 one has @xmath181 . for @xmath138 it is easy to see that all momenta are unstable against perturbations at @xmath120 . notice that for @xmath182 ( i.e. , for the model having only on - site and nearest - neighbor interactions ) eq . reads @xmath183 . the instability regions are depicted in fig . [ fig2 ] as @xmath159 is increased - for @xmath142 , one can identify four regions . for @xmath184 with @xmath185 , then the momenta @xmath80 larger than @xmath186 ( smaller than @xmath186 ) are unstable for all @xmath79 . increasing @xmath111 , one has that for @xmath187 a stable region forms around @xmath188 , while for @xmath189 an unstable region appears close to @xmath190 for momenta @xmath80 between @xmath146 given by eq . and @xmath186 . when @xmath191 , then all the @xmath80 becomes unstable and the instability starts from @xmath190 . therefore , the plane wave @xmath146 is rendered unstable by the wave vector @xmath190 , thus the system is the most sensitive to short wavelength perturbations . the behavior of @xmath146 is plotted in fig . [ fig3 ] . the analytical prediction ( valid for @xmath3 ) is compared for @xmath192 against numerical findings obtained numerically solving the dnlse finding a very good agreement [ for comparison we also plot the analytical result for @xmath182 ] . the numerical solution has been obtained on a finite - size system ( we choose @xmath194 ) whose coherence has been monitored by inspecting the absolute value of the following order parameter @xcite @xmath195 where @xmath196 is the fourier transform of the wave functions . the initial wavefunction @xmath197 is chosen as a plane wave with wave vector @xmath80 perturbed by the highest frequency wavevector @xmath198 , with @xmath190 ( notice that @xmath199 for non - competing interactions , as one can see from fig . [ fig2 ] ) . the ratio between the amplitudes of perturbing and perturbed wave functions is set to be @xmath200 with @xmath201 . as we can see in the left panel of fig . [ fig4 ] , if prepared in a modulationally unstable initial state , the system loses coherence after a time which diverges as we approach the momentum @xmath146 : denoting with @xmath202 the time after which the instability is observed using the order parameter , and noticing that the quantity is vanishing ( at @xmath190 ) as @xmath203 , one can estimate @xmath146 from the numerical data using the dependence @xmath204 ( see fig . [ fig4 ] , right panel ) . this divergence has been used to extract the numerical values of @xmath146 shown in fig . [ fig3 ] . vs @xmath111 from eq . ( in the figure @xmath205 and @xmath192 ) . the dots represent values obtained by direct simulation of the dnlse , see the text for details ( errors are smaller than the symbols ) . notice that for @xmath205 and @xmath182 one has @xmath136 for @xmath206 and @xmath181 for @xmath207 . the dotted line is the analytical value of @xmath146 vs @xmath111 from eq . with @xmath182 and @xmath205 . ] , defined in eq . , for five different values ( from left to right @xmath208 , @xmath209 , @xmath210 , @xmath211 , @xmath212 , @xmath213 in units of @xmath214 with @xmath194 ) of carrier plane - wave wavevector for the dnlse with parameters @xmath205 , @xmath215 , and @xmath192 . the lifetimes of the initial states are depicted in the right part of the figure . these times have been fitted with a function @xmath216 ( dotted line ) to obtain the a numerical estimates of @xmath146 ( vertical line ) . , title="fig : " ] , defined in eq . , for five different values ( from left to right @xmath208 , @xmath209 , @xmath210 , @xmath211 , @xmath212 , @xmath213 in units of @xmath214 with @xmath194 ) of carrier plane - wave wavevector for the dnlse with parameters @xmath205 , @xmath215 , and @xmath192 . the lifetimes of the initial states are depicted in the right part of the figure . these times have been fitted with a function @xmath216 ( dotted line ) to obtain the a numerical estimates of @xmath146 ( vertical line ) . , title="fig : " ] for @xmath218 a necessary condition for stability is given by @xmath219 that is obtained by analyzing the stability at @xmath120 . the critical momentum is given by @xmath220}\left[\arccos\left(\frac{\bar u + 2\bar v \ell_\alpha(q)}{2\sin^2(\frac{q}{2 } ) } \right)\right].\ ] ] all momenta @xmath221 are stable . an important feature of the case with competing interactions is that the most unstable perturbations can arise for a value of @xmath222 different from @xmath223 and @xmath145 ( i.e. , @xmath224 ) . in the following we determine for what conditions at the critical value @xmath146 there is an instability at a @xmath225 . we will refer to these values of @xmath222 as to _ finite _ values for the occurrence of modulational instability : this is because the instability develops on a length scale @xmath226 . it is intended that if the instability arises at @xmath190 , this develops on a length scale of the order the lattice unit , while an instability at @xmath120 involves length of the order of the lattice size : the later of these will be the case for non - local hoppings with weak - long - range exponents @xmath227 . an example of a finite @xmath222 is shown in fig . [ fig5 ] . to see how this can arise we analyze for simplicity the situation at @xmath228 , which can be easily generalized to a finite value in the interval @xmath229 . to observe an instability region like the one in fig . [ fig5 ] we have to impose that the equation @xmath230 with @xmath231 obtained from eq . by the substitution @xmath228 , has exactly one finite solution for fixed values of the parameters @xmath232 and no solutions for smaller values of @xmath156 . after some algebra one can derive a set of conditions on the coefficients @xmath233 such that the curve @xmath234 possesses one minimum . it turns out that , if @xmath235 , then there is a @xmath236 such that the instability region is tangent to the @xmath228 axis . for smaller values of @xmath156 the previous equation @xmath237 does not have any solution and one obtains a finite value of @xmath146 given by eq . . we can thus assert that the presence of competing interactions may originate the wavelength @xmath238 smaller than @xmath145 ( and larger than @xmath145 ) , unlike the case of the noncompeting interaction examined in the previous section for which the system is most sensible to perturbations at @xmath190 . this is a general feature of systems with competing interactions acting on different scales which , if properly tuned , give rise to the birth of a new intermediate lengthscale ( in this case of the order of @xmath239 ) . for similar phenomena ultimately leading to stripe formation and more generally spatially modulated patterns in different contexts see , e.g. , @xcite . ) for the case of competing interactions with @xmath240 , @xmath241 and @xmath242 . the value of the most unstable @xmath79 which defines @xmath146 is denoted by @xmath238 . ] for completeness we list the conditions on the parameters @xmath233 such that the function @xmath234 has a minimum for a finite value @xmath222 of the perturbing wavevector : * for @xmath243 @xmath244 * for @xmath245 @xmath246 * for @xmath247 @xmath248 ( similar results are found for @xmath249 ) . the value @xmath250 in eq . is given by @xmath251 , the unique solution of the equation : @xmath252 the condition ensuring the stability for every @xmath253 $ ] is given by : @xmath254 ) for the case of long - range hopping and local interaction @xmath205 and @xmath255 . panels ( a ) , ( b ) , ( c ) , and ( d ) refer to the values of @xmath256 , @xmath257 , @xmath65 , @xmath258.,title="fig : " ] ) for the case of long - range hopping and local interaction @xmath205 and @xmath255 . panels ( a ) , ( b ) , ( c ) , and ( d ) refer to the values of @xmath256 , @xmath257 , @xmath65 , @xmath258.,title="fig : " ] ) for the case of long - range hopping and local interaction @xmath205 and @xmath255 . panels ( a ) , ( b ) , ( c ) , and ( d ) refer to the values of @xmath256 , @xmath257 , @xmath65 , @xmath258.,title="fig : " ] ) for the case of long - range hopping and local interaction @xmath205 and @xmath255 . panels ( a ) , ( b ) , ( c ) , and ( d ) refer to the values of @xmath256 , @xmath257 , @xmath65 , @xmath258.,title="fig : " ] in this section we consider the case of power - law nearest - neighbor hoppings , with exponent @xmath259 . one has @xmath260 while in the cases considered in sec . [ mi_lr_interaction ] the instabilities arise in the higher frequency range of @xmath79 [ @xmath190 ] for non - competing interaction or at a finite value of @xmath79 [ @xmath262 for competing interactions , with non - local hoppings even the long - wavelength perturbations can affect significantly the stability properties of the system . this can be verified by an inspection of the behavior of @xmath139 for small @xmath79 . one finds for @xmath263 @xmath264 the investigation of the behavior of the second derivative of @xmath265 reveals that @xmath266 is positive for @xmath227 for each @xmath80 . it follows that @xmath267 i.e. , the modulational stability regions shrink to zero in the weak - long - range regime , irrespective of @xmath110 and @xmath111 ( we assume for simplicity in this section @xmath268 and @xmath269 ) . the same analysis shows that for @xmath132 the critical value @xmath146 does not depend on the specific value of @xmath110 for @xmath270 . the above scenario is confirmed by fig . [ fig6 ] where we plot @xmath129 as @xmath1 is increased . the values of @xmath146 as a function of @xmath1 are plotted in fig . notice that @xmath146 tends to @xmath186 when the hopping exponent approaches the short range limit @xmath271 . as was done in sec . [ mi_lr_interaction ] we compare the analytical results with numerical simulations of the dnlse obtaining a very good agreement . vs the power - law exponent @xmath1 of power - law hoppings with @xmath132 . crosses are numerical data . ] it should be stressed that the lifetimes of the modulationally unstable states in this case is typically longer than the ones encountered in sec . [ mi_lr_interaction ] . this will be supported in the next section where we directly compare instabilities arising from the long - range interaction and hopping , respectively . when we introduce the long - range interaction @xmath272 the instability at @xmath120 due to long - range hopping remains unchanged since it is due to the vanishing of @xmath273 [ see eq . ] , while the instabilities already discussed in sec . [ mi_lr_interaction ] are possibly generated . more precisely , an instability region at a finite value @xmath274 ( @xmath275 for non - competing interactions ) arises in the presence of @xmath159 , however , for @xmath159 much smaller than @xmath141 ( and smaller than a critical value @xmath276 ) then @xmath146 is determined only by the @xmath120 instability driven by non - local hoppings and it is again given by the value at @xmath81 , as shown in fig . when @xmath277 the contribution of both instabilities has to be taken into account to determine @xmath146 , with @xmath120 instabilities generated by long - range hopping and @xmath190 ( or @xmath278 if the competition is sufficiently strong ) instabilities due to the interaction . we comment on these scenarios in the next section . the hopping instability described in sec . [ mi_lr_hopping ] exhibits an important difference with the one arising from the interaction described in sec . [ mi_lr_interaction ] since @xmath146 becomes unstable for @xmath120 perturbations , i.e. , with a size of the order of the system s size . to compare the time scales on which these two kind of instabilities act we have considered a case where we switch on and off alternatively the long - range interaction and hopping ( see fig . [ fig8 ] ) . we prepare the two systems with a planewave @xmath80 slightly inside the instability region perturbed by a @xmath79 equaling @xmath145 and @xmath279 ( the widest available perturbation with a nonvanishing imaginary bogoliubov frequency ) . by inspecting the contour plots [ figs . [ fig8](a ) and ( b ) ] we foresee a longer lifetime for the instability induced by the non - local hoppings . this result is confirmed by numerical simulations , shown in panel @xmath174 of fig . [ fig8](c ) . this is a general feature of the long - range hopping instability : we observe that at the the critical value @xmath146 , by definition , @xmath280 both for non - local interaction and hoppings . however , entering the unstable region gives a vanishing value of @xmath129 for @xmath120 and @xmath281 and @xmath282 for @xmath79 small for long - range hoppings . at variance for long - range interactions as @xmath80 is slightly larger than @xmath146 then @xmath283 generally acquires a finite value at the wavevector @xmath284 at which the instability arises [ with @xmath285 for the non - competing case and @xmath286 for the competing one ] . ( see caption of fig . [ fig1 ] ) for @xmath287 and @xmath288 , @xmath289 ( long - range interaction ) and @xmath290 , @xmath291 ( long - range hopping ) . in panel ( c ) we depict the time evolution of the order parameter defined in eq . for the initial conditions indicated in the contour plots with points . as we can see the long - range hopping instability , upper ( green ) curve , takes a much longer time to set in than the long - range interaction instability , lower ( blue ) curve . , title="fig : " ] ( see caption of fig . [ fig1 ] ) for @xmath287 and @xmath288 , @xmath289 ( long - range interaction ) and @xmath290 , @xmath291 ( long - range hopping ) . in panel ( c ) we depict the time evolution of the order parameter defined in eq . for the initial conditions indicated in the contour plots with points . as we can see the long - range hopping instability , upper ( green ) curve , takes a much longer time to set in than the long - range interaction instability , lower ( blue ) curve . , title="fig : " ] ( see caption of fig . [ fig1 ] ) for @xmath287 and @xmath288 , @xmath289 ( long - range interaction ) and @xmath290 , @xmath291 ( long - range hopping ) . in panel ( c ) we depict the time evolution of the order parameter defined in eq . for the initial conditions indicated in the contour plots with points . as we can see the long - range hopping instability , upper ( green ) curve , takes a much longer time to set in than the long - range interaction instability , lower ( blue ) curve . , title="fig : " ] let us examine the peculiarities of the instabilities due to interaction and hopping and examine whether these may be reproduced with finite - range couplings or hoppings . as far as the long - range interaction is concerned , in the non competing case ( examined in sec . [ mi_lr_noncompeting ] ) , we observe that the situation is not very different from the finite - range one with nearest - neighbor interaction ( @xmath292 ) . we can indeed find an effective nearest - neighbor @xmath293 interaction giving rise to the same @xmath146 [ see eq . ] : @xmath294 similarly in the case where the competition in present ( described in sec . [ mi_lr_competing ] ) the case with @xmath0 finite can be seen to be similar to the one obtained with @xmath295 . in fact one can generate a new wavevector @xmath238 even when @xmath295 , and thus the long - ranged - ness of the interaction is not actually playing a major role . as an example in fig . [ fig9](a ) we have considered a case where a nearest - neighbor and next - to - nearest - neighbor interaction generates a stability diagram closely resembling the one shown in fig . [ fig5 ] . when we move to the hopping instabilities ( examined in sec . [ mi_lr_hopping ] ) we observe that while a finite - range hopping can generate @xmath120 instabilities for @xmath296 , the value of @xmath146 is effectively limited by the range of the hopping and it is not @xmath223 . this can be seen considering the hopping coefficients of range @xmath297 , i.e. , @xmath298 for @xmath299 . an example is given by @xmath300 where we set @xmath301 , @xmath302 . explicit expressions ( not reported here ) for the stability regions can be derived repeating the analysis presented in sec . [ mi_analysis ] . it is possible to show that for a hopping of range @xmath297 then the critical momentum @xmath146 can become as small as @xmath303 due to the @xmath120 instability . thus for @xmath270 one can find a finite - range hopping with range @xmath297 to reproduce the critical value @xmath146 and the stability regions of non - local hoppings with exponent @xmath1 , while this is not the case for @xmath227 . we conclude that the case @xmath227 where @xmath181 is singled out as a case where the instability is _ genuinely _ due to the long - range nature of the hoppings . in fig . [ fig9](b ) we depict the stability diagram of a case with the nearest- and next - to - nearest - neighbor hoppings mimicking the effect of long - range hopping shown in fig . [ fig6](a ) and [ fig6](b ) , with a finite value of @xmath146 . ( see caption of fig . [ fig1 ] ) for models including only on - site interactions ( @xmath141 ) and next- and next - to - nearest - neighbor interaction ( @xmath159 and @xmath304 ) and hoppings ( @xmath305 and @xmath306 ) , respectively . panel ( a ) refers to values @xmath307 , @xmath308 , @xmath309 , and @xmath310 while panel ( b ) refers to values @xmath311 , @xmath312 , and @xmath313 . , title="fig : " ] ( see caption of fig . [ fig1 ] ) for models including only on - site interactions ( @xmath141 ) and next- and next - to - nearest - neighbor interaction ( @xmath159 and @xmath304 ) and hoppings ( @xmath305 and @xmath306 ) , respectively . panel ( a ) refers to values @xmath307 , @xmath308 , @xmath309 , and @xmath310 while panel ( b ) refers to values @xmath311 , @xmath312 , and @xmath313 . , title="fig : " ] the cases we have analyzed can be applied to the study of modulational instability in a variety of physical systems characterized by non - local interaction and hopping coefficients . for example , the excitation transfer energy in molecular crystals and biopolymers exhibits a @xmath11 decay due to the dipole - dipole interaction @xcite . the possibility of the inclusion of a non - local hopping is also present in dna modeling ( e.g. , in the the peyrard - bishop model @xcite ) , where the equation of motion for the transverse stretching of the hydrogen bonds connecting the bases facing each other contains a long - range hopping term due to the dipole - dipole interaction among hydrogen bonds resulting in a value @xmath314 @xcite . another physical system characterized by non - local interactions is provided by ultracold atomic dipolar gases in optical lattices @xcite , for which experimental results recently appeared @xcite . for dipolar bosons as @xmath315 in the bose - einstein condensate phase the low - energy effective hamiltonian is expected to be the dnlse in deep optical lattices with the power - law interaction coefficients having @xmath192 . we also mention that for ultracold bosons in suitably tailored optical lattices it is possible to have non - local hoppings @xmath316 , @xmath317 @xcite . to be specific , in the following we study the instability threshold for the case of @xmath315 trapped bosons in a quasi - one - dimensional geometry . we provide estimates of the dnlse parameters @xmath318 , @xmath141 , @xmath159 in terms of the physical parameters , showing that the critical value @xmath146 does not crucially depend on the exponent @xmath0 . a similar computation is presented for a model having non - local hoppings @xmath316 , @xmath317 , as the one described in @xcite . also in this case we find a relatively small quantitative effect on the value of the critical value @xmath146 . as discussed in secs . [ mi_lr_interaction ] to [ mi_interaction_hopping_comparison ] , there is no qualitative difference between a finite value of @xmath0 or @xmath1 and the corresponding results for @xmath182 or @xmath82 , as soon as that @xmath3 or @xmath270 . the results presented in this section further show that the quantitative difference ( say , comparing @xmath192 and @xmath182 results ) is rather small . the gross - pitaevskii hamiltonian for a bose - einstein condensate of dipolar bosons in a quasi - one - dimensional trap in presence of an optical lattice is given by @xmath319 @xmath320 in eq . @xmath321 is the condensate wavefunction and @xmath322 is the periodic potential due to the optical lattice along the @xmath323 direction , reading as @xmath324 where @xmath325 ( @xmath326 is the lattice spacing ) . the strength @xmath327 is usually measured in units of the recoil energy @xmath328 : we set @xmath329 . the condensate wavefunction is normalized to the total number of particles @xmath330 and the filling @xmath76 is defined by @xmath331 , where @xmath332 is the number of wells . moreover , in eq . @xmath333 is the effective one - dimensional coupling constant , which is determined in terms of the 3d @xmath334-wave scattering length @xmath172 and the size of the transverse confinement @xmath335 @xcite , which in turn depends on the radial confinement frequency @xmath336 ( we will consider values of @xmath337 much smaller than one , far from the confinement - induced resonance ) . the dipole - dipole interaction induces a non - local two - body potential decaying as @xmath11 . the dipole - dipole scattering length is defined as @xmath338 , where @xmath17 is the dipole moment and @xmath339 is the vacuum permeability @xcite . for dipolar gases in quasi - one - dimensional geometries it is possible to obtain an expression in the single - mode approximation for the effective non - local interaction @xmath340 by integrating the transverse ground - state @xcite : one obtains @xmath341 $ ] , where @xmath342 and @xmath343 is the complementary error function ; the dimensionless constant @xmath344 depends on the angle @xmath345 the dipoles form with the @xmath323 axis and it may vary between @xmath64 ( corresponding to @xmath346 ) and @xmath347 ( @xmath348 ) . the dnlse is found for deep lattices by using a tight - binding ansatz @xcite of the form @xmath349 where @xmath350 is the wannier function centered in the @xmath351-th well ( and assumed in the following estimates to have a shape independent of @xmath351 ) . the coefficients @xmath318 , @xmath141 and @xmath159 are expressed as suitable overlap integrals of wannier functions and can be estimated by using a gaussian form for the @xmath352 s , with the width being a parameter to be variationally determined ( see the discussions in @xcite ) . we mention that the effect of the inter - site interaction term @xmath159 in the bose - hubbard phase diagram was recently investigated in @xcite . the critical momentum @xmath353 is an important quantity which has been theoretically investigated in @xcite and experimentally detected in @xcite for a system with short - range interactions . we recall that experimentally the modulational instability may be triggered by subjecting the system to a sudden shift of the optical lattice or of confining harmonic trap ( as done in @xcite ) . + our results for the critical value @xmath146 , after computing the parameters @xmath318 , @xmath141 , and @xmath159 , are drawn in fig . [ fig10 ] for a set of typical experimental parameters : we fixed @xmath354 , @xmath355 , @xmath356 ( corresponding to repulsion ) and @xmath357 and we varied @xmath358 for @xmath315 atoms for two different values of the filling @xmath76 ( we choose @xmath201 and @xmath359 ) . in the inset of fig . [ fig10 ] we plot the ratio @xmath360 versus @xmath358 to quantify how much the interaction is non - local for a typical value of the parameters . one sees that deviations are observed from the critical value @xmath136 obtained without non - local interactions and these deviations are not crucially dependent on @xmath0 . in fig . [ fig11 ] we finally plot the critical value @xmath146 for a model having @xmath316 and @xmath317 ( i.e. , @xmath300 ) with @xmath81 : the critical value @xmath146 does not depend on @xmath141 and smoothly passes from @xmath186 ( for @xmath310 ) to @xmath361 ( for @xmath362 ) . vs @xmath358 for a dipolar gas of @xmath315 atoms . solid lines refer to @xmath192 ( top solid line : @xmath201 , bottom solid line : @xmath359 ) , while dotted lines refer to @xmath182 ( top dashed line : @xmath201 , bottom dashed line : @xmath359 ) . parameters are @xmath354 , @xmath355 , @xmath356 and @xmath363 . inset : corresponding value of @xmath364 vs @xmath358 for @xmath201 , solid line , and @xmath359 , dotted line ( the two lines practically coinciding with one another ) . ] vs @xmath365 for a model having nearest - neighbor and next - nearest - neighbor hoppings @xmath316 and @xmath317 ( @xmath81 ) . ] in this paper we studied the occurrence of modulational instabilities in nonlinear lattices with long - range hoppings and interactions . we were motivated by experiments of ( di)polar gases in optical lattices and by the interest in the study of dynamical regimes in systems with long - range interactions . using the discrete nonlinear schrdinger equation in one dimension , we considered power - law decaying interactions ( with exponent @xmath0 ) and hoppings ( with exponent @xmath1 ) . we showed that the effect of long - range interactions is that of shifting the onset of the modulational instability region for @xmath3 ( corresponding to an extensive energy ) . at a critical value of the interaction strength , the modulational stable region shrinks to zero . similar results are found for short - range non - local hoppings ( @xmath270 ) . at variance , for longer - ranged hoppings ( @xmath227 ) there is no longer any modulational stability . explicit estimates for the critical values of the momentum at which the system becomes unstable are presented for a quasi - one - dimensional ultracold dipolar gas in a deep optical lattice . instabilities due to the interaction generally differ from the ones due to hopping since the first ones are sensitive to finite wavelength perturbations while the second ones are most sensitive to perturbation of the order of the system size . such hopping generated instability turns out to have generally longer timescales than the interaction generated ones . if we allow interactions acting on different scales to compete we may generate instabilities with longer , but finite , wavelengths , in analogy with what is met in other systems with competing interactions @xcite . the instabilities met in the long - range interacting and long - range hopping for @xmath270 are not specific to the long - ranged nature of the interaction or hopping . in fact , their effects are , in principle , not different from finite range cases . as far as very long - ranged ( @xmath227 ) hopping is concerned , we found that it gives rise to _ genuinely _ long - range instabilities , since they can not be reproduced with suitably chosen finite - range hoppings . _ acknowledgements _ we thank m. iazzi , a. smerzi , g. de ninno and f. staniscia for very useful discussions . the analysis of the stability regions presented in the main text is based on the study of the quantity @xmath139 defined in , which in turn contains the function @xmath367 defined in eq . : @xmath368 ( with @xmath3 ) . we are interested in the domain @xmath176 $ ] . from the definition it follows that @xmath369 ; it is also @xmath370 the plot of @xmath371 and @xmath372 as a function of @xmath0 is drawn in fig . [ app_1 ] . notice that the behavior of @xmath371 for @xmath53 and @xmath182 is given respectively by @xmath373 and @xmath374 . for @xmath372 one has @xmath375 and @xmath376 . the second derivative of @xmath382 can be computed explicitly and it gives : @xmath383 for @xmath384 then @xmath385 is a positive function and we have @xmath386 , constant over @xmath387 $ ] . the second derivative takes the following values at the extrema of the interval @xmath388 $ ] for @xmath3 : @xmath389 finally at @xmath190 we have @xmath390 for every @xmath3 .

the nucleon s second spin dependent structure function @xmath0 is of considerable phenomenological interest . the most important theoretical tool for its analysis is the operator product expansion ( ope ) @xcite . in leading order of @xmath1 , @xmath0 receives contributions from both twist two and twist three operators . it thus offers the unique possibility of directly assessing higher twist effects . the twist three operator probes the transverse momentum distribution of the quarks in the nucleon , and has no simple parton model interpretation . in leading order of @xmath1 , and for massless quarks , the moments of @xmath0 are given by @xmath3 \end{split } \label{mom}\ ] ] for even @xmath4 in the flavor nonsinglet sector . here @xmath5 runs over the light quark flavors . the reduced matrix elements @xmath6 and @xmath7 , taken in a nucleon state with momentum @xmath8 and spin vector @xmath9 , are defined by @xcite @xmath10 \ , , \label{twist2 } \\ \langle p , s| { { \mathcal { o}}}^{5 ( f ) } _ { [ \sigma \ { \mu_1 ] \cdots \mu_n \ } } | p , s \rangle & = & \frac{1}{n+1}d_n^{(f ) } \ , [ ( s_\sigma p_{\mu_1 } - s_{\mu_1 } p_\sigma ) p_{\mu_2}\cdots p_{\mu_n } + \cdots -\mbox{traces } ] \ , , \label{twist3 } \\ { { \mathcal { o}}}^{5 ( f)}_{\sigma\mu_1\cdots\mu_n } & = & \left(\frac{\mathrm i}{2}\right)^n\bar{\psi}\gamma_{\sigma } \gamma_5 { \overset{\leftrightarrow}{d}_{\mu_1 } } \cdots { \overset{\leftrightarrow}{d}_{\mu_n } } \psi -\mbox{traces } \,.\end{aligned}\ ] ] here @xmath11 denotes the renormalization scale . the wilson coefficients @xmath12 , @xmath13 depend on the ratio of scales @xmath14 and on the running coupling constant @xmath15 . the tree level values of the wilson coefficients for electroproduction are given by the quark charges @xmath16 : @xmath17 the symbol @xmath18 ( @xmath19 $ ] ) indicates symmetrization ( antisymmetrization ) with @xmath20 the operator ( [ twist2 ] ) has twist two , whereas the operator ( [ twist3 ] ) has twist three . the twist two contribution in ( [ mom ] ) is also known as the wandzura wilczek contribution @xcite . note for comparison that in leading order of @xmath1 the moments of @xmath21 are given by the twist two matrix elements @xmath22 : @xmath23 both the wilson coefficients and the operators are renormalized at the scale @xmath11 . it is assumed that the wilson coefficients can be computed perturbatively . the reduced matrix elements @xmath22 and @xmath24 , on the other hand , are non perturbative quantities and hence a problem for the lattice . ( note that some authors use a different definition of @xmath25 and @xmath26 , e.g. the values given in refs . @xcite have to be multiplied by 2 to agree with our conventions . ) in the following we shall drop the flavor indices , unless they are necessary . a few years ago we computed the lowest non trivial moment of @xmath0 on the lattice @xcite . this calculation splits into two separate tasks . the first task is to compute the nucleon matrix elements of the appropriate lattice operators . this was described in detail in @xcite . the second task is to renormalize the operators . renormalization effects are a major source of systematic error . an essential feature of our previous calculation was that the renormalization was done in perturbation theory and hence mixing with lower dimensional operators could not be taken into account . in that approach the twist three contribution turned out to be the dominant contribution to both the proton and the neutron structure functions . this result has been recently confirmed by dolgov et al.@xcite . in the meantime , it has become possible to study renormalization non perturbatively on the lattice , see e.g. @xcite . this approach allows us to consider mixing with lower dimensional operators . if present , it will be the dominant mixing effect in the continuum limit . since the twist three operators ( [ twist3 ] ) can suffer from such mixing , we shall extend our previous work by employing non perturbative renormalization . in a recent paper @xcite we have started a non perturbative calculation of the renormalization constants associated with the structure functions @xmath27 , @xmath28 and @xmath21 in the flavor nonsinglet sector . here we consider the case of the structure function @xmath0 restricting ourselves to @xmath29 , the lowest moment of @xmath0 for which the ope makes a statement . a preliminary version of this work based on lower statistics at a single value of the bare coupling has already been presented in ref . the renormalization of the operators which contribute to the moments of @xmath0 has been studied by several authors in continuum perturbation theory @xcite . since the more recent paper by kodaira et al . @xcite ( see also @xcite ) is closest to the methods applied on the lattice , let us briefly recapitulate the main findings of these authors . they consider the case @xmath30 in the flavor nonsinglet sector and start from the operators @xmath31 - \mbox{traces } \ , , \\ r_1^{\sigma \mu \nu } & = & \frac{1}{12 } g \left [ \epsilon^{\sigma \mu \alpha \beta } \bar{\psi } f_{\alpha \beta } \gamma^{\nu } \psi + \epsilon^{\sigma \nu \alpha \beta } \bar{\psi } f_{\alpha \beta } \gamma^{\mu } \psi \right ] - \mbox{traces } \ , , \\ r_m^{\sigma \mu \nu } & = & \mbox{i } m \bar{\psi } \gamma_5 \gamma^\sigma d^{\ { \mu } \gamma^{\nu \ } } \psi - \mbox{traces } \ , , \\ r_{\mathrm { eq}}^{\sigma \mu \nu } & = & \frac{\mbox{i}}{3 } \left [ \bar{\psi } \gamma_5 \gamma^\sigma d^{\ { \mu } \gamma^{\nu \ } } ( \mbox{i } { \not\!\!{d}}- m ) \psi + \bar{\psi } ( \mbox{i } { \not\!\!{d}}- m ) \gamma_5 \gamma^\sigma d^{\ { \mu } \gamma^{\nu \ } } \psi \right ] - \mbox{traces } \,.\end{aligned}\ ] ] here @xmath32 denotes the gluon field strength tensor , which could alternatively be expressed as a commutator of two covariant derivatives . due to the relation @xmath33 it is possible to eliminate one of the above operators . a one loop calculation of the quark quark gluon three point functions with a single insertion of each of these operators reveals the necessity of taking one more operator into account in the process of renormalization , namely the gauge variant operator @xmath34 - \mbox{traces } \,.\ ] ] of course , in physical matrix elements neither @xmath35 nor @xmath36 will contribute . they show up , however , in off shell vertex functions and influence the renormalization factors . kodaira et al . choose @xmath37 and @xmath38 as the physical operators . in the chiral limit @xmath39 @xmath38 is neglected , and they obtain for the scale dependence of the twist three piece @xmath40 where for @xmath41 colors and @xmath42 flavors @xmath43 in agreement with earlier calculations . using @xmath44 and @xmath37 as the physical operators , one finds that in the large@xmath41 limit the operator @xmath44 dominates the renormalization group evolution of the nucleon matrix elements . this has been shown by ali , braun , and hiller @xcite , and was rederived in the present framework by sasaki @xcite . in a lattice calculation , the first step is the analytic continuation to imaginary times , leading from the physical minkowski space to euclidean space . from now on , all expressions are written for the euclidean case ( for the details of our conventions see appendix a of ref . hence we have to study operators of the form @xmath45 we shall neglect quark masses , i.e. we consider only the chiral limit . in our earlier work @xcite we have computed the renormalization constants in perturbation theory to one loop order . however , it is believed that perturbation theory can not give reliable values for the mixing with lower dimensional operators because non perturbative effects are expected to be important . for a multiplicatively renormalizable operator , i.e. in the absence of mixing , we can write @xmath46 where @xmath47 is the lattice spacing . the renormalization constant @xmath48 is fixed by a suitable condition . as in the continuum , we impose the ( mom like ) renormalization condition @xmath49 on the corresponding quark quark vertex function in the landau gauge . here @xmath50 denotes the born or tree level contribution to the vertex function . the renormalized vertex function @xmath51 and its bare precursor @xmath52 are related by multiplicative renormalization : @xmath53 where @xmath54 is the quark wave function renormalization constant defined as in ref . @xcite . as before @xcite , we give the nucleon a momentum in the 1direction and choose the polarization in the 2direction . with these choices we use the operator @xmath55 for the twist two matrix element @xmath56 . it belongs to the representation @xmath57 of the hypercubic group @xmath58 @xcite and this property protects it from mixing with operators of equal or lower dimension . hence it is multiplicatively renormalizable , and the operator renormalized at the scale @xmath11 is written as @xmath59 as the operator for the twist three matrix element @xmath2 we take @xmath60 4\ } } & = & { \mbox{\small $ \frac{1}{3}$}}\left ( 2 { { \mathcal { o}}}^5_{2\{14\ } } - { { \mathcal { o}}}^5_{1\{24\ } } - { { \mathcal { o}}}^5_{4\{12\ } } \right ) \nonumber \\ & = & \mbox{\small $ \frac{1}{12}$ } \bar{\psi}\big(\gamma_2 { \overset{\leftrightarrow}{d}_{1 } } { \overset{\leftrightarrow}{d}_{4 } } + \gamma_2 { \overset{\leftrightarrow}{d}_{4 } } { \overset{\leftrightarrow}{d}_{1 } } - { \mbox{\small $ \frac{1}{2}$}}\gamma_1 { \overset{\leftrightarrow}{d}_{2 } } { \overset{\leftrightarrow}{d}_{4 } } - { \mbox{\small $ \frac{1}{2}$}}\gamma_1 { \overset{\leftrightarrow}{d}_{4 } } { \overset{\leftrightarrow}{d}_{2 } } \nonumber \\ & & \quad \quad \quad - { \mbox{\small $ \frac{1}{2}$}}\gamma_4 { \overset{\leftrightarrow}{d}_{1 } } { \overset{\leftrightarrow}{d}_{2 } } - { \mbox{\small $ \frac{1}{2}$}}\gamma_4 { \overset{\leftrightarrow}{d}_{2 } } { \overset{\leftrightarrow}{d}_{1}}\big ) \gamma_5 \psi \nonumber \\ & = : & { { \mathcal { o}}}^{[5 ] } \ , , \label{o5}\end{aligned}\ ] ] which belongs to the representation @xmath61 of @xmath58 . the operator ( [ o5 ] ) has dimension five and @xmath62parity @xmath63 and is the euclidean counterpart of the minkowski operator @xmath44 . it turns out that there exist two more operators of dimension four and five , respectively , transforming identically under @xmath58 and having the same @xmath62parity , with which ( [ o5 ] ) can mix : @xmath64 @xmath65 we use the definition @xmath66 $ ] . the operator @xmath67 is the euclidean analog of @xmath37 with the field strength replaced by a commutator of two covariant derivatives , and @xmath68 corresponds to @xmath38 . in continuum perturbation theory @xmath38 can be neglected in the chiral limit . on the lattice , the explicit breaking of chiral symmetry induced by wilson type fermions , which we shall use , persists even when the quarks are massless . for dimensional reasons , we expect that @xmath68 contributes with a coefficient @xmath69 and hence has to be kept . the operator @xmath67 , on the other hand , being of the same dimension as @xmath70}$ ] , mixes with a coefficient of order @xmath71 , which should be small . therefore we discard @xmath67 as well as possible lattice counterparts of @xmath35 and @xmath36 , which are also of dimension five and hence are also multiplied by a factor of order @xmath71 . the above mentioned observation @xcite that @xmath44 dominates over @xmath37 in the renormalization group evolution as @xmath72 may be taken as another indication that neglecting @xmath67 is not unreasonable . however , this dominance holds only in physical matrix elements and does not apply to the mixing with the operators @xmath35 and @xmath36 . so we make the following ansatz for @xmath70}$ ] renormalized at the scale @xmath11 : @xmath73}_{\mathrm r}(\mu ) = z^{[5]}(a\mu ) { { \mathcal { o}}}^{[5]}(a ) + \frac{1}{a } z^\sigma(a\mu ) { { \mathcal { o}}}^\sigma(a ) \ , . \label{renorm}\ ] ] the renormalization constant @xmath74}$ ] and the mixing coefficient @xmath75 are determined from the conditions @xmath76}_{\mathrm r } ( p ) \gamma^{[5]}_{\mathrm { born } } ( p ) ^\dagger \right ) & \underset{p^2 = \mu^2}{= } & { \rm tr } \left ( \gamma^{[5]}_{\mathrm { born } } ( p ) \gamma^{[5]}_{\mathrm { born } } ( p ) ^\dagger \right ) \ , , \\ { \rm tr } \left ( \gamma^{[5]}_{\mathrm r } ( p ) \gamma^\sigma_{\mathrm { born } } ( p ) ^\dagger \right ) & \underset{p^2 = \mu^2}{= } & { \rm tr } \left ( \gamma^{[5]}_{\mathrm { born } } ( p ) \gamma^\sigma_{\mathrm { born } } ( p ) ^\dagger \right ) = 0 \,,\end{aligned}\ ] ] which are straightforward generalizations of eq . ( [ mom1 ] ) . note that the operator ( [ o0 ] ) vanishes in the born approximation between quark states , which is another reason why we do not take it into account . rewriting eq . ( [ renorm ] ) as @xmath73}_{\mathrm r } ( \mu ) = z^{[5]}(a\mu ) \left ( { { \mathcal { o}}}^{[5]}(a ) + \frac{1}{a } \frac{z^\sigma(a\mu)}{z^{[5]}(a\mu ) } { { \mathcal { o}}}^\sigma(a ) \right)\ ] ] we see that @xmath70}(\mu)$ ] will have a multiplicative dependence on @xmath11 ( cf . ( [ op1 ] ) ) only if the ratio @xmath77}(a\mu)$ ] does not depend on @xmath11 . the scale dependence will then completely reside in @xmath74}$ ] . we have obtained numerical results for matrix elements and @xmath78 factors in quenched simulations at @xmath79 , 6.2 , and 6.4 ( @xmath80 bare coupling constant on the lattice ) . whereas our original calculation @xcite at @xmath81 used wilson fermions , we have meanwhile switched to non perturbatively improved fermions ( clover fermions ) in order to reduce @xmath82 effects . the value of the clover coefficient @xmath83 is taken from ref . @xcite . since we have not improved the operators , there will still be residual @xmath82 effects in the matrix elements and the renormalization factors . a few details of our computations are collected in table [ param ] . the matrix elements are calculated from several hundred configurations for each @xmath84 . to compute the renormalization factors we use a momentum source @xcite . therefore the statistical error is @xmath85 for @xmath86 configurations on a lattice of volume @xmath87 , and we already get small statistical uncertainties even from a small number of configurations , four in our case . ( there is of course a price to be paid , the calculation for each momentum is independent , so the number of inversions of the fermion matrix is proportional to the number of momentum values . ) the main source of statistical uncertainty in our final results is from the matrix elements , not the @xmath78 values . the momenta in the vertex functions used for the evaluation of the renormalization factors have been chosen close to the diagonal in the brillouin zone in order to keep cut off effects as small as possible . one should bear in mind that this diagonal extends up to @xmath88 , but we use only momenta with @xmath89 . in each case , the calculations are done at three ( or more ) values of the hopping parameter @xmath90 determining the bare quark mass so that we can extrapolate our results ( both the bare matrix elements and renormalization factors ) to the chiral limit . the extrapolation is performed linearly in @xmath91 , the square of the pion mass . the bare reduced matrix elements are calculated from three point functions in the standard fashion ( see , e.g. , ref . @xcite ) . in eq.(25 ) of ref . @xcite the ratios of three to two point functions for the @xmath56 operator @xmath92 and the @xmath2 operator @xmath70}$ ] are given . for the operators @xmath67 and @xmath68 the ratios and ratio factors are the same as for the @xmath2 operator . the matrix elements are collated in tables [ rmeclo60 ] , [ rmeclo62 ] , [ rmeclo64 ] separately for @xmath93 and @xmath94 quarks in the proton . here @xmath95}$ ] and @xmath96 correspond to the operators ( [ o5 ] ) and ( [ osigma ] ) , respectively . in addition we give the pion masses ( in lattice units ) which we use in the chiral extrapolations . they are mostly taken from ref . note that all our errors are purely statistical . they were determined by the jackknife procedure . let us begin the more detailed presentation of our numerical results with the renormalization factor @xmath97 of the multiplicatively renormalizable @xmath56 operator ( [ oa2 ] ) . we convert our mom numbers to the @xmath98 scheme using 1loop continuum perturbation theory as described in @xcite . in fig . [ fig.za2 ] we show the @xmath11 dependence of @xmath99 extrapolated to the chiral limit . results for wilson fermions can be found in ref . note that at scales @xmath100 exceeding a few times the lattice cut off @xmath101 strong lattice artifacts may be present so that the corresponding results should not be taken too seriously . turning to the more subtle renormalization of the @xmath2 operator ( [ o5 ] ) we must note that the conversion factor from our mom scheme to the @xmath98 scheme has not yet been calculated because of the complications caused by the mixing effects . therefore we stick to the mom numbers . let us first consider the ratio @xmath77}(a\mu)$ ] . as discussed above , @xmath77}(a\mu)$ ] should be independent of the renormalization scale @xmath11 if the renormalized operator is to depend on @xmath11 multiplicatively . in fig . [ fig.zrat ] we show this ratio for our three @xmath84 values . it becomes approximately flat for scales @xmath11 larger than about 3.5 gev . while a scale of 3.5 gev might seem to be somewhat too close to the cut off for @xmath81 and perhaps also for @xmath102 , it enters the region where lattice artifacts die out in the case of @xmath103 . therefore we feel encouraged to apply the @xmath78 factors around @xmath104 in order to evaluate structure function moments in the next section . in fig . [ fig.z5 ] we plot @xmath74}(a\mu)$ ] in order to show the size of the multiplicative renormalization in our approach . let us now discuss our nucleon matrix elements . in figs . [ fig.chiex.a2u ] , [ fig.chiex.d25u ] we show the chiral extrapolations of the bare values of @xmath105 and @xmath106\,(u ) } $ ] , respectively . unfortunately , like in all other current qcd simulations our quark masses are rather large so that the extrapolation has to bridge quite some gap . a striking feature of the data is that the bare @xmath105 values at @xmath102 are rather different from the values at the other two @xmath84 s . we can interpret this only as an unpleasantly large statistical fluctuation . the bare matrix elements in the chiral limit will be combined with the renormalization factors of the preceding section to yield estimates of the renormalized matrix elements . we start with the twist two matrix element @xmath56 . in the @xmath98 scheme with anticommuting @xmath107 the corresponding wilson coefficient is given by ( see , e.g. , ref . @xcite ) @xmath108 \end{array}\ ] ] with @xmath109 , @xmath110 , @xmath111 , @xmath112 . ( these are the numbers for @xmath113 flavors , appropriate for the quenched approximation . ) the renormalized reduced matrix element is obtained from the bare value ( extrapolated to the chiral limit ) after multiplication with the non perturbative renormalization factor converted to the @xmath98 scheme . from eq . ( [ g1moments ] ) we can then calculate @xmath114 . to avoid large logarithms in the wilson coefficient we put @xmath115 . in fig . [ fig.g1 ] we show the results for the proton and compare with the experimental value @xcite . while our lattice results at @xmath81 agree surprisingly well with the experimental number , the above mentioned fluctuation makes them considerably larger at @xmath102 . fortunately , they drop again at @xmath103 . for the neutron there is a similar effect , but due to the larger errors it is less significant . let us now turn to our results for the twist3 matrix elements . in this case it is unclear how to convert our mom results to the @xmath98 scheme due to the mixing effects . therefore we do not make use of the @xmath98 wilson coefficient , which has recently been calculated @xcite ( with the t hooft veltman @xmath107 ) . it would change the final results for @xmath2 by @xmath116 . instead we use only the lowest order approximation for the coefficient functions , i.e. the tree level coefficients ( [ wilcoeff ] ) , which are the same in all schemes , and define ( by a slight abuse of notation ) @xmath117 for the proton and the neutron , respectively . the renormalized values of @xmath118 for @xmath119 in the proton are calculated from @xmath120 } d_2^{[5]\,(f ) } + \frac{1}{a } z^\sigma d_2^{\sigma \,(f ) } \,.\ ] ] remember that , besides the twist three matrix element @xmath2 , @xmath121 also contains a twist two piece , the wandzura wilczek contribution , see eq . ( [ mom ] ) . to be consistent we restrict ourselves to the tree level wilson coefficients and the mom matrix elements also in this contribution when computing @xmath121 from ( cf . ( [ mom ] ) and ( [ g1moments ] ) ) @xmath122 the moment @xmath121 is plotted in fig . [ fig.g2 ] for the proton , where we have again identified @xmath115 . the experimental value is obtained by combining @xmath114 from ref . @xcite with @xmath2 from ref . again we see the effect of the `` fluctuation '' at @xmath102 . comparing the proton results shown in fig . [ fig.g2 ] with the numbers presented in fig . [ fig.g1 ] one sees that @xmath121 is dominated by the twist two operator . there is little room left for the twist three operator , and one obtains rather small values for @xmath2 as shown in fig . [ fig.d2 ] for the proton . in the neutron , @xmath2 is even smaller in magnitude and hardly different from zero within the statistical errors . in tables [ res5 ] and [ res10 ] we present results at the scales 5 gev@xmath123 and 10 gev@xmath123 , respectively . note that here @xmath114 includes the one loop wilson coefficient as well as the conversion factor to the @xmath98 scheme , both of which were neglected in the calculation of @xmath121 for the reason explained above . the difference between the two sides of eq . ( [ x2g2 ] ) when evaluated with the numbers taken from the tables gives therefore an impression of the uncertainties originating from our incomplete knowledge of the perturbative corrections . in figs . [ fig.contlim.g1 ] , [ fig.contlim.g2 ] , [ fig.contlim.d2 ] we fix the scale at 5 gev@xmath123 and plot our results for the proton as well as for the neutron versus the lattice spacing @xmath47 . although an extrapolation to the continuum limit appears to be problematic , it is reassuring to see that we are getting close to the experimental numbers shown at @xmath124 . of course , we should not forget that our computation suffers from various uncertainties . apart from the fact that our treatment of the operator mixing is still incomplete , these concern , e.g. , the influence of the quenched approximation , the extrapolation to the chiral limit , and the size of the lattice artifacts . sea quark effects are expected to be concentrated at small @xmath125 , hence they should be suppressed by the factor @xmath126 in the moment which we have considered . therefore we may hope that the quenched approximation is reasonable in the case at hand . if indeed the valence quarks dominate , then it should also be justified to neglect flavor singlet contributions ( like disconnected insertions and pure gluon operators ) , and it makes sense to consider proton and neutron matrix elements separately ( as we have done ) and not only flavor non singlet combinations like @xmath127 . the quark mass dependence of our results is rather mild for the range of ( relatively large ) masses that we studied . therefore the extrapolation to the chiral limit looks quite safe , although , of course , unexpectedly large effects at truly small masses can not be excluded . lattice artifacts are obvious in our renormalization factors ( see , e.g. , fig . [ fig.zrat ] ) . we have to expect them also in the nucleon matrix elements . since we are working with non perturbatively improved fermions we could in principle reduce their size by using improved operators . unfortunately , the non perturbative improvement of operators of the kind needed here is not straightforward and has yet to be worked out . in particular , improvement of the renormalization factors requires off shell improvement . although there are some ideas on how to solve this non trivial problem ( see , e.g. , ref . @xcite ) , an implementation for the operators considered here is beyond our present possibilities . in this paper we have tried to obtain a more reliable lattice estimate of the twist three nucleon matrix element @xmath2 improving on our first calculation @xcite in several respects . we have made a serious attempt to take into acccount the most important part of the operator mixing which occurs in this case , namely the mixing with lower dimensional operators . this could only be done non perturbatively and led to a significant change in the results for @xmath2 moving them close to the experimental numbers . thus the mixing with lower dimensional operators seems to account for a large part of the difference between our previous computation and the experimental data . the calculations have been performed in the quenched approximation at three different values of @xmath84 corresponding to three different values of the lattice spacing . while our results are still not good enough to allow for a meaningful extrapolation to the continuum limit , the mutual consistency of the values obtained for @xmath2 at the various @xmath84 s indicates that discretization effects are smaller than our statistical errors and corroborates our conclusion that the twist three nucleon matrix element is rather small , in agreement with the experimental findings . we consider the computations at @xmath103 , i.e. at our smallest lattice spacing , to be most reliable . at @xmath128 they yield the structure function moments ( cf . table [ res5 ] ) @xmath129 for @xmath21 and @xmath130 for @xmath0 . these numbers are to be compared with the experimental results @xcite @xmath131 and @xmath132 respectively . the numerical calculations have been done on the ape / quadrics computers at desy ( zeuthen ) and the cray t3e at nic ( jlich ) and zib ( berlin ) . we thank the operating staff for support . this work was supported in part by the deutsche forschungsgemeinschaft and by bmbf . jaffe , comments nucl . * 19 * , 239 ( 1990 ) ; r.l . jaffe and x. ji , phys . d * 43 * , 724 ( 1991 ) . s. wandzura and f. wilczek , phys . b * 72 * , 195 ( 1977 ) . k. abe et al . ( e143 collaboration ) , phys . d * 58 * , 112003 ( 1998 ) . anthony et al . ( e155 collaboration ) , phys . b * 458 * , 529 ( 1999 ) . m. gckeler , r. horsley , e .- ilgenfritz , h. perlt , p. rakow , g. schierholz , and a. schiller , phys . rev . d * 53 * , 2317 ( 1996 ) . d. dolgov , r. brower , s. capitani , j.w . negele , a. pochinsky , d. renner , n. eicker , t. lippert , k. schilling , r.g . edwards , and u.m . heller , hep - lat/0011010 . g. martinelli , c. pittori , c.t . sachrajda , m. testa , and a. vladikas , nucl . * b445 * , 81 ( 1995 ) . k. jansen , c. liu , m. lscher , h. simma , s. sint , r. sommer , p. weisz , and u. wolff , phys . lett . b * 372 * , 275 ( 1996 ) . m. gckeler , r. horsley , h. oelrich , h. perlt , d. petters , p. rakow , a. schfer , and g. schierholz , nucl . b544 * , 699 ( 1999 ) . m. gckeler , r. horsley , w. krzinger , h. oelrich , p. rakow , and g. schierholz , hep - ph/9909253 . e.v . shuryak and a.i . vainshtein , nucl . * b201 * , 141 ( 1982 ) ; a.p . bukhvostov , e.a . kuraev , and l.n . lipatov , jetp lett . * 37 * , 482 ( 1983 ) ; sov . jetp * 60 * , 22 ( 1984 ) ; p.g . ratcliffe , nucl . * b264 * , 493 ( 1986 ) ; i.i . balitsky and v.m . braun , nucl . * b311 * , 541 ( 1988/89 ) ; x. ji and c. chou , phys . d * 42 * , 3637 ( 1990 ) . j. kodaira , y. yasui , and t. uematsu , phys . b * 344 * , 348 ( 1995 ) . j. kodaira , y. yasui , k. tanaka , and t. uematsu , phys . b * 387 * , 855 ( 1996 ) . a. ali , v.m . braun , and g. hiller , phys . b * 266 * , 117 ( 1991 ) . k. sasaki , phys . d * 58 * , 094007 ( 1998 ) . c. best , m. gckeler , r. horsley , e.m . ilgenfritz , h. perlt , p. rakow , a. schfer , g. schierholz , a. schiller , and s. schramm , phys . d * 56 * , 2743 ( 1997 ) . m. gckeler , r. horsley , e .- ilgenfritz , h. perlt , p. rakow , g. schierholz , and a. schiller , nucl . * b472 * , 309 ( 1996 ) . m. baake , b. gemnden , and r. oedingen , j. math . phys . * 23 * , 944 ( 1982 ) ; j. mandula , g. zweig , and j. govaerts , nucl . b228 * , 109 ( 1983 ) . m. gckeler , r. horsley , e .- ilgenfritz , h. perlt , p. rakow , g. schierholz , and a. schiller , phys . d * 54 * , 5705 ( 1996 ) . m. lscher , s. sint , r. sommer , p. weisz , and u. wolff , nucl . * b491 * , 323 ( 1997 ) . m. gckeler , r. horsley , h. oelrich , d. petters , d. pleiter , p.e.l . rakow , g. schierholz , and p. stephenson , phys . d * 62 * , 054504 ( 2000 ) . m. stratmann , a. weber , and w. vogelsang , phys . d * 53 * , 138 ( 1996 ) . x. ji , w. lu , j. osborne , and x. song , hep - ph/0006121 . s. capitani , m. gckeler , r. horsley , h. perlt , p.e.l . rakow , g. schierholz , and a. schiller , hep - lat/0007004 . m. guagnelli , r. sommer , and h. wittig , nucl . * b535 * , 389 ( 1998 ) . dccdd @xmath84 & lattice & & @xmath133[gev ] & @xmath83 + 6.0 & @xmath134 & me & 2.12 & 1.769 + 6.0 & @xmath135 & z & 2.12 & 1.769 + 6.2 & @xmath135 & me & 2.90 & 1.614 + 6.2 & @xmath136 & z & 2.90 & 1.614 + 6.4 & @xmath137 & me & 3.85 & 1.526 + 6.4 & @xmath138 & z & 3.85 & 1.526 cdddddd & + & 0.132 & 0.1324 & 0.1333 & 0.1338 & 0.1342 & @xmath139 + @xmath140 & 0.5412(9 ) & 0.5042(7 ) & 0.4122(9 ) & 0.3549(12 ) & 0.3012(10 ) & 0.0 + & & & & & & + @xmath141 & 0.114(8 ) & 0.114(8 ) & 0.107(11 ) & 0.09(2 ) & 0.08(2 ) & 0.08(2 ) + @xmath142 & @xmath1430.029(3 ) & @xmath1430.032(4 ) & @xmath1430.037(7 ) & @xmath1430.032(12 ) & @xmath1430.047(16 ) & @xmath1430.046(11 ) + & & & & & & + @xmath106\,(u ) } $ ] & 0.0063(12 ) & 0.0028(13 ) & @xmath1430.010(2 ) & @xmath1430.023(5 ) & @xmath1430.028(6 ) & @xmath1430.037(4 ) + @xmath106\,(d ) } $ ] & @xmath1430.0041(6 ) & @xmath1430.0027(7 ) & @xmath1430.0008(14 ) & @xmath1430.005(2 ) & @xmath1430.001(4 ) & 0.001(2 ) + & & & & & & + @xmath144 & @xmath1430.216(12 ) & @xmath1430.228(12 ) & @xmath1430.246(18 ) & @xmath1430.29(4 ) & @xmath1430.27(3 ) & @xmath1430.30(3 ) + @xmath145 & 0.050(4 ) & 0.050(3 ) & 0.052(6 ) & 0.064(12 ) & 0.044(17 ) & 0.057(12 ) cddddd & + & 0.1333 & 0.1339 & 0.1344 & 0.1349 & @xmath139 + @xmath140 & 0.4136(6 ) & 0.3570(10 ) & 0.3034(6 ) & 0.2431(7 ) & 0.0 + & & & & & + @xmath141 & 0.142(10 ) & 0.137(15 ) & 0.157(17 ) & 0.16(3 ) & 0.17(3 ) + @xmath142 & @xmath1430.033(4 ) & @xmath1430.030(5 ) & @xmath1430.034(9 ) & @xmath1430.031(15 ) & @xmath1430.030(13 ) + & & & & & + @xmath106\,(u ) } $ ] & @xmath1430.0017(14 ) & @xmath1430.017(2 ) & @xmath1430.031(5 ) & @xmath1430.051(11 ) & @xmath1430.065(7 ) + @xmath106\,(d ) } $ ] & @xmath1430.0027(6 ) & 0.0006(11 ) & @xmath1430.0003(17 ) & @xmath1430.000(4 ) & 0.004(3 ) + & & & & & + @xmath144 & @xmath1430.34(2 ) & @xmath1430.37(3 ) & @xmath1430.44(4 ) & @xmath1430.50(7 ) & @xmath1430.54(7 ) + @xmath145 & 0.067(6 ) & 0.065(8 ) & 0.072(13 ) & 0.07(3 ) & 0.07(2 ) cdddddd & + & 0.1338 & 0.1342 & 0.1346 & 0.135 & 0.1353 & @xmath139 + @xmath140 & 0.3213(8 ) & 0.2836(9 ) & 0.2402(8 ) & 0.1933(7 ) & 0.1507(8 ) & 0.0 + & & & & & & + @xmath141 & 0.123(8 ) & 0.092(13 ) & 0.114(13 ) & 0.102(17 ) & 0.14(4 ) & 0.096(19 ) + @xmath142 & @xmath1430.030(4 ) & @xmath1430.018(5 ) & @xmath1430.032(7 ) & @xmath1430.033(10 ) & @xmath1430.01(2 ) & @xmath1430.023(10 ) + & & & & & & + @xmath106\,(u ) } $ ] & @xmath1430.0149(14 ) & @xmath1430.024(3 ) & @xmath1430.038(4 ) & @xmath1430.055(6 ) & @xmath1430.055(15 ) & @xmath1430.068(6 ) + @xmath106\,(d ) } $ ] & @xmath146.0014(7 ) & @xmath146.0021(13 ) & @xmath146.0062(17 ) & @xmath146.012(3 ) & @xmath146.015(8 ) & 0.013(3 ) + & & & & & & + @xmath144 & @xmath1430.40(2 ) & @xmath1430.41(4 ) & @xmath1430.47(4 ) & @xmath1430.51(5 ) & @xmath1430.55(13 ) & @xmath1430.56(5 ) + @xmath145 & 0.088(6 ) & 0.075(12 ) & 0.105(13 ) & 0.12(2 ) & 0.15(5 ) & 0.12(2 ) cddd & + & 6.0 & 6.2 & 6.4 + @xmath147 & 0.046(13 ) & 0.11(2 ) & 0.066(14 ) + @xmath148 & 0.008(4 ) & 0.017(9 ) & 0.017(7 ) + @xmath149 & 0.012(3 ) & 0.029(5 ) & 0.017(4 ) + @xmath150 & @xmath1430.007(3 ) & @xmath1430.019(4 ) & @xmath1430.010(3 ) + & & & + @xmath151 & @xmath1430.017(8 ) & 0.009(11 ) & 0.000(8 ) + @xmath152 & @xmath1430.003(2 ) & @xmath1430.001(4 ) & @xmath1430.001(3 ) + @xmath153 & @xmath1430.004(2 ) & 0.002(3 ) & 0.000(2 ) + @xmath154 & 0.0026(16 ) & @xmath1430.002(2 ) & @xmath1430.0002(17 ) cddd & + & 6.0 & 6.2 & 6.4 + @xmath147 & 0.040(12 ) & 0.098(18 ) & 0.057(12 ) + @xmath148 & 0.002(4 ) & 0.006(7 ) & 0.008(6 ) + @xmath149 & 0.010(3 ) & 0.025(5 ) & 0.015(3 ) + @xmath150 & @xmath1430.007(2 ) & @xmath1430.017(4 ) & @xmath1430.009(3 ) + & & & + @xmath151 & @xmath1430.015(7 ) & 0.007(9 ) & 0.000(7 ) + @xmath152 & @xmath1430.003(2 ) & @xmath1430.002(3 ) & @xmath1430.001(3 ) + @xmath153 & @xmath1430.0037(18 ) & 0.002(2 ) & 0.0001(19 ) + @xmath154 & 0.0021(14 ) & @xmath1430.0016(18 ) & @xmath1430.0002(14 )

supernovae of type ii are those that have obvious hydrogen lines in their optical spectra . type iib supernovae have obvious hydrogen lines around the time when they reach their maximum brightness ( hereafter just `` maximum light '' ) but later the hydrogen lines become weak or even disappear . type ib supernovae do not have obvious hydrogen lines but they do develop conspicuous he i lines after maximum light . neither hydrogen nor he i lines are conspicuous in the spectra of type ic supernovae . most or all events of these four types ii , iib , ib , and ic are thought to result from core collapse in massive stars . ( type ia supernovae , whose spectra lack hydrogen and have a strong absorption feature produced by si ii @xmath0 , are thought to have a fundamentally different origin , as thermonuclear disruptions of accreting or merging white dwarfs . ) for a recent review of supernova spectral classification , including its historical development and with illustrations of spectra of each type , see filippenko ( 1997 ) . in this paper we are concerned with optical photospheric phase spectra of type ib supernovae ( sne ib ) . until recently , good photospheric phase spectra had been published for only two sne ib : sn 1983n ( richtler & sadler 1983 ; harkness et al . 1987 ) and sn 1984l ( harkness et al . 1987 ) . the situation has improved substantially now that matheson et al . ( 2001 ) have published spectra of the sne ib that were observed at the lick observatory during the 1990s . these newly available spectra , together with some additional previously unpublished spectra that are presented in this paper , motivated us to carry out a comparative study of sn ib spectra . ( matheson et al . also present spectra of sne iib and ic , the study of which we defer to separate papers . ) our method is to compare the observed sn ib spectra with synthetic spectra that we generate with the fast , parameterized , supernova spectrum synthesis code , synow . we refer to this approach , in which the goal is to extract some constraints on the ejected matter from the observations in an empirical spirit , as `` direct '' analysis to distinguish it from the process of making very detailed non local thermodynamic equilibrium ( non lte ) calculations of synthetic spectra based on supernova hydrodynamical models ( e.g. , baron et al . 1999 ) . issues that we can explore by means of our direct analysis include ( 1 ) line identifications ; ( 2 ) the extent to which synthetic spectra calculated on the basis of simple assumptions can or can not account for observed sn ib spectra ; ( 3 ) the degree to which the rather homogeneous appearance of sn ib spectra , pointed out by matheson et al . ( 2001 ) , reflects a genuine physical homogeneity ; ( 4 ) the velocities at which the he i lines form , compared to the velocity at the photosphere as determined by the fe ii lines ; ( 5 ) whether hydrogen lines are present and , when they are , the velocities at which they form . previous work on the interpretation of photospheric phase spectra of sne ib is briefly summarized in 2 . the observed spectra that were selected for this project are discussed in 3 and those that have not been published previously are displayed . the synthetic spectrum calculations are described in 4 , and comparisons with observed spectra are presented in 5 . the results are summarized and discussed in 6 . the classic early paper on the interpretation of the photospheric phase spectra of sne ib was that of harkness et al . ( 1987 ) , who calculated local thermondynamic equilibrium ( lte ) synthetic spectra for parameterized model supernovae having power law density structures and homogeneous chemical compositions , and compared them to observed spectra of sne 1984l and 1983n . some observed features that appear in spectra of all supernova types were readily attributed to ca ii and fe ii lines . most of the remaining conspicuous features were convincingly attributed to lines of he i , even though large ad hoc overpopulations of the highly excited lower levels of the he i lines had to be invoked in order to account for their presence . the explanation for the overpopulations was later shown to be nonthermal excitation and ionization caused by the decay products of radioactive @xmath1ni and @xmath1co ( lucy 1991 ; swartz et al . 1993 ) . a feature of special interest in the spectra of sne 1984l and 1983n was an absorption near 6300 that could not be attributed to fe ii , ca ii , or he i lines . [ harkness et al . ( 1987 ) referred to this as `` the 6300 absorption '' and so will we , although its wavelength can be as short as 6200 at early times . ] harkness et al . tentatively attributed this absorption to c ii @xmath2 , forming in outer high velocity ( @xmath314,000 ) layers of the ejected matter . later , on the basis of lte synthetic spectra calculated for radially stratified chemical compositions , wheeler et al . ( 1994 ) suggested the absorption to be h@xmath4 , forming at @xmath313,000 . the issue of whether hydrogen is present in sne ib is very important because of its implications for the nature and appearance of the progenitor stars , but it has been a difficult issue to resolve because c ii 6580 , being less than 800 to the red of h@xmath4 , is usually a plausible alternative identification . woosley & eastman ( 1997 ) presented a comparison of a non lte synthetic spectrum based on a particular explosion model with a photospheric phase spectrum of sn 1984l . overall , the synthetic spectrum accounted rather well for the major features in the observed spectrum . the explosion model did not contain any hydrogen , and in the synthetic spectrum the absorption nearest to the 6300 feature was produced by si ii , but as we will see below this can not be the actual identification of the 6300 absorption . this is the only comparison of an non lte synthetic spectrum with an observed sn ib photospheric phase spectrum to be published so far . no other sn ib was well observed until sn 1999dn . deng et al . ( 2000 ) used the same synow code that we use in this paper to make a detailed study of line identifications in three spectra that were obtained at the beijing astronomical observatory at times of about 10 days before , at , and 14 days after maximum light . in addition to fe ii , ca ii , and he i lines , deng et al . explored the possible role of lines of other ions ( c i , o i , c ii , [ o ii ] , na i , mg ii , si ii , ca i , and ni ii ) in shaping the spectra of sn 1999dn . they attributed the 6300 absorption in the latest of their three spectra of sn 1999dn to c ii @xmath2 , forming at @xmath310,000 , but they suggested that in the earlier two spectra the observed feature was more likely to be h@xmath4 , forming at higher velocity . the 11 sne ib that were selected for this study are listed in table 1 . an asterisk preceding the recession velocity , @xmath5 , indicates that it is the value given by matheson et al . ( 2001 ) for an h ii region near the site of the supernova ; otherwise the listed value is that of the parent galaxy , from the asiago supernova catalog ( barbon et al . 1999 ; updates are available at _ http://merlino.pd.astro.it/supern/_ ) . all observed spectra displayed in this paper are corrected for redshift using the values of @xmath5 listed in table 1 . an asterisk preceding the date of maximum light in the @xmath6 band , @xmath7 , indicates that only the date of discovery is listed , because the date of maximum light is unknown . six of these events sne 1991ar , 1997dc , 1998 t , 1998dt , 1999di , and 1999dn were selected from matheson et al . ( 2001 ) because they do nt have obvious hydrogen lines while they do have conspicuous he i lines . [ sn 1991d , which also may be a type ib but with fairly weak he i lines , is discussed in a separate paper ( s. benetti et al . , in preparation ) . ] sne 1998dt , 1999di , and 1999dn are especially useful for our study because on the basis of photometry obtained at the lick observatory , matheson et al . were able to estimate the dates of maximum light in the @xmath8 band , which we will assume to peak at the same time as the @xmath6 band [ as was the case for the type iib sn 1996cb ( qiu et al . the three spectra of sn 1999dn that appeared in deng et al . ( 2000 ) also are included in this study . the spectra of sn 1983n are from richter & sadler ( 1983 ) and harkness et al . ( 1987 ) , and the adopted date of maximum light , 1983 july 17 , ( in the _ iue _ fes band , which is roughly like the @xmath6 band ) is from an unpublished manuscript that was circulated by n. panagia et al . in 1984 . the spectra of sn 1984l are from harkness et al . tsvetkov ( 1987 ) estimated that sn 1984l reached maximum light in the @xmath9 band on 1984 august @xmath10 days . we assume that the @xmath6 band peaked two days later ( as was the case for sn 1996cb ) and adopt august 22 as the date of maximum light in the @xmath6 band . sn 1954a is a special case because only photographic spectra , obtained by n. u. mayall at the lick observatory and by r. minkowski at the mount wilson and palomar observatories , are available . microphotometer tracings of the spectra of sn 1954a and many other supernovae observed at the lick observatory and the mount wilson and palomar observatories between 1937 and 1971 have been digitized and displayed by casebeer et al . ( 2000 ) and blaylock et al . ( 2000 ) . in the asiago catalog the date of maximum light of sn 1954a in the @xmath9 band is estimated as 1954 april 19 , so we will adopt april 21 for the @xmath6 band . one previously unpublished spectrum of sn 1996n is included in this study . the spectrum , obtained at the anglo australian telescope on 1996 march 23 ( germany et al . 2000 ) , 11 days after discovery , is shown in figure 1 . the date of maximum light of sn 1996n is unknown . the spectrum appears to be that of a typical sn ib not long after maximum light . six previously unpublished spectra65.h0292 , and at the asiago observatory ] of sn 2000h ( pastorello et al . 2000 ; benetti et al . 2000 ) also are included . these are shown in figure 2 . the spectra of sn 2000h resemble those of a typical sn ib except for an unusually deep 6300 absorption in the first four spectra , as well as a weak , narrow absorption near 4650 in at least the second and third spectra . benetti et al . attributed these absorptions to h@xmath4 and h@xmath11 . ( the h@xmath11 feature will be seen more clearly in subsequent figures . ) this identification of hydrogen lines might raise the question of whether sn 2000h should be regarded as a type iib , but we do not favor such a classification because even at the earliest observed times the presence of hydrogen lines was not obvious , as evidenced by initial classifications of sn 2000h as a peculiar type ia ( garnavich et al . 2000 ) and a type ic ( pastorello et al . 2000 ) . from unpublished eso photometry of sn 2000h we estimate that the date of maximum light in the @xmath9 band was 2000 february @xmath12 days , so we adopt february 11 as the date of maximum light in the @xmath6 band . calculations have been carried out with the fast , parameterized supernova spectrum synthesis code , synow . recent applications to type ic supernovae and brief descriptions of synow can be found in millard et al . ( 1999 ) and branch ( 2001 ) , and technical details of the code are in fisher ( 2000 ) . an extensive discussion and illustration of the elements of supernova line formation appears in jeffery & branch ( 1990 ) . the basic assumptions of synow are spherical symmetry ; velocity proportional to radius ; a sharp photosphere ; and line formation by resonant scattering , treated in the sobolev approximation . various fitting parameters are available . the parameter @xmath13 is the temperature of the blackbody continuum from the photosphere . the values used in this paper range from 8500 to 3600 k and typically are @xmath146500 k around the time of maximum light and @xmath145000 k beginning roughly two weeks after maximum . we do not attach much physical significance to these values because ( for one thing ) the observed spectra have not been corrected for interstellar extinction . for each ion whose lines are introduced , the optical depth at the photosphere of a `` reference line '' is a fitting parameter , and the optical depths of the other lines of the ion are calculated assuming boltzmann excitation at excitation temperature @xmath15 . in this paper , to reduce the number of free parameters , we simply fix @xmath15 at a nominal sn ib value of 7000 k. the relevant lines of a given ion do nt have widely differing excitation potentials so their relative optical depths do nt depend strongly on @xmath15 , within the range of temperatures that are relevant here . the line optical depths are taken to vary with ejection velocity as @xmath16 . again for simplicity , in this paper we always use @xmath17 , except for one illustration of the effects of using @xmath18 instead . in the analysis of the sn 1999dn spectra by deng et al . ( 2000 ) , the line optical depths were taken to vary as @xmath19 , with @xmath20 . since the exponential distribution has an effective power law index of @xmath21 , the distribution used by deng et al . falls off more steeply than @xmath17 for @xmath22 and less steeply for @xmath23 . this leads to some differences in the values of @xmath24 and the reference line optical depths used by deng et al . and by us to match the same observed spectra . the maximum velocity of the line forming region is set high enough so that effectively there is no outer boundary . the default minimum velocity of the line forming region is the velocity at the photosphere ; when an ion is assigned a higher minimum velocity , that ion is said to be detached from the photosphere . reasons that synow spectra can not be expected to provide exact fits to observed spectra are numerous and obvious : the calculations are based on many simplifiying assumptions , including spherical symmetry , and the oscillator strengths ( kurucz 1993 ) are good but not perfect . in this paper we are not concerned with proposing a line identification for every weak observed feature . we are more interested in establishing the identities of the major features and then concentrating on a comparative analysis to investigate the degree to which the sne ib of our sample are physically similar , and to look for differences . we begin the comparisons of observed and synthetic spectra by concentrating on a `` fiducial '' sn ib spectrum a spectrum of a typical sn ib that has good signal to noise ratio and broad wavelength coverage , and in which most of the major spectral features are well developed . the best available spectrum for this purpose is the matheson et al . ( 2001 ) spectrum of sn 1999dn obtained on 1999 september 17 , 17 days after maximum light . in figure 3 , this spectrum is compared with a synthetic spectrum that has = 6000 and = 4800 k , and contains lines of fe ii , he i , o i , ca ii , ti ii , and sc ii . almost all of the features in the observed spectrum can be attributed to these ions . the discrepancies will be discussed as we look at the contribution of each ion to the synthetic spectrum . as always when fitting observed spectra with synow spectra , we are more concerned with discrepancies in the wavelengths of absorption features than with discrepancies in flux ; the latter are inevitable given the simplicity of our spectrum calculations . figure 4 is like figure 3 but with nothing but the fe ii lines in the synthetic spectrum . the optical depth of the reference line , 5018 , is 7 . the fe ii lines are mainly responsible for the spectral features from about 4300 to 5300 , and they have additional effects at shorter wavelengths . at this post maximum time they may also be responsible for the observed absorptions near 6100 and 6300 . ( these two features are not strong enough in this particular synthetic spectrum , but a higher value of @xmath15 would increase their strengths relative to the reference line . ) for this reason we do nt use h@xmath4 or c ii 6580 to account for the weak 6300 absorption in this observed spectrum . around maximum light , however , fe ii lines are not strong enough to account for the 6300 absorption that is observed at that time . in this paper we always determine the value of @xmath24 on the basis of the fe ii features in the region from about 4300 to 5300 . figure 5 shows a comparison of two fe ii synthetic spectra that have = 5000 and 10,000 . this figure shows that the spectral signature of fe ii in this wavelength range is quite sensitive to . our fitting uncertainty in @xmath24 is about 1000 . the top panel of figure 6 is like figure 3 but with only the he i lines in the synthetic spectrum . the he i lines are detached at 8000 ( recall that = 6000 ) , where the optical depth of the reference line , 5876 , is 10 . it is likely that two optical he i lines , 6678 and 7065 , are almost entirely responsible for their corresponding observed features . two other lines , 5876 and 4472 , are mainly responsible for their corresponding observed features but they may be blended with the na i d lines ( 5890 , 5896 ) and mg ii 4481 , respectively . in this spectrum he i 7281 accounts very nicely for an observed feature , but in some other spectra the fit is not so good . the remaining optical he i lines are weaker and in the synthetic spectrum of figure 3 they are overwhelmed by lines of other ions . in figures 3 and 6 the blue edge of the synthetic absorption produced by he i 10830 is not blue enough to account for the sharp drop in the observed spectrum near 9000 , but we show how to remedy this below . in figure 3 and the top panel of figure 6 the he i lines are detached at 8000 in order to fit the wavelengths of the corresponding observed absorption features . the detachment causes the flat tops of the synthetic he i emission components ( which are superimposed on a sloping continuum ) . the rounded emission peak that is observed near 5900 could easily be achieved in the synthetic spectrum by including undetached na i d lines . to illustrate the necessity of detaching the he i lines , the bottom panel of figure 6 shows how they appear when they are undetached , i.e. , when they are allowed to form down to the photospheric velocity of 6000 . these synthetic absorptions obviously are insufficently blueshifted . the top panel of figure 7 is like figure 3 but with only the o i lines . the optical depth of the reference line , 7773 , is 1 . the 7773 line accounts for at least most of an observed feature ; in some of the other observed spectra this feature may be partly produced by mg ii 7890 . the o i 9264 line may be responsible for a weak observed feature , while the 8446 feature usually is overwhelmed by the ca ii infrared triplet in sne ib . whenever we use o i lines in the synthetic spectra of this paper , the optical depth of the reference line is near 1 . the bottom panel of figure 7 is like figure 3 but with only the ca ii lines . the optical depth of the reference line , 3933 , is 300 . only the h&k lines ( 3933 , 3968 ) and the infrared triplet ( 8542 , 8662 , 8498 ) produce observable features , both of which are very strong . the notch in the synthetic spectrum near 8400 nicely matches an observed feature . in this synthetic spectrum the ca ii lines are detached to 7000 to match the infrared triplet , but in most of our synthetic spectra the ca ii lines are undetached . the top panel of figure 8 is like figure 3 but with only the ti ii lines . the optical depth of the reference line , 4550 , is 1 . the ti ii lines are used to help match the broad observed absorption trough between 4100 and 4500 . we consider the presence of ti ii lines in the observed spectrum to be probable but not definite . [ we do consider them to be definite in peculiar subluminous sne ia such as sn 1991bg ( filippenko et al . 1992 ) and sn 1999by ( garnavich et al . 2001 ) , and in the type ic sn 1994i ( millard et al . whenever we use ti ii lines in this paper , the optical depth is near 1 . none of our other conclusions would be affected by omitting the ti ii lines . the bottom panel of figure 8 is like figure 3 but with only the sc ii lines . the optical depth of the reference line , 4247 , is 0.5 . the sc ii lines are used mainly to get a peak in the synthetic spectrum near 5500 . the price to be paid is an overly strong synthetic absorption produced by 4247 . sc ii lines are plausibly present in sne ib because they are expected ( in lte ) to appear at low temperatures ( hatano et al . 1999 ) and they appear in sne ii . harkness et al . ( 1987 ) suggested that in sn 1984l the observed emission peak near 5470 was caused by the early emergence of blueshifted [ o i ] 5577 nebular phase emission , but swartz et al . ( 1993 ) found this to be unlikely for the type ic sn 1987 m . for a discussion of the possibility of an early emergence of blueshifted [ o i ] 5577 emission in the type iib sn 1996cb , see qiu et al . ( 1999 ) and deng , qiu , & hu ( 2001 ) . in our view the 5500 emission in the fiducial spectrum of sn 1999dn probably , but not definitely , is produced by sc ii lines . whenever we use them , the optical depth of the reference line is near 1 . none of our other conclusions would be affected by omitting the sc ii lines . figure 9 is like figure 3 except that the power law index @xmath25 has been reduced from 8 to 5 ( and the optical depths of the reference lines have been correspondingly reduced to keep the synthetic features from becoming too strong ) . now the blue wings of the synthetic absorptions produced by the ca ii infrared triplet , ca ii h&k , and he i 10830 fit better than in figure 3 , but the synthetic absorptions produced by 5876 and 6678 extend too far to the blue . this reflects the fact that in real supernovae , contrary to our assumption , the line optical depths do not all follow the same power law ( or any power law ) . a slower radial decline of the optical depth of 10830 line , compared to the optical he i lines , is expected on the basis of the nonthermal excitation calculations of lucy ( 1991 ; his figure 3 ) and swartz et al . ( 1993 ; their figure 11 ) . note that if the blue edge of the 10830 absorption really extends to 9000 , as it does in the synthetic spectrum of figure 9 , then the line is forming all the way out to 50,000 . matheson et al . ( 2001 ) demonstrated that the absorption produced by 6678 ( the lower level of which is 1s2p @xmath26p@xmath27 ) becomes weaker with time relative to the absorptions produced by 5876 and 7065 ( both 1s2p @xmath28p@xmath27 ) ; this also is to be expected on the basis of the results of lucy and of swartz et al . , because the singlet resonance transitions to the ground state become less opaque as the ejecta density decreases through expansion . as mentioned above , deng et al . ( 2000 ) identified c ii lines in their september 14 spectrum of sn 1999dn , obtained only three days before our fiducial spectrum of september 17 . the identification of c ii 6580 for the 6300 absorption was supported by attributing a weak absorption near 4580 to c ii 4738 , 4745 . the reasons that we do nt introduce c ii lines for the fiducial spectrum ( apart from the fact that the 4580 absorption does nt appear distinctly in the fiducial spectrum ) are that ( 1 ) as mentioned above , fe ii lines could be responsible for the 6300 absorption , and ( 2 ) the absorption near 4580 in the beijing spectrum of september 14 might be produced by he i 4731 ( see the top panel of figure 6 ) and/or lines of sc ii ( see the bottom panel of figure 8) . now we turn to sn 2000h , an event that has conspicuous he i lines but that according to benetti et al . ( 2000 ) also has hydrogen lines . figure 10 compares the @xmath29 day spectrum of sn 2000h with a synthetic spectrum that has = 8000 and = 6500 k , and contains hydrogen lines in addition to the ions used above for the fiducial spectrum of sn 1999dn . the he i lines are detached at 9000 , where the optical depth of the reference line is 2 . the hydrogen lines are detached at 13,000 where the optical depth of the reference line , h@xmath4 , is 2.5 . with this detachment velocity , h@xmath4 accounts for at least most of the 6300 absorption and h@xmath11 accounts for the unusual notch in the emission peak near 4650 . a closer view of the h@xmath11 region is provided in figure 11 . the presence of an absorption that is attributable to h@xmath11 provides strong support for the presence of hydrogen in sn 2000h . matheson et al . ( 2001 ) noted the presence of an unusually deep 6300 absorption in sn 1999di , and mentioned that it could be si ii 6355 , c ii 6580 , or h@xmath4 . si ii can now be rejected because its absorption would be blueshifted by only 2600 , which is too much lower than than our value of = 6000 . figure 12 compares our earliest spectrum of sn 1999di , obtained 21 days after maximum , with the @xmath30 day spectrum of sn 2000h . the similarity of these two spectra is remarkable ( apart from differences at wavelengths longer than 9000 where both spectra are noisy ) . the narrow h@xmath11 absorption of sn 2000h also can be seen in sn 1999di . figure 13 compares the @xmath31 day spectrum of sn 1999di with a synthetic spectrum that has = 7000 and = 4500 k and contains the same ions as figure 10 for sn 2000h . hydrogen is detached at 12,000 . figure 14 shows a closer view of the h@xmath11 region . could the 6300 absorption in sne 2000h and 1999di be produced by c ii 6580 rather than h@xmath4 ? we think not . it would be surprising to see a deep c ii 6580 absorption in sne ib , when even in sne ic this line never forms a deep absorption and , if present at all , is hard to identify . it also would be surprising that the c ii feature would be so detached in sne ib . the top panel of figure 15 is like figure 13 except that only c ii lines , detached at 13,000 , are used . although 6580 can account for the 6300 absorption as well as h@xmath4 does , and 7236 , 7231 are not a problem because , being more detached than he i , they fall near or within the strong feature produced by he i 7065 , the absorption produced by c ii 4738 , 4745 is much too strong ( at least in lte at 7000 k ) . could the 6300 absorption in sne 2000h and 1999di be produced by ne i @xmath326402 rather than h@xmath4 ? no , it can not . the bottom panel of figure 15 shows that undetached ne i fails in two ways : ( 1 ) the absorption produced by @xmath326402 is too far to the blue , and ( 2 ) even though the synthetic absorption produced by @xmath326402 has not been made strong enough , other unwanted features already are present . in view of these difficulties with c ii and ne i , and the apparent presence of h@xmath11 in the observed spectra , we consider the identification of hydrogen lines in sne 2000h and 1999di to be definite . sn 1954a appears to have been spectroscopically akin to sne 2000h and 1999di . mclaughlin ( 1963 ) and branch ( 1972 ) identified he i lines in photographic spectra of sn 1954a obtained at the lick observatory and the mount wilson and palomar observatories , respectively . consequently sn 1954a usually has been regarded to have been a type ib supernova , but on occasion doubts has been expressed about the classification because of the low quality of the photographic spectra compared to modern observations . figure 16 compares microphotometer tracings of the two earliest spectra of sn 1954a ( blaylock et al . 2000 ) , obtained on blue and red sensitive emulsions 46 days after maximum light , with the @xmath30 day spectrum of sn 2000h . only this one spectrum of sn 1954 was obtained on a red sensitive emulsion . the sn 1954a spectra are not actually relative flux but merely a measure of the transmission through the photographic plate . parts of the spectra were overexposed so the features are distorted ; in particular , emission peaks tend to be suppressed . the shapes of these spectra also are strongly influenced by the wavelength dependences of the emulsion sensitivities ; e.g. , neither emulsion is sensitive around 5200 and the sensitivity falls off steeply between 6500 and 7000 . nevertheless , some of the absorptions in sn 1954a can be located and they correspond well with the absorptions in sn 2000h , including the two attributed to h@xmath4 and h@xmath11 . branch ( 1972 ) considered the possibility of h@xmath4 and h@xmath11 absorptions in sn 1954a , blueshifted by 10,800 in the observer s frame ; the value of @xmath5 for the parent galaxy is now known to be 291 , so the hydrogen lines are blueshifted by about 11,000 in the supernova frame . branch also considered the possibility of ne i lines in sn 1954a . this is perhaps still not ruled out , but given the apparent resemblance of sn 1954a to sne 2000h , we prefer the hydrogen identification . in terms of apparent magnitude , sn 1954a was the fourth brightest supernova of the twentieth century , surpassed only by the type ii sn 1987a and the type ia sne 1972e and 1937c ( barbon et al . sn 1954a occurred in the star bursting dwarf galaxy ngc 4214 , at a distance of only about 4 mpc ( leitherer et al . 1996 ) , more than 10 times closer than sne 2000h and 1999di , so it is more amenable to studies of the environment in which it exploded . we note that van dyk , hamuy , & filippenko ( 1996 ) found sn 1954a to be unusual among sne ib in that it was not near any visible h ii region , and that a deep vla search for radio emission at the site of sn 1954a carried out by j. cowan and d. branch in may , 1986 , resulted in a three sigma upper limit to the flux density at 20 cm of 0.068 mjy ( eck 1998 ) , which corresponds to a monochromatic luminosity of less than one twentieth of cas a. figure 17 compares a spectrum of sn 1984l obtained 9 days after maximum with a synthetic spectrum that has = 8000 and = 6500 k , and includes hydrogen lines detached at 15,000 . the fe ii lines fit very well , and the fit to the other features is satisfactory . given the presence of h@xmath4 in sne 2000h , 1999di , and 1954a , we assume that the 6300 absorption is produced by h@xmath4 . however , because the h@xmath4 optical depth at the detachment velocity is only 0.6 , there is no support for this identification from h@xmath11 because it is too weak to see . ( the oscillator strength of h@xmath11 is about one fifth that of h@xmath4 . ) figure 18 compares the @xmath33 day spectrum of sn 1984l with the @xmath29 day spectrum of sn 2000h . the spectra are similar except that in sn 1984l the 6300 absorption is weaker and the 4560 absorption is not visible . this shows that while our assumption that the 6300 absorption in sn 1984l is produced by h@xmath4 is reasonable , it is not proven . it is conceivable that c ii 6580 or ( more plausibly because it would nt need to be detached ) ne i @xmath326402 could be responsible for the 6300 absorption in sn 1984l and other typical sne ib , and be overwhelmed by h@xmath4 only in events such as sne 2000h , 1999di , and 1954a . we proceed on the assumption that at early times the 6300 absorption is produced by h@xmath4 in all the sne ib of our sample . as an example of a comparison at a earlier time when is higher , figure 19 compares the earliest spectrum for which good wavelength coverage is available , the beijing spectrum of sn 1999dn 10 days before maximum , with a synthetic spectrum that has = 14,000 and = 6500 k. now hydrogen lines are detached at 18,000 , where the h@xmath4 optical depth is 1.3 . the fit is good , except near 6600 . figure 20 compares a spectrum of sn 1998dt obtained 32 days after maximum with a synthetic spectrum that has = 9000 and = 5000 k. this comparison is shown because , as will be seen below , although the spectrum has a typical sn ib appearance the inferred value of is unusually high for a sn ib this long after maximum . the fit to the fe ii features is unusually poor , but this value of does give the best fit in the 4300 to 5300 region , and a significantly lower value would give a noticeably worse fit . ( the narrow h@xmath4 emission from an h ii region in the parent galaxy is very close to 6563 , which shows that the high required value of is not due to the observed spectrum having been inadvertently overcorrected for parent galaxy redshift . ) now we briefly consider the events of our sample for which the times of maximum light are unknown . the spectrum of sn 1996n indicates that it was discovered not long after maximum . because we use hydrogen lines in the synthetic spectrum for sn 1996n , the comparison with the observed spectrum is shown in figure 21 . helium lines are undetached and the optical depth of the reference line is 5 ; hydrogen lines are detached at 17,000 where the optical depth of h@xmath4 is 0.5 . overall , the fit is good . sne 1991ar and 1997dc do not appear to be unusual provided that they were discovered well after maximum light . as discussed by matheson et al . ( 2001 ) , the spectra of sn 1998 t are seriously contaminated by light from the parent galaxy ; taking this into consideration , sn 1998 t does not appear to be unusual provided that it was discovered about a week after maximum . as far as we can tell , sne 1991ar , 1997dc , and 1998 t were typical sne ib , but they were not observed early enough to check on the presence of h@xmath4 . the most important of the fitting parameters that have been used for the synthetic spectra are collected in table 2 . the spectra are listed in order of time with respect to maximum light so only the supernovae for which we have an estimate of the time of maximum light appear in the table . ( the date 0703 , for example , refers to july 3 . ) for the six sne ib for which we have estimates of the time of maximum light , is plotted against time in figure 22 . the tightness of the relationship is striking . sn 1998dt at 32 days after maximum seems to stand out ; otherwise , the scatter about the mean curve is about what should be expected from our nominal errors of 1000 in and a few days in the dates of maximum light . for the simple case of constant opacity and a @xmath16 density distribution , the velocity at the photosphere would decrease with time as @xmath34 . the line in figure 22 , the best power law fit to the data ( excluding sn 1998dt at 32 days ) , corresponds to @xmath35 . considering that the opacity is not really constant , that the actual density distribution does not really follow a single power law over a wide velocity range , and that the best power law index for fitting the spectra is not well constrained , not much significance should be attched to the difference between @xmath35 and @xmath36 . our adopted values of can be used to make rough estimates of the mass and kinetic energy above the photosphere . for spherical symmetry and a @xmath16 density distribution , the mass ( in @xmath37 ) and the kinetic energy ( in 10@xmath38 ergs ) above the electron scattering optical depth @xmath39 are ( millard et al . 1999 ) where @xmath42 is in units of 10,000 , @xmath43 is the time since explosion in days , @xmath44 is the mean molecular weight per free electron , and the integration is carried out to arbitrarily high velocity . if we assume that maximum light occurs 20 days after explosion , that @xmath36 , that @xmath45 ( e.g. , half ionized helium or singly ionized oxygen ) , and that is at @xmath46 , then at maximum light = 10,000 ( figure 22 ) gives @xmath47 and @xmath48 ergs . at 20 days after maximum , using = 7000 and keeping the other parameters the same gives @xmath49 and @xmath50 ergs . in reality , of course , the kinetic energy above the 7000 photosphere must be greater than that above the 10,000 photosphere . if we use @xmath51 instead of @xmath36 between 7000 and 10,000 then we obtain a total of @xmath52 and @xmath53 ergs above the 7000 photosphere . figure 22 provides some constraints on models of sne ib . the hydrodynamics must account for the velocity at the photosphere as a function of time , and the ensemble of sn ib progenitors must be consistent with the tightness of the relationship . the small scatter suggests that the masses and the kinetic energies of the sne ib of our sample are similar , and it does not leave much room for the influence on of departures from spherical symmetry . figure 23 shows the minimum velocities of the he i lines ( squares when undetached and diamonds when detached ) . there appears to be a standard pattern ( again with the possible exception of sn 1998dt at 32 days ) . before and near the time of maximum the he i lines tend to be undetached . after maximum the lines tend to be detached , but the detachment velocities tend to decrease with time , from about 10,000 to 7000 . this means that the fraction of helium in this velocity range that is in the lower levels of the optical he i lines is increasing with time faster than @xmath54 . ( the matter density is decreasing as @xmath55 by expansion but the sobolev optical depth also is proportional to @xmath56 because it is inversely proportional to the velocity gradient . ) the increasing fraction of helium in the excited levels may be understandable in terms of the decreasing column depth between the nickel core and the helium layers , and the decreasing detachment velocity may mean that the fractional helium abundance is lower at lower velocities . in any case , some helium is present at least down to 7000 . our estimate above for the total mass above the 7000 photosphere was 1.1 to 1.5 m@xmath57 , which is a rough upper limit on the mass of helium above 7000 . there could be more helium below 7000 . figure 23 provides more constraints on models of sne ib . the radial profile of the helium abundance , together with that of the @xmath1ni that is responsible for exciting it , should account for the he i velocities and optical depths ( table 2 ) . figure 23 also shows the minimum velocities of the hydrogen lines ( circles ) , which always are detached . in the three events for which we are convinced of the hydrogen identifications sne 2000h , 1999dn , and 1954a the minimum hydrogen velocity is between 11,000 and 13,000 . in the events in which we assume h@xmath4 to be present at early times sne 1983n , 1984l , 1999dn , and 1996n ( the latter is not shown in figure 23 because the date of maximum light is unknown ) the detachment velocities tend to decrease with time but they are consistent with similar minimum velocities of the hydrogen . thus the available evidence is consistent with the proposition that sne ib in general have hydrogen down to @xmath58 . [ for comparison , in the type iib sn 1993j the characteristic velocity of the ejected hydrogen was about 9000 ( patat et al . 1995 ; utrobin 1996 ; houck & fransson 1996 ) , and in the type iib sn 1996cb it was about 10,000 ( deng et al . a challenge for those who study the complicated evolution of massive stars in binary systems ( e.g. , podsiadlowski , joss , & hsu 1992 ; nomoto , iwamoto , & suzuki 1995 ; wellstein , langer , & braun 2001 ) is to understand why many or perhaps even all stellar explosions that develop strong he i lines should eject at least a small amount of hydrogen . the hydrogen mass that is required to give an h@xmath4 optical depth of unity depends on the fraction of hydrogen that is in the balmer level . in lte , with an electron density of @xmath59 @xmath60 , the balmer fraction peaks around @xmath61 near 6000 k. in this case we estimate that a hydrogen mass on the order of @xmath62 would be required . non lte calculations that take nonthermal excitation into account are needed for a more reliable estimate of the hydrogen mass . the optical depths of the helium lines , and especially the hydrogen lines , are not very high , even though the corresponding absorption features are distinct and fairly deep . this reflects a simple geometric aspect of supernova line formation : as explained in jeffery & branch ( 1990 ) , absorption features formed by detached lines are deeper than those formed by undetached lines . this point is illustrated in figure 24 , which shows that when hydrogen is detached from the photosphere by a factor of two and h@xmath4 has @xmath63 , its absorption feature is deeper than that of an undetached h@xmath4 that has @xmath64 . the h@xmath4 optical depths in our synthetic spectra for sne 2000h , 1999di , and 1954a are not high , so if they were only mildly lower these events would look like typical sne ib . similarly , moderately lower he i line optical depths would transform a sn ib into a sn ic . figure 24 also illustrates how undetached hydrogen lines are more `` obvious '' than detached lines . first , undetached lines have conspicuous narrow , rounded emission peaks while detached lines have inconspicuous broad , flat peaks . note also that although the h@xmath4 absorption is deeper in the detached spectrum , the h@xmath11 absorption is deeper in the undetached spectrum . this is because in the undetached spectrum the optical depth at the photosphere of h@xmath11 is about 2 while in the detached spectrum the optical depth at the detachment velocity is only about 0.4 . an optical depth as low as 0.4 can produce only a shallow absorption , even when the line is detached . for these reasons , supernovae that have undetached hydrogen lines have obvious hydrogen lines and are classified as type ii . supernovae that have detached hydrogen lines are classified as type ib because the presence of hydrogen is not immediately obvious , even when the h@xmath4 absorption is as deep as it is in sne 2000h , 1999di , and 1954a . sne iib are those that have undetached hydrogen lines when they are first observed . in some cases , whether an event is classified as ib or iib may depend on how early the first spectrum is obtained . the implication of the previous paragraphs is that the spectroscopic differences between sne iib , the sne ib that have deep h@xmath4 absorptions , and typical sne ib may be caused mainly by mild differences in the hydrogen mass . for a given kinetic energy , the lower the hydrogen mass the higher the minimum velocity of the ejected hydrogen . similarly , the spectroscopic differences between typical sne ic and sne ib could be caused mainly by moderate differences in the helium mass . for example , matheson et al . ( 2001 ) found higher blueshifts of the o i @xmath327773 line in sne ic than in sne ib . for a given kinetic energy , the lower the helium mass the higher the minimum velocity of the ejected helium , and therefore the higher the velocity of the ejected oxygen . these suggestions are not original to this paper , but they are strengthened by our finding that the h i and he i optical depths in sne ib are not very high . these suggestions also are not inconsistent with arguments , based on light curves , for the existence of different physical classes of hydrogen poor events that cut across the conventional spectroscopic types ( e.g. , clocchiatti & wheeler 1997 ) . the number of sne ib for which good spectral coverage is available is still relatively small . more events should be observed to explore the degree of the spectral homogeneity and to find out whether there is a continuum of hydrogen line strengths . also needed are detailed non lte spectrum calculations for supernova models having radially stratified compositions , to determine the the hydrogen and helium masses and the distribution of the @xmath1ni that is required to excite the helium . the possibility that nonthermally excited ne i can produce spectral features strong enough to be seen needs to be investigated . detailed non lte calculations for parameterized sn ib models , using the phoenix code ( e.g. , baron et al . 1999 ) , are underway . this material is based upon work supported by the national science foundation under grants no . ast9986965 and ast9731450 at oklahoma and ast9987438 at berkeley . is grateful to the guggenheim foundation for a fellowship . 1954a & ngc 4214 & 291 & april 21 + 1983n & ngc 5236 & 513 & july 17 + 1984l & ngc 991 & 1534 & august 22 + 1991ar & ic 49 & * 4520 & * september 2 + 1996n & ngc 1398 & 1491 & * march 12 + 1997dc & ngc 7678 & 3480 & * august 5 + 1998 t & ngc 3690 & * 3080 & * march 3 + 1998dt & ngc 945 & * 4580 & september 12 + 1999di & ngc 776 & * 4920 & july 27 + 1999dn & ngc 7714 & 2700 & august 31 + 2000h & ic 454 & 3894 & february 11 + 1983n & 0703 & -14 & 17000 & & & & + 1983n & 0706 & -11 & 13000 & & & & + 1999dn & 0821 & -10 & 14000 & 2.0 & 14000 & 1.5 & 18000 + 1983n & 0713 & -4 & 11000 & & & & + 1983n & 0717 & 0 & 11000 & 2.5 & 11000 & 0.8 & 15000 + 2000h & 0211 & 0 & 11000 & 1 & 11000 & 6.0 & 13000 + 1999dn & 0831 & 0 & 10000 & 5.0 & 11000 & 1.0 & 14000 + 1983n & 0719 & 2 & 10000 & 4.0 & 10000 & 0.7 & 14000 + 2000h & 0216 & 5 & 8000 & 2.0 & 9000 & 2.5 & 13000 + 1984l & 0830 & 8 & 9000 & 2.0 & 10000 & 0.6 & 15000 + 1998dt & 0920 & 8 & 9000 & 4.0 & 11000 & & + 1984l & 0831 & 9 & 8000 & 1.5 & 10000 & 0.6 & 15000 + 1983n & 0727 & 10 & 7000 & 7.0 & 8000 & 0.6 & 12000 + 1999dn & 0910 & 10 & 7000 & 3.0 & 8000 & & + 1984l & 0903 & 12 & 7000 & 1.0 & 9000 & 0.5 & 14000 + 1999dn & 0914 & 14 & 6000 & 10 & 7000 & & + 1999dn & 0917 & 17 & 6000 & 10 & 8000 & & + 2000h & 0302 & 19 & 6000 & 5.0 & 8000 & 2.0 & 13000 + 1999di & 0817 & 21 & 7000 & 10 & 8500 & 4.0 & 12000 + 1984l & 0919 & 28 & 5000 & & & & + 2000h & 0313 & 30 & 5000 & 3.0 & 7000 & 1.5 & 13000 + 1984l & 0923 & 32 & 5000 & 5.0 & 7000 & & + 1998dt & 1015 & 33 & 9000 & 10 & 9000 & & + 1984l & 0928 & 37 & 5000 & 10 & 6000 & & + 1999dn & 1008 & 38 & 6000 & 10 & 7000 & & + 1999di & 0910 & 45 & 6000 & 10 & 7000 & 2.0 & 12000 + 2000h & 0330 & 47 & 5000 & 1.0 & 7000 & 0.5 & 12000 + 1999di & 0917 & 52 & 6000 & 10 & 7000 & 1.0 & 12000 + 2000h & 0408 & 56 & 4000 & & & & + 1984l & 1018 & 57 & 4000 & & & & +

the last four years more understanding has been gained of the interplay between instantons and monopoles in non - abelian gauge theories , based on the ability to construct exact caloron solutions , i.e. instantons at finite temperature for which @xmath3 approaches a constant at spatial infinity @xcite . this last condition is best expressed by specifying the polyakov loop to approach a constant value at infinity , also called the holonomy . it is the finite action that demands the field strength to go to zero at infinity , and guarantees the polyakov loop to be independent of the direction in which we approach infinity . one can parametrize this polyakov loop in terms of its eigenvalues @xmath4 and a gauge rotation @xmath5 , such that = _ x(_0^a_0(x , t)dt)=g^(2i(_1,_2, ,_n))g , which can be arranged such that @xmath6 , and @xmath7 , with @xmath8 . this is in the gauge where the gauge fields , assumed to be anti - hermitian matrices taking values in the algebra of @xmath1 , are periodic in the time direction @xmath9 . in our conventions the field strength is given by @xmath10 $ ] . we will find it convenient to construct the multi - caloron solutions from the adhm - nahm fourier construction @xcite based on taking instantons in @xmath11 , which periodically repeat in the time direction ( see also ref . @xcite which uses directly the nahm transformation @xcite as the starting point for the charge 1 construction ) . to allow for non - trivial holonomy , the periodicity is only up to a constant gauge rotation ( which is the holonomy ) . in this so - called algebraic gauge , all gauge field components vanish at spatial infinity , and we may approximately superpose these calorons by simply adding the gauge fields . when each gauge field is periodic up to the same constant gauge transformation , the sum satisfies the same property . it should be noted that we are not allowed to add gauge fields with different holonomy ; in an _ infinite _ volume the holonomy is fixed by the boundary condition . as to the topological charge , we recall that in the atiyah - drinfeld - hitchin - manin ( adhm ) construction @xcite it is supported by gauge singularities ( the algebraic gauge is for this reason also called the singular gauge ) , as opposed to at infinity being a pure gauge with the gauge function having the appropriate winding number . to deal with the gauge singularity of one instanton , when adding the field of the others , one has to smoothly deform the gauge field of the latter to vanish near the gauge singularity . as long as the singularities are not too close , this can be done without a significant increase in the action . on the lattice this problem does not occur , when hiding singularities between the meshes of the lattice . 4.1 cm calorons , however , have an additional feature . when squeezed in the imaginary time direction ( the size @xmath12 becoming bigger than the period @xmath13 ) , they split in constituent monopoles with masses @xmath14 ( @xmath15 ) . outside the cores of these monopoles the gauge field becomes abelian . ignoring the charged components , which decay exponentially outside the cores of the constituent monopoles , the field is described in terms of self - dual dirac monopoles . the singularity of a dirac string can only be avoided by _ not _ neglecting the contributions coming from the charged components , even when far away from the constituents . for a single caloron we would not care , since the dirac string is not seen in gauge invariant quantities . it does , however , involve a rather subtle interplay between the charged and neutral components of the gauge field in the vicinity of the would - be dirac string @xcite . it is this subtle interplay that is disturbed when we add gauge fields of various calorons together . unlike for the gauge singularity , the combined field will not diverge , but it shows a narrow and steep enhancement at the location of the would - be dirac string as illustrated in fig . [ fig : string ] , where we added two @xmath16 calorons . also here one may shield the dirac strings from these tails . but one always pays the price that the dirac string no longer can be hidden , and carries energy . let us stress again that these are genuine gauge invariant non - singular features in the configuration , even though they are of course a consequence of our particular way of constructing a superposition . a visible would - be dirac string presents a formidable obstacle to considering the constituents as independent objects ; they remember to which caloron they belong . insisting the abelian field far from the constituents to be exactly additive under the approximate superposition leaves us little room for considering other possible superpositions , apart from carefully fine - tuning the charged components of the configuration . solving for the exact self - dual caloron solutions of higher charge we wish to show that a visible dirac string is an artifact of the particular procedure to construct approximate caloron solutions . for the exact charge @xmath0 caloron solutions we expect @xmath17 constituents and from the point of view of the parameter space these can be expected to be independent ( as long the constituents do not get too close together ) . 3.0 cm in this paper we develop the formalism and give exact solutions for a class of axially symmetric solutions . for @xmath16 an example of such an exact solution with charge 2 is presented in fig . [ fig : wellsep ] . we compare the exact solution ( left ) , with the one obtained by adding two charge 1 calorons , showing the would - be dirac strings ( middle ) and with the exact abelian solution determined from ( self - dual ) dirac monopoles placed at the location of the constituents ( right ) . indeed , the would - be dirac strings are no longer visible for the exact non - abelian solution . this gives us good reasons to expect that moving away from the requirement of axial symmetry , the exact solutions will exhibit the constituent monopoles as independent objects . this is an important prerequisite for attempting to formulate the long distance features of qcd in terms of these monopoles . non - linearities will remain important when constituents overlap , like for instantons at zero - temperature , as will be illustrated through suitable examples . our results can also be used to extract information on multi - monopole solutions . in the charge 1 case it has been shown @xcite that sending one of the constituents to infinity one is left with exact monopole solutions . perhaps somewhat surprisingly , it has been notoriously difficult to construct approximate superpositions of magnetic monopoles in such a way the interaction energy decreases with their separation . within our approach this is due to the difficulty ( particularly when sending some constituents to infinity ) in keeping dirac strings hidden . it would therefore still be important to find approximate superpositions that achieve this . nevertheless , it shows how subtle it is to consider abelian fields as embedded in non - abelian ones . there is much more to t hooft s abelian projection @xcite than meets the eye at first glance . the rest of the paper will be organized as follows . in sect . 2 we formulate the fourier analysis of the adhm data for calorons of higher charge . this allows us in sect . 2.1 to relate to the nahm equations , providing the connection between the adhm and nahm data . the nahm equation for charge @xmath0 calorons @xcite expresses self - duality of dual @xmath18 gauge fields on a circle , but with singularities . in sect . 2.2 we introduce the `` master '' green s function in terms of which the gauge field ( sect . 2.3 ) and the action density ( sect . 2.4 ) can be explicitly computed , relying heavily on some beautiful results @xcite derived in the context of the adhm construction . this can all be achieved without explicitly solving the nahm equation . in sect . 3 we will , however , solve it for some special cases with axial symmetry . there we also illustrate some interesting features that appear when constituents overlap , similar to what was observed at zero temperature @xcite . sect . 4 is devoted to the far - field limit ( related to the high - temperature limit ) , where up to exponential corrections only the abelian component of the field survives . again , expressions can be derived without explicitly solving the nahm equation . we apply the formalism to the exact axially symmetric solutions in sect . 4.3.1 , where we prove that in the far - field region they are in a precise sense described by ( self - dual ) dirac monopoles , which is conjectured to be the case for any exact solution . we end in sect . 5 with a discussion of lattice results that have been obtained by various groups , some open questions and suggestions for future studies . in constructing the higher charge caloron solutions similar steps are followed as for charge 1 . we distinguish two constructions , @xmath19 based on quaternions , and @xmath1 based on complex spinors . with @xmath20 , this implies that for @xmath16 two constructions are possible . for @xmath21 it was easily shown these are identical , but for @xmath22 this is more complicated due to the quadratic adhm constraint @xcite which can not be solved in all generality . since the essential part of the construction that employs the impurity scattering technique is identical in both approaches , we will give a unified presentation . the @xmath1 adhm formalism for charge @xmath0 instantons @xcite employs a @xmath0 dimensional vector @xmath23 , where @xmath24 is a two - component spinor in the @xmath25 representation of @xmath1 . alternatively , @xmath26 can be seen as a @xmath27 complex matrix . in addition one has four complex hermitian @xmath28 matrices @xmath29 , combined into a @xmath30 complex matrix @xmath31 , using the unit quaternions @xmath32 and @xmath33 , where @xmath34 are the pauli matrices . with some abuse of notation , we often write @xmath35 . together @xmath26 and @xmath36 constitute the @xmath37 dimensional matrix @xmath38 , to which is associated a complex @xmath39 dimensional normalized zero mode vector @xmath40 , ( x)= ( + b(x)),b(x)=b - x,^(x)v(x)=0,v^(x)v(x)=_n . here the quaternion @xmath41 denotes the position ( a @xmath28 unit matrix is implicit ) and @xmath40 can be solved explicitly in terms of the adhm data by v(x)=(-_n + u(x))^-,u(x)= ( b^-x^)^-1^,(x)=_n+u^(x)u(x ) , as @xmath42 is an @xmath43 positive hermitian matrix , its square root @xmath44 is well - defined . the gauge field is given by a_(x)=v^(x)_v(x)=^-(x)(u^(x ) _ u(x))^-(x)+^(x)_^-(x ) . [ eq : aadhm ] the @xmath45 adhm formalism for charge @xmath0 instantons @xcite employs also a @xmath0 dimensional vector @xmath23 , where now @xmath46 is a quaternion ( 4 real parameters ) . again , @xmath26 can be seen as an @xmath47 complex matrix ( but with @xmath48 real , as opposed to complex , parameters ) . now the four @xmath28 matrices @xmath29 are required to be real and symmetric , still to be combined into a @xmath30 complex matrix @xmath35 . the @xmath49 dimensional matrix @xmath38 is constructed as before . it is immediately obvious that now @xmath42 is proportional to @xmath50 , simplifying the expression for the gauge field to a_(x)=v^(x)_v(x)=(u^(x)_u(x))/(x ) . for @xmath51 to be a self - dual connection , @xmath38 has to satisfy the quadratic adhm constraint , which states that @xmath52 ( considered as a @xmath28 complex quaternionic matrix ) has to commute with the quaternions , or equivalently ^(x)(x)=_0f^-1_x,[eq : adhmconstr ] defining @xmath53 as a hermitian ( resp . symmetric ) @xmath28 green s function . the ity follows by computing the curvature f_=2 ^ -(x)u^(x)_f_x u(x)^-(x ) , making essential use of the fact that @xmath53 commutes with the quaternions , and @xmath54}$ ] being ( @xmath55}$ ] is ) . the quadratic constraint can be formulated as @xmath56 , where @xmath57 , and one obtains |_b_b_+_a _ 2(_a^)= 0,[eq : radhmcon ] where @xmath58 is the spinorial trace . note that this implies that @xmath59 is traceless for @xmath60 . to count the number of instanton parameters we observe that the transformation @xmath61 , @xmath62 , with @xmath63 ( resp . @xmath64 ) leaves the gauge field and the adhm constraint untouched . taking this symmetry into account , one checks the dimension of the instanton moduli space to be @xmath65 dimensional . we have @xmath65 and @xmath66 real parameters from @xmath26 and @xmath29 . the @xmath67 symmetry removes @xmath68 real parameters and finally the quadratic adhm constraint gives @xmath69 real equations . on the other hand , for the quaternionic construction there are @xmath48 and @xmath70 parameters from @xmath26 and @xmath29 , of which @xmath71 are removed by the @xmath72 symmetry , and the quadratic adhm constraint gives @xmath73 equations . global gauge transformations are realized by @xmath74 , with @xmath75 and are here included in the parameter count . for calorons with non - trivial holonomy the dimension of the gauge invariant parameter space is minimally reduced by @xmath76 ( maximal symmetry breaking ) and maximally by @xmath77 ( trivial holonomy ) . we note that for @xmath16 the quadratic adhm constraint and symmetry of the adhm data for high charge differ considerably . for the caloron we have to deal in a sense with infinite topological charge , but finite within each ( imaginary ) time interval of length @xmath13 . this infinity is resolved by fourier transformation , relating it to the nahm formalism @xcite , but the difference between the @xmath67 and @xmath72 symmetries ( the infinity is moved to making these gauge symmetries local ) remains , as well as of course the nature of the nahm data ( the fourier transformation of the adhm data ) . henceforth we put @xmath78 , which can always be achieved by a rescaling . like for charge one @xcite the caloron with polyakov loop @xmath79 at infinity is built out of a periodic array of instantons , twisted by @xmath79 . this is implemented in the adhm formalism by requiring ( suppressing color and spinor indices , respectively quaternion indices ) u_pk+k+a(x+1)=u_pk+a(x)_^-1 [ eq : ucocyc ] with @xmath80 ( the fourier index ) , and @xmath81 ( associated to the non - abelian nature of the nahm data ) . using that @xmath82 , leads to the required periodicity . demanding _ pk+k+a=_pk+a , b_pk+a , qk+b = b_pk - k+a , qk - k+b+ _ 0_pq_ab , suitably implements and is partially solved by imposing _ pk+a=^p__a , b_pk+a , qk+b = p_0_pq _ ab+a^ab_p - q , with @xmath83 still to be determined to account for . it is useful to introduce the @xmath84 projectors @xmath85 on the @xmath86th eigenvalue of @xmath79 , such that @xmath87 and @xmath88 . we now perform the fourier transformation to the nahm setting @xcite , which casts @xmath36 into a weyl operator and @xmath89 into a singularity structure on @xmath90 , & & _ p , q b_pk+a , qk+b(x ) e^2i(pz - qz)= d^ab(z),2i_p e^2ipza^ab_p = a^ab(z ) , + & & _ p , q^_pk+a e^2i(pz - qz)^ _ qk+b=(z - z)_ab(z),_p e^-2 ipz_pk+a=_a(z ) , with @xmath91 , we can write @xmath92 and @xmath93 , where & & d^ab_x(z)d^ab(z)-2ix^ab= ^ab(_0 - 2ix)+a^ab(z ) , + & & _ ab(z)=_m(z-_m)_a^p_m_b , _ a(z)=_m(z-_m)p_m_a.it should be noted that @xmath29 hermitian , implies that @xmath94 is hermitian ( as a @xmath28 matrix ) , whereas a real symmetric @xmath29 in addition implies @xmath95 . the @xmath96 matrix @xmath97 can always be decomposed as _ s^ab_m-^ab_m).[eq : lam ] on the diagonal one can show , as for @xmath21 , that @xmath98 , but in general the relation between @xmath99 and @xmath100 is more complicated , s_m^abs_m^cd+s_m^ads_m^cb=_m^ab _ m^cd+_m^ad_m^cb . furthermore , @xmath101 is traceless implies that @xmath102 . both these conditions are equally valid for the @xmath45 construction . but in the latter case , since through the nahm equation @xmath103 determines the discontinuities in @xmath104 , compatibility with @xmath105 requires @xmath106 , which will be verified below . the nahm equation is obtained by fourier transforming the quadratic adhm constraint [ d_(z),d_(z)]|_=4 ^ 2(z ) , or in more familiar form ( as a @xmath28 matrix equation ) a_j(z)+[a_0(z),a_j(z)]+_jk[a_k(z ) , a_(z)]=2i_m(z-_m)_m^j,[eq : nahm ] ( 10,1.5)(-1.6,-1.0 ) ( 0.,0)@xmath107 ( 1.2,0)(1,0)0.2 ( 1.6,0)(1,0)0.2 ( 1.25,-0.5)@xmath108 ( 2,0)(1,0)3 ( 2,-0.25)(0,1)0.5 ( 2,0 ) ( 2,-0.5)@xmath109 ( 1.7,0.5)@xmath110 ( 3.75,-0.25)(0,1)0.5 ( 3.75,0 ) ( 3.45,-0.5)@xmath111 ( 3.45,0.5)@xmath112 ( 5,-0.25)(0,1)0.5 ( 5,0 ) ( 4.7,-0.5)@xmath113 ( 4.7,0.5)@xmath114 ( 5.2,0)(1,0)0.2 ( 5.6,0)(1,0)0.2 ( 6.0,0)(1,0)0.2 ( 6.4,0)(1,0)0.2 ( 6.8,0)(1,0)0.2 ( 7.25,0)(1,0)0.2 ( 7.7,0)(1,0)0.2 ( 8.1,0)(1,0)0.2 ( 8.5,0)(1,0)2.5 ( 8.5,-0.25)(0,1)0.5 ( 8.5,0 ) ( 7.8,-.5)@xmath115 ( 7.8,.5)@xmath116 ( 9.5,-0.25)(0,1)0.5 ( 9.5,0 ) ( 9.2,-.5)@xmath117 ( 9.2,.5)@xmath118 ( 11.0,0 ) ( 11.,-0.25)(0,1)0.5 ( 10.7,-.5)@xmath119 ( 10.7,.5)@xmath120 ( 11.0 , 0)(1,0)1.0 ( 12.0,-0.25)(0,1)0.5 ( 12.0,0 ) ( 11.6,-.5)@xmath121 ( 11.7,.5)@xmath122 ( 12.2 , 0.)(1,0)0.2 ( 12.6 , 0.)(1,0)0.2 ( 14.3,0 . ) . + where the figure illustrates the jumps of @xmath123 at each of the singularities ( an overall factor of @xmath124 is not shown ) . the @xmath125 symmetry in the adhm construction translates into a @xmath67 gauge symmetry on @xmath90 , which allows one to set @xmath126 to be constant , where @xmath127 is a hermitian matrix which can be made diagonal by a constant gauge transformation . its trace part can be absorbed in @xmath128 . central to the adhm construction is the green s function @xmath53 , which when fourier transformed to @xmath129 is a solution to the differential equation \{()^2 + _ m(z-_m)s_m}f_x(z , z)= _ k(z - z),[eq : df ] ( 10,2.0)(-1.5,-1.1 ) ( 0.,0.)@xmath130 ( 1.2,0)(1,0)0.2 ( 1.6,0)(1,0)0.2 ( 1.25,-0.5)@xmath108 ( 2,0)(1,0)3 ( 2,-0.25)(0,1)0.5 ( 2,0 ) ( 2,-0.5)@xmath109 ( 1.7,0.5)@xmath131 ( 3.75,-0.25)(0,1)0.5 ( 3.75,0 ) ( 3.45,-0.5)@xmath111 ( 3.45,0.5)@xmath132 ( 5,-0.25)(0,1)0.5 ( 5,0 ) ( 4.7,-0.5)@xmath113 ( 4.7,0.5)@xmath133 ( 5.2,0)(1,0)0.2 ( 5.6,0)(1,0)0.2 ( 6.0,0)(1,0)0.2 ( 6.4,0)(1,0)0.2 ( 6.5,-0.25)(0,1)0.5 ( 6.5,0 ) ( 6.1,.5)@xmath134 ( 6.5,-0.5)@xmath135 ( 6.8,0)(1,0)0.2 ( 7.25,0)(1,0)0.2 ( 7.7,0)(1,0)0.2 ( 8.1,0)(1,0)0.2 ( 8.5,0)(1,0)2.5 ( 8.5,-0.25)(0,1)0.5 ( 8.5,0 ) ( 7.8,-.5)@xmath115 ( 7.8,.5)@xmath136 ( 9.5,-0.25)(0,1)0.5 ( 9.5,0 ) ( 9.2,-.5)@xmath117 ( 9.2,.5)@xmath137 ( 11.0,0 ) ( 11.,-0.25)(0,1)0.5 ( 10.7,-.5)@xmath119 ( 10.7,.5)@xmath138 ( 11.0 , 0)(1,0)1.0 ( 12.0,-0.25)(0,1)0.5 ( 12.0,0 ) ( 11.6,-.5)@xmath121 ( 11.7,.5)@xmath139 ( 12.2 , 0.)(1,0)0.2 ( 12.6 , 0.)(1,0)0.2 ( 14.3,0 . ) . + where the figure illustrates the jumps for the _ derivative _ of @xmath140 ( itself being continuous ) at each of the singularities , which now includes @xmath141 ( an overall factor of @xmath142 is omitted ) . with help of the impurity scattering formalism we will be able to express the solution in a simple form , without assuming explicit knowledge of @xmath104 . as a bonus this also gives a more transparent derivation for @xmath21 . the equation for the green s function is valid for the @xmath1 and @xmath45 formulations alike . in general the matrix @xmath53 is complex hermitian , but for @xmath45 it is real symmetric , which implies that f_x^ab(z , z)=f_x^ba(z,z)^*,f_x^ab(z , z)=f_x^ba(-z,-z ) . [ eq : fsymm ] for @xmath45 consistency requires , @xmath143 , which is to be compared with the condition we found in the previous section , @xmath144 . to demonstrate the validity of these relations we make use of the fact that @xmath145 , with @xmath146 real ( such that @xmath147 ) . furthermore , we note that with @xmath148 , @xmath149 , whereas @xmath150 and @xmath151 . introducing @xmath152 , which is an imaginary unit quaternion ( i.e. @xmath153 , @xmath154 ) , we find with the help of _ 0s^ab_m-^ab_m&=&_a^|_(_0-i(-1)^m)^_b_[eq : rhos ] + & = & _ a^_b^(+ ) + & = & _ 0+,for which we used that @xmath155}$ ] is a real quaternion , therefore given by @xmath156})={{\scriptstyle{{1\over 2}}}}\sigma_0\tr_2(\omega\eta_{\nu\mu})=\sigma_0\hat\omega \cdot\vec\eta_{\mu\nu}$ ] . we directly read off @xmath100 and @xmath103 and verify that indeed @xmath157 and @xmath158 . we wrote the result so as to easily make contact with the @xmath21 results @xcite ( for which the 2nd term in @xmath100 and the 1st term in @xmath103 do not contribute ) . in formulating the impurity scattering we note that the @xmath67 gauge transformation g(z)=(2i(_0-x_0_k)z ) turns @xmath159 into the ordinary derivative and conjugates all other objects in , such that \{-+v(z;x ) } f_x(z , z)=4 ^ 2 _ k(z - z),[eq : dftrans ] with @xmath160 and @xmath161 given by f_x(z , z)g(z)f_x(z , z)g^(z),v(z;x ) 4 ^ 2r^2(z;x)+2_m(z-_m)s_m,[eq : ftofh ] where @xmath162 and @xmath163 are defined by r_j(z;x)x_j_k - g(z)a_j(z)g^(z),s_mg(_m)s_mg^(_m).[eq : rdef ] periodicity is now only up to a gauge transformation . note that @xmath164 is a hermitian @xmath28 matrix , and that @xmath161 does _ not _ depend on @xmath128 . by combining @xmath53 and its derivative into a vector , f_x(z , z ) , we can turn the second order equation in to a first order equation \{-}f_x ( z , z)=c(z - z),c-4 ^ 2 its solution is f_x(z , z)=w(z ) , w(z)=w(z ) where @xmath165 is a @xmath96 matrix , whereas @xmath166 is a two - component vector ( like @xmath167 ) , all components being hermitian @xmath28 matrices . the theta function takes care of the inhomogeneous part of the equation . however , we have to restrict to @xmath168 $ ] since the delta function is periodic . the solution can be extended beyond this range using the periodicity , which when imposed , as we will see , also determines @xmath169 . the solution for @xmath170 can be ( formally ) written as a path ordered exponential integral in the usual way . to show that this indeed makes sense , we should specify how to deal with the delta functions in @xmath171 . from the definition of the path ordered exponential integral we find @xmath172 with t_m== , which correctly reflects the matching conditions due to the impurity " at @xmath173 , since @xmath160 is continuous across @xmath173 , whereas the derivative jumps with @xmath174 and both conditions can be summarized by @xmath175 . note that @xmath176 evolves with @xmath177 as @xmath178 , with w(z_2)w^-1(z_1)=w(z_2,z_1).[eq : wdef ] in particular , when @xmath179 and @xmath180 this gives the propagation " between two neighboring impurities and we can write h_mw(_m+1 - 0,_m+0)=.[eq : hm ] neither @xmath181 nor @xmath182 require us to specify a boundary condition for @xmath170 . a change in boundary condition , however , affects @xmath169 . to avoid such ambiguities , we define @xmath183 such that f_x(z , z)=w(z , z_0).[eq : fxz0 ] we determine @xmath184 by scattering around " the circle determined from the boundary conditions of @xmath176 . using , the fact that @xmath185 is strictly periodic and that @xmath186 , one finds @xmath187 and @xmath188 ( where the order of the _ matrix _ multiplication , defined componentwise , is important ) . the latter condition can be used to fix @xmath184 over the range of one period , @xmath189 $ ] ( the use of @xmath185 requires @xmath168 $ ] , that of @xmath190 further restricts the range for @xmath135 to @xmath191 ) . with the help of the periodicity properties of @xmath165 it gives @xmath192 . we thus find for _ arbitrary _ @xmath193 f_x(z , z)=w(z , z_0)\{(_2k-_z_0)^-1-(z-z)_2k } w^-1(z,z_0)c,[eq : vecf ] where we introduced the holonomy " @xmath194 _ z_0g^(1)w(z_0 + 1,z_0)=g^(1 ) . the equation for @xmath176 is valid for @xmath195 $ ] , but can be extended with the appropriate periodicity specified above . we note that @xmath196 plays the role of a cocycle , in the gauge where @xmath197 is transformed to 0 , with @xmath194 the full circle scattering " matrix . the @xmath193-independence of @xmath176 follows from _ z_0&=&g^(1)w(z_0 + 1,z_0^+1)w(z^_0 + 1,z_0)= w(z_0,z^_0)g^(1)w(z^_0 + 1,z_0 ) + & = & w(z_0,z^_0)_z_0^w^-1(z_0,z^_0).[eq : frel ] putting things together we therefore find f_x(z , z)&=&g^(z)(_k,0)f_x(z , z ) g(z ) , + d_x^0(z)f_x(z , z)&=&g^(z ) ( 0,_k)f_x(z , z)g(z),[eq : finf ] satisfying all required conditions as can be checked explicitly . interestingly , implies @xmath198 as should of course be the case . in the light of this we also note that ( choosing @xmath199 ) _ _ m&=&g^(1)t_m+nh_m+n-1t_m+n-1h_m+n-2 t_m+1h_m + & = & t_mh_m-1t_2h_1t_1g^(1)h_n t_nh_n-1t_m+1h_m,[eq : ths ] ( 10,2.0)(-2.8,-1.1 ) ( -0.2,0.)@xmath200 ( 0.8,-0.5)@xmath201 ( 2,0)(1,0)2 ( 2,-0.25)(0,1)0.5 ( 2,0 ) ( 1.7,0.5)@xmath181 ( 2.5,0.2)@xmath202 ( 4,0 ) ( 4,-0.25)(0,1)0.5 ( 3.4,-0.5)@xmath203 ( 3.7,0.5)@xmath204 ( 4.0,0)(1,0)0.2 ( 4.4,0)(1,0)0.2 ( 4.8,0)(1,0)0.2 ( 5.2,0)(1,0)0.2 ( 5.6,0)(1,0)0.2 ( 6.0,0)(1,0)0.2 ( 6.4,0 ) ( 6.4,-0.25)(0,1)0.5 ( 5.7,.5)@xmath205 ( 5.8,-0.5)@xmath206 ( 6.4,0)(1,0)2 ( 7.2,0.2)@xmath207 ( 8.4,-0.25)(0,1)0.5 ( 8.4,0 ) ( 8.1,.5)@xmath208 ( 8.1,-0.5)@xmath119 ( 8.4,0)(1,0)0.2 ( 8.8,0)(1,0)0.2 ( 9.2,0)(1,0)0.2 ( 9.6,0)(1,0)0.2 ( 10,0)(1,0)2 ( 10,-0.25)(0,1)0.5 ( 10,0 ) ( 9.7,-.5)@xmath209 ( 9.7,.5)@xmath210 ( 10.0,0 ) ( 10.9,0.2)@xmath182 ( 12.0,0 ) ( 12.,-0.25)(0,1)0.5 ( 11.7,-.5)@xmath211 ( 12.2,0 . ) , + using for the second identity that @xmath212 and @xmath213 . we will use these ingredients further on to relate to the earlier results for @xmath21 , where the positioning of @xmath214 is of course irrelevant . the central role of the green s function @xmath185 becomes clear when one appeals to the fact that it can be used to find the gauge field ( working out ) , whereas its determinant gives a simple expression for the action density . this follows from the general adhm construction @xcite , and can be directly taken over for the caloron @xcite . for the gauge field one finds a_(x)&=&^(x)|__f_x ^^(x)+[^-(x),_^(x ) ] + & = & ^(x)|_^j__j(x ) ^(x)+[^-(x),_^(x ) ] , [ eq : adef ] where @xmath42 and @xmath215 are @xmath43 matrices defined by ( x)(_n - f_x^)^-1 , _ j_j f_x^. to apply this to the caloron all we have to do is perform the fourier transformation , ( x)^-1=_n-_m , mp_m_a f_x^ab(_m,_m ) ^_bp_m,_j=_m , mp_m_a _ j f_x^ab(_m,_m)^_bp_m.[eq : phidef ] for the @xmath45 construction @xmath42 is a real quaternion , and hence a multiple of @xmath50 , after which the gauge field simplifies to a_(x)=(x)|_^j__j(x).[eq : asp1 ] to simplify @xmath42 and @xmath215 we use , together with @xmath216 , such that @xmath217 and @xmath218 . we note that @xmath42 and @xmath215 involve the combinations @xmath219 , which can be split in symmetric and anti - symmetric combinations ( cf . ) _ a_0|_b&=&_0_a^_b^+ _ _ a^_b^ , + _ a_j|_b&=&_0|^j__a^_b^+_\{a_j|_b}. no contributions from @xmath220 and @xmath221 can appear in @xmath42 since these are symmetric in @xmath222 and @xmath223 , selecting from @xmath224 the term proportional to @xmath50 , but @xmath225 and @xmath226 vanish . therefore ( cf . ) & -2cm(x)^-1=_0-f_x^ab(_2,_2)_m=1 ^ 2 ( p_m_a^_b^p_m+(-1)^mp_m__a^_b^p_m ) = [ eq : phis ] + & _ 0)=_0.similarly we can simplify the expression for @xmath215 , which we split in a charged component and abelian , or neutral , component @xmath227 with _ j^ch = f_x^ab(_1,_2)p_1_\{a_j|_b } p_2- and & -3cm_j^abel(x)=f_x^ab(_2,_2)_m=1 ^ 2 ( p_m_\{a_j|_b}p_m+(-1)^mp_m|_^j _ a^_b^p_m)[eq : vphis ] + & = if_x^ab(_2,_2)=-i^-1_k(f_x(_2,_2)_2^j ) , where we used that @xmath228 to correctly identify @xmath229 , see . we may of course express @xmath42 and @xmath230 also in terms of the first impurity , @xmath231 $ ] and @xmath232 , as is easily verified . like for @xmath21 we will show further on that @xmath233 decays exponentially , away from the cores of the constituent monopoles , where only the abelian component survives , a_^abel(x)=-^-1|^j__. note that for @xmath21 we are able to write @xmath234 , with @xmath235 , making use of the fact that @xmath236 . this seems to allow for the dipoles of @xmath0 well - separated calorons to point in different directions . we will discuss this issue further when studying the limiting behavior far from the cores of the constituents , where the field becomes algebraic and constituent locations are readily identified . within the adhm formalism the action density is given by @xcite , -_n f_^2(x)=-_^2_^2 ( x),[eq : osborn ] where @xmath237 equals @xmath238 after regularization to extract an irrelevant overall and for calorons divergent constant . we will be able to find a simple expression for @xmath237 at any @xmath0 , generalizing the result for charge 1 calorons . we use that @xmath239 , where in the last step we performed the fourier transformation , and @xmath240 includes an integration with respect to @xmath177 . the case @xmath241 is treated separately , due to the discontinuity in the @xmath177 derivative of @xmath185 at @xmath141 , which we regularize using point - splitting : ( d_x^0f_x)&=&_0_0 ^ 1dz _ k(f_x(z+,z)+f_x(z-,z ) ) _ z=z + & = & 4i_0 ^ 1dz ( w^-1(z , z_0 ) w(z , z_0 ) ) + & = & 2i_0 ^ 1dz . the tr without an index or hat indicates the full trace over the @xmath242 matrix involved . to see how the total derivative term appears ( not contributing to the integral due to the periodicity of @xmath243 $ ] ) we use that _ k(f_x(z , z))=-4 ^ 2(w^-1(z , z_0)w(z , z_0 ) ) + = -4 ^ 2(w^-1(z , z_0)w(z , z_0 ) ) + = -4 ^ 2 ( w^-1(z , z_0)w(z , z_0 ) ) , with @xmath244 ( as for point - spitting , although one checks that the @xmath245 dependent term actually vanishes ) . we note that @xmath194 depends on @xmath128 only through @xmath246 , and that @xmath247 , such that @xmath248 . with this we find _ 0(f_x)=-_0,(ie^-ix_0(_2k-_z_0)/),[eq : defpsi ] which is _ independent _ of @xmath193 . the factor @xmath249 in the argument of the determinant was inserted just so @xmath250 agrees with the definition introduced earlier for @xmath21 . next we compute @xmath251 , _ 0 ^ 1 dz _ k(d_x^jf_x(z , z))=-2_0 ^ 1 dz _ k(r_j(z;x)f_x(z , z))= + 8 ^ 2_0 ^ 1 dz ( w^-1(z , z_0)w(z , z_0))=-4 mm + ( ( _ 2k-_z_0)^-1_z_0_z_0 ^ 1+z_0dz w^-1(z , z_0)_jw(z , z_0 ) ) , where again @xmath245 can take any value , but for convenience is best set to 1 here . finally noting that @xmath194 only depends on @xmath252 through @xmath253 and using that w^-1(z , z_0)_j w(z , z_0)=_z_0 ^ 1+z_0 dz w^-1(z , z_0 ) _ jw(z , z_0 ) , we verify that @xmath254 for all @xmath255 and @xmath0 . it is amusing to note that this implies the remarkable formula -_n f_^2(x)=-_^2_^2 ( _ 2k - g^(1 ) ) , even though explicit evaluation can be quite cumbersome . not so for some special cases , including the single caloron @xmath21 , where @xmath162 is _ piecewise constant _ as we will discuss next . consider @xmath162 to be piecewise constant , for @xmath256 $ ] defined to be @xmath257 ( cf . ) , with @xmath258 constant hermitian @xmath28 matrices related to the nahm potential by @xmath259 . in this case we can easily deal with the path ordered exponential integrals . to be specific , the propagation " from @xmath211 to @xmath209 defined through @xmath182 in , is given by h_m=,[eq : conhm ] where @xmath260 ( with @xmath261 such that @xmath262 ) and @xmath263 ( a hermitian @xmath28 matrix ) . since @xmath264 and @xmath265 are both quadratic in @xmath266 , actually no square root is involved in this expression for @xmath182 . for charge 1 the nahm equation ( ) has no commutator terms and @xmath162 is always piecewise constant . it reduces to the @xmath267 constituent radius for @xmath268 $ ] , @xmath269 , whereas the prefactor at the impurity becomes @xmath270 . with _ mr_m^-1 , a link to the earlier charge 1 results @xcite is established by noting that _ m = t_m+1h_m ^-1 . with the placing of @xmath214 irrelevant , and the possibility of absorbing @xmath127 in @xmath128 , we therefore find _ _ m^k=1=e^2ix_0_m-1_m-2 _ 1_n_m+1_m^-1 , cf . . in particular this shows that @xmath271 agrees with the result found earlier , @xmath272 . the green s functions can be shown to agree as well , using @xmath273 . arranging @xmath162 to be piecewise constant when @xmath22 requires one to fulfill some constraints . to solve the nahm equation , , in terms of the @xmath274 , the commutator term should vanish . one way to achieve this , is by choosing @xmath275 . the nahm equation relates the discontinuities of @xmath276 to @xmath103 , g^(_m)(y_m - y_m-1)g(_m)e=_m , [ eq : jumps ] which imposes constraints on @xmath277 , see . to seek a solution we choose all @xmath278 to be parallel in group space , @xmath279 ( @xmath280 a positive real number ) this reduces the problem to @xmath21 , since @xmath281 is proportional to @xmath282 . for @xmath16 this already solves the constraint , since for @xmath21 one has @xmath283 . for @xmath284 it has been shown @xcite that @xmath99 can take any value , provided @xmath285 , in particular we may choose all @xmath103 to be proportional to @xmath286 ( by properly choosing @xmath287 ) . for @xmath21 it is convenient to parametrize @xmath288 in terms of constituent locations , @xmath289 . as in sect . [ sec : gaugefield ] we will take @xmath235 . we obtain a larger class of @xmath278 for which the @xmath103 are parallel , by taking @xmath278 to be parallel up to a gauge rotation with an element of the unbroken subgroup @xmath290 which leaves the holonomy unchanged , _ a^m p_m,_n_a=0 . this leads to _ m^ab=^a^b(2i(_b^m-_a^m ) ) y_m , s_m^ab=_m^aby_m/|y_m|.[eq : srho ] note that for @xmath45 we verify that @xmath291 and @xmath292 ( using @xmath293 ) . with @xmath294 for all @xmath86 ( by definition @xmath295 ) we may solve , y_m^ab=(_a+_a^2 y_m)_ab+i(1-_ab)_a_b _ j= 1^ny_j,[eq : ymfull ] where the @xmath296 are arbitrary , @xmath297 and @xmath298 for @xmath299 and @xmath300 for @xmath301 . the eigenvalues of these hermitian matrices determine the constituent locations , all lined - up along @xmath286 . it should be noted that there is no reason to expect that all the @xmath302 can be diagonalized simultaneously . we will come back to this in the following section . returning to the simplest case of parallel gauge orientations , i.e. putting @xmath303 , we may take the limit @xmath304 ( related to vanishing time separations ) to find y_m^ab=(_a+y_c_a^2)_ab+(y_m - y_c)_a_b y_c_m=1^n_my_m.[eq : ym ] the ( @xmath21 ) center of mass " coordinate @xmath305 can be freely chosen and the @xmath306 play the role of center of mass " of each constituent caloron . how exactly this is realized becomes clear when we diagonalize @xmath302 . let us first consider for @xmath16 and charge 2 , @xmath307 , with @xmath308 . without loss of generality we choose @xmath309 , such that the two eigenvalues of @xmath302 are given by @xmath310 , where @xmath311 and @xmath312 . for large and positive @xmath313 we find @xmath314 , representing two charge 1 calorons centered at @xmath313 and @xmath315 , with separations between their constituents monopoles given in terms of @xmath316 . we plot the constituent locations as a function of @xmath313 in fig . [ fig : ydiag ] for @xmath317 and @xmath318 . 7.4 cm @xmath266 1.3 cm 10.5 cm @xmath313 1.7 cm action density profiles are shown in fig . [ fig : wellsep ] ( left ) for @xmath319 and in fig . [ fig : overlap - xz ] for @xmath320 . from the dotted lines in fig . [ fig : ydiag ] one reads off the associated constituent locations . note that the magnetic moments of the two calorons are pointing in the same direction and that we can not freely interchange constituent monopole locations within our axially symmetric ansatz . however , when @xmath313 is small it is more natural to interpret the configuration as a narrow caloron ( i.e. instanton ) with inverted magnetic moment in the background of a large caloron . this is the proper setting to understand the non - trivial time dependence for @xmath321 illustrated in fig . [ fig : overlap - zt ] ( left ) . for @xmath322 a singular caloron arises due to the fusion of two constituents ( with opposite magnetic charge ) . this singularity is avoided when @xmath323 , which can be understood by observing that the eigenvalues of @xmath324 parametrize time - locations . if @xmath325 , with @xmath127 and @xmath313 made small , one will find two calorons ( and their constituents ) to be pushed far from each other . this can be understood as well , in terms of a short - to - long distance duality in the adhm data for an instanton pair with non - parallel group orientation @xcite , but can also be read off from the eigenvalues of @xmath302 defined in . as an example we take again @xmath16 and charge 2 , but now with @xmath326 and @xmath327 in general non - zero . for the case @xmath328 ( perpendicular relative color orientations ) , @xmath329 and @xmath330 ( equal mass constituents ) , the eigenvalues are @xmath331 , whereas for @xmath332 one finds @xmath333 . in the light of this it is interesting to observe , as shown in fig . [ fig : overlap - zt ] , that with @xmath332 and _ increasing _ @xmath127 the constituents are pushed out in the @xmath177 direction as well . when @xmath334 the constituents would otherwise come close together through the periodicity in the time direction . effectively these constituents thus have perpendicular color orientations ( due to our choice of holonomy with @xmath335 ) . the transition from constituents separating in the time direction for @xmath127 near 0 to constituents separating in the @xmath177 direction for @xmath127 near @xmath336 occurs for @xmath337 at approximately @xmath338 . with a little imagination one detects the ring - shaped structure also observed @xcite in the case of instantons at zero temperature , see fig . [ fig : overlap - zt ] ( middle ) . a more direct analogy of course occurs when two calorons ( with @xmath12 small , i.e. instantons with unresolved constituent monopoles ) approach each other . we checked that for @xmath339 and @xmath340 a singular caloron forms due to the overlap of two calorons with parallel gauge orientations , whereas for @xmath341 the two calorons are pushed away ( to infinity ) in the @xmath177 direction as is appropriate for the non - parallel group orientation due to the non - trivial holonomy . at an intermediate value ( @xmath342 for @xmath343 ) one observes a small ring in the @xmath344-@xmath177 plane . choosing @xmath313 large one may check that @xmath345 indeed gives the time location for each caloron . when , however , @xmath127 approaches @xmath336 they can no longer keep parallel gauge orientations due to the non - trivial holonomy . as noted before , this may be described by a solution with @xmath346 and @xmath340 . computing the eigenvalues of @xmath302 therefore allows one to easily predict the behavior of the exact solution . for charge 1 it had been shown @xcite that as soon as one of the constituents is far removed from the others the solution becomes static . for the dimensional reduction " to take place at higher charge this is no longer sufficient . we have seen ( generalization to @xmath1 is straightforward ) that any magnetically neutral cluster of constituents , when small with respect to @xmath13 , will behave like an instanton that is localized in time . for the special case with parallel group orientations , putting all @xmath347 in one would even be left with @xmath348 singular instantons on top of one regular caloron , whose scale parameter is set by @xmath349 , which can be understood from the fact that the matrix @xmath350 has rank 1 . it does , however , give us one opportunity to go beyond the axial symmetry considered so far . when @xmath351 we could solve the nahm equation for parallel gauge orientations by @xmath352 _ without _ insisting all the @xmath353 line - up . this still describes @xmath348 singular instantons on top of one regular caloron , except that now the singular instantons can have _ arbitrary _ locations . even though our ansatz to obtain exact solutions has been restrictive ( as is clear from the axial symmetry ) , we stress that the solutions for the nahm equation we found provide genuine multi - caloron solutions , which reveal @xmath17 isolated lumps for each of its constituent monopoles ( with suitably chosen @xmath306 so the constituents do not overlap ) . this is not only illustrated in fig . [ fig : wellsep ] , but can also be understood analytically for any charge @xmath0 and @xmath1 as follows . we diagonalize each @xmath302 with ( in general _ different _ ) similarity transformations @xmath354 , or @xmath355 . these bring @xmath356 to the diagonal form @xmath357 , with @xmath358 the constituent radii , such that @xmath359 ( @xmath354 acting componentwise ) simplifies to ( cf . ) the action density can now be explicitly expressed in terms of the constituent radii & -_n f_^2(x)=-_^2_^2 ( x),=(ie^-ix_0(_2k-)/ ) , + & ( 2i(x_0_k-_0))u_nh_nt_nh_n-1 t_n-1h_1 u_1^t_1 . ( [ eq : ths],[eq : osborn],[eq : defpsi],[eq : conhm ] ) , where @xmath360 ( @xmath354 again acting componentwise ) . the size of the constituent monopoles is read off to be @xmath361 ( or @xmath362 when @xmath363 ) , and one concludes that the action density will contain @xmath17 lumps for sufficiently well separated constituents . the figures were produced by computing the action density using precisely this method . the non - trivial value of the polyakov loop at spatial infinity ( holonomy ) leads to a spontaneous breaking of the gauge symmetry , but without the need of introducing a higgs field . one may view @xmath3 as the higgs field in the adjoint representation . this is one way to understand why constituent monopoles emerge . the best way to describe the caloron solutions , in case of well separated constituents , is by analyzing the field outside the cores of these constituents , where only the abelian field survives . since in our case the asymptotic polyakov loop value defines a global direction in color space the abelian generator in terms of which we can describe the so - called far - field configuration is fixed , giving rise to a global embedding in the full gauge group . extrapolating the abelian fields back to inside the core of the constituents leads to dirac monopoles . such an extrapolation is well defined in terms of the _ high temperature limit _ , which makes the core of the constituents shrink to zero size and the field to become a smooth abelian gauge field everywhere except for the singularities of the dirac monopoles . thus we anticipate that in this limit the self - dual abelian field is still described by point like constituents , despite the fact that @xmath364 is no longer piecewise constant . any fuzziness " of the constituent location that may result from this , would be confined to the non - abelian core , and not visible from afar . despite the somewhat formal expression for the green s function @xmath185 , one can extract information about the long - range fields from it . in the following we will show how to neglect the exponentially decaying fields in the cores of the monopole constituents , being left with the abelian components of fields which decay algebraically . we only need to consider the bulk " contributions @xmath182 , , which contain all the dependence on @xmath252 . our starting point is restricted to the @xmath86th interval , @xmath365 \{-+4 ^ 2r^2(z;x)}f_m(z)=0 . [ eq : dfm ] to distinguish between exponentially growing and decreasing contributions for this homogeneous equation we take as a basis for @xmath366 functions @xmath367 with the following asymptotic behavior @xmath368 . this prompts us to introduce on each interval the matrix valued functions @xmath369 ( we suppress the dependence on @xmath252 ) such that f_m^(z)=p . from which it follows that @xmath369 is a solution of the riccati equation r_m^(z)^2r_m^(z)=r^2(z;x).[eq : riccati ] we note that for @xmath370 , @xmath371 and that for piecewise constant @xmath162 both @xmath372 and @xmath373 are constant and equal to @xmath356 , introduced in . we can write for @xmath374 the propagator " @xmath375 defined in in terms of @xmath367 as @xmath376 with w_m(z ) . using that @xmath377 and @xmath378 , we find by neglecting the exponentially decreasing factors @xmath379 the required limiting behavior for @xmath182 . paying special attention to the ordering of the @xmath28 matrices @xmath380 , observing that w_m^-1(_m)=(4r_m(_m))^-1,r_m(r_m^+(_m)+r_m^-(_m ) ) , we find well outside the cores of the constituents h_mf_m^+(_m+1 ) ( 4r_m)^-1 . the sparse nature of the matrices involved will be of considerable help to simplify the limiting behavior of the green s function . a crucial ingredient is the combination & & = + & & _ mr_m-1^+(_m)+r_m^-(_m)+s_m,[eq : defsigma ] for clarity summarizing the various ingredients in the following picture + ( 10,3.6)(-0.5,-1.2 ) ( 1.2,0)(1,0)0.2 ( 1.6,0)(1,0)0.2 ( 0.95,-0.5)@xmath108 ( 2,0)(1,0)11 ( 2,-0.25)(0,1)0.5 ( 2,0 ) ( 1.7,-0.5)@xmath381 ( 1.7,1.3)@xmath382 ( 1.7,0.7)@xmath383 ( 3.2,-0.6)@xmath384 ( 7.75,-0.25)(0,1)0.5 ( 7.75,0 ) ( 7.45,-0.5)@xmath211 ( 7.45,1.7)@xmath385 ( 5.7,1.3)@xmath386 ( 7.45,0.8)@xmath387 ( 7.75,0.5)@xmath388 ( 9.1,-0.6)@xmath389 ( 13,-0.25)(0,1)0.5 ( 13,0 ) ( 12.7,-0.5)@xmath209 ( 12.7,1.3)@xmath390 ( 12.7,0.7)@xmath391 ( 13.2,0)(1,0)0.2 ( 13.6,0)(1,0)0.2 ( 14.4,0 . ) . + this leads to the following far - field approximation for @xmath392 ( cf . ) , _ _ mg^(1)_m [ eq : approxf ] where @xmath393 and _ m,m&&f_m-1^+(_m)(2r_m-1)^-1_m-1 f_m-2^+(_m-1)(2r_m-2)^-1_m-2 + & & f_m+1^+(_m+2)(2r_m+1)^-1_m+1 f_m^+(_m+1)(4r_m)^-1.[eq : defgm ] one might have expected a factor @xmath394 on the right , but this is contained in the remaining terms of . for example @xmath395 . as we have seen in eqs . ( [ eq : adef ] ) and ( [ eq : phidef ] ) , the gauge field only requires us to know the green s function at the impurities . without loss of generality we may assume @xmath396 and take @xmath199 , such that ( see eqs . ( [ eq : ftofh],[eq : vecf],[eq : finf ] ) ) f_x(_m,_m)=-4 ^ 2(_k,0)w(_m,_m+0 ) ( _ 2k-__m)^-1 . the matrix @xmath397 has a @xmath96 block structure , and one can verify that in general ^-1=.[eq : blockinv ] identifying the blocks , in the high temperature limit we find a(_2k-__m)_11&&_k-2g^(1)_m r_m^-(_m ) , + b(_2k-__m)_12&&-g^(1)_m , + c(_2k-__m)_21&&-4 ^ 2(r_m-1^+(_m)+ s_m)g^(1)_m r_m^-(_m),[eq : hight ] + d(_2k-__m)_22&&_k-2(r_m-1^+(_m)+s_m ) g^(1)_m , where we used that @xmath398 , cf . ( [ eq : frel],[eq : ths ] ) . evaluating the green s function at the _ same _ impurities , @xmath399 , is simplified by the fact that @xmath400 . this gives the following remarkably simple result in the far - field limit , f_x(_m,_m)=-4 ^ 2(_2k-__m)_12 ^ -14 ^ 2 ( 2_m-_m^-1g(1))^-1 2(_m)^-1.[eq : fdiag ] for the green s function evaluated at _ different _ impurities , @xmath401 , we need to determine @xmath402 , for which we can follow the same method as for @xmath392 w(_m-0,_m+0)=_m,m , with @xmath403 as defined in . this leads to f_x(_m,_m)&&-4 ^ 2_m,m((_2k-__m ) ^-1_22 + 2r_m^-(_m)(_2k-__m)^-1_12 ) + & & _ m,m_m^-1g(1)f_x(_m,_m ) , which is exponentially suppressed since @xmath403 grows as @xmath404 . this can not compensate for the decay of @xmath405 , provided all @xmath211 are unequal , i.e. all constituents have a non - zero mass . massless constituents have a so - called non - abelian cloud @xcite , which has no abelian far - field limit . to determine @xmath250 in the expression for the action density , eqs . ( [ eq : osborn],[eq : defpsi ] ) , we need to compute @xmath406 . using we find & ( _ 2k - f__n)===(b)(ab^-1d - c ) + & ( g^(1)_m)(_m^-1g(1)-2_m ) ( -2g^(1)_m_m ) , such that ( _ m_m)=2 ^ -k_m=1^n\{(f^+_m ( _ m+1))(_m)/(2r_m)}. for @xmath407 , @xmath408 ( see ) and @xmath409 , which implies that @xmath410 = 2^{-k}\exp(2\pi k|\vec x|)$ ] ( recall that @xmath411 ) . therefore , the action is given by @xmath412 , as should be the case for a self - dual charge @xmath0 solution . without the off - diagonal components of the green s function contributing to the far - field region , the functions @xmath413 and @xmath414 in eqs ( [ eq : adef],[eq : phidef ] ) can be further simplified to ( x)^-11-_mf^ab(_m,_m)p_m_a^_b p_m , _ j_mf^ab(_m,_m)p_m_a _ j ^_bp_m , and only the abelian components of the gauge field survive . particularly the case of @xmath45 discussed in sect . [ sec : gaugefield ] is easy to deal with . using eqs . ( [ eq : ftofh],[eq : asp1],[eq : phis],[eq : vphis],[eq : fdiag ] ) we find ^-1(x)&&_0(1-_k[2_2 ^ -1s_2])_0 _ ff^-1(x),(x ) _ k[2_2 ^ -1g(_2)_2g^(_2 ) ] ) , + a_(x)&&(1-_k[2 _ 2 ^ -1s_2])^-1|_^j__k[2_2 ^ -1 g(_2)_2^jg^(_2)]),[eq : approxphi ] where we recall ( see ) that @xmath415 . this is in perfect agreement with the earlier @xmath21 results @xcite . note that the gauge rotation which relates @xmath416 to @xmath160 also relates @xmath100 to @xmath163 ( see eqs . ( [ eq : ftofh],[eq : rdef ] ) ) and therefore does not appear in the final expression for @xmath42 . it is interesting to note that the dipole moment of the abelian gauge field is particularly simple and does not require us to solve for @xmath417 , since @xmath418 such that _ |x|a_(x)= |_^j_.[eq : dip ] hence the dipole moment @xmath419 only involves @xmath277 , and we do except it allows for configurations with a vanishing dipole moment . for higher multipole moments , through @xmath417 , we need to deal with the full quadratic adhm constraint , or equivalently with the riccati and nahm equations . nevertheless , it is remarkable that in the high temperature limit the @xmath252 dependence is restricted to @xmath420 and @xmath421 _ only_. we would like to prove that each of its eigenvalues vanish at an isolated point , as one way to identify the @xmath422 constituent locations . we will defer the study of this interesting issue , and its generalization to @xmath1 , to a future publication . the far - field approximation further simplifies when considering the axially symmetric solutions discussed in sect . [ sec : axial ] . we restrict ourselves here to @xmath45 . since @xmath162 is piecewise constant , the riccati equation is trivial to solve , r^_m(z)=r_m=. the square root involves a positive @xmath28 matrix , and is well - defined . due to the fact that @xmath423 ( see eqs . ( [ eq : rdef],[eq : srho ] ) ) , the abelian component of the gauge field is of the simple form ( @xmath235 , see sect . [ sec : gaugefield ] ) a^abel_(x)=-e_j|^j _ _ ( x ) . in the far field limit @xmath424 ( see ) . since @xmath425 has rank 1 , the matrix @xmath426 has only one non - vanishing column with respect to a suitably chosen basis , which implies that @xmath427 . this allows us to write in the far - field region _ ff(x)= , from which we immediately read off the result @xcite for @xmath21 , in which case it is easy to show that @xmath428 , revealing @xmath51 to be a linear superposition of two oppositely charged self - dual dirac monopoles . we would like @xmath429 to similarly factorize for @xmath22 in @xmath422 dirac monopoles , but since @xmath430 , @xmath431 and @xmath425 in general do not commute , some care is required in demonstrating the factorization . we will rely on the fact that @xmath432 and hence independent of @xmath86 . since by definition @xmath295 ( see sect . [ sec : axial ] ) , one finds that _ ff(x)=. this can now be reorganized according to _ ff(x)&= & + & = & , after which we can separately diagonalize @xmath433 and @xmath434 to find _ ff(x)=_i=_i , where @xmath435 give the locations of the constituent monopoles , with @xmath436 the eigenvalues of @xmath302 and @xmath437 ( the index @xmath86 distinguishing their charge ) . the second expression for the factorized version of @xmath438 uses the fact that the constituent locations can be ordered according to y_1^(1)<y_2^(1)<y_1^(2)<y_2^(2)< <y_1^(k)<y_2^(k ) . it should be noted that this prevents the constituents to pass each other while varying the parameters for the axially symmetric configuration , see also fig . [ fig : ydiag ] and the discussion in sect . [ sec : axial ] . we need to go beyond this simple axially symmetric configuration to allow for the constituents to rearrange themselves more freely . we have presented the general formalism for finding exact instanton solutions at finite temperature ( calorons ) with any non - trivial holonomy and topological charge . in an infinite volume holonomy and charge are fixed . the solution is described by @xmath65 parameters of which @xmath439 give the spatial locations of the @xmath17 constituent monopoles . the remaining parameters are given by @xmath0 time locations and @xmath440 phases associated to gauge rotations in the subgroup that leaves the holonomy unchanged . of these , @xmath76 can be considered as global gauge rotations . the dimension of the moduli space , i.e. the number of gauge invariant parameters , is therefore equal to @xmath441 . our subset of axially symmetric solutions has @xmath442 paramters of which there are @xmath76 global gauge rotations , or @xmath443 gauge invariant paramters . explicit solutions were found for the case of axial symmetry , an important limitation being the difficulty of solving the nahm equation , or equivalently the quadratic adhm constraint . we certainly expect more progress can be made on this in the near future . nevertheless , we already found a rich structure that bodes well for being able to consider the constituent monopoles as independent objects . this is surprisingly subtle , as we have illustrated by the fact that an approximate superposition of charge 1 calorons tends to give rise to a _ visible _ dirac string . this is also related to the difficulty of finding approximate multi - monopole solutions , which can be obtained from the caloron solutions by sending a subset of the constituent monopoles to infinity , as has been well studied in the charge 1 case @xcite . an important tool has been our study of the far - field limit , describing the abelian gauge field far removed from any of the constituent monopoles . this allows for a description of the long distance properties in terms of ( self - dual ) dirac monopoles . much could be extracted concerning its properties without the need to explicitly solve the nahm equation . we conjecture in general to be able to identify the dirac monopole location , but some work remains to be done here . 4.6 cm it may seem that all these results are somewhat academic since until recently none of these constituent monopoles were found in dynamical lattice configurations . first of all one would be tempted to search for them at high temperature , but it should be noted that above the deconfining phase transition the average polyakov loop takes on trivial values , associated to the center of the gauge group , which is not the environment in which a caloron will reveal its constituents . this would give the well - known harrington - shepard solution constructed long ago @xcite . with our present understanding this solution can be seen as having @xmath76 massless constituents which can not be localized . only when sending all these to infinity one is left with a monopole @xcite . furthermore , at high temperature classical configurations will be heavily suppressed due to their boltzmann weight . rather , the hope is that the constituent monopoles play an important role _ below _ the deconfining temperature , where the average of the polyakov loop is non - trivial , and tends to favor _ equal _ mass constituents . this is why in this paper our examples were for that case , see figs . [ fig : wellsep],[fig : overlap - xz],[fig : overlap - zt ] and the discussion in sect . [ sec : axial ] . nevertheless , the formalism developed here gives results for any choice of the holonomy , and a sample of unequal mass constituents is given in fig . [ fig : uneq ] . in general the constituent monopoles can be characterized by their magnetic ( = electric ) charge . for @xmath1 there are @xmath84 different types of abelian charges involved @xcite , and all @xmath0 constituents of a given type have the same mass . we will discuss briefly the lattice evidence for the presence of constituent monopoles that has accumulated the last few years . a first numerical study using cooling was performed with twisted boundary conditions , which implies non - trivial holonomy @xcite . good agreement was found with the infinite volume charge 1 analytic results , in particular so for the fermion zero - modes @xcite which are more localized than the action density . a charge 2 solution was also found , shown in fig . 8 of ref . fermion zero - modes played an intricate role in an extensive numerical study of nahm dualities on the torus @xcite . as suggested in ref . @xcite one may also enforce non - trivial holonomy on the lattice by putting at the spatial boundary of the box all links in the time direction to the same constant value @xmath444 , such that @xmath445 ( @xmath446 the number of lattice sites in the time direction ) . this has been implemented in @xmath16 lattice monte carlo studies as well , where @xmath79 was set to the average value of the polyakov loop , appropriate for the temperature at which the simulations were performed @xcite . cooling was applied to find calorons , including those at higher charge . apart from the configurations that in the continuum would be exactly self - dual , the lattice allows one to also consider configurations in which both self - dual and anti - selfdual lumps appear . this revealed constituent monopoles that seem not directly associated to calorons , called @xmath447 ( as opposed to @xmath448 ) . both objects in such a @xmath447 configuration have fractional topological charged , but opposite in sign . perhaps these arise when two near constituent monopoles , one belonging to a caloron , the other to an anti - caloron , annihilate " . our analytic methods can not directly address this situation due to the lack of self - duality . the same holds for configurations that seem to only carry magnetic fields , which were already seen long ago @xcite . one point of criticism that applies to both methods is that the choice of boundary conditions may force the dissociation " of instantons into constituent monopoles , particularly since volumes can not yet be chosen so large that many instantons are contained within a given configuration . a recent study @xcite has done away with the fixed boundary conditions that enforce the non - trivial holonomy . nevertheless , still one finds in many cases that calorons dissociate " into constituent monopoles below the deconfinement transition temperature . a particularly useful tool has turned out to be the fermionic ( near ) zero - modes to detect the monopole constituents when they are too close together to reveal themselves from the action density @xcite . this relies on the observation that the zero - mode is localized on _ only _ one of the constituent monopoles , determined by the boundary conditions imposed on the fermions in the time direction @xcite . for @xmath16 this is particularly simple , with periodic and anti - periodic boundary conditions of the fermions making the zero - mode switch from one constituent to the other , as is illustrated for a close pair of constituents in fig . [ fig : zmflip ] . in addition one may use the polyakov loop for diagnostic purposes @xcite , for @xmath16 taking the values @xmath449 and @xmath450 near the respective constituent locations ( at these points the gauge symmetry is restored , providing an alternative definition for the center of a constituent monopole ) . 3.7 cm fermion eigenfunctions with eigenvalues near zero have also been used as an alternative to cooling , to filter out the high frequency modes and identify topological lumps . for a recent lattice study , including some discussion of calorons with non - trivial holonomy , see ref . @xcite and references therein . using the near zero - modes as a filter , constituent monopoles have even been identified recently for @xmath451 well below the deconfining temperature @xcite , resembling fig . [ fig : zmflip ] ( see also fig . 14 of ref . @xcite . for the exact axially symmetric multi - caloron solutions constructed in this paper the @xmath0 associated fermion zero - modes ( for charge @xmath0 ) will be derived in the near future . we anticipate that one can choose a basis where each is localized on one of the constituent monopoles , the type of which is determined by the choice of fermionic boundary conditions in the time direction . analyzing these zero - modes is particularly interesting in the light of some puzzles that were presented in a recent study @xcite of the normalizable fermion zero - modes in the background of a collection of so - called bipoles , i.e. pairs of oppositely charged ( but self - dual ) dirac monopoles , which are of interest in a wider context as well . this will be one of the many topics we have access to with our analytic tools . but ultimately our main aim is to develop a reliable method to describe the long distance features of non - abelian gauge theories in terms of monopole constituents to understand both confinement and chiral symmetry breaking . the results of this paper , and in particular the recent lattice results , provide some encouragement in this direction . we thank conor houghton for initial collaboration on the monopole aspects of this work and him as well as chris ford for extensive discussions . pvb also thanks michael mller - preussker and christof gattringer for discussions concerning calorons with non - trivial holonomy on the lattice . furthermore he is grateful to leo stodolsky and valya zakharov for hospitality at the mpi in munich and to poul damgaard , urs heller and jac verbaarschot for inviting him to the ect * workshop `` non - perturbative aspects of qcd '' in trento . he thanks both institutions for their support , while some of the work presented in this paper was performed . fb likes to thank the organizers of the channel meeting on theoretical particle physics " for a well organized and stimulating meeting as well as dimitri diakonov , gerald dunne , alexander gorsky and peter orland for discussions . the research of fb is supported by fom . kraan and p. van baal , phys . lett . b428 ( 1998 ) 268 [ hep - th/9802049 ] ; nucl . phys . b533 ( 1998 ) 627 [ hep - th/9805168 ] . kraan and p. van baal , phys . lett . b435 ( 1998 ) 389 [ hep - th/9806034 ] . atiyah , n.j . hitchin , v.g . drinfeld , yu . i. manin , phys . 65 a ( 1978 ) 185 ; m.f . atiyah , _ geometry of yang - mills fields _ , fermi lectures , ( scuola normale superiore , pisa , 1979 ) . w. nahm , _ self - dual monopoles and calorons _ , in : lect . notes in physics . 201 , eds . g. denardo , e.a . ( 1984 ) p. 189 . k. lee and p. yi , phys . d56 ( 1997 ) 3711 ( hep - th/9702107 ) ; k. lee , phys . lett . b426 ( 1998 ) 323 [ hep - th/9802012 ] ; k. lee and c. lu , phys . d58 ( 1998 ) 025011 ( hep - th/9802108 ) . w. nahm , phys . lett . 90b ( 1980 ) 413 . g. t hooft , nucl . phys . b190 [ fs3 ] ( 1981 ) 455 ; physica scripta 25 ( 1982 ) 133 . corrigan , d.b . fairlie , s. templeton and p. goddard , nucl . b140 ( 1978 ) 31 . h. osborn , nucl . b159 ( 1979 ) 497 . m. garca prez , t.g . kovcs and p. van baal , phys . b472 ( 2000 ) 295 [ hep - ph/9911485 ] . p. van baal , in : lattice fermions and structure of the vacuum , eds . v. mitrjushkin and g. schierholz ( kluwer , dordrecht , 2000 ) , p. 269 [ hep - th/9912035 ] . k. lee , e.j . weinberg and p. yi , phys . b376 ( 1996 ) 97 ( hep - th/9601097 ) ; phys . d54 ( 1996 ) 6351 ( hep - th/9605229 ) ; e.j . weinberg , _ massive and massless monopoles and duality _ , hep - th/9908095 . kraan , comm . 212 ( 2000 ) 503 [ hep - th/9811179 ] . harrington and h.k . shepard , phys . d17 ( 1978 ) 2122 ; ibid . d18 ( 1978 ) 2990 . p. rossi , nucl . b149 ( 1979 ) 170 . m. garca prez , a. gonzlez - arroyo , a. montero and p. van baal , jour . of high energy phys 06 ( 1999 ) 001 [ hep - lat/9903022 ) ] . m. garca prez , a. gonzlez - arroyo , c. pena and p. van baal , phys . rev . d60 ( 1999 ) 031901 [ hep - th/9905016 ] . chernodub , t.c . kraan and p. van baal , nucl . phys . b(proc.suppl . ) 83 - 84 ( 2000 ) 556 [ hep - lat/9907001 ] . m. garca prez , a. gonzlez - arroyo , c. pena and p. van baal , nucl . b564 ( 1999 ) 159 [ hep - th/9905138 ] . ilgenfritz , m. mller - preussker , and a.i . veselov , in : lattice fermions and structure of the vacuum , eds . v. mitrjushkin and g. schierholz ( kluwer , dordrecht , 2000 ) , 345 [ hep - lat/0003025 ] . ilgenfritz , b.v . martemyanov , m. mller - preussker and a.i . veselov , nucl . b(proc.suppl.)94 ( 2001 ) 407 [ hep - lat/0011051 ] ; nucl . suppl.)106 ( 2002 ) 589 [ hep - lat/0110212 ] . laursen and g. schierholz , z. phys . c38 ( 1988 ) 501 . ilgenfritz , b.v . martemyanov , m. mller - preussker , s. shcherendin and a.i . veselov , hep - lat/0206004 . p. van baal , nucl . b(proc.suppl.)106 ( 2002 ) 586 [ hep - lat/0108027 ] . c. gattringer , m. gckeler , p.e.l . rakow , s. schaefer , a. schfer , nucl . b618 ( 2001 ) 205 [ hep - lat/0105023 ] . christof gattringer , private communications . the zero - mode densities for the case of fig . [ fig : zmflip ] were produced for the purpose of illustrating the behavior observed by c. gattringer and co - workers in @xmath451 lattice gauge theory . kraan and p. van baal , nucl . suppl.)73 ( 1999 ) 554 [ hep - lat/9808015 ] . p. van baal , chiral zero - modes for abelian bps dipoles , to appear in confinement , topology , and other non - perturbative aspects of qcd " , eds . j. greensite and s. olejnik ( kluwer ) , hep - th/0202182 .

long before the development of general relativity , john michell wondered about the gravitational influence of objects on the light they emit , and how one might go about finding objects so dense that light could not escape their surfaces . `` [ i]f any other luminous bodies should happen to revolve about them we might still perhaps from the motions of these revolving bodies infer the existence of the central ones with some degree of probability , as this might afford a clue to some of the apparent irregularities of the revolving bodies , which would not be easily explicable on any other hypothesis '' @xcite . more than 200 years later , precisely this method has been employed to determine the mass of the black hole ( bh ) at the center of our milky way . sgr a@xmath10 is one of the most tightly constrained bhs in the universe , with a total mass uncertainty ( statistical and systematic ) of less than 10% ( see * ? ? ? * and references therein ) . using stellar motions to measure the bh masses in more distant galaxies is complicated by our current inability to spatially resolve the orbits of individual stars . instead , one must rely on the luminosity - weighted line - of - sight velocity distribution ( losvd ) within different spatial resolution elements , and model the most likely contribution to the gravitational potential from the unseen bh ( beyond that provided by the stars themselves and any dark matter ) . the most common numerical approach to constraining the bh mass ( @xmath0 ) and mass - to - light ratio of the stars ( @xmath11 ) is the orbit superposition method of @xcite , which simultaneously optimizes the fit to the surface brightness distribution and the line - of - sight kinematics of stars within the nuclear region . application of that technique has yielded more than 50 bh mass estimates ( e.g. , * ? ? ? * ; * ? ? ? a complimentary mass scale has been developed for the bhs in active galactic nuclei ( agns ) , based on the technique of reverberation mapping ( rm ; * ? ? ? * ; * ? ? ? measuring the size - scale and velocity of gas near the bh has provided mass estimates for @xmath1250 bhs @xcite , but systematic uncertainties remain due to the unknown geometry and dynamics of the gas . this uncertainty is traditionally encapsulated in a scaling ( or projection ) factor , @xmath13 an empirical calibration for the rm mass scale @xmath14 has been established by assuming that those agns also lie on the tight correlation between bh mass and bulge stellar velocity dispersion ( the @xmath15 relation ) found in quiescent galaxies . this was first done by @xcite , who obtained an average rm mass scale of @xmath16 . subsequent determinations have ranged from @xmath17 @xcite to @xmath18 @xcite , with the most recent value being @xmath19 @xcite . for ngc 4151 , this gives a current rm - derived bh mass of @xmath20 ( based on the revised estimate of * ? ? ? it is important to note that the error generally quoted on a rm - derived bh mass in an individual galaxy is the formal uncertainty in the virial product and does not include the uncertainty in the nominal mean scale factor @xmath21 . what has been missing is an independent bh mass measurement for a sample of galaxies with rm estimates . early results have suggested that the empirically calibrated rm masses are consistent with the values produced by other techniques @xcite , but the few agns near enough to be probed by stellar dynamics have not yielded particularly tight constraints ( in part , owing to the complicating factor of the bright agn overwhelming the stellar absorption features at the smallest galactic radii ) . while only a few galaxies lend themselves to multiple methods for measuring the masses of their central bhs , such independent bh mass determinations are crucial to isolating the systematic errors inherent in each method . it is only via independent measurements that it is possible to derive robust bh masses . ngc 4151 is a particularly interesting galaxy since it is one of the rare objects for which there already exist two independent dynamical measurements for the mass of the bh . @xcite use optical spectroscopy of stellar absorption lines and orbit superposition modeling to derive an upper limit for the bh mass of @xmath22 when the galactic bulge is assumed to be edge - on . when the bulge is assumed to have the same inclination as the large - scale disk , they derive a bh mass of 4 - 5@xmath23 , which they caution is likely to be a biased solution since it is associated with a very noisy @xmath1 surface that also gives a poor fit to the data ( in terms of absolute @xmath1 value ) . @xcite use adaptive optics ( ao ) to obtain high spatial resolution , near - infrared spectroscopy of the molecular , ionized , and highly ionized gas in ngc 4151 . from the kinematics of gas in the vicinity of the bh , they derive a dynamical mass of @xmath24 for the bh . being somewhat smaller than the value obtained by @xcite , this suggests a need to revisit the stellar dynamical bh mass measurement . two serious concerns with the result of @xcite motivate the need for the high spatial resolution observations that are presented in this paper . first , based on the rm mass and the bh mass derived by @xcite , the sphere of influence of the bh is estimated to be @xmath25 , significantly smaller than the spatial resolution of the spectra used by @xcite . second , low spatial resolution integral field spectroscopy of a larger field of view @xcite shows that the true kinematic major axis of the bulge ( derived from the 2d velocity field ) is oriented @xmath26 from the axis assumed by @xcite . the observations and modeling presented in this paper are designed to overcome both of these drawbacks and to provide a more robust comparison with the gas dynamical measurement of @xcite . to enhance the comparison between the rm and stellar dynamical mass scales for bhs , we have obtained near - ir integral field spectroscopy with ao in order to study the stellar dynamics in the inner regions of the reverberation - mapped agn , ngc 4151 . in conjunction with our previous optical spectroscopy and multi - wavelength imaging data , we are able to place the tightest dynamical constraints to date on the bh mass in ngc 4151 . in section 2 , we describe the observational data . in section 3 , we discuss our stellar dynamical modeling . section 4 describes our results . section 5 contains further discussion and our conclusions . ngc 4151 is a nearby ( @xmath27 ) galaxy with a prominent bulge , faint spiral arms and a large scale bar ( see sections 3.4 and 3.5 for more details on the morphology and distance of the galaxy ) . the galaxy s nuclear activity has been studied extensively since the original paper by seyfert ( 1943 ) ; in particular , the analysis of uv and optical emission lines have been used for determining the bh mass via rm ( e.g. , * ? ? ? the analysis presented here uses new near - ir spectroscopy performed with ao , as well as previously published optical long - slit spectroscopy , and space- and ground - based imaging ( primarily from * ? ? ? * hereafter paper i ) to provide a mass estimate independent of the rm value . observations of ngc 4151 and three giant stars ( used as velocity templates ) were obtained with the near - infrared integral field spectrograph ( nifs ; * ? ? ? * ) on gemini north between 16 feb 2008 and 24 feb 2008 in queue mode ( program i d gn-2008a - q-41 ) . the nifs image - slicing integral field unit ( ifu ) provides @xmath28 near - ir spectra across a @xmath29 field , with a spatial sampling of @xmath30 , or @xmath31 `` spaxels '' ( spatial pixels ) , where the smaller pixel scale is along each of the 29 image slices . because previous observations of the k - band stellar absorption lines in ngc 4151 show the features to be weak in the central regions @xcite , we focus on the co ( 6 - 3 ) bandhead at a rest - frame wavelength of 1.62 @xmath32 m , which has been found to be quite prominent in previous nifs spectra @xcite . the @xmath2-band spectra cover a wavelength range of @xmath33m@xmath34 m ( centered at 1.6490 @xmath32 m ) , with a dispersion of 1.60 pixel@xmath35 across the 2040 spectral pixels , and a velocity resolution of 56.8 . to take advantage of the spatial sampling afforded by nifs , the observations were taken with the altair ao system @xcite , using the agn in ngc 4151 as the ao guide star . the observations employed the altair field lens @xcite to provide a more uniform point spread function ( psf ) across the nifs field . the level of ao correction depends upon the prevailing seeing conditions at the time . the median uncorrected seeing was @xmath36 , and more than 95% of the data were obtained in seeing better than @xmath37 , which provided an ao - corrected nuclear fwhm between @xmath38 and @xmath39 . to estimate the psf , we model the contribution from the agn to the spectrum in each spaxel across the field of view , taking advantage of the agn s strong [ ] @xmath40 m emission line , at an observed wavelength of @xmath41 m . although the [ ] emission is spatially extended in ngc 4151 , the off - nuclear emission line is narrower than the nuclear line profile , and tight constraints can be placed on the amount of agn spectrum that ends up in each spaxel ( see 4.2 of * ? ? ? the analysis was restricted to a wavelength region around the [ ] line ( @xmath42m@xmath43 m ) , and to control the effects of noise , the spectra were box - car smoothed over 5 pixels . a slice was taken through the center of the resulting 2d map , cutting along the smaller pixel scale ( @xmath44 ) . the psf profile shows the standard core+halo structure delivered by ao systems , which we model as a 2-part gaussian , where the core has @xmath45 , the halo has @xmath46 , and each component has equal amplitude . the data for ngc 4151 were acquired at a position angle of @xmath47 , which aligns the nifs slitlets perpendicular to the radio jet ( matching the orientation used by * ? ? ? * ; * ? ? ? * ; * ? ? ? the typical observing sequence included a 9-point dither pattern with separations of 0206 , and offsets to a sky field ( 210 away ) approximately every third frame . individual exposures were 120 s long , and the observations were split into sequences of 80 minutes in duration ( in some instances , shortened by changing weather conditions ) . arc lamps were taken in the middle of each observing sequence , and telluric standards were obtained before and after each sequence . the telluric stars were chosen to be 2mass sources to simultaneously provide photometric calibration : hd 98152 was observed before ngc 4151 , and hd 116405 afterwards . darks , flats , and ronchi mask calibrations were obtained during the daytime after each night s science observations . in total , 253 on - source frames were obtained , for a total exposure time of 30360 s. a total of 112 sky frames were also acquired . the giant stars we observed were included so as to provide a measure of the spectral broadening of the instrument , allowing the intrinsically narrow - featured stars to serve as templates that could be broadened to match the actual stellar velocity distribution of ngc 4151 . these velocity template stars were observed without ao , and were selected to span a range of spectral type : hd 35833 ( g0 ) , hd 40280 ( ) , hip 60145 ( m0 ) . the observations of each star consist of four on - source and two sky exposures , each with an integration time of 5.3 s. the nifs data were reduced with a combination of version 1.9.1 of the gemini iraf package and our own iraf scripts . calibrations were processed independently for each night . the observations of ngc 4151 were grouped by proximity to a telluric standard , typically dividing the observing sequence in half . the data were flat - fielded , dark - subtracted , wavelength - calibrated , and spatially rectified , then combined , corrected for telluric features , and flux - calibrated . the flux calibration provided by the two telluric / flux standards differs systematically by @xmath48% , implying a 0.3 mag discrepancy in the 2mass photometry for the two stars . because the absolute flux levels in the nifs data play no role in our subsequent analysis , we do not attempt to correct for this offset . since the image quality of some frames is significantly worse than others , we select a subsample of individual frames with fwhm values below a certain upper limit . with a threshold of 016 in the fwhm ( measured at a uniform wavelength of 1.6598 @xmath32 m to avoid emission features in the spectra ) , 236 frames are included in the final datacube . we spatially shift the data to account for the dither pattern and then median - combine the images , yielding a datacube containing 29@xmath4969 spatial pixels and 2040 spectral pixels . the near - ir spectra at each spatial position in the final nifs datacube are combined onto a bin of size @xmath1202-square ( 2@xmath495 raw spaxels ) , resulting in a cube of 15@xmath4915 spaxels . this choice of binning the spaxels arose from two main considerations . first , prior to carrying out the axisymmetric dynamical modeling , the kinematic data need to be symmetrized about the kinematic major axis and reflected about the rotation axis , to avoid modeling inconsistencies . this process is best accomplished with square bins , especially when the symmetry axes of the galaxy are not aligned with the axes of the spectrograph ( as is generally the case ) . second , since the halo of the psf has @xmath46 and this component comprises 50% of the overall psf amplitude , the spatial resolution of the data is not significantly altered by using larger bins . we reduce the observations of the telluric / flux standard stars and the velocity template stars with procedures similar to those we apply to the ngc 4151 data . we median - combine the four frames for each star , extract spectra with a fixed aperture of 1@xmath505 , and then we normalize each extracted spectrum for the velocity template stars by a low - order fit to the continuum . in addition to the new data from nifs , we utilize the results from the optical spectra of ngc 4151 that were analyzed in paper i. the data consist of long - slit spectra of the calcium triplet region ( @xmath51 ) taken with the mayall 4 m telescope at kitt peak national observatory in 2001 , and with the 6.5 m mmt observatory at mt . hopkins in 2004 . for the data from kitt peak , the spectroscopic slit was 2wide and was placed at a position angle of 135@xmath52 . to enhance the signal - to - noise , spectra were extracted with an aperture that increased in size away from the galaxy center , extending to radii of @xmath5314 . the spectral resolution was @xmath54 . for the observations with the mmt , the slit was 1 wide , at a position angle of 69@xmath52 . spectra were extracted in @xmath55 bins to a distance of 12on either side of the agn . the spectral resolution was @xmath56 . the seeing conditions for the two datasets were @xmath121.8 and @xmath123 , respectively . we refer the reader to paper i for full details . to model the surface brightness profile of ngc 4151 in the near - ir , we make use of the @xmath2-band imaging from the early data release of the ohio state university bright spiral galaxy survey ( osubsgs ; * ? ? ? the data were obtained with the 1.8 m perkins telescope at lowell observatory . the image has a plate scale of 15 pixel@xmath35 , and we estimate the seeing to be 2 . we photometrically calibrate the osubsgs imaging using 2mass point - sources in the field , deriving a photometric zero - point of @xmath57 mag . to bridge the gap in the surface brightness information between the small field - of - view of the ao - assisted nifs data and the large - scale low resolution imaging data of the osubsgs , we use the _ hubble space telescope _ images from paper i. these data were taken with the f550 m filter on the high resolution channel of the advanced camera for surveys ( acs ) , and cover a field - of - view of @xmath1225@xmath4929 . we refer the reader to paper i for additional details . we also make use of @xmath58- and @xmath59-band optical imaging from the sloan digital sky survey ( sdss ) to assist in constraining the mass - to - light ratio ( @xmath11 ) . the images have an exposure time of 53.9 s in each band , with 0396 pixels and @xmath1210-seeing in both filters . we use the photometrically calibrated and sky - subtracted images that sdss makes available . as in paper i , our modeling relies on simultaneously fitting both the luminosity distribution of the bulge and the observed integrated line - of - sight velocity distributions ( losvd ) of stars obtained with various spectrographs over a range of spatial scales . we detail below the methods of deriving those quantities from the data described above . as discussed in [ sec : nifs_details ] , the nifs instrument produces a grid of spatial pixels ( `` spaxels '' ) of size @xmath30 which were combined onto a bin of size 02-square , resulting in a cube of 15@xmath4915 spaxels . the spectrum from each binned spaxel is then analyzed using the `` penalized pixel fitting '' ( ppxf ) package for idl by @xcite . the ppxf routine computes the velocity shift , the velocity dispersion , and the higher - order moments of the stellar losvds ( parameterized as gauss - hermite ( gh ) polynomials ) that , when convolved with the velocity template star spectrum , provide the best match to the observed data . the software allows emission features to be masked in the fit , and also provides for multiplicative and additive polynomials to be included ( which we use to account for the contribution of the agn continuum ) . our ppxf analysis fits the losvd up to 4th - order gh terms , including 2nd - order multiplicative and additive polynomials , and fits the spectra over the wavelength range @xmath60m@xmath61 m , with four small wavelength gaps around agn emission lines and residual sky features . figure [ fig.spec ] shows four example spectra and ppxf fits , from different positions within the datacube . we obtain the smallest uncertainties on the kinematic parameters when we fit the agn spectra with the m0 velocity template , and allowing simultaneous use of all three templates does not measurably improve the results . we also compared the velocity measurements to those obtained when using nifs velocity templates ( spectral types , , , and a ) acquired by watson et al . ( 2008 ; gemini program gn-2006b - sv-110 ) . the m - type stars provided better matches to the relative strengths of the absorption lines in each ngc 4151 spectrum , and the kinematic uncertainties were smaller than for the k - type stars . compared to the kinematics derived from our m0 velocity template , the differences were typically less than 1@xmath4 , suggesting our fits are robust against the effects of template - mismatch . our results are consistent with those of @xcite , who found better matches to @xmath2-band galaxy spectra with m - type stars than with k - type stars . figure [ fig : vel_fields_orig ] ( top row ) shows maps of the line - of - sight velocity ( @xmath62 ) , velocity dispersion ( @xmath63 ) , and the 3rd and 4th gh moments ( @xmath64 ) in 02 spaxels . the @xmath65 coordinates are aligned with the major and minor kinematic axes , respectively . the color scale is in km / s for @xmath62 and @xmath63 , and in dimensionless units for @xmath66 and @xmath67 . we then bi - symmetrize the 2d losvd map ( fig . [ fig : vel_fields_orig ] bottom row ) around the kinematic minor axis at a position angle of @xmath68 ( @xmath69 relative to the nifs y - axis ) , using an idl routine developed by @xcite . this procedure enhances the s / n by combining the measurements on either side of the kinematic minor axis in a symmetric ( @xmath63 and @xmath67 ) , or anti - symmetric ( @xmath62 and @xmath66 ) way . data point is multiplied by -1 and averaged with the @xmath66 value at a position mirrored on the opposite side of the bh . ] such a procedure is necessary in order to obtain sensible results from an axisymmetric dynamical modeling code , which assumes such an underlying symmetry . the kinematic data were bi - symmetrized in a weighted mean , using the individual uncertainties reported by ppxf to derive the weights . the final kinematic uncertainties are computed as the errors in the weighted mean values . figure [ fig : nifs_sau ] shows the line - of - sight stellar velocity fields from nifs ( left ) and sauron ( right ) . the latter data are from @xcite and show a similar large scale rotational velocity to that seen in the inner 3@xmath493 region observed with nifs . although the large scale line - of - sight velocity fields in both panels show the same global rotational velocity , the central region of the sauron velocity field within 5@xmath49 5 ( right ) is dominated by agn continuum . therefore , the innermost parts allow no velocity measurement , and appear as a black oval region in the image . thus , the sauron data do not provide constraints on the bh mass , although they could provide constraints on the mass - to - light ratio ( @xmath11 ) . however , we do not include the sauron data in our modeling , relying instead on the long - slit kinematics obtained from mmt and kpno to provide the constraints on @xmath11 at large radii . this is partly a pragmatic choice since the use of ifu data on both small and large scales would significantly increase the total number of kinematic constraints that would need to be fitted and , therefore , would require a significantly larger orbit library than we use here , making the problem significantly more computationally expensive than we are able to handle . we calculate the @xmath2-band surface brightness profile using our nested series of images to probe different spatial scales . for large radii ( @xmath70 ) , we use the photometrically calibrated osubsgs images . at intermediate radii ( @xmath71 ) , we use the acs optical image , allowing a significant region of overlap with the osubsgs data . the acs flux level is then rescaled to provide the best match to the @xmath2-band data in the overlap region . our use of the optical data could potentially bias the surface brightness profile we derive if there are significant variations in extinction or stellar population at radii of @xmath72 . however , there are no independent indications of stellar population changes in that part of ngc 4151 , and the radial profile of the acs image is relatively smooth at those distances . those radii are also well isolated from the psf of the agn in the acs image . in the central regions , we derive a stellar flux profile directly from the nifs datacube . one of the outputs of the ppxf modeling is the derived amplitude of the stellar template that best matches the galaxy data . as our template spectra are continuum - normalized , these amplitudes constitute relative measurements of the stellar flux in each spaxel , calculated in a way that naturally omits the agn contribution to the light profile . the unbinned data are used to probe radii of 0@xmath73 ( excluding the agn - dominated central spaxel ) . the overall nifs stellar profile is then scaled in amplitude to match the overlapping acs data from larger radii . we then fit the full @xmath2-band surface brightness profile using the multi - gaussian expansion ( mge ) package @xcite . the mge routine fits the smallest necessary number of nested 2d gaussians with a common center and radially varies their amplitudes and axis ratios to obtain the best possible smooth fit to the observed 2d surface brightness distribution , taking account of the psf . the deprojection of a 2d surface brightness distribution to derive the 3d luminosity distribution suffers from the well known degeneracy arising from `` konus densities '' that are invisible in projection @xcite . a number of different methods have been used to derive the 3d luminosity density , none of which guarantee a unique deprojection . the mge method has the advantage that , once an angle of inclination is assumed , each 2d gaussian has a unique 3d luminosity density whose potential is easily computed and which is positive everywhere @xcite . the code outputs the amplitude ( total counts ) , width in pixels ( @xmath74 ) , and observed axis ratio ( @xmath75 ) for each gaussian . we convert those values to a central surface brightness , @xmath76 , for each gaussian as : @xmath77 where @xmath78 is the @xmath2-band surface brightness of the sun at a distance where 1 pc@xmath79 subtends 1 arcsec@xmath79 . for an absolute @xmath2-band magnitude of 3.32@xmath530.03 mag @xcite , this gives @xmath80 mag arcsec@xmath81 . the surface brightness is computed from the mge fits as : @xmath82 where @xmath83 is the osubsgs @xmath2-band image zero - point of 22.2 mag , @xmath84 is the total counts , @xmath85 is the plate scale of @xmath86 pixel@xmath35 , and the @xmath2-band extinction is @xmath87=0.016 mag ( * ? ? ? * retrieved from the nasa / ipac extragalactic database ) . for the mge modelling , a 1-component psf was used , having an effective width of @xmath88 . the output of our mge fit to the surface brightness distribution is given in table [ tab : mge_sb ] . for each gaussian in the mge fit , the first column gives an identifying number , the second column gives the @xmath2-band luminosity density , the third column gives the standard deviation @xmath4 ( in arcsec ) , and the fourth column gives the flattening @xmath75 ( ellipticity ) . it is clear from the values of @xmath75 in the third column that the bulge surface brightness distribution is close to circular since most of the gaussians have @xmath89 ( except for the innermost and outermost gaussians ) suggesting that the bulge is probably close to spherical . in figure [ fig : sb_profile ] , we show the composite radial surface brightness profile obtained by patching together the three imaging datasets . lcccgaussian & surface density , @xmath90 & gaussian @xmath4 & axis ratio @xmath75 + number & ( @xmath91/pc@xmath92 ) & ( ) & + ( 1 ) & ( 2 ) & ( 3 ) & ( 4 ) + + 1 & 9.3963e+04 & 0.126 & 0.7906 + 2 & 1.3200e+05 & 0.232 & 1.0000 + 3 & 7.8100e+04 & 0.549 & 1.0000 + 4 & 2.0951e+04 & 1.421 & 1.0000 + 5 & 6.7191e+03 & 2.799 & 1.0000 + 6 & 3.7009e+03 & 6.703 & 1.0000 + 7 & 6.8485e+02 & 14.513 & 0.9550 + 8 & 3.1906e+02 & 64.987 & 0.7052 + [ tab : mge_sb ] to provide a rough constraint on the relationship between the luminosity distribution and the stellar mass distribution , we combine optical imaging from sdss with the osubsgs @xmath2-band image . we construct ( @xmath93 ) and ( @xmath94 ) color maps of ngc 4151 , finding the bulge colors of ( @xmath93)=1.1@xmath530.1 mag , and ( @xmath94)=2.4@xmath530.2 mag to be quite uniform across the bulge ( aside from the central few arcseconds , where the agn dominates the emission ) . @xcite calculated @xmath2-band mass - to - light ratios ( @xmath95 ) in ( @xmath93)-(@xmath94 ) color - space , based on a monte carlo analysis of @xcite stellar population synthesis models . examining our measured colors in the @xcite color - space indicates that @xmath8 . our initial exploration with dynamical models spanning the entire range @xmath96 showed that the best fit values were always in the range @xmath97 , hence this range of values is used in the models below . ngc 4151 was classified by @xcite as ( r)sab(rs)ab : a disk galaxy with a pseudo - outer ring ( r ) , a `` mixed '' bar ( ab ) ( i.e. , possibly barred but is not strongly so ) , with mixed spiral / inner ring classification ( rs ) , and an intermediate to large bulge ( ab ) . figure [ fig : slit_orient ] ( left ) shows the palomar observatory sky survey ( poss ) image of ngc 4151 in the red optical filter . in part because of the very circular bulge , the classification of this galaxy as barred has been questioned by authors as far back as @xcite , who speculated that the elongation was not a bar , but was in fact made up of material that was ejected from the nucleus in a former explosive event . however , we believe that ngc 4151 may be an example of a galaxy containing a `` barlens '' , an oval - shaped structure that is likely to be the vertically thick aspect of a bar seen from a particular viewing angle @xcite . in paper i , we assumed that the kinematic major axis is aligned with the bar , since spectral information was only available along two long - slits . the kpno slit ( represented by the red line in fig . [ fig : slit_orient ] left ) was oriented along the bar at a position angle of 135@xmath52 @xcite . the mmt slit ( green line in fig . [ fig : slit_orient ] left ) was placed at a position angle of 69@xmath52 . the middle panel of figure [ fig : slit_orient ] shows the region within the white box from the left panel . the grey box in the middle panel represents the orientation and shape of the nifs field of view . it is quite clear from a comparison of the nifs velocity field and the sauron velocity field in fig . [ fig : nifs_sau ] , that the rotation axis of the disk and bulge are not perpendicular to the bar as was assumed in paper i. this is an important difference between the models presented here and the models presented in paper i. to first order , the bh mass derived by the stellar dynamical modeling depends linearly on the assumed distance to the galaxy where r is the linear radius within which the velocity ( or velocity dispersion ) is measured , and the conversion between angular and linear radii is inversely proportional to distance in the nearby universe . ] . for consistency with the analysis in paper i , we adopt a distance for ngc 4151 of 13.9 mpc ( based on the recession velocity of 998 measured by * ? ? ? however , as noted in paper i , the actual distance to the galaxy is rather uncertain , with published estimates spanning a range of 10 mpc to 30 mpc . recently , @xcite compiled a group - averaged tully - fisher distance of 11.2@xmath531.1 mpc in the extragalactic distance database . however , that group distance is based on just four galaxies , which have a standard deviation in their distance moduli of more than four times the individual distance modulus errors . furthermore , it is unclear whether the contribution to the group average distance from ngc 4151 ( of @xmath124 mpc ) accounts for the agn contribution to the galaxy brightness . thus , we can not regard the @xcite value as being any more reliable than other distances , and until a more accurate distance can be obtained , we continue to use the value of 13.9 mpc . we apply what is now the standard stellar dynamical modeling approach of orbit superposition ( e.g * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) to fit the observed losvds and luminosity distribution of ngc 4151 . the orbit superposition ( or `` schwarzschild '' ) method of dynamical modeling uses line - of - sight kinematical and imaging data to constrain the mass distributions of external galaxies . the implementation of the code used here differs from the method detailed in ( * ? ? ? * hereafter vme04 ) in two small aspects . it is ( a ) designed to fit generalized axisymmetric models that are obtained from direct deprojection of the surface brightness profiles of the individual galaxies using a mge code , following the method outlined in @xcite ; ( b ) modified to fit integral field kinematics in addition to long - slit kinematics . we restrict our models to the surface brightness distribution within 50 i.e. , the bulge region using the mge decomposition from table [ tab : mge_sb ] . for each dynamical model , we assume a constant @xmath2-band mass - to - light ratio ( @xmath95 ) for the stars value for the stars and the best - fit model values of @xmath95 described below are completely consistent with the expectations derived from the stellar photometry in sec . [ sec : mtol ] . even if we were to include such a halo , its contribution to the inner 15 region , where the nifs kinematics exist , would likely be small . as we will show , the nifs data entirely determine our final bh mass , justifying our decision to not include a dark matter halo . ] , with values varying from @xmath98 . the resulting density profile is then converted to a gravitational potential using an axisymmetric multipole expansion scheme @xcite . the gravitational potential of a point mass ( @xmath0 ) , representing the central bh , is then added . models are constructed for every pair of parameters ( @xmath0 , @xmath95 ) , building an orbit library for each combination . the large - scale disk of ngc 4151 is inclined to the line of sight at @xmath99 @xcite . however , since the bulge appears very close to circular in projection ( see fig . [ fig : slit_orient ] and tab . [ tab : mge_sb ] ) , it is probably nearly spherical or being viewed close to face on . thus , we run models on a 2d grid of ( @xmath100 ) values for three different inclination angles @xmath101 ( edge - on ) , @xmath102 and @xmath103 ( close to face - on ) . we also run a limited set of models for which the value of @xmath0 is held fixed while the inclination angle of the model is varied with values @xmath104 , and @xmath105 , in an effort to set constraints on the inclination of the bulge to the line - of - sight . for inclination angles of @xmath101 and @xmath106 , we run orbit libraries for 18 bh masses between @xmath107 , using a baseline @xmath108 . these models are then scaled to create libraries for 18 different @xmath95 values , producing a total of 324 models for each inclination angle . we also run orbit libraries for @xmath102 , using 10 bh masses across the same mass range and 18 @xmath95 values for each @xmath0 ( 180 models ) . the range of bh mass values was set by initial experimentation with a coarse grid on a wide range of values . the lowest two values of @xmath0 are set to 10@xmath109 and 10@xmath110 , both of which consistently give poor fits to the data . the upper limit of @xmath0 was raised until it was clear that models once again gave very poor fits to the data . the combined gravitational potential of the bulge and the bh is used to integrate a large library of @xmath111 stellar orbits selected on a grid with @xmath112 energy values , @xmath113 angular momentum values ( @xmath114 ) at each energy , and @xmath115 pseudo third integral values ( @xmath116 ) at each angular momentum value ( see vme04 for details ) . for instance , models with @xmath117 used @xmath118 , @xmath119 , @xmath120 . each orbit is integrated for 100 orbital periods and its average `` observed '' contribution to each of the `` apertures '' in which kinematical data are available is stored . the orbital kinematics are subjected to the same type of instrumental effects as the real data ( i.e. psf convolution , pixel binning ) . self - consistent orbit - superposition models are constructed by linearly co - adding orbits in each library to simultaneously fit several elements within the observed apertures ( imaging and spectroscopic ) : the 3d mass distribution ( for each value of @xmath95 ) , the projected surface density distribution , and the losvds . the fits use a non - negative least squares optimization algorithm ( nnls ; * ? ? ? * ) , which gives the weighted superposition of the orbits that best reproduce both the self - consistency constraints ( i.e. , the 3d mass distribution and surface brightness distribution ) and the observed kinematical constraints ( @xmath121 ) . vme04 showed that as the ratio of orbits to constraints decreases , the error bars decrease artificially . in the majority of the models presented here we have 1240 constraints that are fitted by a library of 9936 orbits . this size of orbit library gives an orbits - to - constraints ratio of 8 which is only slightly above the ratio ( 5 ) at which vme04 find that solutions begin to be biased . in section [ sec : robust_size ] we carry out a limited exploration with a 50% larger orbit library and find that our results are unchanged . smoothing ( or `` regularization '' ) of the orbital solutions is typically performed in this type of analysis , and a variety of methods have been employed to that end . regularization mainly helps to reduce the sensitivity of the solutions to noise in the data by requiring a smoother sampling of the orbit libraries used in the solution . following @xcite , we run a trial series of models with a regularization scheme that requires local smoothing in phase space . we find that for small smoothing parameters ( @xmath122 ) the resultant best - fit @xmath0 values do not differ from the models without regularization . however , the @xmath1 value for the best - fit model is larger and the error bars on @xmath0 are smaller . vme04 showed that it is impossible ( without computationally expensive approaches such as generalized cross - validation ) to determine the ideal regularization parameter . not regularizing the solutions can yield large error bars , but choosing too large a value of the smoothing parameter can introduce a bias in the best - fit value of @xmath0 . in this paper , we present models without regularization , since imposing regularization constraints significantly increases the total number of constraints and therefore decreases even further the orbits - to - constraints ratio . for the libraries used here , models without regularization provide the most conservative error bars on the estimated value of @xmath0 . since the mass and surface mass density constraints are _ self - consistency constraints _ rather than observational constraints , they do not have observational errors associated with them . there are two ways in which the absence of observational errors are handled in the literature . @xcite and vme04 include self - consistency constraints in the nnls minimization problem with errors corresponding to predefined level of relative accuracy in the fit ( e.g. 5% -10% ) , while others @xcite solve the orbit superposition problem by requiring the self - consistency constraints ( mass and surface brightness ) to be fitted as separate as linear ( in)equality constraints with predetermined absolute accuracy . @xcite state that they require the self - consistency constraints to be fitted to an accuracy of 0.02 . we also attempted to fit self - consistency constraints as linear equalities but find that that since the values of the mass constraints vary by 5 orders of magnitude ( from @xmath123 to 10 ) , using a fixed numerical accuracy fails to fit the mass constraints at small radii ( where the logarithmic radial bins have the smallest mass values per bin and where the fit to mass constraints most strongly affects the estimated @xmath0 ) . for the models presented here we include the self - consistency constraints in the optimization problem and require the optimization problem to fit each of the 472 self - consistency constraints ( 280 mass constraints and 192 aperture surface brightness constraints ) with an error corresponding to a relative accuracy - per - constraint of 5% . for our preferred model with @xmath101 , the fractional accuracy per self - consistency constraint varies from 0.27% ( for the best fit model ) to 2.2% for the worst fit model , i.e. , always better than our required accuracy . figure [ fig : mass_err ] shows contours of constant fractional error in the fit to the self - consistency constraints ( mass and surface brightness in apertures ) over the 2d parameter space ( @xmath0 , @xmath95 ) for models with @xmath101 . the star is the location of the model with 0.27% accuracy per self - consistency constraint , and the contours are spaced at intervals of 0.1% relative accuracy per constraint . in previous applications of our schwarzschild modeling code ( vme04 , * ? ? ? * paper i ) , the error in the fit to the self - consistency constraints was essentially random over the entire 2d parameter space ( @xmath124 ) . although the self - consistency constraints were always included in the @xmath1 , they contributed so little to the total @xmath1 that the best fit solution was always driven by the fit to the kinematical constraints . we will show that this is not true for our current ngc 4151 dataset . despite the fact that the fits to the 3d mass distribution and surface brightness distribution are very good , because the fit vary systematically rather than randomly across the parameter space , the best fit solution is altered by the decision of whether or not to include the fit to the self - consistency constraints in the @xmath1 . based on the arguments above , we compute the @xmath1 in two different ways : ( 1 ) the total @xmath125 includes all the constraints in the nnls optimization problem : 280 mass constraints , 192 surface brightness constraints , and 768 kinematical constraints ( giving a total of @xmath126 constraints ) ; ( 2 ) @xmath127 which only considers the quality of the fit to @xmath128 kinematical constraints ( @xmath129 and @xmath67 in 192 apertures : 136 nifs spaxels , 15 kpno apertures , and 41 mmt apertures ) . for each of the three assumed angles of inclination for the bulge , we construct models by fitting these constraints using an orbit library consisting of @xmath117 orbits . for one inclination angle ( @xmath101 ) we do a limited exploration with a library consisting of @xmath130 orbits . after generating orbit libraries for a grid of parameters ( @xmath131 ) the best fit parameters are determined by the minimum in the 2d @xmath1-contour plot . in this section , we present the results of our modeling . in section [ sec : define_chi2 ] we describe results for inclination angles @xmath101 , @xmath102 and @xmath103 , showing 2d contour plots of @xmath125 and @xmath127 using a full 2d grid of models for each of these inclination angles . in section [ sec : inclination ] we examine the dependence on inclination angle for 4 additional models using a single @xmath0 . in section [ sec : robust ] we explore the robustness of our solutions by examining their sensitivity to the size of the orbit library and to the effect of restricting the set of kinematic constraints to those from the nuclear region only . we also present some of the best - fit solutions and show how well different pairs of @xmath100 fit both the 1d and 2d kinematical data . in section 5.4 , we discuss our results in the context of the relation . figure [ fig:2dchi2 ] shows 2d contour plots of @xmath125 in the plane of model parameters @xmath0 and @xmath95 for @xmath101 ( top ) , @xmath102 ( middle ) , and @xmath103 ( bottom ) . the grid of points shows the model parameters ( @xmath100 ) for which the orbit libraries is constructed and the optimization problem is solved . the first 6 contour levels in all the @xmath1 contour plots that follow correspond to @xmath132 , 68.3% confidence ) , @xmath133 , 90% confidence ) , @xmath134 , 95.4% confidence ) , @xmath135 , 99% confidence ) , @xmath136 , 99.73% ) , and @xmath137 , 99.99% ) , with subsequent contours equally spaced in @xmath138 . for @xmath101 , the minimum value of @xmath125 is obtained for @xmath139 and @xmath140 , with @xmath141 . for inclination angle of @xmath102 , the best fit model has @xmath142 and @xmath143 , with @xmath144 . for inclination angle of @xmath103 , the best fit model has @xmath145 and @xmath140 , with @xmath146 . table [ tab : bestfit ] summarizes the results and includes minimum @xmath1 values and the 1@xmath4 and 3@xmath4 error bars . the errors on a given parameter are obtained by marginalizing the 2d @xmath147 values over the other parameter . figure [ fig:1dchi2 ] ( top ) shows the marginalized 1d @xmath138 versus @xmath0 for the three different inclination angles , as indicated by the legends . the horizontal dotted lines are at @xmath148 ( 1@xmath4 , 68.5% confidence interval for 1 degree - of - freedom ) , @xmath149 ( 2@xmath4 , 90% confidence interval ) , and @xmath150 ( 3@xmath4 , 99% confidence interval ) . the 1@xmath4 and 3@xmath4 errors on @xmath0 and @xmath95 are given in table [ tab : bestfit ] . likewise , figure [ fig:1dchi2 ] ( bottom ) shows the marginalized 1d @xmath138 versus @xmath95 for each of the three inclinations of models to the line - of - sight . it is important to note that with the standard methods used to derive the velocity profiles ( vps ) from the spectra , the errors in the gauss - hermite ( gh ) coefficients are not independent . the errors associated with the even moments ( @xmath4 and @xmath67 ) are correlated , and errors associated with the odd moments ( @xmath151 and @xmath66 ) are also correlated ( see , e.g. , * ? ? ? * ; * ? ? ? @xcite point out that an important consequence of correlated errors in the gh coefficients is that the errors obtained by comparing data to dynamical models using the @xmath1 estimator are significantly underestimated . in this paper we quote both 1-@xmath4 and 3-@xmath4 errors , and often consider the 3-@xmath4 errors as better representations of the true modeling errors . lcccccccc & & & & + & & & & & & & & + ( 1 ) & ( 2 ) & ( 3 ) & ( 4 ) & ( 5 ) & ( 6 ) & ( 7 ) & ( 8) & ( 9 ) + + + 90@xmath52 & 9936 & 709.1 & 4.68 & @xmath152 & @xmath153 & 0.304 & @xmath154 & @xmath155 + & & & & & & & + 60@xmath52 & 9936 & 767.5 & 5.42 & @xmath156 & @xmath157 & 0.305&@xmath158 & @xmath159 + & & & & & & & + 23@xmath52 & 9936 & 1050.8 & 3.69 & @xmath160 & @xmath161 & 0.304 & @xmath158 & @xmath155 + & & & & & & & + + + 90@xmath52 & 9936 & 186.7 & 7.76 & @xmath162 & @xmath163 & 0.332 & @xmath164 & @xmath165 + & & & & & & & + 60@xmath52 & 9936 & 193.1 & 8.62 & @xmath166 & @xmath167 & 0.326&@xmath159 & @xmath168 + & & & & & & & + 23@xmath52 & 9936 & 348.2 & 8.50 & @xmath169 & @xmath170 & 0.329 & @xmath171 & @xmath172 + & & & & & & & + [ tab : bestfit ] the best fit solutions for both @xmath0 and @xmath95 for the three different inclination angles are consistent with each other within 3@xmath4 . however @xmath173 is the smallest for the edge - on model ( @xmath101 ) , increases for @xmath102 and is significantly larger for @xmath103 implying that it may be possible to constrain the inclination of the bulge ( see section [ sec : inclination ] ) . marginalizing @xmath125 over the three values of inclination gives a best fit @xmath174 and @xmath175 . figure [ fig:2dchi2_kin ] shows 2d contour plots of @xmath176 for the same sets of models as in figure [ fig:2dchi2 ] , with inclination angles @xmath177 ( from top to bottom ) . we emphasize that the model solutions that are used to generate figure [ fig:2dchi2_kin ] are identical to those used to generate figure [ fig:2dchi2 ] only the quantities used in computing the @xmath1 of the fit to the data differ . figure [ fig:1dchi2_kin ] shows the same data marginalized over @xmath95 ( top ) and marginalized over @xmath0 ( bottom ) . it is clear from these two figures that the best - fit solutions obtained using @xmath178 give both a larger @xmath179 and a larger @xmath180 . we will discuss the possible causes for this in section [ sec : robust ] and section [ sec : discuss ] . in contrast to the results of paper i , where @xmath101 models gave an upper limit of @xmath181 and @xmath103 models gave a best fit @xmath0 between @xmath182 , figures [ fig:1dchi2 ] and [ fig:1dchi2_kin ] show that best - fit @xmath0 is relatively insensitive to inclination ( although the actual best - fit value depends on how @xmath1 was computed ) . table [ tab : bestfit ] shows that the minimum values of @xmath183 and @xmath184 for @xmath101 are lower than the corresponding @xmath1 values for the other two inclination angles . we now try to constrain the inclination angle of the bulge of ngc 4151 by assuming a constant bh mass of @xmath185 . with this value of @xmath0 we compute orbit libraries for four additional inclination angles : @xmath186 , @xmath187 , @xmath188 , and @xmath105 . figure [ fig : chi_inc ] shows the values of @xmath183 for models with this value of @xmath0 and @xmath189 as a function of the inclination of the bulge to the line - of - sight . it is clear that @xmath183 decreases steadily with increasing inclination angle with a minimum at @xmath101 . ( a similar dependence on inclination angle is obtained for @xmath190 , although all the models give much worse fits as assessed by the @xmath125 values . ) the inclination angle of @xmath103 that is inferred from the velocity field is strongly disfavored . since the disk of this galaxy is clearly seen to be close to face - on , the edge - on orientation of the bulge preferred by our models is rather puzzling : the surface brightness isophotes of the bulge are nearly circular , hence the best - fit edge - on orientation implies that the bulge is nearly spherical in shape ; but in this case the rotation axis of the bulge must be misaligned with the rotation axis of the disk by @xmath191 . previous determinations of the orientations of the various components of ngc 4151 have shown a remarkable diversity . while the circular isophotes of the large - scale disk suggest the nearly face - on orientation with @xmath103 @xcite which motivated our testing of that value of inclination the components associated with the agn have been inferred to lie further from our line of sight . models of the bi - conical narrow - line region ( nlr ) in the uv and optical @xcite , and in the near - ir @xcite have typically found inclinations of @xmath192 . closer to the agn , @xcite fit the absorption using a pc - scale nuclear disk inclined at @xmath193 . on even smaller scales , an inclination of @xmath194 for the agn accretion disk was obtained from modeling the x - ray spectrum @xcite , while @xcite suggest that the radio jet in ngc 4151 lies very close to the plane of the sky and is interacting directly with circumnuclear gas in the plane of the galaxy ( i.e. , @xmath195 ) . this lack of alignment between the different spatial scales could be the consequence of a merger in the galaxy s recent past . it has been claimed that the periodic variability in the nuclear activity of ngc 4151 is indicative of a bh binary , with a separation on the order of @xmath196 au @xcite . while distinguishing between a single bh and a close binary is far beyond the capability of our dataset , the remnants of a merger could also contribute to the complicated dynamics we observe on the size scales of the bulge . in section [ sec : robust ] , we will show that the kinematics in the inner 5 show evidence for a bar in this region , implying that the internal dynamics of the bulge are likely not that of a spherical or axisymmetric system . a full understanding of why the kinematics of our model prefers an edge - on configuration to the more inclined configuration requires the system to be modeled by a bar dynamical modeling code , which does not exist at the present time . since the result of this section show that models with edge - on inclination ( @xmath101 ) are greatly preferred over more face - on models , in the rest of this paper we confine our analysis to models with @xmath101 . lllllcccccc & & & & & & + & & & & & & & & & & + ( 1 ) & ( 2 ) & ( 3 ) & ( 4 ) & ( 5 ) & ( 6 ) & ( 7 ) & ( 8) & ( 9 ) & ( 10 ) & ( 11 ) + + @xmath183 & 15092 & 1240 & 641.9 & 0.52 & 4.68 & @xmath197 & @xmath198 & 0.31 & @xmath158 & @xmath199 + & & & & & & & & & & + @xmath178&15092 & 768 & 170.2 & 0.22 & 7.32 & @xmath200 & @xmath201 & 0.31 & @xmath199 & @xmath202 + & & & & & & & & & & + + + @xmath183 & 9936 & 960 & 583.1 & 0.61 & 3.82 & @xmath203 & @xmath204 & 0.32 & @xmath199 & @xmath205 + & & & & & & & & & & + @xmath178 & 9936 & 680 & 101.7 & 0.14 & 3.70 & @xmath206 & @xmath207 & 0.35 & @xmath199 & @xmath202 + [ tab : robustbestfit ] vme04 showed that the solution derived from the schwarzschild method was sensitive to two important factors : ( a ) the size of the orbit libraries used in obtaining the best - fit bh mass ; and ( b ) the spatial resolution of the kinematic data and whether or not they resolved the sphere - of - influence of the bh . they showed that biased solutions could result when the size of the orbit libraries was too small for the available data . they also showed that when the kinematic data did not resolve the sphere - of - influence of the bh , spurious results could arise . in the next two sections we perform two tests to assess the robustness of our solutions . in particular , our goal is to assess which of the two values of @xmath0 obtained previously ( 5@xmath208 or 8@xmath209 ) is more robust . in this section , we construct a set of 288 models with @xmath101 but with orbit libraries containing @xmath210 orbits ( as compared to @xmath211 used thus far ) to fit the same set of 1240 constraints that were fitted in section [ sec : define_chi2 ] . large orbit libraries are constructed for 16 bh masses between 10@xmath109 and 15@xmath212 and for 18 values of @xmath213 . figure [ fig:1dchi_i9015k ] shows the 1d @xmath183 ( solid curves ) and @xmath178 ( dashed curves ) obtained by marginalizing over the mass - to - light ratio @xmath95 ( top panel ) and obtained by marginalizing over @xmath0 ( bottom panel ) . the resulting best - fit values for @xmath0 and @xmath95 , and their 1@xmath4 and @xmath214 errors , are shown in the top two rows of table [ tab : robustbestfit ] . the reduced @xmath1 values are given in column five . it is clear from a comparison with results in table [ tab : bestfit ] that the best - fit @xmath0 values are relatively insensitive to an increase in the size of the orbit library by @xmath1250% . the best - fit value of @xmath95 obtained by using @xmath178 ( dashed curves in the bottom panel of fig . [ fig:1dchi_i9015k ] ) is somewhat lower than the value of @xmath215 obtained with the smaller library , but is still within 2@xmath4 of this value . the test with the larger orbit libraries confirms the result we obtained with @xmath117 orbits per library , implying that our best - fit parameters are not biased by having too small an orbit library . furthermore , the discrepancy between the best - fit values obtained using @xmath147 and @xmath178 persists even with larger orbit libraries . figure [ fig : kinenifs_90 ] shows examples of model fits to the 2 dimensional nifs velocity fields ( from left to right , , and the gh moments @xmath66 and @xmath67 ) for the edge - on ( @xmath101 ) case with @xmath210 . note that while the best - fit solutions listed in tables [ tab : bestfit ] & [ tab : robustbestfit ] are determined by marginalizing over one parameter in the 2d contour surface ( which itself is smoothed with a kernel ) , the 2d grid of models is discrete . the derived best - fit value ( shown as stars on the 2d @xmath1 maps of figs . [ fig:2dchi2 ] and [ fig:2dchi2_kin ] ) does not overlap with any model on the grid , but lies between 4 grid points . we therefore show velocity fields and 1d kinematics for the model whose parameters are closest to the best - fit solution . in figure [ fig : kinenifs_90 ] , the top row shows velocity fields for @xmath216 ( the model closest to the best - fit obtained with for @xmath210 ) and the middle row shows velocity fields for @xmath217 , @xmath218 ( the model closest to the best - fit obtained with for @xmath210 ) . ( the bottom row of this figure is for a model with @xmath219 , @xmath218 , obtained when only nifs kinematical data are fitted , and will be discussed in section [ sec : robust_nuclear ] . ) the fitted velocity in figure [ fig : kinenifs_90 ] should be compared with the bi - symmetrized velocity fields in the bottom row of figure [ fig : vel_fields_orig ] . the four white pixels in each panel of the figure were not fitted because it was determined that their extremely high velocity dispersion values ( @xmath220 ) were spurious ( see fig . [ fig : vel_fields_orig ] ) since their values were larger than the central velocity dispersion despite being far from the center . in figure [ fig : kinenifs_90 ] the odd - moments of the losvd ( , @xmath66 ) are fairly well fitted by both the low @xmath0 model ( top row ) and the high @xmath0 model ( middle row ) . however , the nifs velocity dispersion for the model with @xmath221 , @xmath108 ( top row ) is much lower than in the data ( bottom of fig . [ fig : vel_fields_orig ] ) . the model with the larger value of @xmath217 does a slightly better job of reproducing the higher values within @xmath222 , but fails to fit the at the edges of the nifs field , where model values are significantly lower ( i.e. bluer pixel colors ) than the observed values seen in figure [ fig : vel_fields_orig ] ( bottom ) . this is because from nifs in the inner 15 region is overall larger than that obtained with kpno ( @xmath223 ) and mmt ( @xmath224 ) . inconsistent velocity dispersion values in the _ same physical region _ drives the solution to fit the data with the smallest error bars ( i.e. , mmt data ) thereby under - estimating from nifs . also note that neither model is able to fit both the low central and high outer @xmath67 values seen in the bi - symmetrized nifs velocity fields ( fig . [ fig : vel_fields_orig ] bottom row , right - most panel ) . the difference between the model fits and the data is more clearly seen in figure [ fig : kinematics_90_1d ] which shows the one - dimensional line - of - sight velocity ( ) , velocity dispersion ( ) and the gh moments of the losvd ( @xmath66 and @xmath67 ) that we obtain for the edge - on ( @xmath101 ) model . the red and blue curves show the best fits for same models as in the top and middle panels of figure [ fig : kinenifs_90 ] with values and @xmath95 indicated by the line legends in the top left panel of the first row ( these two models used libraries with @xmath210 orbits ) . the green curves in the top four nifs panels are obtained when only the nifs data are fitted ( with @xmath211 ) , this model is discussed in the next section . the open circles in the panels show the three different kinematic datasets used in the modeling , along with their 1@xmath4 uncertainties . ( the top four panels show kinematics in the nifs spaxels lying closest to the major axis of the model ; the middle four panels show data from the kpno long - slit ; the bottom four panels show data from the mmt long - slit . ) the most striking feature of the fits is that the red curve for @xmath225 ( red curve ) the model closest to the best - fit solution derived from the minimum in @xmath183 ( from figure . [ fig:1dchi2 ] ) clearly _ underestimates _ the nifs velocity dispersion ( top right panel ) for the central - most points . the uniformly large central @xmath226 values in the inner @xmath227 region are much better fitted by larger values of @xmath228 and @xmath218 ( blue curve ) . both the models do an equally good job of fitting , , and @xmath66 from the mmt data , but overestimate from the kpno data . another important point to note is that the values of the gh parameters @xmath67 are _ negative _ in the innermost regions of each of the three datasets . neither model is able to fit the negative @xmath67 values measured with the mmt spectrograph nor the central dips to negative values in @xmath67 seen in the nifs data and kpno data . in spherical and axisymmetric galaxies , @xmath67 is negative when the velocity distribution is tangentially biased ; it is positive when orbits are predominantly radial ; and it is zero when the velocity distribution in isotropic @xcite . however , negative @xmath67 values have also been associated specifically with the presence of a bar . in recent @xmath229-body simulations of barred galaxies by @xcite , the losvds have negative @xmath67 values in the region where the bar dominates . the negative @xmath67 parameters arise due to the kinematic properties of the special orbits that constitute the bar , and due to the fact that from an external observer s point of view , the bar has a large tangential velocity component due to its pattern speed . even after a bar buckles and forms a boxy bulge , negative @xmath67 values can be seen in face - on systems @xcite . another signature of the kinematics associated with a nuclear bar is undulations in the rising part of the inner rotation curve @xcite , which we also see in the mmt data for ngc 4151 . finally , while @xmath66 is always anti - correlated with the line - of - sight velocity in axisymmetric systems , @xcite show that @xmath66 is _ correlated _ with along the major axis of the bar . for the data obtained with kpno ( where the slit lies along the major axis of the large scale bar ) @xmath66 is correlated with over the radial range @xmath230 ( figure [ fig : kinematics_90_1d ] ) , providing additional kinematical evidence for a bar . therefore it appears that despite the very circular photometric contours of the bulge in ngc 4151 , and the visual appearance of a rather weak bar at radii outside the bulge , there are several clear kinematical signatures of a small scale bar in the vicinity of the nuclear bh . this is consistent with the properties of `` barlens '' features that have recently been identified in real and simulated galaxies @xcite . as pointed out by @xcite , barlenses frequently masquerade as classical bulges because of their nearly circular central isophotes , especially when viewed nearly face - on , but have bar - like kinematics . the combination of bars and bhs can have significant implications for the central dynamics of galaxies . @xcite analyze a suite of @xmath229-body simulations of barred galaxies in which point masses representing bhs were grown adiabatically . they find that the growth of a bh of a given mass causes a 5 - 10% larger increase in @xmath63 in a barred galaxy than in an otherwise identical axisymmetric galaxy . @xcite find that if a disk galaxy with a bulge has a pre - existing black hole , the formation and evolution of a bar can result in a 15 - 40% increase in the central velocity dispersion . both studies find that the increase in the observed line - of - sight velocity dispersion in the barred systems is a consequence of three separate factors : ( a ) mass inflow due to angular momentum transport by the bar , ( b ) velocity anisotropy due to the presence of bar orbits , and ( c ) weak dependence on orientation of position angle of the bar . if ngc 4151 does have a bar and is thus non - axisymmetric , that fact also has implications for our modeling results . in stellar dynamical modeling of axisymmetric galaxies , fitting a negative @xmath67 requires a larger fraction of tangential orbits , which contribute little to the line - of - sight velocity dispersion . fitting a given velocity dispersion with such an orbit population will imply a larger enclosed mass than would be required if @xmath67 was zero or positive ( which is typically the case in axisymmetric systems with bhs ) . @xcite argue that if an axisymmetric stellar dynamical modeling code is used to derive the value of @xmath0 in a barred galaxy with a high central @xmath231 and a negative @xmath67 , the central bh mass will be systematically overestimated . if the large values of @xmath231 in the nuclear regions and the negative @xmath67 values in the mmt data are a consequence a nuclear bar , then the discrepancy between @xmath127 and @xmath125 could reflect this predicted bias in @xmath0 . in this section we examine the consequences of fitting only the nifs kinematical data ( ignoring the kinematical constraints at larger radii obtained from the lower spatial resolution long - slit spectra from kpno and mmt ) . there are three reasons why the nifs data alone may provide better constraints on the mass of the bh . first , the nifs data have higher spatial resolution ( 02 ) than the long slit data ( 1 - 2 ) . for a @xmath232 , the sphere - of - influence of the bh ( assuming @xmath233 ; see section [ sec : msigma ] below ) is @xmath1216 pc , which corresponds to @xmath12 024 for our assumed distance of 13.9 mpc for this galaxy . therefore the nifs data used in our modeling are barely able to resolve the sphere - of - influence of the bh , but are quite close to doing so . if @xmath234 , the sphere - of - influence of the bh would be 26 pc ( 038 ) and the bh sphere - of - influence would be resolved by the nifs data . the long slit data do not provide constraints on @xmath0 and are included mainly to provide constraints on the mass - to - light ratio of the stars , which is assumed to be independent of radius . second , as discussed previously , the nifs velocity dispersion within @xmath235 is significantly larger than the velocity dispersion on same physical scale obtained from the long - slit data . however , since the error bars on the long - slit mmt data are about half those from the nifs data , the optimization code tries to fit them better than the nifs data . in figure [ fig : kinenifs_90 ] ( top two panels ) , the need to fit the lower velocity dispersion values from mmt manifests as low velocity dispersion values ( blue ) at the edges of the nifs field . these low dispersion values outside the sphere - of - influence of the bh also force the model to adopt a lower mass - to - light ratio , which is then compensated for in the inner region by requiring a much larger to fit the nifs velocity dispersion . neglecting the long - slit data would allow the optimization code to raise the mass - to - light ratio in the inner @xmath532 region ( where the velocity dispersion is quite flat ) . finally , in figure [ fig : kinematics_90_1d ] , we see kinematic evidence for the bar in mmt and kpno long - slit data ( negative @xmath67 values and undulations in the rising part of the rotation curve seen in mmt data ; correlation between @xmath66 and @xmath236 in kpno data ) occurring on scales of 3 - 5 , well outside the nuclear region . except for the central value of @xmath67 in the nifs dataset ( which , as we see from fig . [ fig.spec ] , is contaminated by the agn ) , all the other @xmath67 values are positive , which is what is expected for orbits around a central bh . neglecting the long - slit kinematical data could help overcome biases introduced by the bar kinematics . assuming that the bh mass should primarily be constrained by kinematic constraints at the smallest radii we now test the effect of not including the long - slit data in the nnls optimization problem . the expectation is that neglecting the long - slit data should bring the best - fit solutions obtained from @xmath183 and @xmath178 closer to each other . we construct a full sets of models ( 18 values of @xmath0 and 18 values of @xmath95 ) in which we only fitted the nuclear kinematical data obtained with the nifs spectrograph , with an orbit library of @xmath211 . figure [ fig : nifs_only ] shows @xmath138 values obtained by marginalizing over @xmath95 ( top ) and by marginalizing over @xmath0 ( middle ) . the solid curves show @xmath237 while the dashed curves show @xmath238 . the dotted horizontal lines show the 1@xmath4 , 2@xmath4 and 3@xmath4 confidence levels , respectively ( for 1 degree - of - freedom ) . the top plot shows that the minimum @xmath178 ( dashed curve ) has decreased quite significantly and is now in agreement with the minimum @xmath183 ( solid curve ) . the best - fit values of @xmath95 ( middle panel ) are still inconsistent within 1@xmath4 but are consistent within 3@xmath4 . the bottom - most panel shows the error on the fit to the mass and surface brightness distribution . contours are spaced at intervals of 0.1% error per constraint . it is clear that the fit to the self - consistency constraints is uniform over a much larger range of @xmath0 and @xmath95 values implying that the fit to the mass is no longer a significant factor in determining the location of the minimum . the best fit black hole mass and m / l ratio obtained using @xmath237 and @xmath238 , along with the reduced @xmath138 and 1@xmath4 and 3@xmath4 error bars , are shown in the bottom two rows of table [ tab : robustbestfit ] . since the best fit @xmath0 values and their 1@xmath4 error bars from both methods of computing @xmath1 are nearly the same we use the average of the two best - fit values as the solution and sum their 1@xmath4 errors in quadrature to obtain @xmath239 . figure [ fig : nifs_only ] shows that the @xmath95 values obtained from @xmath178 are consistent with the best fit value obtained with @xmath183 at the 3@xmath4 level , but both are well within the expected range @xmath240 . therefore we use the average of the two best - fit m / l values and sum their 3@xmath4 errors in quadrature to get @xmath241 . the bottom panel of figure [ fig : kinenifs_90 ] shows a 2d kinematic map for a model with @xmath219 and @xmath242 , with parameters closest to the best - fit solution obtained with @xmath125 ( see 3rd row of table [ tab : robustbestfit ] ) . this model achieves adequately large values of @xmath231 ( see red / orange regions ) over the central nifs field by raising the m / l ratio very slightly from @xmath243 to @xmath244 . the 1d kinematic fit of this model to the nifs data along the kinematic major axis of the model is shown by the green curves in figure [ fig : kinematics_90_1d ] , from which it is also clear that the lower value of @xmath0 is able to fit the high velocity dispersion values even better than the high bh mass when the model attempted to fit both the nifs and long - slit kinematic data . this test largely confirms our hypothesis that the two main causes of the large value of @xmath0 obtained in the previous section were ( a ) inconsistencies between @xmath231 values from nifs and the long - slit data in the same spatial region , ( b ) the need to fit the low @xmath67 values and high central @xmath231 values with a low m / l ratio . thus , we see that by not including the kinematics data with poorer spatial resolution ( which are clearly unaffected by the bh as can be seen in figure [ fig : kinenifs_90 ] ) we are able to simultaneously fit the nifs kinematics and the mass and surface brightness distributions with a lower value of @xmath0 . in this section , we determine how the new nifs kinematic measurements in the nuclear region of ngc 4151 affect the galaxy s location in the @xmath15 relation . the aperture used to define the appropriate @xmath4 value for the relation is a historically contentious issue ( e.g. * ? ? ? * ; * ? ? ? * ) , but the nifs data naturally lend themselves to one of those standards : @xmath245 , which is the velocity dispersion within an aperture of radius @xmath246/8 , where @xmath246 is the bulge effective radius @xcite . ( despite having to adopt one particular standard , we note that @xcite found no systematic differences between estimates of @xmath245 ( inside of @xmath246/8 ) and estimates of @xmath4 that extended out to @xmath246 , likely because of the steep radial surface brightness gradient in most bulges . ) the @xmath246 value for ngc 4151 has been measured to be @xmath247 @xcite , while the nifs field - of - view extends to a radius of between @xmath86 and @xmath248 . thus , we sum the nifs datacube across all of the spaxels into a single spectrum , and fit for the dispersion using ppxf . from the nifs data , we find @xmath249 . this is somewhat larger than either the @xmath250 measured by @xcite , or the @xmath251 measured by @xcite . both of those earlier measurements used the calcium triplet absorption lines , but whereas the former covered a similar region of the galaxy as our nifs spectra ( @xmath252 ) , the latter used an aperture of @xmath253 . previous studies of bulge kinematics have found near - ir and optical data to give consistent results ( see * ? ? ? * and references therein ) , while the discrepancies seen in some luminous ir galaxies are in the sense of _ smaller _ @xmath4 values from the co - bandheads @xcite . our new measurement of @xmath254 is @xmath1225% above the previous velocity dispersion measurements , while our best - fit stellar dynamical estimate of @xmath255 is 17% below our earlier stellar dynamical estimate @xcite . previously , ngc 4151 had been an outlier from the agn @xmath15 relation in the direction of low-@xmath4 or high-@xmath0 . the new @xmath245 and @xmath0 measurements helps to bring ngc 4151 closer to the best - fit relation of @xcite , though it does remain on the low-@xmath4 side . finally , we note that if the bh masses in barred galaxies have been overestimated due to the use of axisymmetric dynamical modeling codes , such galaxies would lie even further below the standard relation than has been found previously ( * ? ? ? * and references therein ) . this demonstrates the importance of developing robust bar dynamical modeling approaches in the future . we have conducted ao - assisted near - ir integral field spectroscopy of the local seyfert galaxy ngc 4151 , using the nifs instrument on gemini north . we used an axisymmetric orbit - superposition code to fit the observed surface brightness distribution within 50 , nifs kinematics within @xmath5315 , and long slit kinematics along two different position angles in order to estimate the best - fit values of the mass of the bh ( @xmath0 ) and the mass - to - light ratio of the stars ( @xmath95 ) . models were constructed for 10 - 18 values of @xmath0 , 18 values of @xmath95 and 3 different inclination angles . the main results of this paper are summarized below . we use the @xmath1 estimator to determine the model which gives the best - fit to the data but find that the solution depends on whether the @xmath1 includes both self - consistency constraints ( which are not strictly speaking `` data '' ) and kinematic constraints or only kinematic constraints . when we use both types of constraints and all the available kinematical data ( from low - resolution long - slit data and high resolution nifs data ) the best - fit solution is @xmath256 and @xmath175 , for a model with @xmath101 . however , this best fit model gives a central @xmath231 that is too low to fit the data obtained with the nifs instrument . when we use @xmath127 ( determined by only considering the fit to the kinematic data ) , the best fit model has @xmath257 and @xmath258 , and provides a better fit to the nuclear kinematical data from nifs spectrograph , although it gives a slightly worse fit to the mass and surface brightness distributions . an interesting point worth noting is that @xcite found a best - fit value of @xmath259 in ngc 4151 when they fitted the kinematics of the h@xmath260 line - emitting gas within 1 of the center ( see figure 43 in their paper ) . however , if they included the kinematics of gas within 2 they obtained a larger value of @xmath261 , although with a somewhat worse reduced-@xmath1 . it is intriguing that both our stellar dynamical modeling and their gas dynamical modeling , while preferring lower values of @xmath262 , also suggest that _ higher _ values of @xmath0 may be obtained when lower spatial resolution data are included in the fit . models are generated for three different inclination angles of the bulge @xmath263 and show that the best - fit @xmath0 and @xmath95 values are relatively insensitive to inclination angle . we examine an additional 4 inclination angles for fixed bh mass values ( @xmath264 ) and fixed mass - to - light ratio ( @xmath265 ) . we find that edge - on models ( @xmath101 ) give the smallest @xmath1 values , while models with the inclination of the large scale disk ( @xmath103 ) are strongly disfavored . this suggests that the bulge must be nearly spherical but its rotation axis is significantly misaligned from the rotation axis of the disk . however , it is well known from previous work @xcite that inclination of a spheroid is extremely difficult to determine via dynamical modelling , especially in the presence of a bar . this determination of the inclination of the bulge should therefore be treated with caution . a detailed examination of the kinematics in the inner 5 of ngc 4151 shows evidence for bar kinematics , which manifest as high values of @xmath231 , low or negative values of @xmath67 and @xmath66 values which are correlated with @xmath236 . we hypothesize that the discrepancy between the @xmath1 obtained from including all the constraints and that obtained by only including the kinematic constraints is likely a consequence of two factors : ( a ) a stellar bar , which is not possible to model with our axisymmetric code , and ( b ) inconsistencies between the velocity dispersion values obtained from the low - spatial resolution long - slit data and the high spatial resolution nifs data over the same spatial range . this hypothesis was tested by fitting only the nifs data . when we fit only the kinematical data obtained with the nifs spectrograph ( neglecting all kinematical constraints beyond @xmath266 - i.e those data that show kinematical evidence for a bar and data that is discrepant with high - resolution nifs data ) , the non - negative optimization problem was able to find a much better fit to the nifs data and the same best - fit @xmath0 was obtained by both methods of computing @xmath1 . the stellar - dynamical modeling carried out in this paper gives a best - fit value of @xmath267 ( obtained by averaging over the @xmath125 and @xmath127 values in the lower two rows of table [ tab : robustbestfit ] ) which is consistent at the 1@xmath4 level with the reverberation mapping - based mass of @xmath268 ( 1@xmath4 errors ) obtained by using data from @xcite , but relying on a recently updated empirical calibration of the rm mass scale by @xcite . the best fit m / l ratio @xmath269 ( 3@xmath4 error ) is consistent with the photometrically derived mass - to - light ratio of @xmath270 . the new nifs data yields @xmath254 , the velocity dispersion within @xmath271 , which is @xmath1225% larger than previous values , while the new best estimate of @xmath272 17% smaller than but fully consistent with both our previous stellar dynamical upper limit @xcite and the dynamical estimate based on h@xmath260 line - emitting gas @xcite . the larger value of @xmath245 and smaller value of @xmath0 help to bring ngc 4151 closer to the best - fit relation of @xcite . the analysis in this paper demonstrates that biased estimates of bh masses can arise when an axisymmetric orbit superposition code is used to model a galaxy with a weak , but kinematically identifiable barred galaxy , possibly resulting in an over - estimate in @xmath0 if the m / l ratio is constrained primarily by data beyond the sphere - of - influence of the bh . this confirms the prediction made from the recent analysis of @xmath229-body simulations of barred galaxies with bhs by @xcite , and points to the need for new dynamical modeling tools capable of modeling a stellar bar . when such codes are applied to the existing sample of barred galaxies , our results suggest that it will likely enhance the discrepancy between barred and unbarred galaxies in the @xmath15 relation @xcite . our modeling of this complex system suggests that the standard practice of fitting kinematical constraints over a large range of radii with a constant mass - to - light ratio ( for exceptions to this practice , see * ? ? ? * ; * ? ? ? * ) can bias the mass of the central bh and the derived mass - to - light ratio . to some extent , the biases can be overcome by only considering the kinematics very close to the central bh , although this gives larger errors on the estimated solutions of @xmath0 and @xmath95 . our tests of robustness demonstrate that the use of such high spatial resolution nuclear kinematical data in such modeling is very valuable . however , the complications of the stellar dynamical bar imply that additional work is required in order to perform the crucial test of the rm mass scale calibration in this galaxy . only with improved non - axisymmetric modeling methods and/or bh mass measurements in other reverberation - mapped agns will we ultimately be able to assess whether the growth history of the bhs that we see accreting at present is systematically different from those currently in quiescence . we thank remco van den bosch for valuable input and for providing his symmetrization routine . we thank gaelle dumas and eric emsellem for providing us with the sauron integral field kinematics . we thank linda watson for providing copies of her nifs stellar velocity template spectra . we thank the anonymous referee for suggestions that helped to improve our manuscript . we are grateful to michele cappellari for making his analysis routines publicly available and for his continuing efforts in adding software features . m. valluri and jb were supported by national science foundation grant ast-0908346 , and m. valluri acknowledges support from the university of michigan s elizabeth crosby award . bmp and rwp are grateful to the national science foundation for support of this work through grant ast-1008882 to the ohio state university . mb gratefully acknowledges support by the national science foundation under grant ast-1253702 . m. vestergaard acknowledges support from a freja fellowship granted by the dean of the faculty of natural sciences at the university of copenhagen and a marie curie international incoming fellowship . the research leading to these results has received funding from the people programme ( marie curie actions ) of the european union s seventh framework programme fp7/2007 - 2013/ under rea grant agreement no . 300553 ( mv ) . the dark cosmology centre is funded by the danish national research foundation . based on observations obtained at the gemini observatory , which is operated by the association of universities for research in astronomy , inc . , under a cooperative agreement with the nsf on behalf of the gemini partnership : the national science foundation ( united states ) , the science and technology facilities council ( united kingdom ) , the national research council ( canada ) , conicyt ( chile ) , the australian research council ( australia ) , ministrio da cincia e tecnologia ( brazil ) and ministerio de ciencia , tecnologa e innovacin productiva ( argentina ) . this research was supported in part through computational resources and services provided by advanced research computing and the flux cluster at the university of michigan , ann arbor . this work made use of data from the ohio state university bright spiral galaxy survey , which was funded by grants ast-9217716 and ast-9617006 from the united states national science foundation , with additional support from the ohio state university . based on observations made with the nasa / esa hubble space telescope , obtained from the data archive at the space telescope institute . stsci is operated by the association of universities for research in astronomy , inc . under the nasa contract nas 5 - 26555 . funding for the sdss and sdss - ii has been provided by the alfred p. sloan foundation , the participating institutions , the national science foundation , the u.s . department of energy , the national aeronautics and space administration , the japanese monbukagakusho , the max planck society , and the higher education funding council for england . the sdss is managed by the astrophysical research consortium for the participating institutions . the participating institutions are the american museum of natural history , astrophysical institute potsdam , university of basel , university of cambridge , case western reserve university , university of chicago , drexel university , fermilab , the institute for advanced study , the japan participation group , johns hopkins university , the joint institute for nuclear astrophysics , the kavli institute for particle astrophysics and cosmology , the korean scientist group , the chinese academy of sciences ( lamost ) , los alamos national laboratory , the max - planck - institute for astronomy ( mpia ) , the max - planck - institute for astrophysics ( mpa ) , new mexico state university , ohio state university , university of pittsburgh , university of portsmouth , princeton university , the united states naval observatory , and the university of washington . this research has made use of the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration . this research has made use of the simbad database , operated at cds , strasbourg , france . , m. c. , denney , k. d. , cackett , e. m. , dietrich , m. , fogel , j. k. j. , ghosh , h. , horne , k. , kuehn , c. , minezaki , t. , onken , c. a. , peterson , b. m. , pogge , r. w. , pronik , v. i. , richstone , d. o. , sergeev , s. g. , vestergaard , m. , walker , m. g. , & yoshii , y. 2006 , , 651 , 775 , m. c. , walsh , j. l. , barth , a. j. , baliber , n. , bennert , v. n. , canalizo , g. , filippenko , a. v. , ganeshalingam , m. , gates , e. l. , greene , j. e. , hidas , m. g. , hiner , k. d. , lee , n. , li , w. , malkan , m. a. , minezaki , t. , sakata , y. , serduke , f. j. d. , silverman , j. m. , steele , t. n. , stern , d. , street , r. a. , thornton , c. e. , treu , t. , wang , x. , woo , j .- h . , & yoshii , y. 2009 , , 705 , 199 , m. c. , walsh , j. l. , barth , a. j. , yoshii , y. , woo , j .- h . , wang , x. , treu , t. , thornton , c. e. , street , r. a. , steele , t. n. , silverman , j. m. , serduke , f. j. d. , sakata , y. , minezaki , t. , malkan , m. a. , li , w. , lee , n. , hiner , k. d. , hidas , m. g. , greene , j. e. , gates , e. l. , ganeshalingam , m. , filippenko , a. v. , canalizo , g. , bennert , v. n. , & baliber , n. 2010 , , 716 , 993 , g. , de vaucouleurs , a. , corwin , jr . , h. g. , buta , r. j. , paturel , g. , & fouqu , p. 1991 , third reference catalogue of bright galaxies . volume i : explanations and references . volume ii : data for galaxies between 0@xmath273 and 12@xmath273 . volume iii : data for galaxies between 12@xmath273 and 24@xmath273 . ( springer - verlag : new york ) , p. b. , frogel , j. a. , pogge , r. w. , quillen , a. c. , berlind , a. a. , davies , r. l. , depoy , d. l. , gilbert , k. m. , houdashelt , m. l. , kuchinski , l. e. , ramrez , s. v. , sellgren , k. , stutz , a. , terndrup , d. m. , & tiede , g. p. 2002 , , 143 , 73 , k. , richstone , d. , tremaine , s. , lauer , t. r. , bender , r. , bower , g. , dressler , a. , faber , s. m. , filippenko , a. v. , green , r. , grillmair , c. , ho , l. c. , kormendy , j. , magorrian , j. , & pinkney , j. 2003 , , 583 , 92 , c. l. , merritt , d. , olling , r. , valluri , m. , bender , r. , bower , g. , danks , a. , gull , t. , hutchings , j. , kaiser , m. e. , maran , s. , weistrop , d. , woodgate , b. , malumuth , e. , nelson , c. , plait , p. , & lindler , d. 2001 , , 550 , 668 , p .- y . , cappellari , m. , emsellem , e. , bournaud , f. , michel - dansac , l. , alatalo , k. , blitz , l. , bois , m. , bureau , m. , davies , r. l. , davis , t. a. , de zeeuw , p. t. , duc , p .- a . , khochfar , s. , krajnovi , d. , kuntschner , h. , morganti , r. , mcdermid , r. m. , naab , t. , oosterloo , t. , sarzi , m. , scott , n. , serra , p. , weijmans , a .- m . , & young , l. m. 2012 , , 424 , 1495 , p. j. , hart , j. , conroy , p. g. , pfitzner , m. l. , bloxham , g. j. , jones , d. j. , downing , m. d. , dawson , m. , young , p. , jarnyk , m. , & van harmelen , j. 2003 , in society of photo - optical instrumentation engineers ( spie ) conference series , vol . 4841 , society of photo - optical instrumentation engineers ( spie ) conference series , ed . m. iye & a. f. m. moorwood , 15811591 , c. a. , valluri , m. , peterson , b. m. , pogge , r. w. , bentz , m. c. , ferrarese , l. , vestergaard , m. , crenshaw , d. m. , sergeev , s. g. , mchardy , i. m. , merritt , d. , bower , g. a. , heckman , t. m. , & wandel , a. 2007 , , 670 , 105 ( paper i ) , s. , gebhardt , k. , bender , r. , bower , g. , dressler , a. , faber , s. m. , filippenko , a. v. , green , r. , grillmair , c. , ho , l. c. , kormendy , j. , lauer , t. r. , magorrian , j. , pinkney , j. , & richstone , d. 2002 , , 574 , 740 , j .- h . , treu , t. , barth , a. j. , wright , s. a. , walsh , j. l. , bentz , m. c. , martini , p. , bennert , v. n. , canalizo , g. , filippenko , a. v. , gates , e. , greene , j. , li , w. , malkan , m. a. , stern , d. , & minezaki , t. 2010 , , 716 , 269

closed - orbit theory was first introduced by du and delos @xcite and bogomolny @xcite some twenty years ago to interpret the modulations observed in the photo - absorption spectra of hydrogenic rydberg atoms in a magnetic field close to the ionization threshold . since that time , it turned out to be a powerful and flexible tool for the semiclassical interpretation of a variety of spectra . it has been used to describe atoms in electric @xcite as well as parallel @xcite or crossed @xcite electric and magnetic fields . in the case of non - hydrogenic atoms , the influence of the ionic core can be modelled either by means of an effective classical potential @xcite or in terms of quantum defects @xcite . recently , closed - orbit theory has even been shown to be applicable to the spectra of simple molecules in external fields @xcite . a complete description of photo - absorption spectra requires the calculation of the energies @xmath2 of the excited atomic states and the strengths of the spectral lines , which is characterized by the dipole matrix elements @xmath3 between the initial state @xmath4 and the rydberg state @xmath5 , where @xmath6 is the component of the dipole operator describing the polarization of the exciting laser field . these quantities are neatly summarized in the response function @xmath7 where @xmath8 denotes the retarded green s function . closed - orbit theory provides a semiclassical approximation to the quantum response function ( [ gdef ] ) , which splits into a smooth part and an oscillatory part of the form @xmath9 where the sum extends over all classical closed orbits starting from the nucleus and returning to it after having been deflected by the external fields , @xmath10 is the classical action of the closed orbit , and the amplitude @xmath11 describes its stability and its starting and returning directions . its precise form depends on the geometry of the external fields . in section [ sec : smatrix ] , it will be specified for systems with and without a rotational symmetry . although the closed - orbit sum ( [ resoscgen ] ) appears to provide a straightforward means of calculating the response function from the classical closed orbits , this is actually not the case because the sum usually diverges for real energies @xmath12 . thus , the quantal information can not be extracted directly from the semiclassical expansion . one particular and widely applicable method to overcome the convergence problems of the closed - orbit sum is semiclassical quantization by harmonic inversion @xcite . for the hydrogen atom in a magnetic field , this method has been shown @xcite to be capable of extracting semiclassical eigenenergies and transition matrix elements from a closed - orbit sum . in the present paper we will investigate how these results can be generalized to the hydrogen atom in crossed electric and magnetic fields . this problem is considerably harder than the treatment of the diamagnetic hydrogen atom , which possesses a rotational symmetry around the field axis . due to that symmetry , in classical mechanics the angular momentum around the field axis is conserved . so is , in quantum mechanics , the magnetic quantum number @xmath13 . in crossed fields , the rotational symmetry is broken . as a consequence , the selection rules for the @xmath13-quantum number no longer hold , and a multitude of additional lines appears in the quantum spectrum . at the same time , the determination of classical closed orbits gets significantly more difficult because three non - separable degrees of freedom have to be dealt with . a detailed description of the intricate pattern of closed orbits and their bifurcations was given in an accompanying paper @xcite . that data forms the basis of the present work , where the semiclassical treatment of the crossed - fields hydrogen atom will be dealt with , and we will freely use the nomenclature introduced in @xcite . after the essential properties of the crossed - fields hamiltonian have been summarized in section [ sec : classham ] , we start , in section [ sec : smatrix ] , with a derivation of the closed - orbit formula ( [ resoscgen ] ) in the context of the @xmath0-matrix formulation of closed - orbit theory introduced recently by granger and greene @xcite . we show that the novel framework can be extended to the crossed - fields situation , and we clarify some misleading conclusions arrived at in @xcite . section [ sec : xqm ] describes the quantum spectrum under study , and section [ sec : scllo ] compares it to a semiclassical spectrum in low resolution . in section [ sec : sclhi ] , the results of a high - resolution semiclassical quantization using the technique of harmonic inversion are presented . the semiclassical spectrum correctly identifies the strongest spectral lines , but it fails to describe finer details of the quantum spectrum . in section [ sec : sclrec ] , we compare a quantum recurrence spectrum to the classical data to show that the principal source of this difficulty lies in the abundance of closed - orbit bifurcations . uniform approximations provide a tool to cope with the divergences introduced into semiclassical spectra by bifurcations of classical orbits . a general technique for their construction is described in section [ sec : uniform ] , and uniform approximations for the two types of generic codimension - one bifurcations identified in @xcite are derived . finally , in section [ ssec : unifrecur ] we demonstrate how uniform approximations can be incorporated into recurrence spectra , thus paving the way for their inclusion into the high - resolution semiclassical quantization by harmonic inversion . throughout this work , we will assume the magnetic field to be directed along the @xmath14-axis and the electric field to be directed along the @xmath15-axis . in atomic units , the hamiltonian describing the motion of the atomic electron then reads @xmath16 where @xmath17 and @xmath18 denote the magnetic and electric field strengths , respectively , @xmath19 , @xmath20 , and @xmath21 is the @xmath14-component of the angular momentum vector . by virtue of the scaling properties of the hamiltonian ( [ specham ] ) , if all classical quantities are multiplied by suitable powers of the scaling parameter @xmath22 the dynamics can be shown not to depend on the energy @xmath12 and the field strengths @xmath17 and @xmath18 separately , but only on the scaled energy @xmath23 and the scaled electric field strength @xmath24 . in particular , classical actions scale according to @xmath25 . thus , the semiclassical limit of large classical actions corresponds to the limit of large @xmath26 . the way of recording a quantum spectrum which is best suited for semiclassical investigations is scaled - energy spectroscopy . a spectrum then consists of a list of the scaling parameters @xmath27 characterizing the quantum states for given scaled energy @xmath28 and scaled electric field strength @xmath29 . scaled - energy spectroscopy offers the advantage that the underlying classical dynamics does not change across the spectrum . it will be adopted throughout this work . the basic observation fundamental to all of closed - orbit theory is a partition of space into physically distinct regions . in the core region close to the nucleus , the rydberg electron interacts in a complicated manner with all electrons of the ionic core . this interaction is manifestly quantum mechanical in nature , it can not be described in the framework of semiclassical theories . on the other hand , the interaction of the rydberg electron with the external fields is much weaker in the core region than its interaction with the core , so that the fields can safely be neglected . therefore , a description of the core obtained in the field - free case can be used . in particular , the initial state of the photo - absorption process is assumed to be localized in the core region and not to be influenced by the external fields . in the long - range region far away from the nucleus , on the other hand , the external fields play a dominant role , whereas there is no interaction with the ionic core except for the coulomb attraction of its residual charge . in this region , the dynamics of the rydberg electron is well - suited for a semiclassical description . it is independent of the details of the ionic core . in order to establish a link between the dynamics in the core and long - range regions , a matching region is assumed to exist at intermediate distances from the nucleus where both the external fields and the interaction with the core are negligible . thus , in the matching region the simple physics of an electron subject to the residual coulomb field of the core is observed . recently , granger and greene @xcite proposed a novel formulation of the theory based on ideas borrowed from quantum - defect theory . their formulation achieves a clear separation between properties of the external field configuration and the ionic core , which are encoded in separate @xmath0-matrices . suitable approximations to the core and the long - range @xmath0-matrices can be derived independently . therefore , the formalism can be expected to allow a generalization of closed - orbit theory to atoms with ionic cores exhibiting more complicated internal dynamics than have been treated so far . the derivation given by granger and greene treated the case of an atom in a magnetic field only . it will now be extended in such a way that it holds for combined electric and magnetic fields with arbitrary field configurations . to this end , the ansatz and basic formulae of granger and greene s theory will be summarized in this section . a more detailed treatment can be found in their paper @xcite . in subsequent sections , we will then turn to a discussion of the long - range scattering matrices pertinent to different external field configurations . to lay the foundation for a definition of the @xmath0-matrices , we pick a basis set @xmath30 and @xmath31 of wave functions of the rydberg electron valid in the core and long - range regions , respectively , and expand in terms of spherical harmonics @xmath32 the channel index @xmath33 is to be understood as a double index @xmath34 characterizing the spherical harmonics . when studying a complicated atom with more than one relevant state of the core , additional information can be included in the channel functions @xmath35 . in the matching region , the radial function matrices @xmath36 and @xmath37 can both be expressed in terms of radial coulomb functions . we use the functions @xmath38 and @xmath39 satisfying outgoing and incoming wave boundary conditions , respectively , given by @xcite and choose the radial functions to be of the form and @xmath40 used by granger and greene @xcite differ from those used by robicheaux @xcite in that they are energy - normalized in rydberg rather than in hartree units . the radial function matrices given here agree in normalization with those adopted by robicheaux , whereas the matrices used by granger and greene are inconsistent with their equation ( 12 ) . ] @xmath41\;,\ ] ] @xmath42\;.\ ] ] physically , these choices mean that the basis function @xmath30 is a superposition of a single incoming wave in channel @xmath33 and the outgoing waves in different channels produced from it by scattering off the core . similarly , @xmath31 consists of an outgoing wave in channel @xmath33 and the returning waves generated by scattering off the external fields . the scattering matrices @xmath43 and @xmath44 thus summarize the physical properties of the core and the external fields , respectively . they are determined by the condition that the radial functions obey suitable boundary conditions , i.e. @xmath36 is regular at the origin , whereas @xmath37 vanishes or satisfies outgoing - wave boundary conditions at infinity for bound and free states , respectively . for hydrogen , @xmath43 is the identity matrix . following previous work by robicheaux @xcite , granger and greene derive the following expression for the response function ( [ gdef ] ) : @xmath45 where the vector @xmath46 comprises the energy - dependent dipole matrix elements @xmath47 between the initial state and the core - region channel wave functions . for hydrogen they can be computed explicitly ( see , e.g. , @xcite or @xcite ) . the terms of the series ( [ resseries ] ) embody contributions from paths where the rydberg electron takes zero , one , two , etc . trips out into the long - range region and back to the core before interfering with the initial outgoing wave . in the semiclassical approximation , @xmath44 will be given in terms of closed orbits . a returning wave is associated with each returning classical orbit . by a general ionic core , it is scattered into all directions . the parts of the wave scattered into the outgoing direction of a closed orbit will then follow this orbit until they return to the core again . thus , core scattering leads to a concatenation of different closed orbits @xcite . in hydrogen , the coulomb center scatters the incoming wave back into its direction of incidence , so that there is no coupling of closed orbits . terms describing repeated scattering off the external fields are therefore absent from the sum , and the hydrogen response function can be decomposed into a smooth part @xmath48 which is the same as in the field - free case and contains `` direct '' contributions where the electron does not scatter off the external fields at all , and an oscillatory part @xmath49 generated by the electron going out into the long - range region and being scattered back to the nucleus . it is this part which describes the impact of the external fields . the basis for a semiclassical approximation is provided by the retarded green s function @xmath50 describing the propagation of the electron from @xmath51 to @xmath52 at the energy @xmath12 . it can be expanded in terms of the channel functions as @xmath53 with @xmath54 the long - range scattering matrix is related to the green s function matrix by @xcite @xmath55^{-1 } \underline{g}(r_0,r_0 ) [ \underline{f}^-(r_0)]^{-1 } \;,\ ] ] where @xmath56 is the matching radius , @xmath57 is the diagonal matrix @xmath58 comprising the radial wave functions , and @xmath59 denotes the part of @xmath60 satisfying incoming - wave boundary conditions at the final radius @xmath61 . the latter condition ensures that only electron paths approaching the matching radius from the long - range region contribute to @xmath44 , whereas paths that traverse the core region are omitted . to obtain a semiclassical approximation to the long - range scattering matrix , we make use of the semiclassical green s function derived by gutzwiller @xcite @xmath62 where the sum extends over all classical trajectories leading from @xmath51 to @xmath52 at the energy @xmath12 , @xmath1 is the number of degrees of freedom , @xmath0 is the classical action along the trajectory , @xmath63 the number of caustics along the trajectory , and @xmath64 \frac{\partial^2 s}{\partial e \partial{\boldsymbol}x ' } & \frac{\partial^2 s}{\partial e^2 } \end{array } \right)\ ] ] is the amplitude for the contribution of the trajectory . by ( [ greenchannel ] ) , we obtain a semiclassical approximation to the green s function matrix @xmath65 as usual in semiclassics , the integrals will be evaluated in the stationary - phase approximation . it yields a sum over all classical trajectories leaving the matching sphere at a direction given by @xmath66 and returning to it at @xmath67 . the condition that @xmath68 obeys incoming - wave boundary conditions at the final radius translates into the condition that only orbits going out from the matching sphere into the long - range region and then returning to @xmath56 are to be included , whereas orbits passing through the core region are omitted . if all factors in the integrand except for the exponential are assumed to vary slowly , the stationary - phase approximation reads @xmath69 where @xmath70 is the number of negative eigenvalues of the hessian matrix of @xmath0 occurring in the prefactor . because the initial state is assumed to be well localized , it is clear that the outgoing waves generated by the photo - excitation originate in the immediate neighborhood of the nucleus . therefore , only trajectories leaving the matching sphere radially need to be included in ( [ sclgreensp ] ) . by the same token , the trajectories can be assumed to return to the matching radius radially . thus , they are parts of closed orbits starting precisely at the nucleus and returning there . by transforming ( [ greenddef ] ) to spherical coordinates and making use of the relations @xmath71 the amplitude factor @xmath6 for radial trajectories can be simplified to @xmath72 the determinants occurring in ( [ sclgreensp ] ) combine to @xmath73 = & \det\frac{\partial(p_\vartheta',p_\varphi',p_\vartheta , p_\varphi ) } { \partial(\vartheta,\varphi , p_\vartheta , p_\varphi ) } \cdot \left(\det\frac{\partial(-p_\vartheta',-p_\varphi',p_\vartheta , p_\varphi ) } { \partial(\vartheta',\varphi',\vartheta,\varphi ) } \right)^{-1 } \\[.5ex ] = & \det\frac{\partial(\vartheta',\varphi ' ) } { \partial(p_\vartheta , p_\varphi ) } \;. \end{split}\ ] ] with these results , the green s function matrix assumes the form @xmath74 the determinant in the denominator of ( [ sclgreenchannel2 ] ) measures the dependence of the final angular momenta of the trajectory upon the starting angles . as it stands , it suffers from the singularities of the spherical coordinate chart : at the poles , neither the angle @xmath75 nor the angular momenta @xmath76 and @xmath77 are well defined , so that close to the poles , the calculation of the determinant becomes numerically unstable . the determinant can be rewritten in the form @xcite @xmath78 with a @xmath79-determinant @xmath80 devoid of any singularities . the parameter @xmath80 was already used in @xcite to study the bifurcations of closed orbits . we showed there that a closed orbit bifurcates if and only if @xmath81 . with the form ( [ mprime ] ) of the stability determinant , the semiclassical green s function matrix reads @xmath82 which is free of any singularities introduced by the spherical coordinates . by virtue of ( [ sgreen ] ) , the semiclassical long - range scattering matrix reads @xmath83 this expression can be further simplified if , for excited states close to the ionization threshold , the radial wave functions @xmath84 are approximated by the zero - energy wave functions , and the hankel functions are replaced with their asymptotic forms for large arguments @xcite @xmath85 this approximation has proven accurate in many cases of interest , but it was called into question by granger and greene @xcite . it will be discussed further in section [ sec : corot ] , where we will show that there is no reason to doubt its reliability . it leads to @xmath86 because , due to the conservation of energy , @xmath87 if @xmath88 . in equation [ scrossed0 ] , the channel indices @xmath89 are finally written out explicitly . for a radial trajectory in a hydrogen atom going out from the nucleus to @xmath90 at zero energy , the action is @xmath91 , so that @xmath92 is the action of a closed orbit , measured from its start at the nucleus to its return . the semiclassical long - range @xmath0-matrix finally reads @xmath93 both the action @xmath10 and the stability determinant @xmath80 are evaluated at the nucleus rather than on the matching sphere . the response function is given by @xmath94 where the maslov index @xmath95 was increased by 1 to absorb an additional phase , and the function @xmath96 with the core - region matrix elements @xmath97 given by ( [ channelmat ] ) , characterizes the initial state and the exciting photon . through the @xmath97 , the function @xmath98 is energy - dependent . in accordance with the choice of zero - energy radial wave functions in the @xmath0-matrix elements , @xmath98 will be evaluated at zero energy . this approximation has proven accurate in all applications of closed - orbit theory considered in the literature so far . however , from the @xmath0-matrix theory derivation it is obvious that the energy - dependence of both the dipole matrix elements @xmath97 and the @xmath0-matrix elements can easily be included should the need arise . the semiclassical response function ( [ gcrossed ] ) has the anticipated form ( [ resoscgen ] ) with @xmath99 an atom in a single ( electric or magnetic ) external field possesses a rotational symmetry around the field axis , which must be taken into account in the derivation of the closed - orbit formulae . the symmetry gives rise to a conserved magnetic quantum number @xmath13 , so that the angular momentum quantum number @xmath100 remains the only relevant channel index . the semiclassical scattering matrix reads @xcite @xmath101 where @xmath102 @xmath103 is the number of poles of @xmath104 encountered along the trajectory , and the sum includes all classical trajectories with azimuthal angular momentum @xmath13 joining the circles given by polar angles @xmath105 and @xmath106 on the matching sphere . if the radius of the matching sphere is much larger than the extent of the initial state , the trajectories can again be assumed to leave the sphere and return to it radially . strictly speaking , this condition can only be met if @xmath107 , which we will assume in what follows . if @xmath108 , the initial angular velocity @xmath109 must be non - zero , but it will be small if the matching radius is large . in this case , the trajectory will not actually close at the nucleus , but swing by at a short distance . using , as above , the radial wave functions at zero energy , we obtain the semiclassical scattering matrix @xmath110 and the response function @xmath111 with @xmath112 and @xmath113 this result has the form ( [ resoscgen ] ) with @xmath114 it differs from the result obtained previously by du and delos @xcite in that in their work the amplitude factor @xmath104 of ( [ agg ] ) is replaced with @xmath115 this discrepancy was noted and numerically investigated by granger and greene @xcite . they attribute it to the approximation of using zero - energy wave functions , which can easily be avoided in the @xmath0-matrix theory , but is an integral part of the derivation given by du and delos . ) , solid line ) and after du and delos @xcite ( equation ( [ add ] ) , dashed line ) for the closed orbit perpendicular to the magnetic field as a function of the scaled energy . the scaled matching radius is @xmath116 . ] for the closed orbit perpendicular to the field in the diamagnetic kepler problem and a scaled matching radius of @xmath116 , the amplitudes ( [ agg ] ) and ( [ add ] ) are plotted in figure [ ampefig ] . this figure is similar to figure 1 in @xcite , although for the latter the matching radius is not given . the agreement is excellent at scaled energies close to zero , but becomes poor if the energy decreases . however , contrary to their conclusions , the lack of agreement is not due to the zero - energy approximation , but rather to the dependence of the amplitudes on the matching radius . this statement can be verified most conveniently if the motion is described in semiparabolic coordinates @xmath117 if the trajectory is recorded as a function of a parameter @xmath118 related to the time @xmath119 by @xmath120 and a prime denotes differentiation with respect to @xmath118 , for trajectories with vanishing azimuthal angular momentum the equations of motion in the coulomb region read @xmath121 these equations are devoid of any singularities , so that they can conveniently be used to discuss the motion close to the nucleus . the transformation inverse to ( [ sparcoord ] ) is given by @xmath122 the momenta transform according to @xmath123 note that the transformation from semiparabolic to cartesian coordinates is not one - to - one , but that @xmath124 and @xmath125 are fixed up to the choice of sign only . to evaluate ( [ agg ] ) and ( [ add ] ) , the derivatives @xmath126 and @xmath127 , must be calculated and their dependence on the matching radius @xmath61 must be determined . as the radial trajectory specified by a starting angle @xmath105 is independent of the radius where the angle is measured , the @xmath61-dependence of the derivatives is determined by the returning trajectories only . it can be evaluated as follows : we arbitrarily fix the returning time of a closed orbit at @xmath128 , so that @xmath129 . the solution to ( [ speqns ] ) describing a trajectory returning at an angle @xmath106 is given by @xmath130 where the coefficients were chosen to satisfy the conservation of energy and to give the correct returning angle after a transformation to cartesian coordinates . the second expression in each line follows from @xmath131 , whence for @xmath132 @xmath133 equations of motion for the derivatives @xmath134 and @xmath135 are obtained by linearizing ( [ speqns ] ) . since ( [ speqns ] ) is already linear , the derivatives satisfy the same equations of motion as the coordinates themselves as long as the electron moves in the coulomb region . there the solutions read @xmath136 and @xmath137 equation ( [ scr ] ) yields @xmath138 so that the coefficients @xmath139 can be identified with the values of the derivatives obtained at @xmath140 . analogous expressions hold for @xmath135 . from ( [ spmomenta ] ) , the amplitude ( [ agg ] ) @xmath141 can be evaluated . it is independent of @xmath61 , as could have been anticipated from the fact that @xmath76 is a component of the total angular momentum and thus is conserved along the trajectory once the electron has entered the coulomb region . the amplitude @xmath142 can also , up to an immaterial choice of sign , be identified with the monodromy matrix element @xmath143 introduced by bogomolny @xcite to describe the semiclassical amplitudes , so that the amplitudes derived by granger and greene from the @xmath0-matrix theory agree with bogomolny s . similarly , the amplitude ( [ add ] ) used by du and delos reads , by ( [ spinverse ] ) , @xmath144 \\ & = \frac{1}{a } + { \cal o}\left(\sqrt{r}\right ) \;. \end{split}\ ] ] thus , the amplitudes @xmath104 and @xmath145 agree in the limit of vanishing matching radius , but the amplitude @xmath145 proposed by du and delos exhibits a strong dependence on @xmath61 , whereas the amplitude @xmath104 given by granger and greene does not . these findings can also be confirmed numerically . figure [ amprfig ] shows the two amplitudes for the closed orbit perpendicular to the magnetic field at a scaled energy of @xmath146 as a function of the scaled matching radius @xmath147 . the dependence of @xmath145 on @xmath147 is considerable . ) , solid line ) and after du and delos @xcite ( equatiion ( [ add ] ) , dashed line ) for the closed orbit perpendicular to the magnetic field as a function of the matching radius at @xmath146 . ] we have thus shown that , contrary to the conclusion reached by granger and greene , the discrepancy between their semiclassical amplitude and that obtained by du and delos is not due to the zero - energy approximation , but rather due to the choice of a finite matching radius . in addition , the amplitude derived by granger and greene is not specific to the @xmath0-matrix formulation , it agrees with the result derived earlier by bogomolny in the context of a semiclassical wave function formalism . nevertheless , as it eliminates the need to specify a finite matching radius and allows one to calculate all classical quantities at the nucleus , it seems more appropriate than the amplitude given by du and delos , which introduces a certain arbitrariness in the choice of a matching radius . if schrdinger s equation for the crossed - fields hydrogen atom is rewritten in terms of the scaled energy and the scaled electric field strength , a quadratic eigenvalue problem for the scaling parameter @xmath26 is obtained . an exact numerical method of solution for the quadratic eigenvalue problem has become available only recently @xcite . we resort to the method introduced by main @xcite , which relies on an approximate linearization of the eigenvalue problem to compute eigenvalues in a small spectral interval . the accuracy of the linearization can be verified by comparing results calculated using different overlapping intervals . the eigenvalues are obtained to a relative accuracy of at least @xmath148 , which is far beyond the typical accuracy of semiclassical approximations , so that the algorithm is well suited to this study . and the scaled electric field strength @xmath149 . the initial state is @xmath150 , the light is polarized along the magnetic field axis . the plot shows the squared dipole matrix elements , which for graphical reasons are multiplied by @xmath26 . the strengths of the extraordinarily strong lines of the lowest @xmath1-manifolds at @xmath151 are scaled down by a factor of 0.2 . ] in the following we will discuss quantum and semiclassical photo - absorption spectra obtained for the scaled energy @xmath152 and the scaled electric field strength @xmath149 with the initial state @xmath150 and light linearly polarized along the magnetic field axis . a quantum spectrum for these parameter values is shown in figure [ quantumfig ] . as for a semiclassical analysis ( see section [ sec : sclrec ] ) it is essential to have as many eigenvalues available as possible , the calculation was extended up to @xmath153 . the spectrum shown in figure [ quantumfig ] contains nearly 30,000 lines , many of which are too weak to be discernible in the plot . the eigenenergies of the field - free hydrogen atom satisfy @xmath154 so that in the scaled spectrum the unperturbed @xmath1-manifolds appear equidistantly spaced at @xmath155 these spacings can clearly be recognized in figure [ quantumfig ] . at low values of @xmath26 , neighboring @xmath1-manifolds are isolated . furthermore , in this region the magnetic quantum number @xmath13 is nearly conserved . this is apparent from the fact that each @xmath1-manifold contains a central group of strong levels corresponding to @xmath107 , which can be excited even at @xmath156 , and adjacent groups of considerably weaker levels with @xmath157 . levels with higher magnetic quantum numbers are too weak in this region to be seen in the figure . at higher values of @xmath26 , the conservation of @xmath13 is violated , and individual @xmath1-manifolds acquire strong side bands . at even higher @xmath26 , different @xmath1-manifolds strongly overlap . throughout the spectral range shown , groups of strong lines indicating the centers of different @xmath1-manifolds are clearly discernible . a semiclassical approximation to a scaled photo - absorption spectrum is obtained if the closed - orbit theory formulae of section [ sec : cocrossed ] are rewritten in terms of scaled quantities , viz . @xmath158 with @xmath159 when low - resolution photo - absorption spectra are to be calculated from ( [ goscscal ] ) , a method of cut - off must be adopted to deal with the divergence of the semiclassical closed - orbit sum . for this section , we choose a gaussian cut - off , i.e. ( [ goscscal ] ) is replaced with @xmath160 so that orbits with scaled actions larger than the cut - off action @xmath63 are smoothly suppressed . this smoothing corresponds to a convolution of the quantum signal with a gaussian of width @xmath161 . and ( b ) @xmath162 . ] to facilitate the comparison of ( [ goscsmooth ] ) with the convoluted quantum spectrum , we added the smooth part of the spectrum to @xmath163 , which was calculated by convoluting the quantum spectrum with a gaussian of width @xmath164 . this function is broad enough to wipe out the distinction between neighboring principal quantum numbers . results obtained for @xmath165 and @xmath162 are shown in figure [ xlofig ] . in both cases it is apparent that the large - scale structure of equidistant principal quantum numbers is well reproduced by the semiclassical approximation . in the quantum spectra , the substructure of the individual @xmath1-shells can be discerned to a certain degree , given by the smoothing width @xmath161 . in the case of @xmath165 , much of this fine structure is also present in the semiclassical spectrum , but often the agreement is not good quantitatively . in particular , the peaks corresponding to the lowest @xmath1-manifolds are considerably wider in the semiclassical than in the quantum spectrum . if the cut - off action is increased to @xmath162 , finer details are resolved in the quantum spectrum . at the same time , the semiclassical closed - orbit sum becomes more oscillatory to reproduce this fine structure . it appears , however , to be somewhat over - oscillatory , developing structures absent from the quantum spectrum . this type of behavior is typical of closed - orbit sums in non - integrable systems . thus , it can be questioned if the low - resolution closed - orbit sum can meaningfully be extended to even longer orbits . a high - resolution quantization based on the present semiclassical approximation will be presented in the following section . for the calculation of a scaled semiclassical spectrum , the method of semiclassical quantization by harmonic inversion of @xmath166 function signals @xcite can be applied . this technique requires the inclusion of closed orbits up to a maximum scaled action , i.e. it replaces the gaussian cut - off used for the low - resolution semiclassical spectra presented in the previous section with a rectangular cut - off . a rough estimate for the required cut - off action can be obtained by means of perturbation theory @xcite . @xmath167 for the case @xmath168 and @xmath169 , i.e. @xmath170 , this estimate yields @xmath171 . according to ( [ smaxpert ] ) , to compute levels at high quantum numbers @xmath1 a long semiclassical signal is needed , which can be hard or even impossible to obtain . we calculated closed orbits up to @xmath172 , so that the orbital data is available for nearly 18,000 closed - orbit multiplets . however , for reasons to be described in section [ sec : sclrec ] a useful semiclassical signal can be constructed up to @xmath173 only , so that , from the above estimate , the semiclassical calculation can not reach manifolds much higher than @xmath174 . on the other hand , the semiclassical approximation must be expected to yield more accurate results for higher quantum numbers . thus , when a high - resolution semiclassical spectrum is to be calculated , a compromise must be made between the contradictory requirements of describing a spectral region at sufficiently high quantum numbers and with a sufficiently low spectral density . @r|rrr@ @xmath1 & @xmath175 ( scl . ) & @xmath175 ( qm . ) & @xmath176 + & & 9.88321 & 1.3617 + & & 9.91431 & 3.1145 + & & 9.97747 & 1.7474 + & 10.05366 & 10.05912 & 51.0512 + 6 & 10.09551 & 10.09621 & 20.9313 + & 10.15461 & 10.15378 & 7.0060 + & & 10.24076 & 0.9608 + & & 10.26612 & 2.0777 + & & 10.31803 & 1.9385 + & & 11.56497 & 2.5663 + & 11.60898 & 11.60820 & 2.5875 + & 11.66889 & 11.67341 & 2.3104 + & 11.72048 & 11.73128 & 32.8808 + 7 & & 11.75121 & 16.7278 + & & 11.78850 & 10.0092 + & & 11.84856 & 5.6249 + & & 11.92188 & 1.9229 + & & 11.95821 & 1.7923 + & & 12.01338 & 2.4821 + & & 13.23441 & 1.3668 + & & 13.25629 & 2.5141 + & & 13.30255 & 1.9971 + & 13.36921 & 13.36913 & 2.8189 + & 13.40177 & 13.40568 & 30.8875 + 8 & 13.44313 & 13.43744 & 16.0829 + & 13.48737 & 13.48146 & 4.8263 + & & 13.54340 & 4.3111 + & & 13.59258 & 1.0747 + & & 13.61133 & 1.9475 + & & 13.65111 & 1.4081 + & & 13.70866 & 2.9676 + & & 14.91192 & 2.1880 + & & 14.94654 & 2.9922 + & & 14.99711 & 1.4563 + 9 & [ 2ex]15.06960 & 15.06470 & 3.2226 + & & 15.07888 & 25.1866 + & & 15.10074 & 8.4317 + [ cols=">,>,>,>",options="header " , ] for the harmonic analysis of the closed - orbit sum we applied the method of @xmath166 function decimated signal diagonalization @xcite , which yields not only semiclassical eigenvalues and amplitudes , but also an error parameter estimating the precision of the eigenvalues . results obtained for @xmath152 and @xmath149 with a signal length of @xmath177 are compiled in table [ xscltab ] . the table contains the quantum eigenvalues of @xmath26 and their dipole matrix elements for levels satisfying @xmath178 . it is obvious at a glance that out of the multitude of spectral lines with intensities varying over many orders of magnitude ( most of which are not contained in the table ) only the strongest lines were detected in the semiclassical spectrum . the semiclassical eigenvalues given are characterized by having small imaginary parts , small error parameters and large amplitudes as well as being stable with respect to a variation of numerical parameters . the calculation operates at the edge of convergence , and in a few cases one can be in doubt whether a level should be included according to these fairly `` soft '' criteria , but in general a clear decision can be made . semiclassical values for the transition strengths are not given because they are not reasonably well converged and depend strongly on the numerical parameters . one might expect that in each @xmath1-manifold it is the strongest lines that are detected semiclassically , and in general this expectation is confirmed by the numerical data . this can clearly be seen , e.g. , in the manifold @xmath179 , which contains the most stably converged lines in the spectrum . there are , however , a few conspicuous exceptions , e.g. at @xmath180 , where strong lines are missing whereas comparatively weak lines are found . for @xmath181 , no lines at all can be computed from the given semiclassical signal . if the signal length is decreased to @xmath182 , the three strongest lines appear in the spectrum in this manifold . at higher @xmath1 , the number of strong lines in the quantum spectrum increases . so does the number of lines in the semiclassical spectrum until @xmath183 , where only three semiclassical lines are found . they appear rather arbitrarily scattered across the quantum spectrum , and their convergence is notably worse than in lower manifolds . it is clear that in this @xmath1-shell the semiclassical quantization with the given signal is about to break down . at @xmath184 , no lines can be detected semiclassically . as , from the above discussion , this failure was to be expected because the required signal length becomes too large , the obvious way to improve convergence seems to be to use a longer signal . however , if the signal length is increased to @xmath185 , no reasonably converged semiclassical lines can be found in any @xmath1-manifold . neither are results improved if the technique of harmonic inversion of cross - correlated closed - orbit sums @xcite is applied . this method has proven powerful in reducing the signal length required in a semiclassical quantization . in the present case , however , because the cross - correlation increases the total number of frequencies obtained from the harmonic inversion , the true eigenvalues are hidden among a multitude of spurious frequencies , and no useful results can be obtained . for the time being , therefore , the results given in table [ xscltab ] represent what can be achieved in the semiclassical quantization of the crossed - fields hydrogen atom . they confirm the applicability of the closed - orbit theory approach in principle , but they also reveal a fundamental problem in its present formulation . from the analysis of the ideal test signal it is clear that the signal length available is sufficient for a stable signal analysis . thus , if the semiclassical results are not good , the semiclassical signal itself , rather than the signal analysis , must be to blame . this conclusion is confirmed by the observation that an increased signal length destroys the results rather than improves them . we therefore start searching for a flaw in the construction of the semiclassical photo - absorption spectrum . a conspicuous problem lies in the fact that the set of closed orbits available is incomplete . in no series of rotators or vibrators can arbitrarily long orbits be calculated . in the case of vanishing electric field there is a critical angle @xmath186 which the starting angles of both rotators and vibrators approach as the orbits get longer . this convergence indicates that the orbits approach a separatrix between two families of tori in phase space . if sufficiently long orbits are studied , there are many closed orbits with very similar initial conditions , so that the numerical search for closed orbits must eventually fail . , @xmath149 . ] the region of phase space where the unknown orbits are located is lying close to a separatrix , so that it is highly unstable . the orbits can therefore be expected not to contribute much to the semiclassical signal . the magnitude of an orbit s contribution to the closed - orbit sum ( [ gcrossed ] ) is determined mainly by its stability determinant @xmath80 . figure [ stabmfig ] shows the stability determinants of the vibrator orbits for @xmath152 , @xmath149 as a function of the scaled action . different series of vibrators can clearly be discerned in the plot . it is indeed unstable orbits with large @xmath187 that are missing in the data set , but on the other hand the stability determinants of the missing orbits are not large enough to regard the corresponding semiclassical amplitudes as negligibly small . because a vast majority of orbits has small @xmath187 and was found , one can still hope that useful results can be obtained from the semiclassical signal , at least for quantum states not located in the separatrix region in phase space , but it is clear that the quality of the semiclassical signal is reduced by its incompleteness . to assess in detail the detrimental effect of the missing orbits and of any other sources of error that may exist , we carry out a semiclassical analysis of the quantum spectrum . according to equation ( [ goscscal ] ) , in a scaled photo - absorption spectrum every closed orbit contributes a purely sinusoidal modulation to @xmath188 . this contribution can be extracted from the spectrum either by a conventional fourier transform or by means of a high - resolution method . the spectral analysis yields information about classical orbits returning to the nucleus . for this reason , the transformed spectrum is referred to as a recurrence spectrum . high - resolution methods @xcite extract the scaled actions and scaled semiclassical amplitudes of individual orbits and thus yield more complete information about the semiclassical spectrum than the fourier transform , but they fail if the average density of closed orbits per unit of scaled action is too large . by contrast , due to its linearity the fourier transform can be applied to any part of the recurrence spectrum with equal ease and numerical stability , irrespective of the spectral density . in dense regions , it will not be able to identify individual closed orbits , but it will nevertheless yield a recurrence spectrum that can be compared to the classical data . in this section we will present results obtained by both the fourier transform and a high - resolution method . the semiclassical recurrence spectra will be compared to classical results in order to identify the reason why the semiclassical signal is only partially suitable to a semiclassical quantization . using either method , it is essential to note that the semiclassical closed - orbit formula can not be expected to yield accurate results for the lowest levels . thus , the low @xmath1-manifolds must be excluded from the semiclassical analysis , i.e. the analysis is based on the quantum spectrum given in an interval @xmath189 $ ] instead of @xmath190 $ ] . furthermore , to minimize the impact of boundary effects due to the finite length of the semiclassical spectrum , a smooth gaussian cut - off with width @xmath70 centered at @xmath191 is introduced . the smoothing replaces the peaks of the semiclassical recurrence spectrum by gaussians of width @xmath192 . the recurrence spectra presented here were calculated from the quantum spectrum shown in figure [ quantumfig ] , for @xmath152 and @xmath149 , with @xmath193 , @xmath194 , and @xmath195 . for the high - resolution recurrence spectra , the method of @xmath166 function decimated signal diagonalization was used . and @xmath149 . sticks and squares : semiclassical closed - orbit amplitudes , stars : harmonic inversion of the quantum spectrum.,title="fig : " ] and @xmath149 . sticks and squares : semiclassical closed - orbit amplitudes , stars : harmonic inversion of the quantum spectrum.,title="fig : " ] for low scaled actions , where only few closed orbits exist , the high - resolution analysis can be applied . results are shown in figure [ recurreimfig ] , which compares both the scaled actions and the real and imaginary parts of the semiclassical amplitudes extracted from the quantum spectrum to the classical results . for most closed orbits , the agreement is excellent . exceptions occur for the shortest orbits , where the actions of rotator and vibrator orbits are too similar to be resolved by the harmonic inversion . at somewhat larger actions , the three orbits in each group fall apart into two rotator orbits with similar actions and a vibrator orbit with a slightly larger action . these observations can be made even more clearly if the absolute values of the amplitudes are considered . they are shown in figure [ recurabsfig ] , where the results of the high - resolution analysis are also compared to those of the fourier transform . notice that for the fourier transform the semiclassical amplitude is given by the area under a peak rather than the peak height , so that an immediate comparison to the high - resolution results is difficult . in figure [ recurabsfig ] , the fourier transform was arbitrarily scaled so that the peak heights roughly match the values of the high - resolution amplitudes . for isolated orbits identified both in the fourier transform and the high - resolution spectrum , the agreement between the two methods is excellent . where several peaks overlap in the semiclassical spectrum , no direct comparison is possible because the peak phases can not be determined from the figure . figure [ recurabsfig ] also extends the results shown in figure [ recurreimfig ] to higher actions . in this region the density of closed orbits starts to increase because , on the one hand , rotators of the first series exist and , on the other , bifurcations of closed orbits generate additional orbits . apart from the fact that many orbits can not be identified individually even by the high - resolution method , the most conspicuous feature of figure [ recurabsfig ] is that for many orbits the semiclassical amplitudes calculated from the classical data are considerably larger than those extracted from the quantum spectrum . in some cases , this is obvious at a glance , but a closer inspection of the figure reveals that this phenomenon is rather common . some specific cases will be described in detail in section [ sec : uniform ] . the occurrence of exceedingly large semiclassical amplitudes is a well - known problem of both closed - orbit and periodic - orbit theory . it is associated with bifurcations of classical orbits , where , in the case of closed orbits , the stability determinant @xmath80 vanishes and the closed - orbit amplitude ( [ acrossed ] ) diverges . close to the bifurcation , @xmath80 is small . the semiclassical amplitude of the bifurcating orbit is therefore large and exceeds the value determined from the quantum spectrum . in a classical context , we have shown previously @xcite that vanishing @xmath80 is both a necessary and sufficient condition for a bifurcation of closed orbits . in the context of semiclassical closed - orbit theory , it is necessary to overcome the divergence of the closed - orbit formula occurring close to a bifurcation . this problem will be addressed in section [ sec : uniform ] , after the impact of the bifurcations on the semiclassical signal at hand has been investigated further . whereas , in figure [ recurabsfig ] , the vibrator orbits are sufficiently isolated to be resolved by both the harmonic inversion and the fourier transform across the entire range of actions , the rotators occur in groups of several orbits having nearly identical actions . they are not resolved properly by either method . instead , the fourier transform produces peaks describing the collective contribution of the orbits in a group . the harmonic inversion fits this contribution with fewer actions and amplitudes than the actual number of orbits . although the results can be expected to reproduce the quantum spectrum fairly well , the principal virtue of the high - resolution analysis that it is capable of giving individual rather than collective contributions is lost . it is therefore pointless to extend the high - resolution analysis to higher actions unless a significantly longer quantum spectrum can be obtained , and only the fourier transform will be used henceforth . of the recurrence spectrum with @xmath195 ( see text ) . upper part : fourier transform of the quantum spectrum , lower part ( inverted ) : smoothed semiclassical recurrence spectrum . ] figure [ recurlongfig ] displays the fourier recurrence spectrum with smoothing @xmath195 for scaled actions up to @xmath196 and compares it to the semiclassical spectrum . these results extend the semiclassical analysis of quantum spectra to significantly longer orbits than investigated in previous studies . they allow a verification of closed - orbit theory all the way up to the long orbits . it is immediately apparent from the figure that the quantum recurrence spectrum retains its pronounced peak structure . this is to be expected from closed - orbit theory , and indeed the peak locations are given by the actions of closed orbits for long as well as for short orbits . the basic idea of closed - orbit theory that recurrence peaks are related to classical closed orbits is therefore confirmed in principle even for very long orbits . even for the largest actions considered , the quantum and semiclassical recurrence spectra agree quantitatively for some peaks . for most peaks , however , the peak heights in the quantum and semiclassical spectra disagree . there are quantum peaks that are smaller in the semiclassical spectrum or even completely absent . they can be attributed to missing orbits . on the other hand , in many cases the semiclassical peaks are significantly higher than the quantum peaks , sometimes by several orders of magnitude . exceedingly high peaks can be traced back to bifurcations of closed orbits if the possibility is ignored that a quantum peak can be small because orbits missing in the semiclassical spectrum interfere destructively with the orbits present . this latter mechanism becomes the more implausible the larger the semiclassical peak is in comparison to the quantum peak . taken together , the effects of missing orbits and of bifurcating orbits distort the semiclassical recurrence spectrum to the point where it can no longer be expected to provide a suitable basis for a quantization . a close inspection of the recurrence spectrum suggests that the problem posed by bifurcating orbits is more severe . exceedingly high peaks do not only occur frequently , but in addition the very fact that they are high increases their detrimental effect on the semiclassical photo - absorption spectrum . unless a suitable scheme for dealing with bifurcating orbits can be devised , no improvement of the semiclassical signal can be expected . we therefore turn to a description of the semiclassical treatment of bifurcating orbits by means of uniform approximations . exceedingly large contributions of single orbits to a semiclassical spectrum arise when the orbits are too close to a bifurcation to be regarded as isolated , as is implicitly assumed by the stationary - phase approximation used in the derivation of the closed - orbit formula . uniform approximations furnish a collective contribution of all orbits involved in a bifurcation . this solution was first suggested by ozorio de almeida and hannay @xcite in the context of periodic - orbit theory . their original approach was extended by different authors @xcite , so that today uniform approximations are a well - established tool of semiclassical physics . in reference @xcite , we identified two types of generic closed - orbit bifurcations of codimension one . the pertinent uniform semiclassical approximations will be derived in what follows . in most cases of interest , a bifurcation destroys real orbits and turns them into complex ghost orbits that exist in the complexified classical phase space . ghost orbits can yield palpable contributions to semiclassical spectra @xcite . in particular , their knowledge is essential for the construction of uniform approximation . for the generic closed - orbit bifurcations , the ghost orbits were described along with the real orbits in reference @xcite . of particular importance is the observation that bifurcations of codimension higher than one are relevant to semiclassics , although on a classical level they are not generically encountered . they appear as sequences of generic bifurcations , which , if the individual bifurcations are sufficiently close , must be described collectively by a single uniform approximation . several examples of uniform approximations for these complicated bifurcation scenarios have been described in the literature @xcite . the principal requirement a uniform approximation must satisfy is to asymptotically reproduce the known isolated - orbits approximation when the distance from the bifurcation grows large , because in this limit the stationary - phase approximation can be expected to be accurate . in the following , we will advocate a somewhat heuristic technique for the construction of a uniform approximation , which is easy to handle and yields a smooth interpolation between the asymptotic isolated - orbits approximations on either side of the bifurcation . it will first be described in general terms . subsequently , uniform approximations describing the generic types of codimension - one bifurcations of closed orbits will be derived . a bifurcation scenario is described by a normal form @xmath197 depending on @xmath198 variables @xmath119 and @xmath199 parameters @xmath200 such that for any fixed value of the parameters @xmath200 there are stationary points of @xmath197 corresponding to the closed orbits involved in the bifurcation . the parameters @xmath200 must then depend on the energy @xmath12 to reproduce the bifurcations of the closed orbits . for the uniform approximation we make the ansatz @xmath201 with @xmath202 here , the functions @xmath203 and @xmath204 as well as the parameter values @xmath205 have to be determined . all of them must be smooth functions of @xmath12 . to find the asymptotic behavior of the uniform approximation ( [ unifansatz ] ) far from the bifurcations , ( [ unifi ] ) is evaluated in the stationary - phase approximation , which yields @xmath206 where the sum extends over all stationary points @xmath207 of @xmath197 that are real at the given @xmath200 , @xmath208 is the hessian determinant of @xmath209 , and @xmath210 is the number of negative eigenvalues of @xmath211 . this expression is supposed to reproduce the isolated - orbits approximation @xmath212 in this case , the sum extends over all closed orbits involved in the bifurcation that are real at the given energy @xmath12 . if the normal form @xmath197 has been chosen suitably , there is a one - to - one correspondence between these orbits and the stationary points @xmath207 . a comparison of ( [ unifsp ] ) to ( [ unifisol ] ) yields the conditions @xmath213 and @xmath214 these equations must be valid for real orbits . in most bifurcation scenarios , all orbits are real at least at certain energies . in these cases , it appears natural to postulate ( [ unifamp ] ) also to hold for ghost orbits . the parameter values one obtains are then smooth functions of the energy even at the bifurcations where the orbits become ghosts . in some instances , bifurcations involving only ghost orbits occur @xcite . in these cases , the condition ( [ unifamp ] ) still produces smoothly varying parameters and enforces the desired asymptotics . the numbers @xmath210 of negative eigenvalues change discontinuously at a bifurcation . for orbits which are real on either side of the bifurcation , so do the maslov indices contained in the semiclassical amplitudes @xmath215 . these changes must compensate each other if the values @xmath216 are to be continuous across the bifurcation . for these orbits , therefore , the change of maslov index occurring in a bifurcation must be equal to the change in @xmath210 and can be determined from the normal form . for ghost orbits , maslov indices are not well defined classically . they must be chosen such as to make @xmath216 continuous . the normal form parameters @xmath200 and the action @xmath203 can be determined from ( [ unifaction ] ) . they usually turn out to be unique . the amplitude function @xmath204 , on the contrary , is unknown . once the parameters @xmath200 have been found , ( [ unifamp ] ) specifies its values @xmath216 at the stationary points of @xmath197 . these values , of course , do not suffice to identify @xmath204 uniquely , so that there is considerable freedom in the choice of @xmath204 . usually , if there are @xmath33 orbits participating in the bifurcation scenario , we will approximate @xmath204 by a polynomial of degree @xmath217 . this choice is justified by the observation that the uniform approximation is needed only close to a bifurcation , where all orbits are close to @xmath218 . thus , in the spirit of the stationary - phase approximation , the dominant contributions to the integral ( [ unifi ] ) stem from the neighborhood of @xmath218 , whereas the regions of large @xmath119 do not contribute . a suitable approximation to @xmath204 must therefore be precise close to the origin . this is achieved by a taylor series expansion , which leads to the polynomial ansatz . simple as it might appear , however , this choice can bring about a mathematical difficulty : a polynomial @xmath204 diverges as @xmath219 , so that there is no guarantee that the integral ( [ unifi ] ) will converge . if it does not , its divergence is an artefact of the choice of @xmath204 , because by construction the regions of large @xmath119 should not significantly influence the value of the integral . in this case , a suitable regularization scheme must be applied . it can be justified by verifying that the regularized integral possesses the correct asymptotics . a slightly simpler form of the uniform approximation is obtained if the function @xmath204 is assumed to be a constant . this approximation does not exactly reproduce the desired asymptotics , but as the transition across the bifurcation mainly results in a change of the stationary points of @xmath197 rather than essential changes in @xmath204 , it can be expected to capture the principal features . it is clear from the above description that there is a certain arbitrariness in the procedure . this arbitrariness can be reduced to the choice of a suitable amplitude function @xmath204 , because by the splitting lemma and the classification theorems of catastrophe theory @xcite the uniform approximation can always be brought into the form ( [ unifansatz ] ) by a suitable coordinate transformation , provided a normal form @xmath197 equivalent to the actual action function is given . in the following sections , uniform approximations for the two generic codimension - one bifurcations described in @xcite will be derived along the lines given here . they turn out to be analogous to those for isochronous and period - doubling bifurcations of periodic orbits given by schomerus and sieber @xcite . the simplest closed - orbit bifurcation is the creation of two orbits in a tangent bifurcation . it is described by the fold catastrophe @xmath220 this normal form has stationary points at @xmath221 , which are real if @xmath222 . its stationary values are ( [ foldval ] ) @xmath223 by ( [ unifaction ] ) , the actions @xmath224 and @xmath225 of the bifurcating orbits must satisfy @xmath226 for these equations to hold , one must assume @xmath227 if the orbits are real and @xmath228 , @xmath229 if they are ghosts . these conditions determine how the orbits are to be associated with the stationary points of @xmath197 . equation ( [ foldaction ] ) can be solved for @xmath230 and @xmath231 the observation that the bifurcating orbits are real if @xmath222 and ghosts if @xmath232 fixes the sign of @xmath200 . both @xmath203 and @xmath200 have thus be determined . for the semiclassical amplitudes , ( [ unifamp ] ) yields @xmath233 with the ansatz @xmath234 for the amplitude function @xmath204 , we can solve for the parameters @xmath235 and @xmath236 to obtain @xmath237 the uniform approximation thus takes the form @xmath238 with @xmath239 the integral @xmath240 can be evaluated in terms of the airy function @xcite as @xmath241 whereas @xmath242 is given by its derivative @xmath243 with these results , the uniform approximation ( [ uniffold ] ) can be computed once the classical quantities @xmath244 and @xmath245 are known . after some rearrangements , ( [ uniffold ] ) can be found to agree with the uniform approximation derived by schomerus and sieber @xcite for isochronous bifurcations of periodic orbits , although its present form is much simpler . the normal form for the symmetrized cusp catastrophe is given by @xmath246 it has stationary points at @xmath218 and @xmath221 and describes a pitchfork bifurcation , where two asymmetric orbits bifurcate off an orbit invariant under a reflection . we denote their actions and amplitudes by @xmath247 and @xmath248 , respectively , where @xmath249 is understood to be the cumulative amplitude of both asymmetric orbits . as @xmath250 , the reference action @xmath203 must be chosen equal to the action of the symmetric orbit . the action difference is given by the stationary value of @xmath197 , which is @xmath251 , so that @xmath252 and @xmath253 the parameter @xmath200 has to be chosen positive if the asymmetric orbits are real , and negative otherwise . here , @xmath254 was assumed to be positive . if it is not , the normal form @xmath197 must be replaced with @xmath255 , which changes the sign of the stationary values . due to the reflection symmetry , the amplitude function must be an even function of @xmath119 . we make the ansatz @xmath256 and solve ( [ unifamp ] ) for the coefficients @xmath257 the complete uniform approximation reads @xmath258 with @xmath259 the integral @xmath240 can be evaluated analytically in terms of bessel functions @xcite : @xmath260 \;. \end{split}\ ] ] although it is not apparent at first sight , @xmath240 is a smooth function of @xmath200 . this can be verified if the series expansion @xcite @xmath261 with @xmath262 a power series in @xmath263 is used . in terms of @xmath262 , @xmath264 \ ; , \end{split}\ ] ] which is indeed smooth . the second integral @xmath265 can be evaluated from @xmath266 \\ & \phantom{{i}\pi\sqrt{|a|}{e}^{-{i}\frac{a^2}8}\bigg\{}+ \frac a8 \,{e}^{{i}\pi/8 } \left[j_{-5/4}\left(\frac{a^2}8\right ) - j_{3/4}\left(\frac{a^2}8\right ) \right ] \\ & \phantom{{i}\pi\sqrt{|a|}{e}^{-{i}\frac{a^2}8}\bigg\ { } + \operatorname{sign } a \,\frac a8 \,{e}^{-{i}\pi/8 } \left[j_{-3/4}\left(\frac{a^2}8\right ) - j_{5/4}\left(\frac{a^2}8\right ) \right ] \bigg\ } \;. \end{split}\ ] ] this derivation contains an interchange of differentiation and integration which achieves a regularization of the divergent integral @xmath265 . it can be justified by verifying that the asymptotic behavior of ( [ i2cusp ] ) for @xmath267 agrees with the stationary phase approximation to ( [ ikcusp ] ) . the formulae derived in the preceding sections give the uniform approximations directly in terms of the semiclassical actions and amplitudes . this circumstance makes them easy to apply to scaled spectra : we simply put @xmath25 and @xmath268 . as @xmath26 is varied , the bifurcation is not encountered because the classical mechanics does not change , so that the isolated - orbits approximation does not actually diverge . however , if @xmath26 is small , the action differences between the bifurcating orbits are also small , so that the presence of the bifurcation is felt and the isolated - orbits formula produces exceedingly large contributions . for large @xmath26 , the action differences also grow large , so that the isolated - orbits approximation should be recovered in the limit of large @xmath26 . ) for the scaled spectrum at @xmath152 and @xmath269 . ] these findings are illustrated in figure [ unifscalfig ] for a pitchfork bifurcation taking place in the first series of rotators at a repetition number @xmath270 . at @xmath152 , the bifurcation takes place at the scaled electric field strength @xmath271 . the data shown in figure [ unifscalfig ] was calculated for @xmath152 and @xmath272 , which is sufficiently far away from the bifurcation for the asymptotic regieme to be reached within the range of @xmath26 shown . as anticipated , in the limit of @xmath273 the complete uniform approximation agrees with the isolated - orbits formula . the simple approximation also reproduces the beats correctly , but it has a smaller amplitude . , @xmath274 and @xmath149 . ] the scaled uniform approximation can be used to improve the semiclassical recurrence spectrum , but this requires some effort : whereas the isolated - orbits approximation yields @xmath166 function peaks in the recurrence spectrum , which are replaced with gaussians due to the smoothing of the recurrence spectrum ( see section [ sec : sclrec ] ) , the uniform approximation is a complicated function of @xmath26 . it must be subjected to a numerical fourier transform in the same was as the quantum spectrum if its contribution to the recurrence spectrum is to be evaluated . because a bifurcation involves orbits with roughly equal actions , the uniform approximation will produce a recurrence peak at the appropriate action . an example is shown in figure [ unifrecfig ] . it was calculated for the bifurcation already described in figure [ unifscalfig ] . the gaussian smoothing used in section [ sec : sclrec ] was replaced with a rectangular window , so that a number of side peaks appear . in this case , the fourier transform of both the uniform approximation and the isolated - orbits approximation was taken over the rectangular window @xmath275 $ ] . the bifurcating orbits have the scaled action @xmath276 , which is where the fourier peaks are centered in both approximations . the peak produced by the uniform approximation is considerably smaller . if this uniformization procedure is carried out for all excessively high bifurcation peaks , it should be possible to bring the semiclassical recurrence spectrum in figure [ recurlongfig ] into agreement with its quantum counterpart . in practice , however , several obstacles stand in the way . first of all , in many cases ghost orbits must be included in the uniform approximation . they must be found and identified as pertinent to a given bifurcation before the uniformization can be performed . furthermore , even if all relevant orbits are real , those orbits connected with each other in a bifurcation must be recognized in the data set . this is by no means an easy task . for example , if in a given series of rotators and for a given winding number a quartet of orbits appears , there are two different doublet orbits out of which they may have bifurcated , and it is not clear in general which of them must be taken for the uniform approximation . in a single case , this can be found out fairly comfortably by hand . if many orbits are to be classified , however , it is essential to do the grouping automatically . we have not yet been able to devise a practical algorithm for this task , so that an automatized uniformization of all bifurcation peaks is presently impossible . apart from these rather technical difficulties , there are also some obstacles of more fundamental importance . consider , e.g. , the two high semiclassical peaks at @xmath277 in figure [ recurabsfig ] . they are notably too high , and they are well - isolated from neighboring recurrence peaks , so that they may appear to be the ideal testing ground for the uniformization procedure . these peaks are generated by vibrators with repetition numbers @xmath278 and @xmath279 , respectively . the pertinent bifurcation scenarios were described in figures 17 and 18 of @xcite . the `` simple '' scenario taking place at @xmath278 consists of two orbits being generated in the rotational symmetry - breaking at @xmath280 , followed by a tangent bifurcation destroying one of them and a third orbit . to smooth this bifurcation peak , a uniform approximation describing the complete scenario must be found , which requires the construction of a suitable normal form . although a uniform approximation for the symmetry - breaking is available @xcite , the derivation of the pertinent normal form relies on principles different from the catastrophe theory classification used here , and it is not clear how these two can be united into a single normal form . thus , the construction of a uniform approximation for this bifurcation scenario , and even more so for the more complicated scenario at @xmath279 , remains an open problem to be solved in the future . it can be solved within the framework of uniformization presented in section [ ssec : unifconst ] , but will require a novel way of constructing normal forms . the approach to high - resolution semiclassical quantization relies on the harmonic inversion of a fourier transformed semiclassical spectrum , i.e. of a recurrence spectrum . the above method of uniformizing the bifurcation - induced excessively high recurrence peaks in a semiclassical spectrum would therefore , if it could be implemented systematically , also pave the way for the inclusion of uniform approximations into a high - resolution semiclassical quantization , which has not been possible so far . we were able to demonstrate the feasibility of our method by way of example for the hydrogen atom in an electric field @xcite , which is less demanding classically . its application to the crossed - fields hydrogen atom , however , remains open for future work . for the first time , a high - resolution semiclassical quantization of the hydrogen atom in crossed electric and magnetic fields has been presented . it achieved the identification of the strong spectral lines in different @xmath1-manifolds . by means of a detailed semiclassical analysis of the pertinent quantum spectrum , it was shown that bifurcations of closed orbits play a crucial role in the semiclassical spectrum and preclude the resolution of finer details in the semiclassical spectrum . they pose a particular challenge to the semiclassical quantization because they require a special treatment by uniform approximations . a simple heuristic scheme for the construction of uniform approximations has been proposed . its simplicity and efficacy was demonstrated by a derivation of the uniform approximations for the codimension - one generic bifurcations of closed orbits . we have devised a general method for the inclusion of uniform approximations in a high - resolution semiclassical quantization by harmonic inversion . in a recent publication @xcite it was successfully applied to the hydrogen atom in an electric field . in the case of the crossed - fields hydrogen atom , the diversity and complexity of the bifurcation scenarios encountered so far hinders the systematic implementation of the uniformization procedure . the treatment of all relevant bifurcations and the calculation of a detailed semiclassical spectrum thus remain challenging tasks for future studies .

an accurate description of electron correlation is one of the central issues in modern electronic structure theory . in principle , this involves solving the many - electron schrdinger equation , which for a system of @xmath0 interacting electrons is an inseparable @xmath1-dimensional problem . methods based on mean - field approximations of the electron interaction , such as hartree - fock ( hf ) or density - functional theory , @xcite ( dft ) reduce this to @xmath0 independent problems , which we may solve to approximate many physical properties with reasonable accuracy . however , a more complete description of electron correlation is often required to study delicate phenomena such as long - range electron correlation and van der waals interactions . quantum monte carlo ( qmc ) methods provide an important approach for solving the full @xmath1 dimensional problem . @xcite one such method , variational monte carlo ( vmc ) , allows us to estimate expectation values for a given trial wave function . ideally , this wave function is an eigenstate of the many - body hamiltonian . in practice , a parameterized form is used , which approximates the exact eigenstate . the accuracy of these calculations is entirely dependent on the trial wave function , however , and so the development of accurate wave functions is vital both in the accurate estimation of physical properties and in understanding how certain physical phenomena may be simply represented in the wave function . a widely used trial correlated wave function is the jastrow - slater @xcite or feenberg @xcite form , @xmath2 d~. \label{eq : jastrowpsi}\end{aligned}\ ] ] here @xmath3 is a slater determinant of single - particle orbitals and interparticle correlation is introduced with the two - body term @xmath4 in the jastrow factor . the one - body term @xmath5 could in principle be absorbed into the single - particle orbitals of the determinant , but it may be convenient to retain it explicitly in the jastrow factor . in practice , the orbitals in the slater determinant are often determined from a hf or dft calculation . @xcite in this paper , we will focus on the optimization of jastrow - slater wave functions in the context of the electronic structure of periodic solids . we will apply wave function optimization to examine some consequences which emerge from a complete treatment of inhomogeneity in the two - body correlation term for diamond and rhombohedral graphite . the general form of wave function in eq . [ eq : jastrowpsi ] has been used as the starting point of several methods , including fermi hypernetted chain @xcite ( fhnc ) and vmc @xcite calculations . the traditional approach is to use a variational principle on the energy ( or , in the `` variance minimization '' method , @xcite on the fluctuations of the energy ) to define a best approximation to the true eigenstate within the variational freedom allowed by the wave function ansatz . expectation values of the energy and other quantities are calculated from the trial wave function , approximately in the fhnc approach , exactly in vmc ( within statistical error of the sampling ) . while the vmc method has been used in calculations of periodic solids , in practice most calculations have included only homogeneous two - body terms in the jastrow factor @xcite and the optimization of wavefunctions with very large numbers of parameters has remained problematical . to our knowledge , the fhnc approach has not been applied in fully three - dimensional electronic structure calculations . wave functions determined by the vmc approach are often used as guiding , trial functions for the diffusion monte carlo ( dmc ) method . @xcite in the dmc context , the accuracy of the wavefunction affects the numerical efficiency of energy calculations and the accuracy of other physical quantities . when dmc is used in conjunction with non - local pseudopotentials , @xcite as is commonly the case in chemical applications , an accurate trial wavefunction is essential for an accurate calculation of the ground state energy . we present here , for the first time , a numerically robust , rapidly convergent iterative method which minimizes the variational energy with respect to a very general inhomogeneous form of the jastrow factor , including all one- and two - body terms compatible with crystal symmetry . the method remains numerically well - conditioned , even for large systems ( of the order of several hundred electrons ) , where there are more than 3000 independent variational parameters in the jastrow factor . within acceptable computational demands on the monte carlo sampling , the optimal values of parameters in the wave function are found to be well determined , even when their contribution to the total energy is extremely small . the method places no restrictions on the functional form of the anti - symmetric ( determinantal ) part of the many - body wave function and can be used in conjunction with related methods recently developed for energy minimization with respect to all orbitals in the determinant @xcite or with respect to configuration weights in a multi - determinant function . @xcite taken in conjunction with these methods , the approach presented here completes the solution to the problem of energy minimization with respect to the most general variational terms in wavefunctions of the jastrow - slater type , as well as in wave functions where the jastrow factor multiplies a multi - determinant function . although many specific details of the work here refer to periodic systems , the basic method for jastrow factor optimization could also be used in similar calculations of molecular , atomic , or nuclear structure . other general methods exist to achieve energy minimization with respect to parameters in many - body wave functions . when only one or two parameters are optimized , it is possible to perform a systematic search of parameter space , as mcmillan did in his pioneering vmc study of the properties of liquid he . @xcite the stochastic gradient approximation ( sga ) used by harju _ et al . _ @xcite applies control theory to determine iterative corrections to the wave function parameters . et al . _ @xcite explicitly computed analytical derivatives of the energy with respect to variational parameters and used these to optimize their wave function with a newton - style method . however , none of these methods have as yet been applied in systems where a very large number of parameters are to be optimized . in recent years , the variance minimization method of umrigar _ et al . _ , @xcite which optimizes the wave function by reducing the magnitude of the variance of the local energy , has been used much more widely than energy minimization in determining optimal parameters for variational wave functions . although not strictly equivalent to energy minimization for a non - exact wave function , in practice this method has been very successful in improving total energies , providing extremely accurate wave functions in certain atomic systems . @xcite unfortunately , variance minimization is often subject to confinement to local minima and requires much human interaction and experience for successful implementation when large numbers of parameters are to be optimized . variance minimization may be thought of as fitting energy fluctuations to a constant , and attempting to reduce the cost function involved in this fit by direct variation of the wave function parameters . the standard approach is to use a least - squares fit of these energy fluctuations . our method of energy minimization involves the fitting of energy fluctuations to a given ( non - constant ) functional form , which is a linear combination of operators associated with variations of the wave function parameters ( see sec . [ sec.enermin ] ) . this fitting allows us to determine a `` predictor '' , which links changes in the wave function to changes in the fitted energy fluctuations . this predictor iteratively guides our method to a self - consistent solution , where the fitted energy fluctuations are zero . the derivatives of the true many - body energy with respect to all the parameters in the wave function are then also zero ( within statistical sampling error ) for this final solution . our predictor is closely related to the random phase approximation ( rpa ) , introduced by bohm and pines @xcite and recently discussed in the context of inhomogeneous systems . @xcite however , as long as the predictor is sufficiently accurate to ensure stable convergence of the iterations to the self - consistent solution , its exact form does not affect the final solution . our method allows us to surpass the approximations of the rpa and produce explicit trial wave functions of unprecedented accuracy for electrons in periodic solids . we will use the optimized wave functions to study the effects of charge inhomogeneity on the correlation factors in diamond , as a prototype of strongly bonded insulating systems , and rhombohedral graphite , as a prototype of highly anisotropic , inhomogeneous solids . although earlier studies @xcite of these and related systems would indicate that homogeneous two - body correlation factors gain a large fraction of the correlation energy in solids and that inhomogeneous correlation factors are unlikely to lower the total variational energy greatly , substantial differences in correlation factors often give rise to relatively little change in energy . this is particularly true when one is interested in long - range correlation , which is energetically very delicate . the rest of this paper is organized as follows : in sec . [ sec.wavefn ] , we present the detailed numerical form of variational wave functions to be used in these calculations . the approach of fitting energy fluctuations and its use in guiding the iterative solution of the energy minimization problem is discussed in sec . [ sec.enermin ] . in sec . [ sec.results ] , we present some results on the application of the method to the periodic solids , diamond and rhombohedral graphite , and examine the effects of charge density inhomogeneity on the correlation factors of these systems . in sec . [ sec.discussion ] we discuss the results and some computational details of the method , illustrating the justification for certain aspects of our approach with some tests . finally , in sec . [ sec.conclusion ] , we present the overall conclusions of this study . the jastrow two - body correlation factor @xmath6 in eq . [ eq : jastrowpsi ] can in principle always be expressed as : @xmath7 where @xmath8 form a complete set of functions and @xmath9 are expansion coefficients . ( we use @xmath10 to indicate complex conjugation throughout this paper ) . similarly , the one - body function @xmath11 may also be expanded in a basis set of complete functions @xmath12 , as : @xmath13 apart from an additional term to handle the electron - electron cusp , @xcite we will express the jastrow factor in the general form given by eqs . [ eq:2bodyexpansion ] and [ eq:1bodyexpansion ] . values of the parameters @xmath14 will then be determined so that the energy of the system is stationary with respect to all variations @xmath15 . due to the divergence of the coulomb interaction between two electrons , the correct two - body correlation term @xmath6 has a cusp where @xmath16 , leading to slow convergence of any expansion in smooth functions of the form in eq . [ eq:2bodyexpansion ] . @xcite for this reason , it is numerically convenient to re - write the two - body function @xmath4 in the form : @xmath17 where @xmath18 is a short - ranged ( homogeneous ) function which has the correct electron - electron cusp as @xmath19 and @xmath6 is now a smooth , cuspless function . we use a form of the short - ranged function @xmath20 which is generated from a numerical solution of the electron - electron scattering problem . a discussion of the generation of @xmath20 is provided in appendices a and c of ref . . we define @xmath21 , where @xmath22 is defined within appendix c of ref . . the expansion of the remaining function @xmath6 in the form given in eq . [ eq:2bodyexpansion ] then converges much more rapidly than that of the original function , for any set of smooth functions @xmath8 . in this paper , we generate the short - range function @xmath20 in a spin - dependent form , to maintain the cusp conditions . @xcite the cuspless function @xmath6 used in our work is independent of electron spin , although we expect that a spin - dependent form is as easily optimized . summing @xmath6 over all electron pairs in the basis @xmath8 leads to @xmath23 this may be thought of as a two - body expansion of the correlation term . however , it is important to realize that any separation of `` one - body '' and `` two - body '' terms in the jastrow factor is somewhat arbitrary . removal of terms where @xmath24 , as in eq . [ eq:2bodycorrelation ] , is not sufficient to decouple one- and two - body terms completely . to see this , consider the transformation of each of the basis functions obtained by subtracting a constant , @xmath25 . the set remains complete and the function @xmath26 u_{\alpha \beta } [ f_{\beta}({\bf r ' } ) - c_{\beta } ] ~ , \label{eq : arbconstexpansion}\end{aligned}\ ] ] may be interpreted as a `` two - body '' function in the basis set @xmath27 . however , expanding this correlation factor over all electron pairs , we find that , in terms of the original basis @xmath8 , an additional one - body contribution appears : @xmath28 \label{eq:12bodyseparation } \\ & + & ~{\rm constant } ~. \nonumber\end{aligned}\ ] ] we may regard this as a transformation of the one - body function in eq . [ eq : jastrowpsi ] : @xmath29 , where the additional term comes from the second line of eq . [ eq:12bodyseparation ] : @xmath30 f_{\alpha}({\bf r})^ * + \left [ \sum_{\beta } u_{\beta \alpha } c_{\beta}^ * \right ] f_{\alpha}({\bf r } ) \right\ } \nonumber \end{aligned}\ ] ] to make a definite numerical separation of our one- and two - body expansions , we need to consider an appropriate choice of the arbitrary constants @xmath31 , noting that @xmath32 is intended primarily to affect correlation properties ( i.e. , two - body properties ) of the system , leaving single - particle properties unchanged . for example , the mean - field methods that produce @xmath3 [ hf , dft under the local density approximation @xcite ( lda ) , etc . ] normally give very accurate single - particle densities , which can be altered substantially by inclusion of an arbitrary function @xmath6 , in @xmath33 ( eq . [ eq : jastrowpsi ] ) . any substantial change in the density from the hf solution is likely to be energetically very costly , so that ideally we would like to decouple changes in @xmath6 from changes in the density . necessary changes in the density may be allowed for by optimization of the explicit one - body term @xmath5 in the jastrow factor ( eq . [ eq:1bodyexpansion ] ) or by the methods of ref . . therefore , we would like to choose the constants @xmath31 such that the average of any one - body operator ( such as the density ) for @xmath33 remains stationary with respect to variations in the coefficients @xmath9 , at least in the absence of interparticle correlation . if we define the one - body operator @xmath34 for the many - body configuration @xmath35 , then its expectation value is @xmath36 where @xmath37 is the single particle density . since the @xmath38 are fixed functions , variations in the expectation values @xmath39 correspond to variations in @xmath37 . the derivative of @xmath39 with respect to variations of the @xmath9 in eq . [ eq : arbconstexpansion ] is ( see appendix [ app.deriv ] ) @xmath40 \sum_{i \neq j}[f_{\alpha}({\bf r}_i)^ * - c_{\alpha}^ * ] [ f_{\beta}({\bf r}_j ) - c_{\beta } ] \rangle \\ = \sum_k \sum_{i \neq j } \langle [ f_{\gamma}({\bf r}_k)^ * - \langle f_{\gamma } \rangle^ * ] [ f_{\alpha}({\bf r}_i)^ * - c_{\alpha}^ * ] [ f_{\beta}({\bf r}_j ) - c_{\beta } ] \rangle ~ .\end{aligned}\ ] ] in the absence of interparticle correlation , @xmath41 the second summation excludes @xmath24 , so averages of triple products never arise . thus , in the absence of correlation , the derivative becomes @xmath42 [ f_{\alpha}({\bf r}_i)^ * - c_{\alpha}^ * ] \rangle \langle f_{\beta}({\bf r}_j ) - c_{\beta } \rangle \nonumber \\ & + & \langle [ f_{\gamma}({\bf r}_j)^ * - \langle f_{\gamma } \rangle^ * ] [ f_{\beta}({\bf r}_j ) - c_{\beta } ] \rangle \langle f_{\alpha}({\bf r}_i)^ * - c_{\alpha}^ * \rangle ~. \nonumber\end{aligned}\ ] ] we can guarantee that the right hand side of eq . [ eq : approxderiv ] is zero if @xmath43 , for all @xmath44 . in other words , one - body expectation values remain approximately unaffected by the presence of the correlation factor @xmath32 , provided we expand @xmath32 in a basis of `` fluctuation functions '' , @xmath45 . equivalently , we may retain the original basis @xmath8 and form a one - body term @xmath46 from eq . [ eq : natural1body ] with @xmath43 . this may then be inserted into the wave function of eq . [ eq : jastrowpsi ] and varied as the parameters @xmath9 are varied . correlation effects are of course present in the actual wave function used . however , we find that eq . [ eq : approxderiv ] remains approximately true , as previously observed . @xcite for the energy minimization problem , we find that mixed derivatives of the energy @xmath47 are approximately zero when @xmath43 . this gives the numerical advantage that minimization of the energy with respect to @xmath9 approximately decouples from minimization with respect to @xmath48 . we note that this approximate decoupling holds , even if the relation @xmath43 is not exactly true . thus we may use @xmath49 , ( i.e. , lda or hf averages of @xmath8 ) , in place of @xmath50 , while still maintaining the numerical advantages of approximately satisfying @xmath51 . in the context of periodic systems , it is natural to expand the correlation function @xmath32 as a fourier series , where the basis functions are @xmath52 $ ] , for each wave vector @xmath53 . we note that @xmath54 = \sum_i f_{\bf q}({\bf r}_i)^ * \label{eq : fouriercharge}\end{aligned}\ ] ] is the fourier coefficient of the instantaneous charge density , given the electron configuration @xmath55 . ( in atomic units the electronic charge @xmath56 ) . also , summing over pairs leads to a quadratic product of fourier coefficients , with some modification to remove terms with @xmath57 : @xmath58}. \label{eq:1bodyremoved}\end{aligned}\ ] ] in order to approximately remove the effect of the two - body terms on the single - particle density , following sec . [ sec.sep12 ] , we subtract appropriate constants from each basis function to produce a new basis , the collective `` charge fluctuation '' coordinates , @xmath59 these provide a suitable expansion of the two - body correlation factor that approximately preserves single - particle densities . a correlation factor using such coordinates was first suggested by bohm and pines @xcite for the homogeneous electron gas , and has recently been discussed by malatesta _ et al . _ @xcite and gaudoin _ et al . _ @xcite in the context of inhomogeneous systems . in homogeneous systems the expectation value @xmath60 of each charge density fourier coefficient is zero , and so charge density fluctuations are simply @xmath61 . for inhomogeneous systems , in general @xmath62 when @xmath63 , a reciprocal lattice vector . the alternative to using fluctuation coordinates @xmath64 is to incorporate the equivalent one - body term in the jastrow factor , which in periodic systems is of the form @xmath65 with the coefficients coming from the two - body term @xmath66 as discussed by malatesta _ et al . _ @xcite and gaudoin _ et al . _ @xcite the properties of the fourier basis @xmath67 and the correlation factor @xmath4 lead to some convenient symmetry properties for the coefficients @xmath68 . the complex conjugate @xmath69 , just as @xmath70 . the exchange symmetry of u , i.e. , @xmath71 implies that @xmath72 . and , if u possesses inversion symmetry , i.e. @xmath73 , then each @xmath68 is a real number . in periodic systems , @xmath74 for any bravais lattice vector @xmath75 this implies that all fourier coefficients @xmath68 are zero unless @xmath76 , a reciprocal lattice vector . thus , translation symmetry greatly reduces the number of variational parameters in the two - body terms . we arrange the wave function in the form @xmath77 isolating the short range component of the jastrow factor as @xmath78 ~.\end{aligned}\ ] ] the one - body term here is derived from the short range correlation factor @xmath20 of eq . [ eq : shortlongu ] , as @xmath79 , where @xmath80 is the fourier transform of @xmath20 for the reciprocal lattice vector @xmath81 . the prefactor of @xmath82 , which should be present from eq . [ eq:1bodycoeffs ] , approaches unity for large systems and so is neglected . for computational efficiency we leave this one - body term in its fourier space representation and @xmath83 is a cut - off chosen for the fourier sum such that it is converged within a required accuracy . the remaining inhomogeneous part of the two - body jastrow factor is expanded using charge density fluctuation coordinates : @xmath84~,\end{aligned}\ ] ] where @xmath85}$ ] , using the notation @xmath86}$ ] , as defined in eq . [ eq:1bodyremoved ] and the definition of @xmath87 in eq . [ eq : deltarho ] . in practice , this double sum is truncated by using vectors @xmath88 of magnitude less than a suitably chosen cut - off @xmath89 . we also allow for one - body optimization through the use of the explicit one - body jastrow factor . this one - body jastrow factor is also expanded in fluctuation coordinates , @xmath90,\end{aligned}\ ] ] since including the constant average values @xmath91 merely adjusts the normalization of the wave function . the coefficients @xmath92 and @xmath93 , defined in eqs . [ eq : inhomjast ] and [ eq:1bjast ] , are the final variational parameters of our wave function . the remaining sections of this paper describe the method we use to optimize these parameters such that the total energy of a particular electronic system is stationary . a typical calculation presented below involves the simultaneous optimization of over 3000 parameters . we wish to optimize the wave function @xmath94 where @xmath95 a vector of parameters , by solving the euler - lagrange equations , @xmath96 we note that the hamiltonian for the system is @xmath97 where each sum is over the electrons in the system . solving eq . [ eq : eleqns ] is equivalent ( see appendix [ app.deriv ] ) to solving the following system of equations @xmath98 where , given a many - body configuration @xmath99 , we define : @xmath100 , for any operator @xmath101 ; and the local values of the operators @xmath102 as , @xmath103 we shall refer to the local value of the hamiltonian operator as the `` local energy '' , @xmath104 we approach the problem of solving the euler - lagrange equations ( eq . [ eq : eleqns ] ) indirectly , by considering systematic fluctuations of the energy for a given trial wave function @xmath105 . consider fitting the local energy @xmath106 to the functional form @xmath107 in the least - squares sense , where @xmath108 is the set of functions with which we fit the energy , and @xmath109 is the vector of fitting coefficients . the least - squares problem reduces to minimizing the integral @xmath110 which is equivalent ( see appendix [ app.leastsq ] ) to solving the linear system @xmath111 we recognize immediately that if the functions @xmath102 are those functions associated with variations of the wave function parameters @xmath112 ( eq . [ eq : operdefn ] ) , then the right - hand side of eq . [ eq : lstsqsys ] _ is _ the vector of euler - lagrange derivatives in eq . [ eq : elsystem ] . therefore , the euler - lagrange equations ( eq . [ eq : eleqns ] ) are solved if all the fitted coefficients @xmath113 are zero . is singular , or at least numerically so ( i.e. , has a very small eigenvalue ) , singular value decomposition is recommended to set the appropriate fitting coefficients @xmath114 to zero . such problems do not render the wave function parameters @xmath115 ill - conditioned , as the small size of the corresponding expectation values @xmath116 implies that changes in @xmath115 have little effect on the total energy . ] as an illustrative example , consider @xmath117 , an eigenstate of @xmath118 . indeed , the local energy is a constant , independent of @xmath55 , and so we would find that each of the fitted coefficients @xmath113 is zero . therefore , the euler - lagrange derivatives @xmath119 are all zero , and so the energy must be stationary with respect to variations in @xmath117 , as we would expect for an eigenstate . for the trial wave function @xmath105 , no choice of @xmath120 gives an _ exact _ eigenstate of the hamiltonian . however , for a particular choice of parameters , the absence of _ systematic _ variations of the energy ( i.e. , variations correlated with the variations of the functions @xmath102 ) ensures that the fitting coefficients @xmath113 are all zero and that the average energy is stationary with respect to all variations of the parameters @xmath120 . within the parametric freedom of the trial wave function @xmath33 , this is our best approximation to an eigenstate . we now describe a procedure which aims , by appropriate choice of the parameters @xmath120 , to set the fitted coefficients @xmath113 of the total energy @xmath121 to zero . as defined in eq . [ eq : lstsqsys ] , these @xmath113 depend on the wave function @xmath105 and so are functions of the parameters @xmath120 . however , the functional dependence of @xmath113 on @xmath120 is not available in its exact analytic form and we are unable to solve directly the system @xmath122 , i.e. , to find the root @xmath120 which will guarantee the solution to the corresponding euler - lagrange equations . instead , using the wave function @xmath123 , for a particular choice of the parameters @xmath124 , we construct a predictor function @xmath125 , which approximates the unknown function @xmath126 for general values of @xmath120 . more precisely , we construct @xmath125 so that @xmath127 ( i.e. , @xmath128 is exact when @xmath129 ) and , @xmath130 for all relevant values of @xmath120 . to determine this predictor , we use the specific form of the hamiltonian ( eq . [ eq : hamiltonian ] ) and trial wave function @xmath33 ( sec . [ sec.wavefn ] ) , and we partition the local energy @xmath106 into a sum of contributions , @xmath131 where the @xmath132 come from various terms in the kinetic and potential energy ( see below ) . each contribution @xmath133 is approximated with the functional form @xmath134 for some terms , using the specific form of the local energy for @xmath105 , we can expand @xmath132 analytically as in eq . [ eq : enerfuncform ] , enabling us to determine , exactly or approximately , the function @xmath135 . in this analytic form , @xmath135 is independent of the choice of @xmath124 and so remains equally valid for all @xmath120 . where analytic expressions are too complex to derive , we may approximate @xmath135 by fitting @xmath132 to eq . [ eq : enerfuncform ] . the fitting coefficients for @xmath132 , are found by solving the analogue of eq . [ eq : lstsqsys ] , @xmath136 where @xmath137 is determined by using @xmath123 to evaluate the required expectation values . this produces the value of each coefficient at @xmath124 . we may also fit the derivatives of @xmath132 with respect to @xmath120 , to determine the linear dependence of each @xmath135 . we may then approximate the function @xmath138 where the values of the derivatives are found by fitting @xmath139 to eq . [ eq : enerfuncform ] using @xmath123 . in evaluating the term @xmath140 , we consider only the explicit variation of the term @xmath141 with @xmath115 . we do not include the implicit variation due to the dependence of the probability distribution @xmath142 on @xmath143 . in practice , the only term for which we need to fit @xmath139 is explicitly linear in the parameters @xmath143 and so the linear expansion in eq . [ eq : taylorexp ] is valid over a wide range of values of @xmath120 . just as the energy contributions @xmath133 partition the local energy , we may regard the analytic and fitted coefficients @xmath144 as an approximate partition of the local energy coefficients @xmath113 . we define this partition as follows : @xmath145 where the sum of analytically derived coefficients is @xmath146 and the sum of numerically determined coefficients , evaluated using @xmath123 and eq . [ eq : contributionlstsqsys ] , is @xmath147 we construct the predictor @xmath125 such that it satisfies eq . [ eq : predictordefn ] , i.e. , @xmath148 where @xmath149 are found by solving eq . [ eq : lstsqsys ] . we define our iterative approach to determining the parameters @xmath120 , for which the coefficients @xmath126 are zero , as follows : 1 . given the set of parameters @xmath150 , construct the wave function @xmath151 . [ iter : constrwavefn ] 2 . evaluate the required expectation values in eqs . [ eq : lstsqsys ] and [ eq : contributionlstsqsys ] , using @xmath151 , to find the fitting coefficients @xmath152 and numerical functions @xmath153 . [ iter : fitting ] 3 . if _ the total energy fitting coefficients are zero , i.e. , @xmath154 , _ then _ we are done , _ otherwise _ continue . [ iter : donecheck ] 4 . construct the predictor function @xmath155 in eq . [ eq : predictor ] using the fitted terms from step . [ iter : fitting ] , and find the solution @xmath156 to the system @xmath157 using the newton - raphson method @xcite ( see appendix [ app.newtonraphson ] ) . use this set of parameters @xmath156 in step [ iter : constrwavefn ] . [ iter : findroot ] iterations continue until the total energy coefficients @xmath152 tend to zero , and the values of the parameters @xmath150 converge . note that even though the predictor @xmath155 is only an approximation to the exact function @xmath126 , eq . [ eq : predictordefn ] guarantees that at the converged solution @xmath158 in other words , the parameter set @xmath120 solves the euler - lagrange equations for @xmath159 exactly . clearly , the larger the neighbourhood within which the approximate relation in eq . [ eq : predictorapprox ] holds , the faster this iterative procedure will converge . in the trivial case , if the exact analytic form of @xmath126 were known _ a priori _ , then we could just solve the euler - lagrange equations in one step using a suitable root - finding method . rather than starting the procedure from an initial guess @xmath160 , we may begin at step [ iter : findroot ] using only the analytic terms in the predictor , since these are independent of the wave function and so do not require fitting . that is , we solve the system @xmath161 the solution set @xmath162 is then used in step [ iter : constrwavefn ] . we note that the potential energy operators @xmath163 and @xmath164 in the hamiltonian @xmath118 , defined in eq . [ eq : hamiltonian ] , are multiplicative , and therefore their contributions to @xmath106 are constant with respect to variations of the wave function parameters @xmath120 . variations of @xmath120 affect only the contributions of the differential kinetic energy operator . thus , solving the euler - lagrange equations for @xmath159 ( eq . [ eq : eleqns ] ) amounts to adjusting the systematic fluctuations of the kinetic energy to cancel those in the potential energy exactly . if we extract a variational part @xmath165 of @xmath33 , such that @xmath166 , where @xmath167 is dependent on the set of parameters @xmath120 , and @xmath168 is independent of them . then we may partition @xmath106 as follows : @xmath169 where we define @xmath170 clearly , @xmath171 is constant with respect to variations of @xmath120 . further analysis of each of @xmath172 and @xmath173 is necessary to determine how each depends on @xmath120 . however , for @xmath167 expressible in the form of the jastrow factors in sec . [ sec.wavefn ] , i.e. , @xmath174~,\end{aligned}\ ] ] where @xmath175 are defined in eq . [ eq : operdefn ] , we see that : ( i ) @xmath173 is at most linear in @xmath120 , but involves terms coming from @xmath168 , so that it may be impossible to determine analytic expressions for the coefficients @xmath176 ; ( ii ) @xmath172 is at most quadratic in @xmath120 , and involves only @xmath167 , for which we have an analytic expression , and therefore may derive analytic approximations to the coefficients @xmath177 ( see sec . [ sec.predictoranalapprox ] ) . the initial predictor @xmath178 used in the iterative method involves only analytic terms . we determine these by direct expansion of particular local energy contributions , given the analytic form of the wave function . we now consider the contributions of the one- and two - body terms individually . replacing @xmath179 in sec . [ sec.energypartition ] with @xmath180 , we derive an analytic expression for the energy contribution @xmath172 ( see appendix [ app.1benercon ] ) . from this we may extract those coefficients @xmath181 of the functions @xmath182 : @xmath183 by assumption , the one - body contribution of @xmath171 is zero . that is , the mean field methods used to calculate the determinant @xmath3 should remove ( approximately ) all systematic one - body fluctuations in the local energy , and the use of `` fluctuation functions '' @xmath184 in the two - body jastrow factor approximately removes its effect on one - body operators . therefore , we assume initially that the coefficients @xmath185 . we are unable to derive an analytic expression for the coefficients @xmath186 of the energy contribution @xmath173 and so , for the moment , we leave these aside . constructing the full analytic approximation @xmath187 we see that the roots of @xmath188 are trivially @xmath189 , as we would expect . according to sec . [ sec.wavefn ] , the two - body jastrow correlation factor is divided into a short - range term @xmath22 and an inhomogeneous term @xmath190 . we optimize the variational parameters @xmath191 in @xmath190 . as for the one - body jastrow , we expand the corresponding energy contribution @xmath172 which depends on @xmath190 alone ( see appendix [ app.2benercon ] ) . this leads us to an approximation to the coefficients of the functions @xmath192 in @xmath172 : @xmath193 again , we are unable to derive expressions for the coefficients corresponding to @xmath173 . we extract a two - body contribution from the constant contribution @xmath171 ( see appendix [ app.2benercon ] ) , as @xmath194 , for the electron interaction @xmath164 . we replace the true interaction in this expression with a pseudointeraction @xmath195 , which is generated for a given cut - off radius @xmath196 and reference eigenvalue @xmath197 , as explained in ref . . @xmath195 is used to generate the short - range two - body function @xmath20 ( eq . [ eq : shortlongu ] ) , used in @xmath22 . the purpose of this modification of @xmath198 is to account for the presence of @xmath22 in @xmath33 and is explained in appendix [ app.shortjconseq ] . from these contributions we construct the analytic predictor @xmath199 we notice that for periodic systems , this function is separable in the points @xmath53 of the first brillouin zone ( bz ) . for each @xmath53 in bz , we may expand @xmath200 as a function of @xmath201 , with no coupling to parameters @xmath202 for @xmath203 . this block diagonal form of the analytic predictor allows us to find the roots @xmath204 by solving for each block ( i.e. , each @xmath53 ) individually . we do this using the newton - raphson method . a reliable initial guess for the newton - raphson method , rather than using @xmath205 , is the homogeneous solution of @xmath206 . if we regard the system as homogeneous , i.e. , @xmath207 for @xmath208 , then the solution is @xmath209 ~. \label{eq : homosoln}\end{aligned}\ ] ] this @xmath68 is used as the starting point for the newton - raphson iterations , and for all systems studied , this initial guess produced convergent roots of @xmath206 . to complete the construction of the predictor @xmath210 defined in eq . [ eq : predictor ] , for a given set of variational parameters @xmath124 , we must determine some terms numerically by fitting fluctuations in the energy of the system . we calculate @xmath149 in eq . [ eq : predictor ] by fitting the entire local energy , using eq . [ eq : lstsqsys ] . also , we use the fitting method to numerically determine the functions @xmath211 given in eq . [ eq : taylorexp ] , by fitting the energy contribution @xmath173 and each of its derivatives with respect to the parameters to be optimized . again , we note that @xmath173 is explicitly linear in the parameters @xmath120 and the approximation in eq . [ eq : taylorexp ] is exact in this case . to do all this numerical work we use monte carlo sampling to determine the required expectation values : @xmath212 for @xmath213 ranging over the number of parameters @xmath214 in the set @xmath120 . however , for simultaneous optimization of one- and two - body terms , we make some simplifications to reduce the computational workload . if we assume that the parameter @xmath112 can be varied independently of @xmath215 , this corresponds to assuming that @xmath216 . in the expansion of the analytic local energy terms coming from @xmath190 , outlined in appendix [ app.2benercon ] , we saw that these consisted of one- and two - body terms . however , the one - body terms contained fourier coefficients of the average charge density @xmath60 , which are zero for @xmath217 , a reciprocal lattice vector . also , the analytic form of the predictor ( sec . [ sec.predictoranalapprox ] ) is separable in the k - points of the first brillouin zone . therefore , we make the following approximations to the covariance matrix @xmath218 : @xmath219 for reciprocal lattice vectors @xmath220 . we regard one- and two - body optimization as independent for non - zero k - points in the first brillouin zone . if we also exclude the covariance terms between one- and two - body operators for @xmath221 , we find that this slows the convergence of the method for smaller systems . for larger systems , this exclusion prevents the convergence of the newton - raphson method at the first iteration of our method , thus halting the optimization process . however , we are not , in any way , confined to using the newton - raphson method to find the roots of the predictor , and other , more robust , root - finding methods might overcome this problem . note that @xmath173 depends on the variational part @xmath167 of the wave function which is being optimized . the variational components are @xmath222 and @xmath223 for each @xmath53 in bz , where we define @xmath224~,\end{aligned}\ ] ] so that @xmath225 . this greatly reduces the complexity of the predictor , without sacrificing convergence of the method for the systems studied here . ultimately , the predictor @xmath210 is itself only an approximation of the true function @xmath126 , but by definition matches this function at the current values of the parameters being optimized . therefore , approximations in the predictor affect only the rate of convergence of the method . the expressions for @xmath173 and its derivatives @xmath226 for the one - body jastrow factor @xmath222 are given in appendix [ app.1benercon ] . upon fitting these terms to the operators @xmath182 , we may construct the function @xmath211 . this defines the fitted terms @xmath227 in the predictor ( sec . [ sec.iterproc ] ) . for a given @xmath53 in bz , we use the expressions for @xmath173 and the derivatives @xmath228 , given in appendix [ app.2benercon ] , corresponding to the two - body jastrow factor @xmath223 defined above . we construct the function @xmath229 , where @xmath230 , and define the numerical predictor terms @xmath231 , which contribute to the linear dependence of the predictor on @xmath232 . we now apply this optimization method to diamond and rhombohedral graphite . we shall compare the correlation factors determined in both these systems , given the inhomogeneity and anisotropy of the electron charge density in graphite , relative to diamond . the convergence criterion of our iterative optimization : @xmath233 , requires the examination of possibly thousands of parameters , and it is difficult to visualize the overall convergence of the method . for illustrative purposes , we use the coefficients @xmath152 to construct a single function @xmath234 . in the chosen fourier basis , this amounts to reconstructing the real - space function @xmath234 from its fourier coefficients . for one - body optimizations @xmath235 and for two - body optimizations @xmath236 two - body functions , such as @xmath237 and the jastrow correlation function @xmath6 , are functions of six variables and so , extracting useful information from them is difficult . for illustrative purposes , we indicate in fig . [ fig : superlattice ] two points , @xmath101 and @xmath238 , in both the diamond and rhombohedral graphite structures , corresponding to high and low electron charge density regions , respectively . @xmath101 lies midway between two bonded carbon atoms and @xmath238 lies midway between two layers of carbon atoms . we shall position the first electron at either @xmath101 or @xmath238 . the second electron shall be moved away from this position along one of the following line segments ( indicated by heavy black lines in fig . [ fig : superlattice ] ) : @xmath239 lying within a layer of carbon atoms ; @xmath240 perpendicular to the layers ; @xmath241 lying between two layers ; and @xmath242 perpendicular to the layers . lies at the midpoint of a carbon - carbon bond , with the lines @xmath239 and @xmath240 extending within a layer and perpendicular to the layers , respectively . the point @xmath238 lies midway between two layers ( at a hexagonal interstitial point in diamond ) and the lines @xmath241 and @xmath242 extend between the layers and perpendicular to the layers , respectively . , width=143 ] lies at the midpoint of a carbon - carbon bond , with the lines @xmath239 and @xmath240 extending within a layer and perpendicular to the layers , respectively . the point @xmath238 lies midway between two layers ( at a hexagonal interstitial point in diamond ) and the lines @xmath241 and @xmath242 extend between the layers and perpendicular to the layers , respectively . , width=158 ] \(a ) diamond \(b ) graphite by this means we may plot inhomogeneous two - body functions in terms of the relative separation of the electrons in the system . in particular , we may draw some conclusions about electron correlation in the system by examination of @xmath4 . we may determine the isotropy of @xmath4 by comparing plots of @xmath4 with the first electron kept at the same point but the second electron moved in perpendicular directions , e.g. , by comparing plots designated by @xmath239 and @xmath240 . the homogeneity of @xmath4 may be seen by comparing plots of @xmath4 with the second electron moving in parallel directions from different positions of the first electron , e.g. , by comparing @xmath239 and @xmath241 . any differences between these plots of @xmath4 are attributable to the inhomogeneity and anisotropy of the electron correlation factors in the systems studied . we use periodic boundary conditions ( pbc ) to approximate the infinite crystal . @xcite the simulation cells consist of an @xmath243 unit cell arrangement . the unit cells in each system are defined by the bravais lattice basis vectors . for diamond , we use the basis : @xmath244 , where @xmath245 a.u . , corresponding to a carbon bond - length of @xmath246 a.u . for rhombohedral graphite we use the basis : @xmath247 , where @xmath248 a.u . is the bond - length within the layers , and @xmath249 a.u . is the layer separation . we construct the slater determinant @xmath3 for both systems using dft calculations in the local density approximation ( lda ) . @xcite the lda orbitals were expanded using a linear combination of atomic orbitals comprising gaussians centred on each of the two carbon atoms in the unit cell . @xcite in sec . [ sec.eleccusp ] , we discussed the advantages of removing the short - range cusp from the function @xmath4 in order to improve its representability as a linear combination of smooth functions . figure [ fig : pseudousr](a ) compares the pseudointeraction @xmath195 , generated using a cut - off radius of @xmath250 a.u . and energy eigenvalue @xmath251 hartree ( as discussed in ref . ) , with the coulomb interaction @xmath252 . ( in atomic units @xmath253 ) . the pseudointeraction is used to generate a short - range jastrow funtion @xmath20 , which is shown in fig . [ fig : pseudousr](b ) for the relative angular momenta @xmath254 and @xmath255 corresponding to parallel spin and anti - parallel spin correlation , respectively . the short range jastrow factor used in all subsequent calculations is that generated with these particular values of @xmath196 and @xmath197 . subsequent figures in this paper , which involve @xmath20 , represent anti - parallel spin correlation only . the cut - off required for a convergent fourier expansion of a smooth cuspless function should be much less than that required for a function with a short - range cusp . therefore , using a cuspless form greatly reduces the number of terms required to represent the inhomogeneous form of the jastrow factor accurately in fourier space . we illustrate this point using a simple example . in fig . [ fig : cuspremove ] we plot @xmath256 times the fourier transform , for wave vector @xmath257 , of the yukawa - style homogeneous correlation factor , @xmath258 which has been used by many authors to approximate electron correlation in a variety of systems . @xcite we set @xmath259 a.u . , ( @xmath260 is determined from @xmath101 to satisfy the cusp conditions @xcite and depends on the relative spin of the electrons ) . also shown in fig . [ fig : cuspremove ] is @xmath256 times the fourier transform of the cuspless difference @xmath261 . we assume that , for large electron separations , electron correlation is approximately spin - independent . therefore , for each @xmath257 , we plot the mean of the parallel spin and anti - parallel spin values of the functions . in practice , we use the first zero of the fourier transform of @xmath195 as the fourier space cut - off @xmath89 used in the definition of @xmath190 in eq . [ eq : inhomjast ] . for @xmath262 we use the cut - off @xmath263 , beyond which the fourier transform of the cuspless @xmath4 function is approximately zero ( fig . [ fig : cuspremove ] ) . combining both the short - range and inhomogeneous forms of the jastrow factor using this scheme produces a form that is approximately independent of the cut - off , since decreasing @xmath196 increases the reciprocal space cut - off @xmath89 . for comparison with the inhomogeneous @xmath4 functions determined in the following sections we use the yukawa - style homogeneous function @xmath264 of eq . [ eq : yukawa ] and construct a homogeneous trial wave function of the form @xmath265 . as with the inhomogeneous trial wave function , we represent short - range correlation ( i.e. , the cusp ) using @xmath22 and represent the remaining correlation using a homogeneous jastrow factor @xmath266~,\end{aligned}\ ] ] where we define @xmath267 e^{-i{\bf q \cdot r } } d{\bf r}~.\end{aligned}\ ] ] for the uniform electron gas , bohm and pines @xcite predicted that the true function @xmath4 should decay as @xmath268 , at large separation @xmath269 , where @xmath270 is the plasma frequency . rather than use this limiting value @xmath271 in eq . [ eq : yukawa ] , it is common to treat @xmath101 as a free parameter such that the energy is minimized . using variational calculations we can determine the optimal value of @xmath101 . @xcite we optimize the one - body jastrow factor @xmath222 in eq . [ eq:1bjast ] , using our iterative method . the jastrow factor @xmath272 , combined with the slater determinant @xmath3 , is our best approximation of the true many - body eigenstate using a homogeneous two - body jastrow factor and is comparable with similar wave functions optimized using variance minimization . @xcite in fig . [ fig : uhom ] we plot @xmath264 and compare it with the reconstructed function @xmath273 for @xmath274 unit cell simulations of diamond and rhombohedral graphite . periodic boundary conditions and the anisotropy of the unit cell make this reconstructed form appear quite different to the original isotropic function @xmath264 . in particular , for rhombohedral graphite the unit cell used is quite anisotropic , leading to marked differences in the reconstructed function along the perpendicular line segments @xmath239 and @xmath240 . ( the homogeneity of @xmath264 is preserved and so we only plot the function for point @xmath101 , since all other points are equivalent . ) note that the jastrow @xmath4 function is defined up to an arbitrary constant , much like a potential , since this constant affects only the wave function normalization and contributes nothing to the description of correlation . therefore , it is of no consequence that the functional form of @xmath264 in eq . [ eq : yukawa ] appears shifted above the reconstructed forms compatible with pbc . this is due to the removal of the constant fourier coefficient of the correlation function @xmath275 from the expansion of the reconstructed function in eq . [ eq : yukawarecon ] . we note that the cusp conditions @xcite are maintained by all forms . we optimize the one - body jastrow @xmath222 using the method described in sec . [ sec.enermin ] . for the diamond simulations we used a fourier space cut - off @xmath276 , giving 84 variational one - body parameters . this is more than enough for an accurate representation of one - body terms in the wave function ( see sec . [ sec.energycorrections ] ) . for graphite simulations we used @xmath277 , giving 30 variational one - body parameters . the values of the electronic energy per atom for various optimizations of this homogeneous trial wave function are shown in figs . [ fig : diamondener ] and [ fig : graphiteener ] . each iteration involved averaging over @xmath278 monte carlo samples . however , this amount of averaging was more than enough for an accurate implementation of the method . the optimization of the one - body terms in the @xmath274 simulation of graphite ( fig . [ fig : graphiteener ] ) involved only @xmath279 samples per iteration and is well converged . on average , the gain in energy following one - body optimization is approximately @xmath280 mhartree / atom for diamond and @xmath281 mhartree / atom for graphite . we note that the necessity for a one - body correction is a consequence of the inhomogeneity of the electronic charge density in the system , @xcite and it is not surprising that the gain in energy is larger for the more inhomogeneous system , graphite , than for diamond . the slater determinant used in the diamond calculations of fig . [ fig : diamondener ] is composed of single - particle orbitals generated from an lda calculation . the lda orbitals are linear combinations of gaussian basis functions of @xmath282 , @xmath283 and @xmath284 symmetry , using three decays of 0.24 , 0.797 , and 2.65 . the exchange - correlation functional used was of the ceperley - alder @xcite form . the cut - off of the fourier space expansion of the charge density @xcite in the lda calculation was 64 rydberg . the graphite calculations shown in fig . [ fig : graphiteener ] use a slater determinant generated using a gaussian basis - set with @xmath282 and @xmath283 symmetry only . four decays were used : 0.19 , 0.474 , 1.183 , and 2.95 . the orbitals were generated from lda calculations incorporating the hedin - lundqvist @xcite exchange - correlation functional , and a cut - off of 36 rydberg for the fourier space expansion of the charge density . ( see sec . [ sec.energycorrections ] for a discussion of basis - set convergence of the total energy in graphite . ) figure [ fig:1boptchi ] illustrates the convergence during optimization of the one - body function @xmath285 in eq . [ eq : jastrowpsi ] ( where @xmath5 is the accumulation of all one - body terms from each of the jastrow factors ) for the @xmath286 diamond calculation . the optimized function is statistically well - determined , and the change from the initial one - body function , @xmath287 of eq . [ eq : natural1body ] ( as defined in refs . and ) , is well - defined . this alteration of the one - body function may be compared to similar calculations performed using variance minimization . @xcite the number of parameters for optimization could be greatly reduced through exploitation of the crystal point - group symmetry of the structures involved . however , it is worth noting that the optimization process preserves the natural symmetry of the system ( within statistical error ) without such measures , illustrating that for non - symmetric systems with large numbers of parameters , this optimization process should be quite robust . the one - body function @xmath288 , reconstructed from the coefficients associated with the local energy @xmath289 at iteration @xmath290 according to eq . [ eq:1bodyv ] , is shown in fig . [ fig:1boptel ] . clearly , this function decreases in magnitude , indicating a decrease in the magnitudes of the euler - lagrange derivatives . beyond the first iteration , @xmath288 is of the same order of magnitude as its associated standard error , and so is statistically insignificant . therefore , the method has essentially converged after only one iteration . the noisiest regions of @xmath288 correspond to regions of low density where the monte carlo sampling is less frequent . the analytic guess for the two - body predictor functions @xmath291 , outlined in sec . [ sec.predictoranalapprox.j2b ] , leads to an inhomogeneous generalization of the rpa equations . the solution to these equations is the function @xmath292 , shown in fig . [ fig : urpa ] for the @xmath274 graphite simulation . we notice some inhomogeneity in @xmath293 at intermediate- and long - range electron separations . a more homogeneous and isotropic @xmath293 was found for diamond , as we would expect since diamond possesses a more uniform electron density than graphite . in figs . [ fig : diamondener ] and [ fig : graphiteener ] , the first point on all solid curves indicates the total energy per atom in each simulation calculated using @xmath293 . in comparison with the energy calculated using the homogeneous jastrow function @xmath264 , we see that the inhomogeneous rpa trial wave function is at best comparable in accuracy with the optimized trial wave function with homogeneous two - body jastrow factor , and often less accurate . these results are different from those of gaudoin _ et al . _ @xcite for model systems : they find that their inhomogeneous generalization of the rpa produces wave functions that yield lower energies than the homogeneous form . we simultaneously optimized the parameters in both the one - body jastrow factor @xmath222 and the fully inhomogeneous form of the two - body jastrow factor @xmath190 , using our iterative method . the convergence of the total energy per atom for various simulations of diamond and rhombohedral graphite are shown in figs . [ fig : diamondener ] and [ fig : graphiteener ] . ( the slater determinants used in combination with the homogeneous jastrow factor @xmath294 in sec . [ sec.homoj ] are also used here . ) convergence of the total energy is achieved in approximately three iterations in all cases and is stable . this is remarkable given that the system sizes range from 8 electrons and 140 independent parameters to 216 electrons and 3136 independent parameters . also , we used the same amount of monte carlo sampling , viz . , @xmath278 samples per iteration , to determine the required expectation values for all simulations . in all cases the fully inhomogeneous form of @xmath4 allows us to determine more accurate trial wave functions with substantially lower energies than the homogeneous trial functions . in general , the gain in energy through using an inhomogeneous rather than a homogeneous wave function is of the order of @xmath295 mhartree / atom for both diamond and rhombohedral graphite . figure [ fig : graphuconv ] illustrates the rapid convergence of the two - body wave function parameters @xmath191 to their optimal values during the largest graphite optimization ( @xmath274 ) and is typical of all the optimizations performed in both diamond and graphite . beyond the third iteration , no clear distinction exists between subsequent sets of parameters . the optimal jastrow function @xmath4 is significantly different from both the rpa function @xmath296 and the homogeneous form @xmath297 of eq . [ eq : yukawa ] . the proof that this optimization succeeds in minimizing the energy expectation value may be seen in fig . [ fig : graphvconv ] ( which comes from the same graphite calculation as fig . [ fig : graphuconv ] ) . here , we plot the iterative decay of the two - body function @xmath237 reconstructed from the total energy coefficients determined at each iteration , according to eq . [ eq:2bodyv ] . this clearly indicates the reduction to zero ( within statistical noise ) of the derivatives in the euler - lagrange equations , thus solving the energy minimization problem . [ fig : graphuconv ] for the largest simulation cells studied ( @xmath274 unit cell arrangement containing 216 electrons ) , we compare the optimal jastrow correlation functions @xmath4 of diamond and rhombohedral graphite . figures [ fig : diamuopt ] and [ fig : graphuopt ] show the function @xmath4 plotted with respect to electron separation on various line segments in the corresponding crystal structures , as already explained . [ fig : diamuopt ] for a direct comparison of the calculated energies of diamond and rhombohedral graphite , we should ( a ) make some corrections based on the trial wave functions used and ( b ) include finite size and zero - point phonon energy corrections for the expected energy of the real infinite solid . the corrected energies for diamond and graphite are listed in table [ tab : enercorr ] . . energies and energy corrections of diamond and rhombohedral graphite ( in hartree / atom ) . ( a ) total energy , determined by vmc and our optimization method , for a @xmath274 unit cell simulation region , using wave functions of similar variational freedom ( see text ) . ( b ) finite size correction equal to the energy difference between an lda calculation for the @xmath274 simulation region and an lda calculation using a fully - converged k - point set ( see text ) . ( c ) correction for the zero - point phonon energy . @xcite ( d ) total energy including the corrections . the numbers in parentheses indicate the statistical error in the last digits of the corresponding energy . [ tab : enercorr ] [ cols= " < , > , < , > , < " , ] the jastrow factors used in both solids are comparable in their variational freedom , the only significant difference being the size of the cut - off used for @xmath298 . however , a vmc calculation for diamond , using the same cut - off as in graphite ( @xmath299 ( a.u.)@xmath300 , corresponding to 25 variational parameters ) , resulted in an increase in energy of only @xmath301 mhartree / atom for diamond . therefore , the high fourier coefficients of @xmath298 contribute little to the total energy of the system . the diamond vmc energy for the @xmath274 simulation quoted in table [ tab : enercorr ] was determined using @xmath299 ( a.u.)@xmath300 as the cut - off for one - body terms . in graphite , the exclusion of @xmath284 symmetry from the basis set used to construct the lda orbitals in the slater determinant is energetically more important . in addition to the calculations described in secs . [ sec.homoj ] and [ sec.inhomjopt ] , we also performed calculations for graphite using a gaussian basis set with @xmath282 , @xmath283 and @xmath284 symmetry , and three gaussian decays : 0.22 , 0.766 , 2.67 . for the @xmath274 simulation , including @xmath284 symmetry reduces the total vmc energy by @xmath302 mhartree / atom and also reduces the variance in the total energy by 16% . use of the ceperley - alder exchange correlation functional to generate the single - particle orbitals in graphite , rather than the hedin - lundqvist form , made no difference ( within statistical error ) to the vmc energies , and neither did an increase in the cut - off for the fourier space expansion of the lda charge density from 36 to 64 rydberg . the graphite vmc energy for the @xmath274 simulation listed in table [ tab : enercorr ] was calculated using a trial wave function very similar to that used for the calculation of the corresponding diamond vmc energy . the cut - off for the one - body jastrow factor was @xmath299 ( a.u.)@xmath300 and the slater determinant comprised lda orbitals obtained using : ( i ) a basis set with @xmath284 symmetry ( as outlined above ) ; ( ii ) the ceperley - alder exchange correlation functional ; and ( iii ) a 64 rydberg cut - off for the lda charge density expansion . we generate finite size corrections for the @xmath274 unit cell simulations of diamond and graphite , by calculating the difference in energy between an lda calculation which uses k - points compatible with periodic boundary conditions of a @xmath274 simulation region and an lda calculation using a fully converged k - point set . @xcite comparing the change in energy between a @xmath303 calculation and a @xmath274 calculation in both diamond and graphite , using lda and vmc , we see that the change in vmc energy is about 80% of the lda energy change in diamond , and 70% in rhombohedral graphite . perhaps more accurate estimates of the energy of the infinite solid may be obtained by implementation of a model periodic coulomb interaction developed recently . tests of this approach have dramatically reduced finite - size effects in the interaction energy . @xcite for diamond , we estimated the finite size correction to be @xmath304 mhartree / atom , using a converged lda calculation with 220 k - points in the irreducible brillouin zone . for rhombohedral graphite , incorporating @xmath284 symmetry in the basis set ( as described above ) , and using 189 k - points in the lda calculation , we found the finite size correction to be @xmath305 mhartree / atom . we also include the calculated zero - point phonon energies of diamond and graphite , which are @xmath306 and @xmath307 mhartree / atom , respectively . @xcite adding all these corrections to the calculated vmc energies ( table [ tab : enercorr ] ) , we estimate the energies of the infinite solids to be @xmath308 hartree / atom for diamond and @xmath309 hartree / atom for rhombohedral graphite . this appears to indicate that rhombohedral graphite is less stable than diamond . however , given the approximation of using lda finite size corrections , we might expect a systematic error of the order of 2 mhartree / atom in each of these results . this indicates that at the vmc level , the solids diamond and rhombohedral graphite have very similar total energies . we note that in the atomic pseudopotential used in the calculations presented here @xmath283 and higher angular momentum scattering are all included in the local potential . it is possible that the use of a separate @xmath284 pseudopotential might slightly affect the relative energies in both systems . in order to determine the cohesive energy of a solid , we should subtract the energy per atom of the solid from the energy of the isolated atom , @xmath310 . however , when using approximate eigenfunctions , a reasonable estimate of the cohesive energy is obtainable only by subtracting the energies estimated using similar trial wave functions . vmc energies are available for the carbon atom where the trial wave function is of the jastrow - slater form . @xcite the orbitals of the slater determinant are optimized using energy minimization and a sophisticated jastrow factor is optimized using variance minimization , yielding a vmc energy for the atom of @xmath311 mhartree . using this energy , we find the cohesive energies to be @xmath312 hartree / atom for diamond and @xmath313 hartree / atom for graphite . we regard this atomic trial wave function to be close in form and accuracy to our solid trial wave function . however , to remain consistent with the inclusion of @xmath284 symmetry in the basis set of our lda calculations , we could also refer to a multiconfiguration trial wave function for the carbon atom which includes @xmath284 excitations . the vmc energy of the atom , using this wave function , is @xmath314 hartree , @xcite leading to estimates of the cohesive energies of diamond and rhombohedral graphite of @xmath315 hartree / atom and @xmath316 hartree / atom , respectively . the experimental values are 0.271 hartree / atom for diamond , and 0.272 hartree / atom for graphite . @xcite considering the significant gain in energy obtained using an inhomogeneous jastrow factor rather than a homogeneous form , one might wonder how much the difference between vmc and dmc energies has been reduced . using jastrow correlation factors which include one - body terms and homogeneous two - body terms , similar in form to that used in sec . [ sec.homoj ] , for bulk carbon @xcite and silicon , @xcite typically leads to vmc correlation energies that are approximately 90% of the corresponding dmc correlation energies , i.e. , @xmath317 where @xmath318 is the energy calculated using vmc , @xmath319 is the energy calculated using dmc , and @xmath320 is the hartree - fock energy . ( @xmath320 has been estimated using just the lda slater determinant as trial wave function . @xcite ) we have observed that including inhomogeneity in the correlation factor leads to a gain in energy of approximately 10 mhartree / atom in the largest simulations of both diamond and graphite . if we estimate that the vmc correlation energy , determined using our optimized wave function with homogeneous correlation terms , obtains 90% of @xmath321 , then using the optimized wave function with inhomogeneous correlation terms obtains 96% of the dmc correlation energy . this would indicate that the energy difference between vmc and dmc for these solids has been decreased by 60% . it is important to emphasize that , whatever physical conclusions we would like to draw from these calculations , our principal aim has been to optimize the trial wave function for a given hamiltonian , such that the expectation value of the total energy is minimized . this aim has been achieved for all the systems studied . diamond , with a relatively homogeneous and isotropic electron charge density , exhibits an approximately homogeneous , isotropic jastrow correlation function @xmath4 ( fig . [ fig : diamuopt ] ) . at large electron separations ( beyond 6 a.u . ) we see that the electron correlation factor @xmath4 in diamond is well approximated by homogeneous and isotropic functions , as all the curves plotted are quite similar . only slight deviations from homogeneity exist at short and intermediate electron separations . this inhomogeneity may be seen by comparing @xmath4 plotted with its fixed coordinate at different points : @xmath239 and @xmath240 are quite similar at short range , but clearly distinct from @xmath241 and @xmath242 in the same region . this inhomogeneity may have significant effects on the short - range pair - correlation functions calculated for diamond - like systems using vmc methods . @xcite graphite , with clear regions of high electron charge density and well - defined regions of very low electron charge density between its layers , is a highly inhomogeneous and anisotropic structure . this is borne out in fig . [ fig : graphuopt ] , where the function @xmath4 differs considerably in various regions and in various directions . at short - range , @xmath4 is surprisingly homogeneous in comparison with diamond . at intermediate separations the function displays both inhomogeneous and anisotropic behaviour . given that the layer separation in these simulations is 6.33 a.u . , this indicates that inhomogeneous correlation between adjacent layers in the system is not insignificant . this may prove important for van der waals interactions between the layers in graphite . at long range , the anisotropy of the graphite correlation factor is clearly shown in fig . 13 , where the correlation factors for electron separation vectors @xmath322 lying parallel to the graphite planes ( @xmath239 and @xmath241 ) are distinctly different from those with the separation vector perpendicular to the planes ( @xmath240 and @xmath242 ) . inhomogeneity ( i.e. an explicit dependence on the position @xmath323 of the first electron ) is displayed at long range in the differences between the function along @xmath239 and @xmath241 . we might expect this , given that the line aa lies within the graphite planes , where the charge density is concentrated , whereas @xmath241 lies in the very low charge density region between the planes [ see fig . [ fig : superlattice](b ) ] . on the other hand , when the separation vector is oriented perpendicular to the graphite planes ( @xmath240 and @xmath242 ) , inhomogeneous effects are substantially smaller at long range in order to reduce the complexity of the physical results in sec . [ sec.results ] , some computational details of the method were not discussed . we present and discuss some of these details in this section . \(1 ) this iterative method is trivially parallelizable . to determine the expectation values required to construct the predictor , monte carlo sampling may be performed independently on many workstations and the results combined . to obtain total energies of the accuracy presented in this paper requires @xmath324 monte carlo samples . however , for optimization of the wave function using our method , this amount of sampling is also sufficient for accurate estimations of the expectation values required to construct the predictor ( sec . [ sec.predictornumer ] ) . the extra time required to accumulate the various contributions to the predictor represents less than @xmath325 of the computational time needed to calculate each energy sample . there are two reasons for this : firstly , because we have expressed the variational components of the jastrow factor as linear combinations of the operators @xmath102 of sec . [ sec.enermin ] . these @xmath102 need only be evaluated once , for each electron configuration , in order to construct both the jastrow factor and the predictor . secondly , and more importantly , since we must calculate the total energy as a sum of various contributions defined by the hamiltonian of the system , all the energy contributions needed to construct the predictor are already available , either directly or by some simple manipulation . it is also important to note that , despite the complex form of the jastrow factor outlined in sec . [ sec.wavefn ] , the amount of computational time spent evaluating it is still less than or equal to that spent evaluating the slater determinant , for a given electron configuration . the largest calculations presented in this paper were performed on a beowulf cluster of fifteen 500 mhz dual - processor workstations . for a 5 iteration optimization using @xmath278 samples per iteration , these calculations took about 50 hours on this cluster . however , perhaps half this amount of sampling would have produced comparable results [ see ( 3 ) below ] . the required memory for storing all the expectation values necessary for this calculation was approximately 20 mb . \(2 ) the newton - raphson method ( see appendix [ app.newtonraphson ] ) , while quadratically convergent near a root of a multidimensional system , does possess some convergence problems far from the root . in the calculations presented in this paper , we found that below a certain minimum amount of sampling , the noise in the estimated expectation values used to construct the predictor caused divergence of the solution to eq . [ eq : generateparam ] using the newton - raphson method . this problem might be solved through the use of a more robust root - finding method for the predictor function . \(3 ) from the analysis in appendix [ app.convmeth ] , we see that noise in the iterative method comes from the finite sampling used to estimate those expectation values ( listed in sec . [ sec.predictornumer ] ) required to construct the predictor . the effect of this noise on the wave function accuracy is not clear . to reduce the computational cost of these calculations we would prefer to do the least amount of sampling necessary to produce the required results . figure [ fig : esamp ] illustrates the effect of various amounts of sampling on the convergence of the total energy in the optimization of the @xmath303 simulation of rhombohedral graphite . at @xmath326 samples per iteration , the estimation of the required expectation values , during the first iteration of our method , is too crude to produce a convergent root of the predictor using the newton - raphson method . this leads to a wave function , used in the second iteration , with many noisy parameters which are more difficult to optimize , as the convergence of the energy shows . however , for sampling involving @xmath327 samples per iteration , or more , we see that the convergence of the total energy is identical ( within statistical accuracy ) . comparison of the optimized wave function parameters also reveals only small differences , indicating that the only means of determining the true benefits of more sampling is by examination of the fitted coefficients @xmath152 at each iteration @xmath290 . in fig . [ fig : vsamp ] we see , from the reconstruction @xmath328 of the fitted coefficients for two - body optimization at the fifth and final iteration of the method , that the most sampling ( @xmath329 samples per iteration ) reduces the magnitude of the euler - lagrange derivatives the most , indicating that these wave function parameters are the most accurate . however , for practical purposes , there is little distinction between the accuracy of the wave function once we increase the sampling beyond @xmath327 samples per iteration . thus , the optimizations presented in sec . [ sec.results ] may have used more computational time than was strictly necessary . however , more testing is required to determine the minimum amount of sampling as a function of system size . we have developed a generalized form of electron correlation factor for trial many - body wave functions of electrons in periodic solids . this form allows us to represent fully inhomogeneous electron correlation in real physical periodic systems . it is computationally efficient to evaluate , since the electron cusp , which we express as a homogeneous correlation factor , is separated from the fully inhomogeneous form . we have also developed a rapidly convergent iterative method for the optimization of all variational parameters in these wave functions , minimizing the total energy of the given system . it uses the accurate techniques of quantum monte carlo sampling to achieve this optimization and has allowed new insights into the form of many - electron correlation in systems with highly inhomogeneous charge densities . we estimate the difference in energies calculated using the optimal inhomogeneous two - body correlation factor and the optimal homogeneous correlation factor to be approximately 60% of the difference between the dmc and vmc energies obtained with homogeneous 2-body correlation terms . in diamond , the optimal correlation factor is approximately homogeneous and isotropic , with some inhomogeneity at short- and intermediate - range electron separations . this is consistent with its comparatively homogeneous and isotropic electron charge density . graphite has an optimal correlation factor which is quite homogeneous at short - range electron separation , but is significantly inhomogeneous and anisotropic at intermediate- and long - range electron separations , as one might expect from its highly inhomogeneous and anisotropic electron charge density . nevertheless , it is remarkable that despite very large inhomogeneity in the electron pair - correlation function , found by previous authors , @xcite the ideal inhomogeneous jastrow two - body term , calculated for the first time here for diamond and graphite , displays relatively small inhomogeneity . whether this conclusion can be extended to other systems ( e.g. involving strongly correlated @xmath284-electrons ) remains an open question . this work has been supported by enterprise ireland grant number sc/98/748 and by the irish higher education authority . the work of s.f . at the national microelectronics research centre is supported by science foundation ireland . the derivative of the expectation value of an observable @xmath101 , with respect to a parameter @xmath330 of the parameterized wave function @xmath105 , may be written as @xmath331 we assume that @xmath101 is independent of the parameters @xmath120 . also , for a time independent system , we may in general express @xmath33 as a real function . differentiating , we find that @xmath332 where we define the operator associated with variations of @xmath112 as @xmath333 , with local value @xmath334 for the many - body configuration @xmath55 . therefore , we may express the derivative of @xmath335 as @xmath336 where @xmath337 and @xmath338 . our task is to minimize the integral @xmath339 by choosing the appropriate parameters @xmath340 and @xmath341 . we note first that , at the minimum , @xmath342 where @xmath343 is the expectation value of the total energy and @xmath344 is the expectation value of the operator @xmath345 . to fulfill the minimization , we must set all the remaining first derivatives of @xmath346 to zero , i.e. , @xmath347 for each @xmath348 . upon substitution of the minimum value of @xmath340 , this leads to @xmath349 \ .\end{aligned}\ ] ] since , for any operators @xmath350 , @xmath351 we may say that @xmath352 where @xmath353 , then we are lead to the conclusion that the least squares fitting is equivalent to solving the linear system @xmath354 for each @xmath348 . an integral part of the iterative procedure outlined in sec . [ sec.iterproc ] is the determination of the parameters @xmath120 that solve the system in eq . [ eq : generateparam ] . the determination of the roots of any multidimensional function can be troublesome . in all the optimizations presented in this paper , the predictor function @xmath210 is a quadratic function of @xmath120 , whose coefficients are determined analytically , or by numerical fitting at the point @xmath124 . we ignore the implicit dependence of @xmath128 on @xmath124 for the purpose of finding a root , and solve the system @xmath355 for all @xmath356 . we use the newton - raphson method @xcite to determine the roots . this is an iterative method of improving successive guesses for a root of a function . it involves the computation of the function @xmath128 and its jacobian matrix of derivatives with respect to @xmath120 , at each guess . the iterations continue until convergence of the solution is achieved within a predefined tolerance . the success of the newton - raphson method for multi - dimensional systems relies heavily on the proximity of the initial guess to the root we seek . for this reason , the initial guess chosen is normally the parameter set of the previous iteration of the procedure outlined in sec . [ sec.iterproc ] . to find the first set of parameters , by finding the roots of the analytic predictor @xmath357 , we require a good initial guess of the root . since the one - body parameters @xmath93 are expected to be small by construction , an initial guess of zero for all @xmath81 was found to be sufficient to produce a convergent solution to the first application of the newton - raphson method . for the two - body problem , we rescale the variables @xmath68 to improve the convergence of the root - finding method . according to the rpa , @xcite the long - range behaviour of the @xmath4 function should take the form @xmath358 , where the plasma frequency for a homogeneous system with electron charge density @xmath290 is @xmath359 . the charge density @xmath290 is determined in the simulation region to be @xmath360 , for @xmath0 the number of electrons in the simulation volume @xmath361 . therefore , for small @xmath53 in fourier space , @xmath4 behaves like @xmath362 this indicates large relative differences between values of @xmath363 for small @xmath53 . therefore , in inhomogeneous systems , it would be appropriate to rescale the parameters @xmath68 by multiplying by @xmath364 , thus rendering all the variables of the same order of magnitude as the plasma frequency . this leads to a less pathological numerical problem for the newton - raphson method . appropriate scaling must also be applied to the predictor function @xmath365 in sec . [ sec.enermin ] . for the initial analytic guess of the roots of @xmath366 we use the homogeneous solution for @xmath68 outlined in eq . [ eq : homosoln ] . the variational part of @xmath33 associated with one - body terms is the jastrow factor @xmath222 ( eq . [ eq:1bjast ] ) with parameters @xmath367 . we expand the energy contributions @xmath172 and @xmath173 defined in sec . [ sec.energypartition ] . for @xmath222 , we have that @xmath368 the constant terms are not required , so we ignore them . the energy contribution @xmath173 can not be expanded analytically as a linear combination of the functions @xmath369 since @xmath370 is not explicitly a function of these coordinates . however , @xmath173 is explicitly linear in the parameters @xmath372 , with derivatives @xmath373 which are independent of @xmath372 . the variational part of @xmath33 associated with two - body energy contributions is the inhomogeneous jastrow factor @xmath190 of eq . [ eq : inhomjast ] , with parameters @xmath374 . ( for periodic systems we use only @xmath375 . ) the energy contributions @xmath132 of sec . [ sec.energypartition ] are expanded here . the contribution @xmath172 is dependent only on the form of @xmath190 and is expanded as @xmath376 we find that @xmath377 \delta \rho_{\bf q}^ * + \mbox{constant}~. \end{aligned}\ ] ] we retain only the linear combination of the functions @xmath378 . the constant terms we may ignore , and the one - body terms ( linear combinations of @xmath379 ) we assume are compensated by terms in the one - body jastrow @xmath222 . if we ignore the removal of one - body terms from the function @xmath380}$ ] and consider using the function @xmath381 instead , then we find that @xmath382 where we have ignored constant and one - body terms in the expansion . the product @xmath383 contains both two- and three - body terms , since we may rewrite @xmath384 as @xmath385 . we intend here to remove two - body fluctuations and regard three - body fluctuations as much less significant , so we retain only the two - body term @xmath386 from the charge fluctuation products in eq . [ eq : fullquad ] , i.e. , @xmath387 now we make the assumption that removing one - body terms from this expression , i.e. , replacing @xmath381 with @xmath378 is a good approximation , and obtain the expression for @xmath388 given in sec . [ sec.predictoranalapprox.j2b ] . the contribution @xmath173 can not be expanded analytically in the basis of fluctuation functions @xmath378 , however , it is clear that @xmath173 is linear in @xmath191 , since , for @xmath389 , @xmath390 and has first derivatives @xmath391 which are independent of the parameters @xmath191 . the final energy contribution @xmath171 we attempt to express analytically in terms of the two - body fluctuation functions @xmath378 . if @xmath392 , the lda slater determinant , then the sum of contributions from the external potential and the kinetic energy term @xmath393 is a one - body contribution defined by the kohn - sham hamiltonian , @xcite since @xmath394 = \sum_i [ \epsilon^{\rm ks}_i - v_{\rm h}({\bf r}_i ) - v_{\rm xc}({\bf r}_i ) ] ~,\end{aligned}\ ] ] where @xmath395 are the kohn - sham eigenvalues , @xmath396 is the hartree potential and @xmath397 is the exchange and correlation potential . therefore , the two - body contribution of these terms is approximately zero . we are left with the contribution of the electron - electron interaction @xmath164 . for a two - body potential @xmath398 , we may expand the sum over electron pairs as @xmath399}~,\end{aligned}\ ] ] for fourier coefficients @xmath400 . in terms of the fluctuation functions @xmath378 , this may be rewritten as @xmath401 \delta \rho_{\bf q}^ * + ~\mbox{constant}~.\end{aligned}\ ] ] ( we assume that @xmath398 possesses exchange symmetry , so that @xmath402 . ) therefore , both one- and two - body fluctuations arise from a two - body potential in the `` charge fluctuation '' coordinate system . note that the one - body fluctuations are expressible in terms of the hartree potential , since @xmath403 if the two - body potential is homogeneous , i.e. @xmath404 , then we may simplify the fluctuations since @xmath405 , where the fourier transform of the homogeneous function @xmath406 is @xmath407 for a system volume @xmath361 . if @xmath408 then @xmath409 . therefore , @xmath410 again we ignore the one - body and constant terms in this context . the short range jastrow factor @xmath22 is constructed from the pseudointeraction @xmath195 in the following way . for the isolated two - electron scattering problem , we may find an eigenstate @xmath411 of the true two - electron hamilatonian @xmath412 for a given energy eigenvalue @xmath197 . upon replacing the true coulomb interaction @xmath164 with a generated pseudointeraction @xmath195 , we construct the modified hamiltonian @xmath413 with eigenstate @xmath414 corresponding to @xmath197 . we construct @xmath22 such that @xmath415 now , in a many - electron environment , we know that using @xmath22 allows for good approximation of short - range correlations . @xcite we might imagine that for a many - electron system with hamiltonian @xmath118 and many - electron trial wave function @xmath416 , the true energy eigenvalue may be well approximated by @xmath417 where @xmath418 $ ] . now , disguising the true interaction @xmath164 with @xmath419 , and given the transferability of @xmath195 over a wide range of energies , we see that @xmath420 dividing two - body correlation into short - range and inhomogeneous terms , we use a jastrow factor of the form @xmath421 . the local energy determined using the trial wave function @xmath422 , where @xmath423 , may be expanded as @xmath424 for this reason , we use @xmath195 in the expansion of the local energy for @xmath190 to implicitly include the short - range jastrow factor @xmath22 . the convergence criterion @xmath122 , is numerically never exactly achieved . given that the predictor @xmath425 contains some terms determined by statistical fitting , finite sampling errors exist , and this noise is passed on to the fitted parameters in @xmath425 from eqs . [ eq : lstsqsys ] and [ eq : contributionlstsqsys ] . we consider @xmath426 , the fitted coefficients of the local energy at the @xmath427 iteration of the method . the method may be regarded as an iterative map @xmath428 , such that the coefficients are determined via @xmath429 . there are two sources of noise in @xmath430 : ( i ) noise inherited from @xmath431 , which produced the parameters @xmath156 , which were used to construct the wave function @xmath432 , with which we evaluated the expectation values used to calculate @xmath430 ; and ( ii ) finite sampling noise in the evaluation of the expectation values in eqs . [ eq : lstsqsys ] and [ eq : contributionlstsqsys ] via monte carlo sampling . therefore , we associate a set of variances @xmath433 , arising from these two sources of noise , to the coefficients @xmath434 . the variance due to finite sampling alone , at each step @xmath290 , is @xmath435 and we use the initial condition @xmath436 . this implies that the variance obeys the following iterative map : @xmath437 where @xmath438 . for a convergent map @xmath428 , we are guaranteed that @xmath439 at convergence . @xmath440 is a measure of the convergence rate of the map @xmath428 , with @xmath441 implying fast convergence and @xmath442 implying slow convergence . note that if @xmath443 the map is divergent . we conclude that , given a convergent method @xmath428 , the presence of statistical noise does not lead to successively more ill - determined parameters @xmath120 , since their variance is also convergent . note that eq . [ eq : convvar ] indicates that the variance in the fitted coefficients @xmath126 is always greater than or equal to the variance estimated using finite sampling . however , our implementation of the method represented by the map @xmath428 indicates that @xmath449 , since we find that the majority of the coefficients @xmath126 ultimately end up with magnitudes approximately equal to their finite sampling errors , signifying that statistically they are zero . therefore , the final conclusion to be drawn from eq . [ eq : convvar ] is that the accuracy of our optimization method depends ultimately on the finite sampling error . therefore , increasing the computational workload , by increasing the amount of sampling , will result in more accurate optimizations of the wave function . this is demonstrated by the results in sec . [ sec.discussion ] .

we had two distinct objectives in undertaking this work , dominantly to investigate the merit and reliability of k giant stars as absolute fir and millimeter calibration sources , but secondarily to determine just how shallow in depth simple radiative equilibrium cool star model atmospheres might be trusted . we have chosen @xmath0 tau and @xmath0 boo for this initial study because these stars , being both bright and nearby , are easily the two most completely and thoroughly studied red giants . this makes them the ideal test cases for our purposes . it should be noted that they both lie on the cool , non - coronal side of the linsky - haisch dividing line " @xcite which separates coronal from non - coronal giants , although @xcite have found evidence for buried high temperature emission in the case of @xmath0 boo . cohen and colleagues have presented a self - consistent context for absolute calibration in the ir , ideal for use by ground - based , airborne , and spaceborne spectrometers and radiometers @xcite . their approach is based upon a pair of absolutely calibrated , ir - customized models of vega and sirius calculated by kurucz . these efforts have furnished the absolute stellar spectrum of sirius that underpins cobe / dirbe bands @xmath3 @xcite , the spectra of k0-m0iiis for the on - orbit calibration of the near- and mid - infrared spectrometers on the joint nasa / isas infrared telescope in space @xcite , likewise for esa s infrared space observatory ( iso ) @xcite instruments ( e.g. , the short wavelength spectrometer ) @xcite , for the spatial infrared imaging telescope ( spirit - iii ) on the us midcourse space experiment ( msx ) @xcite , and for 2mass @xcite . an extension of this work to substantially fainter stars has provided calibrators for the _ spitzer _ space telescope @xcite . most recently , @xcite have absolutely validated the two calibrated model spectra , and the empirical spectra of the brighter k giants such as @xmath0 tau and @xmath0 boo , in six bands in the mid - infrared ( mir ) , using precision measurements from the msx . the absolute calibration of isophot @xcite was supported by planets , asteroids , and stars , ranked in order of decreasing flux density @xcite . the stellar calibrators @xcite necessitated long - wavelength extrapolations , as far as 300 @xmath2 m , of the observed ( 1.235 @xmath2 m ) absolute stellar spectra by model atmosphere spectra ( computed by one of us [ dfc ] : @xcite ) . we note that the bright , secondary reference stars that have been selected for ir calibration purposes are precisely those quiet " giants in which it was suggested that radiatively - cooled regions very largely dominate the stellar surfaces so that single component atmospheric models were most likely to be valid . examples of these are @xmath0 boo , @xmath0 hya , @xmath0 tau , and @xmath4 dra @xcite . consequently , one of our objectives is to establish whether these stars radiate as predicted , thereby validating the scheme used for iso at the longest wavelengths , or possess long - wavelength chromospheres . the iso long wavelength spectrometer ( lws ) @xcite used uranus as a calibration standard but also observed neptune , and mars when it became available . although uranus has remained the lws primary calibrator , the mars spectrum has been shown to be fully consistent with this uranus framework @xcite . in the case of neptune , a new model based on the lws data has been adopted @xcite . lws was able to observe some stellar calibrators @xcite , although these observations were difficult to make due to poor signal - to - noise in the relatively narrow lws bandwidths . however , the stellar spectra did demonstrate consistency with the uranus - mars framework , in the lws 45170 @xmath2 m region . in this paper we present the totality of results from the iso lws spectra obtained on @xmath0 tau and @xmath0 boo . the long - wavelength stellar spectral extrapolations are based upon the assumption that these calibrator stars are not attended by extensive chromospheres . radiometric testing with isophot @xcite suggests that these extrapolations were meaningful , i.e. consistent with calibrations based on asteroids and planets . it would be of great interest , both scientific and pragmatic , to determine independently whether these same stellar calibrators radiate as predicted in the millimeter ( mm ) domain . the basis for mm - wave absolute flux calibration is mars . therefore , we observed the stellar calibrators against mars , or against secondary microwave standards that were ultimately traceable to a comparison with mars , uranus , or neptune . by this means we hoped to provide a link between the reference flux density scales of planets and stars and to unify flux calibration across the ir - mm regime . early efforts in this direction at berkeley were described by @xcite . @xcite and @xcite have continued this radio absolute calibration work . our secondary goal is to probe the structure of the outer atmospheres in k and , eventually , m giants . @xcite demonstrated in a classic paper how observations of the solar fir continua , atomic lines , and uv continua could be brought together to construct a single - component , time - independent mean solar temperature structure . fir continuum fluxes were of crucial importance in helping to establish this mean structure from the upper photosphere , through the temperature minimum region , and into the chromosphere . using the fir continuum to probe the shallow depths in solar and cooler stars is possible because the primary ir continuous opacity , h@xmath5 free - free , increases as @xmath6 , pushing the depth of continuum formation into the outermost stellar layers . until now these layers in cool stars have been accessible only through non - lte analyses of the cores of strong atomic and molecular lines ( e.g. , @xcite , @xcite , @xcite , @xcite ) . while chromospheric models have become more complex in the years since the @xcite study , the fir and millimeter fluxes still have the potential to be valuable diagnostics of the shallow stellar layers . in accurately determining the fir and millimeter fluxes of selected red giants , we hope to provide useful information for testing models of the cool star stellar photospheres and chromospheres . we are particularly interested in determining the location of the transition from observed fluxes , reproducible by purely radiative equilibrium models , to fluxes that require invoking a non - radiatively heated chromosphere . in addition , we wish to trace the observed run of decreasing fluxes with increasing wavelength from the upper photosphere and determine how closely current stellar model atmosphere calculations can match this run . this information on the location of the temperature reversal and the quality of fit to the fluxes originating in the upper photosphere are vital in defining the wavelength range for which simple radiative models may be used for fir calibrations . table [ obs ] summarizes our various sets of mm observations , listing array configurations and frequencies used , integration times on the stars , phase and flux calibrators ( chosen to be as close to each star s direction in the sky as possible ) , and the primary or secondary mm - calibrator observed with each set of stellar observations . a few measurements were taken in the b - array , with which planetary calibrators like mars were partially resolved , necessitating an extra step in the calibration process to calculate the flux expected on the shortest baseline pairs ( see below ) . consequently , we preferred measurements in c- or d - array configurations . angular diameters of these stars in the optical to mir regions are of order 20 mas , rendering them unresolved targets in all these configurations , as were the local calibrators chosen . on the basis of estimates of the bima performance at 1 and 3 mm , our observing scripts called for typical tracks of several hours duration , of which between roughly two and four hours were spent on the stars themselves . this was a continuum experiment , and we selected observing frequencies from considerations of the expected stellar spectra , receiver performance , and atmospheric transmission . whenever possible we included measurements of mars with each set of stellar and local calibration source observations , in an attempt to unify the calibration . for those few data sets taken in an array configuration that resolved mars , we calculated visibilities as a function of baseline ( in kilo@xmath7 ) and applied these partial - resolution factors to the mars model @xcite flux densities using our shortest baseline pairs . sometimes no planet was visible close in time to our observing tracks . then we used mwc 349 when it was available . hat creek has a long series of observations of this star , it is relatively bright , has shown no statistically significant variability in our archives , and has been modeled in detail by @xcite so that flux extrapolation and interpolation to our frequencies is viable . further , @xcite have obtained its mm flux densities to a precision of 1% absolute . for our final observations , in december 2000 , we used w3oh as reference for our @xmath0 tau track , again because of its relative proximity to this star , long - term temporal stability of signal , and the available multifrequency hat creek archives which enable interpolation between known flux densities at other frequencies to our own . we treated w3oh as having a canonical hii region spectrum and included the frequency dependence of the gaunt factor . other secondary references such as 1415 + 133 and 0530 + 135 were likewise observed when no other obvious flux reference presented itself . independent observations of these were extracted from the hat creek archives , preferably with respect to a primary or acceptable secondary standard . in such circumstances , table [ obs ] carries the primary standard to which the flux pedigree of such a secondary is traceable . upper and lower sidebands were processed independently , maps made of the stars at each frequency , and peak flux densities extracted ( see the final column of table [ obs ] ) . sometimes , weather conditions were adequate in the lower sideband but led to unacceptably large noise in the upper sideband . in such cases we eliminated the noisy datasets which were readily recognized by the presence of artifacts in the maps , such as global striping , or large negative - going blobs . to estimate the uncertainties in the tabulated peak flux densities derived from any map , we used the standard deviation calculated from substantial regions of the map that excluded the target star itself . to combine the simultaneous upper and lower sideband maps at one epoch , we used inverse - variance weighting of the two maps based upon these noise estimators , relying upon the miriad task maths . likewise , we combined results across epochs . phase calibration of interferometer data relies on transferring phases accurately from a point source ( usually a qso ) to the target source . the accuracy of the calibration depends on the distance between source and calibrator , baseline errors , atmospheric phase stability , and the signal - to - noise ratio ( snr ) in the phase measurement of the calibrator . all sources of phase error result in decorrelation of the calibrated image ; i.e. , a lower amplitude in the final map compared with perfectly calibrated data . self - calibration offers a way to eliminate these phase errors , but is limited by the timescale on which phase can be accurately measured on the target source itself . in the case of @xmath0 boo , the measured peak flux density in the normally calibrated image is about 50 mjy . given the system temperature ( 350 k ) , bandwidth ( 670 mhz , reduced from 800 mhz due to edge - channel flagging in the wideband average ) , and a correlator efficiency of 0.88 ( 2-bit ) for these observations , the expected noise in a single - interferometer measurement is about 70 mjy in 10 min . using self - calibration with a 9-element array improves the snr for antenna - based phases by about a factor of 3 , due to baseline averaging . the expected r.m.s . noise in the antenna - based phases using self - calibration is , therefore , about 25 mjy in 10 min . assuming the true flux of @xmath0 boo to be 80 mjy , we would expect an snr of about 3 using 10-min averaging on the source . the expected phase error for an snr of 3 is @xmath8 for each antenna ( arcsin(1/3 ) ) . the decorrelation expected for an r.m.s . phase error of 20@xmath9 is about 6% ( reduced amplitude = exp(@xmath10 ) , where @xmath11=the r.m.s . phase error in radians ) . it would , therefore , appear safe to use an interval of 10 min or longer to self - calibrate the 1-mm @xmath0 boo data in phase . of course , any additional phase error ( e.g. , atmospheric ) incurred during the self - calibration interval will increase the decorrelation in the final image . we can estimate the amount of atmospheric decorrelation expected . the average r.m.s . delay path in 10 min for a 100 m baseline on the day of the @xmath0 boo 1-mm observations ( 01 jun 2000 ) was 210 @xmath2 m , or @xmath121 rad at 1360 @xmath2 m wavelength . in d - array , the maximum baseline is about 30 m. assuming that the atmospheric r.m.s . phase scales as ( baseline)@xmath13 , the r.m.s . phase becomes 0.37 rad on the longest baseline [ ( 30 m/100 m)@xmath13 * 1 rad ] , and about 0.2 rad for the median baseline of 15 m. the expected atmospheric decorrelation in 10 min for the 30 m baseline ( @xmath11=0.37 rad ) is 0.85 , and for the 15 m baseline ( @xmath11=0.2 rad ) is 0.98 . thus , atmospheric decorrelation on 10-min timescales is not expected to be very important , and phase variations on the longer timescales will be removed by the self - calibration . we validated these ideas on the june 2000 observations of @xmath0 boo by experimenting with self - calibration timescales in the reduction of the lower side - band data . the total cleaned flux in the resulting images of this star showed a spread of only 3.5% , while the more relevant quantity , the stellar peak flux densities , were 82.1 , 81.5 , and 81.1 mjy for a 15- , 20- , and 30-min timescale . we settled on a 20-min self - calibration timescale to reduce the data for @xmath0 boo , and 30 min for @xmath0 tau . the final rows in table [ obs ] present the combined peak flux density information based on the totality of the available , good quality , stellar data for @xmath0 tau and @xmath0 boo . the corresponding combined images from which these final flux densities were determined appear as figure [ bima13 ] . observations of @xmath0 tau with full wavelength coverage were made in three revolutions : 818 ( tdt 81803302 ) , 848 ( tdt 84801701 ) , and 861 ( tdt 86101101 ) . the observations of @xmath0 tau each consisted of 28 scans , 14 in each direction with a spectral oversampling of 4 . each observation had about 4630 s integration . lws made two aot observations of @xmath0 boo in revolutions 448 ( tdt 44800303 ) and 608 ( tdt 60800904 ) , each with about 2300 s integration . these observations consisted of 12 scans , 6 in each direction , taken with an oversampling of 4 . each lws spectrum was processed with offline processing version 10 ( olp10 ) and all the data were reduced using the infrared spectral analysis package ( isap ) . the reduction steps consisted of removing outliers and averaging over individual detectors at maximum resolution , clipping at 2.5@xmath14 . no attempt was made to scale the detectors to make the sub - spectra overlap in the reductions . we accomplished this outside isap using the splicing techniques described by @xcite for constructing absolutely calibrated spectra . care was taken to use the same number of scans in each direction , so the effects due to transients in the data are minimal , and no transient correction was applied . the fundamental uranus model for calibrating the lws is a synthesis of globally averaged voyager infrared interferometer spectrometer ( iris ) data up to 50 @xmath2 m and james clerk maxwell telescope data covering the 0.352 mm wavelength range @xcite with a radiative transfer model similar to the one described by @xcite . a formal off - source , full - grating , lws integration is available only for @xmath0 boo ( tdt 60801005 ) , and two fixed - grating ( sampling only ten wavelengths with no detector scans ) observations exist ( tdts 21701113 and 27202302 ) . however , it is vital to recognize that corrections for off - source emission in lws are inappropriate for faint objects like normal stars in low - cirrus regions . all observed lws spectra have had the instrumental dark currents subtracted from them during the reductions . in our olp10 reductions , we used the fixed dark currents " , as determined in dedicated calibration observations . these lws dark current measurements were essentially dark background measurements . the fabry - perot was placed in the beam and offset . therefore , there was flux on the detectors , radiated by the 3.5-k blank , but this was negligible . the signals measured remained constant throughout the mission , and were subtracted from the measured photocurrents on - source . when the dark signal , measured from the blank , was compared with the signal measured on the dark patch of sky , no difference was found . therefore , for dark sky , no additional sky " subtraction is necessary . for regions with high cirrus , there is a sky signal in the lws beam and , if a point source is being observed , an off " position is needed to determine the local sky background . in dark regions , the sky flux predicted by cobe can form a substantial fraction of the lws dark signal . however , this is deemed undetected , because no additional signal is detected over the signal from the blank . consequently , we must assess the measured sky background around the two k giants at those times when the lws spectra were taken . if those backgrounds were below the fixed dark values , then there is no need for further subtraction of off - source lws spectra . indeed , that subtraction would produce meaningless stellar spectra by removing the dark currents twice . both lws and isophot , off - source , measure the sum of zodiacal ( hereafter zodi " ) , cirrus , and extragalactic emission , as did dirbe . irsky " shows that extragalactic and cirrus confusion noises are far below zodi , so the dominant background for the two k giants is zodi . we have used two methods to evaluate the fir background near @xmath0 tau and @xmath0 boo : actual measurements in off - source pixels when these stars were observed in the fir arrays of isophot ; and , as a coarse check , dirbe predictions for the lws observations based on their solar elongation angles . the computation of isophot background sky measurements near @xmath0 tau and @xmath0 boo is summarized in tables [ ataupht ] and [ aboopht ] . these present the relevant isophot observation by filter , nominal wavelength , tdt and measurement number , and the background brightness and its 1-@xmath14 uncertainty . our conversion from point source flux to surface brightness is different from the one given in the isophot handbook @xcite . the supporting analysis and justification are given by schulz . when multiple data exist for a filter , inverse - variance weighted combinations of those data are also given . conversion from mjy sr@xmath15 to f@xmath16 ( in w @xmath17 @xmath2m@xmath18 ) was accomplished using the published values for lws detector beam solid angles , effective apertures , and bandwidths . table [ bgnd ] presents these backgrounds , and compares them with the fixed dark currents and dirbe predictions ( interpolated between 140 and 240 @xmath2 m to assess the 160-@xmath2 m values ) , matching each isophot band to the nearest lws detector . the dirbe predictions are generally in accord with isophot measurements . clearly , the measured sky backgrounds around the two stars are all below the fixed dark currents , confirming that only low level cirrus appears in their vicinities ( with the sole exception of @xmath0 boo in lw2 , where isophot suggests 50% more sky brightness than the lw2 dark current ) . the dirbe predictions are generally in accord with the isophot measurements . _ subtraction of off - source lws spectra would be inappropriate for these k giants_. figures [ ataulws ] and [ aboolws ] illustrate our reduced , spliced lws spectrum for each star , presenting the mean spectrum over all tdts , with the @xmath19 bounds for @xmath0 tau ( which has more lws data , longer integrations , and higher snr ) and @xmath20 bounds for @xmath0 boo ( with its noisier data sets ) . the sizeable triangular excursions in both spectra near 53 and 106 @xmath2 m represent the difficulty , for nir - bright normal stars , of mitigating the substantial leaks of 1.6-@xmath2 m radiation to which lws is prone . we have cleaned the observed spectra by 30-point boxcar smoothing of the data ( equivalent to about a 2-@xmath2 m interval ) . @xmath0 tau s combined spectrum contained 1181 wavelength points ; @xmath0 boo s 706 points . each figure incorporates the absolutely calibrated model continuum spectra that we have computed ( see below ) as the almost flat long - dashed lines that cross the figures , and the mean@xmath21 bounds on our older calibrated energy distributions ( dash - triple - dotted lines ) delivered in january 1996 for the calibration of isophot . to convert the model surface fluxes to predicted fluxes at the earth , we have adopted the angular diameter , 20.88@xmath220.10 mas , of @xcite ( using lunar occultation ) for @xmath0 tau and the angular diameter , 21.0@xmath220.2 mas , of @xcite for @xmath0 boo ( measured by optical intensity interferometry ) . our new , more detailed , calculations of stellar continuum spectra fall well within the uncertainties associated with the calibrated spectra originally delivered to iso . we wish to quantify the degree of accord between measured and modeled spectra in the fir for each star . the grating simultaneously illuminated the ten lws detectors , and each detector was separately wavelength and flux calibrated . therefore , lws consists of ten spectrally independent detectors . the degree of oversampling of these stellar spectra was 4 ; consequently , adjacent points in figures [ ataulws ] and [ aboolws ] are not independent . therefore , to assess the consistency of observations and predictions , and to handle the dependence of adjacent lws points in the fully - plotted spectra correctly , we include the data binned by detector as a series of bold crosses in figures [ ataulws ] and [ aboolws ] . filled circles represent the inverse - variance weighted average of all the meaningful data in each lws detector ( e.g. after the rejection of data contaminated by @xmath23-band leaks ) . error bars show the associated 1-@xmath14 uncertainties . asymmetric horizontal error bars represent the actual wavelength ranges usable for each detector and star . filled circles are plotted at the average wavelength of each detector s set of usable data for that star . the binned detector data lie within 1@xmath14 of the newly modeled energy distribution for @xmath0 tau , and within 1@xmath14 for @xmath0 boo . there is good agreement between the underpinning uranus calibration of lws , the totally independently calibrated stellar spectra made from the empirical spectra by our older model extrapolations based on the available observations between 8 and 23 @xmath2 m , and the newly calculated , absolutely calibrated , emergent stellar energy distributions . averaging the results for the independent detectors for each star with inverse - variance weighting , we determine that , for @xmath0 tau , this mean scale factor is 1.000@xmath220.011 ( over nine detectors ) while , for @xmath0 boo , this factor is 1.012@xmath220.025 ( over seven detectors ) . therefore , empirical stellar and planetary calibrations used by iso , and stellar modeling , are absolutely reconciled to 3.3% ( 3@xmath14 ) for the better measured @xmath0 tau , and 7.5% ( 3@xmath14 ) for the more poorly measured @xmath0 boo . the spectrum calculations were carried out with the source model atmosphere program in essentially the same fashion as described in @xcite . specifics pertinent to the current calculations are described in this section . for @xmath0 tau , we adopted an effective temperature of 3920 k @xcite and a log(g ) of 1.5 @xcite . for the isotopic abundances of c , n and o , we adopted log @xmath24(c ) = 8.40 , log @xmath24(n ) = 8.20 , log @xmath24(o ) = 8.78 , @xmath25c/@xmath26c = 10 , @xmath27o/@xmath28o = 562 , and @xmath27o/@xmath29o = 475 @xcite . the remaining metal abundances were chosen to be solar , following the lead of @xcite . the preceding abundances were derived using the moog spectrum synthesis code @xcite and model atmospheres @xcite constructed with the atlas ( e.g. , @xcite ) model atmosphere code and opacity sampling . our choices of t@xmath30 , log(g ) , and abundances are entirely consistent with current determinations of the parameters of @xmath0 tau . in particular , the values are well within the error bars of the values recently derived directly by decin and her collaborators @xcite from fits to stellar spectral energy distributions generated by using the marcs @xcite model atmosphere code . bertrand plez @xcite kindly computed an @xmath0 tau model atmosphere for us using the aforementioned parameters and the programs described in @xcite . thus , this model was computed with essentially the same model atmosphere program and opacities as employed by decin . for @xmath0 boo , we adopted a slightly revised version ( @xcite : hereafter pdk ) of the @xcite model atmosphere which has more depths than the original published model . this atmosphere has an effective temperature of 4300 k , log(g ) of 1.5 , [ fe / h ] = -0.5 , and other elemental abundances as determined by @xcite . the pdk model parameters derived by @xcite were derived entirely from the observed stellar spectrum . as was the case for the decin et al . investigation of @xmath0 tau , no assumptions were made regarding stellar temperature or gravity . we adopted all of the @xcite values for our study . the parameters t@xmath30 , gravity , microturbulent velocity , and [ fe / h ] derived by @xcite for @xmath0 boo are essentially identical to those derived later and independently by @xcite , while the c , n , and o abundances agree within the error bounds . the alert reader will observe that we have adopted model atmospheres for @xmath0 tau and @xmath0 boo that were derived using entirely different model atmosphere codes ( atlas and marcs ) . we shortly will describe results obtained by using these atmospheres as input to a third model atmosphere code , our source . we do not feel that significant improvement would result from insisting that _ only _ atlas _ or _ marcs models be used in our study . all three codes have been vetted against the solar case and should give very comparable results in the g - k star domain . we cite as support for this stance the striking agreement between the results of @xcite and @xcite for @xmath0 boo . this occurs despite @xcite having used a completely different model atmosphere program and blanketing representation ( marcs with opacity sampling ) to compute their @xmath0 boo model than used by @xcite ( atlas with opacity distribution functions ) . the agreement is even more impressive when one considers that these two investigations derived their results from two entirely different spectral regions , 0.50.9 @xmath2 m _ versus _ 2.3812.0 @xmath2 m . we regard this as evidence that different well - vetted model atmosphere codes and opacity representations now produce comparable model atmospheres and visible - infrared spectra for k - giants * _ within the limited approximations _ * of static , homogeneous atmospheres , lte , and mixing - length - theory radiative / convective equilibrium . that does * _ not _ * mean that the resultant models are accurate , or even adequate , representations of a real star . differences between observed and computed spectra are now much more likely to arise from the presence of physics ( _ e.g. _ , nlte , non - radiative heat deposition , and inhomogeneous dynamical structure ) * _ not treated _ * in the simple models we use here . we will discuss this point in more detail as we proceed . the column mass ( g @xmath17 ) _ vs. _ temperature relations for both the plez @xmath0 tau and pdk @xmath0 boo models were linearly extrapolated to shallower depths as required to obtain optical transparency at the longest wavelengths . for the plez @xmath0 tau model , the extrapolated layers are shallower than 0.020 g @xmath17 ; for the pdk @xmath0 boo model , shallower than 0.010 g @xmath17 . the resultant model atmospheres are shown in figure [ atau_aboo_models ] . it should be stressed that both of the above adopted model atmospheres have purely radiative equilibrium surface temperature structures . specifically , they do _ not _ have chromospheric temperature rises , a point whose importance will become evident later . we have included in figure [ atau_aboo_models ] , for comparison , the @xcite ( kel ) @xmath0 tau chromospheric model b and the @xcite ( al ) @xmath0 boo chromospheric model . these chromospheric models are semiempirical , derived from non - lte analyses of and line fluxes . while these chromospheric models are now over 20 years old , they are very similar from the upper photosphere , through the temperature minimum region , and into the lower chromosphere to models of @xmath0 tau and @xmath0 boo used by @xcite , and the @xmath0 tau model derived by @xcite . using the plez @xmath0 tau atmosphere and parameters detailed above , we have computed the synthetic spectrum for @xmath0 tau over the entire lws range and beyond , 43 to 665 @xmath2 m . the spectrum was computed using a full set of molecular line opacities which includes all the principal isotopomers of co , sio , oh , and @xmath31 and using a variable wavelength mesh which locally provides @xmath322.5 points per doppler width . to allow comparison with the lws observations , we convolved the high - resolution result with a gaussian instrumental profile function having a fwhm equal to 0.29 @xmath2 m for @xmath33 90.5 @xmath2 m , and 0.60 @xmath2 m , @xmath34 90.5 @xmath2 m ( e.g. , @xcite ) . the resulting spectrum , normalized to the continuum and on a greatly magnified ordinate scale , is shown in figure [ atau_spectrum ] . a detailed description of the displayed spectrum may be found in @xcite . we do note here that almost all the absorption beyond @xmath32 50 @xmath2 m is due to the isotopomers of sio . at shorter wavelengths the principal contributor is oh with some h@xmath35o and co as well . it is clear that , _ at the iso lws and the bima spectral resolutions _ , all differences between the pure continuum and the synthetic spectrum computed with the full set of molecular line opacities are less than 2% ; beyond 130 @xmath2 m , they are less than 1% . this is _ much less _ than the noise level in the observed @xmath0 tau spectra in this region . as a consequence , there is no significant advantage gained by using the full synthetic spectrum . for clarity in our presentation , we will use only the continuous spectrum in our comparisons with the @xmath0 tau observations . since @xmath0 boo is both hotter and has lower metallicity , we shall use the continuum alone for comparisons there as well . for both the plez @xmath0 tau and the pdk @xmath0 boo models , the dominant continuous opacity in the continuum - forming layers for wavelengths redward of the h@xmath5 opacity minimum at 1.64 @xmath2 m arises from h@xmath5 free - free . this opacity increases in strength as @xmath6 . as a result , the emergent continuum flux is formed at progressively shallower layers in the stellar atmosphere as one proceeds to longer wavelengths . figure [ atau_aboo_models ] illustrates this point for the plez @xmath0 tau and the pdk @xmath0 boo models . in this figure we have marked , for the radiative equilibrium models , the atmospheric depths for which the continuum optical depth , @xmath36 , reaches unity at the indicated wavelengths . that is , we have marked the column masses at which @xmath36 = 1.0 ; roughly speaking , in the fir , depths near this value and shallower make the greatest contribution to the emergent continuum flux at wavelength @xmath7 . note , for example , that the continuum fluxes between 1 mm and 3 mm are sampling very different parts of the atmospheric temperature structure in both radiative equilibrium models than , say , the continuum fluxes at 2.3 @xmath2 m . indeed , the longest wavelengths are sampling , _ in the continuum _ , atmospheric levels reached only at the cores of lines in bluer portions of the spectrum . only when one proceeds to the vacuum ultraviolet and its strong photoionization continua of metals is it possible to reach in the continuum the shallow atmospheric levels probed by the continuum at long wavelengths ( e.g. figure 1 of @xcite ) . thus , the fir continuum is an effective probe of the shallowest stellar layers which can supplement information derived from line cores and uv continua . as noted above , there is very little difference between the full synthetic spectrum and the spectrum of the continuum . figure [ atau_aboo_models ] provides an important clue as to why spectral lines generally become less prominent at longer wavelengths . it is evident that , as one goes to shallower depths in the models , the slope of the temperature - column mass relation becomes smaller . the significance of this can be appreciated by considering an absorption line in the @xmath0 tau model with a characteristic , depth - independent , line core / continuum opacity ratio of 1000 . if the absorption line occurs at a wavelength of 2.3 @xmath2 m , the continuum will be formed at @xmath325200 k and the line core will be formed at @xmath323200 k. the same line at a wavelength of 200 @xmath2 m will have the continuum formed at @xmath323100 k and the line core at @xmath322600 k , a much smaller temperature difference between line core and continuum than at 2.3 @xmath2 m . couple this with the decreasing temperature sensitivity of the planck function as one goes to longer wavelengths and one has a recipe for progressively less prominent spectral features . figure [ mass_bt ] provides another window into the wavelength dependence of continuum formation . in the upper panel of figure [ mass_bt ] we show the column masses for which @xmath36 = 1.0 over the spectral range 0.5 @xmath2 m to 200 @xmath2 m . as noted above , there is a steady march to progressively shallower depths of continuum formation as one goes to longer wavelengths . since the dominant h@xmath5 bound - free opacity peaks at roughly 0.825 @xmath2 m , continuum fluxes at wavelengths between 0.825 @xmath2 m and 1.64 @xmath2 m probe the same atmospheric layers as continuum fluxes between 1.64 @xmath2 m and 6.5 @xmath2 m . longward of 6.5 @xmath2 m , the continuum flux probes atmospheric layers accessible only in line cores and in the metal bound - free continua of the vacuum ultraviolet . in the lower panel of figure [ mass_bt ] we show the continuum brightness temperature " over the same spectral range for both models . the brightness temperature is defined as the temperature of a blackbody which gives the same flux as the model atmosphere at the indicated wavelength ( a precise definition is given in section [ allbt ] ) . three points are important . first , neither @xmath0 tau nor @xmath0 boo could have its continuum fluxes over any large spectral range matched by a blackbody of a single temperature . second , the blackbody temperatures that do match the model fluxes at specific wavelengths are generally quite different than the effective temperatures of the models ( or the stars ) themselves . third , comparing with figure [ atau_aboo_models ] , one sees that the brightness temperature is systematically smaller than t(@xmath36 = 1.0 ) , the temperature of the atmospheric layer at which optical depth unity is reached in the continuum at wavelength @xmath7 . ( for an atmosphere with a linear source function at wavelength @xmath7 , the brightness temperature at @xmath7 and t(@xmath36 = 2/3 ) would be identical . unfortunately , in the particular cases of the pdk and plez models , the fir source functions are not linear in @xmath36 but , rather , quadratic . ) the differences between t(@xmath36 = 1.0 ) and brightness temperature vary systematically with wavelength . the differences are greatest at 1.64 @xmath2 m and decrease to the red . _ beyond 40 @xmath2 m , the differences are always less than 100 k. _ ( had we chosen for illustration the more correct , but _ still _ approximate , t(@xmath36 = 2/3 ) , the temperature differences would be smaller yet . ) thus , the brightness temperature is a good first guess at t(@xmath36 @xmath32 1.0 ) at the longer wavelengths . later , when we discuss differences between observed and predicted brightness temperatures , the reader may use this approximation and figure [ atau_aboo_models ] to roughly estimate the implied errors in the radiative equilibrium model temperature structures . table [ compare ] shows the comparison between the observed fluxes and the predictions of the radiative equilibrium models . only for the case of @xmath0 tau at 1.4 mm does there appear to be approximate agreement between the observed flux and the flux predicted from the radiative equilibrium model . the remainder of the observed fluxes are all significantly _ larger _ than the fluxes predicted by the models . the discrepancies are particularly striking in the case of @xmath0 boo . we do not believe that the flux computation for the radiative equilibrium models could have errors sufficiently large to produce the marked differences between computed and observed millimeter fluxes . our model atmosphere code has been carefully vetted and the dominant continuous opacities are themselves quite well determined ( e.g. , @xcite ) . ( as an example of evidence for this , we refer the reader forward in the text to figure 11 . there we show the continuum for the plez @xmath0 tau model computed with an entirely independent model code . ) on the observational side , our derived errors for the observed 1.4-mm and 2.8-mm fluxes preclude dismissing the differences between observed and computed fluxes as simply due to measurement error . we believe other explanations must be considered . since the adopted angular diameters for these two stars appear to be extremely well determined , with standard deviations of @xmath321% or less , they would not seem to be a significant source of error . moreover , a major change in the angular diameters would compromise flux agreement for the @xmath33 200 @xmath2 m region . nevertheless , it should be noted that the @xcite diameter is an arithmetic mean of values measured from 0.45 @xmath2 m to 2.2 @xmath2 m and the @xcite diameter is a error - weighted mean biased toward values obtained at 1.6 @xmath2 m and 3.8 @xmath2 m . at any particular monochromatic wavelength , an observed angular diameter represents the apparent diameter of the @xmath37 @xmath32 1.0 surface of the star . since , as shown in figures [ atau_aboo_models ] and [ mass_bt ] , there is a progressive march in depth of continuum formation to ever shallower atmospheric layers as one moves redward of 1.64 @xmath2 m , the angular diameters themselves should increase with wavelength . if the angular diameters at 1.4 mm and 2.8 mm are substantially larger than those at the wavelengths of the @xcite and @xcite observations , our predicted fluxes in table [ compare ] could be seriously underestimated . we can roughly estimate the magnitude of this source of error , at least in the case of the radiative equilibrium models . in table [ extents ] we show the physical radii of @xmath0 tau and @xmath0 boo based on the indicated angular diameters and hipparcos parallaxes @xcite . we also show the physical distances in the radiative equilibrium models between @xmath38 = 1.0 and @xmath39 = 1.0 . given the tabulated numbers , it is clear that , for the pdk and plez models , the atmospheric extent in the continuum at 2.8 mm is not sufficient to seriously affect our predicted fluxes . we find a similar result in the case of the only model chromosphere with a published height scale , mcmurry s chromospheric model for @xmath0 tau @xcite . the mcmurry atmosphere is plane - parallel and hydrostatic , constructed with log g = 1.25 . note that this gravity leads to a more distended atmosphere than does the higher gravity we adopted . the mcmurry atmosphere has an extent of @xmath322.6 @xmath40 10@xmath41 cm between the layers where the 1.64 @xmath2 m continuum is formed and the layer with t = 7120 k in the chromosphere . this extent is only 8% of the stellar radius and again seems too small to explain the discrepancy between computed and observed fluxes . finally , as we will describe in section [ allbt ] , @xcite determined that @xmath0 boo has essentially the photospheric radius at 2 cm . given the expected increase in angular diameter with wavelength , this would imply that @xmath0 boo would have the photospheric radius at @xmath42 2 cm as well . based on the above arguments , we believe that the flux excesses observed at 1.4 mm and 2.8 mm , relative to the predictions of the purely radiative models , are real and significant . furthermore , as we shall show , there is good evidence that these excesses are signatures of the chromospheric temperature rise in the two red giants . figures [ ataummcm ] and [ aboommcm ] present our millimeter flux densities ( filled circles ) and the color - corrected iras measurements of each star at 12 , 25 , 60 , and 100 @xmath2 m ( crosses ) . each figure includes the plez @xmath0 tau or pdk @xmath0 boo spectrum , as appropriate , from 10 @xmath2 m to 3 mm . there are additional high - frequency radio continuum observations of these two k giants in the literature . @xcite made single - dish 86-ghz measurements of @xmath0 boo ( 21.4@xmath227.5 mjy ) with the iram 30-m telescope , while @xcite likewise detected both @xmath0 tau and @xmath0 boo at 250 ghz , obtaining 51@xmath226 and 78@xmath228 mjy , respectively . there are also vla centimeter measurements on both stars at 4.9 , 8.4 , and 14.9 ghz by @xcite , and unpublished data by drake , linsky , & judge from the compilation of radio data by @xcite . these long wavelength measurements all have substantial uncertainties , in part due to the faintness of the stars , in part to the potential for confusion with extragalactic background objects . nonetheless , in combination with our own radio data ( figures [ ataummcm ] , [ aboommcm ] ) one sees that these high - frequency data ( open diamonds ) are in good agreement with our aperture synthesis measurements ( filled circles ) . it is difficult to fully appreciate the significance of the excess fluxes from the logarithmic scales of figures [ ataummcm ] and [ aboommcm ] . an insightful alternative view may be obtained by combining the mm and cm fluxes from other observers with our data and looking at the whole dataset as brightness temperatures over a broad spectral range . in figures [ atauallbt ] and [ abooallbt ] we show the brightness temperatures corresponding to all the observed flux data between 43 @xmath2 m and 6 cm used in our previous figures . for comparison we have also plotted the predictions of the radiative equilibrium models . in deriving brightness temperatures from the observed fluxes , we used the angular diameters of table [ extents ] in the following relation : @xmath43 where @xmath44 is the observed flux in jy , @xmath0 is the angular diameter in mas , @xmath7 is the wavelength in microns , and @xmath45 is in kelvins . the error bars in figures [ atauallbt ] and [ abooallbt ] correspond _ only _ to the uncertainties in the observed fluxes . in the case of the @xmath0 tau data at 2 cm and 3.6 cm , no error values were reported by @xcite and we used our own rough estimates of the plausible errors . before proceeding , a caveat needs to be noted regarding the longer wavelength points especially . the observed fluxes presumably come from progressively shallower stellar layers as @xmath7 increases . all of the brightness temperatures in figures [ atauallbt ] and [ abooallbt ] have been deduced assuming the photospheric angular diameters of table [ extents ] . atmospheric extension may play a role in the chromospheric layers of @xmath0 tau and @xmath0 boo . @xcite estimated the extension of the chromospheres of these stars . they argued that the uv 0.01 lines which they studied were formed in chromospheric layers with temperatures of @xmath3210000 k in @xmath0 tau and @xmath328000 k in @xmath0 boo . they deduced that these chromospheric layers had radii of @xmath322.0 r@xmath46 and @xmath321.4 r@xmath46 , respectively , where r@xmath46 is the stellar photospheric radius . although these estimates are , in the authors words , `` relatively crude '' , they do suggest that atmospheric extension may play a role at least in the hotter layers of the chromosphere . different evidence of extension was found by @xcite in their study of 2 cm and 6 cm fluxes . they treated all the radio emission as coming from an optically thick ionized wind for which a `` half - power radius '' could be deduced . in the case of @xmath0 boo they found the 2 cm half - power radius was @xmath32r@xmath46 ; at 6 cm , the half - power radius had grown to @xmath321.7 r@xmath46 . taking the long wavelength approximation to the brightness temperature expression above , we see that , for a given observed flux , @xmath47 , where @xmath0 is the angular diameter . since the shallowest layers may have angular diameters significantly larger than those of the photosphere , the corresponding brightness temperatures in figures [ atauallbt ] and [ abooallbt ] should be regarded as upper limits . we now describe why we believe that the flux excesses at the longer wavelengths arise from the presence of stellar chromospheres in @xmath0 boo and @xmath0 tau . circumstantial evidence suggests the possibility that stellar chromospheres could explain the flux excesses . this evidence comes by examining the depths at which @xmath36 = 1.0 . in the case of @xmath0 boo , as shown in figure [ atau_aboo_models ] , the al chromospheric model departs from the purely radiative equilibrium model at a column mass of @xmath32 0.75 g @xmath17 @xcite . this is just in the column mass range of the pdk radiatve equilibrium model where optical depth unity is reached between 200 @xmath2 m and 1 mm . given this , it is not surprising that the pdk model might significantly _ underpredict _ the flux at both 1.4 mm and 2.8 mm in @xmath0 boo . we believe that the progressively larger differences between the predictions of the pdk model and the observations beyond 120 @xmath2 m are due to the contributions of @xmath0 boo s chromosphere . the circumstantial evidence in the case of @xmath0 tau is not so clear . as shown in figure [ atau_aboo_models ] , the kel @xmath0 tau chromospheric model begins its temperature rise at a column mass of @xmath480.3 g @xmath17 @xcite . this is slightly _ shallower _ in the plez @xmath0 tau model than where @xmath49 = 1.0 occurs . nevertheless , the 2.8 mm observed brightness temperature is distinctly higher than that predicted by the plez model in figure [ atauallbt ] . in fact , with the exception of our 1.4 mm point , the general run of observed points beyond 200 @xmath2 m suggests that some enhanced ( chromospheric ? ) emission may be present at 1 mm or even shorter wavelengths . whether this is true or not will take additional observations in this spectral region with better snr . while nlte chromospheric modeling is well beyond the intent of this investigation , one of us ( adm ) has computed the infrared fluxes predicted by his chromospheric model of @xmath0 tau . details of the model and the nlte computation may be found in @xcite . this model is essentially _ identical _ to the kel model from the photosphere through the lower and middle chromosphere , differing from the kel model only at column masses shallower than 10@xmath50 g @xmath17 ( see fig . 5 , @xcite ) . the nlte radiative transfer computation was carried out with the independent code multi . we show in figure [ atau_nlte ] the results of this nlte computation with a chromospheric model compared with the observations , and with the predictions of the plez model . it is immediately apparent that , again with the exception of our 1.4 mm point , the observed data beyond 200 @xmath2 m is better fit by the nlte chromosphere model than the lte plez radiative equilibrium model . we believe this is good evidence that the stellar chromosphere of @xmath0 tau has a significant impact on the observed fluxes beyond 200 @xmath2 m . indeed , it should be noted that the fit is even improved between 100 @xmath2 m and 200 @xmath2 m . for @xmath0 boo ( figure [ abooallbt ] ) it appears that the transition from purely radiative occurs beyond 120 @xmath2 m , and certainly well before 1.4 mm . in our opinion this makes any purely radiative model for @xmath0 boo a poor choice for calibration in the sub - mm spectral region . this appears to be the case for @xmath0 tau as well , especially considering the predictions of the mcmurry chromospheric model ( figure [ atau_nlte ] ) . given the evidence described above , we do not believe that it is prudent to use cool star radiative models for calibration purposes at the longer wavelengths _ unless alternatively calibrated observations already confirm _ the validity of the purely radiative models . one must be certain that the flux predicted by the purely radiative model at the long wavelength end of the region being calibrated by the theoretical stellar spectrum does indeed correspond to the actual star . in other words , we believe that when one is approaching the spectral regions where a chromospheric contribution can be significant , one can safely use purely radiative cool star models only as an interpolation tool for calibrations , not as an extrapolation tool . we would like to emphasize the importance of the spectral region from @xmath3230 @xmath2 m to several millimeters to furthering an understanding of cool stellar atmospheres . fluxes from this spectral region reflect the mean time- and spatial - averaged emission from the upper photosphere and chromosphere for stars like those we consider here ( see figure [ atau_aboo_models ] ) . accurate fluxes measured through this spectral region should allow one to test directly model atmosphere structures , even complex ones that include nlte and dynamics , over the very interesting depth range in which non - radiative processes like shocks begin to alter the structure . by selectively sampling progressively longer wavelengths , one can follow the important transition from photosphere to chromosphere , locating the depth at which purely radiative models like the pdk and plez models begin to fail . our original , theoretically extrapolated , continuum spectra for the two k giants , supplied to isophot as absolutely calibrated energy distributions , are in extremely good agreement with our new calibrated synthetic spectra ( figures [ ataulws ] , [ aboolws ] ) , well within the @xmath226% uncertainties assigned to the original calibrators @xcite . the lws spectra , reduced with respect to uranus as calibrator , are consistent with the new ( and old ) synthetic spectra , thereby uniting fir planetary and stellar calibrations at the few percent level ( figures [ ataulws ] , [ aboolws ] ) . there are independent data on @xmath0 boo , derived from the validation of isophot s final pipeline ( olp10 ) @xcite , that show it conforms to our 1996 model spectrum . in addition , @xcite has described the performance through the iso mission of a significant number of isophot calibrators . by removing the contributions made by any particular source to the calibration process , and calibrating that source with respect to all other calibrators , one can assess the ratio of observed to modeled irradiance . using the database described by @xcite , we have been able to extend the range over which our new model of @xmath0 boo can be tested , and to validate @xmath0 tau independently of our lws data . for @xmath0 boo , we have determined the observed to modeled irradiance ratios for the isophot c100 , c120 , c135 , c160 , c180 , and c200 filters ( c " denotes the fir camera within the isophot instrument , and each filter is designated by its reference wavelength in @xmath2 m ) . expressing these observed irradiance ratios with respect to our new modeled spectrum for @xmath0 boo we find : c100 , 1.02@xmath220.02 ; c120 , 1.14@xmath220.08 ; c135 , 1.01@xmath220.02 ; c160 , 1.06@xmath220.02 ; c180 , 1.00@xmath220.03 ; c200 , 1.00@xmath220.26 . for @xmath0 tau , measurements are available solely in the c160 filter and yield a ratio of 0.96@xmath220.05 , with respect to our new model . we conclude that the separate absolute calibrations of lws and isophot are in accord within the uncertainty of the observations , despite their different dependences on planets , asteroids , and stars . this should not be interpreted to mean that more accurate observations might not reveal discrepancies between fir observations and the predictions of static , lte model atmospheres computed in simple radiative / convective equilibrium . our primary objective was to investigate whether purely radiative models for two normal k giants could reproduce the observed fir and microwave fluxes . we have found that for both @xmath0 tau and @xmath0 boo the radiative equilibrium temperature structures of the upper photospheres produce fluxes in agreement with the observations , but only up to a certain point . for @xmath0 boo and @xmath0 tau the agreement extends to at least 125 @xmath2 m , corresponding to a column mass of @xmath325 gm @xmath17 in the pdk atmosphere and @xmath3210 gm @xmath17 in the plez atmosphere . both k giants clearly show definite chromospheric emission at 2.8 mm . in @xmath0 boo this is evident by 1.4 mm . only additional observations can pinpoint precisely where between 125 @xmath2 m and 2.8 mm the chromospheric contribution becomes significant . for this reason , we can not rely on purely radiative model atmospheres to form a basis for stellar calibration in the sub - mm region . purely radiative models of the continuum might be used to provide calibrations _ between _ those shorter fir wavelengths where the validity of the model fluxes already has been independently verified by observations . we wish to thank bertrand plez for computing an @xmath0 tau model for our particular set of input parameters and ruth peterson for the extended version of the @xcite @xmath0 boo model . their help is greatly appreciated . it is a pleasure to thank ullie klaas and phil richards of the isophot team for sending us data on their validation of olp10 as it bears on @xmath0 boo s spectrum , and sumita jayaraman for assistance in extracting dirbe data for the elongation angles at the times of lws and isophot observations of these k giants . mc acknowledges support for this work through subcontract uf99025 between the university of florida and uc berkeley , supporting the calibration of isophot . dfc gratefully acknowledges access to the resources of the nasa advanced supercomputing ( nas ) division , nasa / ames research center , moffett field , ca , as well as the support of the nas division chief walter brooks and nas branch chiefs dochan kwak and william w. thigpen . sgi / cray c90 supercomputer cycles for this research were provided through the consolidated supercomputing management office ( cosmo ) at nasa / ames . this material is based upon work by dfc and dg supported by the national aeronautics and space administration under nra-99 - 01-adp-073 issued through the astrophysics data program . we acknowledge support for bima and carma through nsf grants ast-9981308 and ast-0228963 . altenhoff , w. j. , huchtmeier , w. k. , schmidt , j. , schraml , j. b. , & stumpff , p. 1986 , , 164 , 227 altenhoff , w. j. , thum , c. , & wendker , h. j. 1994 , , 281 , 161 ayres , t.r . , brown , a. , harper , g.m . , bennett , p.d . , linsky , j.l . , carpenter , k.g . , & robinson , r.d . 1997 , , 491 , 876 ayres , t.r . & linsky , j.l . 1975 , , 200 , 660 blackwell , d.e . , lynas - gray , a.e . , & petford , a.d . 1991 , , 245 , 567 van der bliek , n.s . , gustafsson , b. & eriksson , k. 1996 , , 309 , 849 carbon , d.f . , langer , g.e . , butler , d. , kraft , r.p . , trefzger , ch.f . , suntzeff , n.b . , kemper , e. , & romanishin , w. 1982 , , 49 , 207 carbon , d.f . , idehara , h. , & goorvitch , d. 2005 , aj , in preparation carpenter , k.g . , brown , a. , & stencel , r.e . 1985 , , 289 , 676 clegg , p. e. et al . 1996 , , 315 , l38 cohen , m. 1998 , , 115 , 2092 cohen , m. & davies , j. k. 1995 , mnras , 276 , 715 cohen , m. , megeath , t.g . , hammersley , p.l . , martin - luis , f. , & stauffer , j. 2003 , , 125 , 2645 cohen , m. , wheaton , w.a . , & megeath , t.g . 2003 , , 126 , 1090 cohen , m. , walker , r.g . , barlow , m.j . , & deacon , j.r . , 1992a , , 104 , 1650 cohen , m. , walker , r.g . , carter , b. , hammersley , p. , kidger , m. , & noguchi , k. , 1999 , , 117 , 1864 cohen , m. , walker , r.g . , jayaraman , s. , barker , e. , & price , s.d . 2001 , , 121 , 1180 cohen , m. , walker , r. g. , & witteborn , f. c. 1992b , , 104 , 2030 cohen , m. , witteborn , bregman , j. d. , wooden , d.h . , salama , a. , and metcalfe , l. 1996 , , 112 , 241 cohen , m. , witteborn , f. c. , carbon , d. f. , augason , g. c. , wooden , d.h . , bregman , j. d. , & goorvitch , d. 1992c , , 104 , 2045 cohen , m. , witteborn , f. c. , carbon , d. f. , davies , j. k. , bregman , j. d. , & wooden , d. h. 1996b , , 112 , 2274 cohen , m. , witteborn , f.c . , walker , r.g . , bregman , j.d . , & wooden , d.h . 1995 , , 110 , 275 decin , l. , waelkens , c. , eriksson , k. , gustafsson , b. , plez , b. , sauval , a.j . , van assche , w. , & vandenbussche , b. 2000 , , 364 , 137 decin , l. , vandenbussche , b. , waelkens , c. , decin , g. , eriksson , k. , gustafsson , b. , plez , b. , & sauval , a.j . 2003 , , 400 , 709 drake , s. a. & linsky , j. l. 1986 , , 91 , 602 dreher , j.w . & welch , w.j . 1983 , , 88 , 1014 esa 1997 , the hipparcos and tycho catalogues , esa sp-1200 , or http://astro.estec.esa.nl/hipparcos/catalog.html gibson , j. , welch , w. j. , & de pater , i. 2005 , icarus , in press griffin , m. & orton , g. 1993 , icarus , 105,537 gry , c. et al . 2000 , in the iso handbook , vol . iii : lws - the long - wavelength spectrometer " ( esa sp-1262 ) gustafsson , b. , bell , r. a. eriksson , k. , & nordlund , a. 1975 , , 42 , 407 harris , m.j . & lambert , d.l . 1984 , , 285 , 674 johnson , h. r. bernat , a. p. krupp , b. m. 1980 , , 42 , 501 laureijs , r. j. , klaas , u. , richards , p. j. , schulz , b. , & abraham , p. 2003 , the iso handbook vol . iv : pht - the imaging photo - polarimeter " ( esa sp-1262 ) kelch , w.l . , linsky , j.l . , basri , g.s . , chiu , h .- y . , chang , s .- h . , maran , s.p . , & furenlid , i. 1978 , , 220 , 962 kessler , m.f . 1996 , , 315 , l27 kurucz , r. 1993 , atlas9 stellar atmosphere programs 2 km / s grid . kurucz cd - rom cambridge , mass . : smithsonian astrophysical observatory , 1993 lemke , d. et al . 1996 , , 315 , l64 lim , t. , swinyard , b.m . , liu , x .- w . , burgdorf , m. , gry , c. , pezzuto , s. , & tommasi , e. 1997 , first iso workshop on analytical spectroscopy " , eds . a.m. heras , k. leech , n.r . trams & m. perry , esa sp-419 , pg . 281 linsky , j.l . , & haisch , b.m . 1979 , , 229 , l27 mcmurry , a.d . 1999 , mnras , 302,37 mill , j. et al . 1994 , aiaa , 31 , 900 mitchell , k. j. et al . 1996 , in unveiling the cosmic background " , ed . e. dwek ( aip , cp 348 ) , p.301 muller , t.g . & lagerros , j.s.v . 1998 , , 338 , 340 murakami , h. et al . 1996 , pas japan , 48 , l41 orton , g. , burgdorf , m. , davis , g. , sidher , s. , griffin , m. , swinyard , b. , & feuchtgruber , h. 2000 , aas , dps meeting # 32 , # 11.06 peterson , r.c . 1995 , private communication peterson , r.c . , dalle ore , c.m . & kurucz , r.l . 1993 , , 404 , 333 . plez , b. 1999 , private communication . plez , b. , brett , j.m . & nordlund , a. 1992 , , 256 , 551 . price , s.d . , paxson , c. , engelke , c. & murdock , t.l . 2004 , , 128 , 889 quirrenbach , a. , mozurkewich , d. , buscher , d.f . , hummel , c.a . , & armstrong , j.t . 1996 , , 312 , 160 richards , p.j . 2003 , in the calibration legacy of the iso mission " , eds . l. metcalfe , a. , salama , s.b . peschke , m.f . kessler ( esa sp-481 ) , p.279 ridgway , s.t . , jacoby , g.h . , joyce , r.r . , siegel , m.j . , & wells , d.c . 1982 , , 87 , 1044 schaeidt , s. et al . 1996 , , 315 , l55 schulz , b. 2003 , in the calibration legacy of the iso mission " , eds . l. metcalfe , a. , salama , s.b . peschke , m.f . kessler ( esa sp-481 ) , p.83 schulz , b. et al . 2002 , , 381 , 1110 sidher , s.d . , swinyard , b m. , griffin , m.j . , orton , g.s . , burgdorf , m. , & lim , t.l . 2003 , the calibration legacy of the iso mission " , eds . l. metcalfe , a. , salama , s.b . peschke , m.f . kessler ( esa sp-481 ) , p.153 smith , v.v . & lambert , d.l . 1985 , , 294 , 326 smith , v.v . & lambert , d.l . 1990 , , 72 , 387 sneden , c. 1973 , ph.d . thesis , univ . of texas , austin swinyard , b. et al . 1996 , , 315 , l43 vernazza , j. e. , avrett , e. h. , & loeser , r. 1976 , , 30 , 1 welch , w. j. , & gibson , j. 2005 , in preparation wendker , h. 1995 , vizier catalog ii/199a wiedemann , g. , ayres , t. y. , jennings , d. e. , & saar , s. h. 1994 , , 423 , 806 wright , e. l. 1976 , , 210 , 250 wrixon , g. t. , welch , w. j. , & thornton , d. t. 1971 , , 169 , 171 cccccccccc @xmath0 tau & 14nov97 & c & 106.6845 & 4.32 hr & 0449 + 113 & 3c111 & mars , & & 17.14@xmath222.19 + & & & & & & & uranus , neptune & & + @xmath0 boo & 15nov97 & c & 106.6845 & 4.22 hr & 1415 + 133 & 3c111 & mars , & & 20.27@xmath222.95 + & & & & & 1357 + 193 & & uranus , neptune & & + @xmath0 boo & 15nov97 & c & 110.1282 & 4.22hr & 1415 + 133 & 3c111 & mars , & & 18.47@xmath222.73 + & & & & & 1357 + 193 & & uranus , neptune & & + @xmath0 boo & 15nov97 & c & both & & & & & & 19.73@xmath222.16 + @xmath0 tau & 06jun98 & c & 106.6845 & 2.99hr & 0449 + 113 & 3c111 & mars & & 13.10@xmath223.40 + @xmath0 boo & 11mar99 & b & 106.6845 & & 1357 + 193 & 1415 + 133 & & mars & 24.57@xmath221.20 + @xmath0 boo & 11mar99 & b & 110.1282 & 4.37 hr & 1357 + 193 & 1415 + 133 & & mars & 19.68@xmath221.23 + @xmath0 boo & 11mar99 & b & both & & & & & & 22.03@xmath220.97 + @xmath0 boo & 12mar99 & b & 106.6845 & 3.13 hr & 1357 + 193 & 1415 + 133 & mwc 349 , mars & mars & 21.02@xmath221.74 + @xmath0 boo & 12mar99 & b & 110.1282 & & 1357 + 193 & 1415 + 133 & mwc 349 , mars & mars & 18.90@xmath221.20 + @xmath0 boo & 12mar99 & b & both & & & & & & 19.92@xmath221.02 + @xmath0 tau & 05sep99 & d & 216.0982 & 1.85 hr & 0530 + 135 & 0530 + 135 & & mwc 349/mars & 24.23@xmath226.87 + @xmath0 tau & 05sep99 & d & 219.5418 & & 0530 + 135 & 0530 + 135 & & mwc 349/mars & 27.34@xmath228.12 + @xmath0 tau & 05sep99 & d & both & & & & & & 25.78@xmath225.64 + @xmath0 boo & 01jun00 & d & 216.0982 & 2.91 hr & 3c273 & 3c273 & mwc 349 & mars & 81.55@xmath222.35 + @xmath0 boo & 01jun00 & d & 219.5418 & 2.91 hr & 3c273 & 3c273 & mwc 349 & mars & 85.45@xmath222.44 + @xmath0 boo & 01jun00 & d & both & & & & & & 83.50@xmath221.71 + @xmath0 tau & 09dec00 & c & 106.6845 & 3.19 hr & 3c111 & 3c111 & w3oh , mars , & & 12.55@xmath222.16 + & & & & & & & uranus & & + @xmath0 tau & 09dec00 & c & 110.1282 & 3.19 hr & 3c111 & 3c111 & w3oh , mars , & & 12.77@xmath222.95 + & & & & & & & uranus & & + @xmath0 tau & 09dec00 & c & both & & & & & & 12.40@xmath221.91 + @xmath0 tau & all & all & 108.40 & & & & & & 13.97@xmath221.46 + @xmath0 tau & all & all & 217.82 & & & & & & 25.78@xmath225.64 + @xmath0 boo & all & all & 108.40 & & & & & & 20.09@xmath220.69 + @xmath0 boo & all & all & 217.82 & & & & & & 83.50@xmath221.71 + lcccc c1 - 100 & 100 & 86401601 & 2 & 42.32@xmath223.57 + c1 - 100 & 100 & 86401602 & 2 & 44.53@xmath222.68 + c2 - 160 & 170 & 86002102 & 1 & 43.06@xmath224.39 + c1 - 60 & 60 & 84300501 & 5 & 34.54@xmath224.68 + c1 - 100 & 100 & 84300501 & 2 & 36.69@xmath224.14 + c1 - 60 & 60 & 84300501 & 5 & 40.89@xmath225.53 + c1 - 60 & 60 & combined & & 37.19@xmath223.57 + c1 - 100 & 100 & combined & & 42.24@xmath221.90 + lcccc c1 - 50 & 50 & 27501511 & 1 & 18.16@xmath221.42 + c1 - 90 & 90 & 27502117 & 1 & 14.62@xmath220.87 + c1 - 105 & 105 & 27500602 & 1 & 12.54@xmath220.54 + c1 - 105 & 105 & 27500602 & 1 & 11.56@xmath220.94 + c2 - 120 & 120 & 27503008 & 1 & 9.10@xmath221.16 + c2 - 135 & 135 & 27503311 & 1 & 8.73@xmath221.12 + c2 - 160 & 170 & 27503614 & 1 & 9.23@xmath220.96 + c2 - 160 & 170 & 27503614 & 1 & 8.34@xmath220.27 + c2 - 200 & 200 & 27502402 & 1 & 6.61@xmath220.39 + c1 - 105 & 105 & combined & & 12.30@xmath220.47 + c2 - 160 & 170 & combined & & 8.41@xmath220.26 + lcllll @xmath0 tau & c1 - 60 & sw2 & 4.2e-18 & 4.5e-19 & 3.8e-19 + @xmath0 tau & c1 - 100 & lw1 & 8.0e-19 & 1.3e-19 & 1.1e-19 + @xmath0 tau & c2 - 160 & lw4 & 9.1e-20 & 4.5e-20 & 5.9e-20 + @xmath0 boo & c1 - 50 & sw2 & 4.2e-18 & 2.0e-19 & @xmath511.3e-19 + @xmath0 boo & c1 - 90 & sw5 & 3.2e-18 & 8.7e-20 & + @xmath0 boo & c1 - 105 & lw1 & 8.0e-19 & 5.2e-20 & 2.7e-20 + @xmath0 boo & c2 - 120 & lw2 & 7.7e-21 & 1.2e-20 & 1.2e-20 + @xmath0 boo & c2 - 135 & lw3 & 2.0e-20 & 1.7e-20 & 7.0e-21 + @xmath0 boo & c2 - 160 & lw4 & 9.1e-20 & 1.5e-20 & 1.8e-20 + @xmath0 boo & c2 - 200 & lw5 & 4.4e-19 & 8.6e-21 & + llcc 217.820 ghz ( 1.376 mm ) & observed flux ( mjy ) & 83.50@xmath221.71 & 25.78@xmath225.64 + & predicted flux ( mjy ) & 33.43 & 32.38 + & @xmath52predicted observed flux@xmath52/@xmath14(observed ) & 29.3 & 1.17 + & & & + & observed brightness temperature ( k ) & 7040@xmath22140 & 2200@xmath22480 + & predicted brightness temperature ( k ) & 2820 & 2760 + 108.406 ghz ( 2.765 mm ) & observed flux ( mjy ) & 20.09@xmath220.69 & 13.97@xmath221.46 + & predicted flux ( mjy ) & 7.94 & 7.80 + & @xmath52predicted observed flux@xmath52/@xmath14(observed ) & 17.6 & 4.23 + & & & + & observed brightness temperature ( k ) & 6840@xmath22230 & 4810@xmath22500 + & predicted brightness temperature ( k ) & 2700 & 2690 + llcc adopted angular diameter ( mas ) & 21.0@xmath220.2 & 20.88@xmath220.10 + adopted parallax ( mas ) & 88.85@xmath220.74 & 50.09@xmath220.95 + deduced stellar radius ( cm ) & 1.77 x 10@xmath25 & 3.12 x 10@xmath25 + 1.64 @xmath2 m - 2.8 mm atmospheric thickness ( cm ) & 4.38 x 10@xmath53 & 4.22 x 10@xmath53 + percentage of stellar radius & 2.5% & 1.4% +

the precise shape of the quark - gluon vertex plays an essential role in the understanding of many important properties of the infrared behavior of qcd @xcite . this vertex is in particular fundamental in the analysis of the origin of dynamical chiral symmetry breaking in the context of truncated dyson - schwinger ( ds ) equations . indeed , the ds equation for the quark self - energy requires the knowledge of this vertex , together with the gluon two - point vertex function . even if this kind of object depends on the gauge - fixing , one expects that the gauge - invariant physical content of the theory is encoded on its behavior . for example , the meson spectrum can be studied by injecting such vertices and two - point functions in appropriate bethe - salpeter equations , see for example @xcite . for these reasons , the determination of the quark - gluon vertex has been a topic of intensive work in the past . it has been first studied in the perturbative domain . a general parametrization based on its symmetries and a one - loop calculation of the abelian part in feynman gauge was performed in @xcite . some years later , the complete one loop calculation in an arbitrary linear covariant gauge and in arbitrary dimension was performed in @xcite ( where previous partial results are reviewed also ) . more recently the complete 2-loop off - shell calculation has been performed in @xmath0 @xcite . these perturbative studies lead to a satisfactory description of this vertex at high momenta . however , the infrared sector , which is of direct interest for the understanding of chiral symmetry breaking , is not accessible to perturbation theory . to cope with problem , mainly two methods have been used : lattice simulations and ds equations . in both of these approaches , the most convenient gauge is the landau gauge . from the lattice simulation side , it is well - adapted because the gauge - fixing operation can be expressed in terms of an extremization , which can be implemented efficiently on the lattice . from the analytic side , this gauge has several advantages . first , it is a linear gauge condition that manifestly preserves lorentz invariance . second , among all linear covariant gauges , it is the only one that includes the symmetries of the curci - ferrari gauge @xcite . this implies , in particular , the existence of some non - renormalization theorems , that simplify considerably the algebraic work @xcite . third , in this gauge , the gluon propagator is transverse . consequently , only the vertices where the gluon lines are multiplied by transverse projectors contribute to the correlation functions . accordingly , all consideration below will be done in this gauge fixing . there have been many analysis of the infrared behavior of the quark - gluon vertex based in the study of truncated ds equations ( for a review , see @xcite ; for more recent work on the subject , see @xcite ) . in parallel the intermediate and infrared regime of the vertex has been studied by lattice simulations . many tensorial components have been analyzed @xcite for some particular configuration of momenta . moreover , for the tensorial component that is already present at bare level , which is expected to be `` dominant '' , more general configurations of momenta have been considered @xcite . in continuum analytical approaches , the gauge fixing is generally implemented through the faddeev - popov construction , which is a perfectly valid procedure in the ultraviolet regime , where standard perturbation theory works well and where the faddeev - popov construction is well justified . however , for momenta smaller or of the order of 1 gev ( which is the typical energy scale of the strong interaction ) this procedure is not fully justified . the standard lore is that it becomes inappropriate because the coupling grows and perturbation theory eventually becomes a very bad approximation scheme . this argument must be tempered because the real expansion parameter of qcd is of the order of @xmath1 and because lattice simulations , at least in some renormalization schemes , indicate that this parameter remains moderate ( see , for example @xcite ) . accordingly it is reasonable to believe that some sort of perturbation theory should give results that are in qualitative agreement with experiments or lattice simulations in that regime . there is , however , another reason why standard perturbation theory should become inappropriate in the infrared regime : the faddeev - popov procedure itself becomes questionable . this originates in the existence of gribov copies @xcite in covariant gauge - fixings of non - abelian gauge theories that are ignored in the faddeev - popov construction . these copies play no significant role in the ultraviolet regime but a systematic treatment of the associated ambiguity remains an open problem in the infrared . the most developed procedure which aims at taking into account the effects of gribov copies is probably the gribov - zwanziger ( gz ) construction @xcite . the purpose of this approach is to restrict the functional integral to the first gribov region where the faddeev - popov operator is positive definite . gribov showed that in this region there are no copies that are an infinitesimal transformation of one another . unfortunately it was shown later that there are _ finite _ gauge transformations that preserve the landau condition @xcite . consequently the gz procedure does not remove all copies . moreover , the implementation of the restriction of the functional integral to the first gribov region is not completely justified from first principles . despite these difficulties , first studies of the gz predicted that the gluon propagator would vanish at zero momentum , instead of diverging as in the bare theory . a similar behavior was found also as solutions of truncated ds equations @xcite and in non - perturbative renormalization group ( nprg ) calculations @xcite ( for a general review on this formalism , see @xcite ) . this kind of solution is called `` scaling solution '' . the suppression of the gluon propagator was rapidly confirmed by lattice simulations . however , when lattice simulations became more precise they showed that the gluon propagator in landau gauge did not behave precisely as the `` scaling solution '' that ds had suggested , neither in quenched lattice simulations @xcite nor in the presence of dynamical quarks @xcite . in fact , lattice simulations favored a gluon propagator which tends to a finite value at low momentum in @xmath0 . this massive behavior ( sometimes called `` decoupling '' solution ) gives a propagator which is suppressed with respect to the bare one but not as strongly as in the `` scaling solution '' of ds equations . in parallel such decoupling solutions were also obtained from ds and nprg equations @xcite , as well as in a `` refined '' version of the gz construction @xcite . nowadays , all approaches finally converge toward a massive - like gluon propagator in yang - mills theory . moreover , in what concerns the ghost sector , there is also a consensus in favor of a massless behavior , with a slight enhancement at low energy but with the tree level power - law . finally , we note that ds and nprg equations favor a coupling constant bounded in the infrared at not too large values , see for instance @xcite . all these observations led two of us to test wether a simple perturbative analysis where a massive gluon propagator was introduced at the bare level @xcite could reproduce quantitatively the lattice data for several correlation functions . this lagrangian is invariant under a modified becchi - rouet - stora - tuytin ( brst ) symmetry that ensures the renormalizability of the model @xcite . however , because of the gluon mass , the associated transformation is not nilpotent . as a consequence , the textbook construction of a physical space based on the co - homology of the brst charge does not apply . it must be stressed that this difficulty is common to all methods that go beyond the standard perturbative fp method . in particular , the standard definition of the physical space does not apply to the gz model but even stronger , does not apply to the procedure implemented on landau - gauge lattice simulations . in fact the standard perturbative definition based on the cohomology of the brst operator includes in the physical space the transverse gluons . but this is not satisfactory for two reasons . first , on physical grounds , we know that gluons are confined and are not part of the physical spectrum . second , if trasverse gluons were in the physical space , the spectral density associated with their two - point function should be positive and lattice simulations cleary show positivity violations @xcite . all these considerations show that a proper definition of a physical space is an open and important problem . in this sense , the lack of a nilpotent brst charge in the curci - ferrari should not prevent us from studying it in relation with the lattice simulations of correlation functions . it is an open question to know if one can define a physical space for the curcci - ferrari model where all states would have a positive norm . in the present article we exploit this massive extension of the faddeev - popov lagrangian in order to study the quark - gluon vertex in landau gauge . our aim is twofold . first , we want to compare our findings with lattice simulation in order to have a further test of the ability of the curci - ferrari model to reproduce lattice simulations . as we show below , we are able to compute the _ full _ vertex at one loop in any dimension , for all tensorial structures and for any configuration of momenta . the comparison of this one loop result is in reasonable agreement with all available lattice data ( in many cases the agreement is very good ) . in particular , we discuss in detail the results of a certain scalar function ( @xmath2 in the terminology of @xcite ) where a significant discrepancy has been observed between lattice and ds results @xcite . we discuss this discrepancy in detail giving a very simple explanation of its origin . second ( and more importantly ) as explained before , this vertex is very important in various physical applications of ds and bethe - salpeter equations . we expect that the present result will be useful in future physical applications in these equations . we want to stress here that , as it stands , the simple one - loop calculation presented in this article does not encode a spontaneous breaking of the chiral symmetry . however , the renormalization - group flow leads to a strong enhancement of the quark mass in the infrared , which mimicks , for all practical purposes , the behavior induced by this symmetry breaking . this is a key ingredient for reproducing the infrared behavior of the quark - gluon vertex . the rest of the article is organized as follows . in sec . [ sec_model ] we describe in more details our model and shortly review the systematic comparison of one - loop computations with lattice data . we give some details on the one - loop calculations in sec . [ sec_one_loop ] and finally present our results in sec . [ sec_results ] . in this article , we pursue our systematic analysis of the correlation functions of qcd in the landau gauge , in a scheme where perturbative calculations can be performed in a controlled way for all energies . as explained in the introduction we consider the ( euclidean ) curci - ferrari lagrangian @xcite : @xmath3 where @xmath4 is the gauge field , @xmath5 and @xmath6 are ghost and antighost fields , @xmath7 is a lagrange multiplier that enforces the landau gauge condition @xmath8 , @xmath9 is the coupling constant and @xmath10 it is important to recall that mass is introduced at the level of the gauge - fixed action . as a consequence , the bare propagators decrease at large momenta as @xmath11 and the mass induces a soft breaking of the brst symmetry . as a consequence , all the ultraviolet properties of the standard faddeev - popov action are preserved to all orders in perturbation theory , in particular renormalizability and the behavior of correlation functions . one of the main consequences of the mass term is that the low energy properties of the system are more regular than within the usual faddeev - popov approach . in particular , it is possible to find renormalization schemes in which the coupling constant remains finite down to the deep infrared @xcite , at odds with the findings of standard perturbation theory which presents a divergence of the coupling constant at some energy of the order of 1 gev . the absence of such a landau pole opens the way to a consistent treatment of the low energy regime of the theory within perturbation theory . the general idea was tested in @xcite where the quenched gluon and ghost propagators were computed at one - loop order . the comparison with lattice data is very satisfactory , with a discrepancy of at most 10% in the whole range of energy . the analysis of the 3-point correlation functions ( three gluons and ghost - gluon ) was performed under the same line of investigation in @xcite . the comparison with lattice data is also very good although the lattice data are more noisy in this case . more recently , we have considered the influence of dynamical quarks on the theory , which are governed by the lagrangian : @xmath12 where @xmath13 are euclidean dirac matrices satisfying @xmath14 , the flavor index @xmath15 runs over the @xmath16 quark flavors . ] and the covariant derivative of the quark field reads @xmath17 in @xcite we have included the one - loop contribution of the quarks to the previously computed gluon propagator and have computed the quark propagator to the same order . again , the gluon propagator compares very well with the lattice results . functions , which gives an indirect effect on the ghost propagator . again , the comparison of the ghost propagator with lattice data is very satisfactory . ] the quark propagator can be decomposed in two independent structures in spinor indices , often called the mass ( proportional to the identity in spinor indices ) and field renormalization ( proportional to a dirac matrix ) . the one - loop results for the quark mass is in good agreement with the lattice data . in particular , it reproduces the strong enhancement at energies smaller than roughly 2 gev , which is reminiscent of the chiral symmetry breaking . below . ] however , we do not expect that the present simple one - loop calculation can break spontaneously the chiral symmetry . on the other hand , the field renormalization of the quark is not reproduced correctly , even at a qualitative level . this can be attributed to the fact that the one - loop contribution to this quantity is unusually small @xcite . indeed , when the gluon mass is set to zero , the field renormalization has no contribution at one loop but it does have contributions at two - loops , that , in consequence , become dominant . this model has also been used in another line of research in a recent series of articles : the extension of yang - mills theory at finite temperature was considered within the same framework with the goal of studying the deconfinement transition @xcite . it was found that a simple one - loop calculation can reproduce the known phenomenology [ second order transition for the su(2 ) group and first order transition for su(3 ) ] . a next - to leading calculation in the su(2 ) case leads to a quantitative agreement for the critical temperature . the determination of the quark - gluon vertex at one loop requires to compute the two diagrams depicted in fig . [ fig_diags ] . the result can be decomposed in twelve independent tensorial structures . we follow here the convention of @xcite and express the quark - gluon vertex as @xmath18 with @xmath19 we give in table [ tab_struc ] the tensorial structures @xmath20 and @xmath21 , together with some informations concerning the scalar functions @xmath22 and @xmath23 . [ cols="^,<,^,^",options="header " , ] the calculations are performed by following the same procedure as described in @xcite . the feynman rules are implemented in a mathematica procedure and the integrals over momentum written in terms of the passarino - veltman integrals @xcite : @xmath24[(p+q)^2+m_2 ^ 2]}\\ & c_0(p_1,p_2,m_1,m_2,m_3)=\\ & \nonumber\int \frac { d^dq}{(2\pi)^d}\frac 1{[q^2+m_1 ^ 2][(p_1+q)^2+m_2 ^ 2][(p_1+p_2+q)^2+m_3 ^ 2]}\end{aligned}\ ] ] our expressions for the different scalar functions expressed in terms of these integrals are available in the supplemental material @xcite . in actual calculations , it is convenient to use the feynman trick to perform the integrals . in order to reduce the number of feynman parameters , we use the identity : @xmath25 as many times as possible to reduce the number of propagators appearing in the diagrams . the price to pay is that there appears differences of masses in the denominator . for general momentum configurations , after performing the integrals over the internal momentum , we are left with a double integral over feynman parameters . however , for integer space dimension and for some momentum configurations that are considered in next section , all integrals can be realized analytically . we thus obtain analytic expressions for the one - loop vertex for those configurations . we have made several checks of our one - loop expressions . we have compared our results with those of @xcite which were obtained in the limit of vanishing gluon mass . our check consists in comparing numerically our results for vanishing gluon mass with those of @xcite . this was done for several momentum configurations and we find a perfect agreement . therefore , in the regime where the external momenta are large compared with the gluon mass , our results reproduce those of standard perturbation theory ( including anomalous dimensions and all angular dependences ) . as a consequence , we do not study this regime further below . another check comes from the behavior of the coupling constants in the chiral limit ( limit of vanishing quark mass ) . in this limit , we easily derive the ward identity associated with the chiral symmetry : @xmath26 where @xmath27 . by using the anticommutation relations of dirac matrices , it is easy to check that , in the chiral limit , the structures @xmath28 , @xmath29 , @xmath30 , @xmath31 , @xmath32 and @xmath33 are not compatible with this ward identity . we therefore conclude that the associated coupling constants must vanish in the chiral limit . we have explicitly verified that our one - loop expressions are compatible with this result . we regularize the theory by using the infrared - safe ( is ) scheme @xcite . as usual , we introduce renormalized quantities , which are related to the bare ones by renormalization factors : @xmath34 the different renormalization factors are defined by imposing : @xmath35 the last two conditions are consequences of non - renormalization theorems which relate the divergent parts of the different renormalization factors @xcite . here , these relations are imposed also for the finite parts . the last renormalization condition corresponds to the taylor scheme and is obtained by defining the coupling constant through the ghost - gluon vertex function with vanishing momentum for the external ghost . the condition ( [ nonrenmcond ] ) can not be expressed directly in terms of vertex ( or correlation ) functions . in particular , the mass parameter thus defined is not directly related to the value of the gluon propagator at vanishing momentum . expressions for the renormalization factors can be found in @xcite . in the following , we use the renormalization - group improvement for the quark - gluon vertex . it is obtained by integrating the rg equation for the quark - gluon vertex function @xmath36\gamma_\mu=0 , \end{split}\ ] ] where @xmath37 by integrating eq . ( [ eq_rg_eq ] ) we relate the quark - gluon vertex at different scales through the equation : @xmath38 here @xmath39 , @xmath40 and @xmath41 are obtained by integrating the beta functions with initial conditions given at some scale @xmath42 and : @xmath43 observe that the running quark mass appears explicitely in the previous expressions . this leads to a back - reaction of this running mass on the vertex functions . as a consequence , the mass takes very different values at @xmath44 and @xmath45 gev . even at a perturbative level , this is a large effect and , as discussed in the next section , is at the origin of the enhancement of some scalar components of the vertex . in the is scheme , for the initial conditions of interest for comparing with the lattice data , the coupling constant does not present a landau pole . in order to avoid large logarithms , we evaluate the right - hand - side of eq . ( [ eq_int_rg ] ) with @xmath46 of the order of the external momenta when they are larger than the mass . in practice , we used @xmath47 . it is important to stress that contrarily to what happens in the pure glue sector , there is an significative dependence on the scheme in the present calculation . in fact , it is important to chose the renormalization point at a scale that remain fixed when all momenta go to zero as done in @xcite . in the following , we compare our results with lattice data which were obtained in the quenched approximation . consequently , the beta functions for @xmath9 and @xmath48 as well as the anomalous dimensions for @xmath49 and @xmath50 are evaluated at @xmath51 . therefore , the quark mass does not back - react on these quantities , but it does back - react on the vertex . the results presented in the previous section were obtained for an arbitrary momentum configuration and for all tensor structures . there have been lattice studies of the quark gluon vertex for general kinematics @xcite for a particular tensorial structure in the quenched approximation . our results are in general qualitative agreement with those of @xcite at the level of precision of the data . for some particular momentum configurations , lattice data were obtained with larger statistics and therefore smaller error bars . in the following , we concentrate on these configurations which give more stringent tests of the quality of our results . we perform a comparison with lattice data in three momentum configurations . the first one corresponds to configurations where the gluon has zero momentum with arbitrary momentum of the quark . in this configuration many tensorial structures were obtained in the quenched approximation @xcite and we compare below our results for all available functions . the second configuration corresponds to equal momenta for the quark and antiquark ( and , accordingly , gluon momentum equal to minus the double of them ) . again , many tensorial structures were obtained in the quenched approximation for this configuration @xcite and we study all of them below . on top of these two families of configurations , we consider a more general configuration with equal momentum in modulus of quarks and anti - quarks but for various possible angles between them ( or , equivalently , different values of the modulus of the gluon momentum ) . this configuration includes as particular cases the previous ones but for general angles only a single tensorial structure has been simulated @xcite ( also in the quenched approximation ) . for the comparison with the simulations , we focus on two values of modulus of the gluon momenta and also for the completely symmetric configuration where the incoming momentum of the gluon is equal in modulus to the quark and antiquark momenta . it must be pointed out that the employed data are more that ten years old . more recent data with improved lattice parameter could allow to perform a more precise test of our curves but they do not exist for the moment . in the end of this section we finally discuss the effect of including dynamical quarks for the present vertex . even if there are no simulations of this vertex in the unquenched case , there are some simulations of 2-point functions @xcite that allows to fix the parameters and , in consequence , we are able to make a prediction of the behaviour of the quark - gluon vertex for the conditions that are present in those simulations . these could be compared to future unquenched simulations of the quark - gluon vertex . let us first consider the case of zero gluon momentum . in this case , the tensor structure simplifies to the following form @xmath52.\ ] ] moreover , all integrals appearing in the vertex functions can be performed analytically . by introducing @xmath53 @xmath54,\ ] ] @xmath55\]]and @xmath56 we obtain in @xmath57 @xmath58\nonumber\\ & + \frac{g^2}{192 \pi ^2 m^4 n p^4 } \log\left(\frac{m}{m}\right ) \left\{-6 m^8 \left(n^2 - 1\right)+m^6 \left[m^2 \left(10 n^2 - 9\right)-\left(14 n^2 + 3\right ) p^2\right]\right.\nonumber\\ & \left.-3 m^4 \left[m^4 n^2 - 14 m^2 p^2 + 3 \left(n^2 + 2\right ) p^4\right]+3 m^2 \left(m^6 - 7 m^4 p^2 + 15 m^2 p^4-p^6\right)\right.\nonumber\\ & \left .- n^2 \left(m^2+p^2\right)^4\right\}-\frac{g^2}{192 \pi ^2 m^4 n p^4}\log \left(\frac{m^2+p^2}{m^2}\right)\left(m^2+p^2\right)^3 \left[n^2 \left(m^2+p^2\right)-3 m^2\right]\nonumber\\ & + \frac{g^2}{384 \pi ^2 m^4 n p^4 y_1}\left\ { m^2 p^2 y_1 \left[12 m^4 \left(n^2 - 1\right)+m^2 m^2 \left(6 - 8 n^2\right)+m^2 p^2\left(19 n^2 + 6\right ) \right.\right.\nonumber\\ & \left.-6 \mathbb{l}_2 \left(2 m^2 - 5 m^2+p^2\right ) \left(m^2-m^2+p^2\right)-2 n^2 \left(m^2+p^2\right)^2\right]\nonumber\\ & + \mathbb{l}_1 \left[m^4 + 2 m^2 \left(p^2-m^2\right)+\left(m^2+p^2\right)^2\right ] \left[6 m^6 \left(n^2 - 1\right)+ m^4 m^2 \left(3 - 4 n^2\right)\right.\nonumber\\ & + m^4\left.\left(8 n^2 - 3\right ) p^2\right)\left.\left .- m^4m^2 \left(n^2 - 3\right)+6m^2 m^2 p^2+m^2\left(n^2 + 3\right ) p^4-n^2 \left(m^2+p^2\right)^3 \right]\right\},\end{aligned}\ ] ] @xmath59\nonumber\\ & -\frac{g^2}{384 \pi ^2 m^4 n p^6}\log\left(\frac{m}{m}\right)\left\{12 m^8 \left(n^2 - 1\right)-m^6 \left(10 n^2 - 9\right ) \left(2 m^2-p^2\right)+6 m^4 \left(m^4 n^2+p^4\right)\right.\nonumber\\ & + m^2 \left[-6 m^6 - 9 m^4 p^2 + 6 m^2 \left(5 n^2 - 1\right ) p^4 + 9 p^6\right]\nonumber\\ & \left.+2 n^2 \left(m^8 + 4 m^6 p^2 - 9 m^4 p^4 + 19 m^2 p^6+p^8\right)\right\}\nonumber\\ & -\frac{g^2}{384 \pi ^2 m^4 n p^6}\log \left(\frac{p^2}{m^2}+1\right)\left(m^2+p^2\right)^2 \left[m^2 \left(3 p^2 - 6 m^2\right)+2 n^2 \left(m^2+p^2\right)^2\right]\nonumber\\ & + \frac{g^2}{768 \pi ^2 m^4 n p^8}\frac{\mathbb{l}_1}{y_1}\left\{12 m^{10 } \left(n^2 - 1\right ) p^2+m^8 p^2 \left[m^2 \left(30 - 32 n^2\right)+\left(22 n^2 - 21\right ) p^2\right]\right.\nonumber\\ & + m^6 p^2 \left[2 m^4 \left(13 n^2 - 9\right)+3 m^2 \left(1 - 2 n^2\right ) p^2+\left(10 n^2 - 9\right ) p^4\right]\nonumber\\ & -m^4 \left[150 m^8 n^2 + 6 m^6 \left(n^2 + 1\right ) p^2+m^4 \left(10 n^2 - 3\right ) p^4 + 2 m^2 \left(2 n^2 - 3\right ) p^6 + 3 p^8\right]\nonumber\\ & + m^2 \left[300 m^{10 } n^2 + 2 m^8 \left(3 - 149 n^2\right ) p^2+m^6 \left(4 n^2 + 15\right ) p^4 + 9 m^4 p^6-m^2 \left(4 n^2 + 3\right ) p^8\right.\nonumber\\ & \left.\left.-\left(2 n^2 + 3\right ) p^{10}\right]+150 m^8 n^2 y_1 ^ 2 - 2 n^2 \left(m^2+p^2\right)^2 \left(75 m^8+m^6 p^2 + 3 m^4 p^4 + 3 m^2 p^6+p^8\right)\right\}\nonumber\\ & -\frac{g^2}{128 \pi ^2 m^4 n p^2}\mathbb{l}_2 \left(m^2 + 5 m^2 n^2\right ) \left(m^2-m^2+p^2\right ) , \end{aligned}\ ] ] and @xmath60 - 6 m^4 \left(n^2 - 1\right ) p^2\nonumber\\ & -\frac{3 \mathbb{l}_1}{y_1 } m^4 \left(n^2 - 1\right ) \left[p^2 \left(m^2+m^2\right)+\left(m^2-m^2\right)^2\right]\nonumber\\ & -p^2 \mathbb{l}_2 \left(m^2-m^2+p^2\right ) \left[m^2+n^2 \left(p^2 - 3 m^2\right)\right]\left.\right\}. \end{aligned}\ ] ] when performing lattice simulations , these functions are obtained from the simulated vertex @xmath61 by projecting in the various tensor structures in the following way @xcite : @xmath62 @xmath63+\lambda_1(p^2)\right)\ ] ] and @xmath64\ ] ] where , in these expressions , no implicit sum over repeated indices is meant . it is interesting to note that , in order to extract the function @xmath65 , a contribution of @xmath66 must be added . this issue will be discussed in more detail below . to compare our results with the lattice simulation of @xcite we renormalize the model in the is scheme and implement the corresponding renormalization group as discussed in sect . [ sec_one_loop ] . it is important to stress that the simulations of @xcite have been done in the quenched approximation . accordingly , we use the beta - functions of the pure yang - mills theory derived in @xcite for the gluon mass and coupling constant . this implies that the coupling constant and the gluon mass can be fixed once the gluon and ghost 2-point functions have been fitted . the only remaining free parameter is the valence quark mass ( and the overall normalization factor ) . to determine the best fit parameters for the ghost and gluon propagators , we consider the errors functions : @xmath67 we represent these functions in fig . [ fig_error_aacc ] . for the gluon propagator ( upper pannel , for 19% , 22% and 24% ) and @xmath68 for the ghost propagator ( lower pannel , for 4% , 7% and 10%).,title="fig : " ] for the gluon propagator ( upper pannel , for 19% , 22% and 24% ) and @xmath68 for the ghost propagator ( lower pannel , for 4% , 7% and 10%).,title="fig : " ] the best fits are obtained for @xmath69 and mass @xmath70 gev at the scale 1 gev which compare well with those found in @xcite . as in that reference , this value has been compared to other values of the coupling by taking into account the rg running , finding a good agreement with best estimates up to a 10 % error . the valence quark mass has been fixed in order to fit properly the quenched quark propagator . more precisely we choose the quark mass in order for the constituent quark mass to match the lattice value . doing so , one obtains a mass function ( the scalar part of the quark propagator , see @xcite ) that reasonably agree with the lattice data from @xcite as can be seen in fig [ fig_mass ] . one see that this curve mimick the enhancement of the mass in the infrared but not the spontaneous chiral symmetry breaking . indeed , the constituent quark mass vanishes when the current mass at 1 gev tends to zero . the corresponding value of the quark mass is @xmath71 gev at the scale @xmath72 , and it gives a constituent quark mass @xmath73 gev . these values tend to overestimate the mass function @xmath74 in the uv by approximatively 30% ( for example , at 3 gev we obtain 0.13 gev instead of 0.10 gev ) and to underestimate the propagator in the ir ( below 1 gev ) by a maximum value of approximatively 18% . having said that , the overall behavior is approximatively correct . it must be pointed out when fixing the parameters in order to compare both to the quark mass function and to the vertex functions , a practical difficulty takes place . the simulations have been done for both quantities with different mass parameters . for the quark mass function many bare masses were employed . in the vertex case , the employed values for the bare valence quark mass were @xmath75 mev and @xmath76 mev @xcite . however the authors claim that the vertex functions are almost insensitive to the bare quark mass and employ the data corresponding to @xmath75 mev . in order to have a single set of data we choose the bare valence quark nearest to this value that has been used for propagators which is @xmath77 mev this introduces a systematic error but that do not seem to be large because for relatively small masses like those , the mass function do not seem to depend on @xmath78 too much ( see @xcite and fig . [ fig_mass ] ) . one can also analyze the dependence on the valence quark mass of the vertex by studying the error level for the function @xmath66 defined as @xmath79 ^ 2}\left[\lambda_{1,\rm lt.}(p_i^2)-\lambda_{1,\rm th.}(p_i^2)\right]^2,\ ] ] we also consider the analogous error level for the function @xmath80 . for this last function the total error may be misleading as discussed below because it average some momenta where the the curve is essentially within error bars with the lowest momenta where errors are much larger . is problematic . consequently , we do not use this function for fitting the parameters . ] they are both shown in fig . [ errorl1 ] . contour levels for the quantity @xmath81 ( upper panel ) and @xmath82 ( lower panel ) in the plane of gluon mass vs. quark mass . the contour lines correspond to @xmath83 , @xmath84 and @xmath85 for @xmath81 and @xmath86 , @xmath87 and @xmath88 for @xmath82.,title="fig : " ] contour levels for the quantity @xmath81 ( upper panel ) and @xmath82 ( lower panel ) in the plane of gluon mass vs. quark mass . the contour lines correspond to @xmath83 , @xmath84 and @xmath85 for @xmath81 and @xmath86 , @xmath87 and @xmath88 for @xmath82.,title="fig : " ] we observe that , even if the optimum value of the masses do not coincide when fitting @xmath66 and @xmath80 , there is a region of parameters where both errors are reasonably small ( that is less than @xmath86 ) . moreover , the value of the gluon mass extracted from the ghost and gluon propagators ( @xmath70 gev ) is within an error region with less than @xmath87 . it is important to stress that once the parameters @xmath9 , @xmath48 and @xmath89 have been fixed , they are used for the calculation of the quark - gluon vertex for all momentum configurations . accordingly , they become pure predictions without free parameters of the present calculation . , @xmath70 gev , and @xmath71 gev . lattice data @xcite are marqued with dots and parametrized by their bare masses : @xmath90 mev ( squares ) , @xmath91 mev ( triangles ) , @xmath77 mev ( circles ) , @xmath92 mev ( diamond ) . ] the resulting curves for the functions @xmath66 , @xmath65 and @xmath80 can bee seen in fig . [ funesck0 ] . scalar functions of the quark - gluon vertex : @xmath93 ( top ) , @xmath2 ( middle ) and @xmath94 ( bottom ) for a vanishing gluon momentum , as a function of the quark momentum . the line are the one - loop results with @xmath71 gev . the dots correspond to the lattice data of @xcite.,title="fig : " ] scalar functions of the quark - gluon vertex : @xmath93 ( top ) , @xmath2 ( middle ) and @xmath94 ( bottom ) for a vanishing gluon momentum , as a function of the quark momentum . the line are the one - loop results with @xmath71 gev . the dots correspond to the lattice data of @xcite.,title="fig : " ] scalar functions of the quark - gluon vertex : @xmath93 ( top ) , @xmath2 ( middle ) and @xmath94 ( bottom ) for a vanishing gluon momentum , as a function of the quark momentum . the line are the one - loop results with @xmath71 gev . the dots correspond to the lattice data of @xcite.,title="fig : " ] we find a good agreement for the function @xmath66 . all the point are below a 16% of error except the first point ( @xmath95 mev ) where a larger error is observed ( 27% ) . let us point out , however that our curve is systematically below the lattice one . the behavior for @xmath80 is qualitatively reproduced , with a strong increase ( in absolute value ) at momenta of the order of 1 gev this last curve depends more strongly on the quark mass . this can be understood by the observation that @xmath94 tends to zero in the chiral limit , see table [ tab_struc ] . the function is very well reproduced ( essentially within error bars ) except for the two lowest momenta ( @xmath96 140 and 400 mev ) where the errors are larger ( 37% and 33% respectively when comparing to the lattice central value ) . let us , however , point out that the statistical error bars of these lattice points are large ( 22% in both points ) . we therefore find that for a unique choice of parameters @xmath9 , @xmath48 and @xmath89 , we reproduce very well the ghost and gluon propagators and reasonably well the mass function @xmath74 and the vertex functions @xmath93 and @xmath94 , which is a highly nontrivial result . the situation looks qualitatively different for the function @xmath65 , which tends to a constant at zero momentum in our approach while it diverges in the lattice data . this disagreement was already observed in previous analytical treatments of this function @xcite and was interpreted as a consequence of the non - perturbative features of the theory . we want to propose a completely different explanation for this phenomenon . let us go back to eq . ( [ eql2 ] ) which is used to extract @xmath2 from the lattice data . we observe that we must subtract from the function @xmath66 a quantity @xmath97.\ ] ] which is directly extracted from lattice data . the difference is then divided by @xmath98 . in our calculation , the fact that @xmath2 tends to a constant originates from a compensation of these two terms in the limit @xmath99 , the difference being of order @xmath98 . we can interpret this constant as a compensation of large numbers , a situation which is difficult to treat numerically . indeed , a small error in one of these two terms would lead to a divergence of @xmath65 when @xmath99 but this would be certainly an artifact . note that such a divergence in @xmath65 would imply a nonanalytic behavior of the vertex function at small momenta [ see eq . ( [ zerogluonmomentum ] ) ] since the @xmath99 limit would depend on the direction ( in momentum space ) which is used to perform this limit . in our calculations , which are done in the framework of the action ( [ eq_cf])([eq_quarks ] ) , the only non - analytic behaviors observed so far are a consequence of ghost loops ( see , for example , @xcite ) . such contributions are not present at one - loop order in the present vertex and most probably their contributions of higher order should be extremely small . if the disagreement takes it origin in partial compensations in the extraction of the @xmath65 function from the formula ( [ eql2 ] ) , the comparison should be better for the quantity that is directly extracted from lattice data ( that is @xmath100 ) than that for @xmath2 itself . in fig . [ funesck0l1l2comb ] we compare our prediction for @xmath100 to the corresponding lattice values . we observe a very good agreement in both expressions : our curve is almost within the lattice error bars except for the lowest momentum where the error is around the 17% with respect to the central value of simulations . we must , however point out that the curve is systematically above the central values of lattice data . the agreement between the @xmath100 with lattice data suggests that the true origin of the disagreement between analytical and lattice data for @xmath2 lies in the difficulty of obtaining a function from differences of large numbers . let us point out that the origin of the error is certainly not statistical . if it were so , the signal obtained from the lattice data for @xmath65 would be extremely noisy . most probably the origin of the error is more systematical . now it is important to recall that the lattice extraction of these scalar function requires a careful treatment of various lattice artifacts ( see @xcite ) . a mismatch in the treatment of lattice artifact of the longitudinal and transverse parts of the vertex could easily introduce a systematic error that , when divided by @xmath98 , gives a large spurious contribution in the estimate of @xmath65 . in any case , [ funesck0l1l2comb ] shows clearly that not only the component transverse to the external momentum but also the longitudinal one are correctly reproduced by the present one - loop calculation . comparison of the quantities @xmath93 and @xmath101 as a function of the momentum [ see eq . [ eq_def_lambda2 t ] ] . our one - loop results for @xmath93 ( full line ) are almost degenerate with those for @xmath102 ( dashed line ) . the higher ( blue ) points correspond to @xmath93 and the lower ( grey ) ones to @xmath103 , extracted from @xcite . in the latter case , our estimation of the error bar is probably pessimistic . ] in this section we present the results for the kinematic configuration given by equal momenta of quark and anti - quark ( @xmath104 ) and , accordingly , @xmath105 . in this kinematics the quark - gluon vertex simplifies considerably @xmath106\end{aligned}\ ] ] and the transverse projected vertex ( that is @xmath107 ) < has the form @xmath108\ ] ] where we have introduced @xmath109 . in spite of this simplification , for this configuration of momenta , integrals can not be done analytically and some remaining feynman parameter integrals must be performed numerically . fortunately , having fixed the parameters from propagators , the calculation must be done only once for a given choice of parameters . in this kinematic configuration , the scalar functions @xmath110 and @xmath111 are extracted from the vertex as @xmath112 and @xmath113 in this case , no subtraction is needed to extract @xmath110 and @xmath111 from the lattice data . in fig . [ funessim ] we compare our one - loop results with the lattice data using the same initial condition of the coupling constant and masses as in the previous case , @xmath69 , @xmath70 gev and for the value of @xmath71 gev at @xmath114 gev . we observe that the function @xmath115 is well reproduced by our results ( mostly within the lattice error bars , except for the lowest momentum point where the error is of the order of 25@xmath116 ) . the other function @xmath111 is seen to be very sensitive to the choice of the quark mass at 1 gev . again , this is a consequence of the fact that , in the chiral limit , this function tends to zero , see table [ tab_struc ] . we observe that the qualitative behavior for this function is reproduced but the agreement is not quantitative . in particular , our one - loop results do not reach such large values in the deep infrared where for the lowest value of momenta the error reach 45% . moreover , the curve is systematically above the lattice curve . this is similar to what has already been observed in the ds results of @xcite . scalar function of the quark - gluon vertex : @xmath115 ( top ) and @xmath111 ( bottom ) for equal momentum ( p ) for the quark and antiquark . the full line are the one - loop results and the dots correspond to the lattice data of @xcite.,title="fig : " ] scalar function of the quark - gluon vertex : @xmath115 ( top ) and @xmath111 ( bottom ) for equal momentum ( p ) for the quark and antiquark . the full line are the one - loop results and the dots correspond to the lattice data of @xcite.,title="fig : " ] in this subsection we consider the case where the quark and antiquark have momenta with the same modulus but can have arbitrary angle between them ( or , equivalently , that one can vary the value of the gluon momentum ) . it is clear that the two previous kinematic configurations correspond to particular cases of the more general situation considered here . however , more general configurations of momenta were studied in the lattice @xcite , but only for a particular tensorial structure . more precisely , in terms of the tensorial decomposition ( [ eq : decomp_2 ] ) , the function that is extracted in the lattice is @xmath109 as in the previous case . we limit to three particular kinematic configurations . in the first two , the modulus of the gluon momentum taking values @xmath117 gev and @xmath118 gev . the third corresponds to the completely symmetric case where the modulus of the three momenta are equal @xmath119 . as before , initial conditions of the renormalization group equations correspond again to @xmath69 , @xmath70 gev and @xmath71 gev at @xmath114 gev . the results are presented in figs . [ fun_fully_sym ] and [ fun_fully_sym_2 ] . for @xmath117 gev the maximal error is of 12@xmath116 . for @xmath118 gev , our results are within lattice error bars with the exception of the point of lowest momentum ( @xmath120 gev ) where the error is of 18@xmath116 . in the completely symmetric configuration the curve is almost within error bars except for the lowest momentum where the error is 25% . at the end of this section we discuss on possible origins of these discrepancies . scalar function of the quark - gluon vertex @xmath109 for @xmath121 and @xmath122 gev ( top ) and @xmath123 gev ( bottom ) . the full line are the one loop results and the dots the lattice data of @xcite.,title="fig : " ] scalar function of the quark - gluon vertex @xmath109 for @xmath121 and @xmath122 gev ( top ) and @xmath123 gev ( bottom ) . the full line are the one loop results and the dots the lattice data of @xcite.,title="fig : " ] scalar function of the quark - gluon vertex @xmath109 for the completely symmetric configuration @xmath119 . the full line are the one loop results and the dots the lattice data of @xcite interpolated to reconstruct the symmetric configuration . ] we conclude our study by introducing the effect of dynamic quarks in our calculations . this influences the renormalization factor for the gluon @xmath124 and for the gluon mass @xmath125 , which both have a contribution from a fermion closed loop . on the contrary , the first contribution of dynamic quarks to @xmath126 and to the quark - gluon vertex comes at two loops . we conclude that the effect of dynamic quarks only influence indirectly our calculation , through the modification of the @xmath127 functions for the coupling constant and the gluon mass and for the field renormalization @xmath128 . the unquenched propagators for the quarks , gluons and ghosts where studied in @xcite , in the framework of the curci - ferrari action . we use here the parameters that were obtained in that article , as best fits when comparing with the lattice data in the only case where the mass quark can be fixed from the a quark mass function , that is @xmath129 . the best fit parameters in this case is @xmath130 , @xmath131 gev and @xmath132 gev at @xmath114 gev . the corresponding values of the couplings were compared in @xcite with standard estimates of the coupling giving an agreement with an error of the order of 17 % , which is similar to the typical overall error in the infrared of the one - loop approximation . using these values of the parameter we can now predict the vertex in the unquenched case . it is interesting to note that in the present procedure this is an extremely simple calculation once the beta functions have been calculated and the parameters fixed by comparing to the 2-point functions . this is in contrast to lattice simulations that are extremely costly in the unquenched case . this is probably the reason why there are , to our knowledge , no lattice data for this vertex . we focus here on the zero gluon momentum but other tensorial structures and kinematic configuration could be presented as well . we present in fig . [ unquenchedcurves ] the three scalar functions @xmath93 , @xmath103 and @xmath94 for different number of flavors . the presence of dynamic quarks clearly tends to increase these functions at low energy . scalar functions of the quark - gluon vertex in the unquenched case . @xmath93 ( top ) , @xmath133 ( middle ) and @xmath94 ( bottom ) for a vanishing gluon momentum , as a function of the quark momentum . the curves corresponds to for @xmath51 ( full line ) and @xmath16=2 + 1 ( small dashes).,title="fig : " ] scalar functions of the quark - gluon vertex in the unquenched case . @xmath93 ( top ) , @xmath133 ( middle ) and @xmath94 ( bottom ) for a vanishing gluon momentum , as a function of the quark momentum . the curves corresponds to for @xmath51 ( full line ) and @xmath16=2 + 1 ( small dashes).,title="fig : " ] scalar functions of the quark - gluon vertex in the unquenched case . @xmath93 ( top ) , @xmath133 ( middle ) and @xmath94 ( bottom ) for a vanishing gluon momentum , as a function of the quark momentum . the curves corresponds to for @xmath51 ( full line ) and @xmath16=2 + 1 ( small dashes).,title="fig : " ] the overall agreement of the present one - loop calculation is reasonably good . most of the lattice data are well reproduced with a good accuracy . it is important to stress that most of the calculations are pure predictions without free parameters ( that are already adjusted via 2-point functions ) and that they are all obtained by one loop calculations . however , some discrepancies are present and we analyze them in the present subsection . putting aside the @xmath134 , the largest discrepancy concerns the @xmath135 function in the equal momenta configuration . even if in this case the order of magnitude of the function and general behavior is well reproduced , there is not quantitative agreement . finally , one observes some quantitative disagreement for the function @xmath136 function in the symmetric configuration at small momenta . we can say that there is a good general agreement with lattice data but the quantitative agreement is not as good as for gluon and ghost vertices . a similar behavior is observed in the study of propagators @xcite . we give now two possible ( probably complementary ) explanations for this . first , one of the reasons of the success of a perturbative analysis in the gluon and ghost sector is that the relevant couplings ( the value of the 3-gluon and ghost - gluon couplings ) are moderate except , for the 3-gluon vertex for very small values of the coupling where the presence of the gluon mass , has already suppressed fluctuations . however , in the quark sector , even if the coupling does not presents a landau pole , it reaches significantly larger values than in the pure - glue sector . one measure of this is the value of the coupling @xmath137 or @xmath136 . one can define a natural running coupling function for this functions as @xmath138 . it is important to stress that even if for @xmath139 , all the definitions of the running coupling give a universal beta function , this is not the case for momenta @xmath140 . in consequence , the moderate values of the coupling in a certain scheme does not exclude that it may be large in some other scheme . in particular , one observes that for small values of @xmath141 the quark - gluon coupling is larger by a factor of 2.4 with respect to the ghost - gluon coupling ( giving a larger expansion parameter by a factor @xmath142 . in a sense , this is to be expected because it is well established ( see , for example , @xcite ) that the quark - gluon coupling that is required for generating dynamical chiral symmetry breaking ( dscb ) is larger that the one that is extracted from the ghost - gluon vertex . now , the expansion parameter of the perturbative expansion in the present model in the ghost - gluon sector ( precisely discussed in @xcite ) may reach values of the order of 0.2 . in consequence , the expansion parameter in the quark - gluon sector seem to put the perturbative expansion in trouble at very low momenta . this may indicate that even in the present model where the ghost - gluon sector seem to be treatable perturbatively , some sort of partial resummation is needed in the quark sector ( or , at least , in part of it ) . these facts also reflects on a certain tension when the parameters have to be chosen in order to fit all the data at the same time . even if the overall agreement is good with a single set of parameters , a precise renormalization scheme had to be chosen because at very low momenta an important sensitivity is observed . this last point is related to a second possible origin of discrepancies . in the quark - gluon vertex , the quantities that are not sensitive to the dcsb are much better es- timated that those that are sensitive to it . the function @xmath137 , gives better agreement than the @xmath143 or the @xmath135 ( which are sensitive to the dsbc ) . this manifests also in the fact that those quantities seem to have a larger sensitivity on the renormalization conditions . now , it is not clear in what measure we have properly taken into account the role of the dsbc . even if rg effect gives an important enhancement of the quark mass in the infrared that quantitatively reasonably reproduces the mass function measured in the lattice , it is unclear at the moment if a proper treatment dsbc will require to use more sophisticated techniques as the introduction of composite fields for the chiral condensate . this may be at the origin of the lower quality of quantities that are sensitive to the dsbc . in this article , we have presented the calculation of the quark - gluon vertex at one loop in the massive ( curci - ferrari ) extension of the landau gauge . the results were decomposed on the twelve independent tensorial structures . we implement the renormalization - group improvement so as to avoid large logarithms . we find that these analytic results compare qualitatively well with the quenched lattice data . in one case ( @xmath2 for a vanishing gluon mass ) , the difference between lattice data ( that predicts a divergence at low momentum ) and our ( which are finite at small momenta ) may be attributed to a difference of large numbers when extracting this function from lattice data . at the quantitative level , we observe that the quantities that are not forbidden by chiral symmetry are reproduced at the level of 10% except for momenta of the order of 100 - 200 mev where larger errors are found , as detailed in the results section . the others are much more difficult to reproduce and show strong dependence on the renormalization - group scheme . this is probably to be attributed to the rather crude treatment of chiral symmetry breaking in our calculation . moreover , in order to fit simultaneously both the mass function coming from the quark propagator and the vertex function , a certain tension in the choose of parameters is observed and in order for this to be fulfilled it is necessary to choose a precise renormalization scheme . it is interesting to point out that even if , at the end , the agreement of the present one - loop calculation is reasonably good , the precision is not as good as in the ghost - gluon sector . this may be related to a larger value of the associated coupling and some sort of resummation in this sector could be useful . this issues will be addressed in the future . m. s. bhagwat , a. holl , a. krassnigg , c. d. roberts and p. c. tandy , phys . c * 70 * , 035205 ( 2004 ) . r. alkofer , c. s. fischer , f. j. llanes - estrada and k. schwenzer , annals phys . * 324 * , 106 ( 2009 ) . h. h. matevosyan , a. w. thomas and p. c. tandy , phys . rev . c * 75 * ( 2007 ) 045201 . v. a. karmanov and p. maris , few body syst . * 46 * , 95 ( 2009 ) [ arxiv:0811.1100 [ hep - ph ] ] . a. krassnigg , phys . d * 80 * , 114010 ( 2009 ) [ arxiv:0909.4016 [ hep - ph ] ] . d. nicmorus , g. eichmann and r. alkofer , phys . d * 82 * , 114017 ( 2010 ) [ arxiv:1008.3184 [ hep - ph ] ] . h. l. l. roberts , l. chang , i. c. cloet and c. d. roberts , few body syst . * 51 * , 1 ( 2011 ) [ arxiv:1101.4244 [ nucl - th ] ] . h. sanchis - alepuz , r. williams and r. alkofer , phys . d * 87 * , no . 9 , 096015 ( 2013 ) [ arxiv:1302.6048 [ hep - ph ] ] . c. s. fischer , s. kubrak and r. williams , eur . phys . j. a * 50 * , 126 ( 2014 ) [ arxiv:1406.4370 [ hep - ph ] ] . d. dudal , h. verschelde and s. p. sorella , phys . b * 555 * , 126 ( 2003 ) . n. wschebor , int . j. mod . phys . a * 23 * , 2961 ( 2008 ) . j. a. gracey , phys . b * 552 * , 101 ( 2003 ) . r. williams , arxiv:1404.2545 [ hep - ph ] . a. c. aguilar , d. binosi , d. ibaez and j. papavassiliou , phys . d * 90 * , no . 6 , 065027 ( 2014 ) . m. mitter , j. m. pawlowski and n. strodthoff , phys . d * 91 * , 054035 ( 2015 ) [ arxiv:1411.7978 [ hep - ph ] ] . j. skullerud and a. kizilersu , jhep * 0209 * , 013 ( 2002 ) . j. i. skullerud , p. o. bowman , a. kizilersu , d. b. leinweber and a. g. williams , jhep * 0304 * , 047 ( 2003 ) . d. zwanziger , nucl . b * 323 * , 513 ( 1989 ) . d. zwanziger , nucl . b * 399 * , 477 ( 1993 ) . d. zwanziger , phys . rev . d * 65 * , 094039 ( 2002 ) . d. dudal , j. a. gracey , s. p. sorella , n. vandersickel and h. verschelde , phys . d * 78 * , 125012 ( 2008 ) . j. berges , n. tetradis and c. wetterich , phys . * 363 * , 223 ( 2002 ) [ hep - ph/0005122 ] . a. sternbeck , l. von smekal , d. b. leinweber and a. g. williams , pos lat * 2007 * , 340 ( 2007 ) . a. cucchieri and t. mendes , phys . lett . * 100 * , 241601 ( 2008 ) . a. cucchieri and t. mendes , phys . d * 78 * , 094503 ( 2008 ) . a. sternbeck and l. von smekal , eur . j. c * 68 * , 487 ( 2010 ) . a. cucchieri and t. mendes , phys . d * 81 * , 016005 ( 2010 ) . i. l. bogolubsky , e. m. ilgenfritz , m. muller - preussker and a. sternbeck , phys . b * 676 * , 69 ( 2009 ) . d. dudal , o. oliveira and n. vandersickel , phys . d * 81 * , 074505 ( 2010 ) . o. bowman , u. m. heller , d. b. leinweber , m. b. parappilly and a. g. williams , phys . d * 70 * , 034509 ( 2004 ) . m. q. huber and l. von smekal , jhep * 04 * , 149 ( 2013 ) . c. becchi , a. rouet and r. stora , commun . math . phys . * 42 * , 127 ( 1975 ) . c. becchi , a. rouet and r. stora , annals phys . * 98 * , 287 ( 1976 ) . i. v. tyutin , arxiv:0812.0580 [ hep - th ] . a. cucchieri , t. mendes and a. r. taurines , phys . d * 71 * , 051902 ( 2005 ) [ hep - lat/0406020 ] . o. bowman , u. m. heller , d. b. leinweber , m. b. parappilly , a. sternbeck , l. von smekal , a. g. williams and j. b. zhang , phys . d * 76 * , 094505 ( 2007 ) [ hep - lat/0703022 [ hep - lat ] ] . m. tissier and n. wschebor , phys . d * 84 * ( 2011 ) 045018 . j. i. skullerud , p. o. bowman , a. kizilersu , d. b. leinweber and a. g. williams , nucl . suppl . * 141 * , 244 ( 2005 ) . a. sternbeck , k. maltman , m. muller - preussker and l. von smekal , pos lattice * 2012 * ( 2012 ) 243 .

understanding the origin of spiral arms , bars , and rings is key for discerning the long - term evolution of disk galaxies . the density wave ( dw ) theory , originally conceived by @xcite and developed in a linear form by @xcite , has been commonly brought up to explain spiral structure . one fundamental prediction of this theory is the presence of offsets between tracers of the different stages of star formation and evolution @xcite . a manifestation of these offsets would be age / color gradients across arms , as a result of the interaction between a spiral pattern with an angular speed that is nearly constant with radius , and stars and gas with differential rotation . the gas overtakes the spiral pattern inside corotation ( cr , i.e. , the radius where the pattern and orbital angular velocities are the same ) , and is left behind outside cr . as gas feels the gravitational potential near the spiral arms , it shocks @xcite . this produces the piling up of dust and molecular gas that is commonly found in the concave side of the arms , if the process occurs inside cr and spiral arms trail . if the shock triggers star formation , an age ( or color ) gradient is expected across spiral arms as the new stars age and drift away from the arms . young stars would be located immediately downstream the dust lanes , followed by increasingly older stars in the direction of rotation . in part due to the difficulties encountered to detect such gradients , the relevance of spiral dws for disk galaxies has been repeatedly challenged . recent examples are @xcite , @xcite , and @xcite . in a series of papers , we have detected and studied azimuthal color / age gradients across spiral arms and bars , thus establishing a relation between spiral or bar dynamics and star formation @xcite . our studies suggest that azimuthal gradients can be found in almost every galaxy , but are difficult to detect and do not appear widespread within a single galaxy . @xcite has noticed an inverse correlation between the signatures of a spiral shock , and the presence of chains of star complexes ( `` beads on a string '' ) in spiral arms ; he has also observed that spiral shocks seem to be associated with an irregular magnetic field @xcite , whereas the beads on a string " configuration correlates with the existence of a regular magnetic field . pc , irrespective of central column density @xcite . ] the `` infant mortality '' of star clusters @xcite , that leads to their disruption in the first 10 - 30 myr , may prevent the formation of age gradients even if shocks trigger star formation . stellar winds and supernovae may drive out the gas not used in star formation , causing a change in the gravitational potential . this would lead to the disintegration of the clusters , as stars become unbound . coherent age gradients must first survive the `` infant mortality '' process ; then , after @xmath4 myr , the `` dissolution of stellar groups '' scenario proposed by @xcite can also take place near spiral arms . these last process can explain the apparent `` downstream decline '' of the gradients already noticed by us ( cf . * ; * ? ? ? * ) . [ omegavar ] recently , based on the analysis with the radial tremaine - weinberg ( rtw ) method of co and hi data of several galaxies , it has been proposed that the spiral pattern speed @xmath5 may actually increase with decreasing radius in some objects @xcite . according to @xcite and @xcite , this behavior is also seen in simulations . occurs only in simulations with flocculent , transient arms . ] if this is indeed the case , there would be no real long - lived pattern with solid body rotation , and hence no significant offsets between the tracers of the different stages of star formation would be expected ( see , for example , * ? ? ? in particular , azimuthal color gradients across the arms would not be observed . in @xcite , we investigated the effect of having ( and neglecting ) non - circular streaming motions , that is , of using a purely circular dynamic model to derive spiral pattern speeds from observations of color gradients . summarizing , based on semi - analytical solutions and mhd simulations , gas orbits were obtained for models with a fixed pattern speed . stellar population synthesis ( sps ) models of age gradients were incorporated to these gas orbits , and synthetic photometric observations were performed . the method of ( * ? ? ? * hereafter gg96 ) was then applied to measure @xmath5 . in spite of having a real pattern speed , constant with radius , higher pattern speeds were measured at smaller radii ( see figure [ noncircular ] ) . as mentioned in @xcite , the reason for this bias is that gas streamlines in a steady rotating spiral shock turn somewhat along the arm after passing through the shock . hence , stars take longer to move away from the arm , and the observer will think there is a smaller difference between the pattern speed and the orbital frequency . @xmath5 will seem to follow @xmath6 , just like in the previously mentioned applications of the radial rtw method , except that in our case we can be sure that the trend is caused by the streaming motions of the young stars , rather than by a radial variation of @xmath5 . the bias is also found in real observations , as shown in @xcite . in an ideal scenario where @xmath5 is indeed constant for all radii and spiral arms trigger star formation in all their extension , color gradients are predicted to be located all along the arms ( except near cr ) . in reality , age gradients are detected only in a few small regions per galaxy , but they do seem linked to disk dynamics ( gg96 ; * ? ? ? * ; * ? ? ? * ; * ? ? ? so far , we have detected gradients through visual inspection . in order to locate gradients in a more objective way , and to attempt to identify some that do not meet the theoretical expectations of coherence or smoothness , in what follows we will adopt fourier techniques to study normal ( or weakly barred ) spirals . our data consist of deep photometric images of the 13 sa and sab spirals . ] described in @xcite . the data include mosaics in the optical @xmath7 , @xmath8 , and @xmath9 bands ( * ? ? ? * ; * ? ? ? * see table [ tbl - filters ] ) , the near - infrared ( nir ) @xmath0 band , and the nir @xmath10 , @xmath11 , @xmath12 or @xmath13 filters . we have now added six new objects , whose observation log is shown in table [ tbl - obslog ] . ngc 1703 and ngc 6907 , classified as sb by rc3 , were included in the analysis because ( apparently ) they have no prominent bar structure , as appreciated in the images . the data were reduced and calibrated with the same standard procedures used in @xcite , and @xcite . the optical ( @xmath7 , @xmath8 , and @xmath9 ) calibration was done in the thuan - gunn system @xcite . the zero point of this photometric system is chosen such that the standard star bd+17@xmath144708 has @xmath15 mag . the nir @xmath0 data were calibrated with images from the two micron all sky survey ( 2mass ; * ? ? ? * ) . @xmath7 _ & 5000 & 830 + _ @xmath8 _ & 6800 & 1330 + _ @xmath9 _ & 7800 & 1420 + _ @xmath0 _ & 1.25 & 0.29 + cccccl ngc 1703 & @xmath7 & 5100 & ctio 0.9 m & 3/94 , 3/95 & type : sb(r)b + & @xmath8 & 4200 & & & pa ( deg ) : 104 + & @xmath9 & 5400 & & & @xmath16 ( deg ) : 27.0 @xmath17 12.1 + & @xmath0 & 2717 & ctio 1.5 m & 2/94 & @xmath18 ( km s@xmath19 ) : 55 @xmath17 20 + & @xmath11 & 1281 & & & dist ( mpc ) : 20.4 @xmath17 1.7 + & @xmath10 & 2758 & & & + ngc 3001 & @xmath7 & 4500 & ctio 0.9 m & 3/94 , 3/95 & type : sab(rs)bc + & @xmath8 & 5100 & & & pa ( deg ) : 6 + & @xmath9 & 4500 & & & @xmath16 ( deg ) : 47.5 @xmath17 4.9 + & @xmath0 & 2718 & ctio 1.5 m & 2/94 & @xmath18 ( km s@xmath19 ) : 248 @xmath17 20 + & @xmath10 & 2700 & & & dist ( mpc ) : 35.6 @xmath17 3.0 + ngc 3338 & @xmath7 & 3600 & lick 1 m & 4/94 , 11/94 & type : sa(s)c + & @xmath8 & 4500 & & & pa ( deg ) : 100 + & @xmath9 & 4200 & & 2/94 , 4/94 , 11/94 & @xmath16 ( deg ) : 51.9 @xmath17 2.1 + & @xmath0 & 1200 & lick 1 m & 12/94 , 2/95 & @xmath18 ( km s@xmath19 ) : 194 @xmath17 9 + & @xmath0 & 1002 & kitt peak 1.3 m & 3/94 , 11/94 & dist ( mpc ) : 23.7 @xmath17 2.1 + & @xmath20 & 926 & & & + ngc 4603 & @xmath7 & 5400 & ctio 0.9 m & 3/94 , 3/95 & type : sa(s)c + & @xmath8 & 4500 & & & pa ( deg ) : 27 + & @xmath9 & 4200 & & & @xmath16 ( deg ) : 43.6 @xmath17 5.7 + & @xmath0 & 1393 & ctio 1.5 m & 2/94 & @xmath18 ( km s@xmath19 ) : 242 @xmath17 32 + & @xmath10 & 1399 & & & dist ( mpc ) : 29.4 @xmath17 2.6 + ngc 6907 & @xmath7 & 3600 & ctio 0.9 m & 9/94 & type : sb(s)bc + & @xmath8 & 3900 & & & pa ( deg ) : 46 + & @xmath9 & 3600 & & & @xmath16 ( deg ) : 35.6 @xmath17 3.7 + & @xmath0 & 570 & ctio 1.5 m & 9/94 , 9/95 & @xmath18 ( km s@xmath19 ) : 244 @xmath17 41 + & @xmath20 & 355 & & & dist ( mpc ) : 47.7 @xmath17 4.1 + ngc 7083 & @xmath7 & 3600 & ctio 0.9 m & 9/94 & type : sa(s)bc + & @xmath8 & 3900 & & & pa ( deg ) : 5 + & @xmath9 & 3900 & & & @xmath16 ( deg ) : 52.9 @xmath17 3.0 + & @xmath0 & 840 & ctio 1.5 m & 9/94 & @xmath18 ( km s@xmath19 ) : 226 @xmath17 10 + & @xmath20 & 392 & & & dist ( mpc ) : 45.4 @xmath17 3.9 + the dw theory predicts the existence of age gradients across spiral arms . multi - armed or flocculent galaxies , however , are not described by this theory . the short arms of the latter are probably unbound star complexes sheared by galactic differential rotation , whereas the `` knee - like '' . ] shaped arms of m101-type galaxies might be formed by the gravitational instability of galactic disks ; if so , they are transient features @xcite . the azimuthal profiles ( intensity versus @xmath21 ) of such arms should be symmetric about the gravitational potential well , where the gas is yet to be transformed into stars ; this configuration is indeed sometimes observed in certain galaxies ( for example , in ngc 4535 and m61 , a narrow dark lane is seen right at the center of a stellar arm ) . to analyze the validity of the dw theory , one should study galaxies whose structure is likely to be explained by it . mainly , from the point of view of their spiral arms , there are three different types of galaxies ( e.g. , * ? ? ? these are ( 1 ) symmetric grand - design spirals ( e.g. , m81 ) , with mass arms due to dws ; ( 2 ) multi - armed or `` knee - like '' spirals ( e.g. , m101 ) ; and ( 3 ) flocculent spiral galaxies ( e.g. , ngc 2841 ) . however , even in the same galaxy , different arms may be formed by different mechanisms . thus , in the classical multi - armed galaxy m101 the inner arms are rather regular , and in our own galaxy the quite symmetrical inner arms transform into a polygonal multi - armed structure in the outer regions @xcite . albeit flocculent and multi - armed structures have a low mass - to - light ratio ( @xmath22 ) and are seen mainly in the optical ( see , for example , the classical example of ngc 309 in block et al . 1994 ) , in order to determine whether the arms in the nir are indeed mass dws it is important to disentangle the contribution of young stars and clusters at these longer wavelengths ; such contribution can reach up to 20%-30% ( e.g. , rix & rieke 1993 ; gg96 ; rhoads 1998 ; james & seigar 1999 ; patsis et al . 2001 ; grosbl et al . 2006 ) . with the aim of determining the actual structural type of the spiral arms , resolved stellar mass maps are required . mass maps get around the problem that the @xmath22 , at any wavelength , is not constant across the disk and near spiral arms . in order to build resolved maps of stellar mass of our sample galaxies , we use the method of @xcite . the method relies on a monte carlo library of 50,000 stellar population spectra , constructed from the sps models of @xcite , and charlot & bruzual ( 2007 , private communication ) . each spectrum is computed by randomly drawing the model parameters ( star formation history , metallicity , and dust attenuation ) from adequate physical distributions ( see also * ? ? ? we use the `` magphys '' code of @xcite to extract colors from the spectral library . each element in a theoretical two - dimensional color - color plot can be produced by several combinations of model parameters . the median @xmath22 of all these combinations is then found , and attributed to the data with the same position in the observed two - dimensional color - color plot . resolution elements in two - dimensional color space ( in our case , @xmath2 versus @xmath3 ) have a size of @xmath23 mag@xmath24 , in order to account for the observational error . we compute the median @xmath22 in the @xmath0-band ( see figure [ colorsml ] ) , and compare the resulting distributions with the colors obtained from the observations , zibetti / software/ adaptsmooth.html the code enhances the s / n with a minimum loss of effective resolution , and keeps the photometric fluxes unaltered . the models are reddened before comparing with the observations , to take into account foreground galactic extinction @xcite . ] on a pixel - by - pixel basis . this comparison produces a resolved map of @xmath1 . finally , a relative mass map " is obtained , equal to the product of the @xmath1 map by the @xmath0-band image . absolute mass maps can be worked out , taking into account the distance to the galaxies , but are not necessary for our purposes . the results are shown in figures [ j_vs_mass ] and [ j_vs_mass_cont ] , where we compare the @xmath7 and @xmath0 frames with the mass maps at the same relative scale . we then proceed to select the objects in our sample that retain a spiral structure that may be explained by the dw theory . from this _ visual _ inspection we reject the following objects prior to the fourier analysis : 1 . _ ngc 3162_. there is no two - armed spiral structure appreciable in the mass map . ngc 3938_. the mass map has a very low quality , due to a low signal - to - noise ratio ( s / n ) of the @xmath7-band mosaic . ngc 5371_. ring structures rather than spirals are discerned in the mass map . ngc 7083_. the spiral structure looks flocculent in the mass map . _ ngc 7126_. no clear spirals can be appreciated in the mass map . _ ngc 7753_. spiral structure is not observed in the mass map . the remaining 13 objects were analyzed as described below . [ fourier ] [ q_fourier ] the first step in our fourier analysis consists in generating @xmath25 mosaics , following the method of gg96 . @xmath26 is defined as follows : @xmath27 where @xmath28 is the color excess term . this photometric index is reddening - free for screen absorption , and also for a mixture of dust and stars , as long as @xmath29 . at the same time , @xmath25 is sensitive to supergiant stars , and therefore traces star formation . sps models @xcite predict that , following a continuous burst of star formation that lasts for @xmath30 yr , the @xmath25 index increases its value during @xmath31 yr and then starts to decline ( cf . , gg96 ; * ? ? ? * ) . by mass ) and old stars ( with an age of @xmath32 yr ) . both populations have solar metallicity and a salpeter initial mass function ( imf ) , with @xmath33 0.1 @xmath34 , and @xmath35 10 @xmath34 . the choice of @xmath36 is justified by the observed inverse correlation between the locations where color gradients are detected , and those with important @xmath37 emission ( gg96 ; * ? ? ? * ; * ? ? ? ] @xmath25 mosaics of our sample were transformed to polar coordinates . vs. ln @xmath38 ; logarithmic arms appear as straight lines in such a map , with slope @xmath39 , where @xmath9 is the arm pitch angle . ] spiral arms were then `` straightened '' , by adding to the @xmath21 coordinate a phase as a function of @xmath40 , until the arms appear horizontal . unlike our previous studies , where the `` straightening slope '' was different depending on the specific arm and region under analysis , here we adopt the same `` straightening slope '' for both ( or all the ) arms in the same object . for each galaxy , this slope is obtained from the median of the pitch angles measured in different wavelengths ( see section [ pitch_angles ] and table [ tbl - pitchs ] ) . , @xmath8 , @xmath9 , and @xmath0 from the lowest to the highest value , and averaging together the two middle values . ] we divide the `` straightened '' images in vertical sections , equivalent to concentric annuli in a deprojected image of the galaxy , in order to perform a fourier analysis . here we aim to compare the observed @xmath25 index profiles with the sps models adopting fourier techniques . sps models give @xmath25 as a function of age , @xmath41 , while observations provide @xmath25 versus azimuthal angle ( or azimuthal distance after assuming a distance to the galaxy ) . the age gradient signal can be thought of as an asymmetric `` pulse '' ( see figure [ qmodel ] ) , whose width is a function of radius ; the width approaches zero at cr and has larger values at smaller radii . the width of this pulse is analogous to a period , @xmath42 , of a sinusoidal signal . hence , the pulse can also be characterized by its angular frequency @xmath43 , and analyzed with fourier techniques . we compute the fourier transform of the @xmath25 index profiles in the annuli , @xmath44 where @xmath45 in general , @xmath46 is complex , say , of the form @xmath47 . the fourier spectrum is obtained as @xmath48 . we search the spectrum for frequencies with amplitudes between 0.02 and 0.06 mag . this range of amplitudes corresponds to the ones achieved by the @xmath25 index of our models , ) . model `` a '' is obtained with the sps models only . model `` b '' includes both the dissolution of stellar groups @xcite after 50 myr , and stellar orbit diffusion . according to @xcite , the diffusion of stellar orbits can enhance the dissolution of young stellar groups by increasing their internal velocity dispersion . ] including the photometric error of the observed @xmath25 profiles , @xmath49 . if more than one frequency is found , we then select the one with the highest frequency value . profile , whereas higher frequencies correspond to small - scale structure . ] the expected radial behavior of @xmath50 after applying the procedure to the mhd simulation of @xcite is shown in figure [ wplot_prediction ] , with a dotted line . we remind the reader that the simulation has a spiral pattern with a constant angular speed that does not depend on radius . the plot also shows ( solid line ) the behavior of @xmath50 when newborn stars move in perfectly circular orbits . in both cases , since the width of the pulse goes to zero at the cr radius , its reciprocal , @xmath50 , diverges there ; @xmath51 . figures [ wplusphases_1 ] through [ wplusphases_7 ] , left panels , show the results of this analysis when applied to our sample of objects . data @xmath52 values are graphed versus @xmath53 , where @xmath54 is the mean galactocentric radius , and @xmath55 is the spiral maximum radial extent . the latter is determined visually in the nir band indicated in table [ tbl - armend ] . llr ngc 578 & 95.7 @xmath17 2.9 ( @xmath12 ) & 10.5 @xmath17 0.9 + ngc 918 & 75.4 @xmath17 5.8 ( @xmath13 ) & 7.9 @xmath17 0.7 + ngc 1417 & 60.0 @xmath17 5.0 ( @xmath0 ) & 16.6 @xmath17 1.4 + ngc 1421 & 92.8 @xmath17 2.9 ( @xmath12 ) & 13.2 @xmath17 1.1 + ngc 1703 & 27.5 @xmath17 1.4 ( @xmath10 ) & 2.7 @xmath17 0.3 + ngc 3001 & 44.0 @xmath17 2.8 ( @xmath10 ) & 7.6 @xmath17 0.8 + ngc 3338 & 145.0 @xmath17 5.0 ( @xmath20 ) & 16.7 @xmath17 1.6 + ngc 4254 & 157.5 @xmath17 7.5 ( @xmath12 ) & 12.6 @xmath17 0.8 + ngc 4603 & 45.4 @xmath17 1.4 ( @xmath10 ) & 6.5 @xmath17 0.6 + ngc 4939 & 145.0 @xmath17 5.0 ( @xmath12 ) & 32.7 @xmath17 2.8 + ngc 6907 & 55.1 @xmath17 2.9 ( @xmath20 ) & 12.7 @xmath17 1.3 + ngc 6951 & 105.0 @xmath17 5.0 ( @xmath12 ) & 12.7 @xmath17 1.1 + ngc 7125 & 95.7 @xmath17 2.9 ( @xmath0 ) & 20.7 @xmath17 1.8 + [ phases ] if dws indeed produce shocks and induce star formation , we expect increasingly older stellar populations ( with intensity peaks at longer wavelengths ) to be sequentially located downstream ( in the direction of rotation ) from the shock position within cr , and to be sequentially placed upstream ( opposite rotation ) outside cr . no gradients should be observed at cr . we use also fourier techniques to analyze our sample in a systematic manner , at all galactic radii , in search of this effect . we adopt the method of @xcite to determine the phases ( basically , the conjugate of the azimuthal positions ) of stellar populations of different ages , as traced in the @xmath7 , @xmath8 , @xmath9 , and @xmath0 bands . the method is based on computing the fourier transform of the form @xmath56 where @xmath57 is the intensity of radiation , and with phase @xmath58}{\mathrm{im}[\hat{f}(m ) ] } \right\}.\ ] ] given that we have selected objects with a strong two - armed pattern , @xmath59 . this means that 2@xmath60 radians include two periods of the analyzed signal , i.e. , there is a @xmath60 radian symmetry . although this may not be exactly true for real arms in spirals ( p. grosbl 2012 , private communication ) , it is a very good approximation for our purposes . we analyze the same concentric annuli in the `` straightened '' images as before . for the data analysis we assume all spirals in our sample trail , and hence that the spiral opens in the direction opposite rotation . furthermore , we adopt the convention that the angular coordinate increases with rotation , regardless of whether the arms have an `` s '' or a `` z '' on - the - sky view . under these premises , for a wave of the form @xmath61 , where @xmath62 is the phase , @xmath62 should decrease with wavelength inside cr and increase outside cr . there should be no phase difference with wavelength at cr ( see figure [ phaseplot_prediction ] ) . the results of this exercise are shown in figures [ wplusphases_1 ] through [ wplusphases_7 ] , right panels . for seven objects ( ngc 918 , ngc 1417 , ngc 1421 , ngc 3001 , ngc 3338 , ngc 4254 , and ngc 7125 , see table [ tbl - crll ] ) , we find similarities between the theoretical expectations and the observations , regarding both the radial distribution of @xmath50 and the radial run of intensity phases with wavelength , at least for some range of radii . however , all the objects also show discrepancies . for example , in ngc 918 , ngc 1417 , ngc 1421 , ngc 3338 , ngc 4254 , and ngc 7125 , the phases of all four passbands overlap between @xmath63 0.2 and 0.4 @xmath64 , but then behave in accordance to theory up to what is likely the cr radius , as judged by the position where @xmath50 diverges and where , again , phases are the same for all wavelengths . in ngc 918 and ngc 4254 , phase differences agree with theory within cr , but then do not change direction . this could be due to a weaker spiral potential and , hence , weaker shocks at radii beyond cr . conversely , in ngc 3001 the phases show the expected behavior only outside the cr position , if it is indeed located where the @xmath52 plot shows a clear divergence . the intensity phases in ngc 1703 ( which may be an sb type galaxy ) behave as predicted by theory , but @xmath52 does not diverge at the cr position expected from the phases . from this analysis , there is not even one object for which both diagnostics agree perfectly with theory at all radii . these findings confirm more objectively a previous result from visual inspection : age / color textbook gradients that run coherently along entire arms and change direction at cr are very hard to detect in practice . possible causes may be a low s / n of the data , or masking of the gradients by the intense and clumpy ( continuum and line ) radiation from massive star clusters . in the case of the intensity phases method , dusty environments can attenuate the shorter wavelengths , hindering the detection of an ordered age sequence .. ] it is also possible that the gradients are perturbed by the `` infant mortality '' of star clusters @xcite , or by the destruction of an ordered age front by supernovae shocks and stellar winds @xcite . on the other hand , even in the presence of dws , it is possible that no gradients are produced . for example , there may be inadequate physical conditions for the creation of the gradients , related to the strengths of the shock or of the magnetic field @xcite . there is , of course , the possibility that a real constant pattern speed for all radii does not exist , i.e. , that the spiral structure is not explained by the dw theory . our subsample , however , was selected explicitly to minimize this eventuality . furthermore , we had previously established the link of the gradients in the sample , taken as an ensemble , to disk dynamics . as confirmed by the comparison with an mhd simulation with a constant spiral pattern speed , an apparent radial dependency of @xmath5 is caused by neglecting non - circular streaming motions in the analysis of the gradients . conversely , there could be color gradients that are not linked to the dynamics of the disk . one speculative scenario is provided by the igimf ( the integrated galaxial igimf ) theory of @xcite and @xcite . this theory is able to explain the h@xmath16 cutoff in galaxy disks without a corresponding cutoff in star formation ( as deduced from the non - ionizing far - ultraviolet flux given out by b - type stars ) . the igimf theory posits that maximum star cluster mass depends on gas surface density ( or , equivalently , on star formation rate ) , and that maximum stellar mass depends on cluster mass . if this is indeed the case , one could obtain a color gradient from coeval star clusters across a region with a gas surface density gradient . at any rate , a statistical analysis , tied to spiral dynamics , of color gradients in several objects is necessary to identify the gradients that are linked to dws . this is even more important since by general rule gradients are painstakingly identified only in selected regions and not along entire arms . it is , however , important to stress that our new results imply that long - lived modes seem to have produced gradients that run for a sizeable part of a spiral arm in at least @xmath65 of the analyzed objects . this is , in itself , remarkable . for the rest of the objects ( ngc 578 , ngc 1703 , ngc 4603 , ngc 4939 , ngc 6907 , and ngc 6951 ) , the fourier methods failed to detect signs of gradients besides the ones that had already been identified by eye . it is interesting to point out , though , that ngc 4603 shows a multi - armed structure in the optical and nir bands , although it was kept in the sample because of its two - armed structure in the mass map . for ngc 6951 ( type sab in the rc3 ) and ngc 6907 ( type sb in the rc3 ) , the cr radius may lie around @xmath66 . this is possibly an indication of color gradients within the bar region , as already found in @xcite . we find no relation between hubble type and success rate in our fourier tests . lccr ngc 918 & 0.63 & @xmath67 & @xmath68 + ngc 1417 & 0.67 & @xmath69 & @xmath70 + ngc 1421 & 0.49 & @xmath71 & @xmath72 + ngc 3001 & 0.55 & @xmath73 & @xmath74 + ngc 3338 & 0.38 & @xmath75 & @xmath76 + ngc 4254 & 0.46 & @xmath77 & @xmath78 + ngc 7125 & 0.67 & @xmath79 & @xmath80 + [ pitch_angles ] as a complementary test of dw presence , we examine the pitch angles of the spiral arms in different wavebands ( @xmath7 , @xmath8 , @xmath9 , and @xmath0 ) , and in the `` resolved mass maps '' previously obtained with the method of @xcite . as already mentioned in the introduction , according to @xcite , age gradients are observed across arm segments that fulfill the conditions for a strong shock to occur . these segments seem to have an irregular magnetic field , while the regions with a regular magnetic field harbor instead weak shocks , and chains of star complexes with a nearly regular spacing . strong shocks are associated with a larger pitch angle of the spiral segment . another prediction involving pitch angles in the dw theory paradigm comes from the local dispersion relation . at a fixed mass surface density , the pitch angle increases proportionally to the square of the velocity dispersion ( see , e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) , that is , with the age of the stellar population and , hence , with image wavelength in the optical and nir . pitch angles were measured with fourier techniques @xcite , in the radial range where the spiral perturbation seems more prominent ( see table [ tbl - pitchs ] ) . this method assumes a logarithmic geometry for the spirals , although this might not be the case for all objects at all radii @xcite . figure [ histo_j_gri ] shows histograms of @xmath81 - @xmath82 , where @xmath83 is the pitch angle and @xmath84 can be @xmath7 , @xmath8 , @xmath9 , or `` mass '' , depending on the image where @xmath83 has been measured . were adopted for analysis . however , @xmath81 values were discarded if their sign does not agree with that of @xmath82 . ( signs indicate whether the galaxy has an `` s '' or a `` z '' on - the - sky view . ) the pitch angles measured for ngc 578 with the fourier method were also rejected , since their nominal sign is at odds with the observed on - the - sky view . ] negative values of the differences indicate that the pitch angle is larger in the nir than in the optical . in the histograms of @xmath85 , @xmath86 , and @xmath87 , there are indeed more objects with negative differences . the median values of each histogram are , respectively , -0.9@xmath14 , -1.1@xmath14 , and 0.0@xmath14 . this result is in agreement with @xcite , who find tighter arms in bluer colors in images of four spirals , suggesting the presence of dws . interestingly , the median value of the @xmath88 histogram is 2.2@xmath14 , so that the mass arms are statistically more open than the arms in nir light . an inspection of the mass maps indicates that the spiral arms loci in the nir images do not always coincide with the location of the arms in the mass maps ( see figures [ j_vs_mass ] and [ j_vs_mass_cont ] ) , a fact that can be explained by the contribution of young stars to the nir light . finally , we compare the median pitch angle distribution in those galaxies that we have classified as having radially extended azimuthal color gradients , against the distribution in the galaxies where we have detected localized gradients only . figure [ histo_longlived ] displays the pitch angle histograms for each data set . the statistical significance is low , due to the small number of objects , but the objects with signs of widespread color gradients may have more open spiral arms ( larger pitch angles ) . crrrrrrc ngc 578 & ( 20.3 - 63.8 ) & -34.70 ( @xmath89=2 ) & -35.72 ( @xmath89=2 ) & -37.92 ( @xmath89=2 ) & -41.68 ( @xmath89=2 ) & -49.45 ( @xmath89=2 ) & 40.97 + ngc 918 & ( 18.6 - 57.2 ) & 20.60 ( @xmath89=2 ) & 21.77 ( @xmath89=2 ) & 22.63 ( @xmath89=2 ) & 23.54 ( @xmath89=2 ) & 29.42 ( @xmath89=2 ) & 22.20 + ngc 1417 & ( 20.0 - 60.0 ) & 36.82 ( @xmath89=2 ) & 33.76 ( @xmath89=2 ) & 36.82 ( @xmath89=2 ) & 41.71 ( @xmath89=2 ) & 39.13 ( @xmath89=2 ) & 36.82 + ngc 1421 & ( 34.4 - 91.6 ) & 36.17 ( @xmath89=2 ) & 37.17 ( @xmath89=2 ) & 38.21 ( @xmath89=2 ) & 39.31 ( @xmath89=2 ) & -81.66 ( @xmath89=2 ) & 37.69 + ngc 1703 & ( 12.4 - 27.5 ) & 16.55 ( @xmath89=2 ) & 16.29 ( @xmath89=2 ) & 15.79 ( @xmath89=2 ) & -36.16 ( @xmath89=1 ) & -55.62 ( @xmath89=2 ) & 16.04 + ngc 3001 & ( 19.2 - 44.0 ) & 31.55 ( @xmath89=2 ) & 31.55 ( @xmath89=2 ) & 32.45 ( @xmath89=2 ) & 32.45 ( @xmath89=2 ) & 34.40 ( @xmath89=2 ) & 32.00 + ngc 3338 & ( 45.0 - 145.0 ) & -13.43 ( @xmath89=2 ) & -13.60 ( @xmath89=2 ) & -13.77 ( @xmath89=2 ) & -13.77 ( @xmath89=2 ) & 83.87 ( @xmath89=2 ) & -13.68 + ngc 4254 & ( 30.0 - 157.5 ) & -29.30 ( @xmath89=3 ) & 18.13 ( @xmath89=1 ) & -29.30 ( @xmath89=3 ) & -29.30 ( @xmath89=3 ) & 22.62 ( @xmath89=1 ) & -29.30 + ngc 4603 & ( 13.8 - 45.4 ) & 25.61 ( @xmath89=2 ) & 26.87 ( @xmath89=2 ) & 27.54 ( @xmath89=2 ) & 28.25 ( @xmath89=2 ) & 30.59 ( @xmath89=2 ) & 27.20 + ngc 4939 & ( 50.0 - 145.0 ) & 10.96 ( @xmath89=2 ) & 11.07 ( @xmath89=2 ) & 11.07 ( @xmath89=2 ) & 10.74 ( @xmath89=2 ) & 29.89 ( @xmath89=3 ) & 11.02 + ngc 6907 & ( 34.8 - 55.1 ) & 24.89 ( @xmath89=2 ) & 24.34 ( @xmath89=2 ) & 27.34 ( @xmath89=2 ) & 27.34 ( @xmath89=2 ) & 45.15 ( @xmath89=2 ) & 26.12 + ngc 6951 & ( 45.0 - 105.0 ) & -31.48 ( @xmath89=2 ) & -33.26 ( @xmath89=2 ) & -32.35 ( @xmath89=2 ) & -34.23 ( @xmath89=2 ) & -36.30 ( @xmath89=2 ) & -32.80 + ngc 7125 & ( 20.3 - 95.7 ) & -23.88 ( @xmath89=2 ) & -26.76 ( @xmath89=2 ) & -35.99 ( @xmath89=2 ) & -40.84 ( @xmath89=2 ) & -52.36 ( @xmath89=2 ) & -31.38 + [ conclusions ] from a fourier analysis of 13 objects , with a spiral arm structure that can be likely explained by dw theory , we find that @xmath90 of them show evidence of color gradients that run for at least a range of radii . for the remaining objects , the evidence of extended gradients is insufficient . this notwithstanding , our previous studies @xcite suggest that azimuthal age / color gradients ( or candidates ) can be found for most galaxies , at least in some regions of the spiral arms . a relation of such very localized gradients with the dynamics of the disk has been previously established by the comparison with an mhd simulation with a constant pattern speed at all radii @xcite . our results are consistent with other observational studies of age patterns ( or offsets ) across spiral arms @xcite . by means of an h@xmath16 to far - uv flux ratio method , @xcite found age gradients across the spiral arms of the grand - design spirals m74 and m100 ( ngc 628 and ngc 4321 , respectively ) . likewise , the `` breaks '' in the radial metallicity distribution found near cr in a sample of @xmath63 20 galaxies by @xcite and @xcite imply that spiral arms must be long - lived structures . a dominant pattern speed must exist with a unique cr radius , or otherwise any discontinuities in the radial metallicity profiles would be smoothed out @xcite . on the other hand , simulations have shown mutually exclusive results . long - lived quasi - steady features have been obtained for some models ( e.g. , * ? ? ? * ; * ? ? ? other works , however ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) , have found a lack of a persistent pattern , and instead recurrent short - lived transient spirals only . any model that intends to explain spiral structure has to avoid the winding dilemma , that ultimately led to the idea of stationary dws with a fixed pattern speed for all radii . for this same reason , the observational results of @xcite and @xcite for some objects , where @xmath91 decreases with radius , imply the existence of transient spiral modes that emerge and disappear at different stages in the disk evolution . such transient arms can not produce stellar age / color gradients , though . a theory that reconciles the existence of age / color gradients with a radially varying pattern speed is still missing . however , as argued in section [ omegavar ] , the observation of a pattern speed that varies with radius may be an artifact produced by the non - circular streaming motions of the stars newly born in shocked material . this effect can explain the coexistence of azimuthal age / color gradients across spiral arms , and of a spiral pattern speed @xmath5 that seems to vary with radius in the same way as the orbital frequency @xmath6 . we acknowledge dr . yuri efremov , the referee , for his comments and suggestions that have greatly improved this paper . special thanks go to ivanio puerari , for helpful discussions about fourier analysis of spiral galaxies . e.m.g . acknowledges a conacyt grant for a postdoctoral fellowship at inaoe , and former postdoctoral financial support from unam ( dgapa ) , mxico . is also grateful for financial support from both conacyt and dgapa - unam . arzoumanian , d. , andr , p. , didelon , p. , et al . 2011 , , 529 , l6 athanassoula , e. , romero - gmez , m. , bosma , a. , & masdemont , j. j. 2010 , , 407 , 1433 block , d. l. , bertin , g. , stockton , a. , et al . 1994 , , 288 , 365 bruzual , g. , & charlot , s. 2003 , , 344 , 1000 clarke , c. , & gittins , d. 2006 , , 371 , 530 charlot , s. , & fall , s. m. 2000 , , 539 , 718 considere , s. , & athanassoula , e. 1982 , , 111 , 28 considere , s. , & athanassoula , e. 1988 , , 76 , 365 da cunha , e. , charlot , s. , & elbaz , d. 2008 , , 388 , 1595 davis , b. l. , berrier , j. c. , shields , d. w. , et al . 2012 , , 199 , 33 de vaucouleurs , g. , de vaucouleurs , a. , corwin , h. g. , jr . , et al . 1991 , third reference catalogue of bright galaxies ( berlin : springer ) , ( rc3 ) dobbs , c. l. , & bonnell , i. a. 2008 , , 385 , 1893 dobbs , c. l. , & price , d. j. 2008 , , 383 , 497 donner , k. j. , & thomasson , m. 1994 , , 290 , 785 efremov , y. n. 1985 , soviet astronomy letters , 11 , 69 efremov , y. n. 2010 , , 405 , 1531 efremov , y. n. 2011 , astronomy reports , 55 , 108 egusa , f. , kohno , k. , sofue , y. , nakanishi , h. , & komugi , s. 2009 , , 697 , 1870 ferreras , i. , cropper , m. , kawata , d. , page , m. , & hoversten , e. a. 2012 , , 424 , 1636 foyle , k. , rix , h .- w . , dobbs , c. l. , leroy , a. k. , & walter , f. 2011 , , 735 , 101 fujii , m. s. , baba , j. , saitoh , t. r. , et al . 2011 , , 730 , 109 gerola , h. , & seiden , p. e. 1978 , , 223 , 129 gittins , d. m. , & clarke , c. j. 2004 , , 349 , 909 goodwin , s. p. , & bastian , n. 2006 , , 373 , 752 gonzlez , r. a. , & graham , j. r. 1996 , , 460 , 651 ( gg96 ) gonzlez - lpezlira , r. a. , pflamm - altenburg , j. , & kroupa , p. 2012 , , 761 , 124 grand , r. j. j. , kawata , d. , & cropper , m. 2012a , , 421 , 1529 grand , r. j. j. , kawata , d. , & cropper , m. 2012b , , 426 , 167 grosbl , p. j. , & patsis , p. a. 1998 , , 336 , 840 grosbl , p. , & dottori , h. 2009 , , 499 , l21 grosbl , p. , dottori , h. , & gredel , r. 2006 , , 453 , l25 hozumi , s. 2003 , in galaxies and chaos , ed . g. contopoulos & n. voglis ( lecture notes in physics , vol . 626 ; berlin : springer ) , 380 james , p. a. , & seigar , m. s. 1999 , , 350 , 791 lada , c. j. , & lada , e. a. 2003 , , 41 , 57 lin , c. c. , & shu , f. h. 1964 , , 140 , 646 lindblad , b. 1963 , stockholms observatoriums annaler , 22 , 5 martnez - garca , e. e. , gonzlez - lpezlira , r. a. , & bruzual - a , g. 2009a , , 694 , 512 martnez - garca , e. e. , gonzlez - lpezlira , r. a. , & gmez , g. c. 2009b , , 707 , 1650 martnez - garca , e. e. , & gonzlez - lpezlira , r. a. 2011 , , 734 , 122 martnez - garca , e. e. 2012 , , 744 , 92 meidt , s. e. , rand , r. j. , & merrifield , m. r. 2009 , , 702 , 277 merrifield , m. r. , rand , r. j. , & meidt , s. e. 2006 , , 366 , l17 mould , j. r. et al . 2000 , , 529 , 786 mueller , m. w. , & arnett , w. d. 1976 , , 210 , 670 patsis , p. a. , hraudeau , p. , & grosbl , p. 2001 , , 370 , 875 paturel , g. , fang , y. , petit , c. , garnier , r. , & rousseau , j. 2000 , , 146 , 19 paturel , g. , theureau , g. , bottinelli , l. , gouguenheim , l. , coudreau - durand , n. , hallet , n. , petit , c. 2003 , , 412 , 57 pflamm - altenburg , j. , & kroupa , p. 2008 , , 455 , 641 pflamm - altenburg , j. , weidner , c. , & kroupa , p. 2009 , , 395 , 394 pilbratt , g. l. , riedinger , j. r. , passvogel , t. , et al . 2010 , , 518 , l1 puerari , i. , & dottori , h. a. 1992 , , 93 , 469 puerari , i. , & dottori , h. 1997 , , 476 , l73 rix , h.w . , & rieke , m. j. 1993 , , 418 , 123 rhoads , j. e. 1998 , , 115 , 472 roberts , w. w. 1969 , , 158 , 123 roca - fbrega , s. , valenzuela , o. , figueras , f. , et al . 2013 , , submitted ringermacher , h. i. , & mead , l. r. 2009 , , 397 , 164 snchez - gil , m. c. , jones , d. h. , prez , e. , et al . 2011 , , 415 , 753 saraiva schroeder , m. f. , pastoriza , m. g. , kepler , s. o. , & puerari , i. 1994 , , 108 , 41 savchenko , s. s. 2012 , astrophysical bulletin , 67 , 310 scarano , s. , jr . , lpine , j. r. d. , & marcon - uchida , m. m. 2011 , , 412 , 1741 scarano , s. , & lpine , j. r. d. 2013 , , 428 , 625 schlegel , d. j. , finkbeiner , d. p. , & davis , m. 1998 , , 500 , 525 sellwood , j. a. 2011 , , 410 , 1637 speights , j. c. , & westpfahl , d. j. 2012 , , 752 , 52 skrutskie , m. f. , cutri , r. m. , stiening , r. , et al . 2006 , , 131 , 1163 tamburro , d. , rix , h .- w . , walter , f. , et al . 2008 , , 136 , 2872 thuan , t. x. , gunn , j. e. 1976 , , 88 , 543 wada , k. , baba , j. , & saitoh , t. r. 2011 , , 735 , 1 wade , r. a. , hoessel , j. g. , elias , j. h. , huchra , j. p. 1979 , , 91 , 35 wielen , r. 1977 , , 60 , 263 zhang , x. 1998 , , 499 , 93 zibetti , s. , charlot , s. , & rix , h .- w . 2009 , , 400 , 1181 zibetti , s. 2009 , arxiv:0911.4956

in classical statistical mechanics , particles with shapes are ubiquitous . their prominence in granular matter @xcite , soft condensed matter @xcite , and , more specifically , biological matter @xcite , is well established . their shapes vary from the highly complex such as folded proteins to the most elementary such as hard spheres . their equilibrium and nonequilibrium properties are investigated by a broad array of experimental and computational probes . analytic approaches in this area of research see their predictive power restricted to fairly simple scenarios regarding shapes , environment , and dimensionality . for rigorous calculations the domains of applicability are further narrowed . this limitation finds ample compensation in their usefulness as benchmarks and anchor points for approximations and simulations . the work reported in the following is motivated by this chain of reasoning . it deals with hard rods in one dimension at thermal equilibrium in external potentials . we consider monodisperse rods of size @xmath1 , @xmath2 on a lattice ( linear array of cells with volume @xmath0 ) . hard rods of size @xmath5 populating a continuum emerge from the limit @xmath3 , @xmath4 with @xmath6 . there are several approaches that facilitate an exact derivation of thermodynamic and structural properties for a homogeneous one - dimensional hard rod system ( tonks gas ) with first - neighbor takahashi - type interactions @xcite . one rather elegant method uses convolution relations between boltzmann factors to determine partition functions @xcite . it produces the free enthalpy and the equation of state ( eos ) for first - neighbor interactions of arbitrary range . many - body - density distribution functions can be calculated by an extension of this approach @xcite . an alternative method of similar scope has recently been developed . it uses statistically interacting vacancy particles ( sivp ) as quasiparticles @xcite . this method also yields the size distribution of vacancies between rods . on the basis of these approaches for homogeneous systems , it is possible to treat inhomogeneous systems by assuming that the eos is valid on a coarse - grained local scale . from the requirement of mechanical equilibrium , spatial variations of pressure and density can then be calculated from the spatial variations of the external potential . this provides a simple and common thermodynamic route for calculating density and pressure profiles @xcite , which generally is a difficult task for an interacting many - particle system . we will refer to this thermodynamic route as the eos method in the following . because the eos method relies on the assumption of the existence of a local eos , it is interesting to gain insight into how far this assumption is justified and whether the method gives useful information even if the underlying assumption does not hold . to tackle this question analytically , exact results for inhomogeneous systems are required . for hard rods with first - neighbor interactions in one dimension , exact treatments are possible via recursion relations for partition functions @xcite or density functional theory ( dft ) @xcite . these methods allow for the exact derivation of density profiles as well as many - body distribution functions . given exact density profiles , pressure profiles can be obtained by resorting to the requirement of mechanical equilibrium as in the eos method , but without assuming a local eos . the exact calculation of local pressures is a more subtle task . generally , the local pressure can be defined via the trace of the local stress tensor , which governs the time evolution of the momentum density in a coarse - grained continuum description @xcite . in case of pair interaction forces , thermodynamic averaging over the corresponding local stress tensor allows for the determination of pressure profiles from the density profiles and pair distribution function . our calculations here will employ two approaches . the dft for lattice fluids @xcite is used to determine exact density profiles in external fields and the sivp approach @xcite is used as a realization of the eos method . the dft and sivp approaches have domains that partially overlap and strengths that complement each other . in the dft , the operational degrees of freedom are the rods themselves . in the sivp approach , the operational degrees of freedom are the vacancies between the rods . we begin by describing the general methodology and background ( sec . [ sec : meth ] ) and then proceed with applications to rods in a uniform gravitational field ( sec . [ sec : grav ] ) and in a power - law trap ( sec . [ sec : powe ] ) . the subtleties regarding average microscopic pressure are addressed in the context of the first application . steric wall effects in lattice systems and their relations to known continuum results are discussed in sec . [ sec : ster ] . in sec . [ sec : conc ] we summarize the main conclusions and outline projected extensions to polydisperse rods . appendices [ sec : appa ] and [ sec : appc ] summarize outlying background materials for use in the main text . appendix [ sec : appb ] presents a highly practical method of calculating exact density profiles within the dft framework for arbitrary external potentials . consider rods on a linear chain @xmath7 of lattice sites , represented in fig . [ fig : sketch - dft ] as an array of cells . each cell has volume @xmath0 . rods of size @xmath8 occupy that many adjacent cells . hard walls at both ends of the chain define the boundary conditions . assuming that cells ( and rods ) have unit cross section we can conveniently use @xmath0 as a microscopic scale for both volume and length . microstates of this system are encoded in a sequence of occupation numbers , @xmath9 , @xmath10 . to a rod that occupies sites @xmath11 , we assign the occupation number @xmath12 . hardcore exclusion imposes the conditions @xmath13 , @xmath14 , and the hard walls imply @xmath15 for @xmath16 and @xmath17 , respectively . at positions @xmath18 on a lattice of size @xmath19 . the relevant external potentials @xmath20 and interaction potentials @xmath21 are stated . one is a contact potential and the other represents an interaction of maximum range @xmath22.,width=321 ] the model system analyzed in this work is specified by the hamiltonian , @xmath23 where @xmath20 is an external potential and @xmath21 an interaction potential restricted to first - neighbor rods . the first - neighbor restriction is naturally ascertained by interactions @xmath21 with a range limited to @xmath24 , where @xmath25 , but an extended range between first - neighbor rods is permitted . in this work we mainly examine the effects of hardcore repulsion in combination with external potentials . contact forces are included in one application ( secs . [ sec : grav - sivp - cont ] and [ sec : grav - dft - cont ] ) . the effects of long - range forces are being analyzed in a separate study @xcite . the analysis carried out in @xcite is based on former work @xcite and expresses the grand potential as a density functional , i.e. a functional of the mean occupation numbers of rods @xmath26 ) , but with @xmath20 replaced by @xmath27 , where @xmath27 is the unique external ` mermin potential ' that ( for given interaction @xmath21 ) would generate the @xmath28 as the mean occupation numbers in equilibrium . ] @xmath29=f[\tilde n_1,\ldots,\tilde n_l]+ \sum_{i=1}^{l}(\mathcal{u}_i-\mu)\tilde n_i\,,\ ] ] where @xmath30 is the chemical potential . the intrinsic free - energy functional in eq . ( [ eq : omega ] ) can be written in the form @xmath31&=\sum_{i=1}^l f_i[\tilde{\mathbf{n } } ] = \sum_{i=1}^l ( e_i[\tilde{\mathbf{n}}]-ts_i[\tilde{\mathbf{n}}])\,,\end{aligned}\ ] ] where @xmath32 is the temperature , and @xmath33=e_i[\tilde{\mathbf{n}}]-ts_i[\tilde{\mathbf{n}}]$ ] , @xmath34 $ ] , and @xmath35 $ ] are local functionals of the intrinsic free energy , internal interaction energy , and entropy . the latter are given by @xmath36=\sum_{j = i+\sigma}^{i+\xi } \mathcal{v}_{i , j}c_{i , j}[\tilde{\mathbf{n}}]\end{aligned}\ ] ] and @xmath37=&-k_\mathrm{b}\bigl\{\phi(a_i[\tilde{\mathbf{n } } ] ) + \phi(b_{i , i}[\tilde{\mathbf{n}}])-\phi(b_{i-1,i}[\tilde{\mathbf{n } } ] ) + \nonumber\\ & \hspace*{-2em}{}+ \sum_{j = i-\xi}^{i-\sigma}\bigl[\phi(c_{i , j}[\tilde{\mathbf{n } } ] ) + \phi(d_{i , j}[\tilde{\mathbf{n}}])- \phi(d_{i-1,j}[\tilde{\mathbf{n}}])\bigr ] \bigr\}\end{aligned}\ ] ] with pair correlators @xmath38=\langle n_in_j\rangle , \label{eq : correlators}\end{aligned}\ ] ] @xmath39 , and , @xmath40 and @xmath41 refer to the probabilities to have zero or one rod in ` one - particle cavities ' , i.e. in intervals , where at most one rod can be placed . ] [ eq : abc ] @xmath42=\tilde n_i -\sum_{j = i-\xi}^{i-\sigma}c_{i , j}[\tilde{\mathbf{n}}]\,,\ ] ] @xmath43= 1-\sum_{k = j-\xi}^i\tilde n_k + \sum_{k = j-\xi+\sigma}^{i}\sum_{l = j-\xi}^{k-\sigma } c_{k , l}[\tilde{\mathbf{n}}]\,,\ ] ] @xmath44=\tilde n_j -\sum_{k = j+\sigma}^ic_{k , j}[\tilde{\mathbf{n}}]\,.\ ] ] the pair correlators ( [ eq : correlators ] ) have their dependence on @xmath45 encoded in the implicit relations , @xmath46 the equilibrium density profile of rods follows from the extremum condition , @xmath47=0 , \quad i=1,\ldots , l.\ ] ] with the solution @xmath48 of ( [ eq : structure - equations ] ) , the functions @xmath49 $ ] , @xmath50 $ ] and @xmath51 $ ] become the intrinsic free energy , internal interaction energy , and entropy per site . the equilibrium density profile @xmath48 and the profiles of the thermodynamic potentials depend on temperature , crowding , interaction , and environment via @xmath52 , @xmath30 , @xmath21 , and @xmath20 , respectively . the cell occupancy ( mass density ) is obtained from the rod occupancy ( number density ) via @xmath53 in homogeneous systems the pressure @xmath54 follows rigorously from the free - energy density @xmath55 via @xmath56 . a natural extension of this relation to systems with external potential has the form @xmath57+\sum_{j = i-\sigma+1}^i \bar{n}_j\frac{\partial f_i[\bar{\mathbf{n}}]}{\partial \bar{n}_j}\,.\end{aligned}\ ] ] and produces pressure profiles on a microscopic length scale . however , there is no guarantee that the pressure thus derived coincides with the average microscopic pressure as commonly defined via the pair - distribution function . more on this question follows in secs . [ sec : meth - loca ] and [ sec : grav - micro ] . exact profiles for @xmath58 and @xmath59 for arbitrary external potentials and interactions limited to hardcore repulsion on the lattice are calculated by the method introduced in appendix [ sec : appb ] . for this comparative study of methods we need a continuum description for rod positions on a microscopic length scale used in the dft approach ( sec . [ sec : meth - exac ] ) that carries over naturally to the mesoscopic length scale used in any of the eos methods , specifically the sivp approach ( sec . [ sec : meth - eosm ] ) . this continuum description of the lattice system is unrelated to the continuum limit . we replace each lattice site @xmath60 with the interval @xmath61 across one cell ( of unit cross section ) , and we define the interaction potential @xmath62 and external potential @xmath63 by setting @xmath64 and @xmath65 . the first - neighbor restriction for the interaction potential becomes @xmath66 for @xmath67 . the local coverage at equilibrium of this interval allows us to define a number density by the piecewise continuous function @xmath68 for a homogeneous situation we have @xmath69 . from any density profile @xmath70 we can calculate the associated pressure profiles @xmath71 by invoking the balance between internal and external forces at thermodynamic equilibrium @xcite , @xmath72 where @xmath73 is the external force field . in ( [ eq : pressure - density - relation ] ) makes @xmath70 dimensionless , consistent with ( [ eq : rho - dft ] ) . ] integration of this differential equation yields @xmath74 in some applications the reference pressure @xmath75 is known , e.g. via the weight of the rods in a uniform gravitational field . in other cases it can be determined from the average number @xmath76 of rods , which we know from summing @xmath77 over all sites , by using the normalization relation @xmath78 if @xmath79 then we can use @xmath80/\partial { x_0}$ ] . if the rods are only subject to hardcore repulsion we can use the fact that at the system boundaries we have kinematic pressure @xmath81 ( sec . [ sec : meth - loca ] ) . in sec . [ sec : grav - dft ] we compare profiles inferred from ( [ eq : pressure - dft - direct2 ] ) with profiles calculated from eq . ( [ eq : pressure - profile - dft ] ) . if one assumes that the eos of a homogeneous system remains valid in a corresponding inhomogeneous system on a coarse - grained local scale in the presence of external potentials , the balance equation ( [ eq : pressure - density - relation ] ) is sufficient to determine density and pressure profiles . depending on the circumstances we use ( [ eq : pressure - density - relation ] ) to calculate the functions @xmath71 and @xmath82 or the functions @xmath70 and @xmath83 . in the former case we have an eos in the form @xmath84 and solve ( [ eq : pressure - density - relation ] ) by separating variables @xmath54 and @xmath85 : @xmath86 where @xmath87 is determined by one of the conditions stated in sec . [ sec : meth - coar ] . in the latter case we proceed analogously via separation of variables @xmath88 and @xmath85 . the eos method is particularly useful if long - range interactions are present . for such cases an exact dft calculation of density profiles tends to be be impracticable and calculations based on recursion relation for partition functions @xcite are cumbersome . for long - range first - neighbor interactions , the sivp method @xcite provides a user - friendly way to derive the eos . the microstates are encoded in a sequence of @xmath89 vacancies of size @xmath90 @xmath91 between consecutive rods , @xmath76 in number . summing over all microstates means summing over all size combinations of @xmath89 vacancies . this sum is free of constraints . the interaction energy of first - neighbor rods at distance @xmath90 is equivalent to part of the excitation energy @xmath92 of the vacancies , @xmath93 where @xmath94 . the vacancies themselves are free of interaction energies and form a set of polydisperse quasiparticles with generalized exclusion statistics . their statistical mechanics has been worked out in @xcite building on a host of foundational work @xcite . this treatment produces exact results for any thermodynamic quantity of interest for spatially homogeneous situations . the free enthalpy @xmath95 per site is given by @xmath96 from which other thermodynamic quantities are inferred via the auxiliary quantities @xcite @xmath97^\gamma e^{-\beta\epsilon_m(p)}.\ ] ] this includes the mean size @xmath98 of vacant cells @xmath99 from which the eos is obtained in the form @xmath100 in continuum mechanics , the ` microscopic ' pressure is defined as one third of the ( negative ) trace of the microscopic stress tensor and the divergence of this tensor determines the time evolution of the momentum density caused by the internal interaction forces . in one - dimension , the local stress tensor reduces to the microscopic pressure @xmath101 and the equation of motion for the momentum density @xmath102 becomes @xmath103 how statistical mechanical expressions for the local pressure @xmath104 in thermodynamic equilibrium ( or the equilibrium - averaged microscopic stress tensor ) are obtained when starting with these euler equations of continuum mechanics was first studied by irving and kirkwood in 1950 @xcite . in appendix [ sec : appa ] we have adapted the elegant derivation by lutsko @xcite to one dimension , which for pair interactions yields @xmath105 here @xmath106 is the pair distribution function and @xmath107 the force of a particle at position @xmath108 on a particle at position @xmath109 . the first term represents the kinematic pressure and the second the interaction pressure . we set @xmath110 throughout sec . [ sec : meth - loca ] . for hard rod systems , the interaction potential @xmath111 is infinite for @xmath112 . this singularity can be accounted for in eq . ( [ eq : p - rho2 ] ) by considering a modified continuous potential @xmath113 , which agrees with @xmath114 for @xmath115 while for @xmath116 it is given by @xmath117 for @xmath118 and @xmath119(r-\sigma)$ ] for @xmath120 . after inserting the corresponding force in eq . ( [ eq : p - rho2 ] ) the pressure is obtained by taking the limit @xmath121 , @xmath122 . for non - interacting hard rods ( @xmath123 for @xmath115 ) in particular , this procedure yields @xmath124 because @xmath125 for @xmath126 or @xmath127 , the range of integration in the second term extends from zero to @xmath85 for @xmath128 , and from @xmath129 to @xmath130 for @xmath131 . this means that the interaction pressure at @xmath132 and @xmath133 ( the effective system boundary ) vanishes . accordingly , the local pressure at these boundary points is just given by the kinematic pressure , as earlier pointed out by ibsen _ @xcite , who derived eqs . ( [ eq : p - rho2 ] ) and ( [ eq : p - rho2-non - interacting ] ) based on the approach in @xcite . for interacting hard rods ( @xmath134 for @xmath115 ) , the interaction pressure generally does not vanish at the system boundaries . it is interesting to note that equating the expressions for the local pressure in eqs . ( [ eq : pressure - profile - dft ] ) and ( [ eq : p - rho2 ] ) [ or eq . ( [ eq : p - rho2-non - interacting ] ) ] yields an integral equation connecting the pair distribution with the density . this could in principle be used to determine @xmath135 . alternatively , the pair distribution can be obtained by solving the inhomogeneous ornstein - zernike relation with the direct correlation function given by the second order derivatives of the density functional . for hard rods with first - neighbor interactions , many - particle - density distribution functions are most conveniently obtained by employing recursion relations for partition functions , both for continuum @xcite and for lattice systems @xcite . we note that for the latter , the integrals for the interaction pressure in eqs . ( [ eq : p - rho2 ] ) and ( [ eq : p - rho2-non - interacting ] ) can be replaced by corresponding sums ( appendix [ sec : appc ] ) . consider a semi - infinite vertical column of cells numbered @xmath136 from the bottom up . a uniform gravitational field @xmath137 acts on rods of mass @xmath138 . we use it here to represent any linear potential . we begin with non - interacting rods of size @xmath8 on a lattice and then take the continuum limit . results from sivp operating on a mesoscopic length scale are compared with those from dft operating on a microscopic length scale . pressure profiles on a microscopic length scale obtained from dft via ( [ eq : pressure - dft - direct2 ] ) are then compared with average microscopic pressure profiles inferred via ( [ eq : p - rho2-non - interacting ] ) . finally , we discuss some effects of repulsive contact interaction as made manifest in one or the other method . for the sake of brevity we limit the discussion to one simple case study of each approach . they can both be adapted to different applications . the gravitational potential to be used in ( [ eq : pressure - density - relation ] ) is @xmath139 convenient scaled variables for position , pressure , and temperature in this application are @xmath140 where @xmath141 is the length of all @xmath76 rods stacked up in a solid column . the pressure at @xmath142 is @xmath143 , independent of @xmath32 . the thermal energy @xmath144 is measured in units of the work @xmath145 required to lift this weight a distance equal to the size of one rod . the sivp analysis ( sec . [ sec : meth - eosm ] ) starts from the expression for the density of vacant cells , @xmath146^{-1},\ ] ] derived in @xcite from ( [ eq : m - sivp ] ) . the scaled mass density ( volume fraction ) inferred from ( [ eq : eos - sivp ] ) then reads @xmath147 performing the integral ( [ eq : p - from - eos - method ] ) with ( [ eq:10 ] ) and ( [ eq:22 ] ) yields the following equation for the pressure profile @xmath148 the density profile follows from ( [ eq:22 ] ) by substitution . on a lattice and the dashed lines rods ( of any size ) in a continuum.,title="fig:",width=162 ] on a lattice and the dashed lines rods ( of any size ) in a continuum.,title="fig:",width=162 ] the solid curves in fig . [ fig : fig1 ] show the profiles @xmath149 and @xmath150 at various values of @xmath151 for rods of size @xmath152 . the variation of pressure with height crosses over from hydrostatic to atmospheric with increasing temperature . the density profile is rectangular in the low - temperature limit and varies like @xmath153 at @xmath154 . these profiles do not vary much with the rod size @xmath8 when expressed by the scaled quantities ( [ eq:1 ] ) . in the limit @xmath4 , @xmath3 with @xmath6 fixed , we have a system of rods of size @xmath5 and mass @xmath138 in a continuum . the continuum versions of ( [ eq:22 ] ) and ( [ eq:23 ] ) read @xmath155 respectively . the pressure and mass density profiles in the continuum are shown as dashed curves in fig . [ fig : fig1 ] . on the length scale of single rods additional features , not resolved by any eos method including sivp , emerge in the profiles for @xmath156 when analyzed via dft . we write @xmath157 where @xmath158 is the position of the center of the lowest cell occupied by a rod . the free energy functional ( [ eq : freeenergy - dft ] ) with no interaction except hardcore repulsion acquires the form ( [ eq : b1 ] ) and the density profile of rods for any given @xmath20 are the solutions of the coupled equations ( [ eq : b2 ] ) . here , for the linear potential ( [ eq:102 ] ) , we set @xmath159 in all expressions imported from appendix [ sec : appb ] . we first examine the case @xmath152 . the solution ( [ eq : b4 ] ) is constructed from an exponential function as follows : @xmath160 where @xmath161 and the average number @xmath162 of rods is controlled by the fugacity @xmath163 , where @xmath164 is the scaled chemical potential . this dft result exactly reproduces the functional dependence of @xmath165 on @xmath166 obtained via sivp and given by ( [ eq:22 ] ) with ( [ eq:23 ] ) for @xmath152 if we use ( [ eq:101 ] ) and set @xmath167 the results from both methods are fully consistent . the sivp solution remains exact even for small numbers of rods . there exist no microscopic features in the density profile that sivp does not resolve . now we turn to the case @xmath168 , where the microscopic length scale does indeed reveal additional structures in the various profiles . these structures are encoded in eqs . ( [ eq : b5 ] ) , which for slowly varying profiles we expand into the form , @xmath169,\ ] ] and note that the first correction is of second order . the leading term alone leads to the density profile , @xmath170 which coincides with ( [ eq:22 ] ) and ( [ eq:23 ] ) for @xmath168 if we set @xmath171 thus full consistency between the two approaches is established on the mesoscopic length scale . finding the microscopic structures in the density profiles of rods and mass requires that we solve eqs . ( [ eq : b5 ] ) and then use ( [ eq:101 ] ) instead of eq . ( [ eq:34 ] ) . the solution as derived in appendix [ sec : appb ] reads @xmath172 with the @xmath173 determined recursively from @xmath174 in a uniform gravitational field at different temperatures . the solid curves represent solutions of ( [ eq:22 ] ) and ( [ eq:23 ] ) as predicted by sivp for a system with @xmath175 . the @xmath58-data on the left ( circles ) originate from ( [ eq:113 ] ) with @xmath176 from ( [ eq:35 ] ) and @xmath177 from ( [ eq:19 ] ) with @xmath178 . these data are transcribed into the circles on the right via ( [ eq:101 ] ) and into the circles of panel ( g ) via ( [ eq:305]).,title="fig:",width=162 ] in a uniform gravitational field at different temperatures . the solid curves represent solutions of ( [ eq:22 ] ) and ( [ eq:23 ] ) as predicted by sivp for a system with @xmath175 . the @xmath58-data on the left ( circles ) originate from ( [ eq:113 ] ) with @xmath176 from ( [ eq:35 ] ) and @xmath177 from ( [ eq:19 ] ) with @xmath178 . these data are transcribed into the circles on the right via ( [ eq:101 ] ) and into the circles of panel ( g ) via ( [ eq:305]).,title="fig:",width=162 ] in a uniform gravitational field at different temperatures . the solid curves represent solutions of ( [ eq:22 ] ) and ( [ eq:23 ] ) as predicted by sivp for a system with @xmath175 . the @xmath58-data on the left ( circles ) originate from ( [ eq:113 ] ) with @xmath176 from ( [ eq:35 ] ) and @xmath177 from ( [ eq:19 ] ) with @xmath178 . these data are transcribed into the circles on the right via ( [ eq:101 ] ) and into the circles of panel ( g ) via ( [ eq:305]).,title="fig:",width=162 ] in a uniform gravitational field at different temperatures . the solid curves represent solutions of ( [ eq:22 ] ) and ( [ eq:23 ] ) as predicted by sivp for a system with @xmath175 . the @xmath58-data on the left ( circles ) originate from ( [ eq:113 ] ) with @xmath176 from ( [ eq:35 ] ) and @xmath177 from ( [ eq:19 ] ) with @xmath178 . these data are transcribed into the circles on the right via ( [ eq:101 ] ) and into the circles of panel ( g ) via ( [ eq:305]).,title="fig:",width=162 ] in a uniform gravitational field at different temperatures . the solid curves represent solutions of ( [ eq:22 ] ) and ( [ eq:23 ] ) as predicted by sivp for a system with @xmath175 . the @xmath58-data on the left ( circles ) originate from ( [ eq:113 ] ) with @xmath176 from ( [ eq:35 ] ) and @xmath177 from ( [ eq:19 ] ) with @xmath178 . these data are transcribed into the circles on the right via ( [ eq:101 ] ) and into the circles of panel ( g ) via ( [ eq:305]).,title="fig:",width=162 ] in a uniform gravitational field at different temperatures . the solid curves represent solutions of ( [ eq:22 ] ) and ( [ eq:23 ] ) as predicted by sivp for a system with @xmath175 . the @xmath58-data on the left ( circles ) originate from ( [ eq:113 ] ) with @xmath176 from ( [ eq:35 ] ) and @xmath177 from ( [ eq:19 ] ) with @xmath178 . these data are transcribed into the circles on the right via ( [ eq:101 ] ) and into the circles of panel ( g ) via ( [ eq:305]).,title="fig:",width=162 ] in a uniform gravitational field at different temperatures . the solid curves represent solutions of ( [ eq:22 ] ) and ( [ eq:23 ] ) as predicted by sivp for a system with @xmath175 . the @xmath58-data on the left ( circles ) originate from ( [ eq:113 ] ) with @xmath176 from ( [ eq:35 ] ) and @xmath177 from ( [ eq:19 ] ) with @xmath178 . these data are transcribed into the circles on the right via ( [ eq:101 ] ) and into the circles of panel ( g ) via ( [ eq:305]).,title="fig:",width=162 ] the circles in the panels on the left in fig . [ fig : fig9 ] are derived from ( [ eq:113 ] ) . the probability @xmath58 that a rod occupies cells @xmath60 and @xmath179 varies with index @xmath60 in a manner that reflects the combined effects of the hardcore exclusion interaction between rods and the presence of a hard floor @xmath180 . the spatial oscillations are mild at high @xmath32 and discernible only very close to the floor . as @xmath32 is lowered , the amplitude becomes stronger and the range wider . in the limit @xmath181 the @xmath58 strictly alternate between one and zero , reflecting a compact stack of rods . when transcribed via ( [ eq:101 ] ) to the mass density , the wall effect is not nearly as strong . the evidence is represented by the circles in the panels on the right , where we set @xmath182 for the discrete data and @xmath166 from ( [ eq:1 ] ) for the curves . at high @xmath32 the effect is still strongest in the immediate vicinity of the floor but that is no longer the case at the lowest @xmath32 used in fig . [ fig : fig9 ] . at @xmath183 the effect disappears altogether . the local pressure predicted by dft as inferred from ( [ eq : pressure - dft - direct2 ] ) with the @xmath58 from ( [ eq : b2 ] ) substituted into the free - energy functional ( [ eq : b1 ] ) becomes @xmath184 with @xmath59 from ( [ eq:101 ] ) . in panel ( g ) of fig . [ fig : fig9 ] we show the pressure profiles thus obtained from the density data of panels ( a)-(f ) . the profiles are monotonically decreasing from @xmath185 . remarkably , the wild oscillations of the @xmath58 are almost completely smoothed by ( [ eq:305 ] ) . however , there does exist a systematic albeit small deviation between the dft and sivp pressure profiles . the dft pressure profile is closer to but not identical with the true microscopic pressure as will be further discussed in sec . [ sec : grav - micro ] . the agreement of the dft and sivp approaches on the mesoscopic length scale defined earlier is underlined by expressions ( [ eq:19 ] ) and ( [ eq:32 ] ) for @xmath152 and by expressions ( [ eq:34 ] ) and ( [ eq:35 ] ) for @xmath168 . it can further be shown that for any @xmath8 the density @xmath150 determined by ( [ eq:22 ] ) and ( [ eq:23 ] ) is the solution of a polynomial equation of order @xmath8 . likewise , the probabilities @xmath58 at @xmath186 inferred from ( [ eq : b2 ] ) with all @xmath77 within the square brackets set equal to each other is also the solution of a polynomial equation of order @xmath8 . the control variables are @xmath151 , @xmath76 in the first polynomial equation ( canonical ensemble ) whereas they are @xmath151 and @xmath176 in the second polynomial equation ( grandcanonical ensemble ) . the two polynomial equations are equivalent if we set @xmath187 the agreement between dft and sivp on the mesoscopic length scale also extends to the pressure profiles . from ( [ eq:305 ] ) with @xmath188 we infer @xmath189 which is equivalent to ( [ eq:22 ] ) . the microscopic features in the density profile remain conspicuous for @xmath190 . the oscillations that are superimposed on profile predicted by sivp are characterized by a ` wavelength ' proportional to @xmath8 . such profiles are readily produced from ( [ eq : b15 ] ) and ( [ eq : b16 ] ) . the exact hard - wall effects for rods of arbitrary size @xmath8 will be investigated in sec . [ sec : ster ] . soft walls as realized in power - law traps also produce structures on a microscopic length scale . some examples will be investigated in sec . [ sec : powe ] . the average local microscopic pressure @xmath191 as inferred via ( [ eq : p - rho2-non - interacting ] ) from the pair distribution function by the method presented in appendix [ sec : appc ] has two parts : a kinematic pressure and an interaction pressure ( in a formal sense ) . in fig . [ fig : microscopic_pres ] we show the profiles of both parts and their sum for rods of size @xmath192 at low temperature @xmath193 . both parts show strong oscillations that are somewhat out of phase . these oscillations are strongly attenuated with distance from the hard floor . they quickly become imperceptible with rising @xmath32 ( see fig . [ fig : microscpic_dft_pres ] ) . the non - monotonic features of @xmath191 remain totally unresolved in the eos pressure profiles discussed earlier but are partially resolved by the dft pressure profiles as illustrated in fig . [ fig : microscpic_dft_pres ] . at high @xmath151 the two profiles are virtually identical and equal to the sivp profile . at low @xmath151 , the additional @xmath191 oscillations are reproduced by dft with remarkable accuracy albeit not exactly . is not equal the total weight of the rods but it has the form @xmath194 , as can be shown from ( [ eq : part_fct_explicit ] ) , where @xmath195 is the pressure at the top of the system . with increasing system size @xmath195 approaches zero . ] for rods of size @xmath192 at low @xmath151 as inferred from ( [ eq : p - rho2-non - interacting ] ) and the pair distribution function as derived in appendix [ sec : appc].,width=264 ] ) in conjunction with ( [ eq : micscopic_pressure ] ) ( circles connected by dashed lines ) and calculated from ( [ eq : pressure - dft - direct2 ] ) ( solid lines ) for high @xmath151 ( left ) and low @xmath151 ( right).,title="fig:",width=158 ] ) in conjunction with ( [ eq : micscopic_pressure ] ) ( circles connected by dashed lines ) and calculated from ( [ eq : pressure - dft - direct2 ] ) ( solid lines ) for high @xmath151 ( left ) and low @xmath151 ( right).,title="fig:",width=158 ] consider a contact interaction potential @xmath196 that is attractive for @xmath197 and repulsive for @xmath198 . as inferred from @xcite , the contact interaction changes the local density of vacant cells from ( [ eq:21 ] ) to @xmath199},\ ] ] where we have introduced the variable @xmath200 and the scaled interaction @xmath201 in addition to the scaled variables ( [ eq:1 ] ) . the mass density @xmath165 is inferred from ( [ eq : eos - sivp ] ) . the integral ( [ eq : p - from - eos - method ] ) can still be evaluated exactly and yields the pressure profile , @xmath202 in generalization of ( [ eq:23 ] ) . an attractive contact interaction affects the profiles in a way similar to what a drop in temperature does ( see fig . [ fig : fig1 ] ) . no significant additional features make their appearance . repulsion is more interesting in that respect as documented in fig . [ fig : fig5 ] for @xmath152 . ) in a uniform gravitational field at temperatures @xmath203 and for ( repulsive ) contact interaction @xmath204 . panel ( c ) shows the @xmath205 versus @xmath166 data near @xmath180 . the dashed line has slope @xmath206.,title="fig:",width=162 ] ) in a uniform gravitational field at temperatures @xmath203 and for ( repulsive ) contact interaction @xmath204 . panel ( c ) shows the @xmath205 versus @xmath166 data near @xmath180 . the dashed line has slope @xmath206.,title="fig:",width=162 ] ) in a uniform gravitational field at temperatures @xmath203 and for ( repulsive ) contact interaction @xmath204 . panel ( c ) shows the @xmath205 versus @xmath166 data near @xmath180 . the dashed line has slope @xmath206.,title="fig:",width=162 ] the configuration of rods becomes stratified at low temperature . the density profile now exhibits an additional layer of intermediate density . the width of that layer increases with the strength of the repulsion . the ( largely hydrostatic ) pressure profile acquires different slopes inside different layers . these profiles depend systematically on @xmath8 without producing any additional features . from @xcite we know that the effects of attractive or repulsive contact interactions of finite strength fade away completely in the continuum limit . one subtle feature of note concerns the initial slope of the hydrostatic pressure as presented in panel ( a ) of fig . [ fig : fig1 ] and then again in panel ( c ) on a much expanded scale . what in panel ( a ) looks like a clean change between slope @xmath207 for interaction strengths @xmath208 and slope @xmath209 for stronger contact repulsion , @xmath210 is , at very small @xmath166 , a two - step change with slope @xmath206 at @xmath211 in the middle . this asymptotic slope is evident in the expansion of ( [ eq:38 ] ) : @xmath212+\mathrm{o}(\hat{z}^{2}).\end{aligned}\ ] ] here we wish to demonstrate how a repulsive contact force of infinite strength effectively increases the size of rods . the inclusion of a scaled contact potential @xmath213 generalizes ( [ eq : b1 ] ) to @xmath214 as shown in @xcite , where the contact interaction @xmath215 is contained in the correlators , @xmath216 @xmath217 that appear in the first three terms of ( [ eq:43 ] ) . the last two terms reflect the hardcore repulsion . the metamorphosis of hard rods of one size into hard rods of a bigger size is most transparent if we consider the case @xmath152 and compare the limits @xmath218 and @xmath219 . the extremum condition ( [ eq : structure - equations ] ) applied to ( [ eq:43 ] ) leads to the following set of relations that determine the density profile @xmath220 for external potential @xmath20 and fugacity @xmath176 : @xmath221[\bar{n}_i - c_{i-1,i}][\bar{n}_i - c_{i , i+1}]}{\bar{n}_i[1-\bar{n}_{i-1}-\bar{n}_{i } + c_{i-1,i}][1-\bar{n}_i-\bar{n}_{i+1 } + c_{i , i+1}]},\nonumber\end{aligned}\ ] ] @xmath222 where @xmath223 and @xmath224 . for @xmath225 we have @xmath226 and ( [ eq:46 ] ) reduces to @xmath227 whereas for @xmath228 we have @xmath229 and ( [ eq:46 ] ) becomes @xmath230 representing rods of size @xmath168 with only hardcore repulsion . in the application to a uniform gravitational field , ( [ eq:48 ] ) is equivalent to ( [ eq:19 ] ) and ( [ eq:49 ] ) is equivalent to ( [ eq : b5 ] ) . the sivp approach of sec . [ sec : grav - sivp - cont ] describes the same crossover from rods of size @xmath8 to rods of size @xmath231 under a repulsive contact interaction of increasing strength . the pressure profile ( [ eq:38 ] ) evaluated for @xmath232 and any @xmath8 reproduces ( [ eq:23 ] ) . when the same expression is evaluated for @xmath233 it connects again with ( [ eq:23 ] ) but now for @xmath231 provided the scaled variables are properly adjusted . optical or magnetic traps of several different designs for atomic or molecular gases produce wells with a range of profiles . how does the pressure at the center of the trap vary with temperature ? how does the shape of the trap potential affect the profiles of density and pressure ? power - law traps are well - suited for our two approaches and can illuminate these questions with answers from an exact analysis . in this application we consider an infinite row of cells numbered @xmath234 at positions @xmath235 . the rods are confined to a region centered at @xmath132 by the symmetric power - law potential @xmath236 with @xmath237 representing a depth and @xmath238 representing ( at least for @xmath239 ) a width of the trap . the analysis proceeds as in sec . [ sec : grav - sivp ] . expression ( [ eq:22 ] ) remains unchanged . however , the pressure profile is now determined by the relation @xmath240 where , in addition to @xmath205 and @xmath151 from ( [ eq:1 ] ) , we use the scaled variables @xmath241 a solid stack of rods extends out to @xmath242 and the pressure at the center becomes @xmath243 . in the continuum limit , @xmath4 , @xmath3 with @xmath6 , eqs . ( [ eq:22 ] ) and ( [ eq:27 ] ) turn into @xmath244 one additional relation is needed to bring closure to ( [ eq:22 ] ) and ( [ eq:27 ] ) , namely @xmath245 reflecting mass conservation . the pressure @xmath246 at @xmath247 can be determined from this relation . for the two extreme rod sizes we thus obtain @xmath248 @xmath249^{\frac{1-\alpha}{\alpha}}\!\!\!=1\quad ( \sigma=\infty),\ ] ] where @xmath250 is the fermi - dirac function . the rods remain confined at any finite temperature : we have @xmath251 if @xmath252 for any @xmath253 . the limit @xmath254 , representing a trap of width @xmath255 with rigid walls , is subtle . relation ( [ eq:303 ] ) reduces to @xmath256 we then have @xmath257 inside the trap , with a uniform density , @xmath258 . we thus recover the familiar eos @xmath259 of the ideal lattice gas . the limit @xmath260 yields @xmath261 and the profiles for pressure and density are @xmath262 respectively . if @xmath239 @xmath263 the pressure @xmath246 at the center of the trap increases ( decreases ) with @xmath151 rising from zero . for the linear potential @xmath264 , which is equivalent to the case of the uniform gravitational field ( sec . [ sec : grav - sivp ] ) , we have @xmath261 for all @xmath151 . in fig . [ fig : fig3 ] we show pressure and density profiles for the two extreme rod sizes in two different power - law potentials . the shape of one potential @xmath265 is concave and that of the other @xmath266 convex . corresponding profiles for a linear potential @xmath264 have already been shown in fig . [ fig : fig1 ] albeit on a somewhat different scale . ( top ) and @xmath267 ( bottom ) at temperatures @xmath268 for rods of sizes @xmath152 ( solid curves ) and @xmath269 ( dashed curves).,title="fig:",width=162 ] ( top ) and @xmath267 ( bottom ) at temperatures @xmath268 for rods of sizes @xmath152 ( solid curves ) and @xmath269 ( dashed curves).,title="fig:",width=162 ] ( top ) and @xmath267 ( bottom ) at temperatures @xmath268 for rods of sizes @xmath152 ( solid curves ) and @xmath269 ( dashed curves).,title="fig:",width=162 ] ( top ) and @xmath267 ( bottom ) at temperatures @xmath268 for rods of sizes @xmath152 ( solid curves ) and @xmath269 ( dashed curves).,title="fig:",width=162 ] the opposite @xmath151-dependence of the pressure near the center of the trap is evident . the curves at the lowest temperature are close to the @xmath270 profiles ( [ eq:50 ] ) . naturally , the pressure decrease with rising @xmath151 in for @xmath271 has a much larger effect on the density than does the pressure increase for @xmath267 . another noteworthy feature is that the difference in profile between the lattice system and the continuum system is much more pronounced for the convex potential than for the concave potential . in sec . [ sec : grav - dft ] we have already identified some hard - floor effects in the form of spatially attenuated oscillations in the density profiles @xmath58 and @xmath59 of rods and mass , respectively . we found ( in fig . [ fig : fig9 ] ) that the effect is very conspicuous in the former and partially averaged out in the latter . a more systematic analysis of hard - wall effects for rods of various sizes on the lattice and for rods in a continuum will be presented in sec . [ sec : ster ] . and mass @xmath59 for rods of size @xmath168 in a harmonic trap @xmath266 and in a power - law trap with firmer walls @xmath272 . in both traps we use @xmath273 . all data are for @xmath274 . the chemical potential @xmath275 has the values ( a ) 1.0 , ( b ) 1.55 , ( c ) 0.3 , ( d ) 0.5.,title="fig:",width=162 ] and mass @xmath59 for rods of size @xmath168 in a harmonic trap @xmath266 and in a power - law trap with firmer walls @xmath272 . in both traps we use @xmath273 . all data are for @xmath274 . the chemical potential @xmath275 has the values ( a ) 1.0 , ( b ) 1.55 , ( c ) 0.3 , ( d ) 0.5.,title="fig:",width=162 ] and mass @xmath59 for rods of size @xmath168 in a harmonic trap @xmath266 and in a power - law trap with firmer walls @xmath272 . in both traps we use @xmath273 . all data are for @xmath274 . the chemical potential @xmath275 has the values ( a ) 1.0 , ( b ) 1.55 , ( c ) 0.3 , ( d ) 0.5.,title="fig:",width=162 ] and mass @xmath59 for rods of size @xmath168 in a harmonic trap @xmath266 and in a power - law trap with firmer walls @xmath272 . in both traps we use @xmath273 . all data are for @xmath274 . the chemical potential @xmath275 has the values ( a ) 1.0 , ( b ) 1.55 , ( c ) 0.3 , ( d ) 0.5.,title="fig:",width=162 ] here we briefly examine the question whether the soft walls of power - law traps also produce patterns of spatial density oscillations . we begin with rods of size @xmath168 . in fig . [ fig : fig15 ] we compare data for traps with walls of two different degrees of softness . these data are produced by using the potential ( [ eq:26 ] ) with @xmath276 and calculating the profiles from ( [ eq : b10 ] ) and ( [ eq:101 ] ) . we use @xmath277 with @xmath278 and a shift that positions the rod with index @xmath279 at the center of the trap . the harmonic trap is , in some sense , the smoothest form of confinement . we see in panels ( a ) and ( b ) that oscillations do make their appearance in the @xmath58-profiles . the amplitudes of these oscillations tend to be rather uniform across the region where the rod population is significant . the general trend is that the amplitudes increase with increasing chemical potential @xmath280 or decreasing temperature @xmath281 . superimposed on this systematic trend is an oscillatory dependence on @xmath275 . the data in the panels ( a ) and ( b ) are for two successive values near maximum and minimum amplitude of @xmath58-oscillations . the minimum amplitude is almost imperceptible . we also note that in the @xmath59-profiles these oscillations are almost completely averaged out . increasing the stiffness of the trap walls produces no dramatic changes . the dependence on @xmath275 of the spatial oscillations remains qualitatively similar . the amplitudes still oscillate as @xmath275 is varied . the data in the panels ( c ) and ( d ) are for @xmath282 and for values of @xmath275 near successive maximum and minimum amplitude . one systematic trend as the trap wall becomes increasingly firm is that the spatial oscillations become weaker near the center of the trap and stronger near the walls . this trend is visible between the data in panels ( a ) and ( c ) , for example . we shall see in sec . [ sec : ster ] that the oscillatory dependence of the amplitudes on @xmath275 disappears when the walls become hard @xmath283 . in fig . [ fig : fig16 ] we show some data for rods of size @xmath284 in a trap with relatively stiff walls @xmath272 and twice the width of the one used before . the calculations are based on eqs . ( [ eq : b15 ] ) and ( [ eq : b16 ] ) . here we only show data of the @xmath58-profiles for two conditions . for rods of size @xmath284 in a power - law trap with @xmath282 . we use @xmath285 . all data are for @xmath274 . the scaled chemical potentials are ( a ) @xmath286 and ( b ) @xmath287 . in each plot the second and third data sets are vertically displaced by 0.2 and 0.4 , respectively.,title="fig:",width=162 ] for rods of size @xmath284 in a power - law trap with @xmath282 . we use @xmath285 . all data are for @xmath274 . the scaled chemical potentials are ( a ) @xmath286 and ( b ) @xmath287 . in each plot the second and third data sets are vertically displaced by 0.2 and 0.4 , respectively.,title="fig:",width=162 ] the average numbers of rods in the data of panel ( a ) are such that the rods easily fit into the space at low potential energy . we observe significant oscillations with a period near but not exactly three . a small change in chemical potential produces shifts in the oscillatory patterns and variations in the average amplitude . we also observe slightly larger amplitudes near the wall compared to the center of the trap . at significantly smaller values of @xmath162 the oscillations are much weaker . the data in panel ( b ) are for circumstances where the rods are squeezed into the trap . here the rod positions are more correlated . the oscillations are closer to period three . we also observe the oscillatory dependence of the amplitude on @xmath275 . for some values there are one or two dominant configurations , producing high amplitude . for other values , there are three configuration that have very similar statistical weight , producing low amplitude . microscopic density profiles of rods or other particles with shapes near hard walls are relevant in the contexts of granular matter , porous solids , and zeolites among others . robledo and rowlinson @xcite studied the effects of confinement on hard rods in a continuum . davis @xcite extended that study to include first - neighbor interactions between rods . more recently , ibsen _ et al _ @xcite reported a general and exact solution for hard rods confined by a gravitational field and a hard floor . a computer simulation study of mehrotra _ et al _ @xcite of hard spheres under the same confinement in 3d produces similar results . here we pick up threads from secs . [ sec : grav - dft ] and [ sec : powe - soft - wall ] to investigate steric wall effects of hard rods on a lattice . one goal is to showcase the versatility of the method of analysis presented in appendix [ sec : appb ] and to establish how it connects to the continuum analysis familiar from previous work . we consider a box with rigid walls and investigate the oscillations in the density profiles produced by the steric interactions between rods of size @xmath156 . the effects of a single wall , relevant in sufficiently wide boxes , can be determined analytically for rods of any size on the lattice and for rods in the continuum . two walls within the distance of a certain coherence length affect the density profile from opposite sides . that coherence length is shown to grow with average density . we combine exact analytic results with results from a rigorous recursive scheme . we begin with the case of a semi - infinite box with one wall at @xmath288 and the other at @xmath289 populated by rods of size @xmath168 to an average mass density @xmath290 . later we move the second wall to @xmath291 . for the semi - infinite box we have found an analytic solution . the result ( with @xmath292 in the present context ) turns out to have a simple structure : @xmath293,\quad j=0,1,2,\ldots\ ] ] the relation between the fugacity and the average mass density is @xmath294.\ ] ] we have derived @xmath295 in ( [ eq:103 ] ) directly from ( [ eq : b11 ] ) with @xmath296 in the form of an infinite continued fraction that is readily evaluated . the @xmath58 for @xmath297 then follow directly from ( [ eq : b5 ] ) . the oscillations in the probability distribution @xmath58 of rod positions thus decay exponentially with distance from the wall . the boundary coherence length , @xmath298 vanishes for @xmath299 and diverges for @xmath300 . next we examine how the oscillations near the wall at @xmath288 are affected by the presence of a second wall at @xmath291 . for that purpose we have solved ( [ eq : b10 ] ) with @xmath176 from ( [ eq:104 ] ) for comparison with the analytic solution ( [ eq:103 ] ) . the results , shown in fig . [ fig : fig12 ] , are almost indistinguishable for @xmath301 . here the coherence length ( [ eq:106 ] ) is much smaller than the distance between the walls . at larger mass density the two sets of results begin to deviate from each other as @xmath302 grows and reaches about half the wall - to - wall distance at @xmath303 . with center at coordinate @xmath304 in a box of width @xmath305 . full circles connected by lines : solution of ( [ eq : b10 ] ) . open circles : result ( [ eq:103 ] ) for a box of infinite width with one wall at @xmath306.,title="fig:",width=162 ] with center at coordinate @xmath304 in a box of width @xmath305 . full circles connected by lines : solution of ( [ eq : b10 ] ) . open circles : result ( [ eq:103 ] ) for a box of infinite width with one wall at @xmath306.,title="fig:",width=162 ] with center at coordinate @xmath304 in a box of width @xmath305 . full circles connected by lines : solution of ( [ eq : b10 ] ) . open circles : result ( [ eq:103 ] ) for a box of infinite width with one wall at @xmath306.,title="fig:",width=162 ] with center at coordinate @xmath304 in a box of width @xmath305 . full circles connected by lines : solution of ( [ eq : b10 ] ) . open circles : result ( [ eq:103 ] ) for a box of infinite width with one wall at @xmath306.,title="fig:",width=162 ] the signature wall effect for @xmath168 manifests itself in the form of attenuated , period-2 spatial oscillations in the @xmath58 . the exponential attenuation is governed by a coherence length that grows with the density of rods in the box . in the limit @xmath300 the oscillations persist across the box as expected . we continue with the analysis of a system of rods of size @xmath307 subject to the potential @xmath308{0mm}{5 mm } \\ 0 & : ~ i>0 , \rule[-2mm]{0mm}{5 mm } \end{array } \right.\ ] ] representing a single rigid wall at @xmath279 . we present the exact solution of eqs . ( [ eq : b2 ] ) adapted to this case , @xmath309 \prod\limits_{k = i}^{i+\sigma-1}\left[1-\!\!\!\sum\limits_{j = k-\sigma+1}^{k}\bar{n}_{j}\right]^{-1}\!\!\!.\ ] ] we have @xmath310 for @xmath311 and assume that far from the wall @xmath312 the @xmath58 approach uniformity , @xmath313 where @xmath290 is the mass density in the bulk ( average cell occupancy ) . the fugacity , the only control variable aside from @xmath8 , depends on the asymptotic solution ( [ eq:116 ] ) as follows : @xmath314^{\sigma-1 } } { \big[1-\sigma \bar{n}^{(\mathrm{mes})}\big]^\sigma } , \quad 0<\bar{n}^{(\mathrm{mes})}<\frac{1}{\sigma}.\ ] ] next we convert ( [ eq:115 ] ) into a recursion relation that expresses the solution at a given site as a function of the solutions at the @xmath315 sites immediately to its left : @xmath316,\end{aligned}\ ] ] for @xmath317 and with @xmath176 from ( [ eq:117 ] ) . this recursion relation depends on the @xmath315 parameters @xmath318 . assuming that the solution with asymptotics ( [ eq:116 ] ) is unique , these parameters can be found with a little guidance from ( [ eq:103 ] ) for the case @xmath168 . they are @xmath319^{i-1 } } { \big[1-(\sigma-1)\bar{n}^{(\mathrm{mes})}\big]^i},\quad i=1,\ldots,\sigma-1.\ ] ] ( big circles ) and @xmath320 ( small circles ) with first cell at position @xmath321 near a rigid wall at @xmath279.,title="fig:",width=162 ] ( big circles ) and @xmath320 ( small circles ) with first cell at position @xmath321 near a rigid wall at @xmath279.,title="fig:",width=162 ] in fig . [ fig : figa1 ] we show data generated recursively from ( [ eq:118 ] ) for the scaled density @xmath322 versus the scaled position @xmath323 for rods of sizes @xmath324 and @xmath320 . the latter size is meant to generate an impression of what to expect in the continuum limit . the data suggest that the dominant wall effect manifests itself again in an attenuated spatial oscillation . successive minima are approximately spaced by @xmath8 . only in the limit @xmath325 , when the attenuation disappears , are the oscillations locked into the wavelength @xmath8 . this raises the interesting question , best analyzed in the continuum limit , what the spectrum of the wall oscillations is and how it depends on @xmath165 . the continuum limit carried out for @xmath326 produces an exponential function as follows : @xmath327 this result can now be extended to @xmath328 by using a continuum version of the fugacity ( [ eq:117 ] ) , @xmath329 and a continuum version of ( [ eq:115 ] ) , @xmath330 where @xmath331 as derived percus @xcite . following vanderlick et al . @xcite , we convert ( [ eq:122 ] ) and ( [ eq : a123 ] ) into difference - differential equation , @xmath332,\ ] ] @xmath333=\bar{\nu}(x)-\bar{\nu}(x+1),\ ] ] respectively . next we integrate ( [ eq:124 ] ) and ( [ eq:125 ] ) in alternating sequence over intervals of unit length , using ( [ eq:120 ] ) and @xmath334 for @xmath126 . we thus obtain the following exact , continuous , and piecewise analytic expression for @xmath335 on successive intervals @xmath336 : @xmath337 with asymptotic value , @xmath338far from the wall , approached more and more slowly with increasing average mass density @xmath165 . the lattice dft analysis thus connects neatly with known results @xcite for monodisperse rods in a continuum . in fig . [ fig : figa2 ] we graphically compare this analytic solution ( [ eq:126 ] ) for the continuum model with the iterative solution ( [ eq:118 ] ) for the lattice model with @xmath339 . the dominant feature of the curves is an attenuated spatial oscillation . the singularities at @xmath340 become progressively weaker with increasing @xmath90 : @xmath341 is continuous at @xmath340 for @xmath342 . except for @xmath343 , the singularities do not coincide with the minima of @xmath344 . on a lattice ( big circles ) and rods of ( scaled ) unit size in a continuum ( solid lines).,title="fig:",width=162 ] on a lattice ( big circles ) and rods of ( scaled ) unit size in a continuum ( solid lines).,title="fig:",width=162 ] . ( b ) spectrum ( [ eq:152 ] ) of the attenuated oscillation ( [ eq:126 ] ) at three values of @xmath165,title="fig:",width=162 ] . ( b ) spectrum ( [ eq:152 ] ) of the attenuated oscillation ( [ eq:126 ] ) at three values of @xmath165,title="fig:",width=162 ] the limit @xmath300 @xmath345 is subtle . as the population of rods becomes more crowded , they begin to line up with increasing probability near the integer positions as illustrated in fig . [ fig : fig17](a ) . in the limit @xmath300 , maxima and minima approach the singularity values @xmath340 in pairs from opposite sides . the maxima diverge and the minima approach zero . the area under the curve between successive singularities approaches unity . the shape of the curve approaches an of infinite height and unit width , effectively the function @xmath346 in fig . [ fig : fig17](b ) we show the spectrum of the attenuated oscillation ( [ eq:126 ] ) , @xmath347\cos(qx).\ ] ] this quantity features a peak at wave number @xmath348 emerging from a broad and flat background . the fourier transform ( [ eq:152 ] ) and the limit @xmath349 are not interchangeable operations . with increasing @xmath165 the peak becomes taller and sharper as it moves toward commensurability at @xmath350 . the dft mass density in the continuum , calculated from ( [ eq:126 ] ) via a continuum version of ( [ eq:101 ] ) , @xmath351 becomes @xmath352 on successive intervals @xmath336 . in fig . [ fig : fig18 ] we show profiles of @xmath353 for the corresponding to the profiles of @xmath344 shown in figs . [ fig : figa2 ] and [ fig : fig17 ] . the oscillatory wall effects are somewhat milder in @xmath353 than in @xmath344 but still fairly conspicuous . in the limit @xmath349 the oscillations in @xmath344 diverge as in ( [ eq:151 ] ) but disappear in @xmath353 , which becomes constant . .,title="fig:",width=162 ] .,title="fig:",width=162 ] in this work we have investigated the equilibrium statistical mechanics of monodisperse hard rods confined by external potentials , producing characteristic spatial profiles of density and pressure . the rods occupy @xmath8 consecutive cells of volume @xmath0 in a linear array . the continuum limit is implemented as @xmath4 , @xmath3 with @xmath1 kept finite and nonzero . the rods interact via hardcore exclusion forces . contact forces have also been considered . we have been employing two distinct methods of analysis , density functionals and generalized exclusion statistics , with partially overlapping domains of applicability and somewhat complementary strengths . the usefulness , soundness , and consistency of the two approaches has been demonstrated in a series of applications that include confinement by rigid walls , gravity , and power - law traps . we have shown that profiles of density , for example , exhibit characteristic features on a mesoscopic length scale that are reproduced identically by both approaches and , for @xmath156 , additional features on a microscopic length scale that are only resolved in the dft analysis . the dft pressure profiles are also compared with profiles of average microscopic pressure inferred from pair distribution functions in the context of one application . finally , we have established contact between our results for the lattice system and prior results for corresponding continuum systems , demonstrating full consistency the statistical mechanical analysis of hard rods of mixed sizes presents itself as a natural and promising extension of this work . significant ground has already been broken via both calculational tools used in this work . the mathematical structure of exact density functionals for hard - rod mixtures has recently been established @xcite and is awaiting analysis in specific applications . hard rod ensembles of mixed sizes on a lattice are mathematically equivalent to magnetic domains in ising chains . the combinatorics and statistical mechanics of such domains analyzed via generalized exclusion statistics have been in place for some time @xcite and is awaiting extensions to heterogeneous environments and new applications of current interest . the completion of the work reported here has paved the way for projects along these lines @xcite . in terms of the particle ( hard rod ) momenta @xmath354 and positions @xmath355 , the microscopic momentum density in eq . ( [ eq : momentum - flow ] ) is given by @xmath356 . the time derivative of its fourier transform @xmath357 reads ( with @xmath138 the mass of the rods , and assuming only pair interaction forces to be present ) @xmath358 here @xmath359 is the force of the @xmath90th on the @xmath360th particle ( @xmath361 ) , @xmath362 the external force on the @xmath360th particle , and we have used @xmath363 and newton s equations @xmath364 . for interaction forces obeying the principle of action and reaction , @xmath365 , the second term on the right hand side of eq . ( [ eq : dpidt ] ) can be written in the form @xmath366 $ ] , where @xmath367 and @xmath368 . comparing eq . ( [ eq : dpidt ] ) with the fourier - transformed right hand side of eq . ( [ eq : momentum - flow ] ) , the microscopic pressure in fourier space becomes @xmath369 to obtain the corresponding expression in real space , we calculate @xmath370 where @xmath371\,,\\[1ex ] 0\ , , & \textrm{else}\,.\end{array}\right.\end{aligned}\ ] ] accordingly , @xmath372 using @xmath373 and eq . ( [ eq : g - function ] ) , the double sum can be written as @xmath374 , and the equilibrium average of eq . ( [ eq : pmic ] ) then yields eq . ( [ eq : p - rho2 ] ) . consider a system of rods of size @xmath8 confined to a finite array of cells numbered @xmath375 . the hardcore repulsion is the only interaction between rods . the external potential @xmath20 is arbitrary . the free energy functional ( [ eq : freeenergy - dft ] ) simplifies into @xmath376 where @xmath377 are the ( yet undetermined ) probabilities of rods at the allowed positions . the minimization ( [ eq : structure - equations ] ) then leads to the following relations that determine the exact density profile of rods , @xmath220 , for an any given external potential @xmath20 at given temperature @xmath32 and chemical potential @xmath30 : @xmath378}{\prod\limits_{k = i}^{i+\sigma-1}\left[1-\sum\limits_{j = k-\sigma+1}^{k}\bar{n}_{j}\right]},\ ] ] for @xmath379 with @xmath310 for @xmath16 and @xmath380 implied . in some applications we use a semi - infinite array @xmath381 with the second boundary condition replaced by a prescribed ( zero or nonzero ) limit @xmath382 . in the following we use the control variables , @xmath383 with the implication that @xmath384 for @xmath16 and @xmath380 . for rods of size @xmath152 the hardcore repulsion does not produce any interference between the @xmath58 at different positions . equations ( [ eq : b2 ] ) remain uncoupled . the density profile @xmath58 of rods , which , in this case , coincides with the mass density profile @xmath59 , reads @xmath385 its only structure is that imposed by the potential @xmath20 . in the case @xmath168 a rod at position @xmath60 obstructs the placement of a rod positions @xmath386 and @xmath179 . this interference is reflected in eqs . ( [ eq : b2 ] ) , which now read @xmath387 with @xmath388 implied . we solve the coupled eqs . ( [ eq : b5 ] ) by a strategy that also works for @xmath190 as we shall see . we reduce the set of nonlinear equations into two sets of recursion relations to be solved in sequence . this method has the benefit of isolating the physical solution . we begin by introducing the auxiliary quantities , @xmath389 we thus convert ( [ eq : b5 ] ) into @xmath390 from which we determine the @xmath173 recursively : @xmath391 the @xmath58 are then , in turn , determined recursively via ( [ eq : b6 ] ) : @xmath392 producing the explicit form @xmath393 a different rendition of that solution is derived from ( [ eq : b9 ] ) with use of ( [ eq : b8 ] ) : @xmath394 where the @xmath395 are generated recursively : @xmath396 if the external potential is symmetric under under reflection , as in power - law traps or boxes with rigid walls , we have @xmath397 , @xmath379 . the density profile of rods must then also exhibit that symmetry . to make this symmetry transparent we recognize that we have @xmath398 , @xmath379 , under these circumstances . we can then transform @xmath399 into @xmath58 as follows : @xmath400 now we generalize this method to rods of unrestricted size on the lattice : @xmath2 . a rod of size @xmath8 at position @xmath60 obstructs the placement of rods at positions @xmath401 . the right - hand side of ( [ eq : b2 ] ) contains @xmath58 at @xmath402 consecutive positions . the coupled equations now involve polynomials of order @xmath8 . the boundary conditions require that we set @xmath310 for @xmath403 and @xmath404 in these @xmath405 coupled equations . the auxiliary quantities @xmath406^{-1},\quad i=1,\ldots , m \ ] ] then convert ( [ eq : b2 ] ) into the set of recursion relations @xmath407 for given @xmath173 the @xmath58 then follow recursively from ( [ eq : b14 ] ) : @xmath408 consider a system of @xmath76 hard rods of size @xmath8 at positions @xmath409 , @xmath410 on a 1d lattice of @xmath411 sites . the system is confined by hard walls at positions @xmath412 and @xmath413 . the lattice partition function @xmath414 with variable change @xmath415 and excess volume defined as @xmath416 becomes @xmath417 and , after summation over @xmath418 , @xmath419 the density and pair distribution functions are then expressible as follows in terms of partial partition functions @xmath420 representing systems of @xmath421 rods confined by hard walls at positions @xmath422 and @xmath90 @xcite : @xmath423 @xmath424 a lattice version of expression ( [ eq : p - rho2-non - interacting ] ) for the average microscopic pressure to be used in sec . [ sec : grav - micro ] for comparison with the dft pressure profiles thus reads @xmath425

numerous observations , from galaxy rotation curves to gravitational lensing , to cosmic microwave background radiation , all point to the existence of dark matter which is not made of ordinary atoms , but , rather , of some new , yet undiscovered particles . in particular , the abundance of dark matter has been precisely determined recently by the wmap @xcite . however , the nature of dark - matter particles remains unknown . one attractive candidate is a sterile neutrino with mass of several kev and a small mixing with the ordinary neutrinos . if such a particle exists , it could be produced in the early universe with the right abundance to be dark matter either from neutrino oscillations @xcite , or from some other processes , for example from inflaton decay @xcite . the same particle can explain the observed velocities of pulsars because its emission from a cooling neutron star would be anisotropic , hence providing the neutron star with a recoil velocity of a sufficient magnitude @xcite . in addition , the neutrino kick can enhance the convection during the first second of the supernova explosion , which can increase the energy of the supernova shock and bring the supernova calculations in better agreement with the observations @xcite . sterile neutrinos can also help the formation of supermassive black holes in the early universe @xcite . although the dark - matter sterile neutrinos have interactions that are too weak to be discovered in the laboratory , some other sterile neutrinos may exist . most models of neutrino masses introduce sterile ( or right - handed ) neutrinos to generate masses of the ordinary neutrinos via the seesaw mechanism @xcite . the name _ sterile _ was coined by @xcite . many seesaw models assume that sterile neutrinos have very large masses , which makes them unobservable . however , a number of theoretical models predict the existence of sterile neutrinos at or below the electroweak scale @xcite . there is an indication that some light sterile mass eigenstates may cause the neutrino flavor transformation seen in the los alamos liquid scintillator neutrino detector ( lsnd ) experiment @xcite . the results of lsnd are being tested by the miniboone experiment . one can expect that there are at least three light sterile neutrinos @xcite , in which case the neutrino oscillations can explain the baryon asymmetry of the universe @xcite . although dark - matter sterile neutrinos are stable on cosmological time scales , they nevertheless decay . the mass matrix is assumed to have the seesaw form with the majorana mass term of the order of several kev and with a much smaller dirac mass term . the diagonalization of such a mass matrix yields the light mostly active majorana mass eigenstates and the heavy sterile majorana eigenstates , which can decay into the light ones . the dominant decay mode , into three light neutrinos , is `` invisible '' because the daughter neutrinos are beyond the detection capabilities of today s experiments . the most prominent `` visible '' mode is decay into one active neutrino and one photon , @xmath0 . assuming two - neutrino mixing for simplicity , one can express the inverse width of such a decay as @xmath1 where @xmath2 is the mass and @xmath3 is the mixing angle . for mass units we set @xmath4 . since this is a two - body decay , the photon energy is half the mass of the sterile neutrino . the monochromatic line from dark matter decays can , in principle , be observed by x - ray telescopes . no such observation has been reported , and some important limits have been derived on the allowed masses and mixing angles @xcite . these constraints are based on different astrophysical objects , from virgo and coma clusters , to large magellanic clouds , to the milky way halo and its components , etc . there are different uncertainties in modeling the dark matter populations in these objects . different groups have also used very different methodologies in deriving these bounds : from a conservative assumption that the dark - matter line should not exceed the signal , to more ambitious approaches that involved modeling the signal or merely fitting it with a smooth curve and requiring that the line - shaped addition not affect the quality of the fit . the x - ray limits are usually shown based on the assumption that the sterile neutrinos constitute the entire dark matter ( @xmath5 , where @xmath6 and @xmath7 refer to the sterile neutrinos and the dark matter , respectively ) . however , the x - ray observations exclude certain masses and mixing angles even if the sterile neutrinos make up only a fraction of dark matter . the production via mixing @xcite provides the lowest possible abundance of sterile neutrinos . in fig . [ fig : models ] we show both the bounds based on the assumption @xmath5 and the model - independent exclusion region @xcite based on the production only via the @xcite mechanism . @xcite used the analytical fit to the numerical calculation of @xcite , and re - scaled the x - ray flux in proportion to the amount of relic neutrinos produced for a given mass and mixing angle . a different set of bounds comes from the observations of the lyman - alpha forest @xcite , which limit the sterile neutrino mass from below . based on the high - redshift data from sdss and some modeling of gas dynamics , one can set a limit as strong as 14 kev @xcite . however , the high - redshift data have systematic errors that are not well understood @xcite , and more conservative approaches , based on the relatively low - redshift data , have led to some less stringent bounds @xcite . more recently , @xcite have reanalyzed the high - redshift data and arrived at the bound @xmath8 kev . the mass bound as quoted depends on the production mechanism in the early universe . the lyman - alpha observations constrain the free - streaming lengths of dark matter particles , not their masses . for each cosmological production mechanism , the relation between the free - streaming length and the mass is different @xcite . for example , the bound @xmath9 kev @xcite applies to the production model due to @xcite . even within a given cosmological scenario , there are uncertainties in the production rates of neutrinos for any given mass and mixing angle @xcite . these uncertainties may further affect the interpretation of the lyman - alpha bounds in terms of the sterile neutrino mass . the x - ray bounds depend on a combination of mass and mixing angle as in eq . ( [ meanlife ] ) , regardless of one s cosmological assumptions , as long as all the dark matter is comprised of sterile neutrinos . however , the x - ray bounds are subject to uncertainties in modeling the dark matter populations of different objects . if the sterile neutrinos make up only a part of dark matter , none of the above bounds apply . also , if inflation ended with a low reheating temperature , the bounds are significantly weaker @xcite . it should also be mentioned that the lyman - alpha bounds appear to contradict the observations of dwarf spheroidal galaxies @xcite , which suggest that dark matter is warm and which would favor the 15 kev mass range for sterile neutrinos . there are several inconsistencies between the predictions of n - body simulations of cold dark matter ( cdm ) and the observations @xcite . each of these problems may find a separate independent solution . perhaps , a better understanding of cdm on small scales will resolve these discrepancies . it is true , however , that warm dark matter in the form of sterile neutrinos is free from all these small - scale problems altogether , while on large scales wdm fits the data as well as cdm . several independent considerations , from the pulsar kicks to small - scale behavior of dark matter , favor sterile neutrinos as the dark matter candidate . it is important , therefore , to examine the effects of sterile - neutrino dark matter on star formation and reionization of the universe . @xcite have attempted to study such effects for some kind of warm dark matter , but they have used a linear matter power spectrum and assumed that the dark matter particles have a thermal distribution , which is not the case for sterile neutrinos @xcite . even more importantly , they have not considered the effects of sterile neutrino decays and the resulting x - ray background . it was shown by @xcite that the x - rays produced by the decays of relic sterile neutrinos can speed up the formation of molecular hydrogen , which can help the cooling of the gas and the star formation , which can lead , in turn , to a reionization of the universe at the redshift consistent with the wmap results . more specifically , the wmap ( three years ) measurement of the reionization redshift @xmath10 @xcite has posed a challenge to theories of star formation . on the one hand , stars have to form early enough to reionize gas at redshift 11 . on the other hand , the spectra of bright distant quasars imply that reionization must be completed by redshift 6 . stars form in clouds of hydrogen which collapse at different times , depending on their sizes : the small clouds collapse first , while the large ones collapse last . if the big clouds must collapse by redshift 6 , then the small halos must undergo the collapse much earlier . it appears that the star formation in these small halos would have occurred at high redshift , when the gas density was very high , and it would have resulted in an excessive thompson optical depth @xcite . to be consistent with wmap , the efficiency for the production of ionizing photons in minihalos must have been at least an order of magnitude lower than expected @xcite . one solution is to suppress the star formation rate in small halos by some dynamical feedback mechanism . the suppression required is by at least an order of magnitude . alternatively , one can consider warm dark matter , in which case the number of small clouds is strongly suppressed . however , for some wdm candidates ( for example , the gravitino ) , the lack of power on small scales in the spectrum of density perturbations leads to an unacceptable delay in the onset of star formation @xcite . this is not the case for sterile neutrinos , which decay slowly during the `` dark ages '' preceding the birth of stars . the decay photons could ionize gas , and , although this ionization is too weak to affect the wmap measurements directly @xcite , the increase in the ionization fraction could cause a rapid production of molecular hydrogen at redshifts as high as 100 @xcite . molecular hydrogen is the essential cooling agent that allows the gas to cool and collapse to form the first stars . @xcite have not considered the effects of gas heating associated with the decays , which may , in some cases , thwart the effects of ionization . the increase in the fraction of molecular hydrogen can help cooling the gas , but if the heating from sterile neutrino decays is significant , it can overcome the radiative cooling and stymie the collapse and star formation . in this paper we will examine thermal evolution of gas clouds in detail , taking into account both effects of the sterile neutrino decays , namely the ionization and the heating of gas . the paper is organized as follows . in section [ sec_h2 ] we will discuss the role of sterile neutrino decays and the production of molecular hydrogen in the cooling of the primordial gas clouds . next , in sections 3 - 6 we will briefly describe the top - hat overdensity simulation . the results are shown in section [ sec_results ] . the existence of the first objects is a direct consequence of the growth of the primordial density fluctuations . at the beginning , there are linear density perturbations which expand with the overall hubble flow . subsequently , these perturbations can grow and form primordial clouds . clouds with enough density contrast decouple from this flow and start to collapse . the kinetic energy of the infalling gas is dissipated through shocks and the cloud becomes pressure supported . the subsequent evolution of the cloud is determined by its ability to cool sufficiently fast . clouds that do not cool fast enough stay in a pressure - supported state and do not form stars . the existence of the efficient cooling mechanism is necessary to continue the collapse of the cloud , its subsequent fragmentation and star formation . in the absence of metals , the most important cooling mechanism is cooling by hydrogen atoms and molecules . this mechanism involves the collisional excitation of h or h@xmath11 , subsequent spontaneous de - excitation , and photon emission . the emitted photons can escape and take away energy , hence reducing the kinetic energy of the collapsing cloud . cooling by h dominates for temperatures above @xmath12 k , while the h@xmath11 cooling is more important at lower temperatures . the temperature of the primordial cloud can be higher than @xmath12 k only during a short period of virialization of the baryonic gas . therefore , h@xmath11 cooling is the most important cooling mechanism during almost entire evolution of the halos , and it is crucial for the star formation . in primordial gas clouds , h@xmath11 molecules can form mainly through the coupled reactions @xmath13 or @xmath14 in which @xmath15 and @xmath16 act only as a catalyst . x - ray radiation can increase the production of the h@xmath11 by enhancement of the ionization fraction . as discussed above , sterile neutrinos with the mass of several kev and a small mixing are stable on cosmological time scales . nevertheless some of them can decay . the decay channel important for us is @xmath17 its inverse width is given in equation ( [ meanlife ] ) . the energy of the photon is equal to @xmath18 where @xmath19 is the mass of the active neutrino . x - ray photons from the sterile neutrinos decays can increase the ionization fraction and the h@xmath11 production . this can subsequently lead to a speed - up of the gas cooling and star formation . we will follow the evolution of the baryonic top - hat overdensity , the gas temperature and its h@xmath11 and @xmath15 fraction . our goal is to juxtapose the evolution of the gas temperature in the primordial clouds in the @xmath20cdm model and the wdm model with kev sterile neutrinos . we assume that the dark matter is smoothly distributed in the region we are considering . we will use the parameters from the best fit to the three years wmap data @xcite , namely @xmath21 km @xmath22 @xmath23 , @xmath24 , @xmath25 , @xmath26 , @xmath27 , @xmath28 , @xmath29 . let us take some spherical region with the top - hat overdensity @xmath30 . it is well known that such an overdensity will evolve according to the following equation @xcite @xmath31 where the parameter @xmath32 is related to the redshift @xmath33 and redshift of virialization @xmath34 through @xmath35 further evolution depends on the type of the matter in the overdense region . if it is the dark matter then after virialization its density remains constant forever . the situation is different in the case of baryons . if there exists an efficient cooling mechanism , then the density gradually increases . otherwise , the density remains constant , and there is no star formation in the halo . in reality an overdense region consists of two components namely , baryonic gas and dark matter . baryonic density fluctuations are suppressed before recombination . in contrast , the dark matter density fluctuations can start to grow well before this period , so its initial overdensity is larger than that of the baryons . dark matter forms the potential wells into which the baryonic matter falls . overall , the baryon density follows the dark matter density in the early universe . therefore , one can consider the primordial fluctuations as the halos consisting of baryonic and dark matter , whose overall overdensity follow the top - hat solution ( eqs . [ th1 ] and [ th2 ] ) . of course , this assumption breaks down at the later stages of the halo evolution , e.g. after the virialization of the dark matter . following @xcite , let us assume that the evolution of the baryonic and dark matter density is described by equations ( [ th1 ] ) and ( [ th2 ] ) until @xmath32 reaches @xmath36 . this occurs at the redshift @xmath37 after this redshift , we assume that the density of the halo @xmath38 stays constant e.g. it is equal to the value @xmath39 where @xmath40 is the critical density of the universe . we note that the overdensity @xmath30 in equation ( [ th1 ] ) is given by @xmath41 where @xmath42 is the combined average density of dark and baryonic matter . we have taken the ratio of the baryonic to the dark matter mass in the halo to be the standard cosmological ratio of @xmath43 . the density of the baryonic matter can be expressed as @xmath44 the gas temperature evolution is governed by the equation of energy conservation @xmath45 where @xmath46 , @xmath47 and @xmath20 are the pressure , the internal energy per unit mass and the cooling / heating function respectively . we use the equation of state of perfect gas @xmath48 where @xmath49 is the adiabatic index . the internal energy per unit mass can be described as @xmath50 where @xmath51 is the boltzmann constant , @xmath52 is the gas temperature , @xmath53 denotes the molecular weight , and @xmath54 is the mass of the hydrogen atom . introducing the number density of non dark matter particles @xmath55 we can write @xmath56 as @xmath57 if we use eqs . ( [ p ] ) , ( [ u ] ) and ( [ p2 ] ) than we can cast eq . ( [ u1 ] ) in the form @xmath58 the time @xmath59 can be translated to the redshift @xmath33 in the following way @xmath60 using eqs . ( [ tt ] ) and ( [ hubble ] ) , we obtain @xmath61 to calculate the gas temperature of the primordial cloud we have to integrate this equation . one can assume that the gas temperature evolution is governed by eq . ( [ tz ] ) except for the redshift range between @xmath62 and @xmath34 . if we simply integrate eq.([tz ] ) to the redshift @xmath34 , than the gas temperature will be much lower than the virial temperature @xmath63 , at least for clouds with mass greater than some critical value @xcite @xmath64 where @xmath65 , @xmath66 and @xmath67 are respectively the mean molecular weight during virialization , the halo mass and the overdensity of the halo at virialization given by @xcite , which is very well approximated by @xmath68 for redshifts @xmath69 . therefore we have to take into account shocks and increase the gas temperature to the virial value . it is reasonable to assume that the gas temperature evolution is linear between the redshift @xmath62 and @xmath34 and is described by the equation @xmath70 where @xmath71 is the temperature at the redshift @xmath62 . let us consider the following five species : h , h@xmath72 , @xmath73 , @xmath15 , and @xmath74 with the mass fraction @xmath75 @xcite . let us denote by @xmath76 ( @xmath77h , h@xmath72 , h@xmath11 , @xmath15 and he ) the number density of baryonic component @xmath78 . we assume that @xmath79 e.g. there is no he or h@xmath11 ionization . the number density of hydrogen can be written as @xmath80 if we introduce the number fraction of component `` @xmath78 '' @xmath81 then the particle number density @xmath55 and the baryon density @xmath56 are equal to @xmath82 and @xmath83 where @xmath84 is the hydrogen mass fraction . using eqs . ( [ p2 ] ) , ( [ nbb ] ) and ( [ rob ] ) we can get @xmath85 at the virialization we have typically @xmath86 and @xmath87 , so one can take the mean molecular weight at this time to be @xmath88 . we can express @xmath49 as @xmath89 where @xmath90 and @xmath91 . to calculate the cooling / heating function @xmath20 in eq . ( [ tz ] ) we have to take into account all of the relevant chemical and thermal processes and follow the number density evolution of different baryonic components . all chemical and thermal processes included in our calculation are summarized in tables [ tab : reaction ] and [ tab : cooling ] . the evolution of @xmath92 and @xmath93 is described by equations @xmath94 @xmath95 where @xmath96 is the production rate of free electrons by the sterile neutrino decays which is calculated in the appendix . in eq . ( [ xe ] ) only the ionization and recombination of hydrogen are included . all the rates @xmath97 are given in table [ tab : reaction ] . we note in passing that neglecting he and h@xmath11 ionization renders the system with fewer free electrons , so the production of h@xmath11 is slower . in addition , cooling by h@xmath98 is stronger than by h@xmath11 ( see last figure in @xcite ) . therefore , the cooling we get in this approximation is somewhat less efficient than it should be in reality . the molecular hydrogen number fraction can be calculated by integration of the following equation @xmath99 where @xmath100 is the molecular hydrogen formation rate via @xmath101 and h@xmath98 channel @xcite . to get eq . ( [ km ] ) we have to make the assumption that after formation of @xmath101 and @xmath102 they are used almost instantaneously in the next reactions ( reactions number 2 and 3 or 5 and 6 respectively , see table [ tab : reaction ] ) . this is a very good approximation , because in reality the @xmath101 and @xmath102 fractions always remain small . the cooling / heating function @xmath20 is given in the appendix ( eq . [ coolfunc ] ) . it includes contributions from the following processes : 1 . excitation of h by collisions with @xmath15 2 . recombination to h 3 . ionization of h by collisions with @xmath15 4 . photo - ionization of h by the cmb radiation 5 . excitation of h@xmath11 by collisions with h and h@xmath11 6 . h@xmath11 formation cooling 7 . dissociation of h@xmath11 by collisions with h , h@xmath11 and @xmath15 8 . photo - dissociation of h@xmath11 by cmb photons 9 . compton cooling 10 . bremsstrahlung 11 . photo - ionization of h by the photons from the sterile neutrino decays and h ionization by the secondary electrons produced due to these photons we have not included the effect of the photo - dissociation of h@xmath11 by the photons from the sterile neutrino decays . this is because the flux of these photons is small in comparison with the cmb radiation in the 10 - 30 ev range , which is the energy range relevant to the photo - dissociation of h@xmath11 . the formulae for the contributions to the @xmath20 function from all of these processes are given in table [ tab : cooling ] . we have assumed the ortho - h@xmath11 to para - h@xmath11 ratio to be 3 : 1 . we start our calculations at the redshift of recombination @xmath103 , taking as initial conditions @xmath104 , @xmath105 and assuming that the gas temperature is equal to the cmb temperature . evolution of the halo is insensitive for the initial value of @xmath106 . we follow the evolution of the density , eqs . ( [ th1 ] ) and ( [ th2 ] ) , and the temperature , eq . ( [ tz ] ) with the cooling / heating function given in eq . ( [ coolfunc ] ) untill the redshift is equal to @xmath62 , _ cf . _ ( [ z3pi ] ) . from that time we assume that the density of the halo stays constant and is equal to its virial value ( [ densvir ] ) . we assume that the gas starts to be shock heated in this period and reaches its virial temperature at @xmath34 . the temperature of the gas is assumed to change linearly with the redshift between @xmath62 and @xmath34 , eq . ( [ telin ] ) . for the redshift lower than @xmath34 the gas temperature is calculated once again by integration of eq . ( [ tz ] ) . following @xcite , we use the following criterion for the successful collapse : @xmath107 where @xmath108 @xcite . the results of our analysis depend , of course , on the parameters we choose for the dark - matter sterile neutrinos . the x - ray flux , and the resulting increase in the ionization fraction depend on the combination ( @xmath109 ) . the power spectrum on small scales , which determines the presence or absence of small clouds , depends on the free - streaming length , which is set by the mass ( in every given cosmological scenario ) . based in part on the existing limits , we have chosen four illustrative benchmark cases to illustrate the effects of ionization and heating due to sterile neutrino decays on the cloud collapse . we have compared the following four models : 1 . cold dark matter with no additional radiation besides cmb ( cdm ) . warm dark matter ( wdm ) with the sterile neutrino mass @xmath110 kev and mixing angle @xmath111 ( wdm1 ) . 3 . wdm with @xmath112 kev and @xmath111 ( wdm2 ) . 4 . wdm with @xmath113 kev and @xmath114 ( wdm3 ) . the values of parameters chosen for these models are shown in fig.[fig : models ] . some comments are in order . the choice of wdm1 is obvious because it satisfies all the constraints , including the x - ray bounds and the lyman - alpha bounds . the masses as large as @xmath110 kev can not be constrained by either chandra or xmm , and the best constraint comes from heao , which is weaker because of its lower energy resolution @xcite . also , wdm1 corresponds to a very small free - streaming length and , on large scales , it creates structure indistinguishable from cdm . the only difference between wdm1 and cdm is the additional x - ray flux . therefore , the comparison of cdm and wdm1 cases provides a clean test of the effects of the increased ionization fraction @xcite . also , wdm1 provides a realistic set of parameters consistent with all the bounds , as well as the pulsar kick mechanism due to resonant active - sterile neutrino conversions @xcite . wdm2 is also consistent with all the existing limits , including the most ambitious lyman - alpha bounds @xcite . ( we remind the reader that @xcite have obtained a 50% weaker limit using the same data . ) the x - ray background is lower in this case than in the case of wdm1 . finally , wdm3 corresponds to masses that would be ruled out by the `` strong '' lyman - alpha bounds obtained by @xcite and by @xcite . however , one has to remember that the relation between the masses and the free - streaming length depends on the production mechanism of sterile neutrinos . for example , the free - streaming length in the @xcite production mechanism is different from the @xcite mechanism . even within the latter framework , additional entropy production can redshift the sterile neutrino momenta and eliminate the lyman - alpha bound @xcite . if dark matter is produced from a coupling between the sterile neutrinos and the inflaton @xcite , the free - streaming length is , again , different . wdm3 requires redshifting of the neutrino momenta by a factor @xmath115 , as compared to the thermal distribution . this can be achieved , for example , by entropy production of the order of @xmath116 , which is comparable to the effect of change in the number of degrees of freedom from the electroweak scale to zero temperature @xcite . hence , wdm3 should not be considered an unrealistic scenario . in addition , wdm3 is of interest because it is consistent with the pulsar kick mechanism via off - resonant active - sterile neutrino oscillations @xcite , even if this sterile neutrino is not the dominant component of dark matter . the results of the calculation are presented in figures [ fig : t ] , [ fig : xe ] and [ fig : xh2 ] . several conclusions can be drawn from these results . we find that large clouds , with masses greater than @xmath117 , collapse regardless of the details of cooling in all the models defined above . one could expect this behavior because the virial temperature of such a cloud is higher than @xmath12 k , so the h cooling mechanism is efficient enough and the additional h@xmath11 does not play an important role . there is , however , a dramatic difference when it comes to smaller clouds . for a virialization redshift @xmath118 , a cloud of @xmath119 collapses in model wdm1 , but not in the cdm or other wdm models . this illustrates the importance of the increased h@xmath120 fraction on the cooling of gas . in wdm2 and wdm3 model , because of the relatively low associated x - ray flux , the temperature evolution is practically unaffected by the photons from the dark matter decay . from fig . [ fig : xh2 ] one can draw another conclusion . in the large clouds , all of the h@xmath11 molecules are destroyed during the virialization process . this is because of the very high virial temperature . in contrast , there is a drastic increase in h@xmath11 fraction during the virialization of the clouds with the mass of @xmath121 . in both cases , most of molecular hydrogen is produced after virialization . for clouds with @xmath122 and @xmath123 the critical h@xmath11 fraction ( see eq . ( 11 ) of @xcite ) is reached at virialization . on the other hand , in the clouds having equivalent mass with @xmath118 , the molecular hydrogen faction @xmath106 attains its virial critical value only for wdm1 model . this happens at @xmath124 . in all the wdm models the ionization and the molecular hydrogen fraction are enhanced in comparison with the cdm model , as suggested by @xcite . in addition , we can see that in wdm models the temperature of the collapsing clouds soon after virialization is lower then that in the cdm case , which speeds up the collapse , at least for the low - mass clouds . we note in passing that the density of the cloud was kept constant after virialization , so the evolution of @xmath52 , @xmath92 and @xmath106 is reliable for the redshifts close to virialization , but not for much smaller redshifts . it appears that wdm3 model is disfavored . the increase in the cooling rate is almost negligible in comparison with the cdm case , while the suppression on the small scales is very strong . therefore , this leads to an unacceptable delay in the onset of star formation , as discussed by @xcite . in the case of wdm1 and wdm2 models , the cooling is enhanced by the additional fraction of molecular hydrogen . however , we can not definitively prove that star formation occurs at the redshift consistent with wmap because of the simplifying assumptions and limitations of our analysis . the main limitation is that the virialization redshift of a particular cloud is a fixed parameter in the simulation . in reality , it can be changed by the enhancement of the cooling , so the cloud can reach its virial stage much earlier . moreover , the density of the cloud is kept constant after virialization , which is a simplifying assumption . we can not calculate the real redshift of the cloud collapse and onset of the star formation . as a result , we can not calculate how many clouds collapse at a given redshift and we can not calculate the redshift of reionization in wdm model reliably . also , we have not included the additional x - rays from the decay of sterile neutrinos inside the collapsing clump . this radiation can cause an additional increase in ionization and heating , which could change our results . also the effects of opacity can be important in large clouds , which we have not taken into account . these limitations will be addressed in the future work . we have performed a detailed analysis of the cooling and collapse of primordial gas in the model with warm dark matter , taking into account both the increase in the fraction of molecular hydrogen @xcite and the heating due to the sterile neutrino decay . as expected , the effect on the largest gas clouds is negligible , but the smaller clouds are , in fact affected . we have performed the analysis for some benchmark cases which arise in realistic scenarios . for the largest clouds , the additional molecular hydrogen made no difference . for smaller clouds , especially for those with masses @xmath125 , the increase in the x - ray background made the successful collapse possible in cases where it could not occur in the absence of sterile neutrino decays . we thank g. gilmore , m. mapelli , m. shaposhnikov , and s. stachniewicz for very helpful discussions and comments . the work of p.l.b . and j.s . was supported by the pierre auger grant 05 cu 5pd1/2 via desy / bmbf . the work of a.k . was supported in part by the doe grant de - fg03 - 91er40662 and by the nasa atp grants nag 5 - 10842 and nag 5 - 13399 . thanks the cern theory unit and epfl for hospitality during his visit . thanks the organizers of marcel grossmann meeting for support . in this section we shall calculate the photon spectrum produced by the radiative decays of the sterile neutrinos ( our derivation is similar to the one in @xcite ) . the dominant channel of radiative decays is through the following reaction @xmath126 where @xmath127 is the sterile neutrino and @xmath128 is an ordinary active neutrino . the photon energy is given by @xmath18 where @xmath129 and @xmath19 are respectively the mass of the sterile and an active neutrino . the cosmic expansion redshifts the photon energy . a photon that at redshift @xmath33 has energy @xmath130 was produced at redshift @xmath131 given by @xmath132 let @xmath133 be the energy flux ( in units of erg @xmath134 @xmath135 ) at redshift z of photons with energy e produced by the sterile neutrino decays . the flux per unit energy and solid angle is given by @xmath136 where @xmath137 is the photon flux at redshift z. it is related to the photon flux at emission redshift @xmath131 through the following equation @xmath138 where we have included the factor of @xmath139 produced by the expansion of the universe . the photon density emitted at @xmath131 is given by the usual decay law @xmath140 where @xmath141 is the mean sterile neutrino lifetime for the decay into an active neutrino and a photon ( see eq . [ meanlife ] ) , @xmath142 is the number density of the sterile neutrino at redshift @xmath131 and @xmath143 is given by @xmath144 where @xmath145 is the hubble expansion rate at photon emission redshift @xmath131 . writing everything together we obtain @xmath146 choosing the time @xmath147 otherwise arbitrary we can write @xmath148 where @xmath149 is the age of the universe at redshift @xmath131 , @xmath150 and @xmath151 would be the number density of sterile neutrinos at redshift @xmath33 if they did not decay . assuming that all of the dark matter consists of sterile neutrinos , @xmath151 can be written as @xmath152 where @xmath153 in the case of a flat universe filled with matter and a nonvanishing cosmological constant the age of the universe at redshift @xmath33 is given by @xcite @xmath154 taking @xmath155 and substituting eq . ( [ neu ] ) into eq . ( [ ff1 ] ) we can get the specific intensity ( in units of erg @xmath134 @xmath135 @xmath156 @xmath157 ) of the photons with frequency @xmath158 from the sterile neutrino decays at redshift @xmath33 @xmath159 we have to multiply this value by @xmath160 in order to correct specific intensity for the absorption by the background neutral hydrogen atoms . the optical depth is given by @xmath161 where @xmath162 and @xmath163 are the cross section for the h ionization at energy @xmath164 and the hydrogen number density of the universe at redshift @xmath165 respectively ; @xmath164 is the energy of the photon at redshift @xmath165 , which was emitted in sterile neutrino decay at redshift @xmath131 and is given by the equation @xmath166 the mean hydrogen number density of the universe at redshift @xmath165 can be written as @xmath167 \left(1+z'\right)^3 \textrm { , } \label{ap2}\ ] ] where @xmath168 is its present value and is equal to @xmath169 the cross section for the h ionization is given by @xmath170 where @xmath171 @xmath172 and @xmath173 ev . neglecting the @xmath174 term in eq . ( [ hubble ] ) , which is a very good approximation at high redshifts , and using eqs . ( [ energy ] ) , ( [ ap1 ] ) , ( [ ap2 ] ) and ( [ ap4 ] ) we can cast eq . ( [ abs ] ) into the form @xmath175\left(1+z'\right)^{-5/2 } dz ' \textrm{.}\ ] ] as @xmath176 , let us take @xmath177 where @xmath178 is the recombination redshift . it gives us @xmath179 \textrm{.}\ ] ] photons emitted before recombination are compton scattered and thermalized . for simplicity we assume that these photons are not visible . the final specific intensity is equal to @xmath180 where @xmath181 @xmath182 we have assumed that the photons are absorbed only by the ly@xmath32 wedge , so the above equations are valid only for @xmath183 ev . to derive the photon flux at energy lower than 13.6 ev we have to take into account another hydrogen absorption lines . the radiation with energy below 13.6 ev start to be important only at lower redshifts ( e.g. for @xmath184 kev at @xmath185 ) so it is not relevant in our calculations . the photons from the decays of the sterile neutrino are mainly absorbed by neutral hydrogen atoms leading to their ionization . the ionization rate due to these photons is enhance almost 100 times due to additional ionization by the secondary electrons which deposit almost 1/3 of its energy into ionization ( depending on the ionization fraction of the medium ) . the energy of the absorbed photons partially goes into ionization and partially into heating and excitations . we have adopted the approximation by @xcite , for which ionization rate and heating due to the photons from the decays of the sterile neutrino are respectively equal to @xmath186 \nonumber \\ & \times & c_{i}\left(1-x_{e}^{a_{i}}\right)^b_{i } + \int_{\nu_{th}^{h}}^{\infty}4\pi\sigma_{h}\left(\nu\right)\frac{i_{\nu } \left(z\right)}{h\nu } d\nu \label{ion}\end{aligned}\ ] ] @xmath187 \nonumber \\ & \times & c_{h}\left [ 1-\left(1-x_{e}^{a_{h}}\right)^b_{h } \right ] n_{h } \label{heat } \textrm{,}\end{aligned}\ ] ] where @xmath188\ ] ] is the cross section for he ionization , @xmath189 ev is its energy threshold ; @xmath190 , @xmath191 , @xmath192 , @xmath193 , @xmath194 , @xmath195 @xcite . the h@xmath11 cooling function ( @xmath196 , @xmath197 and @xmath198 ) given in table [ tab : cooling ] has been computed assuming that the cmb temperature is much smaller than the gas temperature . this approximation is not valid in our case since in the wdm model with kev sterile neutrinos the gas temperature can be very close or even lower than the cmb temperature because of relatively high ionization fraction . the situation is very similar in cdm model , where the gas temperature strictly follows the cmb temperature at least at high redshifts . in such conditions , the level population of h@xmath11 is strongly affected by stimulated emission and absorption . molecules become an effective heating source for the gas , because the rate of collisional de - excitation of the vibrational levels is faster than their radiative decay . in order to correct the cooling / heating function for this effect , we have to take the net cooling rate @xmath199 where @xmath200 represents either @xmath196 , @xmath197 , or @xmath198 . we also note that @xmath201 for @xmath202 . one should do the same for the @xmath203 ( see table [ tab : cooling ] ) . however , the cooling from the collisional excitation of hydrogen atoms h and its subsequent spontaneous de - excitation is only important for the gas temperatures @xmath204 . in our case the cmb temperature is well below @xmath205 k , so we can neglect the heating term for this cooling / heating mechanism . abazajian , k. 2006a , , 73 , 063506 abazajian , k. 2006b , , 73 , 063513 abazajian , k. , fuller , g. m. , & patel , m. 2001 , , 64 , 023501 abazajian , k. , fuller , g. m. , & tucker , w. h. 2001 , , 562 , 593 abazajian , k. , koushiappas , s. m. , & zentner , a. r. 2006 , private communication abel , t. , anninos , p. , zhang , yu . , & norman , m. l. 1997 , , 2 , 181 akhmedov , e. k. , rubakov , v. a. , & smirnov , a. y. 1998 , , 81 , 1359 allgood , b. , flores , r. a. , primack , j. r. , kravtsov , a. v. , wechsler , r. h. , faltenbacher , a. , & bullock , j. s. 2006 , , 367 , 1781 appelquist , t. & shrock , r. 2002 , phys . b 548 , 204 asaka , t. , kusenko , a. , & shaposhnikov , m. 2006 , arxiv : hep - ph/0602150 asaka , t. , laine , m. , & shaposhnikov , m. 2006 , arxiv : hep - ph/0605209 asaka , t. , blanchet , s. , & shaposhnikov , m. 2005 , phys . b , 631 , 151 asaka , t. , & shaposhnikov , m. 2005 , phys . b , 620 , 17 athanassopoulos , c. , et al . ( lsnd ) 1998 , , 81 , 1774 barkovich , m. , dolivo , j. c. , & montemayor , r. 2004 , , 70 , 043005 barkovich , m. , dolivo , j. c. , & montemayor , r. 2005 , in _ new research _ , nova science publishers [ arxiv : hep - ph/0503113 ] belokurov , v. , et al . 2006 , , 642 , l137 biermann , p. l. , & kusenko , a. 2006 , , 96 , 091301 bode , p. , ostriker , j. p. , & turok , n. 2001 , , 556 , 93 van den bosch , f. c. , & swaters , r. a. 2001 , , 325 , 1017 boyarsky , a. , neronov , a. , ruchayskiy , o. , & shaposhnikov , m. 2005 , arxiv : astro - ph/0512509 boyarsky , a. , neronov , a. , ruchayskiy , o. , shaposhnikov , m. , & tkachev , i. 2006c , arxiv : astro - ph/0603660 bryan , g. l. , & norman , m. l. 1998 , , 495 , 80 dalcanton , j. j. , & hogan , c. j. 2001 , , 561 , 35 de gouva , a. , phys . rev . d 72 , 033005 ( 2005 ) . dodelson , s. , & widrow , l. m. 1994 , , 72 , 17 fryer , c. l. , & kusenko , a. 2006 , . , 163 , 335 fuller , g. m. , kusenko , a. , mocioiu , i. , & pascoli , s. 2003 , , 68 , 103002 galli , d. , & palla , f. 1998 , , 335 , 403 gell - mann , m. , ramond , p. , & slansky , r. 1979 , in _ supergravity _ , ed . by freedman , d. , et al . , north holland gelmini , g. , palomares - ruiz , s. , & pascoli , s. 2004 , , 93 , 081302 gilmore , g. , in proceedings of _ 7th ucla symposium on sources and detection of dark matter and dark energy in the universe _ , 22 - 24 feb 2006 , marina de rey , california , to appear . governato , f. , et al . 2004 , , 607 , 688 haiman , z. , thoul , a. a. , & loeb , a. 1996 , , 464 , 523 haiman , z. , & bryan , g. l. 2006 , arxiv : astro - ph/0603541 hansen , s. h. , lesgourgues , j. , pastor , s. , & silk , j. 2002 , , 333 , 544 kitayama , t. , susa , h. , umemura , m. , & ikeuchi , s. 2001 , , 326 , 1353 kormendy , j. , & fisher , d. b. 2005 , in revista mexicana de astronomia y astrofisica conference series , pp . 101 - 108 , arxiv : astro - ph/0507525 kusenko , a. , & segr , g. 1997 , phys . b396 , 197 kusenko , a. , & segr , g. 1999 , , 59 , 061302 kusenko , a. , segr , g. , & vilenkin , a. 1998 , phys . b437 , 359 kusenko , a. 2004 , ijmp d 13 , 2065 kusenko , a. 2006 , arxiv : hep - ph/0609081 jena , t. , et al . 2005 , , 361 , 70 sommer - larsen , j. , & dolgov , a. d. 2001 , , 551 , 608 sorel , m. , conrad , j. m. , & shaevitz , m. 2004 , , 70 , 073004 spergel , d.n . ( wmap ) 2003 , , 148 , 175 spergel , d.n . , et al . ( wmap ) 2006 , arxiv : astro - ph/0603449 strigari , l. e. , bullock , j. s. , kaplinghat , m. , kravtsov , a. v. , gnedin , o. y. , abazajian , k. , & klypin , a. a. 2006 , arxiv : astro - ph/0603775 viel , m. , lesgourgues , j. , haehnelt , m. g. , matarrese , s. , & riotto , a. 2005 , , 71 , 063534 viel , m. , lesgourgues , j. , haehnelt , m. g. , matarrese , s. , & riotto , a. 2006 , arxiv : astro - ph/0605706 watson , c. r. , beacom , j. f. , yuksel , h. & walker , t. p. 2006 , arxiv : astro - ph/0605424 wilkinson , m. i. , kleyna , j. t. , wyn evans , n. , gilmore , g. f. , read , j. i. , koch , a. , grebel , e. k. , & irwin , m. j. 2006 , eas publications series , 20 , 105 willman , b. , governato , f. , wadsley , j. , & quinn , t. 2004 , , 353 , 639 yanagida , t. 1980 , prog . phys . , 64 , 1103 lll + + @xmath209 & @xmath210 & @xcite + @xmath211 & @xmath212 & @xcite + @xmath213 & @xmath214 & @xcite + + + + @xmath215 & @xmath216 & + & @xmath217 & @xcite + h@xmath218 & @xmath219 & @xcite + h@xmath220 & @xmath221 & @xcite + + + + h@xmath222 & @xmath223 + & @xmath224 & @xcite + & @xmath225 $ ] & @xcite + & @xmath226 + & @xmath227 & @xcite + h@xmath228 & @xmath229 & @xcite + & @xmath230 + & @xmath231 + & @xmath232 + & @xmath233 + h@xmath234 & @xmath235 & @xcite + + + + h@xmath236 & @xmath237 & @xcite + & @xmath238 is the cmb radiation flux in erg @xmath134 @xmath135 + & at @xmath239 & + h@xmath240 & @xmath241 & @xcite + & @xmath242 ev , + & @xmath243 is the cmb intensity & + & @xmath244 + & @xmath245 and @xmath246 are the crossections for para- and ortho - h@xmath11 + & respectively , @xmath247 is the ortho- , to para - h@xmath11 ratio + + + + @xmath248 & @xmath249 & @xcite + & @xmath250 ev + & @xmath251 + @xmath252 & @xmath253 & @xcite + @xmath254 & @xmath255 & @xcite + + + + + @xmath248 & see text eq . ( [ ion ] ) & + [ tab : reaction ] ll + + + + a ) excitation of h by collisions with @xmath15 + @xmath256 & @xcite + + b ) recombination to h + @xmath257 & @xcite + + c ) ionization of h by collisions with @xmath15 + @xmath258 & @xcite + + d ) photo - ionization of h + @xmath259 + @xmath260 ev + + + + + + a ) collisional excitation of h@xmath11 + + + @xmath261 & @xcite + + + + @xmath262 & @xcite + + @xmath263 & @xcite + @xmath264 & @xcite + @xmath265 + 3\times10^{-24}\exp\left(-\frac{0.51}{t_{3}}\right)\right\}$ ] & @xcite + @xmath266 $ ] & @xcite + @xmath267 + 0.75\left[\frac{7}{3 } \gamma_{3}^{h_{2}}\exp\left[-\left(\frac{e^{r}_{31}}{kt}\right)\right]e^{r}_{31 } \right]$ ] & @xcite + @xmath268 & @xcite + @xmath269\right\}$ ] & @xcite + @xmath270 k , @xmath271 k & @xcite + + b ) h@xmath11 formation cooling @xmath272 + @xmath273n_{h}n_{e}$ ] & @xcite + @xmath274 + 1.4x_{h_{2}}\exp\left[-\frac{12000}{t+1200}\right]\right\}$ ] & @xcite + + c ) collisional dissociation of h@xmath11 + @xmath275 & @xcite + + d ) photo - dissociation of h@xmath11 + + + @xmath276 & @xcite + + + @xmath277 $ ] & @xcite + @xmath278 ev , @xmath279 ev , @xmath280 is the ortho - h@xmath11 to para - h@xmath11 ratio + + + @xmath281 & @xcite + + + + + + @xmath282 & @xcite + @xmath283 $ ] + + + + + + @xmath284see text eq . ( [ heat ] ) & + [ tab : cooling ]

[ [ stress - fluctuation - formalism . ] ] stress fluctuation formalism . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + among the fundamental properties of any solid material are its elastic constants @xcite . they describe the macroscopic response to weak external deformations and encode information about the potential landscape of the system . in their seminal work @xcite squire , holt and hoover derived an expression for the isothermal elastic constants extending the classical born theory @xcite to canonical ensembles of imposed particle number @xmath11 , volume @xmath12 and _ finite _ ( mean ) temperature @xmath5 . they found a correction term @xmath13 to the born expression for the elastic constants which involves the mean - square fluctuations of the stress @xcite @xmath14 with @xmath15 being the inverse temperature , @xmath16 boltzmann s constant , @xmath17 the instantaneous value of a component of the stress tensor @xmath18 and greek letters denoting the spatial coordinates in @xmath19 dimension . perhaps due to the fact that the demonstration of the stress fluctuation formalism " in ref . @xcite assumed explicitly a well - defined reference position and a displacement field for the particles , it has not been appreciated that this approach is consistent ( at least if the initial pressure is properly taken into account ) with the well - known pressure fluctuation formula for the compression modulus @xmath20 of isotropic fluids @xcite which has been derived several decades earlier , presumably for the first time in the late 1940s by rowlinson @xcite . in any case , the stress fluctuation formalism provides a convenient route to calculating the elastic properties in computer simulations which has been widely used in the past @xcite . it has been generalized to systems with nonzero initial stress @xcite , hard - sphere interactions @xcite and arbitrary continuous potentials @xcite , or to the calculation of local mechanical properties @xcite . [ [ low - temperature - limit - and - non - affine - displacements . ] ] low - temperature limit and non - affine displacements . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in particular from ref . @xcite it has become clear that the stress fluctuation correction to the leading born term does not necessarily vanish in the zero - temperature limit . this is due to the fact that when a system is subjected to a homogeneous deformation , the ensuing particle displacements need not follow the macroscopic strain affinely @xcite . the stress fluctuation term quantifies the extent of these non - affine displacements , whereas the born term reflects the affine part of the particle displacements . how important the non - affine motions are , depends on the system under consideration @xcite . while the elastic properties of crystals with one atom per unit cell are given by the born term only , stress fluctuations are significant for crystals with more complex unit cells @xcite . they become particularly pronounced for polymer like soft materials @xcite and amorphous solids @xcite . this is revealed by recent simulation studies of various glass formers @xcite , like lennard - jones ( lj ) mixtures @xcite , polymer glasses @xcite or silica melts @xcite . many of these simulation studies concentrate on the mechanical properties deep in the glassy state , exploring for instance correlations between the non - affine displacement field and vibrational anomalies of the glass ( boson peak " ) @xcite , the mechanical heterogeneity at the nanoscale @xcite or the onset of molecular plasticity in the regime where the macroscopic deformation is still elastic @xcite . [ [ shear - stress - and - shear - modulus . ] ] shear stress and shear modulus . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + surprisingly , the temperature dependence of the elastic constants of glassy materials does not appear to have been investigated extensively using the stress fluctuation formalism @xcite . following the pioneering work by barrat _ @xcite , we present in this work such a study where we focus on the behavior of the shear modulus @xmath0 of two isotropic colloidal glass - formers in @xmath1 and @xmath2 dimensions sampled , respectively , by means of molecular dynamics ( md ) and monte carlo ( mc ) simulations @xcite . since we are interested in isotropic systems one can avoid the inconvenient tensorial notation of the full stress fluctuation formalism as shown in fig . [ fig_sketch ] . focusing on the shear in the @xmath21-plane we use @xmath22 for the mean shear stress and @xmath23 for its instantaneous value . as an important contribution to the shear modulus we shall monitor the shear stress fluctuations @xmath24 as a function of temperature @xmath5 . quite generally , the shear modulus @xmath0 may then be computed according to the stress fluctuation formula @xcite @xmath25 where @xmath26 stands for the so - called affine shear elasticity " corresponding to an assumed _ affine _ response to an external shear strain @xmath3 . for pairwise interaction potentials it can be shown that @xmath27 i.e. @xmath26 comprises the so - called born - lam coefficient " @xmath28 , a simple moment of the first and the second derivative of the pair potential with respect to the distance of the interacting particles @xcite , and the excess contribution @xmath29 to the total pressure @xmath6 . these notations will be properly defined in sec . [ theo_stressfluctu ] below where a simple derivation will be given . since eq . ( [ eq_gmodtt ] ) is not the only operational definition of the shear modulus discussed in this work , an index has been added reminding that the shear stress @xmath30 is used for the determination of @xmath0 . [ [ physical - motivation - and - systems - of - interest . ] ] physical motivation and systems of interest . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + if an infinitesimal shear stress @xmath4 is applied at time @xmath31 , an isotropic solid exhibits , quite generally , an elastic response quantified in strength by a finite static ( @xmath32-independent ) shear modulus ( @xmath33 ) as shown by the top curve in fig . [ fig_sketch](b ) , whereas a liquid has a vanishing modulus ( @xmath34 ) and flows on long time scales as shown by the bottom curve . the shear modulus @xmath0 can thus serve as an order parameter " distinguishing liquid and solid states @xcite . upon melting , the shear modulus of crystalline solids is known to vanish discontinuously with @xmath5 . a natural question which arises is that of the behavior of @xmath35 for amorphous solids and glasses in the vicinity of the glass transition temperature @xmath9 . interestingly , qualitative different theoretical predictions have been put forward by mode - coupling theory ( mct ) @xcite and some versions of replica theory @xcite predicting a _ discontinuous jump _ at the glass transition whereas other versions of replica theory @xcite suggest a _ continuous _ transition in the static limit of large sampling times @xmath32 . ( see ref . @xcite for a topical recent discussion of the replica approach . ) the numerical characterization of @xmath35 for colloidal glass - former is thus of high current interest . since the stress fluctuation formalism can readily be added to the standard simulation codes of colloidal glass - formers @xcite at imposed particle number @xmath11 , box volume @xmath12 , box shape ( @xmath36 ) and temperature @xmath5 , called @xmath37-ensembles , this suggests the use of eq . ( [ eq_gmodtt ] ) to tackle this issue . note that the latter stress fluctuation relation should also be useful for wide range of related systems , beyond the scope of the present work , including the glass transition of polymeric liquids @xcite , colloidal gels @xcite and self - assembled networks @xcite as observed experimentally for hyperbranched polymer chains with sticky end - groups @xcite or bridged networks of telechelic polymers in water - oil emulsions @xcite . [ [ validity - through - a - solid - liquid - transition . ] ] validity through a solid - liquid transition . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + this begs the delicate question of whether the general stress fluctuation formalism and especially eq . ( [ eq_gmodtt ] ) for the shear modulus only holds for elastic solids with a well - defined displacement field , as suggested in the early literature @xcite , or if the approach may actually be also applied _ through _ a solid - liquid transition up to the liquid state @xcite with @xmath38 and thus @xmath39 as it should for a liquid . by reworking the theoretical arguments leading to eq . ( [ eq_gmodtt ] ) it will be seen that this is in principal the case , at least if standard thermodynamics can be assumed to apply . the latter point is indeed not self - evident for the colloidal glasses we are interested in where _ ( i ) _ some degrees of freedom are frozen on the time scale probed and _ ( ii ) _ the measured response to an imposed infinitesimal shear @xmath4 may depend on the sampling time @xmath32 , especially for temperatures @xmath5 around @xmath9 as shown in fig . [ fig_sketch](b ) . to be on the safe side , it thus appears to be appropriate to test numerically the applicability of the stress fluctuation formula , eq . ( [ eq_gmodtt ] ) , using the corresponding strain fluctuation relations which are more fundamental on experimental grounds . [ [ strain - fluctuation - relations . ] ] strain fluctuation relations . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + panel ( a ) of fig . [ fig_sketch ] shows a schematic setup for probing experimentally the shear modulus @xmath0 of isotropic systems confined between two rigid walls . we assume that the bottom wall is fixed and the particle number @xmath11 , the ( mean ) normal pressure @xmath6 and the ( mean ) temperature @xmath5 are kept constant . in linear response either a shear strain @xmath40 or a ( mean ) shear stress @xmath4 may be imposed . the thin solid line indicates the original unstrained reference system with @xmath36 and @xmath41 , the dashed line the deformed state . from both the mechanical and the thermodynamical point of view , the shear modulus is defined as @xmath42 where the derivative may also be taken at @xmath41 . in practice , one obtains @xmath0 by fitting a linear slope to the @xmath43-data at vanishing strain . obviously , the noise level is normally large for small strains . from the numerical point of view it would thus be more appropriate to work at imposed mean stress @xmath41 , to let the strain freely fluctuate and to sample the _ instantaneous _ values @xmath44 . the shear modulus @xmath0 is then simply obtained by linear regression @xmath45 where the index indicates that both the instantaneous values @xmath46 and @xmath30 are used . ( note that imposing the strain @xmath3 also fixes the instantaneous strain @xmath46 which thus can not fluctuate . ) the associated dimensionless correlation coefficient @xmath47 allows to determine the quality of the fit . we do not know whether eq . ( [ eq_gmodgt ] ) and eq . ( [ eq_cmodgt ] ) have actually been used in a real experiment , however since no difficulty arises to work in a computer simulation at constant @xmath4 and to record @xmath44 , we use this linear regression as our key operational definition of @xmath0 which is used to judge the correctness of the stress fluctuation relation , eq . ( [ eq_gmodtt ] ) , computed at constant @xmath3 . to avoid additional physics at the walls , we use of course periodic boundary conditions @xcite ( the central image being sketched ) as shown by the panels on the right - hand side of fig . [ fig_sketch ] . as discussed in sec . [ theo_legendre ] , it is possible to determine @xmath48 at constant volume ( @xmath49-ensemble ) as shown by panel ( c ) or constant normal pressure ( @xmath50-ensemble ) as shown by panel ( e ) . the latter ensemble has the advantage that one can easily impose in addition the same normal pressure @xmath6 for all temperatures @xmath5 as it is justified for experimental reasons . if thermodynamics holds , it is known ( sec . [ theo_strainfluctu ] ) that @xmath51 should hold @xcite . using eq . ( [ eq_gmodgt ] ) this implies a second operational definition @xmath52 which has the advantage that only the instantaneous shear strain @xmath46 has to be recorded . the main technical point put forward in this paper is that all operational definitions yield similar results @xmath53 even for temperatures where a strong dependence on the sampling time @xmath32 is observed . in other words , eq . ( [ eq_gmods ] ) states that this time dependence applies similarly ( albeit perhaps not exactly ) to the various moments computed for the three operational definitions of @xmath0 in different ensembles . [ [ outline . ] ] outline . + + + + + + + + the paper is organized as follows : several theoretical issues are regrouped in sec . [ sec_theo ] . we begin the discussion in sec . [ theo_strainfluctu ] by reminding a few useful general thermodynamic relations . a simple derivation of the stress fluctuation formula , eq . ( [ eq_gmodtt ] ) , for the shear modulus @xmath54 in constant-@xmath3 ensembles is given in sec . [ theo_stressfluctu ] . ( the analogous stress fluctuation formula for the compression modulus @xmath20 in constant-@xmath12 ensembles @xcite is rederived in appendix [ app_kstressfluctu ] . ) shear stress fluctuations @xmath55 in ensembles with different boundary conditions are discussed in sec . [ theo_legendre ] . for constant-@xmath4 ensembles @xmath56 will be shown to reduce to the affine shear elasticity @xmath26 , i.e. the functional @xmath54 must vanish for all temperatures . as pointed out recently @xcite , the truncation of the interaction potentials used in computational studies implies an impulsive correction for the affine shear elasticity @xmath26 . this is summarized in sec . [ theo_trunc ] . the numerical models used in this study are presented in sec . [ sec_algo ] . we turn then in sec . [ sec_res ] to our main numerical results . starting with the high-@xmath5 liquid limit , sec . [ res_thigh ] shows that all our operational definitions correctly yield a vanishing shear modulus . moving to the opposite low-@xmath5 limit we present in sec . [ res_dynmat ] the elastic response for a topologically fixed network of harmonic springs constructed from the dynamical matrix " of our 2d model glass system @xcite . as shown in sec . [ res_tlow ] , the same behavior is found qualitatively for low-@xmath5 glassy bead systems . the temperature dependence of stress fluctuations and shear moduli in different ensembles is then discussed in sec . [ res_tvari ] . as stated above , one central goal of this work is to determine the behavior of the shear modulus around @xmath9 . we conclude the paper in sec . [ sec_conc ] . thermodynamic relations and numerical findings related to the compression modulus @xmath20 are regrouped in appendix [ app_k ] . we assume in the following that standard thermodynamics and thermostatistics @xcite can be applied and remind a few useful relations . let us denote an extensive variable by @xmath57 and its instantaneous value by @xmath58 , the conjugated intensive variable by @xmath59 and its instantaneous value by @xmath60 @xcite . with @xmath61 being the free energy at temperature @xmath5 and imposed @xmath57 , the intensive variable @xmath59 and the associated modulus @xmath62 are given by @xmath63 alternatively , one may consider the legendre transformation @xmath64 for which @xmath65 in our case , the extensive variable is @xmath66 , the intensive variable @xmath67 @xcite and the modulus @xmath68 @xcite . we emphasize that there is an important difference between extensive and intensive variables : if the extensive variable @xmath57 is fixed , @xmath58 can not fluctuate . if , on the other hand , the intensive variable @xmath69 is imposed , not only @xmath58 but also @xmath60 may fluctuate @xcite . in our case this means , that @xmath70 for the @xmath37-ensemble , while the shear stress fluctuations @xmath55 defined above do clearly not vanish in the @xmath49-ensemble shown in fig . [ fig_sketch](d ) . assuming @xmath59 to be imposed , it is well known that @xcite @xmath71 with @xmath15 being the inverse temperature . for our variables @xmath72 and @xmath73 , eq . ( [ eq_dxdi ] ) implies eq . ( [ eq_dgamdtau ] ) from which it is seen that eq . ( [ eq_gmodgt ] ) is consistent with eq . ( [ eq_gmodgg ] ) . the latter relation is also obtained directly using eq . ( [ eq_dxdx ] ) . as discussed in textbooks @xcite a simple average @xmath74 of some observable @xmath75 does not depend of the chosen ensemble , at least not if the system is large enough ( @xmath76 ) . a correlation function @xmath77 of two observables @xmath75 and @xmath78 may differ , however , depending on whether @xmath57 or @xmath59 are imposed . as demonstrated by lebowitz , percus and verlet in 1967 @xcite one verifies that @xmath79 if the extensive variable and at least one of the observables are identical , the left - hand side of eq . ( [ eq_dadb ] ) vanishes : if @xmath80 and @xmath81 , this yields eq . ( [ eq_dxdi ] ) . if @xmath82 , this implies eq . ( [ eq_dxdx ] ) . more importantly , for @xmath83 one obtains using eq . ( [ eq_ftom ] ) @xmath84 if it is possible to probe the intensive variable fluctuations in both ensembles , this does in principle allow to determine the modulus @xmath62 . ( the latter equation might actually be taken as the definition of @xmath62 in cases where the thermodynamic reasoning becomes dubious . ) we emphasize that the modulus @xmath62 is an intrinsic property of the system and does _ not _ depend on the ensemble , albeit it is determined using ensemble depending correlation functions . for a thermodynamically stable system , @xmath85 , eq . ( [ eq_didi ] ) implies that @xmath86 and that both fluctuations must become similar in the limit where the modulus @xmath62 becomes small . for the shear stress fluctuations @xmath55 , eq . ( [ eq_didi ] ) corresponds to the transformation @xmath87 and one thus expects @xmath55 to be larger for @xmath49-systems than for @xmath37-systems , while in the liquid limit where @xmath88 the imposed boundary conditions should not matter . we shall verify eq . ( [ eq_muftransform ] ) numerically below . a glance at the stress fluctuation formula , eq . ( [ eq_gmodtt ] ) , suggests the identity @xmath89 assuming eq . ( [ eq_dgamdtau ] ) and eq . ( [ eq_muamuftau ] ) , it follows for the correlation coefficient @xmath90 defined above that @xmath91 if the measured @xmath44 were indeed perfectly correlated , i.e. @xmath92 , this implies @xmath93 and thus @xmath94 . in fact , for all situations studied here we always have @xmath95 i.e. the affine shear elasticity sets only an upper bound to the shear modulus and a theory which only contains the affine strain response must _ overpredict _ @xmath0 . we show now directly that eq . ( [ eq_gmodtt ] ) and , hence , eqs . ( [ eq_muamuftau],[eq_cmodgt_mua],[eq_gbound ] ) indeed hold and this under quite general conditions . [ [ introduction . ] ] introduction . + + + + + + + + + + + + + as shown in fig . [ fig_sketch](a ) , we demonstrate eq . ( [ eq_gmodtt ] ) from the free energy change associated with an imposed arbitrarily small pure shear strain @xmath3 in the @xmath21-plane assuming a constant particle number @xmath11 , a constant volume @xmath12 and a constant temperature @xmath5 ( @xmath37-ensemble ) . it is supposed that the total system hamiltonian may be written as the sum of a kinetic energy and a potential energy . since for a plain shear strain at constant volume the ideal free energy contribution does not change , i.e. is irrelevant for @xmath4 and @xmath0 , we may focus on the excess free energy contribution @xmath96 due to the conservative interaction energy of the particles . the ( excess ) partition function @xmath97 of the unperturbed system at @xmath36 is the boltzmann - weighted sum over all states @xmath98 of the system which are _ accessible _ within the measurement time @xmath32 . the argument @xmath99 is a short - hand for the unstrained reference . following the derivation of the compression modulus @xmath20 by rowlinson @xcite the partition function @xmath100 of the sheared system is supposed to be the sum over the _ same _ states @xmath98 , but with a different _ metric _ @xcite corresponding to the macroscopic strain which changes the total interaction energy @xmath101 of state @xmath98 and , hence , the weight of the sheared configuration for the averages computed . this is the central hypothesis made . interestingly , it is not necessary to specify explicitly the states of the unperturbed or perturbed system , e.g. , it is irrelevant whether the particles are distinguishable or not or whether they have a well - defined reference position for defining a displacement field . [ [ shear - stress - and - modulus - for - general - potential . ] ] shear stress and modulus for general potential . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + assuming eq . ( [ eq_fexgamma ] ) to hold we compute now the mean shear stress @xmath4 and the shear modulus @xmath0 for a general interaction potential @xmath101 . we note for later convenience that @xmath102 for the derivatives of the free energy and @xmath103 for the derivatives of the excess partition function where a prime denotes the derivative of a function @xmath104 with respect to its argument @xmath105 . using eq . ( [ eq_dzdgam ] ) and taking finally the limit @xmath106 one verifies for the shear stress that @xmath107 _ defining _ the instantaneous shear stress @xcite . note that the average taken is defined as @xmath108 using the weights of the unperturbed system . the shear stress thus measures the average change of the total interaction energy @xmath101 taken at @xmath36 . the shear modulus @xmath0 is obtained using in addition eq . ( [ eq_ddlogzddgam ] ) and eq . ( [ eq_ddzddgam ] ) and taking finally the @xmath106 limit . confirming thus the stress fluctuation formula eq . ( [ eq_gmodtt ] ) stated in the introduction , this yields @xmath109 with @xmath26 being the affine shear elasticity " @xmath110 already mentioned above and @xmath55 as defined in eq . ( [ eq_muf ] ) . the comparison of eq . ( [ eq_gstressfluctu ] ) and eq . ( [ eq_muftransform ] ) confirms eq . ( [ eq_muamuftau ] ) . [ [ comment . ] ] comment . + + + + + + + + the affine shear elasticity @xmath26 corresponds to the change ( second derivative ) of the total energy which would be obtained if one actually strains _ affinely _ in a computer simulation a given state @xmath98 without allowing the particles to relax their position . as shown for athermal ( @xmath111 ) amorphous bodies @xcite , the positions of the particles of such a strained configuration will of course in general change slightly to minimize the interaction energy relaxing thus the elastic moduli . this is also of relevance for thermalized solids where the non - affine displacements of the particles are driven by the minimization of the free energy . it is for this reason that the shear - stress fluctuation term @xmath112 must occur in eq . ( [ eq_gstressfluctu ] ) correcting the _ overprediction _ of the shear modulus @xmath0 by the affine shear elasticity @xmath26 , eq . ( [ eq_gbound ] ) . this point has been overlooked in the early literature @xcite and only appreciated much later @xcite as discussed in barrat s review @xcite . interestingly , as has been shown by lutsko @xcite , @xmath55 and other similarly defined stress fluctuations become temperature independent in the harmonic ground state approximation for @xmath111 . probing the stress fluctuations in a low - temperature simulation allows thus to determine the elastic moduli of athermal solids . please note that up to now we have not used explicitly the coordinate transformations ( metric change ) associated with an affine shear strain . this is needed to obtain operational definitions for the instantaneous shear stress @xmath30 ( and thus for @xmath4 and @xmath55 ) and the affine shear elasticity . [ [ coordinate - transformation . ] ] coordinate transformation . + + + + + + + + + + + + + + + + + + + + + + + + + + as shown in fig . [ fig_sketch](a ) , we assume a pure shear strain which transforms the @xmath105-coordinate of a particle position or the distance between two particles as @xmath113 leaving all other coordinates unchanged @xcite . the squared distance @xmath114 between two particles thus transforms as @xmath115 where we need to keep the @xmath116-term for calculating the shear modulus below . we note for later reference that this implies @xmath117 for , respectively , the first and the second derivative with respect to @xmath3 . let us now consider an arbitrary function @xmath118 . using eq . ( [ eq_strain_rone ] ) it is readily seen that its first derivative with respect to @xmath3 may be written as @xmath119 in the second step we have introduced the components @xmath120 and @xmath121 of the normalized distance vector between both particles . more generally , we denote by @xmath122 the @xmath123-component of a normalized vector with @xmath124 stating only the small-@xmath3 limit for the second derivative of @xmath118 with respect to @xmath3 we note finally that @xmath125 where we have dropped the argument @xmath99 for the distance @xmath126 of the unperturbed reference system . [ [ pair - interaction - potentials . ] ] pair interaction potentials . + + + + + + + + + + + + + + + + + + + + + + + + + + + + we assume now that the interaction energy @xmath101 of a configuration @xmath98 is given by a pair interaction potential @xmath127 between the particles @xcite with @xmath128 being the distance between two particles , i.e. @xmath129 where the index @xmath130 labels the interaction between the particles @xmath131 and @xmath132 with @xmath133 . using the general result eq . ( [ eq_tauhat_us ] ) for the shear stress it follows using eq . ( [ eq_strain_fronea ] ) that @xmath134 where we have dropped the argument @xmath99 in the @xmath106 limit . we have thus rederived for the shear stress ( @xmath135 ) the well - known kirkwood expression for the general excess stress tensor @xcite @xmath136 the affine shear elasticity @xmath26 is obtained using eq . ( [ eq_muadef ] ) and eq . ( [ eq_strain_frtwo ] ) . this yields @xmath137 with the first contribution @xmath28 being the so - called born - lam coefficient " @xcite @xmath138 it corresponds to the first term in eq . ( [ eq_strain_frtwo ] ) . the second contribution @xmath139 stands for the normal excess stress in the @xmath140-direction , i.e. the @xmath141 component of the excess stress tensor . [ [ isotropic - systems . ] ] isotropic systems . + + + + + + + + + + + + + + + + + + in this work we focus on isotropic materials under isotropic external loads . the normal ( excess ) stresses @xmath142 for all directions @xmath123 are thus identical , i.e. @xmath143 with @xmath144 being the excess pressure , @xmath6 the total pressure , @xmath145 the ideal gas pressure and @xmath19 the spatial dimension . this implies that the affine shear elasticity is given by @xmath146 as already stated in the introduction , eq . ( [ eq_muamubpex ] ) . we note that for isotropic @xmath19-dimensional systems it can be shown @xcite that the born - lam coefficient can be simplified as @xmath147 the @xmath19-dependent prefactor stems from the assumed isotropy of the system and the mathematical formula @xcite @xmath148 ( @xmath149 being the kronecker symbol @xcite ) for the components of a unit vector in @xmath19 dimensions pointing into arbitrary directions . by comparing the excess pressure @xmath29 , eq . ( [ eq_sigaapex ] ) , with the underlined second contribution to the born coefficient in eq . ( [ eq_mub ] ) this implies that @xmath150 it is thus inconsistent to neglect the explicit excess pressure in eq . ( [ eq_muamubpex ] ) while keeping the second term of the born - lam coefficient @xmath28 . this approximation is only justified if the excess pressure @xmath29 is negligible ( and not the total pressure @xmath6 ) . [ [ different - ensembles . ] ] different ensembles . + + + + + + + + + + + + + + + + + + + + being simple averages neither @xmath29 , @xmath26 or @xmath28 does depend for sufficiently large systems on the chosen ensemble , i.e. irrespective of whether @xmath3 or @xmath4 is imposed one expects to obtain the same values . as already reminded in sec . [ theo_strainfluctu ] , this is different for fluctuations in general @xcite . for instance , it is obviously pointless to use the shear - strain fluctuation formulae @xmath48 or @xmath151 in a constant-@xmath3 ensemble . a more interesting result is predicted if @xmath152 is computed in the wrong " constant-@xmath4 ensemble . since @xmath153 , eq . ( [ eq_muftransform ] ) , one actually expects to observe @xmath154 for all temperatures @xmath5 if our thermodynamic reasoning holds through the glass transition up to the liquid state . we will test numerically this non - trivial prediction in sec . [ sec_res ] . up to now we have only considered for simplicity of the presentation @xmath37- and @xmath49-ensembles at constant volume @xmath12 . since on general grounds pure deviatoric and dilatational ( volumetric ) strains are decoupled ( both strains commute ) , one expects the observable @xmath54 also to be applicable in the @xmath155-ensemble shown in fig . [ fig_sketch](d ) and the observables @xmath48 and @xmath151 to be applicable in the @xmath50-ensemble as sketched in fig . [ fig_sketch](e ) . in the latter ensemble one also expects to find @xmath156 and @xmath153 . this will also be tested below . [ [ direct - derivation - of - ensuremathmu_mathrmf - for - imposed - tau . ] ] direct derivation of @xmath55 for imposed @xmath4 . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we demonstrate now directly eq . ( [ eq_muamuftau ] ) for the shear stress fluctuations using the general mathematical identity @xmath157 with @xmath105 being an _ unconstrained _ coordinate , @xmath158 some property , @xmath159 a force " with respect to some energy " @xmath160 and the average being boltzmann weighted , i.e. @xmath161 . ( it is assumed in eq . ( [ eq_intbyparts ] ) that @xmath160 decays sufficiently fast at the boundaries . ) following work by zwanzig @xcite we express first the instantaneous shear stress @xmath30 given above by the kirkwood formula , eq . ( [ eq_tauhatkirkwood ] ) , by the alternative virial representation @xcite @xmath162 the second term , sometimes called the inner virial " , stands for a sum over all particles @xmath131 with @xmath163 being the @xmath140-coordinate of the particle and @xmath164 the @xmath105-coordinate of the force acting on the particle . this contribution indeed vanishes on average as can be seen using eq . ( [ eq_intbyparts ] ) , @xmath165 which shows that @xmath166 as it must . the stress fluctuations may thus be expressed as @xmath167 where we have used the integration by part , eq . ( [ eq_tauhatfluctu ] ) . this step requires that all particle positions are unconstrained and independent ( generalized ) coordinates . note that this is possible at imposed @xmath4 , but can not hold at fixed @xmath3 . assuming then pair interactions and using eq . ( [ eq_tauhatkirkwood ] ) one can reformulate eq . ( [ eq_stressfluctuvirial ] ) within a few lines . without further approximation this yields @xmath168 where the sum runs now over all interactions @xmath130 between particles @xmath133 . @xmath169 and @xmath170 refer to components of the distance vector of the interaction @xmath130 and @xmath171 to the @xmath105-component of the central force @xmath172 between a particle pair at a distance @xmath173 . note that in eq . ( [ eq_stressfluctdecoupled ] ) the stress fluctuations stemming from different interactions are decoupled , i.e. @xmath174 characterizes the self- or two - point contributions of directly interacting beads . since @xmath175 and , hence , @xmath176 this is identical to the affine shear elasticity @xmath26 derived at the end of sec . [ theo_stressfluctu ] . we have thus confirmed eq . ( [ eq_muamuftau ] ) . interestingly , the argument can be turned around : 1 . the stress fluctuation contribution term @xmath177 is obtained by the simple derivation given in this paragraph . 2 . using the general legendre transformation , eq . ( [ eq_muftransform ] ) , this implies @xmath178 for the @xmath37- and the @xmath155-ensemble . as already stressed , our thermodynamic derivation of eq . ( [ eq_gmodtt ] ) is rather general , not using in particular a well - defined displacement field for the particle positions . [ [ truncation . ] ] truncation . + + + + + + + + + + + it is common practice in computational condensed matter physics @xcite to truncate a pair interaction potential @xmath127 , with @xmath126 being the distance between two particles @xmath131 and @xmath132 , at a conveniently chosen cutoff @xmath179 . this allows to reduce the number of interactions to be computed and energy or force calculations become @xmath180-processes . however , the truncation also introduces technical difficulties , e.g. , instabilities in the numerical solution of differential equations as investigated especially for the md method @xcite . without restricting much in practice the generality of our results , we assume below that * the pair potential @xmath127 is short - ranged , i.e. that it decays within a few particle diameters , * it scales as @xmath181 with the reduced dimensionless distance @xmath182 where @xmath183 characterizes the length scale of the interaction @xmath130 and * the same reduced cutoff @xmath184 is set for all interactions @xmath130 . for instance , for monodisperse particles with constant diameter @xmath185 , as for the standard lennard - jones ( lj ) potential @xcite , @xmath186 the scaling variable becomes @xmath187 and the reduced cutoff @xmath188 . the effect of introducing @xmath189 is to replace @xmath190 by the truncated potential @xmath191 with @xmath192 being the heaviside function @xcite . even if eq . ( [ eq_utrunc ] ) is taken _ by definition _ as the new system hamiltonian , it is well known that impulsive corrections at the cutoff have to be taken into account in general for the excess pressure @xmath29 and other moments of the first derivatives of the potential @xcite . this is seen by considering the derivative of the truncated potential @xmath193 [ [ shifting - of - truncated - potential . ] ] shifting of truncated potential . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + these corrections with respect to first derivatives can be avoided by considering a properly shifted potential @xcite @xmath194 since @xmath195 . with this choice _ no _ impulsive corrections arise for moments of the instantaneous shear stress @xmath30 and of other components of the excess stress tensor @xmath196 , eq . ( [ eq_sigab ] ) . specifically , if the potential is shifted , all impulsive corrections are avoided for the shear stress fluctuations @xmath55 , eq . ( [ eq_muf ] ) . [ [ correction - to - the - born - coefficient . ] ] correction to the born coefficient . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + as shown in ref . @xcite , the standard shifting of a truncated potential is , however , insufficient in general for properties involving second ( and higher ) derivatives of the potential . this is particulary the case for the born coefficient @xmath28 , eq . ( [ eq_mub ] ) , which is required to compute the affine shear elasticity @xmath26 , eq . ( [ eq_muamubpex ] ) . the second derivative of the truncated and shifted potential being @xmath197 this implies that the born coefficient reads @xmath198 with @xmath199 being the bare born coefficient and @xmath200 the impulsive correction which needs to be taken into account . the latter correction may be readily computed numerically from the configuration ensemble using @xcite @xmath201 being a weighted radial pair distribution function which is related to the standard radial pair distribution function @xmath202 @xcite . [ [ mixtures - and - polydisperse - systems . ] ] mixtures and polydisperse systems . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + below we shall consider model systems for mixtures and polydisperse systems where @xmath183 may differ for each interaction @xmath130 . for such mixed potentials @xmath190 , @xmath203 and @xmath204 and their derivatives take in principal an explicit index @xmath130 , i.e. one should write @xmath205 , @xmath206 , @xmath207 and so on . ( this is often not stated explicitly to keep a concise notation . ) for example one might wish to consider the generalization of the monodisperse lj potential , eq . ( [ eq_lj ] ) , to a mixture or polydisperse system with @xmath208 where @xmath209 and @xmath183 are fixed for each interaction @xmath130 . in practice , each particle @xmath131 may be characterized by an energy scale @xmath210 and a diameter " @xmath211 . the interaction parameters @xmath212 and @xmath213 are then given in terms of specified functions of these properties @xcite . the extensively studied kob - andersen ( ka ) model for binary mixtures of beads of type a and b @xcite , is a particular case of eq . ( [ eq_ljgeneralized ] ) with fixed interaction ranges @xmath214 , @xmath215 and @xmath216 and energy parameters @xmath217 , @xmath218 and @xmath219 characterizing , respectively , aa- , bb- and ab - contacts . the expression eq . ( [ eq_histolame ] ) remains valid for such explicitly @xmath130-dependent potentials . [ [ shear - modulus - and - compression - modulus . ] ] shear modulus and compression modulus . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + since @xmath220 , eq . ( [ eq_histolame ] ) implies in turn @xmath221 with @xmath222 being the uncorrected bare shear modulus . as we shall see in sec . [ res_thigh ] , the correction @xmath200 is of importance for the _ precise _ determination of @xmath54 close to the glass transition where the shear modulus must vanish . an impulsive correction has also to be taken into account for the compression modulus @xmath223 computed from the normal pressure fluctuations in a constant volume ensemble as discussed in appendix [ app_k ] . using the symmetry of isotropic systems , eq . ( [ eq_etabmub ] ) , one verifies @xcite @xmath224 with @xmath225 being the uncorrected compression modulus . [ [ introduction.-1 ] ] introduction . + + + + + + + + + + + + + the computational data presented in this work have been obtained using two extremely well studied models of colloidal glass - formers which are described in detail elsewhere @xcite . we present first both models and the molecular dynamics ( md ) and monte carlo ( mc ) @xcite methods used to compute them in the different ensembles ( @xmath37 , @xmath49 , @xmath155 and @xmath50 ) compared in this study . we comment then on the quench protocol and locate the glass transition temperature @xmath9 for both models via dilatometry , as shown in fig . [ fig_quench_rho ] . this is needed as a prerequisite to our study of the elastic behavior in sec . [ sec_res ] . as sketched in fig . [ fig_sketch ] , periodic boundary conditions are applied in all spatial directions with @xmath226 for the linear dimensions of the @xmath19-dimensional simulation box . time scales are measured in units of the lj time @xmath227 @xcite for our md simulations ( @xmath228 being the particle mass , @xmath185 the reference length scale and @xmath229 the lj energy scale ) and in units of monte carlo steps ( mcs ) for our mc simulations . lj units are used throughout this work ( @xmath230 ) and boltzmann s constant @xmath16 is also set to unity . [ [ kob - andersen - model . ] ] kob - andersen model . + + + + + + + + + + + + + + + + + + + the kob - andersen ( ka ) model @xcite for binary mixtures of lj beads in @xmath1 dimensions has been investigated by means of md simulation @xcite taking advantage of the public domain lammps implementation @xcite . we use @xmath231 beads per simulation box and molar fractions @xmath232 and @xmath233 for both types of beads @xmath234 and @xmath235 . following ref . @xcite we set @xmath236 , @xmath237 and @xmath238 for the interaction range and @xmath239 , @xmath240 and @xmath241 for the lj energy scales . only data for the usual ( reduced ) cutoff @xmath242 are presented . for the @xmath155-ensemble we use the nos - hoover barostat provided by the lammps code ( fix npt command " ) @xcite which is used to impose a constant pressure @xmath243 for all temperatures . starting with an equilibrated configuration at @xmath244 well above the glass transition temperature @xmath245 ( as confirmed in the inset of fig . [ fig_quench_rho ] ) , the system is slowly quenched in the @xmath155-ensemble . the imposed mean temperature @xmath246 varies linearly as @xmath247 with @xmath248 being the constant quench rate . after a tempering time @xmath249 we compute over a sampling time @xmath250 various properties for the isobaric system at fixed temperature such as the moduli @xmath54 represented by the filled spheres in fig . [ fig_gt ] . fixing only then the volume ( @xmath37-ensemble ) the system is again tempered ( @xmath249 ) and various properties as the ones recorded in table [ tab_ka ] are computed using a sampling time @xmath250 . note that in the @xmath37-ensemble the equations of motion are integrated using a velocity verlet algorithm @xcite with a time step @xmath251 and systems are kept at the imposed temperature @xmath5 using a langevin thermostat @xcite with friction coefficient @xmath252 . by redoing for several temperatures around and below @xmath9 the tempering and the production runs we have checked that the presented results remain unchanged ( within numerical accuracy ) and that ageing effects are absolutely negligible . averages are taken over four independent configurations . as one may expect from the discussion in sec . [ theo_legendre ] , very similar results have been found for both ensembles for simple averages , for instance for the affine shear elasticity @xmath26 , and for the shear modulus @xmath54 , eq . ( [ eq_gmodtt ] ) . unfortunately , we are yet unable to present data for the ka model data at imposed shear stress @xmath4 ( @xmath49- , @xmath50-ensembles ) . .some properties for the ka model at normal pressure @xmath243 and shear stress @xmath41 : the mean density @xmath253 , the mean energy per volume @xmath254 , the impulsive correction term @xmath200 ( sec . [ theo_trunc ] ) , the ( corrected ) born - lam coefficient @xmath28 , the affine shear elasticity @xmath255 , the shear modulus @xmath54 obtained in the @xmath37-ensemble , the hypervirial @xmath256 and the affine dilatation elasticity @xmath257 discussed in appendix [ app_k ] and the compression modulus @xmath258 obtained for the @xmath37-ensemble . the affine elasticities @xmath26 and @xmath259 are the natural scales for , respectively , the shear modulus @xmath0 and the compression modulus @xmath20 . [ tab_ka ] [ cols="^,^,^,^,^,^,^,^,^,^,^,^",options="header " , ] [ [ local - mc - moves - for - the - plj - model . ] ] local mc moves for the plj model . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + for all ensembles considered here local mc moves are used ( albeit not exclusively ) where a particle is chosen randomly and a displacement @xmath260 , uniformly distributed over a disk of radius @xmath261 , from the current position @xmath262 of the particle is attempted . the corresponding energy change @xmath263 is calculated and the move is accepted using the standard metropolis criterion @xcite . the maxium displacement distance @xmath261 is chosen such that the acceptance rate @xmath234 remains reasonable , i.e. @xmath264 @xcite . as may be seen from the inset in fig . [ fig_quench_e_plj ] where the acceptance rate @xmath265 is shown as a function of temperature @xmath5 , we find that temperatures below @xmath266 are best sampled using @xmath267 , whereas @xmath268 is a good choice for the interesting temperature regime between @xmath269 and @xmath270 around the glass transition temperature @xmath271 . ( see the main panel of fig . [ fig_quench_rho ] for the determination of @xmath9 . ) various data sets are presented in sec . [ sec_res ] for a fixed value @xmath272 which allows a reasonable comparison of the sampling time dependence for temperatures around the glass transition . this value is not necessarily the best choice for tempering the system and for computing static equilibrium properties at the given temperature , especially below @xmath273 . [ [ collective - plane - wave - mc - moves . ] ] collective plane - wave mc moves . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + local mc jumps in the @xmath37-ensemble become obviously inefficient for relaxing large scale structural properties @xcite for the lowest temperatures computed for the plj model ( @xmath274 ) as may be seen , e.g. , from the finite , albeit very small shear stresses @xmath4 which happen to occur naturally in some of the quenched configurations ( especially if the quench rate @xmath275 is too large ) and can only with difficulty be relaxed using local jumps _ alone _ @xcite . we have thus crosschecked and improved the values obtained using only local mc moves by adding global mc move attempts for all particles with longitudinal and ( more importantly ) transverse plane waves commensurate with the simulation box . the random amplitudes and phases of the waves are assumed to be uniformly distributed . note that the maximum amplitude @xmath276 of each wave is chosen inversely proportional to the length @xmath277 of the wavevector in agreement with continuum theory . this allows to keep the same acceptance rate for each global mode if @xmath278 . as one expects , the plane - wave displacement attempts becomes inappropriate for large @xmath279 where continuum elasticity breaks down @xcite and , hence , the acceptance rate @xmath280 gets too large to be efficient @xcite . note that the maximum amplitude @xmath276 has to be chosen much smaller for the longitudinal waves as for the transverse ones , i.e. since the systems are essentially incompressible , the transverse modes naturally are more effective for quenching and equilibrating the systems . these global plane - wave moves have been added for the tempering of all presented configurations below @xmath270 with @xmath281 and for production runs over sampling times @xmath282 for static properties for @xmath283 . details concerning the collective plane - wave mc moves must be given elsewhere . [ [ mc - sampling - of - different - ensembles . ] ] mc sampling of different ensembles . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in this paper we shall be concerned more with the consequences of affine displacements associated with collective mc moves changing the overall box volume @xmath12 and shear strain @xmath3 . using again a metropolis criterion for accepting a suggested change of @xmath12 or @xmath3 , these pure dilatational and pure shear strain fluctuations are used to impose a mean normal pressure @xmath284 and a mean shear stress @xmath41 . for a shear strain altering mc move one first chooses randomly a strain fluctuation @xmath285 with @xmath286 . assuming an imposed shear stress @xmath4 the suggested @xmath285 is accepted if @xmath287 with @xmath288 being a uniformly distributed random variable with @xmath289 . the energy change @xmath263 associated with the affine strain may be calculated using eq . ( [ eq_strain_r ] ) . since @xmath290 is supposed in this study , the last term in the exponential drops out . ( as noted in sec . [ theo_stressfluctu ] , the volume remains unchanged in a pure shear strain and , hence , no translational entropy contribution appears . ) the maximum shear strain step @xmath291 is fixed such that at the given temperature the acceptance rate remains reasonable @xcite . as described in more detail in ref . @xcite , one similarly chooses a relative volume change @xmath292 with @xmath12 being the current volume @xcite . the volume altering move is accepted at an imposed pressure @xmath6 if @xmath293 where @xmath263 may be computed using eq . ( [ eq_strain_r_epsilon ] ) . the logarithmic contribution corresponds to the change of the translational entropy . using eq . ( [ eq_timposed ] ) after each mcs for local mc moves ( and global plane - wave mc moves if these are added ) realizes the @xmath49-ensemble shown in fig . [ fig_sketch](c ) , using eq . ( [ eq_pimposed ] ) the @xmath155-ensemble shown in fig . [ fig_sketch](d ) and using in turn both strain fluctuations the @xmath50-ensemble in fig . [ fig_sketch](e ) . we remind that @xmath37- and @xmath155-ensembles are used to determine @xmath54 and @xmath49- and @xmath50-ensembles to obtain @xmath48 and @xmath151 . as shown in fig . [ fig_quench_rho ] , we quench the configurations starting from @xmath294 in the @xmath155- and the @xmath50-ensemble using a constant quench rate @xmath275 . note that the smallest quench rate @xmath295 for one configuration in the @xmath155-shown did require alone a run over six months using one core of an intel xeon e5410 processor . interestingly , similar results are obtained with the @xmath50-ensemble only using a rate @xmath296 . ( due to the additional computation for the shear strain fluctuations this is , however , only a factor five faster . ) the configurations created by the @xmath155-quenches are used ( after tempering ) for the sampling in the @xmath37- and @xmath155-ensembles , the configurations obtained by the @xmath50-quenches for the sampling in the @xmath49- and @xmath50-ensembles . only two independent configurations have been sampled following the full protocol for each temperature and each of the four ensembles . instead of increasing further the number of configurations ( which will be done in the future ) we have focused in the present preliminary study on long tempering times and production runs over @xmath282 mcs which have been redone several times for a few temperatures ( summing up to total runs of @xmath297 mcs for @xmath269 and @xmath298 ) to check for equilibration problems and to verify that ageing effects can be ignored . we firmly believe that the breaking of the time translational invariance is not a relevant issue for the strong sampling time dependence reported below for the shear moduli . [ [ algo_sample ] ] glass transition temperature . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the inset of fig . [ fig_quench_rho ] presents for the ka model how the glass transition temperature @xmath9 may be obtained using standard dilatometry where the rescaled density @xmath299 is traced as a function of temperature @xmath5 as in ref . @xcite for a similar polymer glass problem . this reveals a linear low-@xmath5 ( solid line ) and a linear high-@xmath5 regime . the intercept of both lines confirms the well - known value @xmath300 from the literature @xcite . the main panel presents the unscaled number density @xmath253 for plj systems at constant normal pressure @xmath284 as a function of temperature @xmath5 for different quench rates @xmath275 in mcs . only local mc moves with a fixed maximum particle displacement @xmath301 for all temperatures @xmath5 are used for the examples given . the crosses refer to a quench in the @xmath50-ensemble where the box shape is allowed to fluctuate at @xmath41 . all other data refer to the @xmath155-ensemble with @xmath36 . the solid line and the dashed line are linear fits to the best density estimates indicated by the stars obtained for , respectively , low and high temperatures . by matching both lines one determines @xmath302 . a similar value is given if @xmath299 is plotted as a function of @xmath5 . the main panel of fig . [ fig_quench_e_plj ] shows the interaction energy per particle @xmath303 of the plj model as a function of temperature for the same quench protocols as in fig . [ fig_quench_rho ] . the solid line and the dashed line are linear fits to the energy for , respectively , low and high temperatures . this confirms @xmath302 for the plj model @xcite . we begin our discussion by focusing on the high temperature liquid limit where all reasonable operational definitions of the shear modulus @xmath0 must vanish . the results presented in fig . [ fig_gt_thigh ] have been obtained for the plj model at mean pressure @xmath284 , mean shear stress @xmath304 and mean temperature @xmath305 , i.e. far above the glass transition temperature @xmath302 . only local mc monomer displacements with a maximum jump displacement @xmath272 are reported here . instantaneous properties relevant for the moments are written down every @xmath306 mcs and averaged using standard gliding averages @xcite , i.e. we compute mean values and fluctuations for a given time interval @xmath307 $ ] and average over all possible intervals of length @xmath32 . the horizontal axis indicates the latter interval length @xmath32 in units of mcs . the vertical axis is made dimensionless by rescaling the moduli with the ( corrected ) affine shear elasticity @xmath308 from table [ tab_plj ] . ( being a simple average , @xmath26 is found to be identical for all computed ensembles . ) in agreement with eq . ( [ eq_gbound ] ) , the ratio @xmath309 for all operational definitions drops rapidly below unity . the dashed line indicates the asymptotic power - law slope @xmath310 @xcite . this can be understood by noting that at imposed @xmath4 the shear strain freely diffuses in the liquid limit , i.e. @xmath311 , which implies that @xmath312 . for both @xmath49- and @xmath50-ensembles we compare our key definition @xmath48 , eq . ( [ eq_gmodgt ] ) , to the observable @xmath151 , eq . ( [ eq_gmodgg ] ) . in both cases we confirm @xmath313 for larger times as suggested by the thermodynamic relation eq . ( [ eq_dgamdtau ] ) which has been directly checked @xcite . we have also computed the dimensionless correlation coefficient @xmath314 , eq . ( [ eq_cmodgt ] ) , which characterize the correlation of the measured instantaneous shear strains @xmath46 and shear stresses @xmath30 . as indicated for the @xmath50-ensemble ( stars ) , @xmath90 vanishes rapidly with @xmath32 , i.e. @xmath46 and @xmath30 are decorrelated . the dash - dotted line indicates the power - law exponent @xmath315 expected from eq . ( [ eq_cmodgt_mua ] ) and @xmath316 . note that by plotting @xmath314 _ vs. _ @xmath317 , eq . ( [ eq_cmodgt_mua ] ) has been directly verified for both @xmath49- and @xmath50-ensembles ( not shown ) . the stress fluctuation formula , eq . ( [ eq_gmodtt ] ) , is represented in fig . [ fig_gt_thigh ] by squares and diamonds for , respectively , @xmath37- and @xmath155-ensembles . the bare shear modulus @xmath222 _ without _ the impulsive correction @xmath200 for the born coefficient @xmath28 discussed in sec . [ theo_trunc ] is given by small filled symbols . the impulsive correction @xmath318 ( table [ tab_plj ] ) computed independently from the weighted radial pair correlation function , eq . ( [ eq_histolame ] ) , is shown by a horizontal line . the corrected moduli @xmath319 vanish indeed as expected . see ref . @xcite for a systematic numerical characterization of @xmath200 as a function of the reduced cutoff @xmath189 . as may also be seen there , a truncation correction is also necessary for the ka model . the @xmath200 for different temperatures may be found in table [ tab_ka ] for the ka model and in table [ tab_plj ] for the plj model . we assume from now on that the impulsive corrections are properly taken into account without stating this technical point explicitly . [ [ introduction.-2 ] ] introduction . + + + + + + + + + + + + + we turn now to the opposite low-@xmath5 limit focusing specifically on the plj model at @xmath320 . obviously , such deeply quenched colloidal glasses must behave as amorphous solids with a constant shear modulus for large times ( albeit not infinite times ) as sketched in fig . [ fig_sketch](b ) by the top curve . before presenting in sec . [ res_tlow ] the elastic properties of these glasses , we discuss first conceptually simpler substitute systems formed by _ permanent _ spring networks . we do this for illustration purposes since questions related to ergodicity and ageing are by construction irrelevant and the thermodynamical relations reminded in sec . [ sec_theo ] should hold rigorously . [ [ network - hamiltonian . ] ] network hamiltonian . + + + + + + + + + + + + + + + + + + + + assuming ideal harmonic springs the interaction energy reads @xmath321 with @xmath322 being the spring constant and @xmath323 the reference length of spring @xmath130 . the sum runs over all springs @xmath130 between topologically connected vertices @xmath131 and @xmath132 of the network . we assume here a strongly and homogeneously connected network where every vertex @xmath131 is in contact with many neighbors @xmath132 . such a network may be constructed from a plj configuration at @xmath320 keeping the particle positions for the positions of the vertices and replacing each lj interaction @xmath130 by a spring of spring constant @xmath322 and reference length @xmath323 . the simplest choice we have investigated is to set @xmath324 and @xmath325 which corresponds to a well - defined ground state of energy @xmath326 and pressure @xmath327 at @xmath328 ( not shown ) . we discuss here instead a slightly more realistic choice of @xmath322 and @xmath323 which corresponds to the _ same _ dynamical matrix " as the plj configuration at @xmath329 and @xmath284 , i.e. the same second derivative of the total interaction potential with respect to the particle positions @xcite . this choice imposes the setting @xmath330 with @xmath331 being the interaction potential for reduced distances @xmath332 . ( impulsive corrections at @xmath333 are neglected here . ) [ [ sampling . ] ] sampling . + + + + + + + + + taking advantage of the fact that all interacting vertices are known , these networks can be computed using global mc moves with longitudinal and transverse planar waves as described in sec . [ sec_algo ] . no local mc jumps for individual vertices have been added . note that although some @xmath322 and @xmath323 may even be negative , this does not affect the global or local stability of the network . presumably due to the fact that our spring constants are rather large , we have not observed either any buckling instability if volume fluctuations are allowed . for networks at the same pressure @xmath6 , shear stress @xmath4 and temperature @xmath5 as the original plj configuration , we obtain , not surprisingly , the same affine shear elasticity @xmath334 as for the reference . the same applies for other simple averages . [ [ shear - modulus . ] ] shear modulus . + + + + + + + + + + + + + + figure [ fig_gt_dynmat ] presents various measurements of the shear modulus @xmath0 for different ensembles as a function of sampling time @xmath32 . as in sec . [ res_thigh ] , we use gliding averages and the vertical axis has been rescaled by the affine shear elasticity @xmath26 . as can be seen , we obtain in the long - time limit @xmath335 ( bold line ) irrespective of whether the shear modulus is determined from @xmath48 or @xmath151 in the @xmath49- or @xmath50-ensembles or using @xmath54 in the @xmath37- or @xmath155-ensembles at fixed shear strain @xmath3 . note that even in these cases the shear modulus decreases first with time albeit the network is perfectly at thermodynamic equilibrium and no ageing occurs by construction . as shown by the dash - dotted line , @xmath336 for short times , i.e. in this limit the response is affine . as indicated by the stars , we have also computed the correlation coefficient @xmath314 as a function of time . from @xmath92 for small @xmath32 , it decreases somewhat becoming @xmath337 for large times in agreement with eq . ( [ eq_cmodgt_mua ] ) . if @xmath54 is computed in the wrong " @xmath49- or @xmath50-ensembles , it is seen to vanish rapidly with time as shown by the filled squares . this decay is of course expected from eq . ( [ eq_muamuftau ] ) as discussed in sec . [ theo_stressfluctu ] and sec . [ theo_legendre ] . [ [ shear - stress - fluctuations . ] ] shear - stress fluctuations . + + + + + + + + + + + + + + + + + + + + + + + + + + to emphasize the latter point we show in fig . [ fig_muft_dynmat ] the stress fluctuations @xmath338 for different ensembles . this is also done to have a closer look at the time dependence of various contributions to @xmath339 . as before , the vertical axis has been rescaled with @xmath340 . we also indicate the rescaled values of @xmath341 ( thin top line ) and @xmath342 ( dash - dotted line ) computed by a gliding average over time intervals @xmath307 $ ] . being simple averages both properties do not depend on the ensemble probed ( not shown ) and become virtually immediately @xmath32-independent as can be seen from the indicated horizontal lines . this is quite different for the different shear stress fluctuations @xmath55 presented which increase monotonously from @xmath343 for short times to their plateau value for long times . in agreement with eq . ( [ eq_muamuftau ] ) we find that @xmath344 for @xmath49- and @xmath50-ensembles and as predicted by eq . ( [ eq_muftransform ] ) we confirm @xmath345 for @xmath37- and @xmath155-ensembles . the asymptotic limit is reached after about @xmath346 mcs in the first case for imposed @xmath4 , but only after @xmath347 for imposed @xmath3 . this is due to the fact that @xmath174 corresponds to the fluctuations of the self - contribution of individual springs as discussed in sec . [ theo_legendre ] , while @xmath348 also contains contributions from stress correlations of different springs . it should be rewarding to analyze in the future finite system - size effects and the time dependence of three - dimensional ( 3d ) systems using the dynamical matrix associated to the ka model in the @xmath329 limit . we return now to the colloidal glasses where the topology of the interactions is , of course , _ not _ permanently fixed ( at least not at finite temperatures ) and the shear stresses may thus fluctuate more strongly . as an example , we present in fig . [ fig_gt_tlow ] the shear modulus @xmath349 as a function of sampling time @xmath32 for plj beads at mean pressure @xmath284 , mean shear stress @xmath350 and mean temperature @xmath320 , i.e. the same conditions as in the previous subsection . all simple averages , such as the affine shear elasticity @xmath351 , are the same for all ensembles albeit different quench protocols have been used . for all examples given we use local mc jumps with @xmath301 which allows to keep a small , but still reasonable acceptance rate @xmath352 . we have also performed runs with additional global mc moves using longitudinal and transverse plane waves . this yields similar data which is shifted horizontally to the left ( not shown ) . as one would expect , the shear moduli presented in fig . [ fig_gt_tlow ] are similar to the ones shown in fig . [ fig_gt_dynmat ] for the permanent spring network . the filled triangles show @xmath54 computed in the wrong " @xmath50-ensemble . as expected from eq . ( [ eq_muftransform ] ) these values vanish rapidly with time . if on the other hand @xmath48 , @xmath151 and @xmath54 are computed in their natural ensembles , all these measures of @xmath0 are similar and approach , as one expects for such a low temperature , the same finite asymptotic value @xmath353 indicated by the bold horizontal line . the stars indicate the squared correlation coefficient @xmath354 which is seen to collapse perfectly on the rescaled modulus @xmath317 which shows that eq . ( [ eq_cmodgt_mua ] ) even holds for small times before the thermodynamic plateau has been reached . similar reduced shear moduli @xmath355 have been also obtained for the ka model in the low-@xmath5 limit as may be seen from table [ tab_ka ] . these results confirm that about half of the affine shear elasticity is released by the shear stress fluctuations in agreement with refs . interestingly , this value is slightly smaller as the corresponding value for the permanent dynamical matrix model presented above . since some , albeit very small , rearrangement of the interactions must occur for the plj particles at finite temperature , this is a reasonable finding . the rescaled shear stress fluctuations @xmath356 for the same ensembles are presented in fig . [ fig_muft_tlow ] . while the two - point correlations @xmath357 ( thin horizontal line ) and @xmath358 ( dash - dotted horizontal line ) do not dependent on time , one observes again that @xmath356 increases monotonously from zero to its long - time asymptotic plateau value . in agreement with eq . ( [ eq_muamuftau ] ) we find @xmath344 for the @xmath49 and the @xmath50 ensembles , while @xmath345 ( dashed horizontal line ) for the @xmath37 and the @xmath155 ensembles . ( similar behavior has been obtained for the ka model in the @xmath37-ensemble for temperatures @xmath274 . ) as may be better seen for the data set obtained for @xmath359 using the @xmath37-ensemble ( small filled squares ) where @xmath360 , the stress fluctuations @xmath361 do not rigorously become constant , but increase extremely slowly on the logarithmic time scales presented . there is , hence , even at low temperatures always some sampling time dependence if the interactions are not permanently fixed as for the topologically fixed networks discussed above . however , this effect becomes only sizeable above @xmath269 for the ka model and above @xmath266 for the plj model . we shall now attempt to characterize this temperature dependence . [ [ stress - fluctuations . ] ] stress fluctuations . + + + + + + + + + + + + + + + + + + + + we turn now to the characterization of the temperature dependence of the shear modulus @xmath0 and various related properties . we focus first on the best values obtained for the longest simulation runs available . figure [ fig_muamuf_t ] presents the stress fluctuations obtained for the ka model using the @xmath37-ensemble ( open spheres ) and for the plj model using @xmath37- and the @xmath50-ensembles . the affine shear elasticity @xmath26 obtained for each system is represented by small filled symbols . as shown for both models by a bold solid line for the low-@xmath5 and by a dashed line for the high-@xmath5 regime , @xmath8 decays roughly linearly with @xmath5 . interestingly , the slopes match precisely at @xmath362 for both models . as may be seen from the large triangles for the @xmath50-ensemble , we confirm for all temperatures that @xmath363 as expected from eq . ( [ eq_muamuftau ] ) . for systems without box shape fluctuations , it is seen that @xmath364 becomes _ non - monotonous _ with a clear maximum at @xmath365 . in the liquid regime for @xmath366 where the boundary conditions do not matter , we obtain @xmath367 , i.e. all shear stress fluctuations decay with temperature . below @xmath9 the boundary constraint ( @xmath36 ) becomes more and more relevant with decreasing @xmath5 which reduces the shear stress fluctuations . according to eq . ( [ eq_muftransform ] ) , @xmath364 must thus decrease with increasing @xmath0 as the system is further cooled down . that @xmath364 must necessarily be non - monotonous with a maximum at @xmath9 is one of the central results of the presented work . [ [ shear - modulus - as - a - function - of - temperature . ] ] shear modulus as a function of temperature . + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the temperature dependence of the ( unscaled ) shear moduli @xmath35 is presented in fig . [ fig_gt ] . data obtained by md simulation of the ka model ( @xmath368 , @xmath300 ) are indicated for the @xmath37-ensemble ( open spheres ) and the @xmath155-ensemble ( filled spheres ) . all other results refer to the best values obtained by mc simulations of the plj model ( @xmath284 , @xmath302 ) using both local and global moves and different ensembles as indicated . for the plj model the values @xmath54 obtained in the @xmath37 ensemble ( squares ) are seen to be essentially identical for all @xmath5 to the moduli @xmath48 ( crosses ) and @xmath151 ( triangles ) obtained in ensembles with strain fluctuations . as shown by the small filled triangles , @xmath54 vanishes in the @xmath50 ensemble for all @xmath5 in agreement with eq . ( [ eq_muamuftau ] ) . ( we remind that the impulsive correction @xmath200 has to be taken into account if @xmath54 is used . ) decreasing the temperature further below @xmath9 the shear moduli are seen to increase rapidly for both models . as shown by the solid and the dashed lines both models are well fitted by a cusp - singularity @xmath369 with fit constants @xmath370 for the ka model and @xmath371 for the plj model where we have set in both cases @xmath372 . confirming the md simulations by barrat _ @xcite , this suggests that the transition is very sharp , albeit _ continuous _ in qualitative agreement with the predictions from ref . @xcite . if we fit the data close to @xmath9 with an additional off - set @xmath373 , a slightly _ negative _ value is obtained which is not compatible with mct @xcite . however , admittedly both the number of data points close to @xmath9 and their precision ( due to the small number of independent configurations ) are yet not sufficient to rule out completely a positive , albeit ( presumably ) small off - set . a slightly different representation of the data is given in fig . [ fig_gt_rescal ] where the reduced shear modulus @xmath374 is plotted as a function of the reduced temperature @xmath375 for both models . using the independently determined glass transition temperature @xmath9 and affine shear elasticity @xmath8 as scales to make the axes dimensionless , the rescaled data are found to collapse ! this is of course a remarkable and rather unexpected result considering that two different models in two different dimensions have been compared . the bold line indicates the cusp - singularity @xmath376 with a prefactor which is compatible to the ones used in eq . ( [ eq_cuspcont ] ) considering the typical values of @xmath8 in both models . whether this striking collapse is just due to some lucky coincidence or makes manifest a more general universal scaling is impossible to decide at present . ( please note that the corresponding data for the compression modulus shown in fig . [ fig_kt ] does not scale . ) no rescaling of the vertical axis is necessary for the squared correlation coefficient @xmath377 shown in the inset for two different ensembles of the plj model . the same continuous cusp - like decay ( bold line ) is found as for the modulus @xmath0 . this is again consistent with the thermodynamic reasoning put forward at the end of sec . [ theo_strainfluctu ] , eq . ( [ eq_cmodgt_mua ] ) . [ [ sampling - time - dependence . ] ] sampling time dependence . + + + + + + + + + + + + + + + + + + + + + + + + + a word of caution must be placed here . the results presented in figs . [ fig_muamuf_t]-[fig_gt_rescal ] correspond to the longest simulation runs we have at present been able to perform . obviously , this does _ not _ mean that they correspond to the limit for asymptotically long sampling times , if the latter limit exists , or at least an intermediate plateau of respectable period of time . focusing on the plj model in the @xmath50-ensemble we attempt in fig . [ fig_gt_t ] to give a tentative characterization of the sampling time dependence . we plot @xmath378 as a function of @xmath32 for different temperatures as indicated . to make the data comparable all simulations for @xmath379 have been performed with local mc jumps of same maximum distance @xmath272 . as we have already seen in fig . [ fig_gt_thigh ] , the shear modulus decays inversely with time in the liquid regime at high temperatures . this limit is indicated by the thin line . with decreasing @xmath5 the ( unscaled ) shear modulus increases and a shoulder develops . note that even for @xmath298 , i.e. slightly below the glass transition temperature @xmath380 estimated via dilatometry ( fig . [ fig_quench_rho ] ) , @xmath378 decays strongly with time . a more or less @xmath32-independent shoulder ( on the logarithmic scales used ) can only be seen below @xmath273 . we emphasize that we present here a sampling time effect expressing the fact that the configuration space is more and more explored with increasing @xmath32 and _ not _ a time correlation function @xcite . please note also that this time dependence has nothing to do with equilibration problems or ageing effects . time translational invariance is perfectly obeyed in our simulations as we have checked by rerunning the sampling _ after _ additional tempering over at least @xmath347 mcs for all temperatures . similar @xmath32-dependencies for different temperatures @xmath5 have also been observed for @xmath378 and @xmath339 for the plj model and for @xmath339 for the ka model ( not shown ) . a different representation of the data for both models is shown in fig . [ fig_gt_t ] where the reduced shear modulus @xmath381 is plotted against the reduced temperature @xmath382 for different sampling times @xmath32 as indicated . similar behavior is found for both models . the shear modulus decreases with @xmath32 and this the more the larger the temperature @xmath5 . for smaller temperatures the different data sets appear to approach more readily as one expects from the @xmath32-independent shoulder for the plj model in fig . [ fig_gt_t ] . as a consequence the glass transition is seen to become sharper with increasing @xmath32 . the data approach the continuous cusp - singularity indicated by the bold line . however , it is not clear from the current data if the data converge to an at least _ intermediately _ stable behavior . longer simulation runs are clearly necessary for both models to clarify this issue . similar data have also been obtained for the plj model from the independent determination of @xmath378 and @xmath383 in the @xmath49-ensemble and for @xmath339 in the @xmath37- and @xmath155-ensemble . the stress fluctuation formalism is a powerful method for computing elastic moduli for canonical ensembles using simulation boxes of constant volume and shape @xcite . focusing on the shear modulus @xmath0 of two well - known glass - forming colloidal model systems in @xmath1 and @xmath2 dimensions , which we have sampled by means of md and mc simulations , we have addressed the general question of whether the stress fluctuation method may be used through a solid - liquid transition where a well - defined displacement field ceases to be defined . deliberately setting the experimentally motivated linear regression eq . ( [ eq_gmodgt ] ) as the fundamental definition , the shear modulus @xmath0 has been computed from the strain and stress fluctuations ( assuming a fixed finite measurement time window @xmath384 ) in ensembles where either a shear strain @xmath36 ( @xmath37- and @xmath155-ensembles ) or a ( mean ) shear stress @xmath41 ( @xmath49- and @xmath50-ensembles ) have been imposed ( fig . [ fig_sketch ] ) . working at constant mean normal pressure we have computed and compared the temperature dependence for various simple averages and fluctuations contributing to the shear modulus in different ensembles . for the ka model only the currently available results for @xmath37- and @xmath155-ensembles have been presented . as has been stressed in sec . [ theo_stressfluctu ] and sec . [ theo_legendre ] , the stress fluctuation representation @xmath54 at fixed shear strain does not rely _ in principle _ on a solid - like reference state for a displacement field of tagged particles if thermodynamics can be assumed to hold . it thus holds formally through the glass - transition temperature @xmath9 up to the liquid regime ( albeit with a trivial value @xmath34 ) as does the better known relation for the compression modulus @xmath20 ( appendix [ app_k ] ) @xcite . as emphasized in sec . [ theo_trunc ] , impulsive corrections due to the truncation of the pair potentials have to be properly taken into account , especially at high temperatures ( fig . [ fig_gt_thigh ] ) , for the precise determination of the affine shear elasticity @xmath255 @xcite . confirming eq . ( [ eq_gmods ] ) , it has been shown for the plj model that @xmath54 at @xmath36 has the _ same _ long - time behavior as the conceptually more direct observables @xmath48 and @xmath151 obtained using the strain fluctuations at constant mean shear stress @xmath41 . the standard thermodynamic relations comparing different ensembles appear thus to hold also _ in practice _ ( at least within our numerical precision ) through the glass transition for all @xmath5 and this albeit 1 . some degrees of freedom get quenched at low temperatures on the time window @xmath384 computationally accessible , 2 . all operational definitions of @xmath0 are transient in the sense that they vanish for @xmath385 ( figs . [ fig_gt_t ] and [ fig_gt_t ] ) and 3 . it is not self - evident that in the latter limit and for large temperatures @xmath3 and @xmath4 can still be treated as a pair of conjugated thermostatistical variables . the latter point compactly epitomized by the thermodynamic relation eq . ( [ eq_dgamdtau ] ) has been shown to hold , however , with remarkable precision up to very high temperatures for the plj model . as predicted by general legendre transformation , @xmath54 vanishes for all temperatures @xmath5 , if @xmath4 rather than @xmath3 is imposed ( figs . [ fig_gt_dynmat ] , [ fig_gt_tlow ] and [ fig_gt ] ) . the shear - stress fluctuations @xmath174 are thus given by the affine response under an external load @xmath26 ( fig . [ fig_muamuf_t ] ) , i.e. @xmath174 reduces to a simple two - point pair correlation function if pair potentials are considered as discussed in sec . [ theo_legendre ] . the same holds above @xmath9 for constant @xmath3 , since the boundary conditions are irrelevant for the liquid state . this symmetry with respect to the boundary conditions ( @xmath386 ) is broken at the glass transition for a large , but finite @xmath384 : at constant @xmath3 the stress fluctuations reveal a strong non - monotonous behavior with a clear maximum at @xmath9 ( fig . [ fig_muamuf_t ] ) . the shear modulus is the order parameter characterizing this symmetry breaking . alternatively , as we have discussed in sec . [ theo_legendre ] , @xmath387 may be seen as an order parameter comparing the ratio of the non - affine to the affine shear responses . since @xmath388 for @xmath389 , this implies that the stress fluctuations contain higher - order correlations _ reducing _ the free energy of the system . the increase of @xmath0 below @xmath9 is reasonably fitted for both models by a _ continuous _ cusp singularity , eq . ( [ eq_cuspcont ] ) , in qualitative agreement with recent replica calculations @xcite . a jump discontinuity , as suggested by mode - coupling theory @xcite and another replica theory @xcite , is not compatible with the currently available data shown in fig . [ fig_gt ] and fig . [ fig_gt_rescal ] . however , as shown in fig . [ fig_gt_t ] and fig . [ fig_gt_t ] , our data depend strongly on sampling time @xmath32 and the computation of larger @xmath32 could lead to a sharper transition . thus a jump discontinuity can not be ruled out completely . ( we are not aware of any other investigation of the sampling time effect for the shear modulus . ) at present we believe , however , that it would be surprising if our data could be reconciled with a discontinuity at @xmath9 . it seems more likely that below @xmath9 , where the choice of the boundary conditions does matter as shown , the definition of the glassy shear modulus used in refs . @xcite do not correspond to the key operational definition @xmath48 of the shear modulus considered by us . in any case , our work shows that any theory and numerical scheme used to determine the shear modulus using a stress or strain fluctuation relation ( in the low - wavevector limit or at a small finite @xmath279 ) must cleary specify and take into account whether the shear stress or the shear strain are macroscopically imposed . otherwise , such an approach is void . [ [ beyond - the - presented - work . ] ] beyond the presented work . + + + + + + + + + + + + + + + + + + + + + + + + + + future work should clearly focus on the more detailed description of the sampling time dependence . by extrapolating appropriately for the large-@xmath32 asymptotic behavior , this may allow to settle the theoretical debate . as emphasized above , one shortcoming of the present work is that the time - scale used in our mc simulations was slightly arbitrary , especially if additional non - local particle moves are used in the low-@xmath5 limit and if box volume and shape changing strains are included . the comparison of @xmath32-dependent properties for different temperatures @xmath5 becomes thus delicate . our mc study for the 2d soft particles should thus in any case be recomputed with langevin thermostat md dynamics as used for the data of the ka model presented . the @xmath32-dependence of the latter model , which is a better glass former than the 2d plj model @xcite , has still to be worked out for the key operational definition @xmath48 in the @xmath49- or @xmath50-ensembles @xcite . we plan also in the near future to reconsider more carefully our previous investigation of the glass transition of polymer melts @xcite taking properly into account the impulsive truncation corrections and comparing the shear moduli @xmath390 ( @xmath50-ensemble ) and @xmath54 ( @xmath50-ensemble ) around @xmath9 . in addition , it should be rewarding to compare our shear moduli with the values obtained from the displacement of particles following the procedure chosen in ref . this procedure relies on the idea that the particles fluctuate around a well - defined reference position . in the low-@xmath5 limit where the interaction network can be replaced by the dynamical matrix , this should yield the same results . however , for larger temperatures where the harmonic approximation must break down , this approach is questionable ( both for the analysis of experimental and computational data ) and should be compared to the moduli obtained directly from the strain and stress fluctuations of the overall simulation box . [ [ a - simple - scalar - model . ] ] a simple scalar model . + + + + + + + + + + + + + + + + + + + + + + we are currently working on an extremely simplified scalar model for permanent and transient networks in @xmath2 dimensions which may help to clarify the theoretical debate @xcite . as sketched in the panel ( a ) of fig . [ fig_band ] , in this model beads ( filled spheres ) are glued on rigid rods which are only allowed to move horizontally . in its most simple implementation this is done by means of global mc moves using transverse plane waves with a wavevector pointing in the @xmath140-direction . the glued beads are connected by ideal harmonic springs where the spring constants @xmath322 and reference lengths @xmath323 are chosen randomly according to fixed , narrow or broad distributions . the stiffness of the contacts between rods may differ ( even including negative @xmath322 and @xmath323 as for the dynamical matrix studied in sec . [ res_dynmat ] ) . the shear modulus may be computed either from the shear strain fluctuations in the @xmath49- or the shear stress fluctuations in the @xmath37-ensemble . the shear stresses can be relaxed by either allowing the particles to move along the rods using longitudinal plane waves in the @xmath105-direction and/or by breaking and reconnection of springs . the latter step is currently done by applying a chemical potential for the springs . physically , the study of such transient networks is motivated by systems of hyperbranched polymer chains with sticky end - groups @xcite or telechelic polymers in water - oil emulsions @xcite . the associated relaxation frequency @xmath391 for a given spring @xmath130 is assumed to be set by an activation barrier @xmath392 which may differ for each spring . depending on the distribution of this activation barrier the shear modulus @xmath349 thus decays more or less rapidly with sampling time @xmath32 as sketched in panel ( b ) of fig . [ fig_band ] . the shear modulus becomes constant , if the random percolating network is permanent ( @xmath393 , @xmath394 ) or the relaxation negligible ( top solid line ) . the system behaves as a liquid ( bottom line ) for large ( mean ) @xmath395 , while a shoulder is seen in the intermediate frequency range . while this is confirmed by preliminary simulations , we do at present not understand the role of the quenched noise injected in the model which is seen to strongly influence the shear stress fluctuations @xmath55 ( especially if fluffy layers happen to appear ) . also the understanding of finite system - size effects is of course crucial for such a low - dimensional model . ( only systems containing @xmath396 particles have been considered so far . ) as sketched in panel ( c ) the key question is to describe accurately for asymptotically long simulation runs and asymptotically large box sizes the shape of @xmath35 around the glass transition temperature @xmath397 where the shear modulus vanishes and @xmath348 should again have a maximum . thanks the cnrs and the irtg soft matter for supporting her sabbatical stay in strasbourg , p.p . the rgion alsace and the irtg soft matter and f.w . the daad for funding . we are indebted to a. blumen ( freiburg ) and h. meyer , o. benzerara and j. farago ( all ics , strasbourg ) for helpful discussions . we have focused in the main part of this paper on the shear modulus @xmath0 of our glass - forming model systems and the comparison of the three observables @xmath48 , @xmath151 and @xmath54 . being isotropic the complete macroscopic elastic description of our systems only requires the numerical determination of _ one _ additional elastic modulus , the isothermal compression modulus @xcite , @xmath398 characterizing the elastic response with respect to a volumetric ( dilatational ) strain fluctuation as shown in fig . [ fig_sketch](d ) . as discussed in the standard textbooks @xcite , the extensive variable is here the volume @xmath399 and the conjugated intensive variable the negative pressure @xmath400 where @xmath401 and @xmath402 stand for the instantaneous values of the volume and the pressure . with these notations eq . ( [ eq_pkv ] ) is obviously consistent with eq . ( [ eq_ftom ] ) or eq . ( [ eq_htom ] ) . in analogy with @xmath48 for the shear modulus @xmath0 , the compression modulus @xmath20 may be determined in an @xmath155- or an @xmath50-ensemble using the linear regression relation @xmath403 where the index @xmath404 indicates that both the measured instantaneous volume @xmath401 and pressure @xmath402 have been used . the corresponding correlation coefficient is @xmath405 since from eq . ( [ eq_dxdi ] ) we have @xmath406 if @xmath6 is imposed , this implies that @xmath407 should be identical to the better known relation @xcite @xmath408 which corresponds to the operational definition @xmath151 discussed above . the compression modulus @xmath20 may of course also be determined using the pressure fluctuations in systems of imposed volume @xmath12 as sketched in fig . [ fig_sketch](c ) . as ( to our knowledge ) first derived by rowlinson @xcite , the corresponding stress fluctuation formula reads @xmath409 being the fluctuation of the total normal pressure ( corresponding to @xmath55 ) and @xmath259 the affine dilatational elasticity " ( corresponding to @xmath26 ) . we shall give below a proper operational definition for @xmath259 . here we only note that due to the general legendre transformation eq . ( [ eq_didi ] ) for intensive variables one knows immediately that @xmath410 this implies by comparison with eq . ( [ eq_kmodpp ] ) that @xmath411 in analogy to eq . ( [ eq_muamuftau ] ) for the shear , i.e. @xmath259 can be directly be determined from the pressure fluctuations in @xmath155- and @xmath50-ensembles . as before for the correlation coefficient @xmath90 , eq . ( [ eq_cmodgt_mua ] ) , it follows from eq . ( [ eq_etaaetafp ] ) that @xmath412 since @xmath413 , the affine dilatational elasticity sets an upper bound to the compression modulus @xmath20 . we reformulate here briefly the derivation of the stress fluctuation relation @xmath258 for the compression modulus @xmath20 of an isotropic system at imposed volume @xmath12 by rowlinson @xcite . with @xmath414 being the volume of the unperturbed simulation box we assume a small relative volume change @xmath415 with all coordinates equally deformed as , e.g. @xmath416 for the @xmath105-coordinate in @xmath19-dimensions . ( the argument @xmath99 denotes again the reference system . ) using a similar rescaling trick as for the shear modulus @xmath0 in sec . [ theo_stressfluctu ] , the interaction energy @xmath417 of a configuration @xmath98 of the strained system is expressed in terms of the coordinates ( state ) of the unperturbed system and the explicit metric parameter @xmath229 . due to the volume change we have now also to take into account the ideal gas ( kinetic ) contributions . we derive first the excess contribution @xmath29 to the total pressure @xmath418 and the excess contribution @xmath419 to the total compression modulus @xmath420 valid for an arbitrary conservative potential . the general relations eqs . ( [ eq_dlogzdgam]-[eq_ddzddgam ] ) stated above for an imposed shear strain @xmath3 still hold with the relative volume change @xmath229 replacing @xmath3 . using eq . ( [ eq_dzdgam ] ) it follows for the excess pressure that @xmath421 _ defining _ the instantaneous excess pressure @xcite . in the second step we have taken the limit @xmath422 and have dropped the argument @xmath99 . the average is again evaluated using the weights of the unperturbed system , eq . ( [ eq_average ] ) . using eq . ( [ eq_ddlogzddgam ] ) and eq . ( [ eq_ddzddgam ] ) one obtains for the compression modulus @xmath423 which is again understood to be taken at @xmath424 . the first term @xmath425 in eq . ( [ eq_kexstressfluctu ] ) corresponds to the change of the system energy assuming an affine strain transformation for all particle positions . it gives the excess contribution @xmath426 to the total affine dilatational elasticity @xmath427 . the second term corresponds to the excess contribution @xmath428 to the total normal pressure fluctuation @xmath429 mentioned in sec . [ app_kstrainfluctu ] . it corrects the overprediction of the affine strain contribution @xmath426 @xcite . we remind that for an ideal gas the isothermal compression modulus @xmath430 is given by the ideal gas pressure @xmath145 and that the ideal gas pressure fluctuation @xmath431 becomes @xmath432 at constant volume @xcite . interestingly , using again the general legendre transform eq . ( [ eq_didi ] ) one sees that @xmath433 for the ideal gas pressure fluctuations at imposed pressure . due the assumed additivity of the total system hamiltonian , the stress fluctuation formula for the total compression modulus may now be written @xmath434 while this result is useful for constant-@xmath12 ensembles , especially if mc simulations are used , it is yet not in a form exactly equivalent to the general legendre transform eq . ( [ eq_kmodpp ] ) . since the ideal and excess pressure fluctuations are decoupled , @xmath435 , this implies @xmath429 . one may thus rewrite eq . ( [ eq_kconstantv ] ) as @xmath436 with @xmath437 where we have used eq . ( [ eq_etafidp ] ) in the second step . assuming eq . ( [ eq_etaaexid ] ) assures that @xmath259 becomes equivalent to @xmath438 . this has the nice feature that not only the functional @xmath258 , eq . ( [ eq_kmodpp ] ) , should give the correct compression modulus @xmath20 in the @xmath37 and the @xmath49 ensembles , but also that @xmath258 must vanish properly if the wrong " @xmath155 and @xmath155 ensembles are used , just as @xmath54 vanishes for @xmath49- or @xmath50-ensembles . up to now we have not taken advantage of the fact that the conservative interaction potential is assumed to be a pair potential , eq . ( [ eq_upairinteraction ] ) . this will be used now to express @xmath29 and @xmath426 in terms of simple ( ensemble independent ) two - point correlation functions . equation ( [ eq_strain_x_epsilon ] ) implies that the squared distance @xmath114 between two particles transforms as @xmath439 it follows that @xmath440 where we have taken finally for both derivatives the limit @xmath422 . for the first two derivatives of a general function @xmath441 with respect to @xmath229 this implies @xmath442 for @xmath422 and dropping the argument @xmath99 on the right - hand sides . according to the general result eq . ( [ eq_peshat_us ] ) for the normal pressure and using eq . ( [ eq_strain_fronebeps ] ) one confirms for pair interactions the expected virial relation @xcite @xmath443 for the instantaneous excess pressure where the interactions between pairs of particles @xmath133 are again labeled by an index @xmath130 . using eq . ( [ eq_strain_rtwo_epsilon ] ) the excess contribution to the affine dilatational elasticity becomes @xmath444 is sometimes called the hypervirial " contribution to the compression modulus @xcite . it corresponds to the first term indicated in eq . ( [ eq_strain_frtwobeps ] ) . by comparing eq . ( [ eq_etab ] ) with eq . ( [ eq_mub ] ) it is readily seen that for isotropic systems @xmath445 i.e. @xmath256 and @xmath28 and , hence , the affine dilatational elasticity @xmath259 and affine shear elasticity @xmath26 are closely related . ( this justifies to call @xmath256 a born coefficient " . ) please note that eq . ( [ eq_kmodpp ] ) or eq . ( [ eq_kconstantv ] ) together with eq . ( [ eq_etaaex ] ) are perfectly consistent with the relations stated in the literature @xcite . we emphasize finally that , as discussed for the affine shear elasticity @xmath26 in sec . [ theo_stressfluctu ] , it is inconsistent to neglect the explicit excess pressure contribution to @xmath259 in eq . ( [ eq_etaaex ] ) but to keep the underlined term of the born coefficient @xmath256 , eq . ( [ eq_etab ] ) @xcite . the affine dilatational elasticity @xmath446 obtained for both models is presented in the inset of fig . [ fig_kt ] as a function of the reduced temperature @xmath447 . ( the values may also be found in table [ tab_ka ] and table [ tab_plj ] . ) we remind that for determining the born coefficient @xmath256 impulsive corrections need to be considered as discussed at the end of sec . [ theo_trunc ] . as shown by the small filled triangles , @xmath259 compares well with the total pressure fluctuation @xmath448 in an @xmath50-ensemble . the observed data collapse is expected from eq . ( [ eq_etaaetafp ] ) and eq . ( [ eq_etafidp ] ) . as indicated by bold and dashed lines for , respectively , the low- and high-@xmath5 limit @xmath259 decreases essentially linearly with temperature . the coefficients of the lines ( not shown ) are consistent with eq . ( [ eq_etabmub ] ) relating @xmath259 and @xmath26 and the fits given in fig . [ fig_muamuf_t ] for @xmath26 . we emphasize that the affine dilatational elasticity @xmath259 ( and thus @xmath26 ) is not only larger for the ka model but , more importantly that , is also increases more rapidly with decreasing @xmath5 than the plj model . the main panel of fig . [ fig_kt ] shows the best long - time values of the compression modulus @xmath20 for both models . the vertical axis is rescaled by @xmath259 . the rescaled stress fluctuation formula , eq . ( [ eq_kconstantv ] ) , for the @xmath37-ensemble is given for both models . for the plj model we also indicate @xmath407 ( crosses ) and @xmath449 ( large triangles ) for the @xmath50-ensemble . no rescaling with @xmath259 is necessary for the squared correlation coefficient @xmath450 ( stars ) also presented . a perfect collapse of all plj data is found confirming the equivalence of all operational definitions used . for both models we observe a strong increase of @xmath451 around @xmath452 with decreasing temperature . interestingly , the rescaled data of both models is similar but not identical . the mentioned increase around @xmath452 is clearly more pronounced for the ka model . note that if the unscaled @xmath20 is directly plotted against @xmath5 , this effect becomes even stronger ( not shown ) consistently with the scaling of @xmath259 presented in the inset . we note finally that for very low temperatures we observe for both models that the ratio @xmath453 approaches unity , i.e. the compression modulus is essentially dominated by the affine response . we remind that this is different for the reduced shear modulus for both models which is observed to approach @xmath454 in the low-@xmath5 limit ( fig . [ fig_gt_rescal ] ) , i.e. the shear stress fluctuations @xmath364 are more relevant than the pressure fluctuations @xmath455 in this limit . for the plj model we present in fig . [ fig_kt_thigh ] the sampling time dependence of various operation definitions of @xmath20 for the liquid limit at temperature @xmath456 . the horizontal axis indicates the sampling time @xmath32 in units of mcs , the vertical axis is made dimensionless using @xmath259 as natural scale . as shown for @xmath407 and @xmath449 obtained in the @xmath50-ensemble and for @xmath258 obtained using eq . ( [ eq_kconstantv ] ) in the @xmath37-ensemble , all operational definitions applied in their natural ensemble converge _ rapidly _ within @xmath457 mcs to the same plateau . ( this is a general finding for all temperatures . ) the plateau value for @xmath456 is indicated by the bold solid line . note that @xmath451 is smaller unity as expected from the discussion at the end of sec . [ app_kstrainfluctu ] . the small filled symbols refer to @xmath258 computed in the wrong " @xmath50-ensemble . the small diamonds have been computed using directly eq . ( [ eq_kconstantv ] ) . as indicated by the dashed line , this yields @xmath458 for large times . only if we compute @xmath436 using eq . ( [ eq_etaaexid ] ) , it does vanish properly as shown by the small triangles . this shows numerically that eq . ( [ eq_kconstantv ] ) , albeit perfectly fine for constant-@xmath12 ensembles , is just not the correct formulation allowing to make manifest the general legendre transformation eq . ( [ eq_etaftransform ] ) . this is , however , achieved using eq . ( [ eq_etaaexid ] ) . this finding holds rigorously for all temperatures as can be seen from the scaling of the small triangles in the inset of fig . [ fig_kt ] .

quantized vortices are among the hallmarks of superfluidity , and have been observed in the gaseous bose - einstein condensates ( becs ) of @xmath0 @xcite and @xmath1 @xcite , where repulsive interatomic interactions play a crucial role in nucleation and stabilization of vortices . in the case of attractive interactions , however , creating vortices by stirring the condensate in a harmonic trap is difficult because the vortex state is energetically unfavorable against decaying into the non - vortex ground state . a possible experimental scheme for creating a vortex state with attractive interactions is to prepare it in the repulsive regime and then switch the sign of the interaction to attractive by using , e.g. , the feshbach resonance @xcite . another possible scheme to create vortices in an attractive bec is to imprint topological phases on the condensate using either controlled interconversion between two components @xcite or spin rotation by adiabatic change of the magnetic field @xcite . the aim of the present paper is to study the dynamical instabilities and ensuing dynamics of vortex states with attractive interactions in 3d axisymmetric traps . there are two types of instabilities that cause a vortex state to decay : dynamical and thermodynamic instabilities . dynamical instabilities , or modulational instabilities , set in when the bogoliubov spectrum acquires an imaginary part which causes an exponential growth in the corresponding mode . vortex nucleation @xcite , vortex exchange between two components @xcite , splitting of a vortex with attractive interactions @xcite , and disintegration of a doubly - quantized vortex with repulsive interactions @xcite are all attributed to dynamical instabilities . thermodynamic instabilities are caused by interactions between the condensate and the thermal cloud accompanied by dissipation of the energy and angular momentum . reductions in the number of vortices @xcite , vortex bending @xcite , and vortex - lattice formation @xcite are associated with thermodynamic instabilities . the time scale of the latter instabilities at low temperatures is typically @xmath2 s or longer @xcite . in contrast , the time scale of dynamical instabilities can be much faster than that of thermodynamic instabilities since the former grows exponentially in time . we shall therefore focus on dynamical instabilities in this paper . in ref . @xcite , pu _ _ reported that singly- and doubly - quantized vortices in 2d exhibit a series of dynamical instabilities when the strength of interaction exceeds the corresponding critical values . we numerically solved the time - dependent gross - pitaevskii ( gp ) equation for a 2d system , and showed that the vortex splits into two clusters which revolve around the center of the trap , and undergo split - merge cycles or collapse @xcite . in the present paper , we examine such dynamics in full 3d systems and also study the dynamical instability of the doubly - quantized vortex , that was recently realized by the mit group @xcite . our primary findings are the following : + ( 1 ) our 3d analysis reveals that the dynamical instabilities depend crucially on the trap geometry . for a singly - quantized vortex , split - merge cycles can occur in a pancake - shaped trap , whereas the split fragments immediately collapse in a spherical trap . in a cigar - shaped trap , there is a parameter range in which the vortex can collapse into a single cluster due to the dipole instability . + ( 2 ) the doubly - quantized vortex with attractive interactions is always unstable against disintegration of the vortex core . if the strength of interaction is suddenly changed to within a certain range , the vortex splits into three clusters which undergo split - merge cycles before one or possibly more of them collapse . + ( 3 ) even when the lifetime of the condensate is very short , the phenomenon of vortex split can be observed by suddenly changing the interaction to a large attractive value . the split of a doubly - quantized vortex is shown to be observable using a current experimental setup of the mit group . + ( 4 ) using a variational method , we find that both the split instability and the split - merge cycles can be understood by using only three basis functions . + ( 5 ) in the immediate vicinity of the critical strength of interaction for the vortex split , quantum fluctuations like the ones found in a 1d case @xcite become prominent . this paper is organized as follows . section [ s : formulation ] briefly reviews the gp and bogoliubov theories and discusses dynamical instabilities . section [ s : single ] studies dynamical instabilities of a singly - quantized vortex state , and sec . [ s : double ] those of a doubly - quantized vortex state . section [ s : variational ] examines a variational method to understand the dynamical instabilities and split - merge phenomenon . section [ s : lowly ] shows that the split - merge phenomenon is accompanied by the emergence of a symmetry - broken low - lying state . section [ s : conclusion ] concludes this paper . we consider a dilute - gas bec at zero temperature confined in an axisymmetric harmonic trap . the dynamics of the condensate are described by the gp equation @xmath3 where @xmath4 is the atomic mass , @xmath5 with @xmath6 and @xmath7 being the radial and axial trap frequencies , and @xmath8 is the s - wave scattering length . we normalize the length , time , energy , and wave function by @xmath9 , @xmath10 , @xmath11 , and @xmath12 , respectively , where @xmath13 is the number of atoms in the bec , and the wave function is normalized as @xmath14 . the gp equation ( [ gp ] ) is then written as @xmath15 where @xmath16 , @xmath17 , and @xmath18 . the bogoliubov spectrum for a stationary state @xmath19 of the gp equation is obtained from the bogoliubov - de gennes equations [ bogo ] @xmath20 where @xmath21 and @xmath22 is the chemical potential . for the axisymmetric vortex states @xmath23 , each bogoliubov mode @xmath24 couples only to @xmath25 , where @xmath26 and @xmath27 are integers and @xmath28 is the azimuthal angle . in the following discussions , we shall refer to the modes with @xmath29 , @xmath30 , and @xmath31 as the dipole , quadrupole , and octupole modes , respectively . the stability of the system is characterized by the bogoliubov spectrum @xmath32 . when the system is in the ground state , or in a metastable state for which there is an energy barrier that prevents the system from decaying into a lower energy state , all bogoliubov eigenenergies are real and positive . an example of such a metastable state is a bec with an attractive interaction below the critical strength of interaction for collapse @xcite . when there are negative eigenvalues in the bogoliubov spectrum , the system can thermodynamically decay into a lower energy state by dissipating energy to the environment . when complex eigenvalues emerge in the bogoliubov spectrum , the corresponding modes grow exponentially in time and the system becomes dynamically unstable . the time scale of such dynamical instability is given by the inverse of the imaginary part . the difference between thermodynamic and dynamical instabilities is that the former requires energy dissipation , whereas the latter can occur via an energy - conserving process . thus , the dynamical instability can be a dominant decay process at sufficiently low temperature in a high - vacuum chamber in which dissipation is negligible over a short time . the time evolution of a system that deviates slightly from the stationary state @xmath19 of the gp equation can be described in terms of the bogoliubov modes @xmath33 and @xmath34 as @xcite @xmath35,\ ] ] where @xmath36 s ( @xmath37 ) are small coefficients that describe the deviation . when the eigenvalue of a bogoliubov mode becomes complex , the mode grows exponentially in time as @xmath38 \nonumber \\ & = & e^{i ( q \phi - \mu t ) } \bigl\ { \tilde \psi_0 + e^{{\rm im}e_n t } re}e_n t + \theta_n ) + ( \tilde u_n- \tilde v_n ) \sin(\delta m \phi - { \rm re}e_n t + \theta_n ) \bigr ] \bigr\},\end{aligned}\ ] ] where we leave only the bogoliubov mode with the largest imaginary part @xmath39 , and define @xmath40 , @xmath41 , @xmath42 , and @xmath43 . since both @xmath44 and @xmath45 are solutions of eq . ( [ bogo ] ) , one of them satisfies @xmath46 . equation ( [ modegrow ] ) indicates that the @xmath27-fold density modulation grows exponentially in time while rotating at frequency @xmath47 . note , however , that the above discussion , which is based on bogoliubov analysis , can be applied only to an initial stage of the growth of the unstable mode , and we must integrate gp equation ( [ gpn ] ) to follow the time evolution over a longer time . we briefly describe the numerical procedure which will be used to solve eqs . ( [ gpn ] ) and ( [ bogo ] ) . the stationary state @xmath19 of eq . ( [ gpn ] ) is obtained by the newton - raphson method @xcite or the imaginary time method @xcite in the restricted functional subspace of @xmath48 . the bogoliubov spectra are then obtained by diagonalizing eq . ( [ bogo ] ) in each subspace of the angular momentum @xmath27 . time evolution of the wave function is obtained by integrating eq . ( [ gpn ] ) using the crank - nicholson scheme @xcite with a spatial discretization taken typically to be @xmath49 . in studying dynamical instabilities , we add a symmetry - breaking perturbation to the initial state such as @xmath50 where @xmath51 is an integer . while dynamically unstable modes can grow from infinitesimal numerical noise over a sufficiently long time even when we start from an axisymmetric state , we add the perturbation ( [ perturb ] ) in order to simulate experimental noise in a controllable manner so that the results are reproducible . this type of perturbation may be realized experimentally , e.g. , using off - resonant laser beams propagating along the trap axis . the system is allowed to evolve in time until just before the collapse sets in , since numerical simulations of violent collapses and explosions @xcite in 3d space present a formidable computational challenge . in this section , we study the dynamical instabilities in a singly - quantized vortex state @xmath52 . figure [ f : l1 ] shows imaginary parts of the bogoliubov spectra as functions of the interaction parameter @xmath53 for spherical ( @xmath54 ) , pancake - shaped ( @xmath55 ) , and cigar - shaped ( @xmath56 @xcite ) traps . for ( a ) @xmath54 ( spherical trap ) , ( b ) @xmath55 ( pancake - shaped ) , and ( c ) @xmath57 ( cigar - shaped ) . the dotted lines show critical values @xmath58 below which the collapse of the vortex state is caused by the monopole instability . , width=317 ] as seen from figs . [ f : l1 ] ( a ) and ( b ) , for the spherical and pancake - shaped traps , the quadrupole mode ( @xmath59 ) first becomes unstable in the sense that the imaginary part @xmath60 appears for values of @xmath61 smaller than those of other modes . a similar result is obtained in ref . @xcite for a 2d system , which corresponds to the pancake - shaped trap with a large asymmetry parameter @xmath62 @xcite . for the cigar - shaped trap [ fig . [ f : l1 ] ( c ) ] , on the other hand , the dipole mode ( @xmath63 ) first becomes unstable , which occurs for @xmath64 @xcite . this suggests that when the strength of interaction @xmath61 is adiabatically increased , the dipole instability plays a dominant role of decay mechanisms for the cigar - shaped trap with @xmath64 ; otherwise the quadrupole mode is dominant . figure [ f : panl1 ] shows the time evolution of integrated column densities and phase profiles for @xmath55 . ( lower - left insets ) for @xmath65 and @xmath55 . the initial state is the stationary singly - quantized vortex state for @xmath66 disturbed by perturbation ( [ perturb ] ) with @xmath67 . the lower - right insets show gray - scale images of the phase of the integrated wave function @xmath68 , where the phase from @xmath69 to @xmath70 is represented in gray scale from black to white . ( f ) the integrated density @xmath71 at @xmath72 , where the interaction is switched to @xmath73 and the trap is turned off at @xmath74 . the sensitivity of the imaging in ( f ) is 12 times higher than that in ( a)-(e ) . the sizes of the images are @xmath75 in @xmath71 and the phase profiles , and @xmath76 in @xmath77 in ( a)-(e ) , and @xmath78 in ( f ) in units of @xmath79 . , width=317 ] the integrated column density is defined by @xmath80 and the phase of the integrated wave function @xmath81 is used to draw the phase profiles . the main panels in fig . [ f : panl1 ] show top views @xmath71 and the lower - left insets show side views @xmath77 . the interaction parameter is taken to be @xmath65 , which exceeds the critical value for the quadrupole instability @xmath82 , but is below the critical value for the dipole instability @xmath83 . the bogoliubov eigenvalue of the quadrupole mode at @xmath65 is @xmath84 . we add the perturbation ( [ perturb ] ) with @xmath67 to the initial state to trigger the quadrupolar dynamical instability . in fig . [ f : panl1 ] ( b ) , we see that a density fluctuation in the quadrupole mode starts to grow with the lyapunov exponent @xmath85 and rotates at frequency @xmath86 in accordance with eq . ( [ modegrow ] ) . the vortex then splits into two clusters which revolve around the center of the trap [ fig . [ f : panl1 ] ( c ) ] , and subsequently these clusters unite to recover the original vortex [ fig . [ f : panl1 ] ( e ) ] . after this , the split - merge cycle repeats . figure [ f : panl1 ] ( f ) shows an expanded image of @xmath71 at @xmath72 , in which the interaction was switched from @xmath65 to @xmath73 and the trap was turned off at @xmath74 . the fringe pattern is due to interference between two matter waves emanating from the two clusters shown in fig . [ f : panl1 ] ( c ) . the above results confirm that our findings in ref . @xcite , which were obtained in a 2d system , hold true in a pancake - shaped trap and are therefore realizable experimentally . when the vortex splits into @xmath51 clusters , each of these subsequently collapses if the effective strength of interaction per each cluster , i.e. , @xmath87 , exceeds the critical value @xmath88 for the collapse of a condensate without vortices , while the split - merge cycle occurs if @xmath87 is sufficiently smaller than @xmath89 . in fact , in the case of @xmath55 in fig . [ f : panl1 ] , @xmath90 is smaller than @xmath91 , and therefore the split clusters do not collapse , allowing the system to undergo split - merge cycles . however , for the spherical trap , @xmath92 and @xmath93 , and hence the value @xmath94 at which the split occurs is always larger than @xmath89 , indicating that split clusters always collapse . we performed numerical calculations for the spherical trap with @xmath61 slightly larger than @xmath95 , and confirmed that two clusters immediately collapse without undergoing a split - merge cycle ( as shown in figs . 4 ( d)-(f ) of ref . the phenomenon , in which a vortex first splits into fragments and each fragment subsequently collapses , may be called `` fragmented collapse '' . figure [ f : mitl1 ] shows time evolution for the cigar - shaped trap with @xmath57 . ( upper panels ) , @xmath96 ( lower panels ) , and phase profiles ( insets ) for @xmath97 and @xmath57 . the initial state is the stationary singly - quantized vortex state for @xmath98 disturbed by perturbation ( [ perturb ] ) with @xmath99 . the sizes of the images are @xmath75 for @xmath71 and @xmath100 for @xmath77 in units of @xmath79 . , width=317 ] the interaction parameter is taken to be @xmath97 , at which only the dipole mode is unstable [ see fig . [ f : l1 ] ( c ) ] . the initial state is the stationary state for @xmath97 plus the small perturbation ( [ perturb ] ) with @xmath99 . we see that the atoms tend towards one side of the ring due to the dipole instability [ fig . [ f : mitl1 ] ( b ) ] , after which collapse occurs [ fig . [ f : mitl1 ] ( c ) ] . since the dipole mode has angular momenta @xmath101 and @xmath102 , the phase defect that exists from the outset slightly deviates from the center associated with the growth of the @xmath103 component , and another phase defect enters due to the @xmath104 component as seen in the inset of fig . [ f : mitl1 ] ( c ) . for @xmath57 , the critical value at which the dipole mode becomes unstable is @xmath105 , which exceeds @xmath106 , and therefore the dipole instability is always followed by collapse . we note that the dipole instability in fig . [ f : mitl1 ] is distinct from the dissipative vortex spiral - out phenomenon that arises from the dipole mode with negative eigenvalue @xcite . in the former , the quasiparticles that change angular momentum by @xmath107 are excited in pairs , and hence the angular momentum is conserved . the energy is also conserved , since the real parts of the excitation energies have the same magnitude with opposite signs . in the latter , the atoms are transferred only into the non - vortex ground state decreasing the angular momentum and dissipating the energy . a doubly - quantized vortex state @xmath108 of @xmath109 bec has recently been realized by the mit group @xcite using a topological phase - imprinting method @xcite , where the interaction is repulsive . the doubly - quantized vortex with repulsive interactions is predicted to be dynamically unstable against disintegration into two singly - quantized vortices @xcite . in the case of attractive interactions , we will show that not only the vortex disintegration but also various new phenomena , such as a three - fold split - merge process , occur due to dynamical instabilities . the imaginary parts of the bogoliubov eigenvalues for the doubly - quantized vortex states are shown in fig . [ f : l2 ] . for ( a ) @xmath54 , ( b ) @xmath110 , and ( c ) @xmath57 . the dotted lines show the critical value @xmath58 below which the collapse of the vortex state is caused by the monopole instability . , width=317 ] we note that unlike the case of the singly - quantized vortex state , the quadrupole mode ( @xmath59 ) is always dynamically unstable for @xmath111 , indicating that no dynamically stable state exists for the case of attractive interactions . therefore , an adiabatic decrease in @xmath53 from positive to negative always gives rise to the quadrupole instability . figure [ f : sphl2 ] shows time evolutions of the integrated densities @xmath71 ( main panels ) and @xmath112 ( lower - left insets ) and the phase profiles ( lower - right insets ) for the doubly - quantized vortex state in a spherical trap , where the initial state is a stationary state of the gp equation with @xmath97 plus the perturbation given in eq . ( [ perturb ] ) with @xmath67 . ( main panels ) , @xmath113 ( lower - left insets ) , and phase profiles ( lower - right insets ) for @xmath54 . the initial state is the stationary doubly - quantized vortex state for @xmath114 disturbed by perturbation ( [ perturb ] ) with @xmath67 , and the interaction is switched to @xmath97 at @xmath115 . the sizes of the images are @xmath75 for @xmath71 and the phase profiles , and @xmath116 for @xmath112 in units of @xmath117 . , width=317 ] for this strength of interaction , only the quadrupole mode is dynamically unstable [ cf . [ f : l2 ] ( a ) ] . the vortex core disintegrates , as can be seen from the density and phase profiles . the angular momenta of the quadrupole excitations are @xmath118 and @xmath102 . vortex disintegration is attributed to the growth of the @xmath103 component , and the four phase defects in fig . [ f : sphl2 ] ( d ) are due to the @xmath119 component . the density distribution exhibits various changes such as high - density regions occurring in the opposite edge of the vortices [ fig . [ f : sphl2 ] ( c ) ] , then between the two vortices [ fig . [ f : sphl2 ] ( d ) ] , and then the density distribution becomes similar to the split vortex as in the case of a singly - quantized vortex [ fig . [ f : sphl2 ] ( e ) ] . finally , in fig . [ f : sphl2 ] ( f ) the central density exceeds critical density and collapses . another important distinction of fig . [ f : l2 ] from fig . [ f : l1 ] is that there is a range of @xmath53 over which the imaginary part of the eigenvalue of the octupole mode ( @xmath120 ) becomes largest . this implies that if @xmath53 is suddenly changed into that regime , the octupolar instability dominates until the quadrupole mode grows . figure [ f : panl2 ] shows a typical example of such situations with @xmath55 , where the initial state is the noninteracting stationary state with vorticity of two subject to perturbation ( [ perturb ] ) with @xmath121 , and the interaction parameter is changed from @xmath122 to @xmath123 at @xmath115 . ( main panels ) , @xmath96 ( lower - left insets ) , and phase profiles ( lower - right insets ) for the pancake - shaped trap with @xmath55 . the initial state is the stationary doubly - quantized vortex state for @xmath114 disturbed by perturbation ( [ perturb ] ) with @xmath121 , and the interaction is switched to @xmath124 at @xmath115 . ( i ) the integrated density @xmath71 at @xmath125 , where the interaction is switched to @xmath73 and the trap is turned off at @xmath126 . the sensitivity of the imaging is 12 times higher than that in ( a)-(h ) . the sizes of the images are @xmath75 in @xmath71 and the phase profiles , and @xmath76 in @xmath77 in ( a)-(h ) , and @xmath78 in ( i ) in units of @xmath79 . , width=317 ] as the octupole mode grows , the vortex splits into three clusters which revolve around the center of the trap [ fig . [ f : panl2 ] ( c ) ] . interestingly , the three clusters merge to recover the original vortex [ fig . [ f : panl2 ] ( d ) ] as in the two - cluster case shown in fig . [ f : panl1 ] . this is because the effective strength of interaction per each cluster @xmath127 is smaller than the critical value for collapse of the non - vortex state @xmath128 . subsequently , the vortex once again splits into three clusters , and the balance in atomic numbers between these is then broken by the quadrupole instability . it follows that one cluster [ denoted by a in fig . [ f : panl2 ] ( f ) ] grows , then another cluster [ b in fig . [ f : panl2 ] ( g ) ] grows , and finally the remaining cluster [ c in fig . [ f : panl2 ] ( h ) ] grows and eventually collapses . this phenomenon is similar to the seesaw - like oscillation between two clusters due to the dipole instability @xcite . at the initial stage of disintegration in which the octupole mode plays a dominant role [ figs . [ f : panl2 ] ( a)-(e ) ] , the split in the phase defects is not very prominent because the angular momenta of the @xmath129 modes are @xmath130 and @xmath131 . when the quadrupole mode becomes significant , the outward shifts of the phase defects manifest themselves [ figs . [ f : panl2 ] ( g)-(h ) ] due to the @xmath103 component of the quadrupole mode . figure [ f : panl2 ] ( i ) shows the integrated density at @xmath125 , following the interaction being switched from @xmath124 to @xmath73 and the trap being turned off at @xmath126 . we see the three - fold fringe pattern resulting from interference between the matter waves that emanate from the three clusters shown in fig . [ f : panl2 ] ( c ) . in an elongated cigar - shaped trap , dynamics along the trap axis also plays an important role . figure [ f : mitl2 ] shows the integrated density and phase profiles for @xmath57 . ( upper panels ) , phase profiles ( insets ) , @xmath96 ( lower - left panels ) , and @xmath132 ( lower - right panels ) for the cigar - shaped trap with @xmath57 . the initial state is the stationary doubly - quantized vortex state for @xmath114 disturbed by perturbation ( [ perturb ] ) with @xmath67 , and the interaction is switched to @xmath133 at @xmath115 . the sizes of the images are @xmath75 for @xmath71 and the phase profiles , and @xmath134 for @xmath77 and @xmath112 in units of @xmath135 . , width=317 ] the initial state is the noninteracting stationary state with vorticity of two plus perturbation ( [ perturb ] ) with @xmath67 , and the interaction is switched from @xmath122 to @xmath133 at @xmath115 . the atoms first gather along the symmetry axis ( i.e. , the @xmath136 axis ) towards @xmath137 [ fig . [ f : mitl2](b ) ] , then the vortex begins to disintegrate since the density is high enough . subsequently the atoms rebound along the @xmath136 direction [ fig . [ f : mitl2](c ) ] . we see that the disintegrated vortex lines in fig . [ f : mitl2 ] ( c ) are straight , unlike the repulsive case where twisting of vortex lines occurs @xcite . in fig . [ f : mitl2 ] ( d ) , the condensate splits into two high - density regions containing the disintegrated vortices as a consequence of the rebound . in an experiment using @xmath1 , the lifetime of the condensate is as short as 1 or 2 ms near the feshbach resonance @xcite , which indicates that the lifetime is @xmath138 ( in our dimensionless units of time ) for the optical dipole trap used in the mit group ( @xmath139 hz , @xmath140 hz , and @xmath56 ) @xcite . for dynamical instabilities to occur within this short lifetime , the strength of attractive interaction @xmath61 must be much larger than those in the above cases . figure [ f : mitl2exp ] shows the density and phase profiles just before the collapse , where the initial state is the noninteracting stationary state with vorticity of two plus perturbation ( [ perturb ] ) with @xmath141 , @xmath30 , @xmath31 , and @xmath142 , and the interaction is switched from @xmath122 to @xmath143 at @xmath115 . , phase profiles , and @xmath132 ( from top to bottom ) for the cigar - shaped trap with @xmath57 . the initial state is the stationary doubly - quantized vortex state for @xmath114 disturbed by perturbation ( [ perturb ] ) with ( a ) @xmath99 , ( b ) @xmath67 , ( c ) @xmath121 , and ( d ) @xmath144 . the interaction is switched from @xmath122 to @xmath145 at @xmath115 . the sizes of the images are @xmath75 for @xmath71 and the phase profiles , and @xmath146 for @xmath112 in units of @xmath117 . , width=317 ] we find that the collapsing dynamics depend on the initial perturbation @xmath51 , since the system has several multipole instabilities for @xmath145 . we note that the vortex split occurs only around @xmath137 , where the density is high enough . thus , the vortex - split phenomena can be observed for large values of @xmath61 , even when the lifetime of the condensate is very short . to understand the split - merge phenomena shown in the preceding sections , we restrict ourselves to 2d space in this and next sections . we employ a gaussian variational wave function for the condensate as @xmath147 where @xmath26 is the vorticity and @xmath148 is a variational parameter characterizing the size of the condensate . substituting eq . ( [ gaussian0 ] ) into the gp energy functional @xmath149 = \int d{\bf r } \left [ \frac{1}{2 } \psi^ * \left ( -\nabla^2 + \omega^2 r^2 \right ) \psi + \frac{g}{2 } |\psi|^4 \right]\ ] ] and minimizing it with respect to @xmath148 gives @xmath150^{\frac{1}{4}}.\ ] ] for @xmath151 for @xmath152 and @xmath153 for @xmath154 agree reasonably well with the numerically obtained critical values @xmath155 and @xmath156 , respectively . to study the time evolution from eq . ( [ gaussian0 ] ) , we use a variational wave function given by @xmath157 where @xmath158 and @xmath159 is given by eq . ( [ gaussian0 ] ) with @xmath26 replaced by @xmath4 . in eq . ( [ varpsi ] ) we use the same width @xmath148 as eq . ( [ d ] ) for all @xmath159 and neglect the radial degrees of freedom to simplify the following calculation and to obtain the results analytically . substituting eq . ( [ varpsi ] ) into the action @xmath160 = \int dt e[\psi ] - i \int d{\bf r } dt \psi^ * \frac{\partial}{\partial t } \psi\ ] ] yields @xmath161 where @xmath162 , and @xmath163 , which vanishes for @xmath164 . from @xmath165 , the variational equation of motion for @xmath166 becomes @xmath167 we assume @xmath168 and @xmath169 , and neglect terms of the order of @xmath170 . equation ( [ eomc ] ) for @xmath171 reduces to @xmath172 and hence we obtain @xmath173 . using eq . ( [ c1 ] ) and defining @xmath174 , we obtain a closed set of equations @xmath175 from which the eigenfrequencies can be found . the condition under which the eigenfrequencies are real is given by @xmath176 ^ 2 \nonumber \\ & & - 4g^2 \left ( g^{m , 2q - m}_{q , q } \right)^2 \geq 0.\end{aligned}\ ] ] first we consider the singly - quantized vortex ( @xmath152 ) . stability of the quadrupole mode is determined by @xmath177 which becomes negative for @xmath178 . this value is in reasonable agreement with the numerically obtained critical value for the quadrupole mode in 2d , @xmath179 . however , the discriminant @xmath180 is always positive , which contradicts the numerical result that dipole instability does occur for @xmath181 . also , for @xmath182 or @xmath119 , the numerical calculation shows that dynamical instability occurs for @xmath183 , while @xmath184 is always positive for @xmath185 [ @xmath53 must be larger than @xmath186 as seen from eq . ( [ d ] ) ] . in the case of the doubly - quantized vortex ( @xmath154 ) , @xmath187 is always negative , which is consistent with the fact that the quadrupole mode is always dynamically unstable for @xmath188 . however , @xmath189 fails to reproduce the behavior for @xmath190 , where stable and unstable regions appear alternatively @xcite . the discriminant @xmath191 becomes negative for @xmath192 , which is in reasonable agreement with the numerical result that the octupole instability sets in for @xmath193 . however , @xmath194 contradicts the numerical result that dipole instability does occur for @xmath195 . thus , the gaussian ansatz ( [ varpsi ] ) is viable for some cases , whereas more sophisticated trial functions are needed to completely describe the dynamical instabilities . the variational results can be improved by using the different variational width @xmath148 for each @xmath159 in eq . ( [ varpsi ] ) , which may be determined by , e.g. , the method used in ref . @xcite . however , near and above the critical point , this variational method fails , since the norm of the bogoliubov function vanishes when a complex eigenvalue emerges . equation ( [ d13 ] ) is expanded near the critical point as @xmath196 where @xmath197 . just above the critical point , therefore , the imaginary part of the complex eigenvalue is proportional to @xmath198 @xcite . all the imaginary parts shown in figs . [ f : l1 ] and [ f : l2 ] ( except @xmath59 in fig . [ f : l2 ] ) have this @xmath53 dependence near the critical points [ so do the @xmath63 curves in figs . [ f : l1 ] ( c ) and [ f : l2 ] ( c ) in the immediate vicinity of the critical points ] . the @xmath59 lines for @xmath154 in fig . [ f : l2 ] are proportional to @xmath61 near @xmath122 , which agrees with @xmath199 . the split - merge cycles shown in figs . [ f : panl1 ] and [ f : panl2 ] are recurrence phenomena arising from nonlinearity . to understand the behavior in fig . [ f : panl1 ] qualitatively , we reduce the number of modes in eq . ( [ eomc ] ) to three , i.e. , @xmath200 , @xmath31 , and @xmath131 . equation ( [ eomc ] ) then reduces to simultaneous nonlinear differential equations with three variables , which can be solved numerically . figure [ f:3mode ] shows the time evolution of @xmath201 , @xmath202 , and @xmath203 for @xmath204 , where the initial condition is @xmath205 and @xmath206 . ( solid curve ) , @xmath207 ( dashed curve ) , and @xmath208 ( dotted curve ) in the three - mode approximation of eq . ( [ eomc ] ) for @xmath204 . the initial condition is @xmath209 and @xmath210 . the insets show the density and phase profiles at @xmath115 and @xmath211 , where the sizes of the images are @xmath75 in units of @xmath212 . , width=317 ] the vortex can be seen to split at @xmath213 , but subsequently recovers the original vortex during @xmath214 , and then splits again at @xmath215 . thus , the split - merge phenomenon can be qualitatively reproduced by a simple three - mode approximation . the split - merge dynamics shown in fig . [ f : panl1 ] implies the existence of a low - lying state in which two clusters revolve around each other without changing their shapes . to find such a low - lying stationary state , we numerically seek a local minimum of the gp energy functional with the angular momentum held fixed at that of the initial vortex state @xmath216 , where @xmath217 . the gp energy functional with lagrange multipliers is given by @xmath218 = e[\psi ] + \int d{\bf r } \left ( -\mu |\psi|^2 + \omega \psi^ * \hat l_z \psi \right),\ ] ] where @xmath219 $ ] is given in eq . ( [ e ] ) , and the chemical potential @xmath22 and the angular frequency @xmath220 of rotation of the system are introduced as the lagrange multipliers in order to fix the normalization and the angular momentum . using the newton - raphson method @xcite starting from appropriate initial states , we obtain the stationary states in the frame of reference rotating at frequency @xmath220 . figure [ f : lowly ] shows various properties of the axisymmetric vortex state and the low - lying state . ( solid curves ) and chemical potential @xmath22 ( dashed curves ) of the axisymmetric singly - quantized vortex state ( upper branch ) and low - lying soliton state ( lower branch for @xmath221 ) . gray - scale images show the density profile of the axisymmetric vortex state for @xmath222 and those of the low - lying states for @xmath223 and @xmath224 . ( b ) the bogoliubov excitation energies of the quadrupole modes above the axisymmetric vortex state ( @xmath225 ) and those of the corresponding modes above the low - lying soliton state ( @xmath221 ) . ( c ) the number of virtually excited atoms in the bogoliubov modes . the inset is the logarithmic plot of the excitation energies and the number of excited atoms ( both of them refer to the common scale ) as functions of @xmath226 . the vertical dotted lines in ( a)-(c ) show @xmath227 . , width=264 ] when @xmath61 is increased , the stationary state bifurcates at the critical point @xmath228 of the quadrupole instability , at which the low - lying state emerges as represented by the lower branches of @xmath229 and @xmath22 in fig . [ f : lowly ] ( a ) . density profiles of the low - lying states for @xmath223 and @xmath124 are shown as gray - scale images in fig . [ f : lowly ] ( a ) . it is clear from them that the split - merge cycle is due to oscillations between the axisymmetric vortex state ( upper branch ) and the low - lying soliton state ( lower branch ) . it is interesting to note that the energies of the axisymmetric and low - lying states are smoothly connected at the critical point , while the chemical potential has a kink , which implies that the nature of the phase transition between the vortex state and the soliton state is of second order . the excitation energies are plotted in fig . [ f : lowly ] ( b ) , where @xmath230 denotes the energy of the lowest excitation with angular momentum @xmath231 above the axisymmetric vortex state . as @xmath53 approaches @xmath228 in the axisymmetric regime , the excitation energy @xmath232 vanishes , despite the fact that both @xmath233 and @xmath233 are nonzero at @xmath234 . this implies that quasiparticles with @xmath235 can be excited in pairs without changing energy and angular momentum . hence , the two modes @xmath235 may be strongly coupled via the pair excitations . if we follow the axisymmetric branch [ the upper branch in fig . [ f : lowly ] ( a ) ] , the eigenenergies of the two modes become complex . if we follow the low - lying soliton branch , one of the two modes becomes the goldstone mode associated with the axisymmetry breaking , and the energy of the other mode @xmath236 increases with decreasing @xmath53 . figure [ f : lowly ] ( c ) shows the number of virtually excited atoms @xmath237 in the quadrupole modes @xmath238 for @xmath239 and that in the corresponding mode @xmath240 for @xmath241 . at @xmath234 the number of excited atoms diverges , indicating that the bogoliubov approximation breaks down . the inset of fig . [ f : lowly ] ( c ) shows the logarithmic plot of the excitation energies and the numbers of excited atoms . close to the critical point @xmath242 , the excitation energies are found to be proportional to @xmath243 and the numbers of excited atoms to @xmath244 . we note that these behaviors near the critical point ( the kink in @xmath22 , the softening of the excitation mode , the divergence in the number of excited atoms , and their power laws ) are very similar to those occurring in the transition between the uniform - density and soliton states in a 1d ring @xcite . thus , in the immediate vicinity of the critical point at which the vortex split occurs , the mean - field and bogoliubov approximations break down due to large quantum fluctuations , and we must employ , e.g. , the exact diagonalization method @xcite to study this regime . however , the mean - field theory used in the present paper is still valid except for this narrow critical region . we studied dynamical instabilities in singly- and doubly - quantized vortex states with attractive interactions , and found a rich variety of dynamics triggered by the dynamical instabilities . the dynamics depend on the strength of interaction @xmath53 , the trap geometry @xmath62 , the initial perturbations , and the manner ( adiabatic or sudden ) in which @xmath53 changes . a singly - quantized vortex in a trap with @xmath245 first exhibits the quadrupole instability as @xmath53 is adiabatically decreased . the vortex then splits into two clusters revolving around the center of the trap . in a spherical trap the split fragments immediately collapse for any value of @xmath53 that gives rise to the split instability , whereas in a pancake - shaped trap with @xmath55 they unite to recover the original vortex and then the split - merge cycles repeat for @xmath61 slightly exceeding @xmath95 . in a cigar - shaped trap with @xmath64 , the dipole instability causes the atoms to concentrate in one cluster and then the system collapses . the doubly - quantized vortex state is always dynamically unstable against disintegration of the vortex core for @xmath188 , and therefore the disintegration is unavoidable if we adiabatically decrease @xmath53 . if @xmath53 is suddenly changed to some negative value with an appropriate initial perturbation to the condensate , the octupole instability can dominate the quadrupole instability , and the vortex splits into three clusters . in a pancake - shaped trap , these undergo split - merge cycles as in the case of a singly - quantized vortex , but eventually one of them collapses due to the quadrupole instability that always exists . we have also demonstrated that the split phenomena can be observed using the current experimental setup of the mit group . to understand the dynamical instability and the split - merge cycles , we performed a variational analysis . the emergence of a complex eigenvalue in some modes was reproduced by the gaussian trial functions . we were able to reproduce the split - merge cycles by assuming only three relevant modes with angular momenta @xmath246 and @xmath247 , suggesting that only a few modes are associated with the split - merge cycles . in the immediate vicinity of the critical strength of interaction of the dynamical instability , the transition from the axisymmetric vortex state to the split state is accompanied by divergence in the number of virtually excited atoms and by softening of the excitation mode according to the bogoliubov theory , which is very similar to the 1d case discussed in ref . the dynamical instabilities in vortex states of attractive becs therefore involve a fundamental generic issue of the decay of a many - particle quantum system . vortex - split phenomena are attributed to the interplay between the attractive interaction and the topological phase defect , that is , the condensate is bound to break axisymmetry , since they can not gather at the phase defect . the emergence of the dynamical instabilities and ensuing dynamics may be regarded as pattern formation due to symmetry breaking . more complicated and interesting pattern might be formed in the vortex lattice , which will be published elsewhere . dynamics after the collapse were not studied because 3d simulation of the collapse and explosion phenomena exceeds our present computational ability . the burst atoms ejected from each cluster interfere with each other and exhibit some anisotropic pattern if the burst atomic cloud is coherent . therefore experiments on vortex split and fragmented collapse will provide a test of whether or not the burst atoms are coherent . we thank j. k. chin for discussions . this work was supported by the special coordination funds for promoting science and technology and a grant - in - aid for scientific research ( grant no . 15340129 ) by the ministry of education , science , sports , and culture of japan , and by the yamada science foundation .

many solids have a crystalline structure . however , ideal crystals are absent in nature , and most of their physical properties , such as plasticity , melting , growth , etc . , are defined by defects of the crystalline structure . therefore , a study of defects is a topical scientific question of importance for applications in the first place . a broad experimental and theoretical investigations of defects in crystals started in the 1930s and continues nowadays . at present , a fundamental theory of defects is absent in spite of the existence of dozens of monographs and thousands of articles . one of the most promising approaches to the theory of defects is based on riemann cartan geometry , which involves nontrivial metric and torsion . in this approach , a crystal is considered as a continuous elastic medium with a spin structure . if the displacement vector field is a smooth function , then there are only elastic stresses corresponding to diffeomorphisms of the euclidean space . if the displacement vector field has discontinuities , then we are saying that there are defects in the elastic structure . defects in the elastic structure are called dislocations and lead to the appearance of nontrivial geometry . precisely , they correspond to a nonzero torsion tensor , equal to the surface density of the burgers vector . the idea to relate torsion to dislocations appeared in the 1950s [ 14 ] . this approach is still being successfully developed ( note reviews [ 511 ] ) , and is often called the gauge theory of dislocations . a similar approach is also being developed in gravity @xcite . it is interesting to note that e cartan introduced torsion in geometry @xcite having the analogy with mechanics of elastic media in mind . the gauge approach to the theory of defects is being developed successfully , and interesting results are being obtained in this way [ 1417 ] . we note in this connection two respects in which the approach proposed below is essentially different . in the gauge models of dislocations based on the translational group or on the semidirect product of the rotational group with translations , one usually chooses the distortion and displacement fields as independent variables . it is always possible to fix the invariance under local translations such that the displacement field becomes zero because it transforms by simple translation under the action of the translational group . in this sense , the displacement field is the gauge parameter of local translations , and physical observables are independent of it in gauge - invariant models . the other disadvantage of the gauge approach is the equations of equilibrium . einstein type equations are usually considered for distortion or vielbein , with the right hand side depending on the stress tensor . this appears unacceptable from the physical point of view because of the following reason . consider , for example , one straight edge dislocation . in this case , the elastic stress field differs from zero everywhere . then the torsion tensor ( or curvature ) is also nontrivial in the whole space due to the equations of equilibrium . this is wrong from our point of view . indeed , we can consider an arbitrary domain of medium outside the cutting surface and look at the creation process for an edge dislocation . the chosen domain was a part of the euclidean space with identically zero torsion and curvature before the defect creation . it is clear that torsion and curvature remain zero because the process of dislocation formation is a diffeomorphism for the considered domain . in addition , the cutting surface may be chosen arbitrary for the defect creation , leaving the dislocation axis unchanged . then it follows that torsion and curvature must be zero everywhere except at the axis of dislocation . in other words , the elasticity stress tensor can not be the source of dislocations . to avoid the apparent contradiction , we propose a radical way out : we do not use the displacement field as an independent variable at all . this does not mean that the displacement field does not exist in real crystals . in the proposed approach , the displacement field exists and can be computed in those regions of medium that do not contain cores of dislocations . in this case , it satisfies the equations of nonlinear elasticity theory . the proposed geometric approach allows considering other defects that do not relate directly to defects of elastic media . the intensive investigations of other defects were conducted in parallel with the study of dislocations . the point is that many solids have not only elastic properties but also a spin structure . for example , there are ferromagnets , liquid crystals , spin glasses . in this case , there are defects in the spin structure which are called disclinations @xcite . they arise when the director field has discontinuities . the presence of disclinations is also connected to nontrivial geometry . namely , the curvature tensor equals the surface density of the frank vector . the gauge approach based on the rotational group @xmath0 was also used for describing disclinations @xcite . @xmath0-gauge models of spin glasses with defects were considered in @xcite . the geometric theory of static distribution of defects which describes both types of defects dislocation and disclinations from a single standpoint was proposed in @xcite . in contrast to other approaches , it involves the vielbein and @xmath0 connection as the only independent variables . the torsion and curvature tensors have direct physical meaning as the surface densities of dislocations and disclinations , respectively . covariant equations of equilibrium for the vielbein and @xmath0 connection similar to those in a gravity model with torsion are postulated . to define the solution uniquely , we must fix the coordinate system ( fix the gauge ) because any solution of the equations of equilibrium is defined up to general coordinate transformations and local @xmath0 rotations . the elastic gauge for the vielbein @xcite and lorentz gauge for the @xmath0 connection @xcite were proposed recently . we stress that the notions of a displacement vector and rotational angle are completely absent in our approach . these notions can be introduced only in those domains where defects are absent . in this case , equations for vielbein and @xmath0 connection are identically satisfied , the elastic gauge reduces to the equations of nonlinear elasticity theory for the displacement vector , and the lorentz gauge leads to the equations for the principal chiral @xmath0 field . in other words , to fix the coordinate system , we choose two fundamental models : the elasticity theory and the principal chiral field model . to show the advantages of the geometric approach and to compare it with the elasticity theory , we consider in detail a wedge dislocation in the frameworks of the elasticity theory and the proposed geometric model . we show that the explicit expression for the metric in the geometric approach is simpler and coincides with the induced metric obtained within the elasticity theory only for small relative deformations . as an application of the geometric theory of defects , we consider two examples in the last sections of the present review . first , we solve the problem of phonon scattering on a wedge dislocation . the problem of phonon scattering is reduced to the integration of equations for extremals for the metric describing a wedge dislocation ( because phonons move along extremals in the eikonal approximation ) . as a second application , we consider the quantum mechanical problem of impurity or vacancy motion inside a cylinder whose axis coincides with a wedge dislocation . the wave functions and energetic spectrum of the impurity are found explicitly . the presence of defects results in nontrivial riemann cartan geometry . this means that for describing the phenomena that relate ingeniously to elastic media , we must make changes in the corresponding equations . for example , if a phonon propagation in an ideal crystal is described by the wave equation , then the presence of defects is easily taken into account . for this , the flat euclidean metric has to be replaced by a nontrivial metric describing the distribution of defects . the same substitution must be made in the schrdinger equation to describe other quantum effects . it is shown nowadays that the presence of defects essentially influences physical phenomena . the schrdinger equation in the presence of dislocations was considered in [ 2546 ] for different problems . problems related to the wave or laplace equations were considered in [ 4754 ] . the influence of the nontrivial metric related to the presence of defects was investigated in electrodynamics @xcite and hydrodynamics @xcite . scattering of phonons on straight parallel dislocations was studied in [ 5759 ] . another approach to the theory of defects based on affine geometry with nonzero nonmetricity tensor was considered recently in @xcite . we consider infinite three dimensional elastic media . we suppose that the undeformed medium in the defect - free case is invariant under translations and rotations in some coordinate system . then , the medium in this coordinate system @xmath1 , @xmath2 , is described by the euclidean metric @xmath3 , and the system of coordinates is called cartesian . thus , in the undeformed state , we have the euclidean space @xmath4 with a given cartesian coordinate system . we also assume that torsion ( see the appendix ) in the medium equals zero . let a point of the medium has coordinates @xmath1 in the ground state . after deformation , this point has the coordinates @xmath5 .45 in the initial coordinate system , see fig . [ feldef ] . the inverse notation is used in the elasticity theory . one usually writes @xmath6 . these are equivalent because both coordinate systems @xmath7 and @xmath1 cover the whole @xmath4 . however , in the theory of defects considered in the next sections , the situation is different . generally , the elastic medium fills the whole euclidean space only in the final state . here and in what follows , we assume that fields depend on coordinates @xmath8 that are coordinates of points of the medium after the deformation and cover the whole euclidean space @xmath4 . in the presence of dislocations , the coordinates @xmath1 do not cover the whole @xmath4 in the general case because part of the media may be removed or , conversely , added . therefore , the system of coordinates related to points of the medium after an elastic deformation and defect creation is more preferable . in the linear elasticity theory , relative deformations are assumed to be small ( @xmath9 ) . the functions @xmath10 are then components of a vector field that is called the displacement vector field and is the basic variable in elasticity theory . in the absence of defects , we assume that the displacement field is a smooth vector field in the euclidean space @xmath4 . the presence of discontinuities and singularities of the displacement field is interpreted as a presence of defects in elastic media . in what follows , we consider only static deformations with the displacement field @xmath11 independent of time . then the basic equations of equilibrium for small deformations are ( see , e.g. , @xcite ) @xmath12 where @xmath13 is the stress tensor , which is assumed to be symmetric . the tensor of small deformations @xmath14 is given by the symmetrized partial derivative of the displacement vector : @xmath15 lowering and raising of the latin indices is performed with the euclidean metric @xmath16 and its inverse @xmath17 . the letters @xmath18 and @xmath19 denote constants characterizing elastic properties of media and are called lam coefficients . functions @xmath20 describe the total density of nonelastic forces inside the medium . we assume in what follows that such forces are absent : @xmath21 . equation ( [ estati ] ) is newton s law , and eqn ( [ ehook ] ) is hook s law relating stresses to deformations . in a cartesian coordinate system and for small deformations , the difference between upper and lower indices disappears because raising and lowering of indices is performed with the help of the euclidean metric . one usually forgets about this difference due to this reason , and this is fully justified . but in the presence of defects , the notion of cartesian coordinate system and euclidean metric is absent , and the indices are raised and lowered with the help of riemannian metric . therefore , we distinguish the upper and lower indices as is accepted in differential geometry , having the transition to elastic media with defects in mind . the main problem in the linear elasticity theory is the solution of the second - order equations for the displacement vector that arise after substitution of ( [ ehook ] ) into ( [ estati ] ) with some boundary conditions . many known solutions are in good agreement with experiment . therefore , one may say that equations ( [ ehook ] ) , ( [ estati ] ) have a solid experimental background . we now look at the elastic deformations from the standpoint of differential geometry . from the mathematical standpoint , map ( [ eeldef ] ) is itself a diffeomorphism of the euclidean space @xmath4 . the euclidean metric @xmath16 is then induced by the map @xmath22 . it means that in the deformed state , the metric in the linear approximation is given by @xmath23 i.e. , is defined by the tensor of small deformations ( [ edefte ] ) . we note that in the linear approximation , @xmath24 and @xmath25 . in riemannian geometry , the metric uniquely defines the levi civita connection @xmath26 ( christoffel s symbols ) , eqn ( [ etigam ] ) . we can compute curvature tensor ( [ ecurva ] ) for these symbols . this tensor equals identically zero , @xmath27 , because the curvature of the euclidean space is zero , and the map @xmath22 is a diffeomorphism . the torsion tensor is equal to zero for the same reason . thus , an elastic deformation of the medium corresponds to the trivial riemann cartan geometry , with zero curvature and torsion tensors . the physical interpretation of metric ( [ emetri ] ) is as follows . the external observer fixes cartesian coordinate system corresponding to the ground undeformed state of the medium . the medium is then deformed , and external observer discovers that the metric becomes nontrivial in this coordinate system . if we assume that elastic perturbations in the medium ( phonons ) propagate along extremals ( lines of minimal length ) , then their trajectories in the deformed medium are defined by eqns ( [ exteqr ] ) . trajectories of phonons are now not straight lines because the christoffel s symbols are nontrivial ( @xmath28 ) . in this sense , metric ( [ emetri ] ) is observable . here , we see the essential role of the cartesian coordinate system @xmath1 defined by the undeformed state , with which the measurement process is connected . we assume that the metric @xmath29 given in the cartesian coordinates corresponds to some state of elastic media without defects . the displacement vector is then defined by the system of equations ( [ emetri ] ) , and its integrability conditions are the equality of the curvature tensor to zero , in accordance with theorem [ tlocev ] in the appendix . in the linear approximation , these conditions are known in elasticity theory as the saint venant integrability conditions . we make a remark that is important for the following consideration . for appropriate boundary conditions , the solution of the elasticity theory equations ( [ estati ] ) , ( [ ehook ] ) is unique . from the geometric standpoint , this means that elasticity theory fixes diffeomorphisms . this fact is used in the geometric theory of defects . equations of nonlinear elasticity theory written in terms of the metric or vielbein are used for fixing the coordinate system . we start with the description of linear dislocations in elastic media ( see , e.g. , @xcite ) . the simplest and most undespread examples of linear dislocations are shown in fig . [ fdislo ] . we cut the medium along the half - plane @xmath30 , @xmath31 , move the upper part of the medium located over the cut @xmath32 , @xmath31 by the vector @xmath33 towards the dislocation axis @xmath34 , and glue the cutting surfaces . the vector @xmath33 is called the burgers vector . in the general case , the burgers vector may not be constant on the cut . for the edge dislocation , it varies from zero to some constant value @xmath33 as it moves from the dislocation axis . after the gluing , the media comes to the equilibrium state called the edge dislocation , see fig . [ fdislo]_a_. if the burgers vector is parallel to the dislocation line , it is called the screw dislocation ( fig . [ fdislo]_b _ ) . the same dislocation can be made in different ways . for example , if the burgers vector is perpendicular to the cutting plane and directed from it in the considered cases , then the produced cavity must be filled with medium before gluing . it is easy to imagine that the edge dislocation is also obtained as a result , but rotated by the angle @xmath35 around the @xmath34 axis . this example shows that a dislocation is characterized not by the cutting surface but by the dislocation line and the burgers vector . straight linear dislocations . ( _ a _ ) the edge dislocation . the burgers vector @xmath33 is perpendicular to the dislocation line . ( _ b _ ) the screw dislocation . the burgers vector @xmath33 is parallel to the dislocation line.,title="fig : " ] straight linear dislocations . ( _ a _ ) the edge dislocation . the burgers vector @xmath33 is perpendicular to the dislocation line . ( _ b _ ) the screw dislocation . the burgers vector @xmath33 is parallel to the dislocation line.,title="fig : " ] + from the topological standpoint , the medium containing several dislocations or even the infinite number of them is the euclidean space @xmath4 . in contrast to the case of elastic deformations , the displacement vector in the presence of dislocations is no longer a smooth function because of the presence of cutting surfaces . at the same time , we assume that partial derivatives of the displacement vector @xmath36 ( the distortion tensor ) are smooth functions on the cutting surface . this assumption is justified physically because these derivatives define deformation tensor ( [ edefte ] ) . in its turn , partial derivatives of the deformation tensor must exist and be smooth functions in the equilibrium state everywhere except , possibly , the dislocation axis , because otherwise equations of equilibrium ( [ estati ] ) have no meaning . we assume that the metric and vielbein are smooth functions everywhere in @xmath4 except , may be , dislocation axes , because the deformation tensor defines the induced metric ( [ emetri ] ) . the main idea of the geometric approach amounts to the following . to describe single dislocations in the framework of elasticity theory we must solve equations for the displacement vector with some boundary conditions on the cuts . this is possible for small number of dislocations . but , with an increasing number of dislocations , the boundary conditions become so complicated that the solution of the problem becomes unrealistic . besides , one and the same dislocation can be created by different cuts which leads to an ambiguity in the displacement vector field . another shortcoming of this approach is that it can not be applied to the description of a continuous distribution of dislocations because the displacement vector field does not exist in this case at all because it must have discontinuities at every point . in the geometric approach , the basic variable is the vielbein which by assumption is a smooth function everywhere except , possibly , dislocation axes . we postulate new equations for the vielbein ( see section [ sequil ] ) . in the geometric approach , the transition from a finite number of dislocations to their continuous distribution is simple and natural . in that way , the smoothing of singularities occurs on dislocation axes in analogy with smoothing of mass distribution for point particles in passing to continuous media . we now develop the formalism of the geometric approach . in a general defect - present case , we do not have a preferred cartesian coordinate frame in the equilibrium because there is no symmetry . therefore , we consider arbitrary coordinates @xmath37 , @xmath38 , in @xmath4 . we use greek letters for coordinates allowing arbitrary coordinate changes . then the burgers vector can be expressed as the integral of the displacement vector @xmath39 where @xmath40 is a closed contour surrounding the dislocation axis , fig . [ fburco ] . .45 this integral is invariant under arbitrary coordinate transformations @xmath41 and covariant under global @xmath0-rotations of @xmath1 . here , components of the displacement vector field @xmath42 are considered with respect to the orthonormal basis in the tangent space , @xmath43 . if components of the displacement vector field are considered with respect to the coordinate basis @xmath44 , the invariance of the integral ( [ eburge ] ) under general coordinate changes is violated . in the geometric approach , we introduce new independent variable the vielbein instead of partial derivatives @xmath45 @xmath46 the vielbein is a smooth function on the cut by construction . we note that if the vielbein was simply defined as the partial derivative @xmath47 , then it would have the @xmath48-function singularity on the cut because functions @xmath49 have a jump . the burgers vector can be expressed through the integral over a surface @xmath50 having contour @xmath40 as the boundary @xmath51 where @xmath52 is the surface element . as a consequence of the definition of the vielbein in ( [ edevid ] ) , the integrand is equal to zero everywhere except at the dislocation axis . for the edge dislocation with constant burgers vector , the integrand has a @xmath48-function singularity at the origin . the criterion for the presence of a dislocation is a violation of the integrability conditions for the system of equations @xmath53 : @xmath54 if dislocations are absent , then the functions @xmath49 exist and define transformation to a cartesian coordinates frame . in the geometric theory of defects , the field @xmath55 is identified with the vielbein . next , we compare the integrand in ( [ eburg2 ] ) with the expression for the torsion in cartan variables ( [ ecurcv ] ) . they differ only by terms containing the @xmath0 connection . this is the ground for the introduction of the following postulate . in the geometric theory of defects , the burgers vector corresponding to a surface @xmath50 is defined by the integral of the torsion tensor : @xmath56 this definition is invariant with respect to general coordinate transformations of @xmath37 and covariant with respect to global rotations . thus , the torsion tensor has straightforward physical interpretation : it is equal to the surface density of the burgers vector . the physical interpretation of the @xmath0 connection is given in section [ sdiscl ] , and now we show how this definition reduces to the expression for the burgers vector ( [ eburg2 ] ) obtained within elasticity theory . if the curvature tensor for the @xmath0 connection is zero , then , according to theorem [ tlotrs ] , the connection is locally trivial , and there exists such @xmath0 rotation such that @xmath57 . in this case , we return to expression ( [ eburg2 ] ) . if the @xmath0 connection is zero and vielbein is a smooth function , then the burgers vector corresponds uniquely to every contour . it can then be expressed as a surface integral of the torsion tensor . the surface integral depends only on the boundary contour but not on the surface due to the stokes theorem . .45 we have shown that the presence of linear defects results in a nontrivial torsion tensor . in the geometric theory of defects , the equality of the torsion tensor to zero @xmath58 is naturally considered the criterion for the absence of dislocations . then , under the name dislocation fall not only linear dislocations but , in fact , arbitrary defects in elastic media . for example , point defects ; vacancies and impurities , are also dislocations . in the first case we cut out a ball from the euclidean space @xmath4 and then shrink the boundary sphere to a point ( fig . [ fpoide ] ) . in the case of impurity , a point of the euclidean space is blown up to a sphere and the produced cavity is filled with the medium . point defects are characterized by the mass of the removed or added media , which is also defined by the vielbein @xcite @xmath59 where @xmath49 are the transition functions to a cartesian coordinate frame in @xmath4 and @xmath60 is the density of the medium which is supposed to be constant . the mass is defined by the difference of two integrals , each of them being separately divergent . the first integral equals to the volume of the medium with defects and the second is equal to the volume of the euclidean space . the torsion tensor for a vacancy or impurity is zero everywhere except at one point , where it has a @xmath48-function singularity . for point defects , the notion of the burgers vector is absent . according to the given definition , the mass of an impurity is positive because the matter is added to the media , and the mass of a vacancy is negative because part of the medium is removed . the negative sign of the mass causes serious problems for a physical interpretation of solutions of the equations of motion or the schrdinger equation . hence , we make a remark . strictly speaking , the integral ( [ emass ] ) should be called the `` bare '' mass because this expression does not account for elastic stresses arising around a point dislocation . the effective mass of such a defect must contain at least two contributions : the bare mass and the free energy coming from elastic stresses . the question about the sign of the effective mass is not solved yet and demands a separate analysis . surface defects may also exist in three - dimensional space , in addition to point and line dislocations . in the geometric approach , all of them are called dislocations because they correspond to a nontrivial torsion . in the preceeding section , we related dislocations to a nontrivial torsion tensor . for this , we introduced an @xmath0 connection . now we show that the curvature tensor for the @xmath0 connection defines the surface density of the frank vector characterizing other well - known defects disclinations in the spin structure of media @xcite . let a unit vector field @xmath61 @xmath62 be given at all points of the medium . for example , @xmath63 has the meaning of the magnetic moment located at each point of the medium for ferromagnets ( fig . [ fspstr]_a _ ) . for nematic liquid crystals , the unit vector field @xmath63 with the equivalence relation @xmath64 describes the director field ( fig . [ fspstr]_b _ ) . examples of media with the spin structure : ( _ a _ ) ferromagnet , ( _ b _ ) liquid crystal.,title="fig : " ] examples of media with the spin structure : ( _ a _ ) ferromagnet , ( _ b _ ) liquid crystal.,title="fig : " ] + we fix some direction in the medium @xmath65 . then the field @xmath61 at a point @xmath8 can be uniquely defined by the field @xmath66 taking values in the rotation algebra @xmath67 ( the rotation angle ) , @xmath68 where @xmath69 is the rotation matrix corresponding to the algebra element @xmath70 . here , we use the following parameterization of the rotation group @xmath0 by elements of its algebra ( see , e.g. , @xcite ) @xmath71 where @xmath72 and @xmath73 is the modulus of the vector @xmath74 . the pseudovector @xmath75 , where @xmath76 is the totally antisymmetric third - rank tensor , @xmath77 , is directed along the rotation axis and its length equals the rotation angle . we call the field @xmath70 spin structure of the media . if a media has a spin structure , then it may have defects called disclinations . for linear disclinations parallel to the @xmath34 axis , the vector field @xmath78 lies in the perpendicular plain @xmath79 . the simplest examples of linear disclinations are shown in fig . [ fdiscl ] . the vector field distributions on the plane @xmath79 for the linear disclinations parallel to the @xmath34 axis . ( _ a _ ) @xmath80 . ( _ b _ ) @xmath81.,title="fig : " ] the vector field distributions on the plane @xmath79 for the linear disclinations parallel to the @xmath34 axis . ( _ a _ ) @xmath80 . ( _ b _ ) @xmath81.,title="fig : " ] + every linear disclination is characterized by the frank vector @xmath82 where @xmath83 and the integral is taken along closed contour @xmath40 surrounding the disclination axis . the length of the frank vector is equal to the total angle of rotation of the field @xmath63 as it goes around the disclination . the vector field @xmath63 defines a map of the euclidean space to a sphere @xmath84 . for linear disclinations parallel to the @xmath34 axis , this map is restricted to a map of the plane @xmath85 to a circle @xmath86 . in this case , the total rotation angle must obviously be a multiple of @xmath87 . for nematic liquid crystals , we have the equivalence relation @xmath64 . therefore , for linear disclinations parallel to the @xmath34 axis , the director field defines a map of the plane into the projective line @xmath88 . in this case , the length of the frank vector must be a multiple of @xmath89 . the corresponding examples of disclinations are shown in fig . [ fdisc2 ] . the director field distribution in the @xmath79 plane for the linear disclinations parallel to the @xmath34 axis . ( _ a _ ) @xmath90 and ( _ b _ ) @xmath91.,title="fig : " ] the director field distribution in the @xmath79 plane for the linear disclinations parallel to the @xmath34 axis . ( _ a _ ) @xmath90 and ( _ b _ ) @xmath91.,title="fig : " ] + as for the displacement field , the field @xmath92 , taking values in the algebra @xmath67 , is not a smooth function in @xmath4 in the presence of disclinations . we make a cut in @xmath4 bounded by the disclination axis . then the field @xmath92 may be considered smooth in the whole space except the cut . we assume that all partial derivatives of @xmath92 have the same limit as it approaches the cut from both sides . then we define the new field @xmath93 the functions @xmath94 are smooth everywhere by construction except , may be , on the disclination axis . then the frank vector may be expressed as the surface integral @xmath95 where @xmath50 is an arbitrary surface having the contour @xmath40 as the boundary . if the field @xmath94 is given , then the integrability conditions for the system of equations @xmath96 are @xmath97 this noncovariant equality yields the criterion for the absence of disclinations . in the geometric theory of defects , we identify the field @xmath94 with the @xmath0 connection . in the expression for the curvature in ( [ ecucav ] ) , the first two terms coincide with ( [ eomcon ] ) , and we therefore postulate the covariant criterion of the absence of disclinations as the equality of the curvature tensor for the @xmath0 connection to zero : @xmath98 simultaneously , we give the physical interpretation of the curvature tensor as the surface density of the frank vector @xmath99 this definition reduces to the previous expression for the frank vector ( [ efrank ] ) in the case where rotation of the vector @xmath78 occures in a fixed plane . in this case , rotations are restricted by the subgroup @xmath100 . the quadratic terms in the expression for the curvature in ( [ ecucav ] ) disappear because the rotation group @xmath101 is abelian , and we obtain the previous expression for the frank vector ( [ efrank ] ) . thus , we described the media with dislocations ( defects of elastic media ) and disclinations ( defects in the spin structure ) in the framework of riemann cartan geometry , the torsion and curvature tensors being identified with the surface density of dislocations and disclinations , respectively . the relations between physical and geometrical notions are summarized in the table [ tdegeo ] . .[tdegeo ] the relation between physical and geometrical notions in the geometric theory of defects . [ cols="<,>,>",options="header " , ] the same physical interpretation of torsion and curvature was considered in @xcite . several possible functionals for the free energy were also considered . in the next section , we propose a new expression for the free energy . until now , we discussed only the relation between physical and geometrical notions . to complete the construction of the geometric theory of defects , we have to postulate equations of equilibrium describing static distribution of defects in media . the vielbein @xmath55 and @xmath0 connection @xmath94 are basic and independent variables in the geometric approach . in contrast to previous geometric approaches , we completely abandon the displacement field @xmath11 and spin structure @xmath70 as the fields entering the system of equilibrium equations . in a general case of a continuous distribution of defects , they simply do not exist . nevertheless , at some level and under definite circumstances they can be reconstructed , and we discuss this in the following section . the expression for the free energy was derived in @xcite . we assume that equations of equilibrium must be covariant under general coordinate transformations and local rotations , be at most of the second order , and follow from a variational principle . the expression for the free energy leading to the equilibrium equations must then be equal to a volume integral of the scalar function ( the lagrangian ) that is quadratic in torsion and curvature tensors . there are three independent invariants quadratic in the torsion tensor and three independent invariants quadratic in the curvature tensor in three dimensions . it is possible to add the scalar curvature and a `` cosmological '' constant @xmath102 . we thus obtain a general eight - parameter lagrangian @xmath103 where @xmath104 , @xmath105 and @xmath106 are some constants , and we have introduced the trace of torsion tensor @xmath107 and the ricci tensor @xmath108 . here and in what follows , transformation of the greek indices into the latin ones and vice versa is always performed using the vielbein and its inverse . for example , @xmath109 the particular feature of three dimensions is that the full curvature tensor is in a one - to - one correspondence with ricci tensor ( [ ecurir ] ) and has three irreducible components . therefore , the lagrangian contains only three independent invariants quadratic in curvature tensor . we do not need to add the hilbert einstein lagrangian @xmath110 , also yielding second - order equations , to the free energy ( [ egenla ] ) because of identity ( [ eidcur ] ) . thus , the most general lagrangian depends on eight constants and leads to very complicated equations of equilibrium . at present , we do not know precisely what values of the constants describe this or that medium . therefore , we make physically reasonable assumptions to simplify matters . namely , we require that equations of equilibrium must admit the following three types of solutions . 1 . there are solutions describing the media with only dislocations , @xmath111 . 2 . there are solutions describing the media with only disclinations , @xmath112 . 3 . there are solutions describing the media without dislocations and disclinations , @xmath113 . it turns out that these simple assumptions reduce the number of independent parameters from eight to two . we now prove this statement . the lagrangian ( [ egenla ] ) yields the equilibrium equations @xmath114 where the covariant derivative acts with the @xmath0 connection on the latin indices and with the christoffel symbols on the greek ones . for example , @xmath115 the first condition on the class of solutions of the equilibrium equations is that they permit solutions describing the presence of only dislocations in media . this means the existence of solutions with zero curvature tensor corresponding to the absence of disclinations . substitution of the condition @xmath116 into eqn ( [ econne ] ) for the @xmath117 connection yields @xmath118 here @xmath119 , @xmath120 , and @xmath121 are the irreducible components of the torsion tensor , @xmath122 where @xmath123 in a general case of dislocations all irreducible components of torsion tensor differ from zero ( @xmath119 , @xmath120 , @xmath124 ) and eqns ( [ ecotor ] ) have a unique solution @xmath125 for these coupling constants , the first four terms in lagrangian ( [ egenla ] ) are equal to the hilbert einstein lagrangian @xmath126 up to a total divergence due to identity ( [ eidcur ] ) . equation ( [ etriad ] ) then reduces to the einstein equations with a cosmological constant @xmath127 in this way , the first condition is satisfied . according to the second condition , the equations of equilibrium must allow solutions with zero torsion @xmath58 . in this case , the curvature tensor has additional symmetry @xmath128 , and eqn ( [ econne ] ) becomes @xmath129 here , we decompose the ricci tensor onto irreducible components , @xmath130 where @xmath131 note that for zero torsion , the ricci tensor is symmetrical : @xmath132 . contraction of eqn ( [ econre ] ) with @xmath133 leads to the equation @xmath134 in the general case of nonvanishing curvature , the covariant derivatives @xmath135 and @xmath136 differ from zero and are independent . therefore , we obtain two equations for the coupling constants , @xmath137 which have a unique solution @xmath138 in this case , eqn ( [ etriad ] ) for the vielbein corresponding to a nonzero torsion also reduces to einstein equations ( [ eeinst ] ) . the last requirement for the existence of solutions with zero curvature and torsion is satisfied only for the zero cosmological constant @xmath139 . therefore , the simple and physically reasonable requirements define the two parameter lagrangian @xcite @xmath140 which is the sum of the hilbert einstein lagrangian for the vielbein and the square of the antisymmetric part of the ricci tensor . we note that @xmath141 and @xmath142 are constructed for different metrical connections . other quadratic lagrangians were considered , for example , in @xcite . we note that they also contain the displacement vector as an independent variable along with the vielbein . expression ( [ elagfi ] ) defines the free energy density in the geometric theory of defects and leads to the equilibrium equations ( the euler lagrange equations ) . in our geometric approach , the displacement vector and spin structure do not enter expression ( [ elagfi ] ) for the free energy . in the geometric approach , the vielbein @xmath55 and @xmath0 connection @xmath143 are the only variables . the displacement field @xmath11 and the spin structure @xmath144 can be introduced only in those regions of media where defects are absent . as the consequence of the absence of disclinations @xmath145 , the @xmath0 connection is actually a pure gauge ( [ eloctr ] ) , i.e. , the spin structure @xmath144 exists . if , in addition , dislocations are absent ( @xmath58 ) then there is the displacement field such that the vielbein equals its partial derivatives ( [ erepad ] ) . in this and only in this case can we introduce the displacement field and spin structure . we show below that this can be done such that the equations of nonlinear elasticity theory and the principal chiral @xmath0 field are fulfilled . for free energy in ( [ elagfi ] ) , the euler lagrange equations are covariant under general coordinate transformations in @xmath4 and local @xmath0 rotations . this means that any solution of the equilibrium equations is defined up to diffeomorphisms and local rotations . for the geometric theory of defects to make predictions , we have to fix the coordinate system ( to fix the gauge in the language of gauge field theory ) . this allows us to choose a unique representative from each class of equivalent solutions . we say afterwards that this solution of the euler lagrange equations describes the distribution of defects in the laboratory coordinate system . we start with gauge - fixing the diffeomorphisms . for this , we choose the elastic gauge proposed in @xcite . this question is of primary importance , and we discuss it in detail . equations ( [ estati ] ) , ( [ ehook ] ) of elasticity theory for @xmath146 yield the second - order equation for the displacement vector @xmath147 where @xmath148 is the poisson ratio , ( @xmath149 ) , and @xmath150 is the laplace operator . it can be rewritten in terms of the induced metric ( [ emetri ] ) , for which we obtain a first - order equation . we choose precisely this equation as the gauge condition fixing the diffeomorphisms . we note that the gauge condition is not uniquely defined because the induced metric is nonlinear in the displacement vector , and different equations for the metric may have the same linear approximation . we give two possible choices , @xmath151 where we introduced the notation for the trace of metric @xmath152 . gauge conditions ( [ eonead ] ) and ( [ etwoad ] ) are understood in the following way . the metric @xmath153 is the euclidean metric written in an arbitrary coordinate system , for example , in the cylindrical or spherical coordinate system . the covariant derivative @xmath154 is built from the christoffel symbols corresponding to the metric @xmath153 , and @xmath155 as a consequence . the metric @xmath156 is the metric describing dislocations [ an exact solution of the equilibrium equations for the free energy ( [ elagfi ] ) ] . the gauge conditions differ because in the first and second cases the contraction is performed with the metric of dislocation @xmath157 and the euclidean metric @xmath158 , respectively , without changing the linear approximation . both gauge conditions yield eqn ( [ eqsepu ] ) in the linear approximation in the displacement vector ( [ emetri ] ) . this is most easily verified in cartesian coordinates . from the geometric standpoint , we have the following . the medium with dislocations is diffeomorphic to the euclidean space @xmath4 equipped with two metrics @xmath153 and @xmath156 . the metric @xmath153 is a flat euclidean metric written in an arbitrary coordinate system . the metric @xmath156 is not flat and describes the distribution of dislocations in the same coordinate system . in fact , the metric @xmath153 is used only to fix the coordinate system in which the metric @xmath156 is measured . if the solution of the equilibrium equations satisfies one of the gauge conditions ( [ eonead ] ) , ( [ etwoad ] ) , written , for example , in cylindrical coordinate system then we say that the solution is found in the cylindrical coordinates . we suppose here that the distribution of dislocations in elastic media in the laboratory cylindrical coordinate system is described by this particular solution . analogously , we may seek solutions in a cartesian , spherical , or any other coordinate system . gauge conditions may also be written for the vielbein @xmath55 , which is defined by eqn ( [ etetra ] ) . this involves additional arbitrariness because the vielbein is defined up to local rotations . this invariance leads to different linear approximations for the vielbein in terms of the displacement vector . we consider two possibilities in cartesian coordinates : @xmath159 where the index is lowered with the help of the kronecker symbol . for these possibilities and gauge condition ( [ etwoad ] ) we have two gauge conditions for the vielbein @xmath160 where @xmath161 . these conditions differ in the coefficient before the second term . we note that in a curvilinear coordinate system , the covariant derivative @xmath154 must also include the flat @xmath0 connection acting on indices @xmath162 . one can also write other possible gauge conditions having the same linear approximation . the question of the correct choice is unanswered at present and outside the scope of this review . at the moment , we want only to demonstrate that the system of coordinates must be fixed , and that the gauge condition depends on the poisson ratio , which is the experimentally observed quantity . gauge conditions ( [ ethrad])([efouad ] ) are first order equations by themselves and have some arbitrariness . therefore , to fix a solution uniquely , we must impose additional boundary conditions on the vielbein for any given problem . if the defect - free case , @xmath58 , @xmath163 , and the equilibrium equations are satisfied because the euler lagrange equations for ( [ elagfi ] ) are satisfied . in this and only in this case , we can introduce a displacement vector , and the elastic gauge reduces to the equations of nonlinear elasticity theory . in the presence of defects , a displacement field does not exist , and the elastic gauge simply defines the vielbein . in choosing the free energy functional , we required that the conditions @xmath116 and @xmath58 satisfy the euler lagrange equations . this is important because otherwise we would obtain an additional condition on the displacement vector ( the euler lagrange equations ) besides the elasticity theory equations following from the elastic gauge . we stress an important point once again . in the geometric theory of defects , we assume that there is a preferred laboratory coordinate system in which measurements are made . this coordinate system is related to the medium without defects and elastic stresses and corresponds to the flat euclidean space @xmath4 . gauge conditions ( [ eonead ] ) , ( [ etwoad ] ) and ( [ ethrad ] ) , ( [ efouad ] ) are written precisely in this euclidean space @xmath4 and contain a measurable quantity , the poisson ratio @xmath164 . this property essentially distinguishes the geometric theory of defects from the models of gravity in which all coordinate systems are considered equivalent . the elastic gauge is used to fix diffeomorphisms . the expression for the free energy in ( [ elagfi ] ) is also invariant under local @xmath0 rotations , and they must also be fixed . for this , we recently proposed the lorentz gauge for the connection @xcite @xmath165 this gauge is written in the laboratory cartesian coordinate system and has deep physical meaning . that is , let disclinations be absent ( @xmath163 ) . then the @xmath0 connection is a pure gauge : @xmath166 in this case , the lorentz gauge reduces to the principal chiral @xmath0-field equations @xmath167 for the spin structure @xmath92 . principal chiral field models ( see , e.g. , [ 6468 ] ) for different groups and in a different number of dimensions attract much interest in mathematical physics because they admit solutions of topological soliton types and find broad applications in physics . thus , the lorentz gauge ( [ elorga ] ) means the following . in the absence of disclinations , equations of equilibrium are identically satisfied , and there exists a field @xmath70 that satisfies equations for the principal chiral field . by this we mean that the spin structure of the medium is described by the model of the principal chiral field in the defect - free case . the principal chiral field model is not the only one that can be used for fixing local rotations . the skyrme model @xcite can also be used for this purpose . the euler lagrange equations for this model are not difficult to rewrite in terms of the @xmath0 connection and use as the gauge conditions . there are other models for spin structures . for describing the distribution of magnetic moments in ferromagnets or the director field in liquid crystals , one uses the expression for the free energy depending on the vector @xmath78-field itself @xcite . lately , much attention is paid to the faddeev model of the @xmath78-field @xcite . the question whether there are gauge conditions on the @xmath0 connection that yield these models in the absence of disclinations is unanswered at present . thus , we pose the following problem in the geometric theory of defects : to find the solution of the euler lagrange equations for free energy ( [ elagfi ] ) that satisfies the elastic gauge for the vielbein and the lorentz gauge for the @xmath0 connection . in sections [ swedie ] and [ swegeo ] , we solve this problem for the wedge dislocation in the framework of the classical elasticity theory and the geometric theory of defects , respectively , and afterwards compare the obtained results . in the preceeding section , we used the elasticity theory and the principal chiral @xmath0-field model to fix the invariance of free energy ( [ elagfi ] ) in the geometric theory of defects . this is not a unique possibility , because other models may be used for gauge fixing . in the present section , we show how another model asymmetric elasticity theory can be used for fixing diffeomorphisms and local rotations . at the beginning of the last century , the cosserat brothers developed the theory of elastic media , every point of which is characterized not only be its position but also by its orientation in space @xcite , i.e. , a vielbein is specified at every point ( fig.[fcosme ] ) . .45 from physical standpoint , this means that every atom in the crystalline structure is not a point but an extended object having orientation . in this case , the stress tensor is no longer a symmetric tensor , and the corresponding theory is called the asymmetric elasticity theory . contemporary exposition of this approach is given in @xcite . in the present section , we show that the asymmetric theory of elasticity is naturally incorporated into the geometric theory of defects . the main variables in the asymmetric theory of elasticity are the displacement vector @xmath42 and the rotation angle @xmath168 . the direction of the pseudovector @xmath74 coincides with the rotation axis of the medium element and its length is equal to the angle of rotation . the angle of rotation discussed in section [ sdiscl ] is dual to the spin structure field @xmath92 : @xmath169 . the cosserat medium is characterized by the stress tensor @xmath170 ( the density of forces acting on the surface with normal @xmath171 in the direction @xmath172 ) and the torque stress tensor @xmath173 ( the density of torques acting on the surface with normal @xmath171 in the direction @xmath172 ) . the cosserat medium is in equilibrium if forces and torques are balanced at every point , @xmath174 where @xmath20 and @xmath175 are the densities of nonelastic external forces and torques . as a consequence of eqn ( [ emofor ] ) , the stress tensor is symmetric if and only if the condition @xmath176 is satisfied . the displacement field and rotation angle uniquely define the deformation tensor @xmath177 and the twist tensor @xmath178 : @xmath179 in general , the deformation and twist tensors have no symmetry in their indices . hook s law in the cosserat medium is changed to two linear relations connecting the stress and torque tensors with the deformation and twist tensors , @xmath180}+{\lambda}{\delta}_{ij}\e_k{}^k , \\ \label{esizkr } \mu_{ij}&=2\g\kappa_{\lbrace ij\rbrace}+2\e\kappa_{[ij ] } + { \beta}{\delta}_{ij}\kappa_k{}^k,\end{aligned}\ ] ] where @xmath181 are the lam coefficients , and @xmath182 are four new elastic constants characterizing the medium . braces and square brackets denote symmetrization and antisymmetrization of indices , respectively . in the case where @xmath183 the deformation tensor is symmetric and has the previous form ( [ edefte ] ) . equation ( [ emofor ] ) , together with ( [ estrte ] ) and ( [ esizkr ] ) , then reduces to the equation @xmath184 the first term vanishes as a consequence of eqn ( [ eqsepu ] ) . thus , for spin structure ( [ eresye ] ) and @xmath185 we return to the symmetric elasticity theory . equations ( [ enewfo ] ) , ( [ emofor ] ) , ( [ estrte ] ) , and ( [ esizkr ] ) , together with the boundary conditions , define the equilibrium state of cosserat media . we now show how this model is included in the geometric theory . first , we note that in the absence of defects ( @xmath58 , @xmath116 ) , the fields @xmath11 and @xmath70 exist . then the vielbein and the @xmath0 connection are defined by the deformation and twist tensors in the linear approximation : @xmath186 we note that relations ( [ edeizk ] ) can be regarded as equations for the displacement vector and rotation angle . the corresponding integrability conditions were obtained in @xcite . these integrability conditions are the linear approximations of equalities @xmath58 and @xmath116 defining the absence of defects . if nonelastic forces and torques are absent ( @xmath146 , @xmath185 ) , then the asymmetric theory of elasticity reduces to second - order equations for the displacement vector and rotation angle : @xmath187 we rewrite these equations for the vielbein and @xmath0 connection @xmath188 of course , these are not the only equations that coincide with eqns ( [ eqtrfi ] ) and ( [ eqstco ] ) in the linear approximation . at present , we do not have arguments for the unique choice . the derived nonlinear equations of the asymmetric elasticity theory can be used as gauge conditions in the geometric theory of defects . here , we have six equations for fixing diffeomorphisms ( three parameters ) and local @xmath0 rotations ( three parameters ) . thus , the asymmetric theory of elasticity is naturally embedded in the geometric theory of defects . in section [ sfixga ] , we considered the elastic gauge for the vielbein and the lorentz gauge for the @xmath0 connection . in this case , the spin structure variables do not interact with elastic deformations when defects are absent . in the asymmetric elasticity theory , the elastic stresses directly influence the spin structure and vice versa . .45 by wedge dislocation , we understand an elastic medium that is topologically the euclidean space @xmath4 without the @xmath189 axis the core of dislocation , obtained in the following way . we consider the infinite elastic medium without defects and cut out the infinite wedge of angle @xmath190 . for definiteness , we assume that the edge of the wedge coincides with the @xmath191 axis ( fig.[fwedis ] ) . the edges of the cut are then moved symmetrically one to the other and glued together . after that , the medium moves to the equilibrium state under the action of elastic forces . if the wedge is cut out from the medium , then the angle is considered negative : @xmath192 . for positive @xmath193 , the wedge is added . thus , the elastic medium initially occupies a domain greater or lesser than the euclidean space @xmath4 , depending on the sign of the deficit angle @xmath193 ; in the cylindrical coordinates @xmath194 this domain is described by the inequalities @xmath195 we note that the wedge dislocation is often called disclination . in our approach , this term seems unnatural because the wedge dislocation is related to a nontrivial torsion . moreover , the term disclination is used for defects in the spin structure . we now proceed with the mathematical formulation of the problem for the wedge dislocation in the framework of elasticity theory . to avoid divergent expressions arising for infinite media , we suppose that the wedge dislocation is represented by the cylinder of a finite radius @xmath196 . this problem has translational symmetry along the @xmath191 axis and rotational symmetry in the @xmath197 plane . therefore , we use cylindrical coordinate system . let @xmath198 be the components of the displacement covector with respect to orthonormal basis in the cylindrical coordinate system . this covector satisfies the equilibrium equation following the substitution of ( [ ehook ] ) in eqn ( [ estati ] ) in domain ( [ ecylco ] ) , @xmath199 where @xmath200 is the covariant derivative for the flat euclidean metric in the considered coordinate system . for references , we write expressions for the divergence and laplacian for a covector field in the cylindrical coordinate system : @xmath201 taking the symmetry of the problem into account , we seek the solution of eqn ( [ qeeste ] ) in the form @xmath202 where @xmath203 and @xmath204 are two unknown functions depending only on the radius . we impose the boundary conditions : @xmath205 the first four equations are geometrical and correspond to the process of dislocation creation . the last condition has simple physical meaning : the absence of external forces at the boundary of the medium . the unknown function @xmath204 is found from the second to the last boundary condition ( [ ebocow ] ) @xmath206 straightforward substitution then shows that @xmath207 and @xmath191 components of equation of equilibrium ( [ qeeste ] ) are identically satisfied , and the radial component reduces to the equation @xmath208 the general solution of this equation is @xmath209 the constant of integration @xmath210 due to the boundary condition at zero . the constant @xmath211 is found from the last boundary condition in ( [ ebocow ] ) . finally , we obtain the known solution of the considered problem @xcite @xmath212 the letter @xmath213 in the expression for @xmath214 denotes the base of the natural logarithm . we note that the radial component of the displacement vector diverges as @xmath215 . this means that the description of the wedge dislocation requires considering a finite - radius cylinder . the linear elasticity theory is applicable for small relative displacements , which for a wedge dislocation are equal to @xmath216 this means that we are able to expect correct results for the displacement field for small deficit angles ( @xmath217 ) and near the boundary of the cylinder ( @xmath218 ) . we find the metric induced by the wedge dislocation in the linear approximation in the deficit angle @xmath193 . calculations can be performed using the general formulas ( [ emetri ] ) or the known expression for the variation of the form of the metric @xmath219 after simple calculations , we obtain the expression for the metric in the @xmath197 plane : @xmath220 this metric is compared with the metric obtained as the solution of three - dimensional einstein equations in section [ swegeo ] . wedge dislocations are relatively rarely met in nature because they require a much amount of a medium to be added or removed , resulting in a vast quantity of energy expenses . nevertheless , their study is of great importance because other linear dislocations can be expressed as a superposition of wedge dislocations . in this respect , wedge dislocations are elementary . we show this for an edge dislocation one of the most widely spread dislocations , as an example . an edge dislocation with the core coinciding with the @xmath191 axis is shown in fig.[feddis]_a_. directed to the dislocation axis ( _ a _ ) . edge dislocation as the dipole of two wedge dislocations with positive and negative deficit angles ( _ b_).[feddis],title="fig : " ] + it appears as the result of cutting the medium over the half - plane @xmath221 , @xmath222 , moving the lower edge of the cut towards the @xmath191 axis on a constant ( far from the core of the dislocation ) burgers vector @xmath33 , and subsequently gluing the edges . to find the displacement vector field for the edge dislocation , we may solve the corresponding boundary value problem for the equilibrium equations ( [ qeeste ] ) @xcite . however , we follow another way , knowing the explicit form of the displacement vector for a wedge dislocation . the edge dislocation is represented by the dipole of two wedge dislocations with positive , @xmath223 , and negative , @xmath190 , deficit angles as shown in fig.[feddis]_b_. we assume that the axes of the first and second wedge dislocations are parallel to the @xmath191 axis and intersect the @xmath197 plane at points with the respective coordinates @xmath224 and @xmath225 . the distance between the wedge dislocation axes is equal to @xmath226 . it follows from the expression for the displacement field ( [ ewedis ] ) that far away from the origin ( @xmath227 ) , the displacement field for the wedge dislocations has the following form in the first order in small @xmath193 and @xmath228 : @xmath229 , \\ u_y^{(1)}&\approx-\theta\left [ \frac{1 - 2\s}{2(1-\s)}(y - a){{\sf\,ln\,}}\frac{r - a{{\sf\,sin\,}}\vf}{{{\sf\,e}}r } + x\left(\vf-\frac{a{{\sf\,cos\,}}\vf}r\right)\right ] , \end{split } \\ \label{esewdi } \begin{split } u_x^{(2)}&\approx\phantom{-}\theta\left [ \frac{1 - 2\s}{2(1-\s)}x{{\sf\,ln\,}}\frac{r+a{{\sf\,sin\,}}\vf}{{{\sf\,e}}r } -(y+a)\left(\vf+\frac{a{{\sf\,cos\,}}\vf}r\right)\right ] , \\ u_y^{(2)}&\approx\phantom{-}\theta\left [ \frac{1 - 2\s}{2(1-\s)}(y+a){{\sf\,ln\,}}\frac{r+a{{\sf\,sin\,}}\vf}{{{\sf\,e}}r } + x\left(\vf+\frac{a{{\sf\,cos\,}}\vf}r\right)\right ] . \end{split}\end{aligned}\ ] ] it is sufficient to sum displacement fields ( [ efiwdi ] ) and ( [ esewdi ] ) to find the displacement field for the edge dislocation because the elasticity theory equations are linear . after simple calculations , up to the translation of the whole medium by a constant vector along the @xmath230 axis , we obtain @xmath231 , \\ u_y&=-b\left[\frac{1 - 2\s}{2(1-\s)}{{\sf\,ln\,}}\frac r{{{\sf\,e}}r } + \frac1{2(1-\s)}\frac{x^2}{x^2+y^2}\right ] , \end{split}\ ] ] where we have introduced the notation for the modulus of the burgers vector @xmath232 this result coincides with the expression for the displacement field obtained by direct solution of the elasticity theory equations @xcite . thus , we have shown that an edge dislocation is the dipole of two parallel wedge dislocations . we next find the metric induced by the edge dislocation . using formulas ( [ einmec ] ) , we obtain the metric in the @xmath197 plane in the linear approximation in @xmath193 and @xmath228 : @xmath233 we note that the induced metric for an edge dislocation does not depend on @xmath196 . in the absence of disclinations ( @xmath145 ) , the @xmath0 connection is a pure gauge , and equations of equilibrium for the @xmath0 connection ( [ econne ] ) are identically satisfied . in this case , the explicit form of the @xmath0 connection is uniquely determined by the spin structure @xmath70 . the field @xmath70 satisfies equations for the principal chiral field as a consequence of the lorentz gauge in ( [ elorga ] ) . solution of this system of equations defines the trivial @xmath0 connection . thus in the absence of disclinations , the problem is reduced to solution of the einstein equations for the vielbein in the elastic gauge and solution of the principal chiral field model for the spin structure . after that , we can compute the torsion tensor through the formulas ( [ ecurcv ] ) , which defines the surface density of the burgers vector . because disclinations are absent ( the curvature tensor is equal to zero ) , we have the space of absolute parallelism . the whole geometry is then defined by the vielbein @xmath55 , which uniquely defines the torsion tensor via ( [ ecurcv ] ) for a vanishing @xmath0 connection . here , we assume that a trivial @xmath0 connection is equal to zero . the vielbein @xmath55 satisfies the three - dimensional einstein equations with a euclidean signature metric , which follow from the expression for the free energy ( [ elagfi ] ) for @xmath163 , @xmath234 here , we have added the source of dislocations @xmath235 to the right hand side of einstein equations ( it is the energy - momentum tensor in gravity ) . we note that without a source of dislocations , the model would be trivial . indeed , the scalar curvature and ricci tensor are equal to zero : @xmath236 , @xmath237 for @xmath238 as a consequence of einstein equations ( [ eindis ] ) . then , the full curvature tensor without sources is also equal to zero because in three - dimensional space , it is in a one - to - one correspondence with ricci tensor ( [ ecurir ] ) . the vanishing of the full curvature tensor means the triviality of the model because defects are absent in this case . the similar statement in three - dimensional gravity is well known . it is usually formulated as : `` three - dimensional gravity does not describe dynamical , i.e. , propagating degrees of freedom '' . for our purposes , we have to find the solution of einstein equations ( [ eindis ] ) describing one wedge dislocation . the einstein equations are a system of nonlinear second - order partial differential equations . not too many exact solutions are known at present , even in a three - dimensional space . the remarkable exact solution describing an arbitrary static distribution of point particles is well known in three - dimensional gravity for the lorentz signature metric @xmath239 [ 7577 ] . we find this solution for the euclidean signature metric and show that it describes an arbitrary distribution of parallel wedge dislocations in the geometric theory of defects . hence , we first consider the more general case of an arbitrary number of wedge dislocations and then analyze in detail one wedge dislocation which is of interest to us . we do this deliberately because the solution in a more general case does not involve essential complications . at the same time , an arbitrary distribution of wedge dislocations is much more interesting for applications . for example , the edge dislocation was shown in the preceeding section to be represented by a dipole of two wedge dislocations of different signs . we consider elastic medium with arbitrary distributed but parallel wedge dislocations . we choose the coordinate system such that the @xmath189 axis is parallel to dislocations axes , and axes @xmath240 , @xmath241 are perpendicular to the @xmath191 axis . then the metric has the block - diagonal form @xmath242 where @xmath243 is a two - dimensional metric on the @xmath197 plane . a two - dimensional metric @xmath244 and a function @xmath245 are independent of @xmath191 due to translational symmetry along the @xmath191 axis . we can say this more simpler dropping the physical arguments . we consider the block - diagonal metric of form ( [ eqmetr ] ) , which has translational invariance along the @xmath191 axis . afterwards , we show that the corresponding solution of the einstein equations indeed describes an arbitrary distribution of parallel wedge dislocations . the curvature tensor for metric ( [ eqmetr ] ) has the components : @xmath246 where @xmath247 is the curvature tensor for the two - dimensional metric @xmath248 and @xmath249 is the two - dimensional covariant derivative with the christoffel symbols . we choose the source of dislocations as @xmath250 where @xmath251 is the two - dimensional @xmath48-function on the @xmath197 plane with the support at a point @xmath252 . the factor @xmath253 in front of the sum sign is due to the property of the @xmath48-function , which is not a function but a tensor density with respect to general coordinate transformations . we show later that the solution of the einstein equations with such a source describes @xmath254 parallel wedge dislocations with deficit angles @xmath255 , which intersect the @xmath197 plane at the points @xmath256 . in three - dimensional gravity this source corresponds to particles of masses @xmath257 , being at rest at the points @xmath258 . einstein equations ( [ eindis ] ) then reduce to four equations , @xmath259 where @xmath260 is the two - dimensional scalar curvature . the metric of form ( [ eqmetr ] ) is still invariant under coordinate transformations in the @xmath197 plane . using this residual symmetry , we fix the conformal gauge on the plane ( this is always possible locally ) @xmath261 where @xmath262 is some function . in the conformal gauge , eqn ( [ eeinon ] ) becomes @xmath263 for constant boundary conditions for @xmath264 on the boundary of the @xmath197 plane this equation has the unique solution @xmath265 . changing the scale of @xmath191 coordinate , we can set @xmath266 without loss of generality . then eqn ( [ eeintw ] ) reduces to the poisson equation @xmath267 which has the general solution @xmath268 thus the metric in the @xmath197 plane is @xmath269 where the polar coordinates @xmath270 cover the whole plane @xmath85 and not more ( this is important ! ) . any solution of the einstein equations is defined up to choosing the coordinate system because the equations are covariant . using this , we set @xmath271 , which is always possible by choosing the scale of @xmath272 . this is indeed the exact solution of the einstein equations describing arbitrary distribution of parallel wedge dislocations . this statement is made clear from the following consideration . we note that transition to a continuous distribution of dislocations in the geometric approach is simple . for this , we have to substitute a continuous distribution of sources in the right hand side of the einstein equations instead of @xmath48-sources . we consider one wedge dislocation with the source at the origin in more detail to show that metric ( [ edisco ] ) indeed describes an arbitrary distribution of wedge dislocations @xmath273 the corresponding metric ( [ edisco ] ) for @xmath271 is @xmath274 we pass to a new coordinate system @xmath275 in which the metric becomes euclidean @xmath276 but the range of the polar angle differs now from @xmath87 : @xmath277 , and covers the @xmath197 plane with removed or added angle @xmath223 . because the metric coincides with the euclidean one in the primed coordinate system @xmath278 , we have the euclidean plane with a removed or added wedge because the angle @xmath279 varies within the interval @xmath280 . the transformation to coordinates @xmath270 in ( [ ekoorv ] ) means the gluing of the edges of the wedge that has appeared , which produces a cone . therefore , both metrics ( [ eonwed ] ) and ( [ equlco ] ) describe the same geometric object conical singularity . the torsion and curvature tensors are obviously zero everywhere except at the origin . creation of a conical singularity coincides exactly with creation of the wedge dislocation in the geometric theory of defects . it is not difficult to show that general solution ( [ edisco ] ) describes an arbitrary distribution of conical singularities with deficit angles @xmath255 at points @xmath258 . thus , this solution describes an arbitrary distribution of parallel wedge dislocations . in the next section , we consider one wedge dislocation in detail . for this , we perform one more coordinate transformation @xmath281 metric ( [ equlco ] ) then becomes @xmath282 this is one more frequently used form of the metric for a conical singularity . we now consider a wedge dislocation from the geometric standpoint . from the qualitative standpoint , the creation of a wedge dislocation coincides with the definition of conical singularity . however , there is a quantitative difference because metric ( [ ecosth ] ) depends only on the deficit angle @xmath193 and can not coincide with the induced metric ( [ emewed ] ) obtained within the elasticity theory . the difference arises because we require the displacement vector in the equilibrium to satisfy equilibrium equations after removing the wedge and gluing its edges ( creating conical singularity ) in the elasticity theory . at the same time the @xmath197 plane for a conical singularity after the gluing can be deformed in an arbitrary way . formally , this manifests itself in that metric ( [ emewed ] ) obtained within the elasticity depends explicitly on the poisson ratio , which is absent in gravity theory . we proposed the elastic gauge for solving this problem @xcite . we choose elastic gauge ( [ efouad ] ) as the simplest one for a wedge dislocation . this problem can be solved in two ways . first , the gauge condition can be inserted in the einstein equations directly . second , we can find the solution in any suitable coordinate system and then find the coordinate transformation such that the gauge condition is satisfied . it is easier to follow the second way because the solution for the metric is known , eqn ( [ ecosth ] ) . the vielbein can be associated with metric ( [ ecosth ] ) as @xmath283 here , a hat over an index means that it corresponds to the orthonormal coordinate system , and an index without a hat is the coordinate one . components of this vielbein are the square roots of the corresponding metric components and therefore have the symmetric linear approximation ( [ etrsep ] ) . we transform the radial coordinate @xmath284 because a wedge dislocation is symmetric under rotations in the @xmath197 plane . after that transformation the vielbein components take the form @xmath285 where the prime denotes differentiation with respect to @xmath272 . we choose the vielbein corresponding to the euclidean metric as @xmath286 it defines the christoffel symbols @xmath287 and @xmath0 connection @xmath288 , which define the covariant derivative . we write only nontrivial components @xmath289 substitution of the vielbein into gauge condition ( [ efouad ] ) yields the euler differential equation for the transition function @xmath290 its general solution depends on two arbitrary constants @xmath291 , @xmath292 where the exponents @xmath293 are defined by the quadratic equation @xmath294 which has real roots for @xmath295 with different signs : positive root @xmath296 and negative @xmath297 . we recall that there are thermodynamical constraints @xmath149 on the poisson ratio @xcite . to fix the constants , we impose boundary conditions on the vielbein : @xmath298 the first boundary condition corresponds to the last boundary condition for displacement vector ( [ ebocow ] ) ( the absence of external forces on the surface of the cylinder ) , and the second one corresponds to the absence of the angular component of the deformation tensor at the core of dislocation . equations ( [ eborep ] ) define the values of integration constants @xmath299 the obtained vielbein defines the metric @xmath300 where @xmath301 this is the solution of the posed problem . the derived solution is valid for all deficit angles @xmath193 and for all @xmath302 . the obtained metric depends on three constants : @xmath193 , @xmath164 , and @xmath196 . the dependence on the deficit angle @xmath193 is due to its occurrence on the right hand side of einstein equations ( [ eindis ] ) . the dependence on the poisson ratio comes from elastic gauge ( [ efouad ] ) , and , finally , the dependence on the cylinder radius comes from boundary condition ( [ eborep ] ) . if a wedge dislocation is absent , then @xmath303 , @xmath304 , @xmath305 , and metric ( [ emeweg ] ) goes to the euclidean one @xmath306 , as expected . we compare metric ( [ emeweg ] ) obtained within the geometric approach with the induced metric from elasticity theory in eqn ( [ emewed ] ) . first , it has a simpler form . second , in the linear approximation in @xmath193 @xmath307 and metric ( [ emeweg ] ) can be easily shown to coincide precisely with metric ( [ emewed ] ) obtained within the elasticity theory . we see that induced metric ( [ emewed ] ) provides only the linear approximation for the metric obtained within the geometric theory of defects , which , in addition , has a simpler form . beyond the perturbation theory , we see essential differences . in particular , metric ( [ emewed ] ) is singular at the origin , whereas metric ( [ emeweg ] ) obtained beyound the perturbation theory is regular . the stress and deformation tensors are related by hook s law ( [ ehook ] ) . there is an experimental possibility to check formulas ( [ emeweg ] ) because the deformation tensor is the linear approximation for the induced metric . for this one has to measure the stress field for a single wedge dislocation . in this way , the geometric theory of defects can be experimentally confirmed or discarded . the problem of reconstruction of the displacement field for a given metric reduces to solution of differential equations ( [ emetri ] ) with metric ( [ emeweg ] ) on the right hand side and boundary conditions ( [ ebocow ] ) . we do not discuss this problem here . we note that in the geometric theory of defects , a complicated stage of finding the displacement vector where it exists is simply absent and is not necessary . two - dimensional metric ( [ emeweg ] ) describes the conical singularity in the elastic gauge . the relation between conical singularities and wedge dislocations was established in [ 22,7880 ] . in these papers , the metric was used in the other gauges ( coordinate systems ) . elastic oscillations in elastic media without defects are described by a time - dependent vector field @xmath308 that satisfies the wave equation ( see , e.g. , @xcite ) @xmath309 where dots denote differentiation with respect to time and @xmath60 is the mass density of medium , which is assumed to be constant . if the medium contains defects , then the metric of the space becomes nontrivial , @xmath310 . we assume that relative displacements for elastic oscillations are much smaller than stresses induced by defects : @xmath311 then , in the first approximation , we assume that elastic oscillations propagate in a riemannian space with a nontrivial metric induced by dislocations . here , we discard changes in the metric due to elastic oscillation themselves . therefore , for elastic oscillations we postulate the following equation , which is a covariant generalization of ( [ efleld ] ) for spatial variables : @xmath312 where @xmath11 denote components of displacement vector field with respect to an orthonormal space basis @xmath313 ( see the appendix ) , @xmath314 is the covariant laplace beltrami operator built for the vielbein @xmath55 , and @xmath315 is the covariant derivative . the explicit form of the covariant derivative of the displacement field is @xmath316 where @xmath317 is the @xmath0 connection built for zero torsion . we note that the displacement vector field describing elastic oscillations is not the total displacement vector field of points of a medium with dislocations . it was already said in section [ sdislo ] that the displacement field for dislocations can be introduced only in those regions of media where defects are absent . if we denote it by @xmath318 , then the total displacement field in these regions is defined by the sum @xmath319 . there , the vielbein is defined only by the displacement field for dislocations @xmath320 . we note that the smallness of relative deformations ( [ edueco ] ) is also meaningful in those regions of space where displacements @xmath318 are not defined . we now decompose the displacement field covariantly into transversal @xmath321 and longitudinal parts @xmath322 , @xmath323 which are defined by the relations @xmath324 the decomposition of a vector field into longitudinal and transversal parts in three - dimensional space with an accuracy up a constant is unique . we recall that latin indices are lowered with the help of kronecker symbols , @xmath325 , and this operation commutes with covariant differentiation . equation ( [ edelop ] ) can be rewritten as @xmath326 because transformation of latin indices into greek ones commutes with the covariant differentiation , and the christoffel symbols are symmetrical in the first two indices . the last equality means that the 1-form @xmath327 is closed . it is easily confirmed that eqn ( [ ephops ] ) for elastic oscillations is equivalent to two independent equations for transverse and longitudinal oscillations , @xmath328 where @xmath329 are the squares of sound velocities for transverse and longitudinal oscillations . particles arising after the secondary quantization of eqns ( [ eelaos ] ) are called phonons in solids . therefore , strictly speaking , the problem of scattering of phonons on dislocations is a quantum mechanical one . in the present review , we consider only classical aspects of this problem . wave equations ( [ eelaos ] ) contain second and first derivatives of the displacement field . the latter are contained in the covariant laplace beltrami operator @xmath330 . the terms with second derivatives can be written in the four - dimensional form @xmath331 where @xmath332 is the inverse metric to @xmath333 where @xmath334 is either the transverse or longitudinal sound velocity . above , we used the following notations . four - dimensional coordinates are denoted by greek letters from the beginning of the alphabet @xmath335 , and letters from the middle of greek alphabet denote only space coordinates @xmath336 . this rule can be easily remembered by the following inclusions @xmath337 and @xmath338 . christoffel symbols ( [ etigam ] ) can be computed for four - dimensional metric ( [ etigas ] ) , which defines a system of ordinary nonlinear equations for extremals @xmath339 ( lines of extremal lengths that coincide with geodesics in riemannian geometry ) , where dots denote differentiation with respect to canonical parameter @xmath340 . for the block - diagonal metric in ( [ etigas ] ) , these equations decompose : @xmath341 where @xmath342 are the three - dimensional christoffel symbols constructed for the three - dimensional metric @xmath156 which , as we recall , depends only on spatial coordinates for a static distribution of defects . let @xmath343 be an arbitrary extremal in four - dimensional space - time . for metric ( [ etigas ] ) , its natural projection on space @xmath344 is also an extremal but now for the spatial part of the metric @xmath156 . equations for extremals ( [ efextr ] ) , ( [ esextr ] ) are invariant under linear transformations of the canonical parameter @xmath340 . therefore , the canonical parameter can be identified with time , @xmath345 , without loss of generality as a consequence of eqn ( [ efextr ] ) . we assume that a particle moves in space along an extremal @xmath346 with velocity @xmath347 . this means that @xmath348 the length of the tangent vector to the corresponding extremal @xmath349 in four - dimensional space - time is then equal to @xmath350 hence , if the particle moves in space along extremal with velocity less than , equal to , or greater than the speed of sound , then its world line in space - time is timelike , null , or spacelike , respectively . we return now to propagation of phonons in media with defects . as in geometric optics @xcite , there are useful notions of wave fronts and rays in the analysis of the asymptotic form of solutions for wave equations ( [ eelaos ] ) . we do not consider mathematical aspects of this approach , which is nontrivial and complicated @xcite , and instead give only a physical description . in the eikonal ( high frequency ) approximation , phonons propagate along rays coinciding with null extremals for the four - dimensional metric @xmath248 . forms of rays , which are identified with trajectories of phonons , are defined by the three - dimensional metric @xmath156 . this means that in the eikonal approximation , trajectories of transverse and longitudinal phonons in a medium with defects are the same and are defined by eqn ( [ esextr ] ) . the difference reduces to the velocities of propagation for transverse and longitudinal phonons being different and equal to @xmath351 and @xmath352 , respectively . calculations in the present section coincide , in fact , with the analysis performed in section 3 of @xcite . the difference is that in what follows , we use the metric written in the elastic gauge . this is important because we assume that trajectories of phonons seen in an experiment coincide with extremals for the metric precisely in this gauge . in the presence of one wedge dislocation , the space metric in the cylindrical coordinates @xmath194 is @xmath353 0 & { \displaystyle\dfrac{\raise-.7ex\hbox{${\alpha}^2$}}{\g^2 } \dfrac{\raise-.7ex\hbox{$r^{2\g}$}}{r^{2\g-2 } } } & 0 \\[2 mm ] 0 & 0 & 1 \end{pmatrix}\ ] ] where the nontrivial part of the metric in the @xmath270 plane was obtain earlier [ see eqn ( [ emeweg ] ) ] . here , we change @xmath296 to @xmath354 for simplicity of notation . the inverse metric is also diagonal , @xmath355 0 & { \displaystyle\dfrac{\raise-.7ex\hbox{$\g^2$}}{{\alpha}^2 } \dfrac{\raise-.7ex\hbox{$r^{2\g-2}$}}{r^{2\g } } } & 0 \\[2 mm ] 0 & 0 & 1 \end{pmatrix}\ ] ] christoffel symbols for metric ( [ ewemep ] ) are calculated according to formulas ( [ etigam ] ) . as a result , only four christoffel symbols differ from zero : @xmath356 in the preceeding section , we showed that in the eikonal approximation , phonons propagate along extremals @xmath346 defined by eqns ( [ esextr ] ) . in the case considered here , these equations are @xmath357 where the dot denotes differentiation with respect to time @xmath358 . it follows from the last equation that phonons move along the @xmath191 axis with constant velocity , which corresponds to translational invariance along @xmath191 . this means that scattering on a wedge dislocation is reduced to a two - dimensional problem in the @xmath270 plane , as could be expected . the system of equations for @xmath359 and @xmath360 in ( [ extrfw ] ) and ( [ extvfw ] ) can be explicitly integrated . for this , we find two first integrals . first , for any metric , there is the integral for equations for extremals @xmath361 we then have the equality @xmath362 second , the invariance of the metric under rotations about the @xmath191 axis results in the existence of an additional integral . it is constructed as follows . there is a killing vector corresponding to the invariance of the metric , which in cylindrical coordinates has the simple form @xmath363 . straightforward verification proves that @xmath364 in the considered case , this results in the identity @xmath365 we analyze the form of an extremal @xmath366 . first derivatives can be found from eqns ( [ efinte ] ) and ( [ einsew ] ) @xmath367 admissible values of the radial coordinate @xmath272 for which the expression under the square root is nonnegative are to be found later . from the above equations , we obtain the equation defining the form of nonradial ( @xmath368 ) extremals @xmath369 this equation can be easily integrated , and we finally obtain explicit formulas defining the form of an extremal : @xmath370=1,\ ] ] where @xmath371 the constant @xmath372 is positive and defines the minimal distance at which an extremal approaches the core of the dislocation , i.e. , @xmath373 . only for these values of @xmath272 is the expression under the root in eqn ( [ eqfexw ] ) nonnegative . the integration constant @xmath374 is arbitrary and corresponds to the invariance of the problem under rotations around the core of dislocation . equations for extremals ( [ extrfw ] ) and ( [ extvfw ] ) also have degenerate solutions : @xmath375 these extremals correspond to the radial motion of phonons . such trajectories are unstable in the sense that there are no nonradial extremals near them . we note that circular extremals @xmath376 are absent as a consequence of eqn ( [ extrfw ] ) , although integrals of motion ( [ efinte ] ) and ( [ einsew ] ) admit such a solution . this is because eqn ( [ extrfw ] ) was multiplied by @xmath377 in deriving first integral ( [ efinte ] ) . we now analyze the form of nonradial extremals ( [ efoexe ] ) . for any extremal , the radius @xmath272 decreases first from infinity to the minimal value @xmath372 and then increases from @xmath372 to infinity . we can assume here without loss of generality that the argument of sine in ( [ efoexe ] ) varies from 0 to @xmath89 . thus , we obtain the range of changes of the polar angle @xmath378 this means that the extremal comes from infinity at the angle @xmath379 and goes to infinity at the angle @xmath380 . this corresponds to the scattering angle @xmath381 we note that the scattering angle depends only on the deficit angle @xmath193 and does not depend on the elastic properties of the medium . the scattering angle has a simple physical interpretation . for positive @xmath193 , the medium is cut and moved apart . a wedge of the same medium without elastic stresses is inserted in the cavity that has appeared . afterwards , gluing is performed , and the wedge is compressed . the compression coefficient for all circles centered at the origin is equal to @xmath382 due to symmetry considerations . therefore , the scattering angle equals half the deficit angle times the compression coefficient , @xmath383 for @xmath303 , the dislocation is absent , and the scattering angle is equal to zero . for positive deficit angles , the scattering angle is positive , which corresponds to repulsion of phonons from the wedge dislocation . corresponding extremals are shown in fig.[fexwpo]_a _ , and they have asymptotes as @xmath384 . _ extremals for the wedge dislocation with a positive ( @xmath385 ) deficit angle . two extremals and their reflections with respect to the @xmath8 axis are shown for the same @xmath385 but different @xmath372 . _ ( b ) _ extremals for the wedge dislocation with a negative ( @xmath386 ) deficit angle . two extremals are shown for the same @xmath372 but different deficit angles @xmath193.,title="fig : " ] + we note that for positive deficit angles , no two points on the @xmath270 plane can be connected by an extremal , i.e. , there is a domain to the right of the wedge dislocation that can not be reached at all by a phonon falling from the left . all extremals shown in the figures in the present section are calculated numerically . the values of the deficit angle @xmath193 , the scattering angle @xmath387 , and the minimal distance @xmath372 to the dislocation axis are shown in the figures . for definiteness , we choose @xmath388 corresponding to the fall of phonons from the left . for negative deficit angles the scattering angle is negative and is defined by the same formulas ( [ escang ] ) . this corresponds to the attraction of phonons to the dislocation axis . in fig.[fexwpo]_b _ , we show two extremals with the same parameter @xmath372 but for two dislocations with different deficit angles . for @xmath389 , the scattering angle varies from 0 to @xmath87 ( fig.[fexwpo]_b _ ) . for @xmath390 , the scattering angle is equal to @xmath87 . we note that for negative deficit angles , phonons have no asymptotes as @xmath384 , i.e. , for @xmath390 , phonons fall from infinity ( @xmath391 ) and return to the infinity ( @xmath392 ) . when the deficit angle is sufficiently small ( @xmath393 ) , a phonon makes one or several rotations around the dislocation and then goes to infinity . examples of such trajectories are shown in figs.[fwede2][fwede4 ] . for @xmath394 , an extremal makes one rotation around the dislocation and then goes forward in the original direction . the blown - up part of the trajectory is shown in the square to the right.,title="fig : " ] + for @xmath395 , an extremal makes two turns around the dislocation and then goes back . the blown - up part of the trajectory is shown in the square to the right.,title="fig : " ] + for @xmath396 , an extremal makes two and a half rotations around the dislocation and then goes forward in the original direction . subsequent blown - up parts of the trajectory are shown in the two squares to the right.,title="fig : " ] + we consider the asymptotic behavior of nonradial extremals as @xmath397 . as a consequence of eqn ( [ erdotf ] ) , far from the core of dislocation , we obtain @xmath398 it follows from this equation that the dependence of the radius on time is the same as for radial extremals ( [ eradex ] ) . because @xmath399 , an infinite value of @xmath272 corresponds to an infinite value of time @xmath358 . this means that the @xmath270 plane with the given metric is complete at @xmath384 . the origin ( the core of dislocation ) is a singular point . only radial extremals fall into it at a finite moment of time . integrals of motion ( [ efinte ] ) and ( [ einsew ] ) have simple physical meaning . equations for extremals ( [ esextr ] ) follow from the variational principle for the lagrangian @xmath400 describing motion of a free massless point particle in a nontrivial metric @xmath401 . here , the metric is considered as a given external field and is not varied . the energy corresponding to this integral is equal to @xmath402 if the metric , as in our case , does not depend on time explicitly , then the energy is conserved ( @xmath403 ) and its numerical value for the motion in the @xmath270 plane is proportional to the integral of motion @xmath404 . for a wedge dislocation , the metric is independent on the polar angle @xmath207 , and the lagrangian is invariant under rotations : @xmath405 . by the noether theorem , the angular momentum conservation law corresponds to this invariance , @xmath406 as a consequence , the constant of integration @xmath407 is proportional to the angular momentum . we note that the behavior of extremals differs qualitatively from trajectories of point particles moving in flat space with euclidean metric @xmath408 in an external potential field @xmath409 . it can be easily shown that trajectories of point particles described by the lagrangian @xmath410 can not coincide with extremals ( [ esextr ] ) for any function @xmath409 . in the present section , we have showed how the problem of scattering of phonons on the simplest wedge dislocation is solved in the geometric approach . the results of calculations can be checked experimentally . the more general problem of scattering of phonons on an arbitrary distribution of wedge dislocations , including the edge dislocation and continuous distribution of dislocations , is solved in @xcite . solution of this problem in the geometric approach is simpler than in the framework of classical elasticity theory , where we have to solve partial differential equations with complicated boundary conditions . in @xcite , calculations were performed in the conformal gauge for the metric . for comparison with experiment , the results must be recalculated in the elastic gauge , which has physical meaning in the geometric approach . we consider elastic media with one wedge dislocation containing one atom of impurity or vacancy . if we consider the influence of impurity on the distribution of elastic stresses small compared with the elastic stresses induced by the dislocation itself , then the motion of the impurity can be considered as taking place in three - dimensional space with the nontrivial metric ( [ emeweg ] ) . in the geometric approach , we assume that the potential energy of the interaction equals zero , and all interactions are due to changes in the kinetic energy , which depends explicitly on the nontrivial metric . we solve the corresponding quantum mechanical problem . we consider bounded states of impurity moving inside a cylinder of radius @xmath196 in the presence of a wedge dislocation . we assume that the cylinder axis coincides with the core of dislocation . the stationary schrdinger equation is @xmath411 where @xmath412 is the plank constant and @xmath254 , @xmath413 , and @xmath414 are mass , wave function , and energy of the impurity . the nontriviality of the interaction with dislocation reduces to the nontrivial laplace beltrami operator @xmath415 where the metric was found earlier [ see eqn ( [ emeweg ] ) ] , and @xmath416 . taking the symmetry of the problem into account , we solve the schrdinger equation ( [ eshrse ] ) in cylindrical coordinates by the separation of variables . let @xmath417 where we have the two following possibilities for the normalization function @xmath418 . if the impurity moves freely along the @xmath191 axis with momentum @xmath419 , then @xmath420 if its motion is restricted by the planes @xmath421 and @xmath422 , then @xmath423 in what follows , we drop the integer - valued subscript @xmath424 indicating the restricted motion , having both possibilities in mind . the condition for the constant @xmath425 ( the eigenvalue of the projection of the momentum on the @xmath191 axis ) to be an integer appears due to the periodicity condition @xmath426 then for the radial wave function , @xmath427 we then obtain the equation @xmath428 we introduce the new radial coordinate @xmath429 this is the transformation in ( [ eftrra ] ) and ( [ ecoifr ] ) up to a constant . the radial equation is then @xmath430 where @xmath431 this is the bessel equation . we solve it with the boundary condition @xmath432 which corresponds to the motion of an impurity inside the cylinder with an impenetrable boundary . the general solution of the bessel equation ( [ erashe ] ) is @xmath433 where @xmath434 and @xmath435 are bessel and neumann functions of order @xmath436 @xcite . the boundedness of the wave function on the axis of the cylinder requires @xmath437 . the constants of integration @xmath438 are found from the normalization condition @xmath439 boundary condition ( [ eboube ] ) yields the equation for @xmath440 @xmath441 which defines the energy levels of bounded states . it is well known that for real @xmath442 and @xmath443 , this equation has only real roots . positive roots form an infinite countable set and all of them are simple @xcite . this provides the inequality @xmath444 we label the positive roots of eqn ( [ eqenlv ] ) by the index @xmath445 ( principle quantum number ) : @xmath446 . then the spectrum of bounded states is @xmath447 for large radii @xmath448 and @xmath449 we have the asymptotic form : @xmath450 as a result , we obtain explicit expression for the spectrum @xmath451 in the absence of defect , @xmath304 , @xmath452 , @xmath453 , and the radial functions @xmath454 are expressed through the bessel functions of integer order @xmath455 . in this case , the spectrum of bounded states depends only on the sizes of the cylinder . in the presence of the wedge dislocation , bessel functions have a noninteger order in general . in this case , the spectrum of energy levels acquires dependence on the deficit angle @xmath193 and the poisson ratio @xmath456 characterizing elastic properties of the medium . if the mass of impurity and vacancy is defined by integral ( [ emass ] ) , then @xmath457 for impurity and the energy eigenvalues are positive . for vacancy , @xmath458 and the energy eigenvalues are negative . in this case , the energy spectrum is not bounded from below , which causes serious problems for physical interpretation . it seems that one has to insert in the schrdinger equation not the bare mass ( [ emass ] ) but the effective mass , with the contribution of elastic stresses arising around a vacancy taken into account . this question presently remains unanswered . the presentation in the present section is close to that in @xcite . in contrast to that paper , we use the elastic gauge for the metric . therefore , our results depend not only on the deficit angle of a wedge dislocation but also on the elastic properties of the medium . calculations of the energy levels of an impurity in the field of a wedge dislocation are actually equivalent to the calculations of bounded - state energies in the aharonov bohm effect @xcite ( see reviews @xcite ) . the difference reduces only to changing the order of the bessel functions , @xmath459 where @xmath460 is the magnetic flux quantum and @xmath461 is the electron charge . the considered example shows how the influence of the presence of defects is taken into account in the geometric approach in the first approximation . if calculations in some problem were performed in elastic media without defects , then to take the influence of defects into account we have to replace the flat euclidean metric with a nontrivial metric describing the given distribution of defects . this problem may appear complicated mathematically because we have to solve the three - dimensional einstein equations to find the metric . however , there are no principal difficulties : the effect of dislocations reduces to a change in the metric . the geometric theory of defects describes defects in elastic media ( dislocations ) and defects in the spin structure ( disclinations ) from a single point of view . this model can be used for describing single defects as well as their continuous distribution . the geometric theory of defects is based on the riemann - cartan geometry . by definition , torsion and curvature tensors are equal to surface densities of burgers and frank vectors , respectively . equations defining the static distribution of defects are covariant and have the same form as equations of gravity models with dynamical torsion . to choose a solution uniquely , one must fix the coordinate system . for this , the elastic gauge for the vielbein and the lorentz gauge for the @xmath0 connection are proposed . in the defect - free case the displacement vector field and the field of the spin structure can be introduced . equations of equilibrium are then identically satisfied , and the gauge conditions reduce to the equations of elasticity theory and the principal chiral @xmath0-field . in this way , the geometric theory of defects incorporates elasticity theory and the model of principal chiral field . in a certain sense , the elastic gauge represents the equations of the nonlinear elasticity theory . nonlinearity is introduced in elasticity theory in two ways . first , the deformation tensor is defined through the induced metric @xmath462 instead of being defined by linear relation ( [ edefte ] ) . then the stress tensor is given by an infinite series in the displacement vector . second , hook s law can be modified assuming nonlinear dependence of the stress tensor on the deformation tensor . hence , the elastic gauge condition is the equations of the nonlinear elasticity theory where the deformation tensor is assumed to be defined through the induced metric and hook s law is kept linear . a generalization to the nonlinear dependence of the deformation tensor on the stress tensor is obvious . as an example , we considered the wedge dislocation from the standpoint of the elasticity theory and the geometric theory of defects . we showed that the elasticity theory reproduces only the linear approximation of the geometric approach . in contrast to the induced metric obtained within the exact solution of the linear elasticity theory , the expression for the metric obtained as the exact solution of the einstein equations in the elastic gauge is simpler and is defined on the whole space and for all deficit angles . the obtained expression for the metric can be checked experimentally . two problems are considered as applications of the geometric theory of defects . the first is the scattering of phonons on a wedge dislocation . in the eikonal approximation , the problem is reduced to the analysis of extremals for the metric describing a given dislocation . equations for extremals are integrated explicitly , and the scattering angle is found . the second of the considered problems is the construction of the wave functions and energy spectrum of impurity in the presence of a wedge dislocation . this requires solving the schrdinger equation . this problem is mathematically equivalent to solving the schrdinger equation for bound states in the aharonov bohm effect @xcite . the explicit dependence of the spectrum on the deficit angle and elastic properties of the medium is found in the presence of a wedge dislocation . the geometric theory of static distribution of defects can also be constructed for membranes , i.e. , on the plane @xmath85 . for this , one has to consider the euclidean version @xcite of two - dimensional gravity with torsion [ 8890 ] . this model is favored by its integrability [ 9194 ] . the developed geometric construction in the theory of defects can be inverted , and we can consider the gravitational interaction of masses in the universe as the interaction of defects in elastic ether . point masses and cosmic strings @xcite then correspond to point defects ( vacancies and impurities ) and wedge dislocations . in this interpretation of gravity , we have a question about the elastic gauge , which has direct physical meaning in the geometric theory of defects . if we take the standpoint of the theory of defects , then the elastic properties of ether correspond to some value of the poisson ratio , which can be measured experimentally . it seems interesting and important for applications to include time in the considered static approach for describing motion of defects in the medium . such a model is lacking at present . from the geometric standpoint this generalization can be easily performed , at least in principle . it is sufficient to change the euclidean space @xmath4 to the minkowski space @xmath463 and to write a suitable lagrangian quadratic in curvature and torsion , which corresponds to the true gravity model with torsion . one of the arising problems is the physical interpretation of the additional components of the vielbein and lorentz connection that contain the time index . the physical meaning of the time component of the vielbein @xmath464 is simple this is the velocity of a point of the medium . this interpretation is natural from the physical standpoint because the motion of continuously distributed dislocations means a flowing of the medium . in fact , the liquid can be imagined as the elastic media with a continuous distribution of moving dislocations . this means that the dynamical theory of defects based on the riemann cartan geometry must include hydrodynamics . it is not clear at present how this can happen . physical interpretation of the other components of the vielbein and the lorentz connection with the time index also remains unclear . the author is sincerely grateful to i v volovich for numerous discussions on the problems considered in this paper . this work is supported by the rfbr ( grant 05 - 01 - 00884 ) and the program of support for leading academic schools ( grant as-1542.2003.1 ) . in the appendix , we briefly present the main facts from differential geometry and introduce the notation used in this review . the description is given in local coordinates , which is sufficient for our purposes . we recommend @xcite as a textbook on differential geometry . if the metric @xmath465 and the affine connection @xmath466 are given , then we say that the geometry is defined on a differentiable manifold @xmath467 , @xmath468 . we assume that all fields on the manifold are given by smooth @xmath469 functions except , possibly , some singular points , and do not stipulate that in what follows . we also assume that the manifold @xmath467 is topologically trivial , i.e. , diffeomorphic to the euclidean space @xmath470 . in a local coordinate system @xmath37 , @xmath471 , the metric is given by a nondegenerate symmetric covariant second rank tensor @xmath401 , which defines the scalar product of vector fields @xmath472 , @xmath473 @xmath474 in general , the scalar product may be not positive definite . if the scalar product is positive definite , we say that a riemannian metric is given on a manifold . by definition , the metric is a covariant tensor field , i.e. , it transforms under coordinate changes @xmath41 by the tensor law @xmath475 this means that the scalar product of vector fields @xmath476 is a scalar field . in a similar way , contractions with the metric @xmath156 and its inverse @xmath157 , @xmath477 , allows one to build scalar fields from higher - rank tensors or to lower their rank . the metric is also used for lowering and raising of tensor indices . an affine connection on a manifold in a local coordinate system is given by the set of coefficients @xmath478 that transform under the diffeomorphisms as @xmath479 these coefficients do not constitute a tensor field because of the presence of an inhomogeneous term in ( [ econts ] ) . an affine connection on a manifold defines covariant derivatives of tensor fields . in particular , covariant derivatives of a vector field and 1-form @xmath480 have the form @xmath481 the covariant derivative of a scalar field coincides with the partial derivative @xmath482 . the covariant derivative of higher - rank tensors is built in a similar way and contains one term with the plus and minus sign for each contravariant and covariant index , respectively . one can easily check that the covariant derivative of a tensor of an arbitrary type @xmath483 is a tensor field of type @xmath484 , i.e. , it has one additional covariant index . we note that the covariant derivative of a product of tensor fields may contain arbitrary contractions of indices . for example , @xmath485 because the inhomogeneous term in ( [ econts ] ) is symmetric in indices @xmath486 and @xmath487 the antisymmetric part of the affine connection @xmath488}{}^\rho$ ] forms a tensor field of type @xmath489 , which is called the torsion tensor @xmath490 in general , the connection @xmath491 has no symmetry in its indices and is not related to the metric @xmath156 in any way because these notions define different geometric operations on a manifold @xmath467 . we then say that the affine geometry is given on @xmath467 . we emphasize that the metric and the affine connection are defined arbitrarily and are completely independent geometric notions . therefore , in the construction of physical models , they can be considered independent fields having different physical interpretations . if the affine geometry is given on a manifold , then we can construct the nonmetricity tensor @xmath492 that is by definition equal to the covariant derivative of the metric : @xmath493 by construction , the nonmetricity tensor is symmetric with respect to the permutation of the last two indices : @xmath494 . we note that we need both objects , the metric and the connection to define the nonmetricity . thus , for a given metric and connection we constructed two tensor fields : the torsion and the nonmetricity tensor . we prove that for a given metric , torsion , and nonmetricity tensor we can uniquely reconstruct the corresponding affine connection . equation ( [ enonme ] ) can always be solved for the connection @xmath466 . indeed , the linear combination @xmath495 yields the expression for the affine connection with all lowered indices : @xmath496 the right hand side of this equality is symmetric in the indices @xmath19 and @xmath436 except one term @xmath497 , and this is in accord with the definition of the torsion tensor ( [ etorsi ] ) . thus , to define the affine geometry on a manifold @xmath467 , it is necessary and sufficient to define three tensor fields : metric , torsion , and nonmetricity . we stress once again that all three object can be specified in a completely arbitrary way , and they can be considered different dynamical variables in models of mathematical physics . it is easy to compute the number of independent components of connection , torsion , and nonmetricity tensors : @xmath498=m^3,~~~~~~ [ t_{\mu\nu}{}^\rho]=\frac{m^2(m-1)}2,~~~~~~ [ q_{\mu\nu\rho}]=\frac{m^2(m+1)}2.\ ] ] this shows that the total number of independent components of the torsion and nonmetricity tensors equals the number of components of the affine connection . in the attempt to unite gravity and electromagnetism , h. weyl considered the nonmetricity tensor of a special type @xcite @xmath499 where @xmath500 is the weyl form identified with the electromagnetic potential ( here , torsion was assumed to be identically equal to zero ) . we say that the _ riemann cartan weyl geometry _ is defined on a manifold if there are given a metric , torsion , and nonmetricity of special type ( [ ewevec ] ) . if the nonmetricity tensor is identically zero , @xmath501 , but the metric and torsion are nontrivial , then we say that the _ riemann cartan geometry _ is given on a manifold . as a consequence of ( [ elicon ] ) , the affine connection is defined uniquely by the metric and torsion in this case : @xmath502 this connection is called metrical because the covariant derivative of the metric is identically equal to zero : @xmath503 the metricity condition provides commutativity of covariant differentiation and the raising and lowering of indices . if both nonmetricity and torsion tensors are identically equal to zero , @xmath501 , @xmath505 , and the metric is nontrivial , then we say that _ riemannian geometry _ is given on a manifold . in this case , the metrical connection is symmetric with respect to the first two indices and uniquely defined by the metric @xmath506 this connection is called the _ levy - civita connection _ or _ christoffel symbols_. we use the tilde to denote the geometrical objects constructed only for the metric and zero torsion and nonmetricity , i.e. , in a ( pseudo-)riemannian geometry . the prefix pseudo- is used when the metric is not positive definite . from the expression for christoffel symbols ( [ etigam ] ) , we see that @xmath507 this means that for the christoffel symbols to be zero in some coordinate system it is necessary and sufficient that the metric components be constant in these coordinates . in another coordinate system , they may be nontrivial because christoffel symbols are not components of a tensor . for example , the christoffel symbols for euclidean space are zero in the cartesian coordinates but differ from zero , e.g. , in spherical or cylindrical coordinate systems . in the case where the nonmetricity tensor and torsion are identically zero , and there is a coordinate system in which the metric is equal to the diagonal unit matrix in the neighborhood of every point , and , consequently , the christoffel symbols are zero , the geometry is called locally euclidean . the corresponding coordinate system is called _ we say that a vector field @xmath472 is transported parallel along a curve @xmath346 , @xmath508 , if @xmath509 where @xmath510 is the tangent vector to the curve . multiplying this equation by @xmath511 we obtain @xmath512 where @xmath513 . we say that under the parallel transport from a point @xmath514 to a neighboring point @xmath515 , the vector @xmath516 acquires a differential @xmath517 , which is linear in @xmath518 and the components of the vector field . in a similar way , we define parallel transport of arbitrary tensor fields along a curve @xmath346 . the result of parallel transport between two points depends in general on a curve connecting these points . we note that parallel transport along a curve is defined only by an affine connection and has no relation to a metric . there are two types of distinguished curves @xmath346 in an affine geometry : geodesics and extremals . a geodesic line is a curve such that a vector tangent to it remains tangent under parallel transport along the curve . with the parameter @xmath358 along the curve chosen canonically , a geodesic line is defined by the system of ordinary nonlinear differential equations @xmath519 although a geodesic line is defined only by the symmetric part of the affine connection @xmath520 , the latter depends nontrivially on torsion and nonmetricity . explicit expression ( [ elicon ] ) yields @xmath521 the second type of distinguished curves in an affine geometry are extremals or lines of extremal length connecting two points . these lines are exclusively defined by the metric . with the parameter along the curve chosen canonically , extremals are defined by an equation similar to ( [ egeoeq ] ) , @xmath522 however , we now have not a general affine connection on the right - hand side but christoffel symbols constructed only from the metric . we see that geodesics and extremals are in general different curves on a manifold . in a riemann cartan geometry , geodesics and extremals coincide if and only if the torsion tensor is antisymmetric in all three indices . in riemannian geometry , geodesics and extremals always coincide . the primary role in differential geometry is played by the curvature tensor of the affine connection , which arises in different contexts . in local coordinates , it is defined by the expression @xmath523 where the parenthesis @xmath524 denote the previous terms with the permuted indices @xmath19 and @xmath436 . it can be easily verified that curvature ( [ ecurva ] ) is indeed a tensor field . we note that the curvature tensor has no relation to a metric and is defined entirely by the connection . in an affine geometry , the curvature tensor with all of its indices lowered has no symmetry under permutations of indices except antisymmetry in the first two indices . contraction of the curvature tensor in the two indices yields the ricci tensor @xmath525 , which is also defined entirely by the connection . in riemannian geometry , the ricci tensor for the levi - civita connection is symmetric with respect to the permutation of indices : @xmath526 . in a riemann cartan geometry , this symmetry is generally absent for a nonzero torsion tensor . the affine connection is called locally trivial if in the neighborhood of every point one can choose the coordinate system in which all components of the connection are zero . now we formulate two important theorems . the proof of this theorem is reduced to an analysis of the transformation rule for the affine connection components in ( [ econts ] ) . if in the new coordinate system the symmetric part of the connection components is equal to zero , then the system of differential equations for the transition functions appears . the local integrability condition for this system of equations is provided by the equality of the curvature tensor to zero . [ tlocev]if for a given affine geometry the torsion , nonmetricity , and curvature tensors are equal to zero on the whole manifold , then this manifold is isometric either to the whole ( pseudo-)euclidean space @xmath470 or to the quotient space @xmath528 , where @xmath529 is a discreet transformation group acting freely . the riemann cartan geometry defined by the metric and torsion provides the basis for the geometric theory of defects . in the analysis of such models , it is more convenient to introduce the cartan variables : the vielbein and @xmath530 connection instead of the metric and torsion . we assume here that a metric on a manifold is positive definite . if a metric were not positively definite then the @xmath531 connection would appear , where @xmath532 . for a riemannian metric , the orthonormal vielbein @xmath533 , @xmath534 , is defined by the system of quadratic equations @xmath535 where @xmath16 is the kronecker symbol . this relation uniquely defines metric for a given vielbein . conversely , system of equations ( [ etetra ] ) defines the vielbein for a given metric up to local @xmath530 rotations . we distinguish greek and latin indices because different transformation groups act on them . from definition ( [ etetra ] ) , we see that @xmath536 . components of the inverse vielbein @xmath537 , @xmath538 , can be considered components of @xmath425 orthonormal vector fields @xmath539 on a manifold @xmath467 with respect to the metric @xmath465 @xmath540 components of the vielbein @xmath55 define @xmath425 orthonormal 1-forms @xmath541 on a manifold . it is known that any manifold @xmath467 can be equipped with a riemannian metric ( see , for e.g. , @xcite ) . at the same time , the global existence of a vielbein provides , in particular , an orientation of the manifold @xmath467 . thus , a vielbein may exist globally only on orientable manifolds . there are also other topological restrictions for the global existence of a vielbein that we do not discuss here . the coordinate basis @xmath542 of the tangent space in every point of @xmath467 is called holonomic . components of tensors of an arbitrary type may also be considered with respect to the unholonomic basis @xmath313 of tangent spaces defined by the vielbein . for example , a vector field has the components @xmath543 we always assume that the transformation of greek indices into latin ones and vice versa is performed with the help of the vielbein . we now define the @xmath530 connection @xmath544 . this is the name of the connection of the principal fiber bundle with the structure lie group @xmath530 and the base @xmath467 . if the riemann - cartan geometry is given on @xmath467 , and the vielbein is defined , then we can define the @xmath530 connection by the relation @xmath545 we see that under coordinate changes , the components of @xmath546 transform as a covector field in the greek index . thus , they define a 1-form on @xmath467 . for a given vielbein , this relation provides a one - to - one correspondence between components of the connections @xmath491 and @xmath546 . the @xmath530 connection defines covariant derivatives for components of tensor fields relative to an unholonomic basis . for example , @xmath547 the covariant derivative is naturally defined for a tensor field having both greek and latin indices . taking the covariant derivative of eqn ( [ etetra ] ) leads to the antisymmetry of the components @xmath548 . this means that the 1-form @xmath549 takes values in the lie algebra @xmath550 , and this indeed corresponds to the @xmath530 connection . in the general case of an affine geometry , the cartan variables can also be defined by relations ( [ etetra ] ) and ( [ etecov ] ) . for this , we have to replace the kronecker symbol on the right - hand side of ( [ etetra ] ) with an arbitrary nondegenerate symmetric matrix @xmath551 . in this case , relation ( [ etecov ] ) defines a linear @xmath552 connection . in the riemann cartan geometry with the lorentzian - signature metric , we would have the lorentz @xmath553 connection . we consider local rotations with a matrix @xmath554 to show that the components @xmath546 do define an @xmath530 connection in the riemann cartan geometry . by definition , the components of vector fields and 1-forms transform under local rotations according to the rule @xmath555 in order that covariant derivatives ( [ ecodve ] ) have the tensor transformation rule under local rotations , it is necessary and sufficient that the components of the @xmath530 connection transform according to the rule @xmath556 this is the transformation law for the @xmath530 connection , indeed . the same transformation law follows from definition ( [ etecov ] ) if the vielbein @xmath55 transforms as a vector with respect to the index @xmath171 , and the christoffel symbols remain unchanged . of course , we can introduce the metrical and @xmath530 connections on @xmath467 in an independent way . if we require afterwards that the @xmath530 connection acts on tensor field components relative to an unholonomic basis defined by the vielbein , we obtain a one - to - one correspondence between the components of the connections ( [ etecov ] ) . we now express the components of the metric connection @xmath491 through the vielbein @xmath55 and @xmath530 connection @xmath546 with the help of relation ( [ etecov ] ) and substitute them into the definitions of torsion ( [ etorsi ] ) and curvature ( [ ecurva ] ) . as a result , we obtain expressions for torsion and curvature in terms of the cartan variables @xmath557 where @xmath558 torsion ( [ ecurcv ] ) and curvature ( [ ecucav ] ) are 2-forms on the manifold @xmath467 with the values in the vector space and lie algebra @xmath550 , respectively . an @xmath530 connection @xmath546 is called locally trivial if every point has a neighborhood containing this point such that it has the form @xmath559 obviously , after local rotation with the matrix @xmath560 , all components of the connection become zero . this connection is also called a pure gauge . one can easily verify that the curvature of a locally trivial connection identically vanishes . the inverse statement is also valid . this theorem is valid for any structure lie group . the proof is reduced to an analysis of transformation rule ( [ elocta ] ) . if the left - hand side of this equation is zero , then we have a system of equations for the field @xmath561 for which vanishing of curvature tensor is the local integrability condition . the space with zero curvature tensor @xmath163 is called the space of absolute parallelism because the parallel displacement of a vector does not depend on the path connecting two fixed points of the manifold . we can perform a local @xmath530 rotation for the locally trivial @xmath530 connection such that it becomes zero : @xmath562 . then , the zero - torsion equality becomes @xmath563 this equation is the local integrability condition for the system of equations @xmath564 for @xmath425 functions @xmath49 . a solution of this system of equations yields the transition functions to cartesian coordinates . thus , the equalities of curvature and torsion tensors to zero are the necessary and sufficient conditions for the existence of the fields @xmath561 and @xmath49 , i.e. , the existence of such a local rotation and coordinate system where the connection vanishes and the metric becomes euclidean . we note that equality of torsion tensor to zero alone is not enough for the existence of a cartesian coordinate system . three - dimensional space is considered in the geometric theory of defects . we make two remarks concerning this . in lower dimensions , the algebraic structure of the curvature tensor with all lowered indexes becomes much simpler . in two dimensions the full curvature tensor in the riemann cartan geometry is in the one - to - one correspondence with its scalar curvature : @xmath565 in three dimensional space , the full curvature tensor is in the one - to - one correspondence with its ricci tensor @xmath566 these formulas are also correct for nonzero torsion . concluding this appendix , we write the identity that is valid in the riemann cartan geometry in an arbitrary number of dimensions : @xmath567 the riemannian scalar curvature on the right - hand side of this equality is constructed only from the vielbein for zero torsion . i. a. kunin and b. i. kunin . gauge theories in mechanics . in _ trends in application of pure mathematics to mechanics . lecture notes in physics , v.249 . _ , pages 246249 , berlin heidelberg , 1986 . verlag . f. w. hehl , j. d. mccrea , e. w. mielke , and y. neeman . metric - affine gauge theory of gravity : field equations , noether identities , world spinors , and breaking of dilaton invariance . , 258(1&2):1171 , 1995 . v. e. zakharov , s. v. manakov , s. p. novikov , and l. p. pitaevskii . . nauka , moscow , 1980 . ; english transl . : s. p. novikov , s. v. manakov , l. p. pitaevskii , and v. e. zakharov _ the inverse scattering method _ plenum , new york ( 1984 ) . b. a. dubrovin , s. p. novikov , a. t. fomenko . . nauka , moscow , 1998 , fourth edition . [ in russian ] ; english transl . prev . ed . : b. a. dubrovin , a. t. fomenko , and s. p. novikov _ modern geometry : methods and applications , _ part 1 , _ the geometry of surfaces , transformation groups , and fields , _ springer , new york ( 1992 ) .

physics enters biology in two broad and overlapping areas information and mechanics @xcite . this review is about the mechanics , statistical and otherwise , of living matter . this is not a review of soft - matter physics in a biological setting . the aspect of biological matter of interest here is the ability to transduce free energy into systematic movement . this property is the defining characteristic of active matter , and my interest is in the unique mechanical properties that arise as a consequence of such active processes @xcite . this review focusses on the collective behaviour of systems with a large number of constituents , and will thus use the ideas of condensed matter and statistical physics . i will not discuss the physics of molecular motors or the related and vast exclusion - process family of problems @xcite . the aim is to complement and update the perspectives of earlier reviews @xcite and to present the achievements and limitations of this rapidly advancing subfield . i begin with the realizations of active matter to be covered . it is natural for a condensed matter physicist to regard a coherently moving flock of birds , beasts or bacteria as an orientationally ordered phase of living matter . this idea was first implemented @xcite in computer models of interacting particles moving with a fixed speed and trying to align their velocity vectors parallel to those of their neighbours , in the presence of noise . to a student of magnetism these are itinerant , classical , ferromagnetically interacting continuous spins that move in the direction in which they point , which is what makes flocks different from magnets . the models showed a nonequilibrium phase transition from a disordered state to a flock with long - range order @xcite in the particle velocities as the noise strength was decreased or the concentration of particles raised . the nature of order and fluctuations in a flock , the character of the transition to a flock , and the flocking of active particles with an axis but no polarity are among the issues this review deals with . amazingly , the physics of flocking can be imitated by a collection of rods lying on a horizontal surface , agitated vertically @xcite . indeed , vibrated monolayers of macroscopic grains have provided some of the most fruitful realizations of active matter . this review will explain briefly the physics behind underlying this analogy and highlight experiments that exploit it @xcite flocking in a fluid medium introduces physics absent in the simplest flocking models : each swimming creature generates fluid flow which moves and reorients other swimmers far away . including this interaction leads @xcite to a modified liquid - crystal hydrodynamics @xcite , in which the constituent particles carry permanent stresses that stir the fluid . the implications of this interplay of self - propelling activity and fluid flow for order and macroscopic rheology are an important part of this review . the physics of the flocking of organisms in a fluid reappears at a _ _ sub__cellular scale in the cytoskeleton @xcite , the polymeric scaffolding that governs transport , adhesion , movement and division in the living cell . two nonequilibrium processes drive the cytoskeleton : the atp - assisted polymerization and depolymerization known as _ treadmilling _ , which i will not discuss ; and _ contractility _ , illustrated in fig . [ contractfig ] , which arises from the atp - driven movement of motor proteins in specific directions on biofilaments , and is central to this review . the cytoskeleton is thus a suspension of filaments endowed with active internal forces . moreover , there are natural mechanisms that promote the alignment of neighbouring filaments , through excluded volume as well as activity . we must therefore allow for the possibility of orientational order . it is then not surprising that the hydrodynamic equations obtained in the active - gel description @xcite of the cytoskeleton have precisely the same form as those @xcite for collections of swimming organisms , ignoring complications such as permanent crosslinking . a condensed - matter physicist accustomed to symmetry arguments should find it reassuring that identical hydrodynamic descriptions apply for a @xmath0 km fish shoal and a @xmath0 @xmath1 m cytoskeleton . the viewpoint of this review is that living matter can fruitfully be regarded as a kind of material @xcite and studied using the tools of condensed matter physics and statistical mechanics ; that there is a practical way to encode into such a description those features of the living state that are relevant to materials science ; and that the results of such an endeavour will help us better understand , control and perhaps mimic active cellular matter . a community of respectable size has grown around this activity , approaching the problem at different length scales and with a variety of techniques . what do they , and we , hope to gain from this enterprise ? first , active matter is condensed matter in a fundamentally new nonequilibrium regime : ( i ) the energy input takes place directly at the scale of each active particle , and is thus homogeneously distributed through the bulk of the system , unlike sheared fluids or three - dimensional _ bulk _ granular matter , where the forcing is applied at the boundaries . ( ii ) self - propelled motion , unlike sedimentation , is force - free : the forces that particle and fluid exert on each other cancel . ( iii ) the direction of self - propelled motion is set by the orientation of the particle itself , not fixed by an external field . indeed , these can be taken as the practical defining properties of active matter . a comprehensive theory of this ubiquitous type of condensed matter is a natural imperative for the physicist , and should yield a catalogue of the generic behaviours , such as nonequilibrium phases and phase transitions , the nature of correlations and response , and characteristic instabilities . second , therefore , the generic tendencies emerging from the theory of active matter , unless suppressed by specific mechanisms , must arise in vivo , which is why biologists should care about it . last , if we can understand active matter , perhaps we can manufacture faithful imitations of it in chemomechanical systems without components of biological origin @xcite the reader should keep in mind that theories of active matter were formulated not in response to a specific puzzle posed by experiments but rather to incorporate living , metabolizing , spontaneously moving matter into the condensed - matter fold . this was done through minimal models whose consequences are relatively easy for theoreticians to work out . natural realizations of living matter are far from minimal , so comparisons of active - matter theory with experiment are likely to be qualitative until well - controlled model systems are devised . although this is not a historical review , it must be stated here that the active - matter idea is not new . in a prescient article , finlayson and scriven @xcite argued that biological matter could display hydrodynamic instabilities driven by stresses arising from metabolic activity in the bulk of a fluid . unlike in this review , they build active stress - tensor contributions from gradients of existing scalars such as concentration and temperature . they point out that the field poses `` challenges to theoretical rheology '' and is `` virgin territory for experimental research in the physics of complex fluids '' . they conclude that `` continuum .. analysis ... of instability by active stress is probably .... destined to play a large part in understanding some of the engines operating at the cellular level in living systems '' . the developments surveyed in this article prove them right . the body of this review consists of several sections each treating a specific example of an active - matter system . in section [ drypolar ] the simplest models of moving flocks , without solvent flow , are introduced and some key properties discussed . flocks without macroscopic polarity and hence no migration velocity are surveyed in section [ dryapolar ] . the theoretical framework for collective self - propulsion in a fluid medium is found in section [ activesusp ] , with a discussion of general symmetry - based approaches , and the active mechanics of the cytoskeleton in section [ cytohydro ] . non - biological self - driven particles are the subject of section [ artificial ] . the article closes with section [ outlook ] which summarizes the achievements and limitations of the approach adopted here to the modelling of active matter and attempts to set a course for future experiments and theory in this vital and rapidly evolving field . arrows are _ polar _ objects , uniform cylinders are _ apolar_. a flock moving in one direction has polar order , whereas equal numbers of ants moving from south to north and from north to south are macrscopically apolar even though the individual ants are polar . we begin by discussing polar ordered states . in the simplest models @xcite of ordering transitions of active particles , the medium through which the particles move is treated as an inert substrate . each particle carries a velocity vector of fixed magnitude , which it rotates to point parallel to the mean of its neighbours velocities , with an angular tolerance @xmath2 . it then takes a small step in the direction of the updated vector . like the continuous - spin magnets that they resemble , the vicsek family of models display a well - defined phase transition from a disordered phase to a coherent flock as @xmath3 is decreased or the number density is increased @xcite . _ _ un__like planar spins @xcite , flocking models show true long - range order even in two dimensions . how they do so was shown by toner and tu @xcite , whose continuum field - theoretic approach , which amounts to a coarse - graining of the vicsek rules , we sketch below . a coarse - grained dynamical description works with `` slow variables '' @xcite , whose relaxation times increase unboundedly with increasing wavelength . for the vicsek model these are @xcite the number density field @xmath4 , since total number is conserved , and the velocity field @xmath5 , whose fluctuations transverse to the mean ordering direction are the spin - waves or nambu - goldstone @xcite modes of the flock @xcite . with the slow variables in hand , toner and tu @xcite write down the equations of motion , incorporating the physics of the vicsek model , on general grounds of symmetry . first , the dynamics of @xmath5 is described in the equation below : [ vecop ] _ t * p * + + ... = ( - ) * p * + - p(c ) + * f * ignoring all gradients in equation [ vecop ] , a phase transition from the isotropic state @xmath6 to an ordered flock with @xmath7 , spontaneously breaking rotation invariance , should occur as the control parameter @xmath8 is tuned from negative to sufficiently large positive values ; @xmath9 , in general a tensor , controls the elastic @xcite restoring forces to distortions in the ordered phase . the random error @xmath2 of the vicsek update rule is summarized in a nonconserving , gaussian noise @xmath10 , uncorrelated in space and time . the crucial difference from traditional models of dynamic critical phenomena @xcite lies in the navier - stokes - like @xmath11 term @xcite , which says that distortions in @xmath5 are advected by @xmath5 , because @xmath5 is not only an order parameter but a velocity . @xmath12 , a general increasing function of concentration , then embodies the equation of state for the pressure . second , c evolves through the continuity equation [ conteqgen ] _ t c + c * p * = 0 , because @xmath13 is the velocity field of the particles . equations [ vecop ] and [ conteqgen ] can be viewed as the dynamics of a fluid with a preferred speed relative to a background medium , or of a magnet whose spin is a velocity . in the second interpretation , the nonequilibrium character of the toner - tu model enters only through the advective @xmath11 term in equation [ vecop ] and the current in equation [ conteqgen ] . the right - hand side of equation [ vecop ] can be written as @xmath14 plus noise , with a free - energy functional [ polarfree ] f[*p * ] = d^dr [ -2*p * + 4 ( * p * ) ^2 + 2 + * p * p(c ) ] . of ginzburg - landau form , in which the `` pressure gradient '' @xmath15 appears as an orienting field for @xmath5 . in the absence of @xmath11 , the average dynamics is downhill in @xmath16 . a flock governed by equations [ vecop]-[polarfree ] is a strange blend of magnet and fluid . let us rediscover some of its unique properties @xcite , all consequences of the fact that the order parameter is itself a velocity . _ sound modes _ : let us linearize equations [ vecop ] , [ conteqgen ] about a uniform ordered state with @xmath18 , @xmath19 , denote directions along and transverse to @xmath20 by @xmath21 and @xmath22 , and see how small disturbances @xmath23 , @xmath24 travel . as in heisenberg or xy magnets @xcite , we can see from equation [ vecop ] that @xmath25 relaxes rapidly , so that on long timescales @xmath26 . if we take @xmath27 in equation [ vecop ] , a shift to a frame moving with speed @xmath28 eliminates advective terms from equations [ vecop ] and [ conteqgen ] in this linearized analysis without losing the essential physics . to leading order in gradients , and ignoring the noise , equations [ vecop ] and [ conteqgen ] then become [ conceomflock ] _ t c = - c_0 _ _ , [ poleomflock ] _ t _ = -p(c_0 ) _ c leading to propagating modes with frequency [ flockdisp ] = q _ at wavevector @xmath29 . no , we havent rediscovered normal sound waves . long - wavelength sound in a fluid propagates because of momentum conservation , which a flock lacks . the `` sound modes '' in a flock are a consequence of spontaneously broken rotation invariance : they propagate at long wavelength because @xmath30 is not damped at zeroth order in wavenumber . _ giant number fluctuations _ : @xmath30 is the nambu - goldstone mode @xcite of broken rotation invariance , easily excited and slow to decay at small wavenumber @xmath31 . hence when noise is included the steady - state variance @xmath32 of the @xmath31th fourier component should diverge at small wavenumber @xmath31 , like spin waves in an xy magnet @xcite or director fluctuations in a nematic @xcite . but equations [ conceomflock]-[poleomflock ] imply that , for frequencies and wavenumbers related by equation [ flockdisp ] , @xmath33 . therefore , the variance of @xmath23 should also diverge at small @xmath31 . in real space this means that for regions with @xmath34 particles on average , the variance in the number grows faster than @xmath34 : a flock should display giant number fluctuations . _ long - range order in @xmath17 _ : equations [ vecop ] and [ conteqgen ] when linearized yield @xmath35 . taken literally this would mean only quasi - long - range order in dimension @xmath36 , by analogy with the xy model . however , nonlinearities in equations [ vecop ] and [ conceomflock ] are stongly relevant in @xmath36 . and lead to @xmath32 diverging more slowly than @xmath37 for most directions of @xmath38 , thus preserving long - range order @xcite . here is a qualitative explanation of how this happens . consider a flock with @xmath39 at a given instant . because @xmath5 is a velocity , a long wavelength fluctuation @xmath30 will allow creatures in the flock to exert their orienting influence on regions out to a distance @xmath40 in a time @xmath41 . this effective enhancement of the range of influence ultimately leads @xcite to a suppression of orientational fluctuations on large scales , and thus to long - range order in @xmath36 . the effect is absent in @xmath42 , where rotations are a discrete symmetry and there _ is _ no @xmath30 . we will return to this point in section [ 1dflocks ] . i have not discussed microscopic derivations of the toner - tu equations @xcite , dynamics near the onset of polar ordering @xcite , and the application of flocking models @xcite to motor - microtubule extracts @xcite . two types of tests of the physicist s approach to flocking are required . experiments on real flocks are of course the real test . however , it is equally important to check the predictions of coarse - grained theories against computer experiments on microscopic models . we discuss both below , although measurements of the phase - transition variety on real flocks are scarce . quantitative agreement has been found between numerical `` experiments '' on microscopic models and the predictions of the coarse - grained theory , including long - range order in @xmath36 , the form of the propagating modes , anomalous density fluctuations and superdiffusion of tagged particles @xcite . i am not aware of any laboratory or field - based attempts to measure ordered - phase correlation functions for flocks on a substrate . although some studies @xcite claim to see a continuous onset of the ordered phase , as a mean - field treatment of equation [ vecop ] or analogy to continuous - spin magnets would suggest , very large - scale numerical studies of the vicsek model and its relatives show that the transition is in general a discontinuous one , characterized by a complicated coexistence @xcite . see also ref . @xcite . in numerical studies of the vicsek model in one space dimension , long domains of coherently moving particles do appear at low noise and high density . it is occasionally claimed @xcite that this is a true phase transition in the limit of infinite system size , as occurs in vicsek models in dimension @xmath43 . recall that the @xmath44 ising model fails to order because of the proliferation of kinks @xcite . does some magic suppress kinks , i.e. , velocity flips , in @xmath44 flocks ? raymond and evans @xcite estimate the lifetime @xmath45 of a @xmath44 flock of length @xmath46 as the mean time for the appearance of a kink . they show that if @xmath45 is to grow at least as a power of @xmath46 , the number of particles per site must grow , artificially , at least as @xmath47 . accepting the constraint of a finite number density yields a finite lifetime and correlation length for the would - be flock . ultimately , nothing saves @xmath44 flocks from the fate of the @xmath44 ising model because flocking in @xmath44 breaks a _ discrete _ symmetry and is thus not accompanied by nambu - goldstone modes whose long range is responsible for the rescue act in @xmath36 . nevertheless , lowering the noise strength or raising the density at fixed @xmath46 will eventually yield a system with correlation length larger than @xmath46 , and a bistable time - series of the flocking order parameter , i.e. , a finite transient flock with a large lifetime . precisely this phenomenon has been seen in experiments , to which we turn next . 1 . couzin @xcite studies the one - dimensional flocking of desert locusts on an annular track . the onset of coherent motion seen in the experiments as the number of locusts is increased , and the bistable time - series switching between clockwise and counterclockwise circuits of the track are well described by the @xmath44 vicsek model @xcite , to which the paper makes a comparison . one must not of course conclude see reference @xcite and section [ 1dflocks ] that there is a true phase transition here . all the same , the experiment demonstrates convincingly the relevance of simple flocking models to an understanding the behaviour of real organisms . szab @xcite created a polar flock in a petri dish by using motile keratocytes extracted from fish scales . as the cells divide and the areal density of cells crosses a threshold , there is a well - defined onset of a state of macroscopic order in the orientations and velocities of the keratocytes , that can plausibly be interpreted as a flocking phase transition , which should be governed by the local physics of the models we have been discussing , as the cells interact only by contact . the data quality makes it impossible to draw conclusions regarding the order of the transition . the authors make further inferences on the transition from a computational model , not from data . despite limitations the experiment is valuable as a realization of the flocking transition , with potential for higher - precision studies including finite - size scaling . there is a surprising link between vibrated granular monolayers and self - propulsion @xcite . polar rods @xcite , placed on a vertically vibrated horizontal linear track , move on average with small end forward . systematic unidirectional motion of apolar rods @xcite was effected by maintaining them permanently tilted , by overfilling an annular channel with rods . the propulsion mechanism @xcite is illustrated in the cartoon fig . [ shakenrod ] . + + the reader is encouraged to check that this system satisfies all three defining properties of active matter in section [ activematerialsub ] . published studies so far @xcite of polar rods at high coverage on a vibrated surface report order and anomalous density fluctuations on a local scale , and an abrupt flight to the boundary when the vibration amplitude crosses a threshold , but not a macroscopically ordered flock . the case of apolar rods @xcite is discussed in section [ subsub : dryapolarexps ] . at least one example is known @xcite of living , pulsating elongated cells forming a state with spontaneous , uniaxial orientational order with a macroscopic axis @xmath48 with @xmath49 symmetry . this is the spatial symmetry of nematic liquid crystals , and it guarantees that the mean macroscopic velocity of such a `` flock '' is zero . and yet active nematics are profoundly different from dead ones . in particular , here as in polar flocks , giant number fluctuations are predicted @xcite even though the order parameter for the nematic is _ not _ a velocity . below is an explanation of how it works . in the presence of noise , long wavelength fluctuations of the axis of orientation the broken - symmetry mode are abundant . working in two dimensions for simplicity , where distortions are described by a single angle @xmath50 , we see ( fig . [ curvcurrfig ] ) that bend or splay produces a _ polar _ configuration . the absence of time - reversal invariance in a driven state , together with the polarity , means @xcite that particles in the curved region must drift in a direction defined by the curvature as pictured in fig . [ curvcurrfig ] , leading to a current of particles @xmath51 with @xmath52 in two dimensions . in steady state , this active flux must be balanced on average by restoring diffusive currents @xmath53 . thus fluctuations in @xmath4 and @xmath50 must be of the same order . orientational fluctuations , being a broken - symmetry mode , should have a variance of order @xmath54 at wavenumber @xmath31 . therefore , number density fluctuations @xmath55 as well which means the number fluctuations in regions containing @xmath34 particles on average have a standard deviation [ gnf ] n n^1/2 + 1/d ~n d = 2 for large @xmath34 . unlike for polar flocks @xcite , these results of the linearized theory for apolar systems survive _ quantitatively _ when nonlinearities are included @xcite . the first test of the existence of giant number fluctuations ( gnfs ) in active nematics was a computer experiment on a clever apolar generalization @xcite of the two - dimensional vicsek model described in section [ vicsekmodel ] . each particle is assigned an _ axis _ , rather than a unit vector , which it aligns parallel to the mean of its neighbours axes , subject to a small angular noise . the particle then takes a small step preferentially along the axis , forward or back . since the orientation variable and the moves are all defined modulo @xmath56 , the model has no polarity . as noise or mean interparticle separation is decreased , a continuous transition is seen from an isotropic state to a phase with quasilong - range nematic order of the particle axes as in equilibrium @xcite . however , the density fluctuations are giant : @xmath57 grew far more rapidly than @xmath34 and could be fitted to a form @xmath58 in precise agreement with the predictions @xcite whose derivation was sketched above . real - space snapshots showed strong transient inhomogeneity , with a banded structure . if the anisotropic , detailed - balance - violating move is replaced by an isotropically distributed random step , the nematic phase survives , but with normal number fluctuations , because detailed balance now holds . s. mishra @xcite has shown that coarse - graining the model of chat @xcite yields the equations of motion of ref . @xcite , with the active current as in section [ activecurr ] . further numerical studies on a particle model @xcite show a close connection between the giant number fluctuations of an active nematic and the phenomenon of fluctuation - dominated phase ordering @xcite . 1 . _ melanocyte nematics : _ living melanocytes , the cells that spread pigment in skin , have been shown @xcite to form apolar , nematic order in vitro as concentration is increased . the resulting @xmath17 nematic phase has large regions of oriented cells with occasional point topological defects of strength @xmath59 which reinforce the nematic interpretation @xcite . although the authors of @xcite emphasize that their systems are very far from thermal equilibrium the analysis they bring to bear on the problem , especially in the apolar case , is largely inherited from the equilibrium theory of liquid crystals . the focus in @xcite is on modelling cell orientations , in a population - averaged description which can not resolve spatial variations in the density . consistent with this limitation , their experiments do not look at the behaviour of density fluctuations . the purpose of this discussion is to urge experimenters to take another look at the melanocyte system , especially as the strange properties of active nematics @xcite remain to be checked in a biological experiment . 2 . _ active apolar states in granular matter : _ + + narayan @xcite carried out a systematic study of liquid - crystalline order in a vertically agitated monolayer of copper - wire segments . in particular , they measured number density fluctuations in the nematic and the isotropic phases . their findings @xcite confirmed the predictions @xcite of giant number fluctuations in the nematic phase . a typical nematic configuration and a plot of standard deviation _ vs _ mean number are shown in figs . [ nemimagelow ] and [ gnflow ] . + + an additional consequence of the theory that the autocorrelation of the local density should decay as @xmath60 over a large time range @xcite was also confirmed @xcite . lastly , single - particle tracking showed that the rods moved preferentially along their length just like the apolar flockers in the model of chat @xcite . + other experiments of interest , on chiral structures in active granular matter , include the work of tsai @xcite and of blair @xcite . the measurements on granular matter reported above are among the few quantitative tests of active - matter theories . that they are not carried out on living systems only serves to reinforce the universality of the approach reviewed in these pages . the collective motion of swimming organisms can be understood by constructing the coupled dynamical equations for the swimmer concentrations and orientations , and a generalized navier - stokes equation for the velocity field @xmath61 of the suspension . self - propelling activity enters via force densities in the navier - stokes equation . these have no monopole moment , since the mutual forces of swimmer and fluid cancel by newton s 3rd law @xcite . the minimal model of an active particle in a fluid is therefore a permanent force dipole ( see fig . [ bactalg ] ) , whose strength @xmath62 is positive for extensile and negative for contractile swimmers [ see also the modelling of ion pumps in a model membrane by manneville @xcite . ] associating a local orientational order parameter field @xmath63 with the vectorial orientation of the swimmers in a coarse - graining cell around @xmath64 at time @xmath41 , a straightforward calculation @xcite shows , to leading order in a gradient expasion , that a collection of active particles at concentration @xmath65 has an active force density @xmath66 with an active stress [ sigact ] ^a & = & w c(*r*,t ) * p * * p * + & = & w c(*r*,t ) ( * r*,t ) + w 3 c p^2 , where @xmath67 is the unit tensor , and @xmath68 , the trace - free part of @xmath69 , determines the deviatoric stress which is all that concerns us in the incompressible limit @xmath70 . activity thus produces flow , and flow in turn reorients the principal axes of rodlike particles preferentially along the extensional axis ( and platelike along the compressional axis ) , as it would in non - living liquid crystals as well . the resulting equations of motion for @xmath5 are obtained by adding flow - orientation couplings to equation [ vecop ] : [ vecopwet ] d_t * p * + + .... = - f + * f * , where @xmath71 is the time - derivative in a frame locally _ comoving _ and _ corotating _ with the fluid , the coefficient @xmath72 governs the orienting effect of the extensional part of the flow through the term in @xmath73/2 $ ] , and the remaining terms are as defined for equation [ vecop ] , @xmath16 being the free - energy of equation [ polarfree ] . inertia , viscosity @xmath2 , elastic forces from equation [ polarfree ] , and activity compete to determine the dynamics of the suspension velocity field : [ nsactive ] ( _ t + * u * ) * u * = -(^a + ^h ) -^2 * u * - ; = 0 , with incompressibility @xmath70 enforced by the pressure @xmath74 . in equation [ nsactive ] , @xmath75 is the contribution to the stress from equation [ polarfree ] for the free - energy functional @xcite . lastly , the current in the continuity equation for the concentration , in the lab frame , is @xmath76 , because we can simply identify the orientation field @xmath5 with the active - particle velocity with respect to the fluid . we turn next to the dramatic consequences of this dynamics . active suspensions with uniform orientational order are in a state of permanent uniaxial tension or compression and are therefore intrinsically unstable @xcite : an individual contractile particle ( fig . [ bactalg ] ) pulls fluid in from both ends along its main axis . in a perfectly ordered , unbounded , parallel collection of contractile particles , these self - generated flows cancel . a long - wavelength splay fluctuation disrupts this delicate balance , resulting in flows upward on one side and downward on the other , as shown in fig . [ splaybendinstab ] . the ensuing shear will amplify the rightward tilt of the middle portion , hence the instability . the same argument , _ mutatis mutandis _ , implies a _ bending _ instability of extensile filaments @xcite . a detailed solution @xcite of equations [ vecopwet ] , [ nsactive ] bears out this pictorial argument . note that we ve assumed instant , global response of the fluid flow , which amounts to the steady stokesian approximation in which inertia and acceleration are ignored in equation [ nsactive ] . in that limit , in an unbounded system with mean concentration @xmath77 , the growth rate of the instability must be proportional to the only available inverse timescale , namely , the ratio @xmath78 of active stress to viscosity . in a container with smallest dimension @xmath46 , active stresses compete @xcitewith orientational elasticity from equation [ polarfree ] , resulting in a threshold @xmath79 for the instability . far beyond threshold , scaling at wavenumber @xmath31 yields a growth rate @xmath80 for @xmath81 , crossing over to @xmath78 for @xmath82 . the detailed form of the instability depends on the experimental geometry . in ref . @xcite , for example , it appears as an active version of the freedericksz transition of nematic liquid crystals @xcite . novel instability mechanisms recently noted in polar systems include oscillations and banding if a concentration field is included @xcite , and travelling undulations if a deformable free surface is present @xcite . aranson @xcite study free - standing active thin films and introduce _ ad hoc _ a propulsive force simply proportional to the thickness - averaged @xmath5 . such a term can be obtained from the treatment of this section and ref . @xcite if @xmath5 is neither parallel nor perpendicular to the free surface . the stabilizing effect of shear has been discussed by muhuri @xcite . there is detailed evidence for the basic instability in numerical studies of particles self - propelled through a fluid @xcite , who also point out a variety of instabilities in the isotropic phase . studies of the pdes of active hydrodynamics include refs . @xcite who show that the linear instability leads to complex flow patterns and turbulence driven not by reynolds number but an active ericksen number @xmath83 where @xmath84 is the collective relaxation time of the @xmath68 tensor . for small but nonzero reynolds number re@xmath85 on the length scale @xmath86 of a particle , the instability arises for wavenumber @xmath31 with [ rerange ] _ a qa 1 , a range that clearly does not exist for swimmers of macroscopic size and speed . the enhancement or reduction of viscosity by activity @xcite in a suspension in the isotropic phase is one of the most robust predictions of active hydrodynamics . the mechanism is readily understood , pictorially ( see fig . [ activevisc ] ) or verbally : consider a collection of filaments endowed with contractile or extensile force dipoles , in the isotropic phase but at high enough concentration that collective orientational relaxation is slow . a modest imposed shear flow will then produce appreciable alignment . contractile filaments will then pull back on , and extensile filaments push out on , the flow that oriented them . the result is a higher stress / rate ratio , that is , a viscosity , for contractile filaments , and lower for extensile filaments , than would be found for dead particles of the same shape and at the same concentration . upon tuning a parameter such as concentration to increase the collective orientational relaxation time @xmath84 , a suspension of contractile filaments should show strong viscoelasticity , like an equilibrium system approaching a glass transition @xcite and extensile filaments should show enhanced unstable shear - thinning . numerical studies @xcite find a complex active rheology including the predicted @xcite glassy enhancement of viscosities for contractile systems , and shed valuable light on the role of boundary conditions . microscopic models for filaments and motors @xcite and swimmers @xcite recover the predictions of ref . @xcite . the enhancement of biological noise by active processes was studied some years ago in the context of membranes with pumps @xcite . a natural explanation for the enhanced noise temperature of swimming bacteria @xcite was found in the active hydrodynamic framework by including random forces , torques and currents @xcite which are consistent with conservation laws but unrelated a priori to transport coefficients , because these are systems far from thermal equilibrium . lau and colleagues @xcite model the microrheology of bacterial suspensions in the active hydrodynamic framework with noise . they relate the collective noise to the tumbling of bacteria , and show that equation [ sigact ] implies a @xmath87 decay of the stress autocorrelation for times shorter than the collective orientational relaxation time @xmath84 . they also show , as do underhill @xcite , that equal - time correlations of the suspension velocity field on intermediate length scales @xmath88 should decay as @xmath89 . related predictions arise in work by golestanian @xcite in the _ a priori _ different context of catalytically self - propelled colloids @xcite . active noise in the ordered phase is predicted @xcite to lead to diffusivities that depend on sample thickness , through a mechanism related to the giant fluctuations @xcite discussed in sections [ drypolar ] and [ activecurr ] . schools of fish @xcite appear to be spontaneously ordered phases at high reynolds number @xcite and , from equation [ rerange ] , evade the instability of the oriented state of low - re self - propelled particles . as a first step towards applying active hydrodynamics to macroscopic swimmers , ref . @xcite studied linear perturbations about an ordered steady state of uniform number density and propulsion velocity relative to a quiescent fluid . ignoring inertia and retaining active stresses and the navier - stokes acceleration term in equation [ nsactive ] , they found that all modes were propagative , with highly anisotropic speeds of order @xmath90 , and showed that the inclusion of fluid flow did not eliminate the giant number fluctuations in polar flocks predicted by toner and tu @xcite . it would seem worthwhile to test these predictions on fish schools . @xcite , to which we return in section [ fishsofq ] below . . the cytoskeleton with its motors , on timescales long compared to the release time of transient crosslinks , is also an active fluid suspension of polar orientable objects @xcite with macroscopically contractile stresses along the filament axes @xcite . hydrodynamic equations for _ active gels _ @xcite were formulated with a view to modelling the cytoskeleton , mainly on long timescales when it is fluid . the description turns out to be precisely the same as that of ref . @xcite , but the construction uses linear irreversible thermodynamics @xcite for a nonequilibrium steady state _ near equilibrium_. a linear relation is assumed @xcite between `` fluxes '' ( including the stress @xmath91 and the rate @xmath88 of atp hydrolysis ) and `` forces '' including the velocity gradient @xmath92 and the chemical potential difference @xmath93 between atp and its reaction products , allowing for a nondiagonal matrix of kinetic coefficients . the existence of the polar order parameter @xmath5 makes it possible to construct a scalar @xmath94 contribution to @xmath88 , linear in @xmath95 , where @xmath96 is a phenomenological parameter . by the ( anti)symmetry of reversible kinetic coefficients , this implies a piece @xmath97 in the stress @xmath91 , which we recognize as the active stress of equation [ sigact ] . below is some recommended reading on recent applications of active hydrodynamics to the cytoskeleton . active stresses have been shown @xcite to be important in the formation of the contractile ring around the equator of an animal cell about to divide . oscillations in experiments on non - adhering fibroblast cells have been understood through the interplay of active stresses in the cortical actin layer with the entry of calcium through ion channels @xcite . strong departures from the fluctuation - dissipation theorem ( fdt ) @xcite have been seen in a permanently crosslinked actin network studded with myosin motors , and analysed using active hydrodynamics extended to the case of a permanently crosslinked network @xcite . strongly non - fdt behaviour has been predicted @xcite in an active hydrodynamic treatment of a stiff filament in an active medium , and connections drawn to the dynamics of auditory hair cells @xcite and axons @xcite . the idea of actin - based active transport parallel to the cell membrane has been proposed in ref . @xcite . kruse and jlicher @xcite , motivated by actomyosin phenomena , studied active stresses in one - dimensional microscopic models of motors and filament bundles . in an important series of papers , liverpool , marchetti and collaborators @xcite have carried out a systematic construction of the coarse - grained equations of motion of active filaments , for systems on a substrate as well as in bulk suspension , starting from a microscopic description of motors gripping filaments and moving along them . they obtain equations of the form proposed by purely phenomenological theories @xcite , but each parameter in the coarse - grained theories is traced to a microscopic process involving motors and filaments . an important outcome @xcite is phase diagrams including the various well - known instabilities of actin solutions such as bundling , with a connection to experimentally accessible control parameters such as filament concentration and strength of motor activity . the initial approach was motivated by cytoskeletal physics , but more recently @xcite the focus has been on swimmers . the prediction @xcite that activity alters the viscosity of suspensions , as discussed in section [ activesusp ] , has recently been tested by rafa @xcite on the motile alga chlamydomonas , a contractile swimmer , in a conventional rheometer , and by sokolov and aranson @xcite on the bacterium bacillus subtilis , an extensile swimmer , by monitoring the decay of an imposed vortex as well as by measuring the viscous torque on a rotating particle . the idea was to check whether the change in viscosity could be attributed directly to the swimming activity . [ rafaifig ] shows the viscosity of the microbe suspension as a function of concentration @xmath98 , and compares the values obtained with live and dead cells , for @xmath98 upto 25% . the dependence of viscosity on @xmath98 was about twice as strong for live as for dead cells . for b. subtilis in @xcite , the viscosity was suppressed by as much as a factor of 7 in some cases . these experiments not only confirm the predictions of hatwalne @xcite but also provide strong support for the idea that a medium suffused with active processes such as swimming should be thought of as a distinct type of material . there is ample qualitative evidence of the instability of active ordered suspensions predicted in reference @xcite and discussed in section [ instabsec ] . it would appear that collections of bacteria ca nt swim straight even if they set out to do so @xcite , and that their own active stresses are responsible . bacterial turbulence is seen in fig . [ rayfig ] . the role of instabilities in focusing the bacterial concentration has been stressed by goldstein and coworkers @xcite . a remarkable underwater acoustic measurement @xcite of the static structure factor of the number density of fish shoals , for wavenumbers from @xmath99 to @xmath100 m@xmath101 reveals a power law @xmath102 , and a rapid wavelike response of fish to perturbations , both of which are consistent with the toner - tu @xcite work described above . however , to deal with the fact that the fish are in water , a more complete theory of active suspensions at high reynolds number , or at least in the unsteady - stokes description , is clearly needed . power counting on equation [ nsactive ] shows that nonlinearities are relevant in dimension @xmath103 . an actual renormalization - group calculation for this problem is even more daunting than that for the toner - tu model @xcite and therefore has not been done . a useful first step would be a comparison of the fish - shoal observations @xcite to the unsteady - stokes predictions @xcite discussed under `` fish schools '' in section [ activesusp ] . we have already seen one imitation of active matter in section [ subsub : dryapolarexps ] , namely , a vertically agitated horizontal layer of bits of wire . several efforts are underway to make colloidal active matter , including beads or rods half covered with platinum and immersed in h@xmath104o@xmath104 @xcite . the platinum catalyzes 2h@xmath104o@xmath104 @xmath105 2h@xmath104o @xmath106 o@xmath104 , and the asymmetry leads to directed motion as a result of osmotic gradients . these have been analyzed theoretically in references @xcite . electrosmotic propulsion has been explored by lammert @xcite a magnetically driven artificial flagellum has been demonstrated by dreyfus @xcite . the study of self - propelled liquid drops on a surface , a common sight on a hot skillet in the kitchen , is a subfield in itself and can not be reviewed in any detail here . the driving force is generally evaporation of one or more constituents from the liquid . a couple of recent examples are the work of thakur @xcite on nematic drops and chen @xcite on oil drops driven by surface tension gradients . it should also be possible to make motors by forcing asymmetric elastic dimers with nonequilibrium noise @xcite . an intriguing new theoretical development holds out the possibility of using mechanical agitation to make stokesian swimmers from collections of elastic dimers @xcite . in all these cases a well - known general principle @xcite seems to work : polarity together with sustained dissipation leads to motility . what has the active - matter effort achieved ? it has opened a new area of fundamental physics the hydrodynamics of liquid crystals and suspensions in an entirely novel regime . it is part of a larger process : biology is increasingly concerned with the mechanical forces exerted in living systems , and the active - matter approach provides the framework in which to ask these questions . it has made specific predictions , and experimenters are managing to check some of them . the impact on biology is already being felt , from the subcellular scale to starling flocks @xcite . in cell biology , tissue is clearly the next mountain , and some steps on it have already been taken @xcite . the active hydrodynamic framework , suitably adapted , should apply to self - driven particles in more complex media and geometries @xcite . the approach to active matter advocated in this review shares an important weakness with all coarse - grained theories : an abundance of phenomenological parameters . if thermal brownian motion and active processes are present in comparable magnitude , a clear distinction can probably be made between parameters determined primarily by equilibrium physics and arising strictly from activity . however this nice separation is lost if noise and coefficients themselves are functions of activity @xcite , as they probably are in bacteria and certainly are in granular matter . this limits somewhat the predictive power of coarse - grained theories . this is a limitation we re used to in equilibrium physics : ginzburg - landau free - energy functionals predict temperature dependences best near continuous phase transitions , not over an entire phase . in addition , living systems are finite in space and time and thus not ideally suited for gradient - expanded approaches . we must admit these limitations and accept humbly the need for microscopic theories . there are two imperatives if the field is to make a real contribution to biology . future theory work must aim to integrate the combined effects of growth , cross - linking , treadmilling , multiple species of filament , and the coupling of the cytoskeleton to the cell membrane @xcite . future experimental work must aim for quantitative tests of existing predictions of active - hydrodynamics , on controlled experimental systems . support from cefipra project 3504 - 2 and from the dst , india through the ccmt , sr / s4/ms:419/07 and a j c bose fellowship are gratefully acknowledged . i thank salima rafa for a useful discussion and the use of fig . [ rafaifig ] , and ray goldstein for fig . [ rayfig ] .

the chemical composition of a galaxy is one of its most fundamental characteristics . here we focus on disk galaxies . because the chemical enrichment depends on various physical processes , such as the star formation history and the mass exchange between the galaxy and its environment , progress in our understanding of galaxy formation and evolution processes depends to a large extent on improving our knowledge of the detailed chemical properties of galaxies , such as the radial distribution of element abundances across galactic disks . establishing the macroscopic properties of spiral and irregular galaxies that likely govern the distribution of heavy elements across their disks is very important in understanding the ( chemical ) evolution of galaxies . accurate abundance determinations in a sample of galaxies are mandatory for such investigations . the classical @xmath1 method , often referred to as the direct method , is generally considered to provide the most reliable oxygen and nitrogen abundances in hii regions . when this method can not be used ( due to the lack of measurements of the required weak auroral lines in the spectra of hii regions ) then combinations of the strong nebular line intensities in spectra of hii regions can be used as indicators of their oxygen abundances , as was first suggested by @xcite and @xcite . this approach is usually referred to as the `` strong - line method '' and has been widely adopted . the establishment of calibrations ( i.e. , of relations between metallicity - sensitive emission - line combinations and metallicity ) was the subject of numerous studies ( * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * among many others ) . a calibration is defined not only by the adopted indicators but also by the objects that serve as calibrating data points . a sample of hii regions with abundances derived through the @xmath2 method is used to construct an empirical calibration . a set of photoionization models is used to construct a theoretical ( model ) calibration . even if the same indicator is used in the empirical calibration and in the theoretical calibration and if the same spectral measurements in a given hii region are used , those calibrations can produce significantly different abundance estimations . metallicities derived using theoretical calibrations tend to be systematically higher ( up to @xmath30.7 dex ) than those derived using the empirical calibrations ( see reviews by * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) . therefore , oxygen abundances in extragalactic hii regions obtained in different studies using different calibrations can be significantly different . spectroscopic measurements of hii regions within and beyond optical radii of galaxies were carried out in many works ( see list of references below ) . in these studies , usually the hii regions in one or several galaxies are measured and the radial distributions of the element abundances across the disks of those galaxies are estimated . since often different methods for abundance determinations are used in different works , the resulting abundances from these studies are not homogeneous and can not be directly compared to each other . therefore , the abundances in a sample of galaxies can be analyzed only after those abundances are homogenized , i.e. , all the abundances are redetermined in a uniform way . this is the first step in our present investigation . it should be noted that there have been several attempts to use uniform abundances for the determination of radial abundance gradients in a sample of galaxies ; e.g. , @xcite for a sample of 30 galaxies , @xcite for 39 galaxies , @xcite for 11 galaxies , @xcite for 54 galaxies , and @xcite for 21 galaxies . however , those samples contain a relatively small number of galaxies ( whereas our present sample includes 130 galaxies ) . so far , little attention has been paid to the radial distributions of nitrogen abundances in the disks of galaxies , despite the fact that this provides several advantages for the study of the chemical evolution of galaxies . indeed , since at 12+log(o / h ) @xmath4 8.3 , secondary nitrogen becomes dominant and the nitrogen abundance increases at a faster rate than the oxygen abundance ( e.g. , * ? ? ? * ) , the change in nitrogen abundances with galactocentric distance should then show a larger amplitude in comparison to oxygen abundances and , as a consequence , should be easier to measure . furthermore , there is a time delay in nitrogen production as compared to oxygen production @xcite . thus the comparison between the radial distributions of oxygen and nitrogen abundances in the disks of galaxies can shed additional light on the chemical evolution of galaxies . therefore we consider here not only the radial distributions of oxygen abundances but also those of nitrogen abundances . our paper is organized in the following way . we describe the method used for the oxygen and nitrogen abundance determinations in the hii regions of our galaxy sample in section 2 . we describe the observational data that were used to determine the abundances in the hii regions in section 3 . we discuss the abundance properties in the disks of nearby galaxies ( within the optical isophotal radii ) in section 4 . we summarize our results in section 5 . throughout this paper , we will use the following standard notations for the line intensities : + @xmath5 = @xmath6 \lambda 3727 + \lambda 3729 } /i_{{\rm h}\beta } $ ] , + @xmath7 = @xmath8 \lambda 6548 + \lambda 6584 } /i_{{\rm h}\beta } $ ] , + @xmath9 = @xmath10 \lambda 6717 + \lambda 6731 } /i_{{\rm h}\beta } $ ] , + @xmath11 = @xmath12 } \lambda 4959 + \lambda 5007 } /i_{{\rm h}\beta } $ ] . + with these definitions , the excitation parameter @xmath13 can be expressed as : @xmath13 = @xmath11/(@xmath5+@xmath11 ) . a new method ( called the `` @xmath14 method '' ) for oxygen and nitrogen abundance determinations has recently been suggested @xcite . the idea of the @xmath14 method is a very simple . we have compared a several combinations of the strong - line intensities in the spectrum of a given hii region with those in the spectra of a sample of reference hii regions with known abundances in order to find the counterpart for the hii region under study . it is assumed that the oxygen and nitrogen abundances in the studied hii region are the same as in its counterpart . a counterpart can be selected by comparison of four combinations of strong - line intensities : @xmath13 = @xmath11/(@xmath5 + @xmath11 ) ( excitation parameter ) , log@xmath11 , log(@xmath7/@xmath5 ) , and log(@xmath9/@xmath5 ) . however , there are recent measurements of spectra of many hii regions where the intensities of [ oii]@xmath153727+@xmath153729 or [ sii]@xmath156717 , [ sii]@xmath156731 ) lines are not available ( e.g. , in @xcite or in the sloan digital sky survey ( sdss ) ; see @xcite ) . it has been argued that the oxygen and nitrogen abundances in hii regions can be estimated even if the [ sii]@xmath156717+@xmath156731 emission line is not measured @xcite or if the [ oii]@xmath153727+@xmath153729 emission line is not available @xcite . the @xmath14 method can be adapted to such cases @xcite . to find the counterpart for the hii region under study , one does not need to compare the four combinations of strong - line intensities , but can instead also use only three combinations : 1 ) log@xmath11 , @xmath13 , and log(@xmath7/@xmath5 ) , or 2 ) log@xmath11 , log@xmath7 , and log(@xmath7/@xmath9 ) . when this first set of combinations of strong - line intensities is used to find the counterpart then the resulting oxygen and nitrogen abundances will be referred to as ( o / h)@xmath16 and ( n / h)@xmath16 . when the second set of combinations of strong - line intensities is used to find the counterpart then the inferred oxygen and nitrogen abundances will be called ( o / h)@xmath17 and ( n / h)@xmath17 . the data for reference hii regions with @xmath2-based abundances are compiled in @xcite . the very recent spectroscopic observations of @xcite have been added to the compilation . using these combined data we select a sample of reference hii regions for which all the absolute differences for oxygen abundances ( o / h)@xmath16 ( o / h)@xmath18 and ( o / h)@xmath17 ( o / h)@xmath18 and for nitrogen abundances ( n / h)@xmath16 ( n / h)@xmath18 and ( n / h)@xmath17 ( n / h)@xmath18 are less than 0.1 dex . this sample of reference hii regions contains 250 objects and will in the following be used for abundance determinations . this sample will be referred to as @xmath19 sample ( etalon sample 2013 ) below . furthermore , only blue spectra were observed for hii regions in a number of galaxies ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) , i.e. , intensities of [ nii]@xmath156584 and [ sii]@xmath156717 , [ sii]@xmath156731 ) are not available in those cases . the oxygen abundances in those hii regions can be estimated through the @xmath13 calibration where only the oxygen [ oii]@xmath153727+@xmath153729 and [ oiii]@xmath155007 lines are used @xcite . we have constructed a new variant of the @xmath13 calibration . the sample of reference hii regions , @xmath19 , has been used as calibration data points . to enlarge the number of calibration data points , we have added a number of hii regions with ( o / h)@xmath16 abundances that were chosen in the following way . we determined the ( o / h)@xmath16 abundances in hii regions and obtained radial oxygen abundance gradients across the disks of galaxies ( see below ) . the hii regions where the deviations of the ( o / h)@xmath16 abundances from the general radial abundance trend are less than 0.1 dex were added to the sample of reference hii regions . it is well known that the relation between the oxygen abundance and the strong oxygen line intensities is double - valued , with two distinct parts traditionally known as the upper and lower branches of the r@xmath20 o / h diagram . we have delimited the upper and lower branches and the transition zone , adopting 12+log(o / h ) = 8.3 as the boundary between the upper branch and the transition zone and 12+log(o / h ) = 8.0 as the boundary between the transition zone and the lower branch . these boundaries are somewhat arbitrary , but were chosen so as to give the best calibrations with the existing data . two distinct relations between the oxygen abundance and the strong oxygen line intensities will be established in the following , one for the upper branch ( the high - metallicity calibration ) and one for the lower branch ( the low - metallicity calibration ) . the relation between the oxygen abundance @xmath21 @xmath22 12 + log ( o / h)@xmath23 and r@xmath24 and p can be fitted by a polynomial of the form @xcite @xmath25 where we have used the notation z @xmath22 12 + log ( o / h ) for brevity . to take into account the dependence on the excitation parameter @xmath13 , the coefficients of eq . ( [ equation : kr3 ] ) are chosen to have the form @xmath26 the coefficients @xmath27 , @xmath28 , @xmath29 , @xmath30 , @xmath31 , and @xmath32 can then be determined by looking for the best fit to our sample of hii regions . we wish to derive a set of coefficients in eq . ( [ equation : kr3 ] ) which gives the minimum value of @xmath33 = @xmath34 . here @xmath35 is equal to log(o / h)@xmath36 log(o / h)@xmath37 for each hii region in our sample . the quantity @xmath33 is the average value of the differences between the oxygen abundances determined through the @xmath13 calibration and the original ones . a few data points with large deviations , in excess of 0.15 dex , are rejected , and are not used in the determination of the final relation . the obtained upper - branch @xmath13 calibration ( for 12 + log(o / h ) @xmath4 8.3 ) is @xmath38 the obtained lower - branch @xmath13 calibration ( for 12 + log(o / h ) @xmath39 8.0 ) is @xmath40 hence , here we use the empirical metallicity scale defined by the hii regions with abundances derived through the direct method ( @xmath2 method ) . we have carried out a fairly comprehensive compilation of published spectra of hii regions in late - type galaxies . only those galaxies where radial abundance gradients can be estimated were taken into consideration . our final list includes 130 galaxies . table [ table : sample ] lists the general characteristics of each galaxy . the first column gives its name . we have used the most widely used name for each galaxy . the galaxies are listed in order of name category , with the following categories in descending order : + ngc new general catalogue , + ic index catalogue , + ugc uppsala general catalog of galaxies , + pgc catalogue of principal galaxies . + the morphological type of the galaxy and morphological type code @xmath41 from leda are reported in columns 2 and 3 . the right ascension ( r.a . ) and declination ( decl . ) ( j2000.0 ) of each galaxy are given in columns 4 and 5 . the right ascension and declination are taken from the nasa / ipac extragalactic database ( ned ) . the isophotal radius @xmath0 in arcmin of each galaxy is reported in column 6 . unless otherwise stated , the @xmath0 values are taken from ( * ? ? ? * thereafter rc3 ) . the position angle and inclination are listed in columns 7 and 8 , and the sources for these values are given in column 9 . the adopted distance and its reference are reported in columns 10 and 11 . the ned distances use flow corrections for virgo , the great attractor , and shapley supercluster infall . the isophotal radius in kpc , estimated from the data in columns 6 and 10 , is listed in column 12 . we have carried out an extensive search of the literature and compiled a sample of measurements of hii regions in nearby late - type galaxies . we have searched for spectra of hii regions with the requirement that they include the [ oii]@xmath153727+@xmath153729 , [ oiii]@xmath155007 , [ nii]@xmath156584 , and [ sii]@xmath156717+@xmath156731 lines . while we have tried to include as many sources as possible , we do not claim our search to be exhaustive . thus , for each listed spectrum , we record the measured values of [ oii]@xmath153727+@xmath153729 , [ oiii]@xmath155007 , [ nii]@xmath156584 , [ sii]@xmath156717 , and [ sii]@xmath156731 . the intensities of all lines are normalised to the h@xmath42 line flux . the predicted values of the flux ratio of oxygen [ oiii]@xmath155007/[oiii]@xmath154959 and nitrogen [ nii]@xmath156584/[nii]@xmath156548 lines are very close to three @xcite . the measurements of the [ oiii]@xmath155007 and @xmath154959 lines in sdss spectra confirm this value of the flux ratio ( e.g. , * ? ? ? therefore , the value of @xmath11 can be estimated without [ oiii]@xmath154959 line as @xmath43[oiii]@xmath155007 , and , similarly , the values of @xmath7 are estimated without the lines [ nii]@xmath156548 as @xmath7 = 1.33[nii]@xmath156584 . we have taken the de - reddened line intensities as reported by the authors . in some papers only the measured fluxes are reported . in these cases , the measured emission - line fluxes were corrected for interstellar reddening using the theoretical h@xmath44 to h@xmath42 ratio ( i.e. , the standard value of h@xmath44/h@xmath42 = 2.86 ) and the analytical approximation to the whitford interstellar reddening law from @xcite . the spectroscopic data so assembled form the basis of the present study . our total list contains 3904 spectra including 162 spectra of hii regions beyond the isophotal radius @xmath0 . here we will examine the radial oxygen and nitrogen abundance distributions within the isophotal radius in every galaxy . the radial oxygen and nitrogen abundances beyond the isphotal radius will be discussed elsewere . when measurements of the lines [ oii]@xmath153727+@xmath153729 , [ oiii]@xmath155007 , and [ nii]@xmath156584 were available then the oxygen ( o / h)@xmath16 and nitrogen ( n / h)@xmath16 abundances in the hii regions were estimated and used in our study . if the intensity of the line [ oii]@xmath153727+@xmath153729 was not measured but the measurements of the lines [ oiii]@xmath155007 , [ nii]@xmath156584 , [ sii]@xmath156717 , and [ sii]@xmath156731 ) were available then the oxygen ( o / h)@xmath17 and nitrogen ( n / h)@xmath17 abundances of the hii regions were estimated and used . if the measurements of only the oxygen lines [ oii]@xmath153727+@xmath153729 and [ oiii]@xmath155007 were available ( blue spectrum was observed only ) then the oxygen ( o / h)@xmath45 abundance of the hii regions was estimated and adopted . the deprojected radii of the hii regions were computed using their coordinates ( or offsets from the nucleus ) as reported in the original papers , as were the position angle and inclination listed in table [ table : sample ] . in some publications , the positions of the observed hii regions ( or their identifications in catalogues ) were not reported , but the deprojected radii were listed instead . in these cases these deprojected radii were used ( after correction for the galaxy distance adopted here , if necessary ) . the fractional radii ( normalized to the optical isophotal radius @xmath0 ) were obtained with isophotal radii from table [ table : sample ] . the radial oxygen abundance distribution within the isophotal radius in every galaxy was fitted by the following equation : @xmath46 where 12 + log(o / h)@xmath47 is the oxygen abundance at @xmath48 = 0 , i.e. , the extrapolated central oxygen abundance , c@xmath49 , is the slope of the oxygen abundance gradient expressed in terms of dex @xmath50 , and @xmath51/@xmath52 is the fractional radius ( the galactocentric distance normalized to the disk s isophotal radius @xmath0 ) . we also determined the oxygen abundance gradient expressed in terms of dex kpc@xmath53 . if there were data points with large deviations , in excess of 0.2 dex , those points were rejected , and were not used in the determination of the final relation . the derived parameters of the oxygen abundance distributions are presented in table [ table : grad ] . the name of the galaxy is listed in column 1 . the extrapolated central 12 + log(o / h)@xmath47 oxygen abundance and the gradient ( the coefficient c@xmath49 in eq . ( [ equation : grado ] ) ) expressed in terms of dex @xmath54 are listed in columns 2 and 3 . the oxygen abundance gradient expressed in terms of dex kpc@xmath53 is listed in column 4 . the scatter of oxygen abundances around the general radial oxygen abundance trend is reported in column 5 . as in the case of the oxygen abundance , the radial nitrogen abundance distribution in every galaxy was fitted by the following equation : @xmath55 again , if there were data points with large deviations , in excess of 0.3 dex , those points were rejected , and were not used in the determination of the final relation . these different rejection criteria for oxygen abundances , 0.2 dex , and nitrogen abundances , 0.3 dex , are used for the following reason . there is no one - to - one correspondence between nitrogen and oxygen abundances , but instead there is a scatter in n / h at a given o / h . therefore one can expect that the natural scatter in n / h at a given galactocentric distance can be larger than that in o / h assuming similar uncertainties in both abundance determinations . the derived parameters of the nitrogen abundance distributions are presented in table [ table : grad ] . the extrapolated central 12 + log(n / h)@xmath47 nitrogen abundance and the gradient ( the coefficient c@xmath56 in eq . ( [ equation : gradn ] ) ) expressed in terms of dex @xmath50 are listed in columns 6 and 7 . the nitrogen abundance gradient expressed in terms of dex kpc@xmath53 is listed in column 8 . the scatter of nitrogen abundances around the general radial nitrogen abundance trend is reported in column 9 . a list of references for the emission line flux measurements in the hii regions is given in table [ table : reference ] . in the case of ( o / h)@xmath23 abundances , there is the following problem . the relationship between oxygen abundance and strong oxygen line intensities , the @xmath13 calibration , is double - valued with two distinct parts usually known as the `` lower '' and `` upper '' branches of the @xmath57 o / h relationship . the expression for the oxygen abundance determination in high metallicity hii regions , eq . ( [ equation : ohph ] ) , is valid only for hii regions that belong to the upper branch , with 12+log(o / h ) @xmath4 8.3 . thus , one has to know a priori on which of the two branches the hii region lies . we can overcome this problem in the way suggested in @xcite . it has been known for a long time @xcite that disks of spiral galaxies show radial oxygen abundance gradients , in the sense that the oxygen abundance is higher in the central part of the disk and decreases with galactocentric distance . we thus start from the hii regions in the central part of disks and move outward until the radius @xmath58 where the oxygen abundance decreases to 12+log(o / h ) @xmath3 8.3 . it should be noted that it is difficult to establish the exact value of @xmath58 due to the scatter in oxygen abundance values at any fixed radius . an unjustified use of eq . ( [ equation : ohph ] ) in the determination of the oxygen abundance in low - metallicity hii regions beyond @xmath58 would result in overestimated oxygen abundances , and would produce a false turnover in the slope of the abundance gradients @xcite . therefore , hii regions with galactocentric distances larger than @xmath58 , i.e. , those with 12+log(o / h ) @xmath39 8.3 were rejected . the derived radial distributions of the oxygen and nitrogen abundances in 130 galaxies are presented in figure [ figure : gs1 ] . the oxygen abundances for individual hii regions are indicated by points . all the data points ( including points with large deviations , which are rejected and are not used in determination of the final relation ) are shown . the linear best fits ( derived via the least squares method ) to the points are represented by solid lines . the galactocentric distances are normalized to the isophotal radius . the nitrogen abundances for individual hii regions are shown by the plus signs . the linear best fits to the points are indicated by dashed lines . the values of the gradients in a number of galaxies ( e.g. , in ngc 12 , the first galaxy in our list ) are rather small and are comparable ( or even lower ) to the uncertainties of gradients . it can be assumed that there is no abundance gradient in such a galaxy , and its abundance can be specified by the mean of the central abundance and the abundance at the isophotal @xmath0 radius of the galaxy . thus , the radial oxygen and nitrogen abundance distributions across the optical disk in every galaxy are individually fitted by a single relation . it looks like a rather good approximation for the majority of galaxies . however , a small change in slope in the abundance distribution can not be excluded in the disks of several galaxies ( e.g. , ngc 925 , ngc 3184 , ngc 5457 ) . it is interesting to note the following . @xcite studied the surface brightness profiles of a sample of late - type ( sb to sdm ) spiral galaxies using imaging data from the sdss survey . they found that the surface brightness profiles can be divided into three classes . a small fraction of galaxies ( around 10 % ) belong to type i , which comprises those galaxies that have a normal ( standard ) exponential disk down to the noise limit . the surface brightness distribution of the rest of the galaxies is better described as a broken exponential . about 60 % of the galaxies belong to type ii , which means that they show a down - bending profile with steeper outer part . about 30 % of the galaxies belong to type iii , implying that they show an up - bending profile with shallower outer part . thus , the abundance distribution profile does not seem to follow strictly the slope of the surface brightness profile . it should be noted , however , that the number of measured hii regions in many galaxies is too small to allow one to detect a small bend in the radial abundance distribution . * ngc 1512 . * the five hii regions that are located in the `` bridge '' between the interacting galaxies ngc 1512 and ngc 1510 @xcite were excluded from our analysis . @xcite found that there is a significant discrepancy between the kinematically derived inclination of ngc 1512 and its photometric value ( @xmath59 = 65@xmath60 and @xmath61 = 35@xmath60 ) . they noted that the explanation is quite clear : the photometric parameter is mainly representative of the bar which contributes a large part of the light resulting in more edge - on values . however , the outer isophotes are much more face - on . therefore , the kinematic value of the inclination is adopted here . * ngc 2442 . * the peculiar spiral galaxy ngc 2442 is tidally distorted * and references therein ) . * ngc 3227 . * ngc 3227 is a nearby seyfert galaxy that is interacting with its gas - poor dwarf elliptical companion ngc 3226 . @xcite discovered a cloud of hi close to , but physically and kinematically distinct from , the galactic disk of ngc 3227 . they suggested that this cloud ( j1023 + 1952 ) might be a dwarf galaxy that is either preexisting and being accreted by ngc 3227 , or a newly created tidal dwarf galaxy @xcite . however , the oxygen abundance of the cloud ( hii regions 24 and 25 from @xcite and of the hii region measured by @xcite ) follows the general trend of the radial abundance distribution in the disk of ngc 3227 . the oxygen abundances of the hii region in the cloud estimated in different ways using the emission line measurements by @xcite are in satisfactory agreement to each other : 12+log(o / h)@xmath62 = 8.40 , 12+log(o / h)@xmath63 = 8.43 and 12+log(o / h)@xmath45 = 8.34 . hii regions 10 and 12 from @xcite follow also the general trend of the radial abundance distribution in the disk of ngc 3227 although those hii regions may be associated with ngc 3226 . * ngc 3239 . * the irregular galaxy ngc 3239 is a candidate for a merging system @xcite . the oxygen abundances determined in four hii regions from sdss spectra as well as the global abundance determined using the integrated spectra from @xcite are 12+log(o / h ) @xmath3 8.0 @xmath64 8.1 , i.e. , those abundances correspond to the transition zone between the upper and lower branches of the o / h @xmath65 diagram . therefore , the ( o / h)@xmath45 abundances we derived from measurements by @xcite are not reliable and were not used . * ngc 3310 . * ngc 3310 is believed to be an advanced merger and its unusual outer structure is probably the result of a recent merger with a smaller galaxy @xcite . spectroscopic observations of six hii regions in ngc 3310 were presented by @xcite . the auroral line [ oiii]@xmath154363 is detected in four hii regions . the @xmath2-based abundances in those hii regions are in the range from 12+log(o / h ) = 8.13 to 8.25 . however , the coordinates of the hii regions are not published , which prevents us from using those hii regions in radial gradient determinations . * ngc 3359 . * the ngc 3359 is a giant , very strongly barred spiral galaxy . @xcite estimated the radial gradient using the [ oiii]/h@xmath42 and [ nii]/[oiii ] indicators ( one - dimensional calibrations ) . they found the radial abundance gradient break near the corotation radius . @xcite carried out new measurements and determined abundances through the @xmath66 and @xmath67 calibrations following @xcite . they concluded that , with a high degree of confidence , a model with a break fits the data significantly better than one without a break . the @xmath14-based abundances show no gradient across the entire galaxy . the @xmath2-based abundances in three hii regions in ngc 3359 are in agreement with no gradient . * ngc 3718 . * this galaxy is so peculiar that it is difficult to categorize it morphologically @xcite . * ngc 4559 . * the hii regions in ngc 4559 with galactocentric radii larger than 0.4@xmath0 measured by @xcite were excluded since they are in the transition zone . * ngc 4625 . * this is a magellanic spiral with @xmath0 @xmath3 3 kpc and an extended faint disk reaching four times the optical @xmath0 radius of the galaxy in the ultraviolet . * ngc 5668 . * @xcite have found that within @xmath68 of the nucleus the oxygen abundance o / h follows an exponential profile while the outer abundance trend flattens out to an approximately constant value and could even reverse . * ngc 7518 . * this is a hydrogen - deficient galaxy in the pegasus cluster . following @xcite , oxygen lines from `` blue '' spectra ( 360 nm 560 nm ) have been used in the abundance determinations . * ngc 7529 . * this galaxy of the pegasus cluster has a normal hi content . there is no determination of the position angle , but the inclination is small . again following @xcite , oxygen lines from `` blue '' spectra have been used in the abundance determinations . * ngc 7591 . * this galaxy of the pegasus cluster has a normal hi content . as before , following @xcite , oxygen lines from `` blue '' spectra have been used in the abundance determinations . * ic 10 . * due to uncertainties in the position angle and in the inclination of ic 10 , the galactocentric distances of hii regions have been computed without any de - projection ( following * ? ? ? * ic 5309 . * a hydrogen - deficient galaxy of the pegasus cluster . following @xcite , oxygen lines from `` blue '' spectra have been used in the abundance determinations . * ugc 9562 = ii zw 71 . * the blue compact dwarf galaxy ugc 9562 is a probable polar - ring galaxy , and there are a several luminous hii regions along its major axis . the oxygen abundances in four hii regions are around 12+log(o / h ) @xmath3 8.2 , i.e. , those abundances correspond to the transition zone between the upper and lower branches of the o / h @xmath65 diagram . therefore , the ( o / h)@xmath45 abundances obtained with measurements from @xcite were not used . the global oxygen abundances determined from the integrated spectra from @xcite are 12+log(o / h ) @xmath3 8.3 . * pgc 029167 . * the dwarf galaxy `` garland '' or pgc 029167 ( a tidal dwarf candidate ; * ? ? ? * ) lies within the tidal bridges of neutral hydrogen connecting m 81 , m 82 and ngc 3077 . this alignment has led to the suggestion that this dwarf formed recently , as a result of tidal interactions within the group . hii regions in and near garland exhibit enhanced metallicities compared to other galaxies at similar luminosities . notably , the oxygen abundances are similar to abundances measured in m 81 and ngc 3077 ( * ? ? ? * and references therein ) . figure [ figure : ohnh ] shows the nitrogen abundance as a function of oxygen abundance in our sample of galaxies . the plus signs are the nitrogen ( n / h)@xmath47 and oxygen abundances ( o / h)@xmath47 at @xmath48 = 0 , i.e. , the central nitrogen and oxygen abundances . the points show the nitrogen ( n / h)@xmath69 and oxygen ( o / h)@xmath69 abundances at the @xmath0 radius of the galaxies . our data show a well - defined sequence in the o / h n / h plot . this sequence exhibits the well - known turnover in the sense that the slope at low metallicity is shallower than that at high metallicity . the commonly accepted explanation of this change of slope is that nitrogen can be interpreted as having both primary and secondary components . nitrogen production is primary at low metallicity , but for 12 + log(o / h ) @xmath4 8.3 , secondary nitrogen becomes prominent , and nitrogen increases at a faster rate than oxygen @xcite . figure [ figure : ohnh ] shows that at high metallicity , 12 + log(o / h ) @xmath4 8.2 , the relation between the logarithms of nitrogen and oxygen abundances can be approximated by a linear expression . for ( n / h)@xmath47 and ( o / h)@xmath47 abundances , we found the following relation through the least squares method : @xmath70 a similar relation @xmath71 was obtained for the ( n / h)@xmath69 and ( o / h)@xmath69 abundances . a comparison between eq . ( [ equation : nhro ] ) and eq . ( [ equation : nhrg ] ) shows that the relations n / h = @xmath72(o / h ) for the ( n / h)@xmath47 and ( o / h)@xmath47 abundances and for the ( n / h)@xmath69 and ( o / h)@xmath69 abundances agree with each other within the errors . the n / h = @xmath72(o / h ) relation given by eq . ( [ equation : nhro ] ) is presented in figure [ figure : ohnh ] by the solid line . the dashed lines indicate shifts along the y - axis by @xmath730.15 dex . at low metallicity , 12 + log(o / h ) @xmath39 8.0 , the single relation @xmath74 was found using both central abundances and abundances at the @xmath0 radii of the galaxies . this relation is presented in figure [ figure : ohnh ] by the dotted line . figure [ figure : ohnh ] shows that there is an appreciable spread in n / h at a given o / h . the variation in n / h is around a factor of 2 for a given o / h . the scatter can be partially attributed to the errors in the abundance determinations but part of it seems to be true abundance scatter . two major mechanisms have been proposed for generating a spread . one mechanism invokes a time delay between ejections of the freshly manufactured oxygen and nitrogen into the interstellar medium by a given stellar generation . the n / o ratios may be an indicator of the age of a galactic system , indicating the time that has passed since the bulk of star formation activity @xcite . thus , the n / o ratio in a galaxy would then depend on its star formation history . the second mechanism for causing n / h variations at a given o / h is a variation in the efficiency of enriched galactic winds @xcite or / and in the inflow of gas into the galaxy @xcite . it is believed that galactic winds do not play a significant role in the chemical evolution of large spiral galaxies @xcite . the enhanced n / o ratio in individual hii regions can be caused by the local pollution in nitrogen by wolf - rayet stars ( * ? ? ? * and references therein ) . since we consider `` average '' n and o abundances based on the abundances of different hii regions this origin of the scatter in our diagrams seems to be unlikely . it has been known for a long time that galaxies of different morphological types have different star formation histories , i.e. , spiral galaxies with early morphological types have a larger fraction of old stars @xcite . one can then expect that n / h at a given o / h may depend on the morphological type of galaxy expressed in terms of @xmath41 type @xcite . on the other hand , the star formation history of a galaxy also strongly depends on its mass ( or luminosity ) as is epitomized in the galaxy downsizing effect , where the star formation acitvity shifts from high - mass galaxies at early cosmic times to lower - mass galaxies at later epochs @xcite . again , one can expect then that the n / h ratio at a given o / h will depend on the galaxy mass @xcite . there is a relatively tight linear correlation between the absolute magnitudes and the logarithms of the linear diameters of nearby galaxies @xcite . therefore , the linear diameter of a galaxy can be used as indicator of its luminosity ( and mass ) . thus , if the spread in n / h at a given o / h is caused by the time delay between nitrogen and oxygen enrichment and the different star formation histories in different galaxies then the n / h at a given o / h should correlate with morphological @xmath41 type or / and with linear radius of the galaxy . figure [ figure : residual ] shows the residual of eqs . ( [ equation : nhro]),([equation : nhrg ] ) as a function of the morphological @xmath41 type and linear radius of the galaxy @xmath0 . figure [ figure : residual ] indicates that there may exist some correlation between the n / h at a given o / h and morphological type or / and linear radius of the galaxy . to verify this , we found a relation between n / h and o / h for spiral galaxies ( with @xmath41 @xmath75 7.5 ) where @xmath41 and log@xmath0 are `` secondary parameters '' , n / h = @xmath72(o / h,@xmath41,log@xmath0 ) . it should be noted that @xmath41 and log@xmath0 are not a perfectly independent parameters since there is some correlation between them . the relation obtained for central abundances is @xmath76 the values of the coefficients in the terms containing @xmath41 and log@xmath0 are similar . however , the @xmath41 value is a more important second parameter than the log@xmath0 value since the variation in @xmath41 values ( from 1 to 7.5 ) is much larger than the variation of log@xmath0 values ( from @xmath30.6 to @xmath31.5 ) . therefore , the variation in n / h due to variation in the morphological type is larger than that due to variation in the linear radii of galaxies . for example , the sab galaxies ( @xmath41 = 2 ) have nitrogen abundances larger than on average 0.1 dex ( i.e. , by around 30% ) than sd ( @xmath41 = 7 ) galaxies with the same oxygen abundances . the relation n / h = @xmath72(o / h,@xmath41,log@xmath0 ) obtained for abundances at the optical @xmath0 radii of the galaxies is @xmath77 in this case the log@xmath0 value is the more important second parameter than the @xmath41 value . although the morphological type is the more important second parameter in the relation for central abundances while log@xmath0 is the more important second parameter in the relation for abundances at the optical @xmath0 radii of the galaxies , it is difficult to draw a solid conclusion whether this is physically meaningful since there is a correlation between @xmath41 and log@xmath0 . that the relation between n / h and o / h depends on the additional parameter(s ) @xmath41 and/or log@xmath0 suggests that the scatter in n / h at a given o / h can be caused , at least partly , by the time delay between nitrogen and oxygen enrichment and the different star formation histories in different galaxies . the upper left panel of figure [ figure : xhrof ] shows the central oxygen ( o / h)@xmath47 abundance in a galaxy as a function of its morphological @xmath41 type ( data from table [ table : sample ] and table [ table : grad ] ) . the ( o / h)@xmath47 @xmath41 diagram shows that there is a trend in central oxygen abundance with morphological type for galaxies later than sc ( @xmath41 @xmath4 5 ) such that the central oxygen abundances are lower in galaxies of later types . this trend disappears for early - type spiral galaxies ( @xmath41 @xmath39 5 ) . the upper right panel of figure [ figure : xhrof ] shows the central oxygen abundance in a galaxy as a function of its isophotal radius @xmath0 . since there is a relatively tight linear correlation between the absolute magnitudes and the logarithms of the linear diameters of nearby galaxies @xcite , this diagram can be considered as some kind of analog of the standard `` luminosity metallicity '' diagram . the ( o / h)@xmath47 @xmath0 diagram shows that there is a weak correlation between the central oxygen abundance and optical radius for small galaxies ( @xmath0 @xmath39 10 kpc ) in the sense that the smaller galaxies have on average lower oxygen abundances . this correlation disappears for large galaxies ( @xmath0 @xmath4 10 kpc ) , i.e. , the most oxygen - rich galaxies of different radii have similar central oxygen abundances . different versions of the luminosity metallicity diagram have been constructed in earlier studies . @xmath78-band luminosity characteristic oxygen abundance diagrams were considered in @xcite , where the characteristic oxygen abundance is defined as the abundance at @xmath51 = 0.4@xmath0 . @xcite have used the global abundances ( in the sense that their abundances do not correspond to the abundances at a fixed galactocentric distance , but instead are some kind of mean abundance for a fraction of a galaxy within the fiber aperture ) estimated from sdss spectra in constructing the luminosity metallicity diagram . the luminosity central metallicity diagram was examined in @xcite . the flattening of the luminosity metallicity relation at high luminosities ( essentially a plateau ) can be seen in all versions of the diagram . thus the plateau in our ( o / h)@xmath47 @xmath0 diagram at large radii is consistent with previous results . it has been advocated that the constant maximum value of the observed central oxygen abundance in the most oxygen - rich galaxies suggests that the observed oxygen abundance in the centers of those galaxies represents the maximum attainable value of the gas - phase oxygen abundance @xcite . the upper - row panels of figure [ figure : xhrof ] show that the observed central oxygen abundance in the most oxygen - rich galaxies in our sample is 12 + log(o / h)@xmath47 @xmath3 8.85 . the dashed lines in the upper row of panels of figure [ figure : xhrof ] show this value of oxygen abundance , which seems to correspond to the maximum attainable value of the gas - phase oxygen abundance in galaxies . the observed central oxygen abundance in the most oxygen - rich galaxies from our sample is a factor @xmath32 higher than the gas - phase oxygen abundance in the solar neighbourhood , 12 + log(o / h ) @xmath3 8.5 ( e.g. , * ? ? ? the maximum attainable value of the oxygen abundance in galaxies obtained here is in agreement with the value from @xcite . because some fraction of the oxygen ( about 0.1 dex ) is expected to be locked in dust grains ( e.g. , * ? ? ? * ) , the maximum value of the true ( gas + dust ) oxygen abundances in hii regions of spiral galaxies is 12 + log(o / h)@xmath47 @xmath3 8.95 . the lower left panel of figure [ figure : xhrof ] shows the central nitrogen ( n / h)@xmath47 abundance in a galaxy disk as a function of its morphological @xmath41 type . the comparison of the panels in the left column of figure [ figure : xhrof ] shows that the changes in the central oxygen and nitrogen abundances with morphological type of a galaxy are rather similar . the lower right panel of figure [ figure : xhrof ] shows the central nitrogen abundance in a galaxy disk as a function of its optical isophotal radius @xmath0 . again , the comparison between the right column panels of figure [ figure : xhrof ] indicates that the general behaviour of the central nitrogen abundances as a function of optical radius is similar to that for oxygen . according to the relation between nitrogen and oxygen abundances given in eq . ( [ equation : nhro ] ) , the value of 12 + log(n / h)@xmath47 @xmath3 8.42 corresponds to the maximum attainable value of the gas - phase oxygen abundance in galaxies with 12 + log(o / h)@xmath47 @xmath3 8.85 . this value of nitrogen abundance is indicated in the lower row of panels of figure [ figure : xhrof ] by the dashed lines . one can see that the value of 12 + log(n / h)@xmath47 @xmath3 8.42 can be adopted as the maximum value of the observed central nitrogen abundance in the most nitrogen - rich galaxies at the present - day epoch . however , in contrast to the case of oxygen , it is not necessary that this value corresponds to the the maximum attainable value of the gas - phase oxygen abundance in galaxies because of the time - delay between nitrogen and oxygen enrichment of the interstellar medium . the panels in the left column of figure [ figure : xhrgf ] show the oxygen ( o / h)@xmath69 ( upper panel ) and nitrogen ( n / h)@xmath69 ( lower panel ) abundances measured at the isophotal radius , @xmath0 , as a function of morphological @xmath41 type . a comparison between the panels in the left column of figure [ figure : xhrof ] and figure [ figure : xhrgf ] shows that the changes in the central oxygen ( nitrogen ) abundances and in the abundances at the optical edge of the disk along the hubble sequence ( or morphological type ) are more or less similar ( at least qualitatively ) . however , the oxygen and nitrogen abundances at the @xmath0 radius in four of the late - type galaxies ( ngc 4625 , ic 10 , ugc 223 , and pgc 29167 ) are high and those galaxies show a large deviation from the general trends . ngc 4625 is a magellanic spiral with @xmath0 @xmath3 3 kpc and a very extended faint disk ( see below ) . the dwarf irregular galaxy ic 10 has a very small optical radius , @xmath0 = 0.6 kpc , and a positive radial abundance gradient . one may suggest though that any radial gradient in such small galaxies is hard to define based on hii regions . it remains a mystery why the oxygen abundance changes by a factor of about two on the scale of 0.6 kpc in this galaxy . ic 10 is one of several dwarf galaxies in which evidence for localized , inhomogeneous enrichment has been found @xcite perhaps this is a common mode of enrichment in these small objects . in contrast , the irregular galaxy ugc 223 is rather large with @xmath0 = 9.3 kpc . the peculiar dwarf galaxy pgc 29167 ( or `` garland '' ) is commonly considered a tidal dwarf galaxy ( see comment on this galaxy in section 4.2 ) . the panels in the right column of figure [ figure : xhrgf ] show the oxygen ( o / h)@xmath69 ( upper panel ) and nitrogen ( n / h)@xmath69 ( lower panel ) abundances at the optical edge of the disk as defined by the @xmath0 radius as a function of isophotal radius @xmath0 . the positions of the galaxies in the ( o / h)@xmath69 @xmath0 and ( n / h)@xmath69 @xmath0 diagrams do not show any obvious trends . in particular , the ( o / h)@xmath69 and ( n / h)@xmath69 abundances do not show any appreciable correlation with isophotal radius @xmath0 . however , one feature in these diagrams should be noted : the ( o / h)@xmath69 and ( n / h)@xmath69 abundances have a maximum value in galaxies with an isophotal radius @xmath0 @xmath3 10 kpc and decrease when moving from this value both towards smaller or larger optical radii . figure [ figure : gradf ] shows the radial oxygen and nitrogen abundance gradients in units of dex kpc@xmath53 as a function of morphological @xmath41 type ( left panels ) and of isophotal radius @xmath0 of a galaxy ( right panels ) . inspection of the left column panels of figure [ figure : xhrgf ] shows that the values of the abundance gradients in units of dex kpc@xmath53 do not correlate with the morphological type of a galaxy . according to @xcite , the lack of a correlation between gradients in units of dex kpc@xmath53 and the macroscopic properties of late - type galaxies may suggest that the relationship between these parameters is more complex than a simple correlation . indeed , @xcite have concluded that a correlation for non - barred galaxies is seen . the panels in the right column of figure [ figure : xhrgf ] show that shallow gradients can be found both in small and large galaxies while steep gradients are seen only in small galaxies in the sense that the smaller a galaxy the steeper its gradient . we compiled published spectra of hii regions in 130 nearby galaxies . our list contains 3904 spectra including 162 spectra of hii regions beyond the isophotal radius @xmath0 . the oxygen and nitrogen abundances in hii regions were determined on the metallicity scale defined by hii regions with @xmath2-based abundances . the radial gradients of oxygen and nitrogen abundances across the disks of the galaxies were estimated . at the centers of metal - rich galaxies ( i.e. , ( 12 + log(o / h ) @xmath4 8.2 ) , we found the relation between n and o abundances to be ( n / h)@xmath79 @xmath80 ( o / h)@xmath81 . the ( n / h)@xmath69 = @xmath72(o / h)@xmath69 relation between n and o abundances at the @xmath0 isophotal radii of high metallicity galaxies is similar to that for the abundances at their centers . the variation in ( n / h ) at a given ( o / h ) is around 0.3 dex . to test whether the scatter in n / h at a given o / h can be explained by the time delay between nitrogen and oxygen enrichment and the different star formation histories in galaxies of different morphological types and dimensions ( masses ) , we derived a more complex relation between n and o abundances ( n / h ) = @xmath72((o / h),@xmath41,@xmath0 ) . we found that the morphological type , @xmath41 , is a more important `` second parameter '' in the relation for central abundances , while the log@xmath0 is a more important second parameter in the relation for abundances at the @xmath0 radii of our galaxies . since there is a correlation between @xmath41 and log@xmath0 it is as yet unclear whether this difference is meaningful the fact that the relation between n / h and o / h depends on additional parameter(s ) , namely @xmath41 and/or log@xmath0 suggests that the scatter in n / h at a given o / h can be caused , at least partly , by the time delay between nitrogen and oxygen enrichment and the different star formation histories in different galaxies . the best fit to n / h as a function of o / h is close to a linear relation at low metallicity ( 12 + log(o / h ) @xmath39 8.0 ) . the central oxygen abundances ( o / h)@xmath79 show a trend along the hubble sequence of galaxies of late morphological types ( @xmath41 @xmath4 5 ) such that the oxygen abundances are lower in galaxies of later types . this trend disappears for early morphological types . the central oxygen abundance also correlates with optical galaxy radius for small galaxies , @xmath0 @xmath39 10 kpc , being lower in galaxies of smaller radii . the trend disappears for galaxies with large radii . the maximum gas - phase oxygen abundance in large ( 10 kpc @xmath39 r@xmath82 @xmath39 30 kpc ) galaxies ( or in galaxies of early ( 1 @xmath39 @xmath41 @xmath39 5 ) morphological types ) is constant , 12 + log(o / h ) @xmath3 8.85 . this implies that the observed central oxygen abundance of the most oxygen - rich galaxies in our sample is a factor of @xmath32 higher than the gas - phase oxygen abundance in the solar neighbourhood . the central nitrogen abundances ( n / h)@xmath47 show a similar behaviour . the observed central nitrogen abundance in the most nitrogen - rich galaxies of our sample is 12 + log(o / h ) @xmath3 8.42 . the radial o and n abundance gradients ( in units of dex kpc@xmath53 ) within the optical radius do not show any significant correlation with the morphological type and optical radius . however , the spread in the gradients increases with decreasing galaxy radius in the sense that shallow gradients are seen both in small and large galaxies while steep gradients occur only in a small galaxies . the smaller a galaxy the steeper is the gradient that it may show . the abundance data set presented in this paper serves as the foundation for other investigations we are carrying out . in a forthcoming paper @xcite , we examine relations between the radial abundance distribution across the disk and the disk surface brightness profiles in the optical @xmath78 and infrared @xmath83 bands for a sample of nearby galaxies . we are grateful to the referee for his or her constructive comments . l.s.p . and e.k.g . acknowledge support within the framework of sonderforschungsbereich ( sfb 881 ) on `` the milky way system '' ( especially subproject a5 ) , which is funded by the german research foundation ( dfg ) . thanks the astronomisches rechen - institut at heidelberg university where this investigation was carried out for the hospitality . a.y.k . acknowledges the support from the national research foundation ( nrf ) of south africa . + we thank h.j . zahid and f. bresolin for supporting us with some unpublished details of their observations of hii regions in ngc 3359 . + the authors acknowledge the work of the sdss collaboration . funding for sdss - iii has been provided by the alfred p. sloan foundation , the participating institutions , the national science foundation , and the u.s . department of energy office of science . the sdss - iii web site is http://www.sdss3.org/. sdss - iii is managed by the astrophysical research consortium for the participating institutions of the sdss - iii collaboration including the university of arizona , the brazilian participation group , brookhaven national laboratory , university of cambridge , carnegie mellon university , university of florida , the french participation group , the german participation group , harvard university , the instituto de astrofisica de canarias , the michigan state / notre dame / jina participation group , johns hopkins university , lawrence berkeley national laboratory , max planck institute for astrophysics , max planck institute for extraterrestrial physics , new mexico state university , new york university , ohio state university , pennsylvania state university , university of portsmouth , princeton university , the spanish participation group , university of tokyo , university of utah , vanderbilt university , university of virginia , university of washington , and yale university . + we acknowledge the usage of the hyperleda database ( http://leda.univ-lyon1.fr ) . llcrrrrccrcr ngc 0012 & sc & 4.9 & 2.486458 & 4.612528 & 0.85 & 125 & 27 & 1 & 53.20 & 2 & 13.14 + ngc 0055 & sbm & 8.8 & 3.723333 & -39.196639 & 16.18 & 110 & 79 & 3 & 1.91 & 4 & 8.99 + ngc 0099 & sc & 5.9 & 5.997470 & 15.770428 & 0.71 & 0 & 24 & 1 & 77.00 & 5 & 15.82 + ngc 0224 & sb & 3.0 & 10.684792 & 41.269065 & 95.27 & 35 & 77 & 6 & 0.74 & 7 & 20.51 + ngc 0234 & sabc & 5.3 & 10.884977 & 14.342564 & 0.81 & 0 & 0 & 1 & 59.60 & 2 & 14.06 + ngc 0253 & sabc & 5.1 & 11.888000 & -25.288219 & 13.77 & 50 & 76 & 8 & 3.46 & 9 & 13.86 + ngc 0300 & scd & 6.9 & 13.722833 & -37.684389 & 10.94 & 106 & 42 & 10 & 1.94 & 4 & 6.17 + ngc 0428 & sabm & 8.6 & 18.232125 & 0.981557 & 2.04 & 120 & 41 & 1 & 14.80 & 11 & 8.77 + ngc 0450 & sabc & 5.8 & 18.876840 & -0.860974 & 1.54 & 72 & 41 & 1 & 18.70 & 11 & 8.40 + ngc 0493 & sc & 5.9 & 20.537458 & 0.945362 & 1.69 & 58 & 72 & 1 & 23.20 & 11 & 11.43 + ngc 0575 & sc & 5.3 & 22.694333 & 21.440417 & 0.85 & 0 & 21 & 1 & 41.80 & 2 & 10.32 + ngc 0598 & sc & 5.9 & 23.462042 & 30.660222 & 35.40 & 23 & 56 & 12 & 0.76 & 4 & 7.83 + ngc 0628 & sc & 5.2 & 24.173946 & 15.783661 & 5.24 & 25 & 6 & 13 & 9.91 & 14 & 15.09 + ngc 0753 & sabc & 4.9 & 29.425833 & 35.916125 & 1.26 & 128 & 50 & 15 & 67.10 & 16 & 24.52 + ngc 0783 & sc & 5.3 & 30.277542 & 31.882472 & 0.79 & 35 & 27 & 1 & 69.30 & 2 & 15.98 + ngc 0925 & scd & 7.0 & 36.820333 & 33.579167 & 5.24 & 101 & 58 & 17 & 9.29 & 18 & 14.15 + ngc 1055 & sbb & 3.1 & 40.438458 & 0.443168 & 3.79 & 105 & 69 & 1 & 14.40 & 11 & 15.89 + ngc 1058 & sc & 5.1 & 40.874988 & 37.341339 & 1.51 & 145 & 15 & 19 & 10.60 & 20 & 4.66 + ngc 1068 & sb & 3.0 & 40.669629 & -0.013281 & 3.54 & 70 & 32 & 1 & 10.10 & 11 & 10.40 + ngc 1090 & sbc & 3.8 & 41.641401 & -0.247158 & 1.99 & 102 & 64 & 1 & 35.90 & 16 & 20.79 + ngc 1097 & sbb & 3.3 & 41.579375 & -30.274889 & 4.67 & 134 & 46 & 21 & 16.50 & 2 & 22.40 + ngc 1232 & sabc & 5.0 & 47.439625 & -20.579306 & 3.71 & 90 & 30 & 22 & 21.00 & 2 & 22.64 + ngc 1313 & sbcd & 7.0 & 49.566875 & -66.498250 & 4.56 & 4 & 38 & 23 & 4.39 & 24 & 5.82 + ngc 1365 & sb & 3.2 & 53.401546 & -36.140403 & 5.61 & 42 & 46 & 25 & 19.60 & 18 & 32.00 + ngc 1512 & sa & 1.1 & 60.976167 & -43.348861 & 4.46 & 80 & 35 & 26 & 9.50 & 27 & 12.32 + ngc 1598 & sbc & 4.9 & 67.140292 & -47.782556 & 0.72 & 123 & 55 & 1 & 55.80 & 16 & 11.73 + ngc 1637 & sc & 5.0 & 70.367408 & -2.857961 & 1.99 & 31 & 40 & 19 & 12.00 & 18 & 6.95 + ngc 1642 & sc & 5.2 & 70.728792 & 0.618583 & 0.91 & 175 & 29 & 1 & 62.10 & 2 & 16.43 + ngc 1672 & sb & 3.3 & 71.427125 & -59.247278 & 3.30 & 170 & 34 & 1 & 16.70 & 2 & 16.05 + ngc 2336 & sbc & 4.0 & 111.766875 & 80.178083 & 2.51@xmath84 & 175 & 55 & 28 & 34.20 & 2 & 24.95 + ngc 2403 & sabc & 6.0 & 114.214167 & 65.602556 & 10.94 & 126 & 58 & 29 & 3.06 & 18 & 9.74 + ngc 2441 & sabb & 3.3 & 117.978083 & 73.015694 & 1.00 & 0 & 29 & 1 & 50.90 & 2 & 14.77 + ngc 2442 & sbc & 3.7 & 114.099333 & -69.530833 & 2.75 & 27 & 29 & 30 & 17.10 & 31 & 13.67 + ngc 2541 & sabc & 6.0 & 123.667148 & 49.061709 & 3.16 & 165 & 60 & 1 & 12.60 & 18 & 11.56 + ngc 2805 & sabc & 6.9 & 140.085042 & 64.102778 & 3.16 & 125 & 41 & 1 & 28.00 & 31 & 25.70 + ngc 2835 & sc & 5.0 & 139.470458 & -22.354667 & 3.30 & 160 & 45 & 32 & 10.10 & 16 & 9.71 + ngc 2841 & sb & 3.0 & 140.510975 & 50.976519 & 4.06 & 150 & 66 & 33 & 14.10 & 18 & 16.67 + ngc 2903 & sabb & 4.0 & 143.042125 & 21.500833 & 6.30 & 22 & 65 & 34 & 8.90 & 35 & 16.30 + ngc 2997 & sabc & 5.1 & 146.411625 & -31.191083 & 4.46 & 107 & 32 & 36 & 12.20 & 36 & 15.82 + ngc 3020 & sc & 5.9 & 147.527500 & 12.813639 & 1.58 & 105 & 60 & 1 & 21.90 & 11 & 10.07 + ngc 3023 & sabc & 5.5 & 147.469112 & 0.618167 & 1.44 & 70 & 60 & 1 & 29.40 & 2 & 12.33 + ngc 3031 & sab & 2.4 & 148.888221 & 69.065295 & 13.46 & 150 & 59 & 37 & 3.63 & 18 & 14.21 + ngc 3184 & sabc & 5.9 & 154.570258 & 41.424053 & 3.71 & 135 & 21 & 1 & 12.60 & 14 & 13.59 + ngc 3198 & sc & 5.2 & 154.978966 & 45.549623 & 4.26 & 35 & 67 & 1 & 14.50 & 18 & 17.95 + ngc 3227 & saba & 1.5 & 155.877412 & 19.865050 & 2.68 & 155 & 47 & 1 & 20.60 & 31 & 16.09 + ngc 3239 & ib & 9.8 & 156.270375 & 17.163583 & 2.51 & 80 & 49 & 38 & 8.10 & 31 & 5.90 + ngc 3310 & sabb & 4.0 & 159.691083 & 53.503383 & 1.54 & 163 & 39 & 39 & 19.20 & 2 & 8.63 + ngc 3319 & sbc & 6.0 & 159.789414 & 41.686675 & 3.08 & 37 & 57 & 1 & 14.10 & 18 & 12.64 + ngc 3344 & sbc & 4.0 & 160.879792 & 24.922222 & 3.54 & 156 & 25 & 40 & 6.10 & 31 & 6.28 + ngc 3351 & sb & 3.1 & 160.990417 & 11.703806 & 3.71 & 13 & 42 & 26 & 10.50 & 18 & 11.32 + ngc 3359 & sc & 5.2 & 161.653596 & 63.224236 & 3.62 & 172 & 51 & 41 & 14.40 & 42 & 15.17 + ngc 3486 & sc & 5.2 & 165.099452 & 28.975137 & 3.54 & 80 & 42 & 1 & 9.16 & 2 & 9.43 + ngc 3521 & sabb & 4.0 & 166.452421 & -0.035864 & 5.48 & 160 & 73 & 37 & 10.70 & 43 & 17.06 + ngc 3621 & sbcd & 6.9 & 169.568792 & -32.814056 & 4.89@xmath85 & 165 & 65 & 37 & 7.24 & 18 & 10.29 + ngc 3631 & sc & 5.2 & 170.261976 & 53.169569 & 2.51 & 152 & 20 & 86 & 22.2 & 2 & 16.18 + ngc 3718 & sa & 1.1 & 173.145221 & 53.067922 & 3.77@xmath86 & 15 & 65 & 44 & 17.00 & 31 & 18.63 + ngc 3820 & sbc & 3.6 & 175.520442 & 10.384253 & 0.33 & 26 & 42 & 45 & 91.30 & 2 & 8.77 + ngc 3893 & sabc & 5.2 & 177.159125 & 48.710833 & 2.23 & 165 & 52 & 1 & 18.20 & 11 & 11.82 + ngc 3938 & sc & 5.2 & 178.206042 & 44.120722 & 2.68 & 15 & 24 & 1 & 17.90 & 46 & 13.98 + ngc 4030 & sbc & 4.0 & 180.098445 & -1.100095 & 2.08 & 27 & 44 & 1 & 27.20 & 11 & 16.49 + ngc 4088 & sabc & 4.7 & 181.392458 & 50.539028 & 2.88 & 43 & 67 & 1 & 14.30 & 11 & 11.97 + ngc 4109 & sa & 1.1 & 181.713008 & 42.995642 & 0.50 & 147 & 44 & 45 & 102.80 & 2 & 14.95 + ngc 4254 & sc & 5.2 & 184.706683 & 14.416508 & 2.68 & 59 & 29 & 19 & 14.30 & 16 & 11.17 + ngc 4258 & sabb & 4.0 & 184.739600 & 47.303972 & 9.31 & 150 & 72 & 47 & 7.20 & 48 & 19.50 + ngc 4303 & sbc & 4.0 & 185.478729 & 4.473647 & 3.23 & 137 & 27 & 19 & 13.60 & 2 & 12.77 + ngc 4321 & sabb & 4.1 & 185.728462 & 15.821817 & 3.71 & 30 & 32 & 1 & 17.20 & 18 & 18.55 + ngc 4395 & sm & 8.9 & 186.453592 & 33.546928 & 6.59 & 147 & 46 & 49 & 4.30 & 50 & 8.24 + ngc 4490 & sbcd & 7.0 & 187.650996 & 41.643898 & 3.16 & 125 & 61 & 1 & 7.80 & 31 & 7.16 + ngc 4501 & sb & 3.3 & 187.996503 & 14.420387 & 3.46 & 141 & 63 & 51 & 14.10 & 52 & 14.19 + ngc 4535 & sc & 5.0 & 188.584625 & 8.197750 & 3.54 & 0 & 45 & 1 & 17.80 & 18 & 18.33 + ngc 4559 & sc & 6.0 & 188.990195 & 27.959992 & 5.36 & 143 & 67 & 53 & 7.87 & 16 & 12.27 + ngc 4625 & sabm & 8.8 & 190.469671 & 41.273964 & 1.09 & 130 & 27 & 54 & 9.20 & 18 & 2.93 + ngc 4631 & sbcd & 6.6 & 190.533375 & 32.541500 & 7.74 & 86 & 80 & 1 & 7.62 & 55 & 17.17 + ngc 4651 & sc & 5.2 & 190.927625 & 16.393389 & 1.99 & 78 & 48 & 19 & 23.60 & 56 & 13.67 + ngc 4654 & sc & 5.9 & 190.985750 & 13.126681 & 2.45 & 122 & 57 & 19 & 16.10 & 16 & 11.47 + ngc 4656 & sbm & 9.0 & 190.990531 & 32.168150 & 7.57 & 33 & 79 & 1 & 8.20 & 2 & 18.05 + ngc 4713 & scd & 6.8 & 192.491139 & 5.311405 & 1.35 & 50 & 35 & 57 & 14.90 & 11 & 5.83 + ngc 4725 & saba & 2.2 & 192.610755 & 25.500805 & 5.36 & 35 & 45 & 1 & 13.50 & 18 & 21.04 + ngc 4736 & sab & 2.4 & 192.721088 & 41.120458 & 5.61 & 116 & 37 & 58 & 4.66 & 24 & 7.60 + ngc 4861 & sm & 8.9 & 194.759754 & 34.859444 & 1.99 & 23 & 69 & 59 & 7.59 & 11 & 4.40 + ngc 5033 & sc & 5.1 & 198.364441 & 36.569936 & 5.36 & 170 & 62 & 1 & 18.70 & 16 & 29.14 + ngc 5055 & sbc & 4.0 & 198.955542 & 42.029278 & 6.30 & 101 & 64 & 33 & 7.94 & 11 & 14.54 + ngc 5068 & sc & 6.0 & 199.728375 & -21.039111 & 3.62 & 110 & 29 & 1 & 5.40 & 60 & 5.69 + ngc 5194 & sbc & 4.0 & 202.469629 & 47.195172 & 5.61 & 163 & 52 & 1 & 8.90 & 61 & 14.52 + ngc 5236 & sc & 5.0 & 204.253958 & -29.865417 & 6.44 & 46 & 24 & 58 & 4.61 & 18 & 8.64 + ngc 5248 & sabb & 4.0 & 204.383433 & 8.885178 & 3.08 & 110 & 44 & 1 & 22.90 & 2 & 20.54 + ngc 5457 & sabc & 6.0 & 210.802267 & 54.348950 & 14.42 & 37 & 18 & 62 & 6.85 & 18 & 28.73 + ngc 5474 & sc & 6.1 & 211.256708 & 53.662222 & 2.39 & 159 & 21 & 63 & 6.80 & 35 & 4.73 + ngc 5668 & scd & 6.9 & 218.351408 & 4.450458 & 1.66 & 164 & 21 & 64 & 26.90 & 31 & 12.96 + ngc 6384 & sabb & 3.6 & 263.101250 & 7.060278 & 3.08 & 32 & 48 & 19 & 25.90 & 11 & 23.23 + ngc 6691 & sbc & 3.6 & 279.801042 & 55.641806 & 0.79 & 0 & 21 & 1 & 85.10 & 2 & 19.62 + ngc 6744 & sbc & 4.0 & 287.442125 & -63.857528 & 9.98 & 20 & 50 & 65 & 9.40 & 16 & 27.28 + ngc 6946 & sabc & 5.9 & 308.718012 & 60.153914 & 5.74 & 62 & 38 & 66 & 6.00 & 67 & 10.02 + ngc 7331 & sbc & 3.9 & 339.266725 & 34.415519 & 5.24 & 169 & 75 & 28 & 15.10 & 68 & 23.00 + ngc 7495 & sabc & 5.3 & 347.238250 & 12.048028 & 0.91 & 5 & 24 & 1 & 70.00 & 69 & 18.53 + ngc 7518 & saba & 1.1 & 348.303042 & 6.321667 & 0.71 & 109 & 42 & 70 & 50.50 & 70 & 10.38 + ngc 7529 & sbc & 4.4 & 348.513292 & 8.992500 & 0.43 & 0 & 17 & 1 & 62.10 & 2 & 7.69 + ngc 7570 & sa & 1.1 & 349.186134 & 13.483011 & 0.77 & 30 & 54 & 1 & 17.70 & 16 & 3.99 + ngc 7591 & sbbc & 3.6 & 349.567833 & 6.585806 & 0.98 & 148 & 68 & 71 & 60.20 & 16 & 17.07 + ngc 7678 & sc & 4.9 & 352.116250 & 22.421194 & 1.17 & 5 & 45 & 1 & 48.90 & 2 & 16.67 + ngc 7793 & scd & 7.4 & 359.457625 & -32.591028 & 4.67 & 99 & 54 & 10 & 3.40 & 72 & 4.62 + ic 0010 & ib & 9.9 & 5.072250 & 59.303778 & 3.16 & 125 & 27 & 45 & 0.66 & 73 & 0.61 + ic 0193 & sc & 5.3 & 30.629083 & 11.093083 & 0.93 & 154 & 32 & 1 & 61.60 & 2 & 16.68 + ic 0208 & sbc & 3.6 & 32.115583 & 6.394917 & 0.91 & 0 & 12 & 1 & 38.30 & 16 & 10.14 + ic 0342 & sabc & 6.0 & 56.702096 & 68.096367 & 10.69 & 37 & 31 & 74 & 3.30 & 68 & 10.26 + ic 1132 & sc & 5.4 & 235.028171 & 20.680606 & 0.59 & 50 & 18 & 75 & 70.60 & 2 & 12.06 + ic 2204 & sab & 2.5 & 115.325435 & 34.232169 & 0.52 & 0 & 0 & 1 & 66.20 & 76 & 10.08 + ic 5201 & sc & 6.1 & 335.239333 & -46.035861 & 4.26 & 33 & 63 & 1 & 12.80 & 2 & 15.85 + ic 5309 & sb & 3.1 & 349.798542 & 8.109306 & 0.67 & 20 & 63 & 77 & 50.00 & 16 & 9.81 + ugc 00223 & i & 9.9 & 6.177915 & 14.824582 & 0.46 & 0 & 0 & 1 & 70.80 & 2 & 9.39 + ugc 01087 & sc & 5.3 & 22.860833 & 14.277500 & 0.74 & 0 & 21 & 1 & 59.60 & 2 & 12.82 + ugc 01862 & sabc & 6.4 & 36.103375 & -2.162361 & 0.83 & 10 & 39 & 1 & 20.10 & 16 & 4.85 + ugc 02023 & i & 9.7 & 37.572417 & 33.491111 & 0.83 & 59 & 26 & 78 & 10.30 & 79 & 2.49 + ugc 02216 & i & 9.9 & 41.087913 & 0.675381 & 0.64 & 15 & 72 & 1 & 36.50 & 2 & 6.84 + ugc 02345 & sm & 8.8 & 42.969583 & -1.172222 & 1.73 & 160 & 30 & 1 & 20.00 & 2 & 10.09 + ugc 03701 & sc & 6.0 & 107.927833 & 72.169306 & 0.91 & 0 & 0 & 1 & 43.20 & 2 & 11.43 + ugc 04107 & sc & 5.3 & 119.257792 & 49.567361 & 0.71 & 0 & 12 & 1 & 51.50 & 2 & 10.58 + ugc 04305 & i & 9.9 & 124.770750 & 70.720027 & 3.97 & 177 & 40 & 80 & 3.38 & 9 & 3.90 + ugc 05100 & sbb & 3.0 & 143.661008 & 5.841439 & 0.60 & 30 & 53 & 1 & 80.60 & 2 & 14.09 + ugc 06410 & sabc & 5.3 & 171.024500 & 45.811072 & 0.55 & 5 & 39 & 1 & 82.50 & 2 & 13.16 + ugc 08091 & i & 9.8 & 194.668512 & 14.217511 & 0.54 & 170 & 20 & 81 & 2.08 & 9 & 0.32 + ugc 09562 & sd & 8.0 & 222.810048 & 35.542269 & 0.48 & 30 & 27 & 1 & 23.80 & 31 & 3.31 + ugc 09837 & sabc & 5.3 & 230.965311 & 58.052941 & 0.91 & 21 & 18 & 82 & 42.00 & 82 & 11.12 + ugc 10445 & sc & 6.0 & 248.448417 & 28.984778 & 1.38 & 142 & 46 & 83 & 20.00 & 2 & 8.01 + ugc 12709 & sabm & 8.7 & 354.350001 & 0.391670 & 1.51 & 145 & 47 & 1 & 37.00 & 2 & 16.25 + pgc 29167 & i & 10.0 & 150.965417 & 68.690278 & 2.70@xmath87 & 0 & 0 & & 3.82 & 84 & 3.00 + pgc 41318 & s ? & 4.3 & 187.589848 & 56.791542 & 0.29 & 0 & 21 & 45 & 70.10 & 2 & 6.00 + pgc 44772 & s ? & 1.9 & 195.138542 & 10.129969 & 0.40 & 128 & 32 & 45 & 107.80 & 2 & 12.39 + pgc 45195 & sd & 7.7 & 196.129979 & -3.572419 & 1.77 & 63 & 44 & 1 & 24.10 & 2 & 12.44 + pgc 46182 & sc & 4.9 & 199.061225 & 41.494460 & 0.54@xmath88 & 34 & 52 & 85 & 89.40 & 76 & 14.15 + ; lccrrrrcc ngc 0012 & 8.56 @xmath73 0.02 & 0.015 @xmath73 0.042 & 0.0011 @xmath73 0.0032 & 0.033 & 7.74 @xmath73 0.04 & -0.008 @xmath73 0.087 & -0.0006 @xmath73 0.0066 & 0.068 + ngc 0055 & 8.04 @xmath73 0.05 & -0.053 @xmath73 0.094 & -0.0059 @xmath73 0.0104 & 0.075 & 6.50 @xmath73 0.07 & -0.037 @xmath73 0.131 & -0.0042 @xmath73 0.0145 & 0.091 + ngc 0099 & 8.55 @xmath73 0.04 & -0.319 @xmath73 0.072 & -0.0202 @xmath73 0.0045 & 0.055 & 7.70 @xmath73 0.07 & -0.666 @xmath73 0.137 & -0.0422 @xmath73 0.0087 & 0.105 + ngc 0224 & 8.72 @xmath73 0.02 & -0.352 @xmath73 0.039 & -0.0172 @xmath73 0.0019 & 0.077 & 8.17 @xmath73 0.04 & -0.996 @xmath73 0.072 & -0.0486 @xmath73 0.0035 & 0.136 + ngc 0234 & 8.65 @xmath73 0.01 & -0.154 @xmath73 0.022 & -0.0109 @xmath73 0.0015 & 0.027 & 7.91 @xmath73 0.02 & -0.329 @xmath73 0.046 & -0.0234 @xmath73 0.0033 & 0.058 + ngc 0253 & 8.55 @xmath73 0.13 & -0.125 @xmath73 0.164 & -0.0090 @xmath73 0.0119 & 0.055 & 7.60 @xmath73 0.35 & -0.241 @xmath73 0.441 & -0.0174 @xmath73 0.0319 & 0.149 + ngc 0300 & 8.51 @xmath73 0.02 & -0.519 @xmath73 0.040 & -0.0842 @xmath73 0.0065 & 0.069 & 7.47 @xmath73 0.04 & -0.955 @xmath73 0.080 & -0.1548 @xmath73 0.0129 & 0.130 + ngc 0428 & 8.20 @xmath73 0.06 & -0.035 @xmath73 0.109 & -0.0040 @xmath73 0.0124 & 0.059 & 6.95 @xmath73 0.11 & -0.118 @xmath73 0.184 & -0.0135 @xmath73 0.0210 & 0.099 + ngc 0450 & 8.35 @xmath73 0.03 & -0.285 @xmath73 0.057 & -0.0340 @xmath73 0.0067 & 0.046 & 7.24 @xmath73 0.06 & -0.589 @xmath73 0.112 & -0.0702 @xmath73 0.0133 & 0.092 + ngc 0493 & 8.51 @xmath73 0.04 & -0.450 @xmath73 0.070 & -0.0393 @xmath73 0.0061 & 0.053 & 7.53 @xmath73 0.07 & -0.893 @xmath73 0.124 & -0.0780 @xmath73 0.0109 & 0.093 + ngc 0575 & 8.58 @xmath73 0.01 & -0.195 @xmath73 0.026 & -0.0189 @xmath73 0.0025 & 0.026 & 7.82 @xmath73 0.03 & -0.530 @xmath73 0.061 & -0.0514 @xmath73 0.0059 & 0.060 + ngc 0598 & 8.48 @xmath73 0.02 & -0.359 @xmath73 0.041 & -0.0459 @xmath73 0.0053 & 0.073 & 7.47 @xmath73 0.04 & -0.691 @xmath73 0.084 & -0.0883 @xmath73 0.0107 & 0.145 + ngc 0628 & 8.78 @xmath73 0.01 & -0.572 @xmath73 0.027 & -0.0379 @xmath73 0.0018 & 0.060 & 8.17 @xmath73 0.03 & -1.292 @xmath73 0.064 & -0.0856 @xmath73 0.0042 & 0.137 + ngc 0753 & 8.67 @xmath73 0.08 & -0.212 @xmath73 0.106 & -0.0087 @xmath73 0.0043 & 0.065 & 8.01 @xmath73 0.20 & -0.399 @xmath73 0.290 & -0.0163 @xmath73 0.0118 & 0.177 + ngc 0783 & 8.68 @xmath73 0.07 & -0.189 @xmath73 0.104 & -0.0118 @xmath73 0.0056 & 0.031 & 7.91 @xmath73 0.30 & -0.274 @xmath73 0.434 & -0.0171 @xmath73 0.0272 & 0.130 + ngc 0925 & 8.48 @xmath73 0.02 & -0.473 @xmath73 0.037 & -0.0334 @xmath73 0.0026 & 0.067 & 7.45 @xmath73 0.04 & -0.941 @xmath73 0.071 & -0.0665 @xmath73 0.0050 & 0.128 + ngc 1055 & 8.62 @xmath73 0.04 & -0.140 @xmath73 0.100 & -0.0088 @xmath73 0.0063 & 0.059 & 7.89 @xmath73 0.05 & -0.414 @xmath73 0.118 & -0.0261 @xmath73 0.0074 & 0.070 + ngc 1058 & 8.62 @xmath73 0.01 & -0.352 @xmath73 0.027 & -0.0757 @xmath73 0.0058 & 0.055 & 7.72 @xmath73 0.03 & -0.751 @xmath73 0.069 & -0.1613 @xmath73 0.0149 & 0.137 + ngc 1068 & 8.64 @xmath73 0.02 & -0.121 @xmath73 0.051 & -0.0116 @xmath73 0.0049 & 0.070 & 7.94 @xmath73 0.04 & -0.141 @xmath73 0.100 & -0.0136 @xmath73 0.0096 & 0.126 + ngc 1090 & 8.61 @xmath73 0.03 & -0.284 @xmath73 0.103 & -0.0137 @xmath73 0.0050 & 0.026 & 7.89 @xmath73 0.04 & -0.775 @xmath73 0.131 & -0.0373 @xmath73 0.0063 & 0.033 + ngc 1097 & 8.74 @xmath73 0.01 & -0.357 @xmath73 0.054 & -0.0159 @xmath73 0.0024 & 0.030 & 8.25 @xmath73 0.05 & -0.997 @xmath73 0.209 & -0.0445 @xmath73 0.0093 & 0.114 + ngc 1232 & 8.78 @xmath73 0.04 & -0.573 @xmath73 0.068 & -0.0253 @xmath73 0.0030 & 0.040 & 8.12 @xmath73 0.10 & -1.166 @xmath73 0.172 & -0.0515 @xmath73 0.0076 & 0.103 + ngc 1313 & 8.21 @xmath73 0.03 & -0.159 @xmath73 0.051 & -0.0273 @xmath73 0.0087 & 0.071 & 6.84 @xmath73 0.05 & -0.269 @xmath73 0.096 & -0.0462 @xmath73 0.0165 & 0.132 + ngc 1365 & 8.64 @xmath73 0.02 & -0.253 @xmath73 0.033 & -0.0079 @xmath73 0.0010 & 0.064 & 7.91 @xmath73 0.04 & -0.622 @xmath73 0.080 & -0.0193 @xmath73 0.0024 & 0.137 + ngc 1512 & 8.79 @xmath73 0.03 & -0.461 @xmath73 0.048 & -0.0374 @xmath73 0.0039 & 0.066 & 8.42 @xmath73 0.05 & -1.410 @xmath73 0.083 & -0.1144 @xmath73 0.0067 & 0.115 + ngc 1598 & 8.74 @xmath73 0.03 & -0.402 @xmath73 0.062 & -0.0343 @xmath73 0.0053 & 0.042 & 8.20 @xmath73 0.09 & -1.014 @xmath73 0.178 & -0.0865 @xmath73 0.0152 & 0.122 + ngc 1637 & 8.64 @xmath73 0.01 & -0.115 @xmath73 0.023 & -0.0165 @xmath73 0.0034 & 0.062 & 7.78 @xmath73 0.03 & -0.170 @xmath73 0.067 & -0.0236 @xmath73 0.0096 & 0.160 + ngc 1642 & 8.64 @xmath73 0.01 & -0.154 @xmath73 0.024 & -0.0094 @xmath73 0.0015 & 0.030 & 7.86 @xmath73 0.02 & -0.244 @xmath73 0.050 & -0.0148 @xmath73 0.0031 & 0.062 + ngc 1672 & 8.67 @xmath73 0.02 & -0.216 @xmath73 0.070 & -0.0135 @xmath73 0.0044 & 0.055 & 8.00 @xmath73 0.05 & -0.464 @xmath73 0.158 & -0.0290 @xmath73 0.0098 & 0.119 + ngc 2336 & 8.80 @xmath73 0.04 & -0.390 @xmath73 0.056 & -0.0156 @xmath73 0.0022 & 0.041 & 8.14 @xmath73 0.08 & -0.725 @xmath73 0.121 & -0.0290 @xmath73 0.0049 & 0.088 + ngc 2403 & 8.48 @xmath73 0.02 & -0.524 @xmath73 0.043 & -0.0538 @xmath73 0.0044 & 0.079 & 7.47 @xmath73 0.03 & -1.005 @xmath73 0.069 & -0.1032 @xmath73 0.0071 & 0.128 + ngc 2441 & 8.54 @xmath73 0.01 & 0.017 @xmath73 0.040 & 0.0011 @xmath73 0.0027 & 0.036 & 7.67 @xmath73 0.03 & 0.100 @xmath73 0.080 & 0.0068 @xmath73 0.0054 & 0.072 + ngc 2442 & 8.64 @xmath73 0.05 & -0.079 @xmath73 0.072 & -0.0058 @xmath73 0.0052 & 0.030 & 7.87 @xmath73 0.14 & -0.119 @xmath73 0.204 & -0.0087 @xmath73 0.0150 & 0.087 + ngc 2541 & 8.45 @xmath73 0.02 & -0.608 @xmath73 0.053 & -0.0526 @xmath73 0.0046 & 0.030 & 0.00 @xmath73 0.00 & 0.000 @xmath73 0.000 & 0.0000 @xmath73 0.0000 & 0.000 + ngc 2805 & 8.51 @xmath73 0.03 & -0.432 @xmath73 0.045 & -0.0168 @xmath73 0.0017 & 0.040 & 7.53 @xmath73 0.06 & -0.869 @xmath73 0.096 & -0.0338 @xmath73 0.0037 & 0.087 + ngc 2835 & 8.50 @xmath73 0.04 & -0.314 @xmath73 0.060 & -0.0324 @xmath73 0.0062 & 0.057 & 7.51 @xmath73 0.10 & -0.633 @xmath73 0.149 & -0.0652 @xmath73 0.0154 & 0.142 + ngc 2841 & 8.99 @xmath73 0.08 & -0.607 @xmath73 0.172 & -0.0363 @xmath73 0.0103 & 0.022 & 8.81 @xmath73 0.20 & -1.488 @xmath73 0.450 & -0.0890 @xmath73 0.0269 & 0.052 + ngc 2903 & 8.82 @xmath73 0.03 & -0.523 @xmath73 0.072 & -0.0321 @xmath73 0.0044 & 0.061 & 8.51 @xmath73 0.06 & -1.588 @xmath73 0.152 & -0.0975 @xmath73 0.0093 & 0.121 + ngc 2997 & 8.80 @xmath73 0.06 & -0.502 @xmath73 0.121 & -0.0318 @xmath73 0.0076 & 0.060 & 8.44 @xmath73 0.13 & -1.503 @xmath73 0.274 & -0.0950 @xmath73 0.0173 & 0.129 + ngc 3020 & 8.39 @xmath73 0.02 & -0.558 @xmath73 0.041 & -0.0554 @xmath73 0.0041 & 0.011 & 7.27 @xmath73 0.07 & -1.053 @xmath73 0.167 & -0.1045 @xmath73 0.0167 & 0.045 + ngc 3023 & 8.34 @xmath73 0.08 & -0.444 @xmath73 0.201 & -0.0361 @xmath73 0.0163 & 0.054 & 7.20 @xmath73 0.14 & -0.889 @xmath73 0.372 & -0.0722 @xmath73 0.0302 & 0.100 + ngc 3031 & 8.58 @xmath73 0.02 & -0.156 @xmath73 0.045 & -0.0110 @xmath73 0.0032 & 0.043 & 7.84 @xmath73 0.06 & -0.464 @xmath73 0.116 & -0.0327 @xmath73 0.0081 & 0.110 + ngc 3184 & 8.66 @xmath73 0.02 & -0.176 @xmath73 0.037 & -0.0130 @xmath73 0.0027 & 0.066 & 7.86 @xmath73 0.04 & -0.338 @xmath73 0.086 & -0.0249 @xmath73 0.0063 & 0.137 + ngc 3198 & 8.60 @xmath73 0.04 & -0.425 @xmath73 0.076 & -0.0237 @xmath73 0.0042 & 0.063 & 7.94 @xmath73 0.11 & -1.145 @xmath73 0.317 & -0.0638 @xmath73 0.0176 & 0.091 + ngc 3227 & 8.64 @xmath73 0.05 & -0.254 @xmath73 0.077 & -0.0158 @xmath73 0.0048 & 0.056 & 0.00 @xmath73 0.00 & 0.000 @xmath73 0.000 & 0.0000 @xmath73 0.0000 & 0.000 + ngc 3239 & 8.12 @xmath73 0.01 & -0.153 @xmath73 0.012 & -0.0260 @xmath73 0.0021 & 0.005 & 6.78 @xmath73 0.04 & -0.275 @xmath73 0.064 & -0.0467 @xmath73 0.0108 & 0.026 + ngc 3310 & 8.37 @xmath73 0.01 & -0.101 @xmath73 0.023 & -0.0117 @xmath73 0.0027 & 0.040 & 7.27 @xmath73 0.02 & -0.254 @xmath73 0.060 & -0.0295 @xmath73 0.0070 & 0.104 + ngc 3319 & 8.50 @xmath73 0.02 & -0.516 @xmath73 0.043 & -0.0408 @xmath73 0.0034 & 0.024 & 0.00 @xmath73 0.00 & 0.000 @xmath73 0.000 & 0.0000 @xmath73 0.0000 & 0.000 + ngc 3344 & 8.72 @xmath73 0.02 & -0.618 @xmath73 0.039 & -0.0984 @xmath73 0.0063 & 0.029 & 8.08 @xmath73 0.17 & -1.413 @xmath73 0.252 & -0.2254 @xmath73 0.0407 & 0.044 + ngc 3351 & 8.82 @xmath73 0.01 & -0.210 @xmath73 0.031 & -0.0186 @xmath73 0.0027 & 0.038 & 8.51 @xmath73 0.03 & -0.678 @xmath73 0.094 & -0.0599 @xmath73 0.0083 & 0.116 + ngc 3359 & 8.40 @xmath73 0.06 & -0.271 @xmath73 0.108 & -0.0179 @xmath73 0.0071 & 0.099 & 7.43 @xmath73 0.09 & -0.688 @xmath73 0.154 & -0.0453 @xmath73 0.0102 & 0.130 + ngc 3486 & 8.60 @xmath73 0.02 & -0.421 @xmath73 0.043 & -0.0447 @xmath73 0.0046 & 0.018 & 7.80 @xmath73 0.04 & -0.928 @xmath73 0.088 & -0.0985 @xmath73 0.0094 & 0.036 + ngc 3521 & 8.83 @xmath73 0.04 & -0.666 @xmath73 0.096 & -0.0390 @xmath73 0.0056 & 0.048 & 0.00 @xmath73 0.00 & 0.000 @xmath73 0.000 & 0.0000 @xmath73 0.0000 & 0.000 + ngc 3621 & 8.68 @xmath73 0.02 & -0.460 @xmath73 0.035 & -0.0446 @xmath73 0.0034 & 0.050 & 8.05 @xmath73 0.05 & -1.298 @xmath73 0.095 & -0.1260 @xmath73 0.0092 & 0.133 + ngc 3631 & 8.71 @xmath73 0.03 & -0.414 @xmath73 0.077 & -0.0256 @xmath73 0.0048 & 0.024 & 7.94 @xmath73 0.05 & -0.719 @xmath73 0.145 & -0.0445 @xmath73 0.0090 & 0.046 + ngc 3718 & 8.60 @xmath73 0.19 & -0.221 @xmath73 0.253 & -0.0119 @xmath73 0.0136 & 0.066 & 0.00 @xmath73 0.00 & 0.000 @xmath73 0.000 & 0.0000 @xmath73 0.0000 & 0.000 + ngc 3820 & 8.64 @xmath73 0.01 & -0.102 @xmath73 0.019 & -0.0116 @xmath73 0.0022 & 0.019 & 7.92 @xmath73 0.05 & -0.342 @xmath73 0.093 & -0.0389 @xmath73 0.0105 & 0.092 + ngc 3893 & 8.73 @xmath73 0.16 & -0.437 @xmath73 0.254 & -0.0371 @xmath73 0.0215 & 0.028 & 0.00 @xmath73 0.00 & 0.000 @xmath73 0.000 & 0.0000 @xmath73 0.0000 & 0.000 + ngc 3938 & 8.79 @xmath73 0.05 & -0.577 @xmath73 0.092 & -0.0413 @xmath73 0.0066 & 0.031 & 8.15 @xmath73 0.07 & -1.204 @xmath73 0.141 & -0.0861 @xmath73 0.0101 & 0.048 + ngc 4030 & 8.77 @xmath73 0.02 & -0.258 @xmath73 0.036 & -0.0156 @xmath73 0.0022 & 0.010 & 8.05 @xmath73 0.08 & -0.344 @xmath73 0.177 & -0.0208 @xmath73 0.0107 & 0.052 + ngc 4088 & 8.71 @xmath73 0.04 & -0.366 @xmath73 0.093 & -0.0305 @xmath73 0.0078 & 0.025 & 7.97 @xmath73 0.06 & -0.698 @xmath73 0.116 & -0.0582 @xmath73 0.0097 & 0.032 + ngc 4109 & 8.63 @xmath73 0.03 & -0.402 @xmath73 0.099 & -0.0269 @xmath73 0.0066 & 0.056 & 7.87 @xmath73 0.05 & -1.152 @xmath73 0.154 & -0.0770 @xmath73 0.0103 & 0.081 + ngc 4254 & 8.77 @xmath73 0.02 & -0.353 @xmath73 0.034 & -0.0316 @xmath73 0.0031 & 0.036 & 8.09 @xmath73 0.05 & -0.716 @xmath73 0.100 & -0.0641 @xmath73 0.0090 & 0.106 + ngc 4258 & 8.54 @xmath73 0.03 & -0.161 @xmath73 0.073 & -0.0083 @xmath73 0.0037 & 0.031 & 7.59 @xmath73 0.13 & -0.232 @xmath73 0.282 & -0.0119 @xmath73 0.0144 & 0.120 + ngc 4303 & 8.78 @xmath73 0.03 & -0.577 @xmath73 0.073 & -0.0452 @xmath73 0.0057 & 0.062 & 8.24 @xmath73 0.08 & -1.370 @xmath73 0.189 & -0.1073 @xmath73 0.0148 & 0.156 + ngc 4321 & 8.74 @xmath73 0.03 & -0.219 @xmath73 0.048 & -0.0118 @xmath73 0.0026 & 0.038 & 8.04 @xmath73 0.06 & -0.355 @xmath73 0.104 & -0.0192 @xmath73 0.0056 & 0.083 + ngc 4395 & 8.19 @xmath73 0.05 & -0.164 @xmath73 0.082 & -0.0200 @xmath73 0.0100 & 0.079 & 6.89 @xmath73 0.08 & -0.346 @xmath73 0.143 & -0.0420 @xmath73 0.0174 & 0.136 + ngc 4490 & 8.29 @xmath73 0.08 & -0.318 @xmath73 0.254 & -0.0444 @xmath73 0.0355 & 0.058 & 7.09 @xmath73 0.15 & -0.560 @xmath73 0.449 & -0.0781 @xmath73 0.0629 & 0.102 + ngc 4501 & 8.92 @xmath73 0.12 & -0.424 @xmath73 0.242 & -0.0299 @xmath73 0.0171 & 0.054 & 0.00 @xmath73 0.00 & 0.000 @xmath73 0.000 & 0.0000 @xmath73 0.0000 & 0.000 + ngc 4535 & 8.71 @xmath73 0.03 & -0.299 @xmath73 0.064 & -0.0163 @xmath73 0.0035 & 0.025 & 7.97 @xmath73 0.03 & -0.594 @xmath73 0.055 & -0.0324 @xmath73 0.0030 & 0.022 + ngc 4559 & 8.53 @xmath73 0.03 & -0.573 @xmath73 0.056 & -0.0467 @xmath73 0.0046 & 0.047 & 0.00 @xmath73 0.00 & 0.000 @xmath73 0.000 & 0.0000 @xmath73 0.0000 & 0.000 + ngc 4625 & 8.58 @xmath73 0.01 & -0.130 @xmath73 0.044 & -0.0443 @xmath73 0.0151 & 0.042 & 7.84 @xmath73 0.05 & -0.680 @xmath73 0.143 & -0.2318 @xmath73 0.0488 & 0.127 + ngc 4631 & 8.39 @xmath73 0.06 & -0.333 @xmath73 0.125 & -0.0194 @xmath73 0.0073 & 0.077 & 7.18 @xmath73 0.09 & -0.480 @xmath73 0.169 & -0.0280 @xmath73 0.0098 & 0.093 + ngc 4651 & 8.68 @xmath73 0.04 & -0.469 @xmath73 0.096 & -0.0343 @xmath73 0.0070 & 0.049 & 8.02 @xmath73 0.10 & -1.133 @xmath73 0.233 & -0.0829 @xmath73 0.0170 & 0.119 + ngc 4654 & 8.66 @xmath73 0.04 & -0.364 @xmath73 0.083 & -0.0318 @xmath73 0.0073 & 0.044 & 7.90 @xmath73 0.11 & -0.797 @xmath73 0.229 & -0.0695 @xmath73 0.0200 & 0.120 + ngc 4656 & 8.06 @xmath73 0.16 & -0.551 @xmath73 0.280 & -0.0305 @xmath73 0.0155 & 0.084 & 6.62 @xmath73 0.24 & -0.634 @xmath73 0.409 & -0.0351 @xmath73 0.0227 & 0.123 + ngc 4713 & 8.53 @xmath73 0.09 & -0.771 @xmath73 0.250 & -0.1325 @xmath73 0.0428 & 0.095 & 7.60 @xmath73 0.14 & -1.548 @xmath73 0.397 & -0.2662 @xmath73 0.0679 & 0.150 + ngc 4725 & 8.83 @xmath73 0.37 & -0.472 @xmath73 0.853 & -0.0223 @xmath73 0.0404 & 0.070 & 0.00 @xmath73 0.00 & 0.000 @xmath73 0.000 & 0.0000 @xmath73 0.0000 & 0.000 + ngc 4736 & 8.57 @xmath73 0.03 & -0.066 @xmath73 0.170 & -0.0086 @xmath73 0.0224 & 0.065 & 0.00 @xmath73 0.00 & 0.000 @xmath73 0.000 & 0.0000 @xmath73 0.0000 & 0.000 + ngc 4861 & 8.00 @xmath73 0.06 & -0.013 @xmath73 0.092 & -0.0030 @xmath73 0.0210 & 0.090 & 6.54 @xmath73 0.13 & -0.058 @xmath73 0.200 & -0.0132 @xmath73 0.0454 & 0.193 + ngc 5033 & 8.64 @xmath73 0.06 & -0.278 @xmath73 0.094 & -0.0095 @xmath73 0.0032 & 0.090 & 8.00 @xmath73 0.10 & -0.599 @xmath73 0.130 & -0.0206 @xmath73 0.0045 & 0.072 + ngc 5055 & 8.87 @xmath73 0.02 & -0.490 @xmath73 0.031 & -0.0337 @xmath73 0.0021 & 0.007 & 8.42 @xmath73 0.10 & -1.259 @xmath73 0.186 & -0.0867 @xmath73 0.0129 & 0.044 + ngc 5068 & 8.57 @xmath73 0.04 & -0.355 @xmath73 0.098 & -0.0626 @xmath73 0.0173 & 0.050 & 7.53 @xmath73 0.11 & -0.451 @xmath73 0.263 & -0.0794 @xmath73 0.0463 & 0.134 + ngc 5194 & 8.88 @xmath73 0.03 & -0.324 @xmath73 0.053 & -0.0223 @xmath73 0.0037 & 0.062 & 8.52 @xmath73 0.05 & -0.704 @xmath73 0.102 & -0.0484 @xmath73 0.0070 & 0.117 + ngc 5236 & 8.78 @xmath73 0.01 & -0.221 @xmath73 0.028 & -0.0256 @xmath73 0.0032 & 0.050 & 8.34 @xmath73 0.03 & -0.649 @xmath73 0.058 & -0.0752 @xmath73 0.0068 & 0.104 + ngc 5248 & 8.64 @xmath73 0.03 & -0.142 @xmath73 0.076 & -0.0069 @xmath73 0.0037 & 0.072 & 7.74 @xmath73 0.05 & -0.132 @xmath73 0.155 & -0.0064 @xmath73 0.0075 & 0.146 + ngc 5457 & 8.71 @xmath73 0.01 & -0.840 @xmath73 0.029 & -0.0293 @xmath73 0.0010 & 0.076 & 8.01 @xmath73 0.03 & -1.760 @xmath73 0.056 & -0.0613 @xmath73 0.0020 & 0.142 + ngc 5474 & 8.19 @xmath73 0.02 & -0.008 @xmath73 0.086 & -0.0019 @xmath73 0.0182 & 0.100 & 6.82 @xmath73 0.03 & -0.261 @xmath73 0.144 & -0.0553 @xmath73 0.0303 & 0.156 + ngc 5668 & 8.52 @xmath73 0.03 & -0.471 @xmath73 0.061 & -0.0364 @xmath73 0.0047 & 0.075 & 7.48 @xmath73 0.05 & -1.032 @xmath73 0.127 & -0.0796 @xmath73 0.0098 & 0.147 + ngc 6384 & 8.87 @xmath73 0.04 & -0.531 @xmath73 0.068 & -0.0228 @xmath73 0.0029 & 0.029 & 8.41 @xmath73 0.12 & -1.172 @xmath73 0.197 & -0.0504 @xmath73 0.0084 & 0.083 + ngc 6691 & 8.66 @xmath73 0.02 & -0.152 @xmath73 0.052 & -0.0077 @xmath73 0.0026 & 0.050 & 7.98 @xmath73 0.06 & -0.414 @xmath73 0.157 & -0.0211 @xmath73 0.0080 & 0.142 + ngc 6744 & 8.88 @xmath73 0.03 & -0.726 @xmath73 0.060 & -0.0266 @xmath73 0.0022 & 0.034 & 8.66 @xmath73 0.07 & -1.996 @xmath73 0.157 & -0.0731 @xmath73 0.0057 & 0.087 + ngc 6946 & 8.72 @xmath73 0.06 & -0.337 @xmath73 0.099 & -0.0337 @xmath73 0.0099 & 0.061 & 8.05 @xmath73 0.15 & -0.751 @xmath73 0.237 & -0.0751 @xmath73 0.0237 & 0.148 + ngc 7331 & 8.67 @xmath73 0.08 & -0.222 @xmath73 0.193 & -0.0097 @xmath73 0.0084 & 0.044 & 7.75 @xmath73 0.09 & -0.056 @xmath73 0.213 & -0.0024 @xmath73 0.0092 & 0.048 + ngc 7495 & 8.61 @xmath73 0.01 & -0.260 @xmath73 0.027 & -0.0140 @xmath73 0.0015 & 0.028 & 7.83 @xmath73 0.02 & -0.616 @xmath73 0.062 & -0.0333 @xmath73 0.0033 & 0.064 + ngc 7518 & 8.74 @xmath73 0.02 & -0.127 @xmath73 0.045 & -0.0122 @xmath73 0.0043 & 0.034 & 8.30 @xmath73 0.07 & -0.514 @xmath73 0.136 & -0.0496 @xmath73 0.0131 & 0.104 + ngc 7529 & 8.64 @xmath73 0.04 & -0.232 @xmath73 0.067 & -0.0303 @xmath73 0.0087 & 0.040 & 7.97 @xmath73 0.13 & -0.752 @xmath73 0.223 & -0.0979 @xmath73 0.0290 & 0.133 + ngc 7570 & 8.58 @xmath73 0.03 & -0.086 @xmath73 0.045 & -0.0217 @xmath73 0.0114 & 0.029 & 7.78 @xmath73 0.06 & -0.219 @xmath73 0.106 & -0.0549 @xmath73 0.0265 & 0.068 + ngc 7591 & 8.71 @xmath73 0.01 & -0.146 @xmath73 0.029 & -0.0085 @xmath73 0.0017 & 0.024 & 8.13 @xmath73 0.06 & -0.400 @xmath73 0.114 & -0.0234 @xmath73 0.0066 & 0.093 + ngc 7678 & 8.55 @xmath73 0.04 & -0.079 @xmath73 0.078 & -0.0048 @xmath73 0.0047 & 0.030 & 7.64 @xmath73 0.14 & -0.081 @xmath73 0.248 & -0.0049 @xmath73 0.0149 & 0.094 + ngc 7793 & 8.50 @xmath73 0.02 & -0.305 @xmath73 0.048 & -0.0662 @xmath73 0.0104 & 0.071 & 7.52 @xmath73 0.05 & -0.752 @xmath73 0.095 & -0.1631 @xmath73 0.0207 & 0.135 + ic 0010 & 8.14 @xmath73 0.08 & 0.252 @xmath73 0.185 & 0.4294 @xmath73 0.3025 & 0.094 & 6.90 @xmath73 0.15 & 0.355 @xmath73 0.338 & 0.6035 @xmath73 0.5534 & 0.165 + ic 0193 & 8.59 @xmath73 0.02 & -0.139 @xmath73 0.064 & -0.0083 @xmath73 0.0039 & 0.039 & 7.80 @xmath73 0.05 & -0.306 @xmath73 0.151 & -0.0183 @xmath73 0.0090 & 0.092 + ic 0208 & 8.58 @xmath73 0.01 & -0.168 @xmath73 0.035 & -0.0165 @xmath73 0.0034 & 0.036 & 7.78 @xmath73 0.03 & -0.422 @xmath73 0.083 & -0.0416 @xmath73 0.0082 & 0.086 + ic 0342 & 8.83 @xmath73 0.04 & -0.513 @xmath73 0.087 & -0.0500 @xmath73 0.0085 & 0.045 & 8.46 @xmath73 0.07 & -1.465 @xmath73 0.131 & -0.1426 @xmath73 0.0128 & 0.068 + ic 1132 & 8.65 @xmath73 0.03 & -0.390 @xmath73 0.059 & -0.0323 @xmath73 0.0049 & 0.077 & 7.69 @xmath73 0.06 & -0.598 @xmath73 0.106 & -0.0495 @xmath73 0.0087 & 0.128 + ic 2204 & 8.62 @xmath73 0.01 & -0.160 @xmath73 0.033 & -0.0159 @xmath73 0.0033 & 0.024 & 7.89 @xmath73 0.03 & -0.387 @xmath73 0.086 & -0.0384 @xmath73 0.0086 & 0.064 + ic 5201 & 8.35 @xmath73 0.07 & -0.501 @xmath73 0.147 & -0.0316 @xmath73 0.0093 & 0.074 & 7.19 @xmath73 0.12 & -0.908 @xmath73 0.242 & -0.0573 @xmath73 0.0153 & 0.122 + ic 5309 & 8.70 @xmath73 0.03 & -0.058 @xmath73 0.066 & -0.0059 @xmath73 0.0067 & 0.059 & 8.08 @xmath73 0.09 & -0.211 @xmath73 0.206 & -0.0215 @xmath73 0.0210 & 0.170 + ugc 00223 & 8.53 @xmath73 0.02 & -0.099 @xmath73 0.053 & -0.0105 @xmath73 0.0057 & 0.029 & 7.72 @xmath73 0.06 & -0.295 @xmath73 0.152 & -0.0313 @xmath73 0.0162 & 0.084 + ugc 01087 & 8.60 @xmath73 0.01 & -0.246 @xmath73 0.023 & -0.0192 @xmath73 0.0018 & 0.026 & 7.83 @xmath73 0.03 & -0.609 @xmath73 0.058 & -0.0475 @xmath73 0.0045 & 0.066 + ugc 01862 & 8.43 @xmath73 0.02 & -0.094 @xmath73 0.039 & -0.0194 @xmath73 0.0081 & 0.040 & 7.42 @xmath73 0.04 & -0.316 @xmath73 0.090 & -0.0651 @xmath73 0.0186 & 0.092 + ugc 02023 & 8.13 @xmath73 0.19 & -0.088 @xmath73 0.280 & -0.0358 @xmath73 0.1128 & 0.075 & 6.84 @xmath73 0.29 & -0.197 @xmath73 0.434 & -0.0800 @xmath73 0.1752 & 0.115 + ugc 02216 & 8.08 @xmath73 0.02 & -0.068 @xmath73 0.052 & -0.0099 @xmath73 0.0077 & 0.024 & 6.72 @xmath73 0.04 & -0.124 @xmath73 0.094 & -0.0181 @xmath73 0.0137 & 0.041 + ugc 02345 & 8.04 @xmath73 0.03 & -0.893 @xmath73 0.085 & -0.0885 @xmath73 0.0084 & 0.041 & 6.63 @xmath73 0.11 & -0.846 @xmath73 0.255 & -0.0838 @xmath73 0.0253 & 0.131 + ugc 03701 & 8.41 @xmath73 0.01 & -0.087 @xmath73 0.046 & -0.0076 @xmath73 0.0040 & 0.042 & 7.41 @xmath73 0.03 & -0.329 @xmath73 0.102 & -0.0288 @xmath73 0.0089 & 0.092 + ugc 04107 & 8.56 @xmath73 0.01 & -0.231 @xmath73 0.041 & -0.0219 @xmath73 0.0039 & 0.045 & 7.72 @xmath73 0.03 & -0.498 @xmath73 0.087 & -0.0470 @xmath73 0.0083 & 0.096 + ugc 04305 & 7.89 @xmath73 0.07 & -0.225 @xmath73 0.137 & -0.0573 @xmath73 0.0352 & 0.075 & 6.41 @xmath73 0.13 & -0.307 @xmath73 0.246 & -0.0784 @xmath73 0.0629 & 0.139 + ugc 05100 & 8.58 @xmath73 0.05 & -0.078 @xmath73 0.075 & -0.0056 @xmath73 0.0053 & 0.047 & 7.90 @xmath73 0.13 & -0.220 @xmath73 0.204 & -0.0156 @xmath73 0.0145 & 0.119 + ugc 06410 & 8.58 @xmath73 0.02 & -0.148 @xmath73 0.040 & -0.0113 @xmath73 0.0031 & 0.057 & 7.78 @xmath73 0.05 & -0.454 @xmath73 0.093 & -0.0345 @xmath73 0.0071 & 0.123 + ugc 08091 & 7.58 @xmath73 0.04 & 0.177 @xmath73 0.067 & 0.5487 @xmath73 0.2003 & 0.055 & 6.03 @xmath73 0.08 & 0.255 @xmath73 0.129 & 0.7891 @xmath73 0.3905 & 0.108 + ugc 09562 & 8.22 @xmath73 0.05 & -0.193 @xmath73 0.143 & -0.0585 @xmath73 0.0433 & 0.080 & 6.80 @xmath73 0.06 & -0.206 @xmath73 0.144 & -0.0626 @xmath73 0.0437 & 0.068 + ugc 09837 & 8.45 @xmath73 0.02 & -0.412 @xmath73 0.076 & -0.0370 @xmath73 0.0068 & 0.078 & 7.32 @xmath73 0.04 & -0.820 @xmath73 0.140 & -0.0738 @xmath73 0.0126 & 0.141 + ugc 10445 & 8.27 @xmath73 0.04 & -0.082 @xmath73 0.077 & -0.0103 @xmath73 0.0096 & 0.050 & 7.03 @xmath73 0.08 & -0.197 @xmath73 0.141 & -0.0246 @xmath73 0.0176 & 0.091 + ugc 12709 & 8.47 @xmath73 0.08 & -0.706 @xmath73 0.218 & -0.0434 @xmath73 0.0134 & 0.082 & 7.30 @xmath73 0.15 & -1.089 @xmath73 0.418 & -0.0669 @xmath73 0.0257 & 0.157 + pgc 29167 & 8.34 @xmath73 0.10 & 0.152 @xmath73 0.169 & 0.0507 @xmath73 0.0565 & 0.040 & 7.25 @xmath73 0.01 & 0.005 @xmath73 0.020 & 0.0017 @xmath73 0.0065 & 0.005 + pgc 41318 & 8.45 @xmath73 0.04 & -0.168 @xmath73 0.143 & -0.0281 @xmath73 0.0239 & 0.033 & 7.28 @xmath73 0.17 & -0.563 @xmath73 0.568 & -0.0941 @xmath73 0.0945 & 0.130 + pgc 44772 & 8.59 @xmath73 0.03 & -0.268 @xmath73 0.045 & -0.0216 @xmath73 0.0036 & 0.043 & 7.79 @xmath73 0.07 & -0.705 @xmath73 0.115 & -0.0569 @xmath73 0.0093 & 0.109 + pgc 45195 & 8.14 @xmath73 0.04 & -0.152 @xmath73 0.059 & -0.0122 @xmath73 0.0047 & 0.035 & 6.81 @xmath73 0.08 & -0.277 @xmath73 0.104 & -0.0223 @xmath73 0.0084 & 0.061 + pgc 46182 & 8.59 @xmath73 0.03 & -0.126 @xmath73 0.078 & -0.0089 @xmath73 0.0055 & 0.035 & 7.92 @xmath73 0.11 & -0.462 @xmath73 0.313 & -0.0327 @xmath73 0.0221 & 0.141 + ll ngc 0012 & @xcite + ngc 0055 & @xcite , @xcite , @xcite + ngc 0099 & @xcite , sdss + ngc 0224 ( m 31 ) & @xcite , @xcite , @xcite , @xcite , + & @xcite , @xcite , @xcite + ngc 0234 & @xcite , sdss + ngc 0253 & @xcite + ngc 0300 & @xcite , @xcite , @xcite , @xcite + ngc 0428 & sdss + ngc 0450 & @xcite , sdss + ngc 0493 & sdss + ngc 0575 & @xcite + ngc 0598 ( m 33 ) & @xcite , @xcite , @xcite , @xcite , @xcite , + & @xcite , @xcite , @xcite , @xcite , + & @xcite , @xcite , @xcite + ngc 0628 ( m 74 ) & @xcite , @xcite , @xcite , @xcite , + & @xcite , @xcite + ngc 0753 & @xcite + ngc 0783 & @xcite + ngc 0925 & @xcite + ngc 1055 & sdss + ngc 1058 & @xcite , @xcite + ngc 1068 ( m 77 ) & @xcite , @xcite , @xcite , sdss + ngc 1090 & sdss + ngc 1097 & @xcite , @xcite + ngc 1232 & @xcite , @xcite + ngc 1313 & @xcite , @xcite , @xcite + ngc 1365 & @xcite , @xcite , @xcite , @xcite + ngc 1512 & @xcite + ngc 1598 & @xcite + ngc 1637 & @xcite , @xcite + ngc 1642 & @xcite + ngc 1672 & @xcite + ngc 2336 & @xcite + ngc 2403 & @xcite , @xcite , @xcite , @xcite , @xcite , + & @xcite , @xcite , @xcite , sdss + ngc 2441 & @xcite + ngc 2442 & @xcite + ngc 2541 & @xcite , sdss + ngc 2805 & @xcite , sdss + ngc 2835 & @xcite + ngc 2841 & @xcite , @xcite + ngc 2903 & @xcite , @xcite , @xcite , sdss + ngc 2997 & @xcite , @xcite , @xcite + ngc 3020 & sdss + ngc 3023 & sdss + ngc 3031 ( m 81 ) & @xcite , @xcite , @xcite , @xcite , + & @xcite , sdss + ngc 3184 & @xcite , @xcite , @xcite , sdss + ngc 3198 & @xcite , sdss + ngc 3227 & @xcite , @xcite , @xcite + ngc 3239 & sdss + ngc 3310 & @xcite , @xcite + ngc 3319 & @xcite , sdss + ngc 3344 & @xcite , @xcite , @xcite , sdss + ngc 3351 ( m 95 ) & @xcite , @xcite , @xcite , @xcite , sdss + ngc 3359 & @xcite , sdss + ngc 3486 & sdss + ngc 3521 & @xcite , @xcite + ngc 3621 & @xcite , @xcite + ngc 3631 & sdss + ngc 3718 & @xcite + ngc 3820 & @xcite , sdss + ngc 3893 & @xcite + ngc 3938 & sdss + ngc 4030 & sdss + ngc 4088 & sdss + ngc 4109 & @xcite + ngc 4254 ( m 99 ) & @xcite , @xcite , @xcite , sdss + ngc 4258 ( m 106 ) & @xcite , @xcite , sdss + ngc 4303 ( m 61 ) & @xcite , @xcite , sdss + ngc 4321 ( m 100 ) & @xcite , @xcite , sdss + ngc 4395 & @xcite , @xcite , @xcite , sdss + ngc 4490 & sdss + ngc 4501 & @xcite + ngc 4535 & sdss + ngc 4559 & @xcite , sdss + ngc 4625 & @xcite , @xcite , sdss + ngc 4631 & @xcite , sdss + ngc 4651 & @xcite , sdss + ngc 4654 & @xcite , sdss + ngc 4656 & sdss + ngc 4713 & @xcite , sdss + ngc 4725 & @xcite + ngc 4736 ( m 94 ) & @xcite , @xcite , @xcite + ngc 4861 & @xcite , @xcite , @xcite , @xcite , + & @xcite , @xcite , sdss + ngc 5033 & @xcite , sdss + ngc 5055 ( m 63 ) & @xcite + ngc 5068 & @xcite + ngc 5194 ( m 51 ) & @xcite , @xcite , @xcite , @xcite , sdss , + & @xcite + ngc 5236 ( m 83 ) & @xcite , @xcite , @xcite , @xcite , + & @xcite , @xcite + ngc 5248 & @xcite , sdss + ngc 5457 ( m 101 ) & @xcite , @xcite , @xcite , @xcite , @xcite , + & @xcite , @xcite , @xcite , @xcite , + & @xcite , sdss , @xcite , @xcite , @xcite , + & @xcite , @xcite , @xcite + ngc 5474 & @xcite , sdss + ngc 5668 & @xcite , sdss + ngc 6384 & @xcite , @xcite + ngc 6691 & @xcite + ngc 6744 & @xcite + ngc 6946 & @xcite , @xcite , @xcite + ngc 7331 & @xcite , @xcite + ngc 7495 & @xcite + ngc 7518 & @xcite + ngc 7529 & @xcite + ngc 7570 & @xcite + ngc 7591 & @xcite + ngc 7678 & @xcite + ngc 7793 & @xcite , @xcite , @xcite , @xcite + ic 0010 & @xcite , @xcite , @xcite + ic 0193 & @xcite + ic 0208 & @xcite + ic 0342 & @xcite + ic 1132 & @xcite + ic 2204 & @xcite , sdss + ic 5201 & @xcite + ic 5309 & @xcite + ugc 00223 & @xcite , sdss + ugc 01087 & @xcite + ugc 01862 & @xcite + ugc 02023 & @xcite + ugc 02216 & sdss + ugc 02345 & sdss + ugc 03701 & @xcite + ugc 04107 & @xcite , sdss + ugc 04305 & @xcite + ugc 05100 & @xcite + ugc 06410 & @xcite , sdss + ugc 08091 ( gr 8) & @xcite , @xcite , sdss + ugc 09562 & @xcite , sdss + ugc 09837 & @xcite , sdss + ugc 10445 & @xcite , sdss + ugc 12709 & sdss + pgc 029167 & @xcite + pgc 041318 & @xcite + pgc 044772 & @xcite + pgc 045195 & sdss + pgc 046182 & sdss , @xcite +

the purpose of the present work is to improve the recent saclay - lyon ( sl ) study @xcite on the strangeness electromagnetic production from the proton . this latter investigation was based upon an effective hadronic lagrangian in the lowest ( tree ) approximation , often called the isobar approximation . in a number of aspects one might safely say that sl is an improved version of its predecessors dealing with the same strangeness production processes . in particular , it has incorporated the _ s-_channel nucleonic resonances with spin 3/2 and 5/2 , expected to be important should the model keep adequate as energy increases . in ref . @xcite such resonances were also considered . however , there the components of the amplitude growing undesirably with increasing channel energy were taken away by hand . as we will see later , these contributions arise from the non - resonant terms associated to each considered resonance with spin @xmath5 1/2 . in the sl study this was avoided by modifying the vertices and propagators in a manner adopted for spin 3/2 resonances in refs . @xcite : a straightforward extension to higher spins , while preserving the electromagnetic gauge invariance . this modification , however , has introduced an unwanted behavior for spin @xmath6 hyperonic resonances exchanged in the _ u-_channel : the corresponding propagators become singular in the physical region . thus in the sl approach only spin 1/2 hyperons have been considered in the _ u-_channel exchange . the phenomenological success of the sl model might imply that , within the present state of the data , the main contributions from baryonic higher spin resonances come mainly from the _ s_-channel resonances ( we will come back to this point in section iv ) . in the study of pion photoproduction , the rensselaer polytechnic institute ( rpi ) group @xcite has shown that of several different forms of the spin 3/2 propagator in the literature only one of them has a correct inverse . also the authors pointed out that there are extra degrees of freedom associated with the interaction vertices involving a spin 3/2 particle . by exploiting these facts , they successfully fitted the existing photo - pion data by the amplitudes generated from effective hadronic lagrangians , and made predictions for some observables as well as the @xmath7 ratio for the @xmath8 vertex . a similar strategy has been applied also by the rpi group @xcite to the photo- and electro - production of the @xmath9 meson . it seems quite natural then , as an extension of the saclay - lyon approach @xcite , as well as the works of the rpi group @xcite , to exploit this treatment for spin 3/2 particles in the study of the photo- and electro - production of the strangeness off the nucleon . yet , one needs to incorporate properly the _ u-_channel exchanges in the phenomenological approaches . the reasons for such an effort are mainly two - fold : _ ( i ) _ a consistent treatment of the higher spin baryonic resonances in both _ s- _ _ and _ _ u-_channels , _ ( ii ) _ very likely , more sophisticated formalisms will be needed to interpret the forthcoming data from new facilities , _ e.g. _ , the thomas jefferson national accelerator facility ( jlab ) , the electron stretcher accelerator ( elsa ) , the european synchrotron radiation facility ( esrf ) , and the 8 gev synchrotron facility ( spring-8 ) under construction in japan . in this paper , we work out the general expressions valid for the processes with a kaon @xmath10 ( @xmath11 @xmath12 ) and an hyperon @xmath13 ( @xmath14 ) in the final state . a selected set of @xmath15 channel observables for the following processes are also reported : @xmath1 ( @xmath2 2.5 gev ) , @xmath3 , and @xmath4 . similar investigations with the @xmath16 hyperons in the final state , _ i.e. _ @xmath17 and @xmath18 channels , are in progress and the results will be reported elsewhere . in section ii , the approach by the rpi group is extended to the photo- and electroproduction of strangeness through _ s-_channel nucleonic resonances of spin 3/2 . the off - shell parameters are introduced in the interaction lagrangians , and the dependence on these parameters of the non - pole part of the invariant amplitudes is clarified . the approaches used previously where the off - shell effects were ignored are placed in the present context . section iii is devoted to the treatment of spin 3/2 resonances in the _ the direct calculation proceeds along the same line as for the _ s-_channel resonance exchange . the substitution rule which emerges from the direct calculation is worked out , leading to simple rules to obtain the _ u-_channel invariant amplitudes from the _ s-_channel ones . in section iv , we give our results and we discuss the dependence of the relevant observables on the off - shell parameters . the summary and conclusions are presented in the last section . in this section we extend the approach by benmerrouche _ et al . _ @xcite , devoted to the @xmath19 and @xmath9 photoproduction , to obtain the amplitudes for the reactions @xmath20 ( @xmath21 ) for both real ( @xmath22 ) and virtual ( @xmath23 ) photons , through an _ s-_channel nucleonic resonance of spin 3/2 and positive parity . once we obtain the amplitude , it is easy to establish its relation to the corresponding one obtained by renard & renard @xcite as well as to the one in sl @xcite . also one finds that the amplitudes due to nucleonic _ s-_channel @xmath24 resonances may be trivially obtained by simple substitutions . although some parts of this section should appear to be repetitive to those who are familiar with ref.@xcite , we give a comprehensive presentation of the matter for completeness , and present the explicit expressions of the invariant amplitudes for the photo- and electro - production . we first define the nucleon field as @xmath25 and a spin 3/2 ( isospin 1/2 ) vector spin field ( resonance ) as @xmath26 . then the free lagrangian for the spin 3/2 field reads @xmath27 where @xmath28 is the usual dirac conjugate of @xmath29 , and @xmath30,\end{aligned}\ ] ] with @xmath31 the mass of the resonance , and @xmath32 being a free parameter which preserves the invariance of the physical quantities constructed from the field under the point transformation @xmath33 @xmath34 , but otherwise arbitrary . the free spin 3/2 field satisfies the equation of motion and two constraints @xmath35 the above constraints ensure that @xmath36 has spin 3/2 with upper ( positive energy ) and lower ( negative energy ) components . the propagator associated with the @xmath36 field is obtained from the equation @xmath37 we may set @xmath38 to find the simplest form for the propagator @xmath39 , \label{eq : prop32}\ ] ] where @xmath40 is the four momentum of the resonance . it is important to note @xcite that this propagator contains the spin 1/2 contribution , which is a consequence of the fact that the above @xmath41 has the correct inverse . now we introduce the interactions for @xmath42 through the _ s_-channel spin 3/2 resonance discussed above . again following ref . @xcite with some modifications appropriate for the processes under consideration , the most general interaction lagrangian which preserves the symmetry under the point interaction introduced in the previous subsection reads @xmath43 , \label{eq : intlag1 } \\ % { \cal l}_{\gamma p r}^{(1 ) } & = & \frac{i e g_1}{2 m_p } \left[\overline r^{\nu } \theta_{\mu\lambda}(y ) \gamma_{\nu}\gamma^5 n f^{\nu\lambda } + \overline n \gamma^5\gamma_{\nu } \theta_{\lambda\mu}(y ) r^{\mu } f^{\nu\lambda } \right ] , \label{eq : intlag2 } \\ % { \cal l}_{\gamma p r}^{(2 ) } & = & \frac{- e g_2}{4 m_p^2 } \left[\overline r^{\mu } \theta_{\mu\nu}(x ) \gamma^5 ( \partial_{\lambda } n ) f^{\nu\lambda } - ( \partial_{\lambda } \overline n ) \gamma^5 \theta_{\nu\mu}(x ) r^{\mu } f^{\nu\lambda } \right ] . % \label{eq : intlag3}\end{aligned}\ ] ] in expression ( [ eq : intlag1 ] ) , @xmath44 specifies the lagrangian for the strong kaon - hyperon - resonance ( @xmath45 ) vertex in which @xmath10 denotes the iso - doublet @xmath46 @xmath47 and @xmath48 are for the @xmath49 and _ derivative _ electromagnetic coupling terms , respectively . there @xmath50 is the standard electromagnetic field tensor , and @xmath51 is defined as ( for our choice of @xmath38 in the previous subsection ) @xmath52 it is important to stress that in the above lagrangian , @xmath53 are arbitrary parameters which conserve the symmetry of the free lagrangian under the point transformation [ eq . ( [ eq : point ] ) ] , and are often called the _ off - shell parameters_. as will become clear later , we shall exploit this extra freedom to make the kaon electromagnetic production amplitudes well tamed . in what follows we shall rather use @xmath54 , @xmath55 , @xmath56 . the contribution to the @xmath57-matrix from the _ s_-channel resonance pole through the @xmath49 coupling reads @xmath58 where @xmath59 is the time - ordering operator . using the above lagrangians , the matrix element for the @xmath49 term is obtained as @xmath60 u_p({\hbox{\boldmath{$p$}\unboldmath}}_{p } ) , \label{eq : gamma5}\end{aligned}\ ] ] where we have introduced the coupling constant @xmath61 @xmath62 is the polarization vector of the photon , @xmath63 is the total momentum ( @xmath64 ) , and @xmath65 is the spin-3/2 propagator introduced in the last subsection , eq.([eq : prop32 ] ) . using expression ( [ eq : teta ] ) for @xmath66 to calculate the terms on both sides of the propagator , we find @xmath67 p^{\mu\nu}(q ) \nonumber \\ & & \times \left\ { [ \epsilon_{\nu}{p } \hskip -0.16 cm / _ { \gamma } -(p_{\gamma})_{\nu}{\epsilon } \hskip -0.16 cm / ] - \widetilde y \gamma_{\nu } [ { \epsilon } \hskip -0.16 cm / { p } \hskip -0.16 cm / _ { \gamma } -\epsilon { \cdot } p_{\gamma } ] \right\ } \gamma^5 u_p({\hbox{\boldmath{$p$}\unboldmath}}_{p } ) . \label{eq : dirac}\end{aligned}\ ] ] a similar calculation leads to the _ derivative _ coupling contribution corresponding to @xmath48 @xmath68 p^{\mu\nu}(q ) \nonumber \\ & & \times \left\{[\epsilon { \cdot } p_p(p_{\gamma})_{\nu } - p_{\gamma } { \cdot } p_p \,\epsilon_{\nu } ] + \widetilde x \gamma_{\nu } [ p_{\gamma } { \cdot } p_p{\epsilon } \hskip -0.16 cm / - \epsilon { \cdot } p_p{p } \hskip -0.16 cm / _ { \gamma } ] \right\ } \gamma^5 u_p({\hbox{\boldmath{$p$}\unboldmath}}_{p } ) , \label{eq : pauli}\end{aligned}\ ] ] with @xmath69 adding the above two contributions given in eqs . ( [ eq : dirac ] ) and ( [ eq : pauli ] ) , we can write the scattering amplitude @xmath70 corresponding to the _ s-_channel exchange of an @xmath71 resonance as @xmath72 where the @xmath73 vertex reads @xmath74 , \label{eq : vertk}\ ] ] and the @xmath75 vertex is @xmath76 i\gamma^5 . \label{eq : vert32 } \end{aligned}\ ] ] note that , for the general case of electroproduction , the above vertex must be multiplied by @xmath77 , the second dirac form factor of the proton . in the case of photoproduction , this factor reduces to unity , and in addition we have @xmath78 . the lorentz invariant matrix element for electroproduction is written as @xmath79 where @xmath80 and @xmath81 are the spinors of the hyperon and the proton , respectively , @xmath82 s are lorentz invariant scalar functions of the mandelstam variables , and @xmath83 s are the six usual gauge invariant matrices for the electroproduction @xmath84 due to the second term , @xmath85 , the choice of @xmath86 is different from that used in refs . @xcite and @xcite . this results in a few modifications in the expressions of the cgln amplitudes for the electroproduction as given in appendix [ app : cgln ] . the advantage of this choice is its symmetric property under the exchange @xmath87 , thus leading to more transparent relations between the _ s- _ and _ u-_channels amplitudes , as shown in the next section . in the case of photoproduction , ( @xmath88 , @xmath78 ) , only the first four invariant amplitudes in eq . ( [ eq : mj ] ) are needed . using the above expressions for the propagator and vertices , application of the dirac algebra leads to the invariant amplitudes @xmath82 , which are expressed as sums of resonant or pole ( @xmath89 ) and non - pole ( @xmath90 ) contributions . in the case of the photoproduction we find @xmath91 \quad , \quad ( j=1,\ldots,4 ) , \label{eq : ajphoto}\ ] ] the expressions of the @xmath92 coefficients are given in appendix [ app : invamp ] , eqs . ( [ eq : p1j ] ) to ( [ eq : p2j ] ) . the electroproduction amplitudes can be written in a similar form @xmath93 \quad , \quad ( j=1,\ldots,6 ) . \label{eq : ajelec}\ ] ] for @xmath94 , the @xmath95 coefficients are expressed in terms of the above defined photoproduction coefficients @xmath92 as @xmath96 the extra terms @xmath97 are given in appendix [ app : invamp ] , eqs . ( [ eq : r1j ] ) and ( [ eq : r2j ] ) . note that this decomposition is not necessary for @xmath98 . the corresponding coefficients @xmath99 are given in eqs . ( [ eq : e1 - 56 ] ) and ( [ eq : e2 - 56 ] ) of appendix [ app : invamp ] . note that in the calculation of the observables ( sec . iv ) , the following replacement is made in the denominator of the pole contribution in eqs . ( [ eq : ajphoto ] ) and ( [ eq : ajelec ] ) @xmath100 where @xmath101 is the width of the resonance . it should be important to emphasize here that the _ pole _ contributions ( see appendix [ app : invamp ] ) are completely _ independent _ of @xmath102 , hence with no off - shell dependence . so far we have discussed the case in which the parity of the _ s_-channel resonance is positive . for a resonance with negative parity , we have only to make the following replacements : @xmath103 in eq . ( [ eq : vertk ] ) , and @xmath104 in eq . ( [ eq : vert32 ] ) . in the corresponding @xmath70 amplitude [ eq . ( [ eq : mfi ] ) ] , @xmath49 is now acting onto the left of the first vertex . using the anti - commutation property of @xmath49 with @xmath105 , it is easy to move the @xmath49 matrix in the same position as in the positive parity case , namely onto the right of the second vertex . by inspection , we immediately obtain the parity rule for the invariant ( pole and non - pole ) amplitudes @xmath106 the expressions used in ref . @xcite for the propagator is the same as eq . ( [ eq : prop32 ] ) , but in the interaction lagrangian eqs . ( [ eq : intlag1])-([eq : intlag3 ] ) @xmath107 was set equal to @xmath108 . in other words , the authors put @xmath109 in ( [ eq : teta ] ) ( or @xmath110 ) , thus no off - shell effect associated with the spin 3/2 particles was considered . it is therefore clear from eqs . ( [ eq : vertk ] ) and ( [ eq : vert32 ] ) that the corresponding amplitude simplifies considerably . however , some of the _ non - pole _ contributions @xmath111 grow linearly in the _ s-_variable ( see appendix [ app : invamp ] ) , causing an undesirable increase , for example , in the production cross section . for this reason _ all _ the resulting _ non - pole _ contributions were artificially thrown away in ref . @xcite . to avoid the difficulties encountered in the renard & renard model @xcite , adelseck _ et al . _ @xcite have suggested and applied the following prescriptions ( used also in ref . the propagator is written from eq . ( [ eq : prop32 ] ) , with the mass of the resonance @xmath31 replaced by the total invariant energy @xmath112 , except in the denominator where the width of the resonance is introduced @xmath113 . \label{eq : prop32adel}\ ] ] this modification provides an extra damping of the amplitude with increasing channel energy . so together with the corresponding modification in the vertices discussed below , an unwanted growth in the production cross section due to the non - pole contribution could be reduced in the absence of the off - shell freedom ( in terms of @xmath114 ) . in the photoproduction case , the @xmath45 vertex is @xmath115 and the @xmath116 vertex factor for a positive parity resonance is written as @xmath117 \gamma^5 . \label{eq : vn32adel}\ ] ] as stated by adelseck _ , these prescriptions were used to ensure gauge invariance of the scattering amplitude . in fact , expressions ( [ eq : vkln32adel ] ) and ( [ eq : vn32adel ] ) may be reached from eqs . ( [ eq : vertk ] ) and ( [ eq : vert32 ] ) as demonstrated in appendix [ app : nooff ] , where the coupling constants @xmath118 , @xmath119 , and @xmath120 are defined in terms of @xmath121 , @xmath122 , and @xmath123 , respectively . particularly , the photon coupling vertex in this choice contains damping factors in the _ s-_variable . however , regarding the spin 3/2 propagator ( [ eq : prop32adel ] ) , when the same form is used for a _ u-_channel resonance exchange , namely the _ s-_variable replaced by the _ u-_variable , the latter may vanish at certain kinematical situations , leading to an unphysical behavior . note also that as pointed out in @xcite , such propagators do not have inverses , and corresponding wave equations for the the spin-3/2 field can not be defined . thus this approach is not appropriate for a consistent simultaneous description of the _ s- _ and _ u-_channels . in this section we show some basic details on how the lowest order _ u-_channel exchange of a @xmath125 resonance contributes to the amplitude for @xmath126 production on the proton . the exchange of a @xmath125 resonance in the _ u-_channel is treated along the same lines as in section ii for the _ s-_channel resonances exchange . the part of the @xmath57-matrix corresponding to the * @xmath49 * photon coupling is @xmath127 with @xmath128 . the resulting matrix element takes the form @xmath129 \theta_{\lambda\nu}(y ) \nonumber\\ & & \times p^{\nu\mu}(-q ' ) \theta_{\mu\chi}(z ) i p_k^{\chi } u_p({\hbox{\boldmath{$p$}\unboldmath}}_{p}).\end{aligned}\ ] ] the momentum transfer is @xmath130 , with @xmath131 . note that with a correct kinematical consideration it is easy to see that the propagator depends on @xmath132 ( not @xmath133 ! ) . using eq . ( [ eq : teta ] ) just as before , one finds @xmath134 - \widetilde y [ { p } \hskip -0.16 cm / _ { \gamma}{\epsilon } \hskip -0.16 cm / -\epsilon { \cdot } p_{\gamma } ] \gamma_{\nu } \right \ } \nonumber \\ & & \times p^{\nu\mu}(-q ' ) [ ( p_k)_{\mu } - \widetilde z \gamma_{\mu } { p } \hskip -0.16 cm / _ k ] u_p({\hbox{\boldmath{$p$}\unboldmath}}_{p } ) . \label{eq : u - dirac}\end{aligned}\ ] ] the _ derivative _ coupling term is calculated along the same lines , leading to @xmath135 + \widetilde x [ p_{\gamma } { \cdot } p_{y}{\epsilon } \hskip -0.16 cm / - \epsilon { \cdot } p_{y } { p } \hskip -0.16 cm / _ { \gamma } ] \gamma_{\nu } \right\}\nonumber \\ & & \times p^{\nu\mu}(-q ' ) [ ( p_k)_{\mu } - \widetilde z \gamma_{\mu}{p } \hskip -0.16 cm / _ k ] u_p({\hbox{\boldmath{$p$}\unboldmath}}_{p } ) . \label{eq : u - pauli}\end{aligned}\ ] ] in the above expressions , @xmath136 and @xmath137 are the two @xmath138 coupling constants , which are similar to @xmath121 and @xmath122 as in eqs . ( [ eq : intlag2 ] ) and ( [ eq : intlag3 ] ) . note the similarity of the last two expressions with the corresponding ones for an _ s-_channel resonance exchange , eqs . ( [ eq : dirac ] ) and ( [ eq : pauli ] ) . adding the two above contributions , the scattering matrix in the _ u-_channel exchange reads @xmath139 the two vertices are @xmath140 , \label{eq : u - vertk}\ ] ] @xmath141 . \label{eq : u - vert32 } \end{aligned}\ ] ] the propagator reads @xmath142 . \label{eq : u - prop32}\ ] ] using the above expressions for the vertices and propagator , the decomposition of @xmath143 in terms of the gauge invariant matrices defined in eq . ( [ eq : mj ] ) can be done along the same lines as in section ii.d . however , comparing the _ s- _ and _ u-_channels vertices and propagators , it is easy to get out the rules regarding how to obtain the expressions for the _ u-_channel exchange from those for the _ s-_channel , namely : _ 1 ) _ exchange @xmath87 ( including @xmath144 ) , _ 2 ) _ express the products of @xmath145 matrices in a reversed order , _ 3 ) _ change @xmath146 , @xmath147 , @xmath148 , and _ 4 ) _ exchange the two vertices and put the appropriate coupling constants . in fact , these rules result from a substitution rule which is simpler to use , since it allows us to formally derive the invariant amplitudes for the _ u-_channel exchange directly from the corresponding _ s-_channel exchange amplitudes . the derivation of the substitution rule and its application to obtain the invariant amplitudes are given in the next two subsections . we now calculate the _ u-_channel results by substitution rule applied to the _ s-_channel expressions . namely , we introduce in eq . ( [ eq : mfi ] ) the appropriate coupling constants , together with the following replacements : @xmath149 , @xmath150 , @xmath87 , @xmath151 , @xmath152 , with @xmath153 , @xmath154 being the negative energy spinors . the resulting scattering matrix is @xmath155 and the expressions of the vertices and propagator are [ cf . ( [ eq : vertk ] ) , ( [ eq : vert32 ] ) , and ( [ eq : prop32 ] ) ] : @xmath156 , \label{eq : vertk - u}\ ] ] @xmath157 \gamma_5 , \label{eq : vert32-u } \end{aligned}\ ] ] @xmath158 , \label{eq : prop32-u}\ ] ] with @xmath159 , as before . using the relation between the @xmath160 and @xmath161 spinors : @xmath162 , with @xmath163 the charge conjugation operator , and @xmath164 the dirac adjoint of @xmath161 , eq . ( [ eq : mfi - u ] ) is written as : @xmath165 by appropriately inserting @xmath166 , the above equation can be transformed into @xmath167^c \ [ p_{\mu \nu}(q')^t]^c \ [ { \cal { v}}^{\mu}(k p r)^t]^c \ u_p({\hbox{\boldmath{$p$}\unboldmath}}_{p } ) , \label{eq : mfi - u3}\ ] ] where we have defined the @xmath168-transform of @xmath169 as @xmath170^c = c^{-1 } x^t c.\ ] ] now , we exploit the properties of the charge conjugation matrix @xmath168 to calculate the @xmath168-transforms of the vertices and propagator . for example , the @xmath168-transform of the @xmath171 vertex eq.([eq : vertk - u ] ) is @xmath172^c = -\frac{g_{k p r}}{m_k } c^{-1}[\ p_k^{\mu } - { \widetilde z } { p } \hskip -0.16 cm / _ k \gamma^{\mu } ] ^t c. \label{eq : c - vertk - u}\ ] ] from the properties of @xmath168 , we obtain @xmath173 and eq . ( [ eq : c - vertk - u ] ) becomes @xmath172^c = -\frac{g_{k p r}}{m_k } [ \ p_k^{\mu } - { \widetilde z } \gamma^{\mu } { p } \hskip -0.16 cm / _ k ] , \label{eq : c - vertk - u2}\ ] ] a similar calculation leads to the @xmath174 vertex @xmath175^c & = & i \gamma_5 \left [ - \frac{e g'_1}{2m_{y}}{\left ( \epsilon^{\nu}{p } \hskip -0.16 cm / _ { \gamma } - p_{\gamma}^{\nu}{\epsilon } \hskip -0.16 cm / -{\widetilde y } ( { p } \hskip -0.16 cm / _ { \gamma } { \epsilon } \hskip -0.16 cm / - \epsilon { \cdot } p_{\gamma})\gamma^{\nu } \right ) } \right . \cr & -&\left . \frac{e g'_2}{4m^2_{y } } { \left ( \epsilon { \cdot } p_{y } p_{\gamma}^{\nu } - p_\gamma { \cdot } p_{y } \epsilon^{\nu } + { \widetilde x } ( p_\gamma { \cdot } p_{y } { \epsilon } \hskip -0.16 cm / - \epsilon { \cdot } p_{y } { p } \hskip -0.16 cm / _ { \gamma})\gamma^{\nu } \right ) } \right ] , \label{eq : c - vert32-u } \end{aligned}\ ] ] and the propagator takes the form @xmath176^c = \frac{-{q'}\hskip -0.29 cm /+m_r}{3(u - m_r^2 ) } \left[3g_{\nu\mu}-\gamma_{\nu}\gamma_{\mu}- \frac{2}{m_r^2}q'_{\nu}q'_{\mu } + \frac{1}{m_r}(\gamma_{\nu}q'_{\mu}-\gamma_{\mu}q'_{\nu})\right ] . \label{eq : c - prop32-u}\ ] ] comparing with eq . ( [ eq : prop32-u ] ) leads to @xmath176^c = p_{\nu\mu}(-q ' ) . \label{eq : cc - prop32-u}\ ] ] combining eqs . ( [ eq : mfi - u3 ] ) , ( [ eq : c - vertk - u2 ] ) , ( [ eq : c - vert32-u ] ) and ( [ eq : cc - prop32-u ] ) leads to the same result as the direct calculation eqs . ( [ eq : u - mfi])-([eq : u - prop32 ] ) . in this subsection , we apply the substitution rule to obtain the invariant amplitudes ( hereafter denoted as @xmath177 ) for the _ u-_channel exchange directly from the corresponding _ s-_channel exchange amplitudes @xmath82 . let us write eq . ( [ eq : mfigen ] ) with specifying the relevant variables @xmath178 where @xmath83 are the six gauge invariant amplitudes eq.([eq : mj ] ) , and @xmath82 have been made explicit in section ii.d . we now apply the substitution rule ( see the previous subsection ) to eq . ( [ eq : mfi - s ] ) in order to obtain the scattering matrix in the _ u-_channel @xmath179 from eq . ( [ eq : mj ] ) , it is clear that @xmath180 we proceed as in the last subsection , and transform eq . ( [ eq : mfi - u4 ] ) into @xmath181 then , we calculate the @xmath168-transforms of the @xmath182 matrices , which can easily be expressed in terms of the original @xmath183 matrices as @xmath184 substituting these relations into eq . ( [ eq : mfi - u5 ] ) , the scattering matrix in the _ u-_channel can be written as @xmath185 where the invariant amplitudes @xmath177 are related to the @xmath82 amplitudes in the _ s-_channel as follows @xmath186 it is then quite easy to obtain these invariant amplitudes in a form similar to eq . ( [ eq : ajelec ] ) , which we will not present in this article . in this section we _ illustrate _ the sensitivity of different @xmath187 channels observables to the off - shell effects . we need , hence , a reliable dynamical model , with respect to the existing data , as starting point . in the following , we present first how a rather simple model was obtained and then , within the dynamical ingredients required by the available data , we report on the importance of the off - shell effects according to the observables and/or the phase space regions investigated . to build a simple model with a reasonably realistic dynamical content , we take advantage of the sl model @xcite which has emerged from a comprehensive phenomenological study . the underlying dynamics in the sl model is , besides extended born terms , resonances exchanges ( table [ tab : abc ] ) in the following channels : * _ s - channel : _ @xmath188(1/2 ) , @xmath189(3/2 ) , @xmath190(5/2 ) ; where the spin of each nucleonic resonance is given in parenthesis . * _ u - channel : _ @xmath191 , @xmath192 , @xmath193 , @xmath194 ; all spin 1/2 hyperonic resonances . * _ t - channel : _ @xmath195 , @xmath196 ; both of them have also been included in the present work and we will not discuss them any further . in the _ s-_channel , the most relevant resonance , in the frame of the present work , is the spin 3/2 resonance @xmath189 . the @xmath188 resonance , @xmath197(1440 ) , was found to have a coupling compatible with zero ( see table [ tab : abc ] and ref . moreover , a recent model - independent nodal structure analysis @xcite concludes that the present data do not require contributions from the @xmath197 resonances . concerning another nucleonic resonance in the sl model , the @xmath198 , it was shown @xcite that its contribution to the considered underlying dynamics _ is not crucial _ ( see table xi in ref . @xcite ) . for these reasons , _ we removed the @xmath188 and @xmath190 resonances in searching for a simple model to study the role of off - shell effects_. the parameters of this model , hereafter called model a , have been obtained by re - fitting the data . note that the formalism used in this re - fitting is still within the context of adelseck _ et al . _ s treatment for the spin 3/2 resonance @xmath189 . model a is the basis of our numerical results reported in the next subsection . the first step was thus , using model a , to fit the same data base as used to obtain the sl model ( all available 312 data points for photo- , electro - production , as well as for the @xmath199 radiative capture process ) . the coupling constants and the reduced @xmath200 are given in table [ tab : abc ] . although the resultant @xmath200 for model a ( 1.84 ) is slightly larger than that for the sl model ( 1.73 ) , it is still acceptable . anticipating the presentation of the observables in the next subsection , the fit of the data with model a appears at a comparable level of quality as with the sl model , see the dotted and dash - dotted curves in figs . [ fig : ds]-[fig : pol ] , and fig . [ fig : ul ] . hence , these results justify the use of model a as a starting point to investigate the sensitivity of the observables to the off - shell effects . given that model a contains only one spin 3/2 baryonic resonance , we have also investigated possible contributions from other known spin 3/2 nucleonic resonances , namely ] . ] , [ @xmath201[@xmath202 ( @xmath203 ) or [ @xmath204[@xmath202 ( @xmath205 ) . we performed minimizations for all possible configurations including one to three of the spin 3/2 resonances @xmath203 , @xmath205 , and @xmath189 . in these configurations , whenever at least one of the two resonances @xmath203 , and @xmath205 was retained , the corresponding @xmath206s were found significantly larger than that for model a , implying that the existing data base does not require contributions from these resonances . through the numerical investigations mentioned above , we have re - confirmed that model a is indeed a _ reasonable starting model _ for the present study . then we adopted the correct propagator [ eq . ( [ eq : prop32 ] ) ] and introduced the off - shell treatments to the @xmath189 resonance , and fitted again the data to obtained model b ( table [ tab : abc ] ) . finally , for the sake of completeness we included the _ u-_channel spin 3/2 hyperonic resonance [ @xmath207[@xmath208 ( @xmath209 ) with the off - shell effect , and once again fitted the data ( model c in table [ tab : abc ] ) . the choice of this resonance , as in the case of nucleonic resonances mentioned above , comes from the fact that the inclusion of any other spin 3/2 hyperonic resonance , [ @xmath210[@xmath211 ( @xmath212 ) or [ @xmath213[@xmath211 ( @xmath214 ) , deteriorates the reduced @xmath200 significantly . here we would like to point out that by adding any spin 3/2 baryonic resonance we introduce five additional free parameters , namely two coupling constants ( @xmath215 and @xmath216 ) and three off - shell parameters . the fact that the @xmath200 associated with model c comes out larger than that for the model b , indicates that the dynamical content of the phenomenological approach discussed here is reliable enough , since _ additional free parameters _ due to apparently unrelevant resonances _ do not improve _ the @xmath200 ( reduced or per point ) . model d in table [ tab : abc ] , with two of the free parameters fixed , will be discussed in subsec . c. in this subsection , we compare with the data the results of the four dynamical models ( sl , a , b , and c ) summarized in table [ tab : abc ] . here we will adhere closely to the observables reported for the sl model @xcite , where a comprehensive discussion on other available phenomenological results @xcite is also presented . in fig . [ fig : ds ] , angular distributions and excitation functions for the differential cross - section are shown . all the models reproduce the data almost equally well . however , the excitation functions at @xmath218 = 90@xmath219 [ fig . [ fig : ds](b ) ] and 150@xmath219 [ fig . [ fig : ds](c ) ] split the four models into two families above @xmath220 1.5 gev : in the backward hemisphere , both the sl and a models predict significantly larger cross sections than the two others ( b and c ) which embody the off - shell effects . for the angular distributions [ fig . [ fig : ds](d ) , [ fig : ds](e ) , and [ fig : ds](f ) ] , the four models give similar results at @xmath221 = 1.0 and 1.45 gev , while at the highest energy [ @xmath221 = 2.1 gev ; fig . [ fig : ds](f ) ] the off - shell treatments produce drastic effects at backward angles . a striking manifestation of the above behaviors can be seen while investigating the total cross section ( fig . [ fig : st ] ) . the long lasting shortcoming of the phenomenological models based on effective lagrangian approaches is significantly cured by the inclusion of the off - shell effects(f ) and [ fig : st ] . ] . namely , the total cross - section does not any more show a diverging behavior above @xmath220 1.5 gev ( see also fig . 5 in ref . @xcite ) . in the explored phase space region , the excitation functions and angular distributions for single polarization asymmetries ( fig . [ fig : pol ] ) show significant sensitivity to the off - shell treatments above roughly 1.8 gev for the @xmath222-polarization asymmetry ( @xmath89 ) and polarized target asymmetry ( @xmath59 ) . in the case of the linearly polarized beam asymmetry ( @xmath16 ) the effects are more drastic . indeed , above @xmath220 1.6 gev the off - shell treatments produce a sign change with sizeable magnitudes around 2 gev . the angular distributions for double polarization asymmetries , at @xmath221 = 1.45 and 2.1 gev , are shown in fig . [ fig : coxz ] . a general trend for these observables is that significant off - shell effects appear in the backward hemisphere . in the case of @xmath223 asymmetry , this sensitivity gets attenuated with increasing photon energy . for the other asymmetry ( @xmath224 ) with circularly polarized beam , as well as for the two asymmetries ( @xmath225 and @xmath226 ) with linearly polarized beam , the effects are , on the contrary , enhanced with increasing photon energy . it is worth noticing that the two models without off - shell treatments predict almost vanishing values for @xmath225 and @xmath226 asymmetries , while introducing these treatments results in a sign change and sizeable magnitudes for these asymmetries in the backward hemisphere . we note that the curves depicted in figures [ fig : ds]-[fig : coxz ] split in two families depending on whether the off - shell effects are included ( models b and c ) or not ( models sl and a ) . the cross section for the electroproduction process is given by @xmath228 with @xmath229 the angle between the outgoing kaon and the virtual photon , and @xmath230 the azimuthal angle between the kaon production plane and the electron scattering plane . transverse and longitudinal polarization parameters @xmath231 and @xmath232 , respectively , are defined as @xmath233 \quad , \quad \epsilon_l = - \frac{p_{\gamma}^2}{p_{\gamma 0}^2}\ > \epsilon , \ ] ] with @xmath234 the angle between the momenta of the incoming and outgoing electrons . moreover , @xmath235 is the cross section for an unpolarized incident photon beam , and the term containing @xmath236 is the asymmetry contribution of a transversally polarized beam . the cross section of a longitudinally polarized virtual photon is given by @xmath237 , while @xmath238 contains the interference effects between the longitudinal and transverse components of the beam . in the figures shown in the remaining of this section , the electromagnetic form factors used are the same as in the sl model ( see subsec . iv.d in ref @xcite ) . figure [ fig : ul ] shows the unpolarized component of the differential cross section @xmath239 = @xmath240 , [ see eq . ( [ eq : sigel ] ) ] , as a function of the momentum transfer . all four models reproduce the data equally well . we note again that models b and c give almost identical results . the predictions for different components of the cross section [ eq . ( [ eq : sigel ] ) ] are reported in fig . [ fig : ulip ] . the transverse component ( t ) splits also the four curves in the same two families , with the off - shell effects producing significantly smaller values for this observable . on the contrary , these effects enhance the longitudinal ( l ) part . the transverse - longitudinal ( tl ) interference term shows similar sensitivities . among the observables reported here , the ( l ) and ( tl ) terms are the only ones to produce the most sizeable differences between the models sl and a. finally , the transverse - transverse ( tt ) interference term shows rather negligible dependence on the ingredient of the models . because of the above predictions on the suppression of the transverse component and the enhancement of the longitudinal one due to the off - shell treatments , the ratio @xmath241 is an interesting quantity to be investigated . this latter was already found appealing in the sl approach while examining the effects of hadrons electromagnetic form factors . here , the off - shell treatments have sizeable effect ( fig . [ fig : r ] ) : the ratio @xmath242 between @xmath243 0.5 and 1.0 gev@xmath244 is increased by a factor of @xmath245 2 to 4 , due to such treatments . the amplitudes of the strangeness photoproduction can be related by crossing symmetry @xcite to those of @xmath199 radiative capture processes @xmath247 here , the relevant quantity is the branching ratio defined as @xmath248 with stopped kaons . in table [ tab : br ] , the results of the four models are compared with the only available data point . they all agree with the upper bound of the experimental result . although in the sl model the presence of the @xmath189 resonance was found relevant in reproducing the measured branching ratio ( see table xi in ref @xcite ) , the off - shell treatments are not affecting this observable . this may be due to the fact that here we are dealing only with stopped kaons , and the reported behavior might be altered for kaons in flight . before ending this section , we wish to make a few comments on some general features of the findings summarized in table [ tab : abc ] and/or depicted in figures 1 to 7 . the models discussed in this paper embody 12 ( model a ) to 20 ( model c ) free parameters , see table [ tab : abc ] . in this subsection , we emphasize that , in spite of rather large number of free parameters , our approach offers some meaningful insight into the dynamics of the strangeness electromagnetic production processes . in the fitting procedures , the two main coupling constants ( table [ tab : abc ] ) , @xmath249 and @xmath250 , have been allowed to vary within their broken @xmath251-symmetry limits @xcite . given that for the other couplings we do not dispose of any reliable values or constraints , we will discuss their variations , within the corresponding uncertainties , according to the models ingredients and/or off - shell treatments . the values referred to concern models sl , a , b , and c in table [ tab : abc ] . * _ s - channel : _ no significant variations are observed . * _ u - channel : _ in going from the sl model to model a , the couplings of the @xmath193 and @xmath194 resonances undergo variations of factor 2 to 3 . then the inclusion of off - shell effects ( going from model a to b ) brings them back close to their sl model values , stabilizing them for the c model . these two consecutive variations might come from the observation @xcite that in the sl model these two resonances are rather strongly correlated . this fact , in the absence of any constraint , leads to large variations of the @xmath193 and @xmath194 coupling constants . however , the combined contribution of these resonances to the observables does not show any drastic variation . * _ t - channel : _ significant variations are noticed comparing sl model with the other ones . notice that a spin 5/2 resonance present in the sl model has been removed in the other models . the global increase of the _ t-_channel strengths when discarding a spin 5/2 resonance is a manifestation of the duality hypothesis ( the interplay between s- and _ t-_channel strengths ) in the strangeness sector , as discussed in ref . @xcite . the above considerations indicate strongly that the underlying dynamics retained in this work are tightly constrained by the available data base . hence , the reported sensitivities to off - shell treatments are not likely to be altered significantly by the rest of the free parameters of the models introduced here . in obtaining models b and c we have treated the three parameters ( x , y , and z ) as free ones . as shown in table [ tab : abc ] , out of six off - shell parameters related to the n7 and l8 resonances , the largest one by far is the y parameter for the n7 resonance . the z parameter related to this latter resonance comes out to be compatible with zero . moreover , all three parameters of the n7 resonance are stable upon comparing b and c models . notice that one of the main motivations in introducing the off - shell effects is to cure an undesirable increase in the predicted photoproduction total cross - section above roughly 1.5 gev . by examining the non - pole part @xmath252 of the amplitudes in appendix [ app : invamp ] , one finds [ multiblock footnote omitted ] that for the off - shell parameters @xmath253 and @xmath254 , there are contributions to the invariant amplitudes which rise linearly as a function of the _ s-_variable ( observe that @xmath13 does not participate in this matter ) . hence , the cross section increase stated above might be due to the @xmath253 and @xmath254 values , as obtained from the present minimizations ( table [ tab : abc ] ) exploiting the available data . the authors of ref . @xcite have discussed extensively different `` choices '' of these free parameters , and especially fixing two of them , as reported in the literature @xcite . they conclude that there is no physical basis to attribute fixed values to any of these off - shell parameters . however , to numerically estimate the consequences of eliminating the undesirable _ s_-dependent terms by imposing @xmath255 and @xmath256 , we have performed a minimization within the context of model b. the results for the coupling constants and the only adjusted off - shell parameter ( @xmath13 ) are given in table [ tab : abc ] as model d. we see that the only significant variation compared to model b concerns the @xmath13-parameter . notice that for model b we had already @xmath257 . hence , decreasing the magnitude of the @xmath258 parameter by roughly a factor of 2 ( between models b and d ) leads to an increase of about 20% of the magnitude of the @xmath13-parameter . in figure [ fig : xz ] the photoproduction total cross section and the electroproduction ratio @xmath241 are depicted for both b and d models . in both cases the results from the two models are quite close and the photoproduction total cross section comes out to give slightly higher values using the _ ad hoc _ fixed values for @xmath258 and @xmath259 [ fig . [ fig : xz](a ) ] . other observables discussed in this paper show similar behaviors while comparing models b and d. the closeness of the predictions for the observables can be understood by noticing that the @xmath259 parameter in model b has a value compatible with zero , and the contributions due to the @xmath13-dependent terms dominate numerically over those coming from the @xmath259-dependent ones . in the case of pion photoproduction , the rpi - group @xcite found that imposing @xmath255 and @xmath256 leads to a significant increase of the @xmath200 . this is not the case with the present investigation ( table [ tab : abc ] ) . the reason is that the pion photoproduction was studied in the @xmath260 resonance region , where the reaction mechanism is dominated by this spin-3/2 resonance , while in the case of strangeness production none of the resonances has a paramount role . moreover , we recall that the model b ( c ) studied here contains one ( two ) spin-3/2 resonance and five spin-1/2 resonances . to our knowledge , there are _ a priori _ no bounds on the values of the off - shell parameters . however , remembering that the off - shell freedom comes in only from the non - pole terms , and that the principal contribution from a given resonance must correspond to the propagation of its proper spin , we expect that the corresponding non - pole parts might not dominate the pole part . this would give reasonable upper - bound to which values @xmath258 , @xmath13 , and @xmath259 may take . this expectation was verified in the case of the models reported here . finally , in the case of @xmath209 hyperonic resonance ( model c ) , very small values of the off - shell parameters , as well as those of coupling constants resulting from the minimization endorse our previous affirmations : contributions from this resonance are not required by the existing data base , and the smallness of the relevant free parameters explains why its inclusion in the underlying dynamics does not have significant consequences , neither on the @xmath200 nor on the predicted observables . in the present article , focused on the electromagnetic production of strangeness , we have been concerned with the improvement on the effective hadronic lagrangian approaches by incorporating the correct spin-3/2 resonances propagators and what is called off - shell effects entering the vertices connected to these resonances . the work presented here allows us to preserve the gauge invariance of the formalism , to ensure that each propagator associated with a spin-3/2 exchanged baryon has an inverse , and to include simultaneously both @xmath261 and @xmath262 spin-3/2 resonances . applying our approach to the @xmath15 channels observables investigated in ref @xcite , we have emphasized that the photo- and electro - production of @xmath263 observables show significant sensitivity to the off - shell effects , while these effects do not lead to measurable manifestations in the @xmath199 radiative capture branching ratio with stopped kaons . the numerical results reported here are of course heavily based on the existing data . given the inconsistencies @xcite within the present data , the dynamical content of the models reported here will very likely evolve with the forthcoming high quality data from several experiments , both ongoing and planned . hence the presented numerical results should be considered as guidelines for relative effects . the efforts in refining the phenomenological approaches are then meant to provide us with appropriate tools to interpret the upcoming data . applying the formalism derived in this paper to the available data - base , our results show that the photoproduction data , especially polarization asymmetries , are crucial in pinning down the role of off - shell effects . once these effects are under control , the electroproduction channel can be investigated in studying the electromagnetic form factors of the baryons , kaon and their resonances . these conclusions were reached for the @xmath187 channels and we are currently investigating the complementary @xmath264 processes . there are a few items not discussed in our investigation here for which some comments may be due . in none of the reported approaches ( including this work ) the constructed amplitudes embody unitarity . recently , there has been some attempts to unitarize the amplitude in this process . et al . _ @xcite have performed a _ `` feasibility study '' _ within a chiral color dielectric model adopting a simple two - channel case and using the @xmath265 phase shifts to approximate the @xmath263 elastic scattering in the final state . et al . _ @xcite have developed an su(3 ) chiral dynamics with an effective coupled - channel potential . s_-wave approximation approach is limited to the near threshold region . these works put forward some indications on the effects from the unitarization , but they do not offer a definitive conclusion about the importance of the final state interactions . when several final channels are taken into account , to be realistic , a complete unitarization procedure becomes beyond our current capacity . note also that there has not been any unique way the unitarization should be carried on . we thus believe and hope that , since the finite widths of the resonances are included , some parts contributing towards unitarization have been effectively included in the models discussed in this paper . before ending this section , we wish to discuss briefly two recent and more fundamental approaches applied to some of the processes investigated in this paper . \(i ) _ chiral perturbation theory ( chpt ) _ : this approach _ limited to the threshold region _ , incorporates the coupling to baryon multiplets , as has recently been done in ref . @xcite putting more emphasis on the consequence from the strict chiral symmetry in the construction of lagrangian . here we simply note that the predictions of this model reproduce reasonably the low energy total photoproduction cross sections and the recoil polarization @xmath89 , and are qualitatively consistent with our results . to establish the quality of the _ chpt _ predictions extended to three flavors and including baryons , more data near threshold are needed . \(ii ) _ quark models _ : these models ( upon adopting some type of chiral quark model , for example ref . @xcite ) can predict certain observables with less number of free parameters than the effective hadronic lagrangian ( ehl ) approaches , given the consequence of the difference in the underlying quark models is well within some controllable limit . right now _ the electroproduction process is rather hard to deal with _ by the existing quark model approaches to the strangeness production . we observe that in the current stage of development , the quarks models and the ehl approaches are somewhat complementary . with available data and upon minimization , ehl can supply the values of combined coupling constants like @xmath215 and @xmath216 in our present approach . then , the thus obtained amplitudes are able to predict yet unmeasured observables . once those observables are measured , they will serve in constraining the hadronic coupling values . they are then decomposed into pure hadronic and strong - electromagnetic parts , to constrain the underlying sub - hadronic dynamics ( selecting certain quark models , for example ) . some quantities are present in one approach which are absent in the other ( like the off - shell effects ) . we still lack a microscopic covariant approach within quark models which could , in principle , answer questions related to these aspects . concluding , the complementarity between the effective lagrangian approach and other promising investigations @xcite , provide us with powerful theoretical means to interpret the copious and high quality data to come . we would like to thank jean - christophe david , zhenping li and nimai mukhopadhyay for fruitful discussions , and dietmar menze and reinhard schumacher for helpful exchanges on the experimental results and projects . the well - known chew , goldberger , low and nambu ( @xmath266 ) amplitudes entering into the expressions of the photo- and electro - production observables ( see for example ref . @xcite ) are related to the @xmath267 invariant functions as follows : @xmath268 % { \cal{f}}_2 & = & \frac{\vert { \bbox{p}}_{\gamma}\vert \vert { \bbox{p}}_{k}\vert } { ( e_p + m_p)(e_y + m_y ) } \bigg[({\sqrt s } + m_p ) { \cal{a}}_1 + p_{\gamma } { \cdot}p_p { \cal{a}}_3 + p_{\gamma } { \cdot}p_y { \cal{a}}_4 + p^2_{\gamma}\;{\cal{a}}_5\bigg ] , \\ [ 10pt ] % { \cal{f}}_3 & = & \frac{\vert { \bbox{p}}_{\gamma}\vert \vert { \bbox{p}}_{k}\vert } { ( e_p + m_p ) } \bigg[- 2 p_{\gamma } { \cdot}p_p { \cal{a}}_2 + ( { \sqrt s } + m_p ) { \cal{a}}_4 + p^2_{\gamma}\;{\cal{a}}_6 \bigg ] , \\ [ 10pt ] % { \cal{f}}_4 & = & \frac{{\vert { \bbox{p}}_{k}\vert}^2 } { ( e_y + m_y ) } \bigg[2 p_{\gamma } { \cdot}p_p { \cal{a}}_2 + ( { \sqrt s } - m_p ) { \cal{a}}_4 - p^2_{\gamma}\;{\cal{a}}_6 \bigg ] , \\ [ 10pt ] % { \cal{f}}_5 & = & \frac{{\vert { \bbox{p}}_{\gamma}\vert}^2 } { ( e_p + m_p ) } \bigg[- { \cal{a}}_1 + 2 p_{\gamma } { \cdot}p_y { \cal{a}}_2 + ( { \sqrt s } + m_p)({\cal{a}}_3 - { \cal{a}}_5 ) + ( p_{\gamma } { \cdot}p_y - p_{\gamma } { \cdot}p_p - p^2_{\gamma } ) { \cal{a}}_6\bigg ] , \\ [ 10pt ] % { \cal{f}}_6 & = & \frac{\vert { \bbox{p}}_{\gamma}\vert \vert { \bbox{p}}_{k}\vert } { ( e_y + m_y ) } \bigg[- 2 p_{\gamma } { \cdot}p_y { \cal{a}}_2 + ( { \sqrt s } - m_p ) { \cal{a}}_3 - ( p_{\gamma } { \cdot}p_y - p_{\gamma } { \cdot}p_p - p^2_{\gamma } ) { \cal{a}}_6 - \\ & & \hskip 2.3 truecm { 1\over e_p + m_p } \bigg\ { p_{\gamma 0 } { \cal{a}}_1 + p_{\gamma } { \cdot}p_p { \cal{a}}_3 + p_{\gamma } { \cdot}p_y { \cal{a}}_4 + p_{\gamma 0 } ( { \sqrt s } + m_p ) { \cal{a}}_5 \bigg\}\bigg ] .\end{aligned}\ ] ] the only differences with the relations given in appendix d of ref @xcite appear in the amplitudes @xmath269 and @xmath270 ( contributing only in the electroproduction observables ) where we have the following extra term inside the braces : @xmath271 . here we present the concrete form for the invariant amplitudes decomposed into the pole ( p ) and non - pole ( np ) parts as discussed in eqs . ( [ eq : ajphoto ] ) to ( [ eq : decomp ] ) . the calculation has been done both manually , and by using maple to confirm the validity of the former . @xmath272 , \\ % { d}(\,{\widetilde x } , { \widetilde z}\ , ) & = & { \displaystyle \frac { 1}{12 \,m_r^{2}\ , ( \,p_{\gamma}{\cdot}p_y - p_{\gamma}{\cdot}p_p\ , ) } } \left [ \,(\,2\,m_p - m_r\ , ) - 2\,(\,m_r + 2\,m_y + 2\,m_p\,)\,{\widetilde z } \right . \cr & & \mbox{\hspace{185pt } } -\left . 2\,m_r\,{\widetilde x } + 4\,(\,m_y + 2\,m_r\,)\,{\widetilde x}\,{\widetilde z } \ , \right ] , \\ % { e } & = & \,{\displaystyle \frac { 1}{12\,m_r}}\ , \left [ \ , m_k^2 - \,(\ , m_y + m_r \,)^2 \,\right ] , \\ % { f}(\,{\widetilde x } , { \widetilde z}\ , ) & = & { \displaystyle \frac { 1}{12\,m_r^2}}\ , [ \,m_r - 2\ , m_r\ , { \widetilde z } - 2\ , m_r\ , { \widetilde x } + 4\,(\,m_y+ 2\,m_r\,)\,{\widetilde x}\,{\widetilde z } \ , ] . % \end{aligned}\ ] ] where @xmath290 is the projection operator for spin @xmath124 states . thus this choice of the propagator cuts out the propagation of spin 1/2 states . with this the scattering amplitude eq . ( [ eq : mfi ] ) reads we now look at the @xmath304 vertex whose matrix element is defined in eq . ( [ eq : vert32adel ] ) . with no off - shell freedom implemented , the vertex eq . ( [ eq : vert32 ] ) reads @xmath305i\gamma^5 . \label{eq : vert32noff}\ ] ] j. c. david , c. fayard , g. h. lamot , and b. saghai , phys . c * 53 * , 2613 ( 1996 ) . f. m. renard and y. renard , nucl . phys . * b25 * , 490 ( 1971 ) ; y. renard , _ ibid . _ * b40 * , 499 ( 1972 ) ; y. renard , thse de doctorat detat s - sciences physiques , universit des sciences et techniques du languedoc , 1971 ( in french ) . r. a. adelseck and l. e. wright , phys . c * 38 * , 1965 ( 1988 ) ; r. a. adelseck , ph.d . thesis , ohio university , 1988 . r. a. adelseck , c. bennhold , and l. e. wright , phys . c * 32 * , 1681 ( 1985 ) . m. benmerrouche , r. m. davidson , and nimai c. mukhopadhyay , phys . c * 39 * , 2339 ( 1989 ) . r. m. davidson , nimai c. mukhopadhyay , and r. s. wittman phys . d * 43 * , 71 ( 1991 ) . m. benmerrouche , nimai c. mukhopadhyay , and j. f. zhang , phys . d * 51 * , 3237 ( 1995 ) . m. benmerrouche , phd thesis , rensselaer polytechnic institute ( 1992 ) ; nimai c. mukhopadhyay , j. f. zhang , and m. benmerrouche , proceedings of the _ forth cebaf / int workshop on n * physics _ , seattle , wa , 9 - 13 sept . 1996 , editors : t .- s . h. lee and w. roberts , ( world scientific , 1997 ) . j. d. bjorken and s. d. drell , _ relativistic quantum mechanics _ ( mcgraw - hill , new york , 1964 ) . b. saghai and f. tabakin , phys . c * 55 * , 917 ( 1997 ) . r. a. adelseck and b. saghai , phys . c * 42 * , 108 ( 1990 ) . r. a. williams , c. r. ji , and s. r. cotanch , phys . c * 46 * , 1617 ( 1992 ) . m. bockhorst _ _ , z. phys . * c 63 * , 37 ( 1994 ) . p. l. donoho and r. l. walker , phys . rev . * 112 * , 981 ( 1958 ) ; b. d. mcdaniel _ _ , _ ibid _ * 115 * , 1039 ( 1959 ) ; h. m. brody _ _ , _ ibid _ * 119 * , 1710 ( 1960 ) ; r. l. anderson _ et al . * 9 * , 131 ( 1962 ) ; h. thom _ et al . _ , _ ibid _ * 11 * , 434 ( 1963 ) ; c. w. peck , phys . rev . * 135 * , 830 ( 1964 ) ; r. l. anderson _ et al . _ , _ proc . int . symp . on electron and photon interactions at high energies _ , hamburg , ( 1965 ) ; s. mori , ph.d . thesis , cornell university ( 1966 ) ; h. thom , phys . 151 * , 1322 ( 1966 ) ; d. e. groom and j. h. marshall , _ ibid _ * 159 * , 1213 ( 1967 ) ; r. erbe _ et al . _ , _ ibid _ * 188 * , 2060 ( 1969 ) ; a. bleckmann _ et al . _ , z. phys . * 239 * , 1 ( 1970 ) ; t. fujii _ d * 2 * , 439 ( 1970 ) ; d. decamp _ et al . _ , orsay report , lal 1236 ( 1970 ) ; th . fourneron , thse de doctorat detat , universit de paris , report lal 1258 , 1971 , ( in french ) ; h. goeing _ et al . _ , nucl . b*26 * , 121 ( 1971 ) ; p. feller _ et al . _ ibid _ b*39 * , 413 ( 1972 ) . b. borgia _ _ , nuovo cimento * 32 * , 218 ( 1964 ) ; m. grilli _ _ , _ ibid _ * 38 * , 1467 ( 1965 ) ; d. e. groom and j. h. marshall , phys . rev . * 159 * , 1213 ( 1967 ) ; r. hass _ et al . b*137 * , 261 ( 1978 ) . k. h. althoff _ _ , nucl . phys . * b137 * , 269 ( 1978 ) . c. n. brown _ et al . * 28 * , 1086 , ( 1972 ) ; t. azemoon _ et al . b * 95 * , 77 ( 1975 ) ; c. j. bebek _ et al . _ , phys . d * 15 * , 594 ( 1977 ) ; _ ibid . _ d * 15 * , 3082 ( 1977 ) ; p. brauel _ et al . _ , z. phys . c * 3 * , 101 ( 1979 ) . saphir collaboration ( j. barth _ et al . _ ) contributed paper to the _ 6th conference on the intersections of particle and nuclear physics _ ( cipanp 97 ) , big sky , mt , 27 may - 2 june 1997 ( nucl - th/9707025 ) ; d. menze , _ private communication_. c. r. ji , and s. r. cotanch , phys . rev . c * 38 * , 2691 ( 1988 ) . d. a. whitehouse _ et al . _ , * 63 * , 1352 ( 1989 ) . b. saghai and f. tabakin , phys . c * 53 * , 66 ( 1996 ) . l. m. nath and b. bhattacharyya , z. phys . c * 5 * , 9 ( 1980 ) . d. lu , r. h. landau , and s. c. phatak , phys . c * 52 * , 1662 ( 1995 ) . n. kaiser , t. waas , and w. weise , nucl . a * 612 * , 297 ( 1997 ) . s. steininger and u .- g . meiner , phys . lett . * b391 * , 446 ( 1997 ) . zhenping li , phys . c * 52 * , 1648 ( 1995 ) ; zhenping li , hongxing ye , and minghui lu , _ ibid . _ c * 56 * , 1099 ( 1997 ) . .exchanged particles , coupling constants , and off - shell parameters ( @xmath314 ) for @xmath315 channels from models sl [ 1 ] , and this work ( a , b , and c ) . the reduced @xmath200 s are given in the last row . model a is a simplified version of the sl model with @xmath188 ( spin 1/2 ) and @xmath190 ( spin 5/2 ) resonances removed . all the baryonic resonances have spin 1/2 , except @xmath189 and @xmath209 ( spin 3/2 ) for which off - shell treatment is applied ( models b and c ) . model d is identical to model b , with fixed values @xmath316 , @xmath317 , and @xmath13 free . [ cols="<,^,^,^ , > , > , > , > , > " , ]

over the past decades scanning tunneling microscopy ( stm ) and spectroscopy techniques have been established as an important experimental tool to study strongly correlated electron systems with a spatial resolution down to the atomic scale . in these experiments the tunneling current between the sample and the stm tip is measured as a function of its position and the applied voltage . the tunneling current is directly related @xcite to the local density of states ( ldos ) in the sample ; thus stm experiments provide a tool to investigate the spatial dependence of the ldos , which may arise for example due to impurities . these spatial modulations can be analyzed in terms of the fourier transform of the tunneling conductance which is directly proportional to the fourier transform of the ldos . using this method of analyzing stm data one can infer informations about the bulk state of matter like the properties of its quasiparticle excitations . for example , this has been used to study high - temperature superconductors @xcite as well as carbon nanotubes @xcite . theoretical studies of the ldos and stm in the presence of boundaries have been carried out in particular for one - dimensional systems including luttinger liquids @xcite , open hubbard chains @xcite , and charge - density wave ( cdw ) states @xcite . in these systems the coupling to impurities is relevant @xcite , which leads , at sufficiently low energies , to an effective cutting of the system into disconnected pieces . in this sense the effect of an impurity can be modeled by a boundary condition . previous studies of the fourier transform of the ldos in both gapless @xcite and gapped @xcite systems revealed a pinning of the cdw at the impurity as well as individually propagating spin and charge degrees of freedom . here we study a quarter - filled one - dimensional mott insulator in the presence of a strong impurity potential . it has been established @xcite that for sufficiently strong interactions double umklapp processes are relevant and generate a gap in the charge sector . as a microscopic realization one may think of a quarter - filled hubbard model extended by including nearest - neighbor interactions , which possesses @xcite an insulating phase for sufficiently strong repulsions . experimentally these systems are relevant for the description of organic quasi - one - dimensional conductors , i.e. the bechgaard and fabre salts @xcite . in this article we study the low - energy behavior of the ldos by employing the field theoretical description of the mott insulator . the spectrum consists of massless spin excitations ( spinons ) carrying spin @xmath0 but no charge and massive charge excitations ( solitons and antisolitons ) carrying charge @xmath1 but no spin . in this framework the impurity is modeled by a boundary for the collective excitations . in particular , we will focus on signatures of dispersing quasiparticles in the spatial fourier transform of the ldos , as they have been clearly observed in similar studies of luttinger liquids @xcite and half - filled mott insulators as well as cdw states @xcite . the spectral function of the translationally invariant , quarter - filled mott insulator has been previously derived @xcite by essler and tsvelik . from the quantum numbers stated above it is clear that an electron will fractionalize into at least one spinon and two antisolitons . thus the spectral function exhibits a featureless scattering continuum of at least three excitations , which is also consistent with experimental results @xcite from angle - resolved photoelectron spectroscopy on bechgaard salts . this article is organized as follows . in sec . [ sec : model ] we will briefly discuss the field theoretical description of the quarter - filled mott insulator . in sec . [ sec : gf ] we present our results for the single - particle green function . this will then be used to derive an analytic expression for the ldos , which we will discuss in detail in sec . [ sec : ldos ] . our main result is presented in eq . as well as figs . [ fig : plot1 ] and [ fig : plot2 ] , which show the fourier transform of the ldos in the presence of an impurity . surprisingly , we find two dispersing features which follow the dispersion relations and respectively . these features can be interpreted as arising from an antisoliton or a spinon - antisoliton pair propagating between the position of the stm tip and the impurity . in addition , we observe a non - dispersing singularity at momentum @xmath2 which is indicative of a pinning of a spin - density wave ( sdw ) at the boundary . a detailed analysis further reveals dispersing features originating in propagating spinons and antisoliton pairs . the suppression of these features is attributed to the `` quasi - fermionic '' behavior of antisolitons with equal momenta . finally we explain the different findings in the spectral function and the ldos . in sec . [ sec : bbs ] we discuss the effect of possible boundary bound states on the ldos and in sec . [ sec : hfmi ] we compare our results to the half - filled mott insulator studied @xcite previously . finally , in sec . [ sec : discussion ] we set our results in the context of other one - dimensional systems and discuss implications for stm experiments on quasi - one - dimensional materials . readers mainly interested in the qualitative features of the ldos may start with this section and study the detailed results afterwards . technical details have been moved to the appendix . as the underlying microscopic model for the quarter - filled mott insulator one may start with the extended hubbard model @xcite @xmath3\\ & + u\sum_j n_{j,\uparrow}\,n_{j,\downarrow}+v\sum_j n_{j}\,n_{j+1 } , \label{eq : exthm } \end{split}\ ] ] where @xmath4 and @xmath5 are electron creation and annihilation operators at site @xmath6 with spin @xmath7 , @xmath8 , and @xmath9 . at quarter filling and for sufficiently large on - site and nearest - neighbor repulsions @xmath10 this model possesses a mott insulating phase @xcite which can be thought of as a microscopic realization of the field - theoretical system we will study below . the extended hubbard model can be used as an effective model for the description of bechgaard salts @xcite ; its spectral and transport properties have been investigated in detail using various techniques @xcite . instead of working with the microscopic model we will study here the corresponding low - energy field theory . the continuum description is obtained by focusing on the degrees of freedom around the fermi points @xmath11 . we introduce slowly varying right- and left - moving fermi fields as @xmath12 and similarly for the electron creation operators . here @xmath13 denotes the lattice spacing , @xmath14 , and the fermi momentum is given by @xmath15 at quarter filling . the impurity potential is assumed to be strong such that we can model it by a boundary condition on the continuum electron field @xmath16 using standard bosonization of the right- and left - moving fermi fields it was shown @xcite that double umklapp processes result in a cosine term in the charge sector , which is relevant at low energies and generates a gap . thus the effective low - energy hamiltonian is given by @xcite @xmath17\nonumber\\ * & & -\frac{g_\mathrm{c}}{(2\pi)^2}\int_{-\infty}^0 dx\ , \cos\bigl(\beta\phi_\mathrm{c } ) , \label{eq : chargehamiltonian}\\ h_\mathrm{s}&=&\frac{v_\mathrm{s}}{16\pi}\int_{-\infty}^0 dx \bigl[\bigr(\partial_x\phi_\mathrm{s}\bigr)^2 + \bigr(\partial_x\theta_\mathrm{s}\bigr)^2\bigr ] , \label{eq : spinhamiltonian}\end{aligned}\ ] ] where , as a consequence of , the canonical bose fields @xmath18 are subject to the hard - wall boundary conditions @xmath19 the fields @xmath20 are dual to @xmath18 . the charge and spin velocities @xmath21 , the luttinger parameter @xmath22 , and the coupling constant @xmath23 are functions of the hopping and interactions in an underlying microscopic model . within the field theory they can be viewed as phenomenological parameters . experimental estimates @xcite for the luttinger parameter in the bechgaard salts yield @xmath24 . in the bosonization procedure leading to the effective low - energy hamiltonian the explicit relation between the fermi and the bose fields is given by @xcite @xmath25\nonumber\\ * & & \qquad\qquad\times\exp\biggl[\frac{{\text{i}}}{4}f_\sigma\bigl(\phi_\mathrm{s } + \theta_\mathrm{s}\bigr)\biggr],\label{eq : bosonizationr}\\ * l^\dagger_\sigma&=&\frac{\eta_\sigma}{\sqrt{2\pi}}\ , \exp\biggl[-\frac{{\text{i}}}{4}\biggl(\frac{\beta}{2}\phi_\mathrm{c } -\frac{2}{\beta}\theta_\mathrm{c}\biggr)\biggr]\nonumber\\ * & & \qquad\qquad\times\exp\biggl[-\frac{{\text{i}}}{4}f_\sigma\bigl(\phi_\mathrm{s } -\theta_\mathrm{s}\bigr)\biggr],\label{eq : bosonizationl}\end{aligned}\ ] ] where the klein factors @xmath26 satisfy the anticommutation relations @xmath27 and @xmath28 . the charge excitations are described by the sine - gordon model on the half - line ( [ eq : chargehamiltonian ] ) . for @xmath29 the cosine term is relevant and generates a gap . the excitations are massive solitons and antisolitons which carry charges @xmath30 respectively . they possess a relativistic dispersion relation ; we parametrize their energy and momentum in the usual way using the rapidity @xmath31 by @xmath32 the soliton mass @xmath33 is a function @xcite of the bare parameters @xmath23 and @xmath22 but will be viewed here as a phenomenological parameter replacing the coupling constant @xmath23 . in the regime @xmath34 solitons and antisolitons are the only excitations in the charge sector ; for @xmath35 propagating breather ( soliton - antisoliton ) bound states exist as well . at the luther - emery point ( lep ) @xmath36 the charge sector is equivalent to a free massive dirac theory @xcite . the sine - gordon model ( [ eq : chargehamiltonian ] ) with the boundary condition is integrable@xcite , which we will use below to calculate the necessary correlation functions of right- and left - moving fermi fields . the spin sector ( [ eq : spinhamiltonian ] ) describes massless relativistic spinon excitations which propagate with velocity @xmath37 and carry spin @xmath0 . we have already assumed spin rotational invariance , i.e. a possible luttinger parameter in the spin sector is set to one . we note that the half - filled mott insulator studied previously @xcite possesses the same effective low - energy hamiltonian . the difference to the quarter - filled case studied in this article is given @xcite by the bosonization relations and for the right- and left - moving fermi fields . this leads to differing green functions as well as ldos . we will proceed with the presentation of the results for the quarter - filled mott insulator and present a detailed comparison to the half - filled system in sec . [ sec : hfmi ] . in order to derive the ldos below we calculate the time - ordered green function in euclidean space , @xmath38 where @xmath39 is the ground state of in the presence of the boundary and @xmath40 denotes imaginary time . at low energies we can linearize around the fermi points @xmath41 , which results in @xmath42 where e.g. @xmath43 . for the calculation of the ldos we have to set @xmath44 below . in particular , we will focus on the spatial fourier transform of the ldos as physical properties can be more easily identified . the four terms in the decomposition then contribute in different regions in momentum space , i.e. @xmath45 and @xmath46 contribute for @xmath47 while @xmath48 and @xmath49 contribute for @xmath50 . in the translationally invariant system right- and left - moving fields in the spin sector decouple which results in @xmath51 . the presence of a boundary couples right and left sectors and concomitantly the fourier transform of the green function ( [ eq : gflowenergy ] ) acquires a non - zero component at @xmath47 . we will focus on the @xmath52-part of the green function and the ldos in the following , since this provides a particularly clean way of investigating boundary effects . we note that the other components of the green function can be calculated in the same way . due to the spin - charge separated hamiltonian the green function @xmath53 factorizes into a product of correlation functions in the spin and charge sectors . the correlation functions in the spin sector can be straightforwardly derived using standard methods like boundary conformal field theory @xcite . on the other hand , the integrability of the sine - gordon model on the half - line enables us to calculate correlation functions in the charge sector using the boundary state formalism @xcite together with a form - factor expansion @xcite . this yields the following result for the @xmath52-component of the green function @xmath54 the infinite series originates from the applied form - factor expansion , which constitutes an expansion of correlation functions in the number of contributing solitons and antisolitons . in app . [ app : gf ] we derive explicit expressions for the first three terms @xmath55 which correspond to two - particle processes in the charge sector . we note that is valid at zero temperature . the effect of small temperatures @xmath56 on the gapless spin sector can be incorporated using conformal field theory @xcite as discussed for the half - filled mott insulator in ref . . in the next section we use the result to derive the fourier transform of the ldos . the ldos can now be calculated directly from the green function . in order to analyze the physical properties it is useful to consider the spatial fourier transform of the ldos as features like dispersing quasiparticles can be more easily identified . this technique was previously applied to the ldos of luttinger liquids @xcite as well as cdw states and half - filled mott insulators @xcite . the fourier transform of the ldos for positive energies @xmath57 is given by @xcite @xmath58 here the green function has been analytically continued to real times and we have taken the limit @xmath59 . we note that due to the assumed spin rotational invariance the ldos does not depend on @xmath60 . as mentioned before , we will concentrate on the @xmath52-component as it vanishes in the absence of the boundary and hence offers a particularly clean way of investigating boundary effects . for @xmath61 only @xmath62 contributes and starting from we arrive at our main result ( @xmath63 ) @xmath64 where @xmath65 as well as @xmath66 , @xmath67 , and @xmath68 . explicit integral representations for the scattering matrix @xmath69 , the boundary reflection matrix @xmath70 , and the function @xmath71 are given in app . [ app : ff ] . physically @xmath69 describes the scattering of two antisolitons with relative momentum corresponding to the rapidity @xmath31 , @xmath70 is the reflection amplitude of an antisoliton with rapidity @xmath31 off the boundary [ and thus incorporates the information on the boundary condition ] , and @xmath71 appears in the matrix element [ i.e. form factor ] of the charge part of and with two - antisoliton states [ see ] . the parameter @xmath72 is defined as @xmath73 . the normalization constant@xcite @xmath74 has dimension @xmath75 ; thus in order to obtain dimensionless quantities we multiply the ldos by @xmath76 in all figures . the singularity of the integrands in is smeared out by taking @xmath77 small but finite ; we choose @xmath78 unless stated otherwise . this mimics the broadening of singularities in experiments due to the instrumental resolution and the finite temperature . the ldos vanishes for @xmath79 as the fractionalization of an electron creates at least two antisolitons . the three terms constitute the leading contribution from the form - factor expansion . as can be seen from the double integrals they all originate in two - particle processes in the charge sector . the number of boundary reflection matrices @xmath70 indicates that @xmath80 involves @xmath81 interactions with the boundary , i.e. the terms contain no , one , or two reflections of charge excitations . the sub - leading contributions involve at least three particles in the charge sector which either come from a higher number of particles in the intermediate state or from higher - order processes due to the boundary . similar terms were thoroughly analyzed for the ldos in a cdw state at the lep @xcite as well as the spectral function in the ising model with a boundary magnetic field @xcite ; they were found to be negligible in both cases . we further note that the suppression of sub - leading terms in the form - factor expansion for bulk two - point functions is a well - known feature of massive integrable field theories @xcite . we thus expect the higher - order corrections to to be negligible in the low - energy regime we consider here . : @xmath82 and @xmath83 for @xmath84 and @xmath85 ( upper panel ) and @xmath86 ( lower panel ) . both plots are shown on the same scale . we observe a peak at @xmath87 and dispersing features at @xmath88 as well as @xmath89 . for @xmath90 the dispersing features are strongly suppressed . ] the constant energy scan of the fourier transform of the ldos is shown in fig . [ fig : plot1 ] . we observe a singularity at @xmath87 , i.e. @xmath2 , which is indicative of the pinning of the @xmath52-sdw at the boundary ( see sec . [ sec : ldos0 ] ) . furthermore there is a peak at @xmath91 . the constant momentum scans shown in fig . [ fig : plot2 ] reveal that this peak shows dispersing behavior and splits above a critical momentum @xmath92 [ @xmath93 will be defined in eq . ] . more precisely these features follow the dispersion relations and respectively ( see sec . [ sec : ldos1 ] ) . the existence of dispersing features is at first surprising since the spectral function @xcite of the translationally invariant system does not show any dispersing quasiparticles . the processes contributing to the ldos can be thought of as arising from creating an electron at a position @xmath94 , the fractionalization of this electron into at least one spinon and two antisolitons , the propagation of these elementary excitations through the system , and their subsequent recombination at the position @xmath94 . this allows a natural interpretation of the leading terms in the form - factor expansion in terms of propagating antisolitons and spinons . in the next sub - sections we will analyze the different terms separately . afterwards we comment on the different findings in the spectral function and the ldos . for @xmath85 and @xmath84 . the curves are constant @xmath95-scans which have been offset along the y - axis by a constant with respect to one another . the ldos is dominated by the peak at @xmath87 . we further observe dispersing features at @xmath96 as well as @xmath97 for @xmath92 . ] let us start with the analysis of the first term in the form - factor expansion , which is shown in fig . [ fig : plot3 ] . we first note that this term is dominated by a singularity at @xmath87 , i.e. @xmath2 . this singularity is due to the pinning of the @xmath52-sdw at the boundary . as @xmath15 we find a spatial modulation with periodicity @xmath99 as sketched @xcite in fig . [ fig : plot3].b . close to @xmath87 the first term in the form - factor expansion and thus the whole ldos behaves as @xcite @xmath100 we note that the exponent is independent of the luttinger parameter @xmath22 ( see fig . [ fig : plot3].a ) . this singularity is similar to the ones observed in the ldos of luttinger liquids @xcite as well as cdw states @xcite where , however , the exponents depend on the interaction strength and thus the luttinger parameter . for @xmath101 . a ) logarithmic plot close to the singularity at @xmath87 for @xmath102 . the dotted line is a guide to the eye ; it has gradient @xmath103 . b ) sketch @xcite of the @xmath52-sdw pinned at the boundary . c ) @xmath104 for @xmath85 and different momenta . we observe a dispersing feature at @xmath105 indicated by small arrows . d ) sketch of the process giving rise to the propagating feature . the electron decomposes into two zero - momentum antisolitons ( solid lines ) and one spinon ( dashed line ) reflected at the boundary . ] furthermore , @xmath106 possesses a weak dispersing feature following ( see @xcite fig . [ fig : plot3].c ) @xmath107 as already discussed , the created electron fractionalizes into two antisolitons and ( at least ) one spinon which can propagate through the system . in this way of thinking the dispersing feature at @xmath108 arises from two antisolitons with zero momentum ( solid lines in fig . [ fig : plot3].d ) contributing an energy @xmath109 and a massless soliton with momentum @xmath95 ( dashed line in fig . [ fig : plot3].d ) . the appearance of @xmath110 in @xmath108 is due to the fact that the spinon has to propagate to the boundary and back , thus covering the distance @xmath111 in time @xmath112 . for @xmath85 , @xmath101 , and different momenta . we observe dispersing features at @xmath96 indicated by up - arrows and @xmath97 indicated by down - arrows . b ) process resulting in @xmath114 . the electron decomposes into one antisoliton and spinon with zero momentum as well as one antisoliton reflected at the boundary . c ) process resulting in @xmath115 . ] the next term in the form - factor expansion contains one boundary reflection matrix @xmath70 . thus it is natural to interpret it as arising from processes involving the reflection of one antisoliton at the boundary . in fig . [ fig : plot4].a we show @xmath116 for different momenta . we clearly observe a dispersing feature ( indicated by up - arrows ) following @xmath117 this feature originates in the process where one antisoliton propagates with momentum @xmath95 while the other antisoliton and the spinon possess zero momentum ( see fig . [ fig : plot4].b ) . the first term in @xmath114 is the rest mass of the zero - momentum antisoliton while the second term is the energy of the propagating antisoliton . furthermore , when @xmath95 exceeds the critical value @xmath93 a second dispersing feature appears ( indicated by down - arrows in fig . [ fig : plot4].a ) at @xmath118 the critical momentum @xmath93 can be determined from the condition that the spin and charge excitations have the same group velocity @xmath119 which leads to @xmath120 thus this feature can be thought of as arising from an antisoliton with momentum @xmath93 and a spinon carrying momentum @xmath121 , while the second antisoliton stays at rest ( see fig . [ fig : plot4].c ) . the fact that the propagating antisoliton and the spinon have the same group velocity results in an increased probability that both simultaneously arrive at position @xmath94 after reflection at the boundary . the observed splitting of the quasiparticle peak is reminiscent of what is found for the single - particle spectral function in the translationally invariant , half - filled mott insulator @xcite as well as the ldos of a cdw state in the presence of a boundary in the regime of attractive interactions @xcite . the peak splitting is a consequence of the curvature of the antisoliton dispersion ( and thus of the charge gap ) together with the condition @xmath122 . hence it can not be observed in the luttinger liquid case @xcite where both sectors are massless . for @xmath85 , @xmath101 , and different momenta . we observe dispersing features at @xmath124 ( up - arrows ) and @xmath125 ( down - arrows ) . processes resulting in ( b ) @xmath126 and in ( c ) @xmath127 . ] the leading term in second order in the boundary reflection matrix is @xmath128 . as in the previous term we find two dispersing features . first , indicated by up - arrows in fig . [ fig : plot5].a , there is a feature following @xmath129 this feature can be thought of as arising from the propagation of both antisolitons with momenta @xmath130 and thus the same velocity , while the spinon has zero momentum ( see fig . [ fig : plot5].b ) . the second feature is found at @xmath131 provided @xmath95 exceeds the critical momentum @xmath132 . this feature is indicated by down - arrows in fig . [ fig : plot5].a . the interpretation is that both antisolitons carry momentum @xmath93 while the excess momentum @xmath133 is carried by the spinon ( see fig . [ fig : plot5].c ) . we note that at finite momentum @xmath91 the term @xmath116 dominates the ldos as can be seen by comparing the scales in figs . [ fig : plot3].c , [ fig : plot4].a , and [ fig : plot5].a , respectively . more importantly , the dispersing features originating in @xmath116 are much more pronounced than those resulting from @xmath106 and @xmath128 . we attribute this to the fact that , at least in the naive interpretation of figs . [ fig : plot4].b and c , the antisolitons involved in processes contributing to @xmath116 possess different momenta as one of them is at rest while the other propagates through the system . in contrast , the dispersing features in @xmath106 and @xmath128 originate from processes in which the two antisolitons possess equal momenta ( see figs . [ fig : plot3].d as well as [ fig : plot5].b and c , respectively ) . thus their relative rapidity vanishes and the scattering matrix is @xmath134 ( see app . [ app : ff ] ) . the resulting `` quasi - fermionic '' behavior leads to a strong suppression of the dispersing features in @xmath106 and @xmath128 via the pauli principle . finally let us comment on the different findings in the spectral function @xcite of the translationally invariant system and the ldos in the presence of a boundary . the spectral function is obtained from the green function @xmath135 via fourier transformation with respect to @xmath136 and @xmath94 . the contributing processes require the propagation of all excitations , and in particular both antisolitons , from @xmath137 to @xmath138 . thus one may expect a dispersing feature at @xmath139 from the antisolitons propagating with the same momentum @xmath130 . in addition , above the critical momentum @xmath140 a linearly dispersing feature following @xmath141 is expected . indeed these are exactly the conditions @xcite for the threshold below which the spectral function vanishes . above the threshold the spectral function is rather featureless , in particular there are no dispersing peaks associated with the propagation of charge or spin degrees of freedom . the absence of dispersing features may be attributed to the `` quasi - fermionic '' behavior of antisolitons , as all processes contributing to the spectral function will contain antisolitons with equal momenta . in contrast , the processes contributing to the ldos are given by the fractionalization of an electron at position @xmath94 , the propagation of the two antisolitons and the spinon through the system , and the subsequent recombination at the original position @xmath94 . in particular , processes where the momentum @xmath95 is solely carried by one of the antisolitons or a spinon - antisoliton pair will contribute , resulting in the dispersing features at @xmath114 and @xmath115 , respectively . the observation of these features thus underlines that the spectral function and the ldos probe rather different physical processes . while the spectral function requires the propagation of all excitations between two points in the system , the ldos also probes processes where only one or two excitations propagate to the impurity and back . up to now we have concentrated on the boundary conditions for the electron field or equivalently for the canonical bose fields . in this section we will consider more general phase shifts for the electron fields which lead to arbitrary dirichlet boundary conditions for the bose fields . as is well known , such boundary conditions can result in the existence of boundary bound states ( bbs s ) , which are expected @xcite to manifest themselves in the ldos as features within the spectral gap . here we will focus on general dirichlet boundary conditions in the charge sector @xmath142 while in the spin sector we still assume @xmath143 . in the boundary state formalism @xcite applied here the information about the boundary condition is encoded in the boundary reflection matrix @xmath70 . in particular , the existence of a bbs will show up as a pole in the physical strip @xmath144 . as was shown in ref . this will be the case provided @xmath145 the pole at @xmath146 corresponds to a bbs with energy @xmath147 when calculating the green function the pole of the boundary reflection matrix will lead to two additional terms , @xmath148 and @xmath149 , in the form - factor expansion . this in turn yields two additional terms in the ldos . for @xmath150 , @xmath151 , and @xmath84 . the curves are constant @xmath95-scans which have been offset along the y - axis by a constant with respect to one another . the existence of a bbs results in spectral weight within the two - soliton gap @xmath79 . ] to be explicit , let us consider the lep @xmath36 in the following . the three leading terms in the form - factor expansion of the ldos are still given by where the boundary reflection matrix now depends explicitly on @xmath152 @xmath153 this obviously possesses a pole in the physical strip @xmath144 provided @xmath154 corresponding to the existence of a bbs . the resulting two additional terms in the ldos read ( @xmath155 ) @xmath156 where @xmath157 as well as @xmath158 and @xmath159 . we note that @xmath160 has the dimension of an inverse length . the spatial dependence of the terms @xmath148 and @xmath149 ( see app . [ app : gf ] ) is dominated by @xmath161 ( @xmath162 ) , thus it is natural to interpret @xmath163 as the width of the bbs . the terms possess a threshold at @xmath164 corresponding to the creation of the boundary bound state and an additional antisoliton . interestingly , although @xmath165 and @xmath166 separately possess jumps at the threshold , the sum of both terms is continuous . this is reminiscent of the bbs contributions in the ising model with a boundary magnetic field @xcite , where a cancellation of singularities yields a smooth spectral function at the threshold . the ldos in the presence of a bbs is shown in fig . [ fig : plot6 ] . we clearly observe a contribution within the gap @xmath79 . in contrast to bbs s in the cdw state @xcite there is , however , no singularity at @xmath167 as the underlying processes involve the creation of a spinon , an antisoliton , and the bbs . as before the ldos is dominated by a singularity at @xmath87 and we observe dispersing features following and respectively . the ldos of a half - filled mott insulator in the presence of a boundary was analyzed in detail @xcite in ref . . as already mentioned , the effective low - energy hamiltonian is also given by but the expressions for the right- and left - moving fermi fields in terms of the bose fields differ from and . from the physical point of view right- and left - moving fermi fields create and annihilate individual antisolitons in the half - filled case whereas they create and annihilate antisoliton pairs in the quarter - filled system . thus the gap in the ldos is given by @xmath33 in the former and @xmath109 in the latter system . apart from this we find the following similarities and differences . in both cases the ldos possesses a singularity at @xmath87 which behaves as @xmath168 independently @xcite of the luttinger parameter @xmath22 . this singularity is due to the pinning of the @xmath52-sdw at the boundary , which possesses a wave length @xmath169 or @xmath170 respectively . beside this singularity the ldos of the half - filled mott insulator shows dispersing features at @xmath171 , at @xmath172 , and , for @xmath92 , at @xmath173 . an interpretation is obtained by noting that the electron decomposes into one spinon and one antisoliton , which then propagate through the system . as the leading processes involve only individual antisolitons , the `` quasi - fermionic '' behavior of the antisolitons discussed at the end of sec . [ sec : ldos2 ] does not affect the ldos . in contrast , in the quarter - filled system we observe only two dispersing modes following and respectively ( see fig . [ fig : plot2 ] ) . all other dispersing features discussed above are strongly suppressed due to the `` quasi - fermionic '' behavior of the antisolitons . finally , in the presence of a bbs with energy @xmath174 the ldos of the half - filled mott insulator possesses a non - dispersing singularity with @xmath175 which originates from the creation of a spinon and the bbs . in contrast , in the quarter - filled case the ldos is smooth at the threshold @xmath164 which originates from the simultaneous creation of one antisoliton and the bbs . in this section we compare qualitative aspects of the ldos of various one - dimensional systems in the presence of a boundary or strong impurity potential . we discuss the spatial fourier transform @xmath176 defined in . specifically we consider luttinger liquids @xcite , half - filled mott insulators ( or cdw states ) @xcite , and the quarter - filled mott insulator investigated in sec . [ sec : ldos ] . we focus on the @xmath61-component of the ldos as it vanishes in the translationally invariant systems and thus provides a particularly clean way to investigate boundary effects . ( we note that the ldos in the small - momentum regime @xmath50 behaves qualitatively similar . ) the first feature , which all three systems have in common , is a singularity of the ldos @xmath177 at @xmath87 , which originates in the pinning of a charge or spin density wave at the impurity . in the case of a luttinger liquid the strength of this singularity depends on the luttinger parameter @xmath178 and thus on the interactions between the electrons , whereas in both the half - filled and quarter - filled mott insulator the ldos behaves as @xcite @xmath179 . furthermore , as was pointed out in ref . , the gap of the ldos in the half - filled mott insulator is given by the soliton mass @xmath33 , which equals the thermal activation gap and is half of the gap observed in optical measurements . in contrast , in the quarter - filled system the ldos possesses a gap of @xmath109 , which is equal to the optical gap but twice the thermal gap . of the ldos in different one - dimensional systems . a ) in a luttinger liquid one finds two linearly dispersing modes . b ) in a half - filled mott insulator ( mi ) there exists a gap @xmath33 . one observes one linear and one curved dispersion . from the latter a third feature splits off at @xmath180 . c ) the quarter - filled mott insulator possesses a gap @xmath109 . one further observes one curved dispersion from which a linear mode splits off at @xmath180 . in both ( b ) and ( c ) the thermal activation gap is given by @xmath33.,width=321 ] in addition , the ldos possesses various dispersing features ( see fig . [ fig : plot7 ] ) . in the luttinger liquid case one observes two linearly dispersing modes corresponding to individually propagating spin and charge degrees of freedom respectively ( see fig . [ fig : plot7].a ) . in the half - filled mott insulator an electron decomposes into massless spin excitations and one charge excitation with mass @xmath33 . propagating spin excitations give rise to a linearly dispersing mode with velocity @xmath37 , while the propagating charge excitation results in a curved dispersion relation . in addition , there exists a critical momentum @xmath180 at which the group velocity of the charge excitation equals @xmath37 . at this momentum a third linearly dispersing feature splits away from the charge mode , which has its origin in the collective propagation of spin and charge degrees of freedom with equal velocities ( see fig . [ fig : plot7].b ) . we note that the qualitative behavior discussed for the half - filled mott insulator is also expected in other systems possessing gapless excitations in one sector and gapped ones in the other like , for instance , cdw states . finally , in the quarter - filled mott insulator an electron decomposes into massless spin excitations and two charge excitations with masses @xmath33 . this results in a curved dispersion originating in the propagation of one charge excitation while the second one stays at the position of the stm tip . this dispersion splits at the critical momentum @xmath93 due to the collective propagation of one charge excitation and additional spin degrees of freedom ( see fig . [ fig : plot7].c ) . we note that one does not observe individually propagating spin excitations . the relevant processes would require the two charge excitations to possess equal momenta , which is forbidden by their `` quasi - fermionic '' behavior ( see sec . [ sec : ldos2 ] ) . to sum up , at sufficiently large momenta @xmath95 we observe two linearly dispersing modes in a luttinger liquid , two linear and one curved dispersion relation in a half - filled mott insulator , and one linear and one curved dispersion in a quarter - filled mott insulator . this clearly demonstrates that characteristic properties of the bulk state of matter and its electronic excitations can be infered from studying spatial modulations of the ldos in the presence of an impurity . the ldos discussed in this article is directly related to the tunneling current measured in stm experiments . if the density of states in the stm tip is approximately independent of energy , the tunneling conductance will be given by the thermally smeared ldos of the sample at the position of the tip @xcite @xmath181 where @xmath182 denotes the fermi function . @xmath176 can then easily be obtained via spatial fourier transformation . in this context the models discussed above apply to quasi - one - dimensional materials at energies above the cross - over scale to three - dimensional behavior . this situation might be experimentally realized for example in stripe phases of high - temperature superconductors @xcite , carbon nanotubes @xcite , bechgaard and fabre salts @xcite , self - organized atomic gold chains @xcite , and two - leg ladder materials @xcite . as discussed above already the knowledge of qualitative properties such as the presence of a spectral gap and its relation to the thermal activation gap as well as the number and positions of singularities can reveal substantial information about the electronic properties and excitations of quasi - one - dimensional materials . a more detailed analysis of quantitative features like the strength of singularities or the precise positions of dispersing modes can even lead to estimates for the luttinger parameter and the velocities of charge and spin excitations . there are , of course , experimental limitations . obviously the experimental resolution and thermal broadenings have to be sufficiently small to clearly distinguish the different singularities and dispersing modes . moreover , the simplyfied relation does not contain @xcite the matrix elements for the tunneling from the sample into the tip , which may introduce a non - trivial spatial dependence of the tunneling conductance on the ldos and thus hamper a natural interpretation in terms of propagating excitations . in conclusion , we have studied a quarter - filled mott insulator in the presence of an impurity , which was modeled by a boundary condition for the electron fields . in this system we calculated the @xmath52-component of the spatial fourier transform of the ldos using the boundary state formalism together with a form - factor expansion . the ldos is dominated by a singularity at @xmath2 , which is indicative of a pinning of a sdw at the impurity . furthermore , we observed several dispersing features above the two - soliton gap @xmath79 . we identified their physical origin as spin and charge degrees of freedom , which propagate between the stm tip and the impurity , and attributed their relative strength to the `` quasi - fermionic '' behavior of charge excitations with equal momenta . this also explains the absence of dispersing features in the spectral function @xcite of the translationally invariant system and underlines that the spectral function and the ldos probe rather different physical processes . we further showed that a bbs in the charge sector leads to a non - vanishing ldos within the two - soliton gap . finally , we discussed our results in the context of other one - dimensional systems , i.e. luttinger liquids and half - filled mott insulators , and argued that the measurement of the spatial modulations in the ldos can be used to extract detailed informations about the electronic states of quasi - one - dimensional systems . i would like to thank fabian essler , markus garst , markus morgenstern , alexei tsvelik , and particularly volker meden for valuable discussions and comments . this work was supported by the german research foundation ( dfg ) through the emmy noether program . in this appendix we derive the leading terms @xmath55 in the form - factor expansion of the charge part of the green function , i.e. we determine the correlation function @xmath183 in the sine - gordon model on the half line . here the operators @xmath184 and @xmath185 are given by the charge parts of and respectively . we calculate using the boundary state formalism @xcite together with a form - factor expansion @xcite . the same technique was previously applied @xcite to determine the green function of a cdw state in the presence of an impurity . here we restrict ourselves to the main steps of the derivation and concentrate on the differences as compared to ref . , to which we refer implicitly for all details and definitions not discussed here . by performing a rotation in euclidean space we map @xcite the problem onto the sine - gordon model on the real axis . the boundary condition is translated into an initial condition encoded in a boundary state . expanding the latter in powers of the boundary reflection matrix and inserting a resolution of the identity between the operators in we find @xmath186 where we have defined the auxiliary functions @xmath187 & \hspace{-20mm}\times { \left\langle0\right|}o_r(\tau , x_1 ) { \left|\theta_{n},\ldots,\theta_{1}\right\rangle}_{c_n,\ldots , c_1}\ ; ^{c_1,\ldots , c_n}\!{\left\langle\theta_{1},\ldots,\theta_{n}\right| } o_l^\dagger(0,x_2 ) { \left|-\theta'_1,\theta'_1,\ldots,-\theta'_m,\theta'_m\right\rangle } _ { a_1,b_1,\ldots , a_m , b_m}. \end{split}\ ] ] here the states @xmath188 ( and so on ) are scattering states of solitons and antisolitons , the indices @xmath189 take the values @xmath190 and label the corresponding u(1 ) charges , and energy and momentum of solitons and antisolitons are given in terms of the rapidities @xmath191 and @xmath192 by . we note that constitutes an expansion in two parameters : ( i ) the number of particles ( with rapidities @xmath191 ) in the intermediate state and ( ii ) the number of reflections of particles at the boundary [ the number of boundary reflection matrices @xmath193 ) ] . the scattering of solitons or antisolitons with rapidities @xmath194 and @xmath195 is encoded in the scattering matrix @xcite @xmath196 . the boundary reflection matrix @xcite @xmath197 similarly describes the reflection at the boundary , @xmath198 , where @xmath199 for @xmath200 . for dirichlet boundary conditions one has @xcite @xmath201 . furthermore , @xmath202 , where @xmath203 and @xmath204 are the hamiltonian and the total momentum of the sine - gordon model on the infinite line . the operator @xmath205 ( @xmath185 ) creates ( annihilates ) two antisolitons , i.e. increases ( decreases ) the u(1 ) charge by 2 . the necessary form factors were derived by lukyanov and zamolodchikov @xcite . in our conventions they are given by @xmath206 where the normalization constant @xmath74 and the auxiliary function @xmath71 are stated in app . [ app : ff ] . the form factors satisfy the form - factor axioms as stated in ref . with lorentz spin @xmath207 and semi - locality factor @xmath208 . in the following we will evaluate the terms @xmath209 , @xmath210 , and @xmath211 which result in @xmath212 , @xmath213 , and @xmath214 , respectively . the first term to be evaluated . it does not contain any information about the boundary in the charge sector , i.e. it contains no boundary reflection matrices . a similar term was calculated in ref . in the derivation of the spectral function in the translationally invariant system . after shifting the contour of integration , @xmath216 , we directly obtain @xmath217 where the center - of - mass coordinates are defined by @xmath218 and @xmath219 . analytical continuation @xmath220 to real times , taking the limit @xmath221 , and fourier transformation with respect to @xmath112 and @xmath94 yield the first term @xmath222 in the form - factor expansion of the ldos . we start with @xmath210 defined in . as the operator @xmath205 creates two antisolitons we deduce @xmath224 ; @xmath225 yields @xmath226 . the matrix element of @xmath185 contains incoming and outgoing particles and thus possesses kinematical poles . we deal with these terms following smirnov @xcite and analytically continue the form factor as ( see ref . for a detailed discussion of this procedure ) @xmath227 here we have already omitted the positive imaginary parts of the rapidities in the form factors in the 2nd to 5th lines as they do not possess kinematical poles . inserting into @xmath210 yields to different contributions : the first line leads to a term containing three integrations over rapidities and is thus a sub - leading correction ; we will not evaluate it here . the 2nd to 5th lines yield using the boundary cross - unitarity condition @xcite @xmath228 as well as unitarity @xcite @xmath229 @xmath230 in order to proceed let us first assume @xmath231 . this implies that the boundary reflection matrix does not depend on the u(1 ) charge , i.e. @xmath232 , where @xmath70 is explicitly stated in app . [ app : ff ] . in particular , @xmath233 is analytic in the physical strip @xmath234 . now we shift the contours of integration @xmath235 ( we note that the contribution originating in the first term of does not possess any poles in the physical strip @xmath236 ) and apply the crossing relation in the second form factor , @xmath237 where the phase factor @xmath238 is due to the semi - locality @xcite of the operator @xmath239 with respect to the fundamental fields creating the excitations . finally we use to obtain @xmath240 fourier transformation directly yields the second term @xmath241 of the ldos . on the other hand , if @xmath242 , the boundary reflection matrix @xmath243 possesses a pole at @xmath244 . at the lep the residue is given by @xmath245 ( see ) . thus when shifting the contours of integration in we obtain the additional bbs contribution @xmath246 fourier transformation yields @xmath247 . finally we calculate the leading term involving the reflection of two antisolitons , which originates from @xmath211 . the analytic continuation of the form factor of @xmath185 reads @xmath249 where the scalar products are evaluated using the faddeev - zamolodchikov algebra @xcite @xmath250 . \end{split}\ ] ] therefore the 2nd to 5th line in yield ( we also use the scattering axiom for the first form factor ) @xmath251 let us again first assume @xmath231 . if we shift the contours of integration @xmath252 and use the crossing relation as well as a lorentz transformation , @xmath253 we obtain @xmath254 @xmath255 is now readily obtained via fourier transformation . starting from the second term @xmath256 related to the bbs is obtained by picking up the poles of @xmath257 and @xmath258 at @xmath259 respectively . at the lep the result is @xmath260 in this appendix we collect some formulas on the sine - gordon theory used in the present article . the scattering matrix between solitons and antisolitons possesses the integral representation @xcite @xmath261,\ ] ] where @xmath73 . a similar integral representation for the boundary reflection matrix reads @xcite ( for dirichlet boundary conditions with @xmath262 ) @xmath263,\\ \sigma(\theta)\!\!\!&=&\!\!\!\exp\!\left[2{\text{i}}\int_0^\infty\frac{dt}{t}\ , \sin\frac{\theta t}{\pi\xi}\ , \frac{\sinh\bigl(\frac{t}{\xi}(1+{\text{i}}\frac{\theta}{\pi})\bigr ) } { \sinh t\,\cosh\frac{t}{\xi}}\right].\end{aligned}\ ] ] the auxiliary function @xmath71 appearing in the form factors is given by @xcite @xmath264,\\ \mathcal{c}_1&=&\exp\left[-\int_0^\infty\frac{dt}{t}\ , \frac{\sinh^2\frac{t}{2}\,\sinh\bigl(t(\xi-1)\bigr ) } { \sinh(2t)\,\cosh t\,\sinh(t\xi)}\right],\end{aligned}\ ] ] while for the normalization factor @xmath74 one finds numerically from the integral repesentation derived in ref . @xmath265 the expressions at the lep are obtained by setting @xmath266 ( @xmath36 ) , i.e. @xmath267 , @xmath268 , @xmath269 , and @xmath270 . j. e. hoffman , k. mcelroy , d .- h . lee , k. m. lang , h. eisaki , s. uchida , and j. c. davis , science * 297 * , 1148 ( 2002 ) ; k. mcelroy , r. w. simmonds , j. e. hoffman , d .- h . lee , j. orenstein , h. eisaki , s. uchida , and j. c. davis , nature * 422 * , 592 ( 2003 ) ; c. howald , h. eisaki , n. kaneko , m. greven , and a. kapitulnik , phys . b * 67 * , 014533 ( 2003 ) ; m. vershinin , s. misra , s. ono , y. abe , y. ando , and a. yazdani , science * 303 * , 1995 ( 2004 ) . m. fabrizio and a. o. gogolin , phys . b * 51 * , 17827 ( 1995 ) ; s. eggert , h. johannesson , and a. mattsson , phys . rev . lett . * 76 * , 1505 ( 1996 ) ; s. eggert , _ ibid . _ * 84 * , 4413 ( 2000 ) ; m. guigou , t. martin , and a. crpieux , phys . rev . b * 80 * , 045420 ( 2009 ) . f. mila and x. zotos , europhys . * 24 * , 133 ( 1993 ) ; k. penc and f. mila , phys . b * 49 * , 9670 ( 1994 ) ; k. sano and y. no , j. phys . soc . jpn . * 63 * , 1250 ( 1994 ) ; phys . rev . b * 70 * , 155102 ( 2004 ) ; s. ejima , f. gebhard , and s. nishimoto , europhys . lett . * 70 * , 492 ( 2005 ) ; k. sano and y. no , phys . rev . b * 75 * , 113103 ( 2007 ) . d. controzzi , f. h. l. essler , and a. m. tsvelik , phys . 86 * , 680 ( 2001 ) ; f. h. l. essler , f. gebhard , and e. jeckelmann , phys . b * 64 * , 125119 ( 2001 ) ; s. andergassen , t. enss , v. meden , w. metzner , u. schollwck , and k. schnhammer , _ ibid . _ * 73 * , 045125 ( 2006 ) ; t. shirakawa and e. jeckelmann , _ ibid . _ * 79 * , 195121 ( 2009 ) . a. schwartz , m. dressel , g. grner , v. vescoli , l. degiorgi , and t. giamarchi , phys . rev . b * 58 * , 1261 ( 1998 ) ; m. dressel , k. petukhov , b. salameh , p. zornoza , and t. giamarchi , _ ibid . _ * 71 * , 075104 ( 2005 ) . b. korin - hamzi , e. tafra , m. basleti , a. hamzi , and m. dressel , _ ibid . _ * 73 * , 115102 ( 2006 ) . f. h. l. essler and r. m. konik , in _ from fields to strings : circumnavigating theoretical physics ( ian kogan memorial collection ) _ , edited by m. shifman , a. vainshtein , and j. wheater ( world scientific , singapore , 2005 ) , vol . i. j. voit , eur . j. b * 5 * , 505 ( 1998 ) ; f. h. l. essler and a. m. tsvelik , phys . rev . b * 65 * , 115117 ( 2002 ) ; f. h. l. essler and a. m. tsvelik , phys . * 90 * , 126401 ( 2003 ) ; h. benthien , f. gebhard , and e. jeckelmann , _ ibid . _ * 92 * , 256401 ( 2004 ) ; t. ulbricht and p. schmitteckert , europhys . * 89 * , 47001 ( 2010 ) . p. abbamonte , g. blumberg , a. rusydi , a. gozar , p. g. evans , t. siegrist , l. venema , h. eisaki , e. d. isaacs , and g. a. sawatzky , nature * 431 * , 1078 ( 2004 ) ; s. notbohm , p. ribeiro , b. lake , d. a. tennant , k. p. schmidt , g. s. uhrig , c. hess , r. klingeler , g. behr , b. bchner , m. reehuis , r. i. bewley , c. d. frost , p. manuel , and r. s. eccleston , phys . * 98 * , 027403 ( 2007 ) .

the @xmath0he three - atomic system is of considerable interest in various fields of physical chemistry and molecular physics . studies of the helium dimer and trimer represent an important step towards understanding the properties of helium liquid drops , superfluidity in @xmath0he films , and so on ( see , for instance , refs . besides , the helium trimer is probably a unique system where a direct manifestation of the efimov effect @xcite can be observed since the binding energy @xmath2 of the @xmath0he dimer is extremely small ( @xmath3mk @xcite ) even in the molecular scale . for this reason , the helium trimer is certainly of interest for nuclear physicists , too . moreover a theoretical study of the @xmath0he trimer is based just on the same methods of the theory of few - body systems that are used in solving three body nuclear problems . from the standpoint of the general theory of few body systems , the @xmath0he trimer belongs to three body systems that are most difficult for a specific investigation , first , owing to its efimov nature , and second , because it is necessary to take into account the practically hard core in the interatomic he he - interaction @xcite . at the same time the problem of three helium atoms can be considered as an example of an ideal three body quantum problem since the @xmath0he atoms are identical neutral bosons with zero spin and the analysis of this problem is complicated neither by separation of spin isospin variables nor by taking into account the coulomb interaction . there is a great number of experimental and theoretical studies of @xmath0he clusters . however , most of the theoretical investigations consist merely in computing the ground - state energies of clusters of that sort , mainly on the basis of variational methods @xcite . besides , the methods based on hyperspherical expansions of the schrdinger and faddeev equations @xcite in the coordinate representation were used . also , the faddeev integral equations in the momentum representation were employed in refs . @xcite while the results of ref . @xcite are based on a direct solving the two - dimensional faddeev differential equations in configuration space . from the experimental studies we would like to mention those of refs . @xcite where clusters consisting of a small number of noble gase atoms were investigated . though much effort was undertaken for studying molecular clusters various problems related to the @xmath0he trimer remained beyond the scope of thorough consideration . in particular , the elastic scattering phases of a helium atom on a helium dimer and breakup amplitudes ( at ultralow energies ) have been calculated only recently @xcite . these computations were preceded only by the computation of characteristics of the he he@xmath4 scattering at zero energy @xcite and estimation of the recombination rate @xmath5 @xcite . as a matter of fact , we have already pointed out basic reasons for computations of excited states and scattering being especially difficult in the @xmath0he@xmath6 system . first , this is a low energy of the dimer @xmath2 which necessitates to consider very large domains in the configuration space with a characteristic size of hundreds of . second , a very strong repulsive component in the he he interaction produces large errors in the standard approximation of the three atomic hamiltonian at short distances between atoms . the capacities of modern computers do not yet allow one to reach dimensions of grids that would remove both the above - mentioned reasons and would provide stable results with the use of the conventional methods . the present paper is a sequel of studies of the @xmath0he@xmath6 system undertaken in the papers @xcite within an approach that is capable , as we think , to resolve both the above - mentioned numerical problems . in these papers the repulsive component of the he he interaction at short distances between atons is approximated by a hard core . this allows one to investigate the @xmath0he@xmath6 system within a mathematically rigorous method of solving a three - body problem in the boundary - condition model developed in @xcite . an important advantage of such an approach that essentially diminishes computational errors is the necessity to approximate , inside the core domains , only the laplacian operator instead of the sum of this operator and a huge repulsive components of the he he - potentials ( see @xcite ) . in @xcite , such an approach has been successfully applied for calculating not only scattering but also binding energies of the ground and excited states of the helium trimer . investigation made in @xcite has shown that the method proposed in @xcite is well suited for performing three body molecular computations in the case where repulsive components of interatomic interactions are of a hard core nature . there is a series of works @xcite showing that the excited state of the @xmath0he trimer is initiated indeed by the efimov effect @xcite . in these works the various versions of the aziz @xmath0he@xmath0he potential were employed ( hfdhe2 @xcite , hfd - b @xcite , and lm2m2 @xcite ) . however , the basic result of refs . @xcite on the excited state of the helium trimer is the same : this state disappears when the interatomic potential is multiplied by the amplification factor " @xmath7 of order 1.2 . more precisely , if this potential is multiplied by the increasing factor @xmath8 then the following effect is observed . first , the difference @xmath9 between the dimer energy @xmath10 and the energy of the trimer excited state @xmath11 increases . then the behavior of this difference radically changes and with further increase of @xmath7 it monotonously decreases . at @xmath12 the level @xmath13 disappears . it is just such a nonstandard behavior of the energy @xmath11 as the coupling between helium atoms becomes more and more strengthening , points to the efimov nature of the trimer excited state . and vice versa , when @xmath7 slightly decreases ( no more than 2% ) , the second excited state @xmath14 appears in the trimer @xcite . this paper is aimed at elucidating the fate of the trimer excited state upon its disappearance in the physical sheet when @xmath8 and at studying the mechanism of arising of new excited states when @xmath15 . as the interatomic he he potential , we use the potential hfd - b @xcite . we have established that for such he he - interactions the trimer excited level @xmath13 merges with the threshold @xmath2 at @xmath16 and with further decreasing @xmath7 it transforms into a virtual level of the first order ( a simple real pole of the analytic continuation of the scattering matrix ) lying in the unphysical energy sheet adjoining the physical sheet along the spectral interval between @xmath2 and the three body threshold . we trace the position of this level for @xmath7 increasing up to 1.5 . besides , we have found that the excited ( efimov ) levels for @xmath15 also originate from virtual levels of the first order that are formed in pairs . before a pair of virtual levels appears , there occurs a fusion of a pair of conjugate resonances of the first order ( simple complex poles of the analytic continuation of the scattering matrix in the unphysical sheet ) resulting in the virtual level of the second order . as it will be clear from the further exposition ( see sect . [ results ] ) , the above - mentioned resonances are not , generally speaking , genuine resonances of the @xmath0he@xmath6 trimer since they are situated outside of the energy domain for which we can rigorously prove the applicability of the method we are using for computing the resonances . we will call the resonances found outside the range of guaranteed applicability of the method the _ ( quasi)resonances_. the paper is organized as follows . in sect . [ difffs ] , we describe the method of search for resonances in a three body system on the basis of the faddeev differential equations . the idea of the method consists in calculating the analytic continuation of the component @xmath17 ( see formula ( [ s0 ] ) ) of the scattering matrix corresponding to the ( @xmath18 ) process , in the physical sheet with the use of these equations . a particular attention in this section is paid to the description of the parabolic domain on the physical sheet where one can analytically continue the function @xmath19 by numerical solving the coordinate space faddeev partial equations . for the potentials we use , the three body resonances ( including virtual levels ) lying in the unphysical sheet of energy @xmath20 plane adjoining the physical sheet along the interval @xmath21 are the roots of the function @xmath19 in the physical sheet . we have earlier employed this method for computing resonances as roots of @xmath17 in the three nucleon problem @xcite . in sect . [ results ] , we first briefly describe the numerical method we use to solve the @xmath1 scattering problem for the @xmath0he@xmath6 system with going out into the domain of complex energies . then we describe the results of our calculations . some notation used throughout the paper is as follows : by @xmath22 we denote the complex plane ; @xmath23 stands for the main branch of the function @xmath24 , @xmath25 for any @xmath26 ; the symbol @xmath27 is used for the quadrant @xmath28 , @xmath29 ; by @xmath30 we understand the hilbert space of complex valued functions which are integrable on @xmath27 with the absolute value squared ; the symbol @xmath31 stands for the complex number conjugated to @xmath20 . in this paper , we will consider the @xmath0he@xmath6 system in the state with the total angular momentum @xmath32 . first we consider the case where the interatomic interactions are described by conventional smooth potentials that include no hard - core component . in this case , the angular partial analysis reduces the initial faddeev equation for three identical bosons to a system of coupled two - dimensional integro - differential equations @xcite @xmath33f_l(x , y)=-v(x)\psi_l(x , y)\,.\ ] ] here , @xmath34 stand for the standard jacobi variables , @xmath28 and @xmath29 , and @xmath35 for the partial component of the kinetic energy operator . functions from the domain of @xmath36 are assumed to obey the boundary conditions @xmath37 which are quite standard when the expansions over bispherical basis are used . the potential @xmath38 is assumed to be central . in our paper , the energy @xmath20 can get both real and complex values . at @xmath32 the partial angular momentum @xmath39 corresponds both to the dimer and an additional atom . the momentum @xmath39 can assume only even values , @xmath40 the partial wave functions @xmath41 are expressed through the faddeev partial components @xmath42 by the relations @xmath43 where @xmath44 and @xmath45 . the explicit form of the functions @xmath46 can be found , e.g. , in refs . @xcite ( see also @xcite ) . here we only deal with a finite number of equations ( [ fadpart ] ) , assuming that @xmath47 where @xmath48 is a certain fixed even number , @xmath49 . the condition @xmath50 is equivalent to the supposition that the potential @xmath38 only acts in the two - body states with @xmath51 . the spectrum of the schrdinger operator for a system of three identical bosons with such a potential is denoted by @xmath52 . its is well known ( see , e.g. , ref . @xcite ) that if the potential @xmath38 is smooth and decreasing as @xmath53 together with its derivatives not slower than @xmath54 , @xmath55 , then the asymptotic conditions as @xmath56 and/or @xmath57 for the partial faddeev components of the @xmath58 scattering wave functions '' . the asymptotics of these functions in the total three - body configuration space @xmath59 contains , apart from the incident wave , only the so - called outgoing spherical waves ( see , e.g. , @xcite ) . ] for @xmath60 , @xmath61 , read@xmath62\right\ } \\ & & + \displaystyle\frac{\exp({\rm i}\sqrt{z}\rho)}{\sqrt{\rho } } \left[a_l(z,\theta)+o\left(1\right)\right]\ , . \end{array}\ ] ] we assume that the @xmath0he@xmath4 dimer has an only bound state with an energy @xmath2 , @xmath63 , and wave function @xmath64 . this function is assumed to be normalized so that for all @xmath65 values of @xmath64 are real . the notations @xmath66 , @xmath67 , and @xmath68 , @xmath69 , are used for the hyperradius and hyperangle . the coefficient @xmath70 , @xmath60 , for @xmath71 is the elastic scattering amplitude . the functions @xmath72 provide us , at @xmath61 , the corresponding partial faddeev breakup amplitudes . note that for @xmath60 the correction terms @xmath73 in coefficients of outgoing waves @xmath74 , @xmath71 , and @xmath75 , @xmath61 , in ( [ asbcparts ] ) are of the form , respectively , @xmath76 and @xmath77 . this property ensures uniqueness of the solution of the boundary value problem ( [ fadpart][asbcparts ] ) for real scattering energies @xmath71 @xcite . the component of the @xmath78-wave partial scattering matrix for a system of three helium atoms is given for real @xmath60 , @xmath71 , by the expression @xmath79 while the scattering phases read @xmath80 where @xmath81 stands for the momentum conjugated to the jacobi variable @xmath82 . our goal is to study the analytic continuation of the scattering matrix @xmath19 into the complex plane ( the physical sheet ) . as it follows from the results of refs . @xcite , roots of the function @xmath17 in the physical sheet of energy @xmath20 plane correspond to the location of the three - body resonances situated in the unphysical sheet connected with the physical sheet by crossing the spectral interval @xmath21 . this statement is a particular case of more general statements regarding the three - body resonances obtained in @xcite for the case of two - body potentials decreasing in the coordinate space not slower than exponentially . we assume that @xmath38 is just a potential which falls off exponentially and , thus , for all @xmath28 @xmath84 with some positive @xmath85 and @xmath86 . for the sake of simplicity we even assume sometimes that @xmath38 is finite , i.e. , for @xmath87 , @xmath88 . looking ahead , we note that , in fact , in our numerical computations of the @xmath0he@xmath6 system at complex energies we make a cutoff " of the interatomic he he - potential at a sufficiently large radius @xmath89 . it is well known that different representations of the same holomorphic function ( for instance , either by a series or by an integral ) allow one to describe this function only in some parts of its riemann surface . the description @xcite of the holomorphy domains for different truncations of the total three - body scattering matrix in the physical sheet was based on the use of the faddeev integral equations in the momentum representation . in this paper , we make use of the faddeev equations in the configuration space . therefore it is necessary to perform an investigation , independent of @xcite , of domains in the physical sheet where we can analytically continue the faddeev components @xmath90 and the amplitudes @xmath91 and @xmath92 just with the use of the configuration space techniques . let us list briefly the main results of this investigation obtained by us for the @xmath0he@xmath6 system under the assumption ( [ vexp ] ) . to formulate these results we distinguish the following three domains in the complex plane @xmath93 . the domain @xmath95 where the faddeev components @xmath90 ( and , hence , the wave functions @xmath96 ) can be analytically continued in @xmath20 so that the differences @xmath97 at @xmath98 turn out to be elements of @xmath99 . the domain @xmath95 is described by the inequality @xmath100 for fixed @xmath34 the functions @xmath101 are continuous in @xmath20 up to the rims of the cut along the continuous spectrum @xmath102 . the domain @xmath103 where both the elastic scattering amplitude @xmath70 and the faddeev breakup amplitudes @xmath92 can be analytically continued in @xmath20 , @xmath104 , provided that the functions @xmath90 obey the asymptotic formulas ( [ asbcparts ] ) . this domain is described by the inequalities @xmath105 3@xmath94 . and finally , we distinguish the domain @xmath106 , most interesting for us , where the analytic continuation in @xmath20 , @xmath104 , can be only done for the elastic scattering amplitude @xmath70 ( and consequently , for the scattering matrix @xmath19 ) ; the analytic continuabilty of the amplitudes @xmath92 in the whole domain @xmath106 is not required . the set @xmath106 is a geometric locus of points obeying the inequality @xmath107 for the domains @xmath95 , @xmath103 , and @xmath106 , the following chain of inclusions @xmath108 is valid . note that the type ( [ pipsi ] ) or ( [ pis ] ) condition , @xmath109 is equivalent to the inequality @xmath110 therefore , for @xmath111 the set @xmath95 is the domain bounded by the parabola @xmath112 for @xmath113 this set coincides with the domain @xmath114 analogously , if @xmath115 then the domain @xmath106 is described by the inequality ( [ parabmu ] ) ; whereas for @xmath116 by the inequality @xmath117 as to the curves bounding the domains ( [ piamain ] ) and ( [ piamu ] ) , we only notice that their order with respect to the variables @xmath118 and @xmath119 is higher than the second order . it is easy to check that each of these curves is connected , symmetric with respect to the @xmath118 axis and crosses the latter only once . for the first curve this intersection occurs at , the slope angle of the tangent at the point of intersection being independent of @xmath2 , . as @xmath120 , the boundaries ( [ piamain ] ) and ( [ piamu ] ) are asymptotically approximated by the type ( [ parabola ] ) parabolas with coefficients @xmath121 and @xmath122 which can be computed explicitly . to prove the assertion1@xmath94 concerning the domain @xmath95 we note that the functions @xmath101 given by the formulas ( [ flphil ] ) satisfy the equations @xmath123\phi_l(x , y;z ) + v(x)\sum_{l'}\int_{-1}^{+1 } d\eta\,h_{l l'}(x , y,\eta)\,\phi_{l'}(x',y',z)= \chi_l(x , y;z)\,\ ] ] where @xmath124 obviously , for @xmath125 the functions @xmath126 fall off exponentially as @xmath127 . moreover , for all the directions @xmath128 the uniform estimate @xmath129 is valid with @xmath130 consequently , if the condition ( [ pipsi ] ) holds then the inhomogeneous terms @xmath126 considered as functions of the variables @xmath131 and @xmath82 at fixed @xmath20 , are elements of @xmath132 . at the same time , the vectors @xmath133 turn out to be holomorphic functions of @xmath125 with respect to the @xmath30 norm . in the problem under consideration , the spectrum of the faddeev matrix operator defined by the l.h.s . of eqs . ( [ faddxi ] ) and by the boundary conditions ( [ bcstandard ] ) in the hilbert space constituted of the vectors , @xmath134 , coincides with the spectrum @xmath52 of the corresponding three - boson schrdinger operator with two - body potentials @xmath38 only acting in the states with @xmath51 . this means that for any energy @xmath125 lying outside of the spectrum @xmath52 , the inhomogeneous system ( [ faddxi ] ) is uniquely solvable in the class of the functions @xmath135 , @xmath51 . since outside of the set @xmath52 the resolvent of the faddeev operator is a holomorphic operator - valued function of the variable @xmath20 , each of the components @xmath136 of the solution of eqs . ( [ faddxi ] ) also is a holomorphic function of @xmath98 . the bound - state energies of the three - boson system under consideration turn out to be poles of the first order for @xmath136 . thus , the faddeev partial components @xmath90 admit the analytic continuation in @xmath20 in the form ( [ flphil ] ) into the domain @xmath137 . the proof of the assertions of 2@xmath94 and 3@xmath94 regarding the domains @xmath106 and @xmath103 is rather cumbersome . this is why we here only outline its main steps . note that the proof is based on the integral equations method and it is quite standard ( see , for instance , ref . @xcite , chapter v ) . first , the equations ( [ faddxi ] ) are rewritten in the form of the faddeev partial integral equations . to do this , it suffices to reverse the operators @xmath138 in ( [ faddxi ] ) . since the variables @xmath131 and @xmath82 in @xmath36 are separated , the kernels @xmath139 , @xmath140 , @xmath141 , of the respective resolvents @xmath142 are explicitly expressed in terms of the two - body problem . analytic properties in the variable @xmath20 and coordinate asymptotics of the kernels @xmath139 are well known ( see ref . @xcite , chapters iv and v ) . iterations first improve " and then stabilize the asymptotic properties of the iterated kernels and inhomogeneous terms of the faddeev equations . ( in the case under consideration , this stabilization requires only three iterations . ) further , it turns out that , for @xmath143 , the iterated kernels are represented by sums of exponentially decreasing terms admitting , in certain domains of the configuration space , an explicit asymptotic factorization with respect to @xmath144 and @xmath145 . since we are working in the domain where @xmath146 , the corresponding asymptotic factors of these terms , along with the asymptotics of the iterations of the inhomogeneous term , determine the coordinate asymptotics of the functions @xmath101 . therefore , finally we are able to determine the geometric locus of the points @xmath20 in the complex plane for which there exists a ( non - empty ) set in the configuration space such that the leading term of the coordinate asymptotics of the function @xmath101 in this set represents a term of the form @xmath147 , and thus , for these @xmath20 the scattering matrix @xmath19 is well defined . this geometrical locus is just the domain @xmath106 . in this domain , as @xmath148 and/or @xmath127 , the functions @xmath101 admit the asymptotic representation @xmath149+{\rm f}_0(y;z)\right\ } \\ \nonumber & + & \displaystyle\frac{\exp({\rm i}\sqrt{z}\rho)}{\sqrt{\rho } } \left[a_l(z,\theta)+o(1)\right ] + { \rm f}_{1,l}(x , y;z ) \ ] ] with @xmath150 where @xmath151 in a parabolic neighborhood of the @xmath82-axis , the functions @xmath152 are also subjected to the asymptotic estimates @xmath153 where @xmath154 is an arbitrary fixed number smaller than unity , @xmath155 . as to the domain @xmath103 , the leading asymptotic term of each of the functions @xmath152 for @xmath156 is a spherical wave @xmath75 with the amplitude @xmath157 being a differentiable function of the angle @xmath68 . therefore , for @xmath158 the term @xmath152 in the r.h.s . of the formula ( [ phipis ] ) can be added to the asymptotic term with a spherical wave preceding @xmath159 . in the domain @xmath106 , and hence , in a narrower domain @xmath103 the condition @xmath160 holds . consequently , for @xmath156 , the faddeev components @xmath90 do obey the standard asymptotic conditions like ( [ asbcparts ] ) . therefore , for any @xmath155 the dominant term of the asymptotics of the function @xmath161 , @xmath162 , in the domain @xmath163 reads as @xmath147 as @xmath148 . this means that , for @xmath162 , it is always possible by solving the equations ( [ fadpart ] ) to separate explicitly the elastic scattering amplitude @xmath70 and , thus , to construct the analytic continuation of the scattering matrix @xmath19 . outside of the domain @xmath106 the numerical construction of @xmath19 by solving the faddeev differential equations is , in general , impossible since for @xmath163 and @xmath155 both functions @xmath164 and @xmath165 , @xmath166 , include terms decreasing slower than @xmath74 as @xmath148 . in the case of potentials with hard core , the partial faddeev differential equations for a system of three identical bosons at @xmath32 acquire the form @xmath167f_l(x , y)=\left\ { \begin{array}{cl } -v(x)\psi_l(x , y ) , & x > c \\ 0 , & x < c\ , , \end{array}\right.\ ] ] where @xmath168 , @xmath169 , is the core size . the partial wave functions @xmath41 are expressed via faddeev partial components @xmath42 by the formulas ( [ ftconn ] ) . the components @xmath42 satisfy the standard boundary conditions ( [ bcstandard ] ) . the two - body central potential @xmath38 acts only beyond the core domain , i.e. only where @xmath170 . we assume as before that @xmath38 falls off not slower than exponentially as and , hence , it satisfies the condition ( [ vexp ] ) for some @xmath171 and @xmath172 . a main difference between the model with hard core and those with smooth potentials is that the functions @xmath42 in this model satisfy the auxiliary boundary conditions @xmath173 requiring that the wave functions @xmath174 vanish on the boundary @xmath175 of the core domain . it can be shown that in fact the conditions ( [ bccorepart ] ) force the wave functions ( [ ftconn ] ) to vanish also inside the core domain at all energies except for a certain countable set of real values of @xmath20 ( see ref . @xcite and references therein ) . asymptotic conditions for the partial faddeev components @xmath90 of the @xmath58 scattering wave functions as @xmath56 and/or @xmath57 are again of the form ( [ asbcparts ] ) . the only difference is that the dimer wave function @xmath64 is considered as zero in the core domain , i.e. for @xmath176 . in the hard - core model , all the assertions of sect . [ holomorphydomains ] regarding the holomorphy domains of the functions @xmath101 and the scattering matrix @xmath19 in @xmath20 still hold true . we have already noticed that the roots of @xmath19 in the physical sheet of energy @xmath20 plane correspond to the location of the three - body resonances in the unphysical sheet adjoining the physical sheet along the spectral interval @xmath21 . in the case under consideration , this statement is an immediate consequence of the unitarity of the scattering matrix @xmath17 for @xmath60 , @xmath177 , @xmath178 indeed , as we have established , the functions @xmath101 are holomorphic functions of @xmath98 . since the boundary value problem ( [ fadpart][asbcparts ] ) is uniquely solvable , one easily verifies that the boundary values and for these functions on the rims of the cut along are related to each other as @xmath179 since , on the one hand , their asymptotics ( [ phipis ] ) as @xmath148 and/or @xmath127 has the same structure and , on the other hand , @xmath180 since @xmath181 consequently , @xmath182 and @xmath183 therefore , it follows from eq . ( [ s0unitar ] ) that for @xmath177 @xmath184^{-1 } \qquad\mbox{and}\qquad { \rm s}_0(e-{\rm i}0)=[{\rm s}_0(e+{\rm i}0)]^{-1}\,.\ ] ] this means that the function @xmath185 is continued through the cut @xmath186 $ ] into the domain @xmath187 as @xmath188 . in a similar manner , @xmath189 is continued into the domain @xmath190 , again as @xmath188 . all this signifies that the scattering matrix @xmath19 admits analytic continuation at least into the domain @xmath106 of the unphysical energy sheet connected with the physical sheet by crossing the interval @xmath186 $ ] , the value of the continued function @xmath17 at @xmath158 in the unphysical sheet coinciding with the value of @xmath188 at the same @xmath20 but in the physical sheet . recall that those points @xmath20 on unphysical sheets are called resonances where the analytically continued scattering matrix possesses poles . the resonances @xmath20 with zero imaginary part @xmath191 and @xmath192 are called the virtual levels . thus , we have here presented a simple proof of the fact that the resonances including the virtual levels corresponding to poles of the analytic continuation of the scattering matrix @xmath17 in the unphysical sheet connected with the physical one by crossing the spectral interval @xmath186 $ ] are the roots of this matrix in the physical sheet . at the same time , the poles of the function @xmath19 in the physical sheet correspond to bound states of the three - boson system under consideration . concluding the subsection , we note that it follows from eq . ( [ aadj ] ) that and , hence , @xmath193 for any @xmath158 . this means that the roots of the function @xmath19 are situated symmetrically with respect to the real axis . in the present work we make use of the faddeev equations ( [ fadpartcor ] ) considered together with the boundary conditions ( [ bcstandard ] ) , ( [ asbcparts ] ) and ( [ bccorepart ] ) to calculate the values of the @xmath0he@xmath6 scattering matrix @xmath19 in the physical sheet . we search for the resonances including the virtual levels as roots of @xmath19 and for the bound - state energies as positions of poles of @xmath19 . all the results presented below are obtained for the case @xmath194 . in all our calculations , as the interatomic he he - interaction we employed the widely used semiempirical potential hfd - b constructed by r.a.aziz and co - workers @xcite . this potential is of the form @xmath195f(\zeta ) \right \}\,\ ] ] where @xmath196 . the function @xmath197 reads @xmath198 ^ 2 } , & \mbox{if $ \zeta\leq d$ } \cr 1 , & \mbox{if $ \zeta > d $ } \ , . } \ ] ] for completeness the parameters of the potential hfd - b are given in table i. the value of the parameter @xmath168 ( the core diameter " of particles ) is chosen to be so small that its further decrease does not appreciably influence the dimer binding energy @xmath2 and the energy of the trimer ground state @xmath199 . unlike papers @xcite , where @xmath168 was taken to be equal 0.7 , now we take @xmath200 . we have found that such a value of @xmath168 provides at least six reliable figures of @xmath2 and three figures of @xmath199 . since the statements of sect . [ difffs ] are valid , generally speaking , only for the potentials decreasing not slower than exponentially , we cut off the potential hfd - b setting @xmath201 for @xmath87 . we have established that this cutoff for provides the same values of @xmath2 ( @xmath202mk ) , @xmath199 ( @xmath203k ) and phases @xmath204 which were obtained in our earlier calculations @xcite performed with the potential hfd - b . comparison of these results with results of other researchers can be found in refs . in all the calculations of the present work we take . note that if the formulas from sect . [ difffs ] including the parameter @xmath86 are used for finite potentials , one should set @xmath205 . before making numerical approximation of the system of equations ( [ bcstandard ] ) , ( [ fadpartcor ] ) , ( [ bccorepart ] ) at @xmath194 we rewrite it in terms of a new unknown function @xmath161 that is expressed via the faddeev component @xmath206 by the relation ( [ flphil ] ) . note that for @xmath98 the function @xmath161 is square integrable in @xmath34 ( see sect . [ holomorphydomains ] ) . therefore , this function is uniquely determined by the asymptotic condition @xmath207 that can be easily approximated and programmed . one could , for instance , require @xmath208 at a sufficiently large @xmath209 and look for a numerical solution of the system ( [ bcstandard ] ) , ( [ fadpartcor ] ) , ( [ bccorepart ] ) satisfying this condition . further , for @xmath158 , one could , going sufficiently far from @xmath209 into the domain of smaller ( but nevertheless , providing the asymptotics ( [ phipis ] ) ) values of @xmath66 , separate the elastic scattering amplitude @xmath70 , putting , e.g. , @xmath210 , where the value of @xmath131 corresponds to the maximum of the function @xmath64 . such an approach is , however , not effective in view of a relatively slow decrease of the exponentials @xmath211 and @xmath212 as well as of the function @xmath213 in the energy domain of interest for us in @xmath106 . for a proper approximation of the condition ( [ phias0 ] ) , very large values of @xmath209 are to be taken . this is just a reason why one should take into account the asymptotics of the function @xmath161 as @xmath53 and/or @xmath148 . though the asymptotic formula ( [ asbcparts ] ) only holds for @xmath156 , we employ it also for @xmath214 . indeed , when @xmath214 , the leading term of the asymptotics of @xmath161 as @xmath148 and @xmath163 , @xmath155 , is given by the same expression ( see sect . [ holomorphydomains ] ) as in eq . ( [ asbcparts ] ) . outside of the parabola @xmath163 , it suffices to require the condition ( [ phias0 ] ) to be satisfied . the presence , in eq . ( [ asbcparts ] ) , of the spherical wave does not contradict this requirement . therefore , the use of asymptotic condition ( [ asbcparts ] ) is justified even if @xmath214 . when solving the boundary - value problem ( [ bcstandard ] ) , ( [ asbcparts ] ) , ( [ fadpartcor ] ) , ( [ bccorepart ] ) written in terms of the function @xmath215 , we carry out its finite - difference approximation in polar coordinates @xmath66 and @xmath68 . the grid is chosen in such a way that the points of intersection of arcs @xmath216 , @xmath217 , and rays @xmath218 , @xmath219 , with the line @xmath175 turn out automatically to be its knots . the @xmath220 points are chosen according to the formulas @xmath221 where @xmath222 stands for the number of arcs inside the core domain and @xmath223 the nonlinear monotonously increasing function @xmath224 , @xmath225 , satisfying the conditions @xmath226 and @xmath227 is chosen in the form @xmath228\\ \alpha_1\tau+\tau^\nu & , & \tau\in(\tau_0,1 ] \end{array } \right .. \ ] ] the values of @xmath229 , @xmath230 and @xmath231 , @xmath232 are determined via @xmath233 and @xmath154 from the continuity condition for @xmath224 and its derivative at the point @xmath233 . as a rule , we took values of @xmath233 within 0.1 and 0.2 . the value of the power @xmath154 depends on the cutoff radius @xmath234@xmath235 its range being within 2 and 4 in our calculations . the knots @xmath236 at @xmath237 are taken according to @xmath238 . the rest knots @xmath236 , @xmath239 are chosen equidistantly . such a choice of the grid is prescribed by the need to have a higher density of points in the domain where the functions @xmath101 are most rapidly changing , i.e. for small values of @xmath66 and/or @xmath131 and lower in the asymptotic domain . in this work , we used the grids of dimension . the number of the last arc knots in @xmath68 lying in the core domain was usually equal to @xmath240 . the finite - difference approximation of the integro - differential equations ( [ fadpartcor ] ) and boundary conditions ( [ bcstandard ] ) , ( [ bccorepart ] ) for @xmath194 reduces the problem to a system of @xmath241 linear algebraic equations . the finite - difference equations corresponding to the arc @xmath242 include initially the values of the unknown function @xmath243 from the arc @xmath244 . to eliminate them , we express these values through the values of @xmath243 on the arcs @xmath242 and @xmath245 by using the asymptotic formula ( [ asbcparts ] ) , just in the manner described in the concluding part of appendix a of ref . @xcite . in @xcite , this approach was only used for computing the energies of bound states . now we extend it also on the scattering problem . ( note that the formulas ( a10 ) and ( a11 ) in @xcite related to the described approach contain misprints . the values @xmath246 in these formulas should be replaced with inverse values @xmath247 . ) the matrix of the resultant system of equations has a block - three - diagonal form ( see ref . @xcite , appendix a ) . every block has the dimension @xmath248 and consists of the coefficients standing at unknown values of the function @xmath243 in the grid knots belonging to a certain arc @xmath216 . the main diagonal of the matrix consists of @xmath249 such blocks . in contrast to @xcite , in the present paper we solve the block - three - diagonal algebraic system on the basis of the matrix sweep method . this allows us to dispense with writing the system matrix on the hard drive and to carry out all the operations related to its inversion immediately in ram . besides , the matrix sweep method reduces almost by one order the computer time required for computations on the grids of the same dimensions as in @xcite . we searched for the resonances ( roots of the function @xmath17 on the physical sheet ) and bound - state energies ( roots of the function @xmath188 for real @xmath250 ) of the helium trimer by using the complex version of the secant method . within this method , the approximation @xmath251 to a root of a holomorphic function @xmath252 is constructed from the two previous approximations @xmath253 and @xmath254 according to the formula @xmath255 as the relationship ( [ s0sym ] ) implies the symmetry of properties of the scattering matrix @xmath19 with respect to the real axis , we performed all the calculations for @xmath19 only for @xmath256 ( except the tests of the code ) . we start with a study of graph surfaces of the real and imaginary parts of the scattering matrix @xmath19 in the domain of its holomorphy @xmath257 . the root lines of the functions @xmath258 and @xmath259 obtained in the case of the grid parameters and are depicted in fig . 1 . both resonances ( roots of @xmath19 ) and bound - state energies ( poles of @xmath19 ) of the @xmath0he trimer are associated with the intersection points of the curves @xmath260 and @xmath261 . when the roots or poles are simple , these curves intersect each other at the right angle . note that for real @xmath262 the function @xmath19 is real and , thus , @xmath261 . in fig . 1 , along with the root lines we also plot the boundaries of the domains @xmath106 , @xmath103 and @xmath95 . one can observe that a good " domain @xmath106 includes none of the points of intersection of the root lines @xmath260 and @xmath263 . nevertheless , as we will see below , the going beyond this domain is of an interest , even though the asymptotic formula ( [ asbcparts ] ) is not valid for @xmath264 and the function @xmath19 calculated there can not be interpreted as the scattering matrix . the caption for fig . 1 points out positions of the four resonances " , the roots of @xmath19 , found immediately beyond the boundary of the domain @xmath106 . as one could expect , the values of the function @xmath19 at @xmath264 and positions of its roots in @xmath265 turn out to be unstable and strongly depend on the value of the cutoff radius @xmath266 , whereas the dependence on the number of knots is weak . in particular , for @xmath267 , a ( quasi)resonance , closest to the real axis , is situated at the point if at the point if and at the point if the same ( quasi)resonance in fig . 1 ( calculated for ) is situated at the point @xmath268 mk . if @xmath269 is fixed , the increase of @xmath266 up to 800 shifts this point to the point @xmath270mk . all the aforesaid regarding the instability of the function @xmath19 values and positions of its roots beyond the domain @xmath106 bears no relation to its pole at the point , corresponding to a trimer excited - state energy , even though this energy does not belong to @xmath106 . the point is that the position of the pole of @xmath17 is only determined by the position of the root of the determinant of the linear algebraic system we solve , whereas the inhomogeneous term of the system plays no role . therefore , the search for the poles of the grid function @xmath19 is equivalent to the search for the binding energies of the trimer . the grids we have used turn out to be quite sufficient for this purpose . the convergence of our results for @xmath13 with respect to the parameters @xmath271 and their accuracy can be judged from the values of the difference @xmath272 obtained with different grids and shown in table ii . we would like to stress that we do not consider the roots of function @xmath19 drawn in fig . 1 as genuine resonances for the @xmath0he@xmath6 system since they are situated beyond the domain @xmath106 where the applicability of our method is proved . we should rather consider them as artifacts of the method . however it is remarkable that the true " ( i.e. , getting inside @xmath106 ) virtual levels and then the energies of the excited ( efimov ) states appear just due to these ( quasi)resonances when the potential @xmath38 is weakened . this is the object of our further consideration . following @xcite , instead of the initial potential @xmath273 , we will consider the potentials @xmath274 to establish the mechanism of formation of new excited states in the @xmath0he trimer , we have first calculated the scattering matrix @xmath19 for @xmath15 . in table iii for some values of @xmath7 from the interval between 0.995 and 0.975 , we present the positions of roots and poles of @xmath19 , we have obtained at real @xmath275 . we have found that for a value of @xmath7 slightly smaller than @xmath276 , the ( quasi)resonance closest to the real axis ( see fig . 1 ) gets on it and transforms into a virtual level ( the root of @xmath19 ) of the second order corresponding to the energy value where the graph of @xmath19 , @xmath277 , @xmath250 , is tangent to the axis @xmath20 . this virtual level is preceded by the ( quasi)resonances mk for @xmath278 and mk for @xmath279 . the originating virtual level is of the second order since simultaneously with the root of the function @xmath17 , also the conjugate root of this function gets on the real axis . with a subsequent decrease of @xmath7 the virtual level of the second order splits into a pair of the virtual levels @xmath280 and @xmath281 , @xmath282 of the first order which move in opposite directions . a characteristic behavior of the scattering matrix @xmath19 when resonances transform into virtual levels is shown in fig . the virtual level @xmath281 moves towards the threshold @xmath2 and `` collides '' with it at @xmath283 . for @xmath284 the function @xmath19 has no longer the root corresponding to @xmath281 . instead of the root , it acquires a new pole corresponding to the second excited state of the trimer with the energy @xmath14 . note that though the virtual levels @xmath280 and @xmath281 appear beyond the domain @xmath106 , already at @xmath285 the point @xmath281 turns out to be inside this domain . therefore , it should be considered as a `` true '' virtual level of the trimer . we expect that the subsequent efimov levels originate from the virtual levels just according to the same scheme as the level @xmath14 does . the other purpose of the present investigation is to determine the mechanism of disappearance of the excited state of the helium trimer when the two - body interactions become stronger owing to the increasing coupling constant @xmath8 . it turned out that this disappearance proceeds just according to the scheme of the formation of new excited states ; only the order of occurring events is inverse . the results of our computations of the energy @xmath13 when @xmath7 changes from 1.05 to 1.17 are given in table iv . in the interval between @xmath286 and @xmath287 there occurs a jump " of the level @xmath13 on the unphysical sheet and it transforms from the pole of the function @xmath17 into its root , @xmath288 , corresponding to the trimer virtual level . the results of calculation of this virtual level where @xmath7 changes from 1.18 to 1.5 are presented in table v. for all the values of @xmath7 presented in tables iv and v , the dimer possesses an only bound state . we have found that the first excited state of the dimer appears only at @xmath289 . note that in the case of finite potentials the geometric characteristics of the domain @xmath106 where the function @xmath19 can be calculated reliably , are only determined by the value of @xmath290 ( see formula ( [ pipsi ] ) for @xmath205 ) . when @xmath290 increases , the domain @xmath106 is enlarged . it is easy to check that the energies of the excited state level @xmath11 and of the virtual level @xmath291 given in tables iv and v belong to the corresponding domains @xmath292 . for @xmath293 , this results in a weak dependence of the calculated values of @xmath11 and @xmath291 on the parameters @xmath294 , @xmath249 and ( this is especially important ) on the parameter @xmath209 . in essential , we chose the values of the cutoff hyperradius @xmath209 given in tables iii v from the scaling considerations . as a matter of fact , we took the value of @xmath209 following the formula @xmath295 where the constant " @xmath296 corresponds to an appropriate choice of @xmath209 at @xmath297 . it has been established in @xcite that such a choice is ensured if @xmath298400600 . in determining the values of @xmath299 , indicated in table iii , we followed the formula ( [ rhomax ] ) literally . as the constant " @xmath300 , we took its value corresponding to the base value of @xmath301600 . the values of @xmath302 presented in tables iv and v correspond to the choice of @xmath303 in the interval within 400 and 800 . all the results presented in tables iii v have been obtained with the grids parameters @xmath269 . the authors are grateful to prof . v.b.belyaev and prof . h.toki for help and assistance in calculations at the supercomputer of the research center for nuclear physics of the osaka university , japan . also , the autors thank mrs . t.dumbrais for her help in translation of the text into english . one of the authors ( a.k.m . ) is much indebted to prof . w.sandhas for his hospitality at the universitt bonn . the support of this work by the deutsche forschungsgemeinschaft and russian foundation for basic research is gratefully acknowledged . 99 rama krishna m.v . , whaley k.b . rev . lett . p. 1126 . lehman k.k . , scoles g. // science . p. 2065 . grebenev s. , toennies j.p . , vilesov a.f . // science . p. 2083 . efimov v. // luo f. , mcbane g.c . , kim g. , giese c.f . , gentry w.r . // j. chem . phys . p. 3564 . luo f. , giese c.f . , gentry w. r. // j. chem . phys . 1996 . v. 104 . p. 1151 . schllkopf w. , toennies j.p . // science . 1994 . v. 266 . p. 1345 . aziz r.a . , nain v.p.s . , carley j.s . , taylor w.l . , mcconville g.t . // j. chem . phys . 1979 . v. 79 . p. 4330 . aziz r.a . , mccourt f.r.w . , wong c.c.k . 1987 . v. 61 . p. 1487 . , slaman m.j . // j. chem . 1991 . v. 94 . p. 8047 . tang k.t . , toennies j.p . , yiu c.l . // 1995 . v. 74 . p. 1546 . mcmillan w.l . pandharipande v.r . , zabolitzky j.g . , pieper s.c . , wiringa r.b . , helmbrecht u. 1983 . v. 50 . p. 1676 . usmani n. , fantoni s. , pandharipande v.r . p. 6123 . pieper s.c . , wiringa r.b . , pandharipande v.r . // phys . rev . b. 1985 . v. 32 . p. r3341 . rick s.w . , lynch d.l . , doll j.d . // j. chem . phys . 1991 . v. 95 . p. 3506 . levinger j.s . // yadernaya fizika . , lin c.d . , greene c.h . // a. 1996 . v. 54 . nielsen e. , fedorov d.v . , jensen a.s . // lanl e - print physics/9806020 . nakaichi - maeda s. , lim t.k . a. 1983 . v. 28 . cornelius th . , glckle w. // j. chem . phys . 1986 . v. 85 . p. 3906 . carbonell j. , gignoux c. , merkuriev s.p . // few body systems schllkopf w. , toennies j.p . // j. chem . phys . 1996 . v. 104 . p. 1155 . kolganova e.a . , motovilov a.k . , sofianos s.a . a. 1997 . v. 56 . p. 1686 . motovilov a.k . , sofianos s.a . , kolganova e.a . phys . lett . 1997 . v. 275 . kolganova e.a . , motovilov a.k . , sofianos s.a . // j. phys . p. 1279 . fedichev p.o . , reynolds m.w . , shlyapnikov g.v . 1996 . v. 77 . p. 2921 . merkuriev s.p . , motovilov a.k . 1983 . v. 7 . merkuriev s.p . , motovilov a.k . , yakovlev s.l . kolganova e.a . , motovilov a.k . // phys . atom . faddeev l.d . , merkuriev s.p . quantum scattering theory for several particle systems . kluwer academic publishers , doderecht , 1993 . merkuriev s.p . , gignoux c. , laverne a. // ann . phys . 1976 . v. 99 . motovilov a.k . 1996 . v. 107 . motovilov a.k .

astrodynamical space test of relativity using optical devices ( astrod ) @xcite aims at testing relativistic gravity , measuring the solar - system parameters with high precision and detecting gravitational waves from massive black holes and galactic binary stars . the concept of astrod is to put two spacecraft in separate solar orbits and carry out laser interferometic ranging with earth reference stations ( e.g. a spacecraft at the earth - sun l1/l2 points ) . a simple version of astrod , astrod i , has been studied as the first step to astrod @xcite . astrod i employs one spacecraft in a solar orbit and carries out interferometric ranging and pulse ranging with ground stations . the main scientific goals of astrod i are to test relativistic gravity and the fundamental laws of spacetime with three - order - of - magnitude improvement in sensitivity and to improve the solar , planetary and asteroid parameter determination by 1 to 3 orders of magnitude . the technological goal of astrod i is to prepare for the astrod mission . the acceleration disturbance goal of the astrod i proof mass is @xmath0ms@xmath1hz@xmath2 at frequency @xmath3 of 0.1mhz . the power spectral density of the allowed level of the acceleration noise is shown in figure [ fig : curve ] . assuming a 10ps one - way timing accuracy ( 3 mm ranging accuracy ) and the acceleration noise of @xmath0ms@xmath1hz@xmath2 at frequency of about 0.1mhz , a simulation for 400 days ( 350750 days after launch ) showed that astrod i could determine the relativistic parameters @xmath4 and @xmath5 , and the solar quadrupole parameter @xmath6 to levels of 10@xmath7 , 10@xmath7 and 10@xmath8 , respectively @xcite . in the simulation , ( i ) the timing noise is modeled as gaussian random noise ; ( ii ) unknown acceleration noise is modeled to have gaussian random magnitude with zero mean and with standard deviation 10@xmath9ms@xmath1 and to have its direction changed randomly every 4h ( equivalent to 10@xmath10ms@xmath1hz@xmath2 for @xmath3 @xmath11 0.1mhz assumed as the requirement of the drag - free system ) and ( iii ) five range points are taken each day ( at 0.2d interval ) . longer term systematic effects will be studied in a future paper . this simulation agrees with the scientific goals of astrod i. the timing uncertainty of event timer reaches 3ps in satellite laser ranging at present . space qualified versions of similar accuracy are under development . for a ranging uncertainty of 3 mm in a distance of 3 @xmath12 10@xmath13 m ( 2au ) , the laser / clock frequency needs to be known to one part in 10@xmath14 . this can be set as a requirement of the space laser / clock or a requirement for laser frequency monitoring through ground clock and modeling . as to ground station jitter , monitoring to an accuracy of 3 mm is required and can be achieved . the atmospheric effects on laser propagation will be monitored and subtracted to mm - level by using 2-color ( 2-wavelength ) ranging ( one color for pulse ranging and one for interferometric ranging ) . these measurement uncertainties are not cumulative in the range determination while the acceleration disturbances accumulate in time in the geodesic deviations . in order to achieve the acceleration disturbance goal , a drag - free control system using capacitive sensors will be employed . in order to decide on detailed designs of the accelerometer for astrod i , we have to know the sources and magnitude of acceleration disturbances that could arise in the accelerometer . in this paper , we carry out analyses , mainly based on existing literature on acceleration disturbances for other gravitational missions ( lisa @xcite and the lisa pathfinder @xcite ) , to give a preliminary overview of the acceleration disturbance to the astrod i proof mass . based on the analyses , we infer some of the requirements for the designs of the astrod i payload and spacecraft . also , we compare parameter values we have assumed in the analyses with those for lisa to confirm the feasibility of the requirements for astrod i. first , we will give an overview of the astrod i configuration ( section [ configuration ] ) and the control - loop model we assumed ( section [ control - loop ] ) . then , we will estimate the magnitude of acceleration disturbances and requirements for astrod i in sections [ ans ] to [ st : requirements ] , and compare the requirements of astrod i with lisa in section [ comparison with lisa ] . the astrod i spacecraft has a cylindrical shape with diameter 2.5 m and height 2 m . its cylindrical side is covered by solar panels . the cylindrical axis is perpendicular to the orbit plane and the telescope is set to point toward a ground laser station . the effective area of receiving sunlight is about 5m@xmath15 and it can generate power that is larger than 500w . the total mass of the spacecraft is about 350 kg and that of payload is 100120 kg ( see @xcite for more detailed descriptions of the configuration ) . the orbit distance from the sun varies from about 0.5au to 1au ( figure 2 of @xcite ) . the proof mass ( @xmath16 = 1.75 kg ) is a rectangular parallelepiped ( 50 @xmath12 50 @xmath12 35mm@xmath17 ) made from au - pt alloy ( density @xmath18 = 2 @xmath12 10@xmath19kgm@xmath20 ) . the six sides of the proof mass are surrounded by electrodes mounted on the housing for capacitive sensing and control . the gap between each side of the proof mass and the opposing electrode is 2 mm . assumed values for the capacitance and voltages are listed in table [ table : parameter values ] . various acceleration disturbances would act on the proof mass in different ways . in order to infer how these different kinds of acceleration disturbances would contribute to the total acceleration disturbance of the proof mass , we tentatively assume a simple single - mass and single - axis control - loop model @xcite . the diagram is shown in figure [ fig : control - loop ] . the relative difference between the displacement disturbance amplitudes of the proof mass and of the spacecraft , @xmath21 , with position readout noise @xmath22 , is measured by the position displacement sensor . the output of this sensor is converted to acceleration disturbance @xmath23 , by a transfer function @xmath24 ( @xmath25 ) . this acceleration is supplied to the thruster and the output acceleration disturbance with thruster noise @xmath26 is applied to the spacecraft . the spacecraft also experiences acceleration disturbances by coupling to the proof mass with a coupling constant @xmath27 and external environmental disturbances ( @xmath28 ) that work directly on the spacecraft . the total acceleration of the spacecraft @xmath29 is converted to the position noise @xmath30 with a transfer function @xmath31 . we consider only the sensitive axis of the proof mass without actuation . the proof mass would experience disturbances by spacecraft - proof mass coupling and environmental disturbances @xmath32 ( see section [ anp ] ) . the total acceleration disturbance of the proof mass @xmath33 is converted to the displacement disturbance @xmath34 with a transfer function @xmath35 . from this control - loop model , we obtain the following linear loop equations @xcite : @xmath36 by solving these equations for @xmath34 , and assuming @xmath37 and @xmath38 @xmath39 @xmath40 (= @xmath41 ; @xmath42 is nominally 1 and its effects are absorbed in @xmath24 and @xmath26 ) , we obtain ( e.g. @xcite ) : @xmath43 where @xmath44 . this acceleration disturbance has to be less than the acceleration noise goal of @xmath45ms@xmath1hz@xmath2 at @xmath3 = 0.1mhz . we will estimate the values of the direct acceleration disturbances of the spacecraft ( @xmath28 and the thruster noise @xmath46 ) in section [ ans ] and of the proof mass ( @xmath32 ) in sections [ anp ] and [ back - action ] , and the stiffness @xmath27 in section [ stiff ] . we will discuss the requirements for @xmath22 and @xmath38 in section [ st : requirements ] . the spacecraft would be affected by environmental disturbances that stem from , for example , solar radiation pressure , solar wind and micrometeorite impacts . among these sources of disturbances , solar radiation pressure is considered to be the major contributor to the acceleration disturbances ( section 7 of @xcite ) . the contribution from solar wind might be comparable to radiation pressure , but the spectral behavior of the solar wind is not well known . by assuming a perfectly reflecting surface of the spacecraft , acceleration noise caused by fluctuation in solar irradiance @xmath47 is @xmath48 , where @xmath49 is the area of the spacecraft facing the sun , @xmath50 is the mass of the spacecraft and @xmath51 is the speed of light in vacuum . from the data of the virgo experiment on soho @xcite , fractional fluctuation in solar irradiance is @xmath52hz@xmath2 at 0.1mhz ( figure 6 of @xcite ) . assuming that @xmath53 is the same at 0.5au , and taking a total irradiance of 5500wm@xmath1 , the area of 5m@xmath15 and a mass of 350 kg , we obtain @xmath54 = 9.8 @xmath12 10@xmath55ms@xmath1hz@xmath2 . the impact rate of 1-ng meteorites on the astrod i spacecraft ( its surface area is 25.5m@xmath15 ) is about 7 events per day or about 0.08mhz . the average velocity of meteorites is 18kms@xmath56 in an earth frame near the earth ( figure 7.2 - 30 of @xcite ) . an impact of 1-ng meteorite with the average velocity of 18kms@xmath56 on the surface of the spacecraft ( 350 kg ) without reflection produces a linear velocity increment of about 5 @xmath12 10@xmath57ms@xmath56 . smaller meteorites have larger flux ( figure 7.2 - 29 of @xcite ) , but their impacts on velocity changes are smaller . therefore , the contribution to the acceleration disturbances at 0.1mhz seems insignificant , in comparison with the effect of solar radiation pressure . however , because such discrete changes could directly affect the astrod i experiment , the impact effects have to be studied carefully in detail . in addition to the acceleration noise from the environmental disturbances , the spacecraft would suffer from the thruster noise . a force fluctuation of 10@xmath58nhz@xmath2 in thruster corresponds to the acceleration disturbance of @xmath46 = 2.8 @xmath12 10@xmath8ms@xmath1hz@xmath2 . therefore , the rss ( root - sum - square ) of these acceleration disturbances is : @xmath59 at 0.1mhz , dominated by the thruster noise . as per @xcite , lisa requires a force fluctuation to be less than a few @xmath58nhz@xmath2 . however , recent studies for the lisa pathfinder indicate that force noise up to 0.1mnhz@xmath2 at 0.1mhz can be tolerated by increasing the gain @xcite ; the lisa study team is designing high gain loops and the requirement has been relaxed . direct proof - mass acceleration disturbances can be classified into two categories depending on their origins @xcite : environmental disturbances ( @xmath60 ) and proof - mass sensor back - action acceleration disturbances ( @xmath61 ) . the former includes disturbances related to magnetic effects ( @xmath62 , @xmath63 , @xmath64 , @xmath65 and @xmath66 ) , impact effects ( by cosmic ray ( @xmath67 ) and residual gas ( @xmath68 ) ) , temperature dependent effects ( radiometric and outgassing effects ( @xmath69 and @xmath70 ) and thermal radiation pressure ( @xmath71 ) ) and gravity gradients caused by thermal distortion of the spacecraft ( @xmath72 ) . the latter is originated from voltage fluctuations ( @xmath73 and @xmath74 ) , charge fluctuations ( @xmath75 and @xmath76 ) , readout electronics ( @xmath77 ) , patch field voltages ( @xmath78 ) and thermal voltage noise by dielectric losses ( @xmath79 ) . parameter values and physical constants used for the estimations are listed in tables [ table : parameter values ] and [ table : physical constants ] , respectively . table [ table : pm acceleration disturbances ] provides a summary of the expressions used to estimate the direct proof - mass acceleration disturbances and the estimated values . the disturbances noted as @xmath62 , @xmath63 , @xmath65 , @xmath66 , @xmath67 , @xmath68 , @xmath69 , @xmath71 , @xmath72 and @xmath73-@xmath76 in table [ table : pm acceleration disturbances ] correspond to @xmath80-@xmath81 and @xmath82-@xmath83 in @xcite , respectively in @xcite , would not arise in astrod i , as the laser beam is not to be injected directly on the surface of the astrod i proof mass . ] . we will briefly explain each disturbance below . @ll @xmath3 = 0.1 mhz & astrod i + + mass : @xmath16 [ kg ] & 1.75 + density : @xmath18 [ kg m@xmath20 ] & @xmath85 + cross section : @xmath86 [ m@xmath15 ] & 0.050 @xmath12 0.035 + temperature : @xmath88 [ k ] & 293 + fluctuation of temperature difference & + across pm and housing : @xmath89 [ k hz@xmath2 ] & 1.0 @xmath12 @xmath91 + maximum charge build - up : @xmath92 [ c ] & @xmath93 + fluctuation in charge : @xmath94 [ c hz@xmath2 ] & 6.1 @xmath12 10@xmath9 + magnetic susceptibility : @xmath95 & @xmath96 + magnetic remanent moment : @xmath97 [ a m@xmath15 ] & 1 @xmath12 10@xmath7 + residual gas pressure : @xmath35 [ pa ] & @xmath91 + electrostatic shielding factors : @xmath98 & 10 + magnetic shielding factors : @xmath99 & 1 + + mass : @xmath50 [ kg ] & 350 + velocity : @xmath100 [ m s@xmath56 ] & @xmath102 + thruster noise [ @xmath58n hz@xmath2 ] & 10 + area facing the sun : @xmath49 [ m@xmath15 ] & 5 + fluctuation of temperature in sc : @xmath103 [ k hz@xmath2 ] & 0.2 + + capacitance : @xmath104 [ pf ] & 6 + capacitance to ground : @xmath105 [ pf ] & 6 + total capacitance : @xmath106 [ pf ] & ( @xmath107 @xmath108)36 + gap : @xmath109 [ mm ] & 2 + asymmetry in gap across opposite sides of pm : @xmath110 [ @xmath58 m ] & 10 + average voltage across opposite faces : @xmath111 ( @xmath39 @xmath112 ) [ v ] & 0.1 + proof mass bias voltage : @xmath113 [ v ] & 0.6 + voltage difference between @xmath111 and voltage to ground : @xmath114 [ v ] & 0.05 + voltage difference between opposing faces : @xmath115 [ v ] & 0.01 + fluctuation of voltage difference across opposite faces : & + @xmath116 [ v hz@xmath2 ] & @xmath117 + residual dc bias voltage on electrodes : @xmath118 [ v ] & 0.1 + loss angle : @xmath119 & 10@xmath120 + + local magnetic field : @xmath121 [ t ] & 8 @xmath12 10@xmath7 + local magnetic field gradient : @xmath122 [ t m@xmath56 ] & 3 @xmath12 10@xmath123 + fluctuation in local magnetic field : @xmath124 [ t hz@xmath2 ] & 1 @xmath12 @xmath125 + interplanetary magnetic field : @xmath126 [ t ] & @xmath127 + interplanetary magnetic field gradient : @xmath128 [ t hz@xmath2 ] & @xmath129 + gradient of time - varying magnetic field : @xmath130 [ t m@xmath56 hz@xmath2 ] & 4 @xmath12 10@xmath8 + + impact rate : @xmath131 [ s@xmath56 ] & 30 + mass of positron : @xmath132 [ kg ] & @xmath133 + incident energy : @xmath134 [ j ] & ( 200 @xmath135 =) 3.2 @xmath12 10@xmath57 + + source mass : @xmath136 [ kg ] & 1 in @xmath72 and 0.03 in @xmath137 + distance from the source mass : @xmath138 [ m ] & 0.5 in @xmath72 and 0.05 in @xmath137 + thermal expansion coefficient : @xmath139(aluminium ) [ k@xmath56 ] & 2.5 @xmath12 10@xmath120 + + multiplicative factor : @xmath4 & 5 + patch field voltage : @xmath140 [ v ] & 0.1 + @lll & symbols & values used in the estimations + elementary charge & @xmath141 & 1.6 @xmath142 c + permeability of vacuum & @xmath143 & 1.26 @xmath144 n a@xmath1 + permittivity of vacuum & @xmath145 & 8.9 @xmath12 10@xmath146 f m@xmath56 + speed of light in vacuum & @xmath51 & 3.0 @xmath147 m s@xmath56 + boltzmann constant & @xmath148 & 1.38 @xmath149 j k@xmath56 + stefan - boltzmann constant & @xmath150 & 5.7 @xmath151 w m@xmath1 k@xmath152 + gravitational constant & @xmath153 & 6.7 @xmath12 10@xmath57 m@xmath17 kg@xmath56 s@xmath1 + @llll [ @xmath154 m s@xmath1 hz@xmath2 ] & & expressions & astrod i + + @xmath62 & magnetic & @xmath155 & 1.2 + @xmath63 & magnetic & @xmath156 & 3.4 + @xmath64 & magnetic & @xmath157 & 1.6 + @xmath65 & magnetic & @xmath158 & 0.0017 + @xmath66 & magnetic & @xmath159 & 0.91 + @xmath160 & cosmic rays & @xmath161 & 0.0015 + @xmath68 & residual gas & @xmath162 & 0.74 + @xmath69 & radiometric & @xmath163 & 0.17 + @xmath70 & outgassing & 10@xmath69 & 1.7 + @xmath71 & thermal radiation & @xmath164 & 0.13 + @xmath72 & gravity gradients & @xmath165 & 2.7 + & 5.2 + + @xmath73 & @xmath166 & @xmath167 & 1.4 + @xmath74 & @xmath168 & @xmath169 & 4.8 + @xmath75 & @xmath170 & @xmath171 & 2.9 + @xmath76 & @xmath172 & @xmath173 & 0.040 + @xmath77 & readout electronics & ( see table 1 of @xcite ) & 1.8 + @xmath78 & patch fields & @xmath174 & 29 + @xmath79 & dielectric losses & @xmath175 & 1.6 + & 30 + & 30 + the lowest order force on a proof mass with a magnetic moment @xmath176 in an external magnetic field @xmath177 is given by @xmath178 = @xmath179 @xcite . the magnetic moment of the proof mass is a vector sum of the remanent moment @xmath180 and the induced moment @xcite : @xmath181 , where @xmath95 and @xmath182 are magnetic susceptibility and the volume of the proof mass , respectively ; @xmath183na@xmath1 is the permeability of vacuum . the external magnetic field would be given by the superposition of the interplanetary magnetic field @xmath184 and a local magnetic field @xmath185 . dominant terms in acceleration of the proof mass due to the induced magnetic moment produce acceleration disturbances noted @xmath62 and @xmath63 in table [ table : pm acceleration disturbances ] , where @xmath99 is a scaling factor for possible suppression by magnetic shielding . the average interplanetary magnetic field at 1au from the sun varies from @xmath186 to 3.7 @xmath12 10@xmath8 t @xcite . ulysses data obtained near 1au from the sun ( figure 9 of @xcite ) showed a @xmath187 dependence of the variation in the interplanetary magnetic field and @xmath188 can be inferred to be about @xmath125thz@xmath2 at 0.1mhz . as the behavior of @xmath188 at 0.5au from the sun is uncertain , we use a somewhat higher value of @xmath189thz@xmath2 ( see footnote in section [ st : lorentz ] ) . according to the formulation studies for lisa and the implementation work of the lisa pathfinder , batteries and micro - thrusters are the principal suspects of the origins of the local magnetic field in lisa . we need elaborate modeling works to estimate the magnitude of local magnetic field . here , we use the same values used in analyses for lisa @xcite : @xmath190 t , @xmath191 = 10@xmath7thz@xmath2 and @xmath122 @xmath108 3 @xmath192 @xmath108 3 @xmath12 10@xmath123tm@xmath56 ( where @xmath193 = 0.75 m ) . silvestri et al . have reported that the magnetic susceptibility of five samples ranged from @xmath1942.8 @xmath12 10@xmath120 to @xmath1942.1 @xmath12 10@xmath120 for two of them without traceable iron contamination and from + 1.1 @xmath12 10@xmath120 to + 8.8 @xmath12 10@xmath120 for the rest samples with a trace of iron @xcite . lower susceptibility may be achievable by controlling the manufacturing process of the alloy . the requirement for lisa is @xmath95 = 3 @xmath12 10@xmath123 @xcite . we use a somewhat moderate value of @xmath95 = 5 @xmath12 10@xmath120 . assuming @xmath99 = 1 as per @xcite , we obtain @xmath62 = 1.2 @xmath12 10@xmath9ms@xmath1hz@xmath2 and @xmath63 = 3.4 @xmath12 10@xmath9ms@xmath1hz@xmath2 at 0.1mhz . acceleration disturbance due to the magnetic remanent moment is given by @xmath64 listed in table [ table : pm acceleration disturbances ] ( equation 4(c ) of @xcite ) , where @xmath195 is time - varying magnetic field gradients . according to measurements by gill et al . , magnetic remanent moment of three 4.2-cm 70/30 au - pt alloy cubes ( one of them was 73/27 au - pt alloy ) ranged from 3.5 @xmath12 10@xmath196am@xmath15kg@xmath56 to 3.1 @xmath12 10@xmath8am@xmath15kg@xmath56 @xcite . by scaling the largest measured remanent moment by weight , it is @xmath197am@xmath15 for a 1.75-kg proof mass . by using a somewhat relaxed value of 1 @xmath12 10@xmath7am@xmath15 and @xmath198tm@xmath56hz@xmath2 @xcite , which is a factor of four higher than the assumed requirement for lisa in @xcite , we obtain @xmath199ms@xmath1hz@xmath2 at 0.1mhz . in the error estimates for lisa @xcite , magnetic remanent moment of 2 @xmath12 10@xmath8am@xmath15 is used . the proof mass in orbit would be charged up by cosmic - ray impacts @xcite . as the charged proof mass moves through the interplanetary magnetic field ( @xmath184 ) with a velocity @xmath100 of about 4 @xmath12 10@xmath19ms@xmath56 , it experiences the lorentz force : @xmath200 = @xmath201 , where @xmath92 is the built - up charge . acceleration disturbances due to the fluctuation of the charge build - up ( @xmath94 ) in the proof mass and of the average interplanetary magnetic field are given by @xmath65 and @xmath66 ( see table [ table : pm acceleration disturbances ] ) , respectively , where @xmath98 is an electrostatic shielding factor ; in the rest frame of the proof mass , the motion of the proof mass through the interplanetary magnetic field generates an electrostatic field . by assuming the poisson distribution of cosmic - ray impacts , the average charge fluctuation spectral density can be defined as @xmath202 . for the frequency ( @xmath203 ) of 0.1mhz and the effective charging rate ( @xmath204 ) of 288 @xmath205s@xmath56 , which was estimated for a lisa proof - mass ( 46-mm cube ) by a simulation using geant 4 toolkit @xcite , we obtain @xmath206 = 6.1 @xmath12 10@xmath9chz@xmath2 . using this value , @xmath98 = 10 and @xmath126 = 1.2 @xmath12 10@xmath7 t , we obtain @xmath65 = 1.7 @xmath12 10@xmath207ms@xmath1hz@xmath2 at 0.1mhz . the maximum value of @xmath126 at 1au is about 3 @xmath12 10@xmath8 t . we tentatively use four times the maximum value as @xmath126 at 0.5au , by assuming 1/(distance)@xmath15 dependence of @xmath126 10@xmath55 t . the main magnetic field at 0.5au to 1au is due to the influence of solar winds which attenuate with ( distance)@xmath15 . ] . because the volume of the astrod i proof - mass is about 10% smaller than the lisa proof mass , the charging rate for astrod i might be smaller than the value for lisa . bao et al . are working on simulations to estimate the charging rates for astrod i @xcite . stebbins et al . use @xmath98 = 100 , which is one order of magnitude larger than the value we used here , in the current error estimates for lisa @xcite . taking a nominal maximum build - up charge @xmath92 = 10@xmath146c , which is one order of magnitude larger than the value used in the error estimates for lisa @xcite , we obtain @xmath66 = 9.1 @xmath12 10@xmath208ms@xmath1hz@xmath2 . some cosmic rays get stopped in the proof mass and deposit momentum @xcite . assuming the poisson distribution of cosmic ray impact , spectral density of momentum transfer ( @xmath209 ) is @xmath210 , where @xmath131 is the fluctuation in the impact rate @xcite . the acceleration disturbance due to the fluctuation in the impact rate is given by @xmath160 , listed in table [ table : pm acceleration disturbances ] . the impact rate was inferred from simulations done for lisa ; by adding the effects of all stopped particles ( protons and helium ) and taking into account their directions , the acceleration disturbance by momentum transfer was estimated to be @xmath108 2 @xmath12 10@xmath207ms@xmath1hz@xmath2 for a lisa proof mass @xcite . this corresponds to a disturbance due to momentum transfer by protons ( mass @xmath2111.7 @xmath12 10@xmath212 kg ) , with an impact rate of @xmath11 30s@xmath56 , at incident energy @xmath134 = 200mev (= 3.2 @xmath12 10@xmath57j ) . using these values , the acceleration disturbance becomes @xmath160 = 1.5 @xmath12 10@xmath207ms@xmath1hz@xmath2 . from the kinetic theory , the number of residual - gas molecules ( assumed as ideal gas ) that pass an area ( @xmath86 ) of the proof mass per second is given by @xmath213 = @xmath214 , where @xmath215 = @xmath216 is the number density of the molecules and @xmath217 = @xmath218 is the average thermal velocity ( @xmath35 is the pressure of residual gas , @xmath148 = 1.38 @xmath12 10@xmath219jk@xmath56 is the boltzmann constant , @xmath220 is the temperature of the proof - mass housing and @xmath221 = 4.65 @xmath12 10@xmath222 kg is the mass of nitrogen molecules ) . assuming the poisson distribution of the impact rate , we define the spectral density of fluctuations in @xmath213 as @xmath223 @xmath39 @xmath224 . acceleration due to the residual gas impacts is given by @xmath225 and acceleration disturbance due to fluctuations in the impact rate of residual gas is given by @xmath68 listed in table [ table : pm acceleration disturbances ] . for @xmath35 = 10@xmath120pa , @xmath226 = 1.75 @xmath12 10@xmath20m@xmath15 and @xmath220 = 293k , we obtain @xmath68 = 7.4 @xmath12 10@xmath208ms@xmath1hz@xmath2 . stebbins et al . use 3@xmath1210@xmath123pa as the residual gas pressure around the proof mass in the error estimates for lisa @xcite . acceleration disturbance due to the radiometric effect ( e.g. @xcite ) is given by @xmath69 listed in table [ table : pm acceleration disturbances ] , where @xmath89 is fluctuation in temperature difference across the proof mass housing . this value has to be estimated by carrying out thermal modeling . according to thermal analysis for lisa , temperature fluctuation on the optical bench , to which the proof mass housing is mounted , due to power dissipation of amplifiers is about 3.0 @xmath12 10@xmath120khz@xmath2 at 1mhz ( table 6.2 - 28 of @xcite ) . by assuming that the fluctuation rises as @xmath227 , the temperature fluctuation of the optical bench is @xmath228khz@xmath2 at 0.1mhz . at the frequency of 0.1mhz and higher frequencies , the temperature fluctuation on the optical bench would be dominated by the fluctuation in the power dissipation @xcite ; fluctuation due to solar irradiance at 0.1mhz is @xmath229khz@xmath2 ( table 6.2 - 16 of @xcite ) . the ratio @xmath230/@xmath231 would range from 30 to 100 @xcite . by using a value of 30 for the ratio , we obtain @xmath232khz@xmath2 at 0.1mhz . by using this value , we obtain @xmath69 = 1.7 @xmath12 10@xmath208ms@xmath1hz@xmath2 at 0.1mhz . outgassing from walls of the sensor cage is thought to produce greater acceleration noise than the radiometric effect @xcite . an analysis done for the lisa technology package ( ltp ) , assuming a simple model of flow circuit with a linear approximation , shows that the outgassing effect is nearly 10 times the radiometric effect @xcite . by using this estimate and the estimate we made for the radiometric effect in the previous section , we obtain @xmath70 = 1.7 @xmath12 10@xmath9ms@xmath1hz@xmath2 at 0.1mhz for astrod i. by assuming a perfectly reflecting surface of the proof mass , thermal radiation pressure produces acceleration disturbance ( @xmath71 in table [ table : pm acceleration disturbances ] ) due to fluctuations in the temperature difference across the proof - mass housing . in the expression of @xmath71 listed in table [ table : pm acceleration disturbances ] , @xmath150 = 5.7 @xmath12 10@xmath8wm@xmath1k@xmath152 is the stefan - boltzmann constant , @xmath86 is the area of the proof mass and @xmath88 is the temperature of the proof - mass housing . a factor of one - third is multiplied , as done in the estimation for lisa by schumaker @xcite , as a margin for the fact that not all of the radiation momentum is normally incident on the proof mass . for the housing temperature of 293k and the temperature fluctuation of 1.0 @xmath12 10@xmath120khz@xmath2 , we obtain @xmath71 = 1.3 @xmath12 10@xmath208ms@xmath1hz@xmath2 at 0.1mhz . the same value was used for the temperature fluctuation in the error estimates for lisa @xcite . the temperature fluctuation in solar irradiance would cause fluctuation in distortions of the spacecraft : @xmath233 , where @xmath139 and @xmath103 are the coefficient of thermal expansion and temperature fluctuation of the spacecraft , respectively . the inherent fluctuation in solar radiation is @xmath234hz@xmath2 at 0.1mhz @xcite . therefore , the temperature fluctuation is 0.2khz@xmath2 for @xmath88 = 293k . the gravitational disturbance by a 1-kg mass ( @xmath136 ) separated from the proof mass in the sensitive axis by @xmath138 = 0.5 m of aluminium structure would be @xmath235ms@xmath1hz@xmath2 for @xmath103 = 0.2khz@xmath2 and @xmath139 ( of aluminium ) = 2.5 @xmath12 10@xmath120k@xmath56 . the same disturbing mass ( @xmath136 = 1 kg and @xmath138 = 0.5 m ) is assumed in @xcite . in reality , the mass to be involved in thermal distortion would be much larger , but the influence would be largely canceled because of the axial symmetry in the original spacecraft geometry . the inherent fluctuation in solar radiation could be reduced largely by thermal shielding . in the estimation for lisa , @xmath103 = 0.004khz@xmath2 is used @xcite . for a more accurate estimate , gravity effects by thermal and non - thermal distortions of the spacecraft and the payload have to be studied by appropriate modeling . the total mechanical energy of the capacitive sensing system can be expressed as ( e.g. equation ( a.3 ) of @xcite ) : @xmath236 where @xmath92 is a net charge of the proof mass ; @xmath106 is the sum of the capacitances due to the applied voltages on the surrounding electrodes @xmath237 and the potential to ground @xmath238 : @xmath239 , where @xmath240 ; @xmath241 is the voltage induced on the proof mass by the applied voltages on the electrodes : @xmath242 . the first term of @xmath243 is the total energy done on the proof mass by the applied voltages of the surrounding electrodes . the second term is the energy acquired on the proof mass by the image charge on the surrounding electrodes . the third term is the energy stored on the proof mass by the deposit of the free charge on the proof mass . the @xmath138-component of the force on the proof mass is given by differentiation of equation ( [ eq : total mechanical energy ] ) with respect to @xmath138 . for a simplicity , we assume that neither the free charge @xmath92 nor the potentials @xmath244 have appreciable gradients along the @xmath138-axis , as per @xcite . acceleration disturbances due to fluctuations in the applied voltages and charge can be given by @xmath73 - @xmath76 as listed in table [ table : pm acceleration disturbances ] . a detailed description on the deviations of @xmath73 - @xmath76 is given in appendix a of @xcite . the parameter values used in this section and the estimated values are listed in tables [ table : parameter values ] and [ table : pm acceleration disturbances ] , respectively . to simplify the analysis , several assumptions were made in the process of deriving the expressions for the four classes of acceleration disturbances ( @xmath73 - @xmath76 ) @xcite : ( 1 ) @xmath245 are all comparable with each other in magnitude and on the order of @xmath104 @xmath108 6pf (= @xmath246 , where @xmath145 = 8.9 @xmath12 10@xmath146fm@xmath56 is the permittivity of vacuum , @xmath247 is the area of each electrode and @xmath109 = 2 mm is the gap ) ; ( 2 ) only the capacitances @xmath248 and @xmath249 have nonzero gradients along the @xmath138-axis : @xmath250 = @xmath251 , and the gradient of the total capacitance @xmath106 @xmath108 6 @xmath104 is : @xmath252 @xmath108 @xmath253 , where @xmath110 is the gap asymmetry in the @xmath138-direction ; ( 3 ) the average of the potentials on opposing faces is same for all three axes and expressed as @xmath111 ( @xmath254 0.1v ) ; ( 4 ) the magnitude of the fluctuation in potential @xmath255 is all identical and take the value of the average fluctuation of all the potentials for the three axes and the voltage to ground . we express the fluctuation as @xmath256 . it should be noted that we do not consider any cross - talk effects that arise in the sensitive axis due to forces applied to the other degrees of freedom . @xmath73 is associated only with sensing voltages but not the free charge , and its value is @xmath73 @xmath108 1.4 @xmath12 10@xmath9ms@xmath1hz@xmath2 , by assuming @xmath114 @xmath39 @xmath257 = 0.05v . this value is larger than the value used for lisa @xcite by a factor of five . also , we use @xmath116 = 1.0 @xmath12 10@xmath152vhz@xmath2 as fluctuation in voltage imbalance @xmath115 ( @xmath39 @xmath258 = 0.01v ) . this value ( @xmath116 ) is one order of magnitude larger than the value used for lisa@xcite . further , we assume that @xmath259 @xmath108 1/6 , as per @xcite . @xmath74 and @xmath75 arise from fluctuations in the force due to the interaction between the net free charge and applied sensing voltages . @xmath74 is due to fluctuation in voltage imbalance and its value is @xmath74 @xmath108 @xmath1944.8 @xmath12 10@xmath9ms@xmath1hz@xmath2 , where the net free charge on the proof mass is set to the nominal maximum built - up charge , described earlier . @xmath75 is due to fluctuation in charge and its value is @xmath75 @xmath108 @xmath1942.9 @xmath12 10@xmath9ms@xmath1hz@xmath2 . the value of @xmath115 used here is larger than the one used in the error estimates for lisa @xcite by a factor of two . @xmath76 is associated only with the free charge and its value is @xmath76 @xmath108 4.0 @xmath12 10@xmath260ms@xmath1hz@xmath2 . in the estimation , we assume that the gap asymmetry in the @xmath138-direction is @xmath110 = 10@xmath58 m , which is larger than @xcite by one order of magnitude . readout electronics for the capacitive sensing will be similar to ones studied for lisa ( e.g. @xcite ) . for the standard resonant inductive - bridge scheme discussed in @xcite , the sources of disturbances due to readout electronics can be classified into two categories : imperfections in the capacitance bridge and the electric noise in the detecting circuit . the former includes fluctuations of inductance imbalance ( @xmath261 ) , fluctuations of mutual inductance imbalance ( @xmath262 ) , bias oscillator relative amplitude noise ( @xmath263 , where @xmath113 is the 100khz bias voltage capacitively applied to the proof mass by injection electrodes on insensitive faces ( the @xmath264 and/or @xmath265 faces ) of the proof mass ; the sensing electrodes are grounded ) and bias oscillator phase noise ( @xmath266 ) . the latter includes current noise and thermal noise . using the expressions for these sources of the disturbances given in table 1 of @xcite , the main contributions to the acceleration disturbance at 0.1mhz for astrod i are from @xmath267 ( @xmath108 1.5 @xmath12 10@xmath9ms@xmath1hz@xmath2 ) and @xmath266 ( @xmath108 9.1 @xmath12 10@xmath208ms@xmath1hz@xmath2 ) , and the thermal noise ( @xmath108 3.8 @xmath12 10@xmath260ms@xmath1hz@xmath2 ) and the current noise ( @xmath108 1.3 @xmath12 10@xmath260ms@xmath1hz@xmath2 ) . contributions from fluctuations of @xmath268 and @xmath262 are insignificant . in the estimation , we use the following parameters based on experimental results reported in @xcite : @xmath269 , @xmath270 , @xmath271hz@xmath2 , @xmath272hz@xmath2 and the turn ratio of the transformer @xmath215 = 1 ( 40 turns , inductance of 5mh and a quality factor of 165 ) , and a somewhat relaxed residual imbalance of the bridge @xmath273 . this relaxation has resulted in the dominant contributions from the bias oscillator relative amplitude noise and the phase noise . in @xcite , the thermal noise is dominant as they use @xmath274 in their estimation . the rss of these disturbances is @xmath77 @xmath108 1.8 @xmath12 10@xmath9ms@xmath1hz@xmath2 at 0.1mhz and listed in table [ table : pm acceleration disturbances ] . even when the sensing voltages are not applied on the electrodes , differences in local surface properties of the electrodes and the proof mass could lead a potential difference , patch - field voltage , between them @xcite . the charge fluctuations @xmath94 result in acceleration disturbance @xmath78 ( table [ table : pm acceleration disturbances ] ) through the patch field @xcite . by taking the average patch - filed voltage difference between opposing electrodes as @xmath140=0.1v @xcite , we obtain @xmath78 = 2.9 @xmath12 10@xmath275ms@xmath1hz@xmath2 at 0.1mhz . this is the dominant contribution to the total acceleration disturbance of the astrod i proof mass . the lisa study team is investigating the possibility of measuring and compensating voltage differences across capacitors , during the mission commission process , to considerably better than 0.01v @xcite . dielectric losses are thought to stem from surface contamination of electrodes and produces thermal voltage noise @xcite : @xmath276 where @xmath119 is loss angle . the upper limit of @xmath119 is reported to be @xmath91 for al electrodes @xcite . for @xmath277 , this voltage noise is about 6.6 @xmath58v hz@xmath2 at 0.1mhz and produces acceleration disturbance @xmath79 ( see table [ table : pm acceleration disturbances ] for the expression ) , through residual dc bias voltage on electrodes , in the sensitive axis @xcite . by making the same assumption of the average potential difference @xmath278v between a given electrode and the proof mass as @xcite , we obtain @xmath79 @xmath108 1.6 @xmath12 10@xmath9ms@xmath1hz@xmath2 . in the error estimates for lisa @xcite , @xmath279v is used . by adding in quadrature , the rss of the proof - mass environmental acceleration disturbances ( @xmath60 ) and the sensor back - action acceleration disturbances ( @xmath61 ) are 5.2 @xmath12 10@xmath9ms@xmath1hz@xmath2 and 3.0 @xmath12 10@xmath275ms@xmath1hz@xmath2 , respectively . therefore , the total direct proof - mass acceleration disturbance ( @xmath32 ) is 3.0 @xmath12 10@xmath275ms@xmath1hz@xmath2 at 0.1mhz , dominated by the sensor back - action acceleration disturbances . the stiffness @xmath27 is considered to stem from gravity gradients ( @xmath280 ) , fluctuations in sensing capacitance and capacitance gradients ( @xmath281 , @xmath282 and @xmath283 ) , bias voltage ( @xmath284 ) , patch field voltage ( @xmath285 ) and magnetic field gradients ( @xmath286 and @xmath287 ) . table [ table : stiffness ] gives a summary of estimated values for these sources of stiffness . the expressions used in the estimations are briefly described below . a detailed description on deviations of the expressions for @xmath280 , @xmath281 , @xmath282 , @xmath283 and @xmath288 is given in appendix a of @xcite in text is noted as @xmath284 in @xcite . ] . for a given disturbing point mass ( @xmath136 ) at a distance @xmath138 from the center of mass of the proof - mass on the sensitive axis , the amplitude of the acceleration disturbance of the proof - mass caused by a positional fluctuation @xmath289 is @xmath290 , where @xmath280 is given in table [ table : stiffness ] . making the same assumption of @xmath136 = 0.03 kg and @xmath138 = 0.05 m as @xcite , we obtain @xmath280 @xmath108 3.2 @xmath12 10@xmath8s@xmath1 . for a more detailed analysis , the identification of the disturbing mass is necessary . gravitational modeling for astrod i is in progress @xcite . this disturbance arises from any positional fluctuation and is different from the gravity gradient caused by thermal distortion or motion ( @xmath72 ) , which was discussed earlier . the expressions for @xmath281 , @xmath282 and @xmath283 can be obtained in the similar way as @xmath73 , @xmath74 , @xmath75 and @xmath76 in the previous section , under the following assumptions @xcite : fluctuations in the capacitances @xmath248 and @xmath249 and their derivatives produce disturbing forces in the @xmath138-direction , but not the fluctuations in other capacitances or their derivatives ; we ignore the cross - coupling effects . also , we assume @xmath291 = @xmath292 and @xmath293 = @xmath294 . @xmath281 is due to the fluctuations in the coulomb interaction between the charged proof mass and the image charges on the surrounding electrodes . @xmath282 arises from interaction between the net free charge @xmath92 on the proof - mass and the average electrode voltages . @xmath283 is due to the applied voltages across electrodes and the voltage difference across opposite electrodes . the estimated values for @xmath281 , @xmath282 and @xmath283 are given in table [ table : stiffness ] . by assuming that the sensing electrodes are grounded , the dominant term of the readout stiffness along the sensitive axis is given by @xmath284 @xcite , listed in table [ table : stiffness ] . assuming , as per @xcite , that the proof mass is biased to @xmath295v , @xmath284 @xmath108 3.1 @xmath12 10@xmath7s@xmath1 . by assuming a nominal patch - field voltage of @xmath140 = 0.1v @xcite and an overall multiplicative factor of @xmath4 = 5 as per @xcite , the contribution due to the patch field to the proof mass - spacecraft coupling is @xmath285 @xmath108 1.2 @xmath12 10@xmath196s@xmath1 ( table [ table : stiffness ] ) . the magnetic stiffness is given by @xmath296 = @xmath297 $ ] @xcite . the expressions for dominant terms of the stiffness due to the induced magnetic moment and the remanent moment are given by @xmath286 and @xmath287 ( as listed in table [ table : stiffness ] ) , respectively , where @xmath298 = 1.7 @xmath12 10@xmath120tm@xmath1 for @xmath193 = 0.75 m . their contributions to the total stiffness are insignificant . these contributions to the coupling constant @xmath27 are summarized in table [ table : stiffness ] . the rss of the coupling constant is 3.1 @xmath12 10@xmath7s@xmath1 . this is slightly below the requirement for the total stiffness in lisa ( 4 @xmath12 10@xmath7s@xmath1 @xcite ) . @llll [ @xmath186 @xmath299 & & expressions & astrod i + @xmath280 & gravity gradients & @xmath300 & 32 + @xmath281 & image charges & @xmath301 & 0.66 + @xmath282 & @xmath302 & @xmath303 & 0.40 + @xmath283 & applied voltages & @xmath304 & 0.095 + @xmath284 & bias voltage & @xmath305 & 3.1 @xmath12 10@xmath306 + @xmath285 & patch fields & @xmath307 & 1.2 + @xmath286 & induced magnetic moments & @xmath308 & 9.0 @xmath12 10@xmath120 + @xmath287 & magnetic remanent moments & @xmath309 & 6.9 @xmath12 10@xmath152 + & 3.1 @xmath12 10@xmath306 + we have estimated values for the coupling constant @xmath27 , the direct spacecraft acceleration disturbance @xmath28 and the direct proof - mass acceleration disturbance @xmath32 . by using the expression for the total acceleration disturbance of @xmath33 ( equation ( [ eq : control - model ] ) ) , we infer the requirements for the readout sensitivity @xmath22 and the spacecraft control - loop gain @xmath38 . in this process , we allocate an identical magnitude @xmath310 to each term of the expression ; @xmath311 . for astrod i , the noise goal is @xmath0ms@xmath1hz@xmath2 at 0.1mhz and , therefore , @xmath312ms@xmath1hz@xmath2 . the second term of the expression is @xmath32 , which was estimated to be 3.0 @xmath12 10@xmath275ms@xmath1hz@xmath2 ( table [ table : pm acceleration disturbances ] ) ; it is about a factor of 2 smaller than the allocated requirement . with the estimated total stiffness @xmath313s@xmath1 , we obtain @xmath314mhz@xmath2 from the first term and @xmath315 from the last term of the expression . from figure [ fig : curve ] , one can see that the acceleration noise spectral density requirements for astrod i take its lowest value of about 0.4 @xmath12 10@xmath10ms@xmath1hz@xmath2 at 0.3mhz . at this frequency , @xmath310 becomes 2.3 @xmath12 @xmath316ms@xmath1 hz@xmath2 . therefore , the requirement for @xmath22 becomes more stringent at 0.3mhz : @xmath317mhz@xmath2 . as for the second term of @xmath318 , @xmath32 is smaller at higher frequencies @xcite . our estimate of @xmath32 is dominated by the contribution from @xmath78 , which scales as @xmath319 . therefore , at 0.3mhz , @xmath32 would be @xmath11 1 @xmath12 10@xmath275ms@xmath1hz , which is smaller than @xmath320 at 0.3mhz . the last term scales as @xmath321 , and @xmath28 is expected to be smaller at higher frequencies because of the @xmath322 dependence of the fractional fluctuation in solar irradiance ( figure 6 of @xcite ) . therefore , by assuming the same level of the thruster noise , the requirement for @xmath38 at 0.3mhz would be less stringent than that at 0.1mhz by a factor of @xmath11 4 . in summary , the requirements for the readout sensitivity and the control loop gain for astrod i are @xmath317mhz@xmath2 and @xmath323 , respectively . main relaxed parameter - values are listed in table [ table : comparison with lisa ] in comparison with lisa . the values for lisa are quoted from the current error estimates by stebbins et al . @xcite , except the thruster noise quoted from @xcite . recently the lisa requirements for the thruster noise and the residual gas pressure have been relaxed further from the values given in table [ table : comparison with lisa ] ( this is a comment from one of the referees ) . @lll @xmath3 = 0.1 mhz & astrod i & lisa + maximum charge build - up : @xmath92 [ c ] & @xmath93 & @xmath0 + magnetic susceptibility : @xmath95 & @xmath96 & @xmath324 + magnetic remanent moment : @xmath97 [ a m@xmath15 ] & 1 @xmath12 10@xmath7 & 2 @xmath12 10@xmath8 + residual gas pressure : @xmath35 [ pa ] & @xmath91 & @xmath325 + electrostatic shielding factors : @xmath98 & 10 & 100 + thruster noise [ @xmath58n hz@xmath2 ] & 10 & a few + fluctuation of temperature in sc : @xmath103 [ k hz@xmath2 ] & 0.2 & 0.004 + voltage difference between average voltage across & & + opposite faces and voltage to ground : @xmath114 [ v ] & 0.05 & 0.01 + voltage difference between opposing faces : @xmath115 [ v ] & 0.01 & 0.005 + fluctuation of voltage difference across opposite faces : & & + @xmath116 [ v hz@xmath2 ] & @xmath117 & @xmath91 + residual dc bias voltage on electrodes : @xmath118 [ v ] & 0.1 & 0.01 + @xmath326 according to one of the referees , this lisa requirement has been relaxed to 10@xmath120 pa . we have estimated the spacecraft acceleration disturbance @xmath28 ( section [ ans ] ) , the proof - mass acceleration disturbances @xmath32 ( sections [ anp ] and [ back - action ] ) and the stiffness @xmath27 between the spacecraft and the proof mass ( section [ stiff ] ) . by using the expression ( [ eq : control - model ] ) , we have inferred the requirements for the readout sensitivity @xmath22 and the control - loop gain @xmath38 ( section [ st : requirements ] ) . table [ table : disturbances and requirements ] provides a summary of the estimated acceleration disturbances and the stiffness ( section ( b ) ) , and the requirements ( section ( c ) ) . the estimated total acceleration disturbance @xmath33 at 0.1mhz ( section ( a ) of table [ table : disturbances and requirements ] ) is about 13 % less than the noise goal of @xmath0ms@xmath1hz@xmath2 . we have compared the parameter values used in the estimation with lisa in section [ comparison with lisa ] . @ll & astrod i + + @xmath33 [ m s@xmath1 hz@xmath2 ] & 8.7 @xmath12 10@xmath275 + + & @xmath327 + direct pm acceleration disturbances : & + @xmath32 [ m s@xmath1 hz@xmath2 ] & 3.0 @xmath12 @xmath316 + direct sc acceleration disturbances : & + @xmath328 [ m s@xmath1 hz@xmath2 ] & 2.8 @xmath12 @xmath329 + pm - sc spring constant : & + @xmath27 [ s@xmath1 ] & 3.1 @xmath12 @xmath125 + + position readout noise : @xmath22 [ m hz@xmath2 ] & 7.4 @xmath12 @xmath329 + control - loop gain : @xmath38 & 3.8 @xmath12 @xmath331 + the total direct acceleration disturbance of the proof mass ( @xmath32 ) at 0.1mhz was estimated to be nearly a factor of two smaller than the requirement . this @xmath11 50% margin may be allocated for unknown disturbances or disturbances that would arise but have not been studied yet . these unestimated disturbances would be originated from , for instance , cross - talks in the capacitive sensing and magnetic damping of the proof mass . an estimate of acceleration disturbance due to magnetic damping for lisa is about 2 @xmath12 10@xmath208ms@xmath1hz@xmath2 at 0.1mhz @xcite . the contribution from the magnetic damping effect to astrod i would be in the similar order and insignificant . the sensor back - action acceleration disturbances can be reduced by increasing the magnitude of the gap @xmath109 . the total stiffness @xmath27 would be also reduced , for example , by a factor of four by changing the gap to 4 mm . the optimum design for the capacitive sensing is to be discussed based on results from the ongoing laboratory torsion balance experiment for astrod i @xcite . parameter values we used in this paper are mainly based on the results of studies done for lisa and lisa pathfinder . this may be sufficient for the preliminary estimation . more accurate estimation would be obtained by carrying out the following works dedicated for astrod i : ( a ) modeling local magnetic fields of the spacecraft , ( b ) estimating the effective charging rate of the proof - mass , ( c ) estimating the cosmic - ray impact rate of the proof - mass , ( d ) estimating the micrometeorite impact effects , ( e ) thermal modeling of the proof - mass housing and the spacecraft , ( f ) gravitational modeling that includes thermal and non - thermal deformation of the spacecraft and the payload , ( g ) electrostatic modeling for the capacitive sensors and ( h ) estimating environmental factors ( such as the interplanetary magnetic field , solar wind , solar radiation and cosmic rays ) in the varying orbit ( 0.5au to 1au ) . simulations to estimate charging rates for astrod i are in progress @xcite . we have tentatively estimated acceleration disturbances for astrod i. this work has allowed us to set preliminary requirements for astrod i. to improve the current estimation , the disturbances that have not been studied yet have to be included and more detailed modeling works are necessary for astrod i. in comparison with lisa , requirements for astrod i can be largely relaxed . this will make the technological developments for astrod i less demanding to meet the drag - free requirements . this work was funded by the national science council and the foundation of minor planets of purple mountain observatory . we thank a. rdiger , s. vitale , d. k. gill , a. pulido patn , and , especially , the referees for useful information on acceleration disturbances and helpful comments on the manuscript . 00 bec - borsenberger a _ et al _ 2000 _ astrod _ esa f2/f3 mission proposal ; and references therein ni w - t 2002 _ int . j. mod . d _ * 11 * 947 ; and references therein ni w - t _ et al _ 2002 _ int . j. mod . d _ * 11 * 1035 ; and references therein ni w - t _ et al _ 2003 astrod i : mission concept and venus flybys , _ proc . 5th iaa intnl . conf . on low - cost planetary missions , estec , noordwijk , the netherlands , 24 - 26 september 2003 _ , esa sp-542 , pp . 79 - 86 , november 2003 ; _ acta astronaut . _ in press , 2006 xia y , ni w - t , tang c j and li g 2006 orbit design and orbit simulation for astrod i , _ gen . relat . gravit . _ * 38 * , in press anza s _ et al _ 2005 _ class . quantum grav . _ * 22 * s125 2000 esa - sci(2000)11 hammesfahr a _ et al _ 2000 _ lisa - study of the laser interferometer space antenna , final technical report _ esa industrial study at phase a , estec contract no . 13631/99/nl / ms , dss report no . li - rp - ds-009 vitale s _ et al _ 2005 _ science requirements and top - level architecture definition for the lisa technology package ( ltp ) on board lisa pathfinder ( smart-2 ) _ ltpa - utn - ssrd - iss003-rev1 june 30 , 2005 ni w - t , shiomi s and liao a - c 2004 _ class . quantum grav . _ * 21 * 641 vitale s and dolesi r 2000 cp523 _ gravitational waves : third edoardo amaldi conference _ , ed meshkov s , american institute of physics 1 56396 - 944 - 0 231 vitale s _ et al _ 2002 _ nucl . b ( proc . suppl . ) _ * 110 * 209 schumaker b l 2002 overview of disturbance reduction requirements for lisa http://elmer.tapir.caltech.edu/ph237/week18/drs-bls01may02.pdf schumaker b l 2003 _ class . quantum grav . _ * 20 * s239 pap j _ et al _ 1999 _ adv . space res . _ * 24 * , no.2 215 fichter w , gath p , vitale s and bortoluzzi d 2005 _ class . quantum grav . _ * 22 * s139 jackson j d 1998 _ classical electrodynamics _ , 3rd ed . , john wiley & sons , inc . hanson j _ et al _ 2003 _ class . quantum grav . _ * 20 * s109 lang k r 2001 _ the cambridge encyclopedia of the sun _ , cambridge university press balogh a _ et al _ 1992 _ astron . * 221 shaul d n a _ et al _ 2005 _ class . quantum grav . _ * 22 * s297 silvestri z _ et al _ 2003 _ metrologia _ * 40 * 172 - 176 stebbins r t _ et al _ 2004 _ class . quantum grav . _ * 21 * s653 gill d k , private communication , january 2006 jafry y and sumner t j 1997 _ class . quantum grav . _ * 14 * 1567 arajo h m _ et al _ 2004 detailed calculation of test - mass charging in the lisa mission airxiv : astro - ph/0405522 v3 7 oct 2004 , published in _ astroparticle physics _ * 22 * 2005 451 bao g _ et al _ 2006 astrod i charging simulation and disturbances _ gen . gravit . _ * 38 * , in press blaser j p , lockerbie n , paik h j , speake c and vitale s 1996 _ class . quantum grav . _ * 13 * a203 touboul p , rodrigues m , willemenot e and bernard a 1996 _ class . quantum grav . _ * 13 * a67 nobili a m _ et al _ 2001 _ phys . * 63 * 101101(r ) worden p , mester j and torii r 2001 _ class . quantum grav . _ * 18 * 2543 rodrigues m , foulon b , liorzou f and touboul p 2003 _ class . quantum grav . _ * 20 * s291 bender p 2003 _ class . quantum grav . _ * 20 * s301 rdiger a 2002 residual gas effects in space - borne position sensors http://eotvos.dm.unipi.it/nobili/opendiscussion/ruediger_30aprile.pdf dolesi r _ et al _ 2003 _ class . quantum grav . _ * 20 * s99 speake c c and andrews p l 1997 _ class . quantum grav . _ * 14 * 1557 cavalleri a _ et al _ 2001 _ class . quantum grav . _ * 18 * 4133 weber w j _ et al _ 2003 _ proc . spie conf . on astronomical telescopes and instrumentation _ * 4856 * ed cruise m and saulson p ( bellingham wa : spie optical engineering press ) 31 speake c c 1996 _ class . quantum grav . _ * 13 * a291-a297 vitale s and speake c c 1998 _ 2nd lisa symposium aip proceedings _ * 456 * , 172 speake c c , davis r s , quinn t j and richman s j 1999 _ phys . a _ * 263 * 219 shiomi s 2005 acceleration disturbances due to local gravity gradients in astrod i , _ proceedings of the 6th edoardo amaldi conference on gravitational waves _ ( okinawa , japan , june 20 - 24 , 2005 ) ; 2006 _ j. phys.:conf . ser . _ * 32 * 186 zhou z b , gao s w and luo j 2005 _ class . quantum grav . _ * 22 * s537

several problems in engineering and in particular signal and image processing necessitate to estimate binary vectors corrupted by some noise and can be simply addressed using the least squares principle under binarity consraints . the resulting problem is a minimization of a quadratic form over @xmath0 , a problem which is known to be @xmath1-hard in general . one of the main approaches to relax this problem into a convex one is the semi - definite programming relaxation which has been extensively used in classification , pattern recognition and communication systems . some of the main achievements in the study of the sdp relaxation were obtained by goemans and williamson @xcite and @xcite . however , solving a semidefinite program in practice relies on interior point methods which although enjoying nice theoretical convergence properties are limited to problems of size up to 500 @xmath2 . on the other hand , very pratically efficient bundle methods are available for the eigenvalue relaxation of the same binary quadratic optimization problems . we refer the reader to @xcite for a discussion of the pratical superiority of bundle methods for solving certain semi - definite programs such as the ones appearing in the present paper . despite this empirical fact in favor of the eigenvalue relaxation , one of the main reasons most users prefer the sdp relaxation is that good primal binary solutions can be recovered using goemans and williamson s randomized algorithm . the main motivation of the present paper is to show how a solution of the sdp can be recovered from a solution of the eigenvalue relaxation . as a by product , a new geometric interpretation of the randomized algorithm is proposed . penalized binary least squares estimation problems are problems of the form @xmath3 where the vector @xmath4 is the observed data , the matrix @xmath5 represents the `` filter '' , the vector @xmath6 is the signal , or parameter vector , that has to be estimated , and the term @xmath7 is a penalization term that can often be interpreted as an _ a priori _ information in terms of bayesian statistics . this problem belongs to the larger class of minimization of quadratic forms over binary vectors which is known to be @xmath1-hard . much work has been devoted to constructing semi definite programming ( sdp ) based relaxations for general quadratic binary problems . semi - definite programs are linear optimization problems over symmetric matrices with real coefficients and with the additional convex constraint of positive semidefiniteness ; see for instance @xcite or @xcite for excellent introductions to convex programming and in particular sdp . sdp methods have already played an important role in various topics inside signal processing problems and we refer to @xcite for a nice survey on possible applications . a common feature of essentially all the existing relaxations is that they can be obtained using lagrange duality which is a general methodology for obtaining lower bounds to hard minimization problems , as overviewed in @xcite and @xcite . the goal of the paper is to survey what is known about another lagrangian duality based relaxation , namely the eigenvalue relaxation , for this problem . this relaxation was first proposed by delorme and poljak @xcite for the max - cut problem . see also the work of poljak , rendl and wolkowics @xcite for more details . the main advantage of the eigenvalue relaxation over the sdp relaxation is that the eigenvalue relaxation can be solved much faster than the sdp relaxation , as reported for instance in @xcite , @xcite , @xcite and @xcite . this remarkable computational tractability of the eigenvalue relaxation is the main motivation for writing this detailed survey . the content of the paper is as follows . the second section is devoted to a rapid presentation of the relaxation and its relationship with lagrangian duality . we also recall a simple and well known certificate for exactness of the relaxation , i.e. the fact that a globally optimal binary solution is obtained . the third section details the relationships between the semi - definite relaxation and the eigenvalue relaxation . the main result of this section is the following : a solution of the sdp relaxation can be recovered from the solution of the eigenvalue relaxation . the case of inexact solutions to the eigenvalue relaxation is also studied . the forth section deals with the problem of recovering binary primal solutions from the dual scheme . we first give sufficient conditions under which strong duality holds and the eigenvectors of norm @xmath8 associated to the maximum eigenvalue at optimality are binary solutions . next , in the case where strong duality does not hold , we show that goemans and williamson s randomized algorithm has a very natural meaning when viewed in terms of the optimal eigenspace associated to the maximal eigenvalue in the eigenvalue relaxation . in the last section , we propose simulation experiments in the case of binary image denoising and cdma multiuser detection problems . the first of these problems has been previously approached by stochastic methods based on markov chains like simulated annealing and metropolis hastings schemes ; see for instance @xcite and the more recent work of gibbs @xcite . the approach discussed here was presented in @xcite . recently a lot more problems have been addressed using the sdp relaxation in @xcite . the results obtained so far are quite encouraging and the approach performs well on very dirty images . we prove hat strong duality holds for the immage denoising problem , thus recovering back the polynomial solvability result of greig , porteous and seheult as a special case and by a very different path . passing to the second problem , our monte carlo experiments show that the average computational effort for solving the eigenvalue relaxation as a function of the number of users grows slowlier than for the sdp relaxation with the standard implementations available with scilab . * notations*. in the sequel we will use the following notations . the inner product on @xmath9 is denoted by @xmath10 , the set of real symmetric matrices of order @xmath11 are denoted by @xmath12 . the partial order @xmath13 denotes the loewner ordering , i.e. for @xmath14 and @xmath15 in @xmath12 , @xmath16 means that @xmath17 is positive semidefinite . for a set @xmath18 in @xmath9 , @xmath19 denotes the convex hull of @xmath18 and @xmath20 denotes its closure . for a matrix @xmath14 in @xmath12 , @xmath21 denotes its diagonal vector and for @xmath22 in @xmath9 , @xmath23 denotes the diagonal matrix whose diagonal vector is @xmath22 . if an equation number @xmath24 corresponds to an optimization problem , then @xmath25 will denote the optimum value for this problem . we first introduce the eigenvalue relaxation and at the same time , we propose a quick refresher on lagrangian duality , collecting all the results that will play an essential role in the sequel . the proofs of almost all the results presented here can be found in @xcite . the binary least - squares estimation problem is in fact equivalent to the homogenized problem @xmath26 x \text { s.t . } x\in \{-1,1\}^{n+1}. \hspace{2 cm } { \rm ( bls ) } \nonumber\ ] ] indeed , if we add the constraint @xmath27 in ( bls ) , we obtain exactly the binary least squares problem . now , if @xmath28 is a solution of ( bls ) , then @xmath29 is again a solution of of ( bls ) , thus adding the constraint @xmath27 is in fact redundant , which proves the claimed equivalence . set @xmath30 . \nonumber\ ] ] notice further that the constraint @xmath31 is equivalent to @xmath32 for all @xmath33 . thus , to problem ( bls ) , we can associate the lagrangian function @xmath34 now we can add to the problem the implicit spherical constraint @xmath35 which is redundant with the binary constraints . then , optimizing over this sphere , we obtain the lagrangian dual function , i.e. @xmath36 which , using raleigh - ritz variational formulation of the largest eigenvalue of symmetric matrices , can be written @xmath37 finally , the dual problem , i.e. the eigenvalue relaxation , is given by @xmath38 it is important to notice first that the dual function @xmath39 is convex , since it is the maximum over a family parametrized by @xmath40 of linear functions in the variable @xmath41 . the main classical property of the lagrangian dual is weak duality , i.e. @xmath42 where @xmath43 denotes the optimal value . this property explains in part why lagrange duality is used : it provides a bound on the primal optimal value . when equality holds in the weak duality property , we say that strong duality holds . sometimes , like in the case of the max - cut problem , the bound can be proved to be proportional to the optimal original value . more precisely , goemans and williamson proved that the optimum value of the eigenvalue relaxation ( in fact the equivalent sdp formulation ; see the original paper and section [ eqsdp ] below ) is greater than or equal to the optimal original value ( this is just weak duality ) , which itself is always greater than or equal to .876 times the eigenvalue relaxation s optimal value . a quite similar but less tight bound , proved by nesterov applies directly to the present problem . we will recall this bound in section [ gwn ] below . it is well known that there exists an optimal dual solution . this was proved by poljak and wolkowicz in @xcite . the proof given here is more direct . the dual function admits a minimizer . * proof*. let @xmath44 . make the change of variable @xmath45 , i.e. define @xmath46 we now have the property that @xmath47 . we prove that @xmath48 is coercive . take any sequence @xmath49 with @xmath50 as @xmath51 . we can assume that @xmath52 for some @xmath53 because otherwise , the fact that @xmath50 implies that there must exists a sequence @xmath54 with @xmath55 and the fact that @xmath47 gives a contradiction . now , the gershgorin circle around the diagonal element @xmath56 has a constant radius , say @xmath57 and its center goes to @xmath58 . since @xmath59 , this implies that @xmath60 . thus @xmath48 is coercive and since it is continuous , it admits a minimizer that we will denote by @xmath61 . now , for all @xmath62 , @xmath45 , we have @xmath63 but , on the other hand , @xmath64 . therefore , @xmath65 and the proof is complete . @xmath66 the subdifferential @xmath67 of the eigenvalue relaxation has been much studied . recall that for any convex function @xmath68 , the subdifferential @xmath69 is defined by @xmath70 the analysis of @xmath67 is based on the following general theorem . [ eigsubd]@xcite let @xmath71 be an affine operator defined by @xmath72 for some linear operator @xmath73 and some matrix @xmath74 . then , we have @xmath75 with @xmath76 where @xmath77 is the adjoint of @xmath78 , @xmath79 denotes the multiplicity of @xmath80 at @xmath81 and @xmath82 is a matrix whose columns form any orthonormal basis of the eigenspace of @xmath83 associated to @xmath80 . now , if we set @xmath84 , we get @xmath85 , @xmath86 and @xmath87 . for @xmath88 , let @xmath89 be defined by @xmath90 using the previous theorem , we obtain [ subdtheta ] the subdifferential @xmath67 of the dual function @xmath91 is given by @xmath92 following oustry @xcite , the formula for @xmath93 in theorem [ eigsubd ] is proved by showing that the maximum eigenvalue function @xmath94 on @xmath12 is nothing but the support function @xmath95 of @xmath96 , defined by @xmath97 with the scalar product defined by @xmath98 . by definition , the face @xmath99 of @xmath96 exposed by @xmath83 is the set of maximizers in ( [ support ] ) , i.e. @xmath100 knowing that the subdifferential of a support function of a set is exactly the exposed face of this set , we finally get @xmath101 the formula follows after some linear algebra . there is a different path to the subdifferential s formula , which is perhaps more _ a propos _ in the context of duality : it is proved in ( * ? ? ? * chapter xii ) that @xmath102 where @xmath103 denotes the closure of the convex hull . this fact is in fact true for general continuous constrained problems in the case where the underlying space is compact ( for example ) ] and the associated technical condition is called the _ filling property_. the following proposition provides a useful sufficient condition for proving that the relaxation is exact , i.e. strong duality applies . [ multone ] let @xmath104 be a minimizer of the dual eigenvalue relaxation . then , if @xmath105 has multiplicity one , then @xmath106 and any eigenvector @xmath107 of @xmath108 whose squared norm is @xmath8 is a binary solution of ( bls ) . the proof is a direct consequence of ( * ? ? ? * theorem xii.2.3.4 . ) . we provide a specialized proof here because it is short and instructive . * proof*. since the multiplicity of @xmath105 is one , the subdifferential of @xmath109 at @xmath104 is a single vector . thus , @xmath91 is differentiable at @xmath104 and its gradient is simply @xmath110 for any @xmath28 in @xmath111 such that @xmath112 . since , @xmath104 minimizes @xmath91 , we must have @xmath113 . this implies that @xmath114 for all @xmath33 . thus , using weak duality @xmath115 which proves that @xmath28 solves the original problem ( bls ) . @xmath66 we now have a nice criterion for deciding whether our relaxation was exact and if so , we also know how to recover a binary solution from an optimal eigenvector . this approach works for any quadratic binary problem and is extensively used for approximating combinatorial problems . however , the question remains on what to do when the relaxation is not exact , i.e. when the multiplicity at the optimum is greater than one . the next two sections will help answer this crucial question . the purpose of the next two sections is to describe how to recover primal binary solutions from the eigenvector solutions of the dual eigenvalue problem . it was first shown that good binary solution can be generated at random using the sdp solution by goemans and williamson @xcite in the case of the max - cut problem in graph theory . their results were then extended by nesterov to the case of indefinite quadratic binary programming @xcite . those results allowed to conclude that both eigenvalue and sdp relaxations are in a certain precise sense very efficient . however , both relaxations are not equivalent from the computational point of view . recall that one of the main motivations for using the eigenvalue relaxation is its manageable practical complexity which is often favorable compared to the one of solving the sdp relaxation . but what is not clear is how to generate good ( primal ) binary solutions in average with the eigenvalue relaxation only ? the first natural approach to this question is of course to try and recover an optimal sdp solution from the eigenvalue relaxation . thus , we devote this section to this problem . it can be solved as follows : an appropriate convex combination of rank one matrices obtained from a set of optimal eigenvectors is shown to be a solution we are looking for . our approach simplifies the presentation of @xcite . the adaptation of the randomized algorithm of goemans and williamson and the associated bound established by nesterov will be discussed in the next section . in order to obtain the semi - definite programming ( sdp ) relaxation of the the homogenized problem ( bls ) , we begin with the following equivalence relating our problem to a problem on symmetric matrices . we have ] @xmath116 this last problem is itself equivalent to @xmath117 this problem being nonconvex , we drop the rank constraint and obtain the following sdp ( convex ) relaxation @xmath118 whose value is obviously greater than or equal to @xmath119 . an important result of pataki ( * ? ? ? * theorem 2.1 ) gives a bound on the rank of solutions to semi - definite programs . in the case of our semi - definite relaxation , this theorem implies that the rank @xmath120 of an optimal matrix @xmath121 satisfies @xmath122 . it follows from the subdifferential s formula given in corollary [ subdtheta ] that at any minimizer @xmath104 , we have @xmath123 suppose we have in hand a matrix @xmath124 such that @xmath125 it appears that a good guess for a candidate solution @xmath121 to the sdp relaxation in the general case is @xmath126 we just need to check the details to see how it works . this result was initially proved in @xcite but the proof given here is more direct . @xcite [ equiv ] let @xmath104 be the optimal solution of the eigenvalue relaxation let @xmath82 be a matrix whose columns for an orthonormal basis of the eigenspace associated to @xmath105 and let @xmath127 be as in ( [ optz ] ) . then the matrix @xmath128 is an optimal solution of the sdp relaxation . we would like to underline at this point that a more elegant proof of the theorem could be obtained using conic duality but we preferred to keep on with elementary arguments since this is possible in the present context . * proof*. compute the eigenvalue / eigenvector decomposition @xmath129 , set @xmath130 , @xmath131 , let @xmath57 be the multiplicity of @xmath108 and let @xmath132 denote the columns of @xmath133 . recall that from the definition of @xmath127 , we have @xmath134 . then , we get @xmath135 thus , @xmath136 using this fact , we obtain @xmath137 since @xmath134 . thus , the optimal value of the sdp is greater than or equal to the optimal value of the eigenvalue relaxation . on the other hand , it is well known that the optimal value of the eigenvalue relaxation is greater than or equal to the one of the sdp relaxation . we provide a proof here for the sake of completeness . let @xmath138 be an optimal solution to the sdp relaxation . now , for all @xmath41 in @xmath139 , we have @xmath140 by using the fact that @xmath141 . now , compute the eigenvalue / eigenvector decomposition @xmath142 and let @xmath80 be the greatest of these eigenvalues . then , @xmath143 since this is true for all @xmath41 , we obtain that the eigenvalue relaxation majorates the sdp relaxation . thus , both optimal values are equal and this completes the proof of the proposition . @xmath66 of course , it can be hard to find a matrix @xmath144 that works . we will now try to overcome this problem . we first have to specify how the subgradients are obtained in practice . at each point @xmath145 , choose an eigenvector @xmath107 of squared norm equal to @xmath8 associated to @xmath146 . then , using the alternative representation of the subdifferential ( [ subd2 ] ) , a subgradient of @xmath91 at @xmath41 is obtained by setting @xmath147^t$ ] . assume that we have a set of subgradients @xmath148^t\in \partial \theta(u^j)$ ] for some @xmath149 , @xmath150 and such that @xmath151 for some nonnegative @xmath152 s with @xmath153 . this can be performed for @xmath154 as small as we want by using a bundle method . such a method will construct in a finite number of iterations , say @xmath155 , an iterate @xmath156 and a family of @xmath149 s with the desired property , all of them lying in a small neighborhood of @xmath156 . this is one very nice feature of the bundle mechanism which is extensively described in ( * ? ? ? * volume ii ) . moreover , it is a well known fact , called caratheodory s theorem , that only @xmath157 subgradients are sufficient in the expression ( @xmath154opt ) . set @xmath158 then , we have the following result . for any @xmath159 , the matrix @xmath160 defined above satisfies @xmath161 * proof*. let @xmath104 be any minimizer of @xmath91 . then , for each @xmath150 , we have by the definition of the subdifferential @xmath162 but @xmath163 is given by @xmath164 on the other hand , since @xmath165 , @xmath166 thus , we obtain @xmath167 which implies , after multiplying by @xmath152 and summing over @xmath150 @xmath168 using cauchy - schwartz inequality , this gives @xmath169 since the eigenvalue and the sdp relaxation have equal optimal values , we finally obtain @xmath170 which implies the desired result . @xmath66 it is a common idea that the sdp relaxation contains more information than the eigenvalue relaxation . we hope that the results of this section managed to convince the reader that this is in fact not the case and a good approximate solution can be recovered quite easily using subgradient information at the optimum . we now are in position to answer our main question of how to recover a satisfactory although sometimes suboptimal primal binary solution . in the first part of this section , we show that optimal binary solutions can actually be exactly recovered using the eigenvalue relaxation , i.e. strong duality holds , under some simple conditions . then , in the case where the problem does not satisfy these necessary conditions for strong duality , we develop a randomized algorithm based on the optimal eigenspace of the maximum eigenvalue dual function and show that this procedure is equivalent to goemans and williamson s randomized algorithm for max - cut . this provides a new interpretation of goemans and williamson s procedure . we have the following theorem . [ strong ] for almost all @xmath14 in the sense of the lebesgue measure , such that @xmath171 is componentwise negative outside the diagonal . then the eigenvalue relaxation is exact , i.e. strong duality holds . * proof*. fix @xmath145 . let @xmath172 be the vector of the first @xmath11 components of @xmath41 . the fact that @xmath171 is componentwise negative outside the diagonal implies that @xmath173 is componentwise positive . thus , the perron - frobenius theorem implies that the maximum eigenvalue of @xmath173 has multiplicity one . from this , we deduce that the maximum eigenvalue of @xmath174 also has multiplicity one . let @xmath175 be an eigenvalue decomposition of @xmath176 , where we used the subscript @xmath172 in order to remember that whatever the chosen decomposition , it is a nonlinear and non necessarily continuous function of @xmath41 . moreover , since the maximum eigenvalue has multiplicity one , corollary 4 in @xcite says that it is possible to choose the eigenvector associated to the maximum eigenvalue as a continuously differentiable function of @xmath172 . we will denote by @xmath177 this eigenvector . using this parametrization , the matrix @xmath178+d(u ) \nonumber\ ] ] can be rewritten as @xmath179 \left [ \begin{array}{cc } d_{u_1^n } & -v_{u_1^n}^ta^ty \\ -y^ta v_{u_1^n } & y^ty+u_{n+1 } \end{array } \right ] \left [ \begin{array}{cc } v_{u_1^n } & 0 \\ 0 & 1 \end{array } \right]^t , \nonumber\ ] ] where all dimensions can easily be guessed from the previous knowledge on the involved submatrices . let @xmath180 be the codimension one differentiable submanifold defined by @xmath181 let @xmath182 be the optimal set defined by @xmath183 due to the representation @xmath184 the set @xmath182 is the projection onto the cartesian product @xmath185 of the set @xmath186 where @xmath187 is the upper bound of pataki ( see section [ sdp ] ) on the optimal rank of the sdp relaxation ( here @xmath188 for @xmath11 large ) and where @xmath189 is the set @xmath190 whose intersection with @xmath191 corresponds to the parameter set allowing for zero to belong to the subdifferential of the dual function @xmath91 in the case where @xmath192 . now , since the constraint @xmath193 is described by @xmath194 equations , @xmath195 by @xmath196 equations , @xmath197 by one equation and @xmath198 , the dimension of @xmath189 is greater than or equal to @xmath199 . furthermore , notice that since the eigenvalues are continuous fonctions of the entries of @xmath200 , the subset of @xmath201 for which @xmath192 is open in the topology induced by the ambiant space . therefore its projection set onto the cartesian product @xmath185 is of dimension at least @xmath202 which garantees that the projection onto the @xmath14-space @xmath203 of its intersection with @xmath180 is a set of null lebesgue measure . and thus , for almost all @xmath14 , such that @xmath171 is componentwise negative outside the diagonal , @xmath204 . using this result , theorem a about the interlacing property of the eigenvalues for arrow matrices in the appendix implies that the maximum eigenvalue of @xmath205 is greater than the maximum diagonal element of @xmath206 which nothing by @xmath207 and all @xmath11 other eigenvalues are less than @xmath207 . this implies that for allmost all @xmath14 , the maximum eigenvalue of @xmath205 has multiplicity one at the optimum , which implies that @xmath91 is differentiable at the optimum . therefore , using proposition [ multone ] we obtain that strong duality holds for allmost all @xmath14 such that @xmath171 is componentwise negative outside the diagonal . we start this section with some recalls on goemans and williamson s algorithm and nesterov s bound . the method relies on the cholesky factorization of the optimal solution @xmath121 of the sdp relaxation , @xmath208 from theorem [ equiv ] we see that @xmath209 where @xmath79 is the multiplicity of @xmath105 at the chosen corresponding solution @xmath104 of the eigenvalue relaxation . this factorization is important , since it allows to write @xmath210 where @xmath211 is the transpose of @xmath212 row vector of @xmath213 . let @xmath214 be a random variable with uniform distribution on the unit sphere in @xmath215 . [ sdpgw ] * ( goemans and williamson s algorithm ) * \1 . find the cholesky factorization @xmath216 . let @xmath217 be a random vector with uniform distribution on the unit sphere of @xmath218 . the random cut is defined by @xmath219 where the sign function is defined coordinate - wise . draw @xmath11 samples from @xmath220 , say @xmath221 , , @xmath222 and choose the sample giving the best value of the objective function @xmath223 . the key result is that , in average , the vector @xmath220 gives a good binary solution to the original problem . since the best sample will have greater cut value than the average with overwhelming probability , the above procedure should work well . this is made precise by nesterov s theorem . define @xmath224 and @xmath225 then , we have @xmath226}{f^*-f_*}\leq \frac2{\pi}.\ ] ] this result is remarkable despite the fact that the bound @xmath227 is rather large . an important issue for future research is to study such type of bounds for particular subclasses of problems in hope of improving nesterov s result . the main drawback of the former presentation is that using the uniform variable @xmath214 is quite hard to motivate from an optimization viewpoint . let us take a slightly different perspective . assume that we have a solution @xmath104 of the eigenvalue relaxation . as before , let @xmath82 be a matrix whose columns form an orthonormal bases of the eigenspace associated to @xmath105 . moreover , we may require that @xmath228 where @xmath229 is some diagonal matrix with @xmath230 , @xmath231 and @xmath232 . in the case where the multiplicity at the optimum is one , the optimal eigenbasis reduces to a unique vector and we saw in proposition [ multone ] that multiplying this vector by @xmath233 gives a binary solution . now let us turn to the case where there are @xmath234 eigenvectors . to each unit norm eigenvector @xmath235 , we associate a subgradient @xmath236^t$ ] . then , ( [ eigopt ] ) implies that @xmath237 [ opteig ] ( -1,-1)(2,2 ) ( 0,0)(0,.95 ) ( 0,0)(1.3,-.95 ) ( 0,0)(-1.42,-.77 ) ( 0,1)(1.4,-1 ) ( 1.4,-1)(-1.5,-.8 ) ( 0,1)(-1.5,-.8 ) ( 0,1.5)@xmath238 ( 1.9,-1)@xmath239 ( -1.6,-.3)@xmath240 ( .2,.15)@xmath241 now one natural strategy might be the following : pick the best eigenvector , i.e. the eigenvector @xmath242 whose associated coefficient @xmath243 in expression ( [ convcombzero ] ) is the _ greatest _ and round its coordinates to the nearest binary values . there is a second strategy : draw random linear combinations of the @xmath244 s giving preference to the components with higher associated coefficient in ( [ convcombzero ] ) . this can be done by sampling vectors of the type @xmath245 where the @xmath246 s are independent random variables with distribution @xmath247 . for each sample , a feasible solution is obtained by rounding off the components to the nearest binary . we sum up this procedure as follows . * ( randomized algorithm based on optimal eigenvectors ) * [ eiggw ] 1 . find the matrix @xmath82 whose columns form an orthonormal eigenbasis associated to @xmath248 such that ( [ convcombzero ] ) holds for some @xmath152 s satisfying @xmath231 and @xmath249 . let @xmath217 be a random vector with distribution @xmath250 . the random cut is defined by @xmath251 \3 . draw @xmath11 samples from @xmath220 , say @xmath252 , , @xmath253 and choose the sample giving the best value of the objective function @xmath254 . the important result is that this second strategy is equivalent to goemans and williamson s randomized procedure . procedure [ eiggw ] is equivalent to goemans and williamson s algorithm . * proof*. set @xmath255 . then theorem [ equiv ] and equation [ convcombzero ] imply that @xmath216 with @xmath256 , thus retrieving the cholesky factorization of @xmath121 . let @xmath257 . it is clear that @xmath214 has distribution @xmath258 . this proves that the cut @xmath220 obtained by procedure [ eiggw ] is exactly the output of goemans and williamson s procedure . @xmath66 the eigenvalue point of view thus allowed us to provide an alternative and geometric explanation for taking a random cut using a uniformly distributed variable on the sphere in goemans and williamson s methodology . in this section , we provide some results for the concrete problems of image denoising and show how this relaxation applies to the problem of multiuser detection in cdma systems . the first set of simulations is devoted to the denoising problem , in which @xmath14 is simply the identity matrix . this is the problem considered in @xcite , @xcite and @xcite for instance . the original binary image as 26 rows and 62 columns which gives a total number of 1612 variables . for this problem , the penalization matrix @xmath259 is chosen so as to smooth the image . this is achieved by requiring neighboring pixels to be similar in the sense that if @xmath53 and @xmath260 are indices of neighbor pixels , then , we would like the least square cost to be penalized by the quantity @xmath261 . thus , @xmath259 is the matrix associated to the quadratic form @xmath262 where @xmath263 denotes the property of being neighbor indices and the @xmath264 are nonnegative . the neighborhood of each pixel is usually chosen to be the north , south , east and west pixels . the following theorem is the main result of this section . for @xmath265 , the identity matrix and @xmath259 the matrix associated to the quadratic form ( [ reg ] ) , the eigenvalue relaxation is exact . * proof*. the eigenvalue relaxation of the optimization problem corresponding to this binary least square denoising problem is as before @xmath266 consider now the perturbed optimization problem @xmath267 where @xmath268 is negative outside the diagonal . since the @xmath264 are nonnegative , the matrix @xmath259 has only nonpositive off diagonal terms and thus , theorem [ strong ] proves that strong duality holds for this problem and there exists a binary eigenvector that achieves optimality . assume that @xmath268 is chosen so that @xmath269 . then , the optimum value @xmath270 of problem ( denoise ) and the optimum value @xmath271 of problem ( perturbed ) satisfy @xmath272 moreover , by weak duality , we have @xmath273 since strong duality holds for problem ( perturbed ) , denoting by @xmath274 a solution of @xmath275 we have @xmath276 therefore , we obtain @xmath277 which implies @xmath278 now , since @xmath0 is finite , the image @xmath279 of @xmath0 by the function @xmath280 is a finite set . let @xmath281 denote the closest number to @xmath282 in @xmath279 . now , choosing @xmath283 , we obtain @xmath284 which proves that the denoising problem is polynomial time solvable by solving problem ( perturbed ) . @xmath66 this theorem is to be compared with the results of d. m. greig , b. t. porteous and a. h. seheult @xcite which formulates the binary denoising problem as a minimization problem with cost given at the top of page 273 . the objective to be minimized in @xcite can be rearranged so as to minimize a linear cost with same penalization as the one given by ( [ reg ] ) . the main contribution of @xcite is to say that this problem can be solved in polynomial time using a network flow algorithm . notice that our proof works for @xmath285 and any additional linear term added to the penalized objective function to be optimized . since the eigenvalue relaxation can also be optimized in polynomial time , this confirms that the eigenvalue relaxation performs at least as good as previous approaches on a well known problem . on the other hand , the eigenvalue relaxation can be a flexible approach in more complicated cases where @xmath14 is not equal to the identify or other quadratic constraints have to be incorporated such as in @xcite . the experiments reported on below were performed for the case of quite noisy original images . the noise was taken to be additive , independent identically distributed and gaussian @xmath286 and was applied to the symmetrized image with pixel values in @xmath287 . in order to show the influence of the smoothing parameter @xmath288 , we displayed the percentage of misspecified bits vs values of @xmath288 . the recovered image is the one with the choice of @xmath288 giving the best percentage of bits recovered . we found the results very encouraging . indeed , even when the observed image is very noisy , we still recover an image which is readable . this suggested that an appropriate postprocessing might easily allow to recover the original written words , by comparing the letters to a given dictionary . cross validation can be used to estimate @xmath288 . we will not discuss this problem here . instead , it seems reasonable to argue that the choice of @xmath288 can just be made _ a posteriori _ since it consists of tuning the method until a satisfactory solution is obtained . this reduces the hard combinatorial initial problem to a simpler one parameter knobing procedure . the displayed experiment and the numerous simulations not presented here confirm that robust intervals for the values of @xmath288 are not very difficult to identify in practice . this problem was studied by @xcite using the maximum likelihood approach . as we will see , the resulting optimization problem is of the same form as the binary least squares problem . the main difference here is that @xmath289 and @xmath290 . a synchronous k users ds - cdma system is considered with a common single path additive white gaussian noise ( awgn ) channel . the signature waveform of the @xmath155th user is denoted by @xmath291 , a function taking nonzero values in @xmath292 $ ] and being equal to zero outside this interval , and @xmath293 is the information bit transmitted by user @xmath155 . the overall received signal is therefore of the form @xmath294 where @xmath295 is the amplitude of the @xmath155th user s signal and @xmath296 is an additive white gaussian white noise with zero mean and variance @xmath297 . the signal @xmath298 is then filtered using a bank of @xmath299 matched filters . the output of the @xmath155th matched filter is given by @xmath300 in matrix form , this can be written @xmath301 where @xmath302^t$ ] , @xmath187 is the correlation matrix whose components are given by @xmath303 , @xmath304 and @xmath288 is the vector with components @xmath305 . since the gaussian vector has a correlation matrix equal to @xmath306 , the ml estimator is obtained by simply solving the following combinatorial optimization problem . @xmath307 the sdp approach seems to have been first applied for the ds - cdma detection problem in @xcite . since then numerous contributions have appeared using the sdr and comparing it to other methods as in @xcite and @xcite . extension to m - ary phase shift keying symbol constellations is proposed in @xcite . the issue of accelerating the speed of the method is addressed in @xcite . however , as for the former problem , the main drawback of the standard primal semidefinite relaxation is that the size of the problem is greatly increased by using @xmath308 matrices instead of vectors of size @xmath299 . in order to overcome this problem , a better approach using semidefinite programming duality was recently proposed in @xcite . the analysis of the previous sections proves that the eigenvalue relaxation is equally applicable to this problem and maybe a good competitor to the sdp relaxation . the most important point of our analysis is the following : theorem [ strong ] proves that if the correlation matrix @xmath187 is componentwise negative outside the diagonal , then strong duality holds , i.e. the detection problem can be solved exactly in polynomial time . the construction of efficient signatures is the current subject of an active research activity . for instance , the theory of frames allows to consider the problem from an interesting viewpoint as developed in @xcite . our findings suggest in particular that the componentwise negativity of the correlation matrix may be an interesting constraint to look at in future investigations on this problem . [ negcorr ] ( -1,-1)(1,1 ) ( 0,0)(0,1 ) ( 0,0)(.866,-.5 ) ( 0,0)(-.866,-.5 ) ( 0,0)1 ( 0,0).5 - 3090 ( 0,1.2)@xmath309 ( 1.2,-.5)@xmath310 ( -1.2,-.5)@xmath311 ( .6,.6)@xmath312 finally , the eigenvalue relaxation can also be useful even for general signatures because of the weak duality property . indeed , several recent publications prove that clever heuristics can perform better than the sdp relaxation . however , in real situations it is hard to certify that a primal solution provided by such a heuristic is indeed the optimal solution because the original signal is unknown . comparing the dual optimal value to a primal value given by a heuristic can give a precise idea of the error without prior information on the signal . in order to verify this point , we performed monte carlo simulations over 1000 random problems for a number a users varying from 10 to 35 . these computational experiments are reported in figure [ cdma ] where the number of users is on the x - axis and the average computation time is on the y - axis . the computations where performed using the scilab software @xcite . the sdp solver called _ semidef _ interfaces boyd and vandenberghe s sp.c program . the eigenvalue relaxation was solved using the solver _ optim _ with the `` nd '' option for possibly nondifferentiable costs as is the case here . the curves in figure [ cdma ] interpolate the average computation times for messages taken to be sequences of uniform and independant variables taking values in @xmath313 vs. the number of users . the curve with dashed style is for the results of the sdp relaxation while the curve with plain style is for the eigenvalue relaxation . our computations suggest that the eigenvalue relaxation has lower complexity growth as the number of users increases exactly as expected . the reader should be warned that this experiment does not prove that the complexity of the eigenvalue relaxation is lower than the sdp relaxation . the experiment only shows that when a widely used routine for sdp is used , the eigenvalue relaxation , solved using a general purpose bundle method available through a free a well established software , has a lower complexity growth on this problem . arrow matrices are matrices @xmath14 of the form @xmath314,\ ] ] the properties of the eigenvalues of such matrices have been well studied in the past . some of them are summarized in the following theorem . * theorem a*. let @xmath14 be an arrow matrix , with @xmath315 . moreover , assume that all the components of @xmath316 are different from zero . let @xmath317 be its eigenvalues considered in increasing order . then , the characteristic polynomial of @xmath14 is given by @xmath318 then , we have @xmath319 and @xmath320 . moreover , if @xmath321 we have @xmath322 and if @xmath323 , we have @xmath324 . the properties of the eigenvalues of arrow matrices are part of the folkore , especially in the realm of mathematical physics . we give a sketch of the proof of this theorm below in order to give the main ideas underlying the results . * proof of theorem a*. the formula for the characteristic polynomial @xmath325 is easily obtained by reccurence on the dimension . we have to consider two cases : * for some @xmath53 , @xmath321 , * @xmath326 in the first case @xmath327 is a root of @xmath328 . in the second case @xmath329 which is different from zero since we assumed all the @xmath330 s to be different from zero . in this case , the eigenvalues of @xmath14 are the zeros of the function @xmath331 from this formula , we deduce that there is a root in each interval @xmath332 , @xmath333 , for all @xmath334 and @xmath335 . the final conclusions are easily derived by combining the results in the two simple cases discussed above . in this paper , we surveyed the main properties of the eigenvalue relaxation for binary least squares problem . a full connection with the standard sdp relaxation was presented and we showed how to recover a solution of the semi - definite program from the solution of the eigenvalue minimization problem . the problem of recovering primal binary solution was also addressed and we gave simple sufficient conditions for strong duality . in the case where these conditions are not satisfied , the randomized procedure adapted from goemans and williamson s allows to recover binary solutions with garanteed relative approximation ratio due to nesterov s bound . two applications were presented : binary image denoising and detection in multiuser cdma systems . in the case of image denoising , we show that strong duality holds . for the multiuser detection problem , our results prove that strong duality holds when the signature covariance matrix has nonpositive off diagonal components . lemarchal , claude and oustry , franois _ sdp relaxations in combinatorial optimization from a lagrangian viewpoint_. advances in convex analysis and global optimization ( pythagorion , 2000 ) , 119134 , nonconvex optim . 54 , kluwer acad . publ . , dordrecht , 2001 . wolkowicz , henry and anjos , miguel f. _ semidefinite programming for discrete optimization and matrix completion problems_. workshop on discrete optimization , do99 ( piscataway , nj ) . discrete appl . 123 ( 2002 ) , no . 1 - 3 , 513577 . ben - tal , aharon and nemirovski , arkadi _ lectures on modern convex optimization . analysis , algorithms , and engineering applications_. mps / siam series on optimization . society for industrial and applied mathematics ( siam ) , philadelphia , pa ; mathematical programming society ( mps ) , philadelphia , pa , 2001 . helmberg , christoph and oustry , franois _ bundle methods to minimize the maximum eigenvalue function_. handbook of semidefinite programming , 307337 , internat . management sci . , 27 , kluwer acad . boston , ma , 2000 nesterov , yuri , wolkowicz , henry and ye , yinyu _ semidefinite programming relaxations of nonconvex quadratic optimization_. handbook of semidefinite programming , 361419 , internat . management sci . , 27 , kluwer acad . boston , ma , 2000 . keuchel , j. ; schnorr , c. ; schellewald , c. ; cremers , d. ; _ binary partitioning , perceptual grouping , and restoration with semidefinite programming _ , pattern analysis and machine intelligence , ieee transactions on * 25 * , ( 2003 ) , no . 11 , 13641379 . hiriart - urruty , j .- b . ; lemarchal , c. _ convex analysis and minimization algorithms . ii . advanced theory and bundle methods . _ grundlehren der mathematischen wissenschaften , 306 . springer - verlag , berlin , 1993 . goemans , michel x.and williamson , david p. _ improved approximation algorithms for maximum cut and satisfiability problems using semidefinite programming_. j. assoc . 42 ( 1995 ) , no . 6 , 11151145 . garey , m. r. and johnson , d. s. _ computers and intractability . a guide to the theory of np - completeness_. a series of books in the mathematical sciences . w. h. freeman and co. , san francisco , calif . , 1979 . fumihiro hasegawa , jie luo , krishna r. pattipati , peter willett and david pham , _ speed and accuracy comparison of techniques for multiuser detection in synchronous cdma _ , ieee trans . 52 ( 2004 ) , no . 4 , 540545 . moussa abdi , hassan el nahas , alexandre jard and eric moulines , _ semidefinite positive relaxation of the maximum likelihood criterion applied to the multiuser detection in a cdma context _ , ieee sig . letters , 9 , ( 2002 ) , no . 6 , 165167 . d. m. greig , b. t. porteous , and a. h. seheult , _ exact maximum a posteriori estimation for binary images _ , b 51 ( 1989 ) , no . 2 , 271 279 . j. tropp , i. s. dhillon , r. w. heath , jr . , and t. strohmer _ designing structured tight frames via an alternating projection method _ , ieee trans . theory , vol . 51 , ( 2005 ) no . 1 , 188209 .

the observations of distant supernovae @xcite give the evidence that our universe is undergoing accelerated expansion in the present epoch . in principle , there are two different approaches to treat this phenomenon . in the first approach it is postulated that there is some unknown exotic matter which violates the strong energy condition @xmath1 , where @xmath2 is the pressure and @xmath3 is the energy density of perfect fluid . this form of matter is called dark energy . in the past few years different scalar field models like quintessence and more recently the tachyonic scalar field have been conjectured for modelling the dark energy in terms of sub - negative pressure @xmath4 . a scalar field with super - negative pressure @xmath5 called a phantom field can formally be obtained by switching the sign of the kinetic energy in the lagrangian for a standard scalar field . for example in the friedmann - robertson - walker ( frw ) model the phantom field minimally coupled to a gravity field leads to @xmath6 , where @xmath7 , @xmath8 , and @xmath9 is the phantom potential . such a field was called the phantom field by caldwell @xcite who proposed it as a possible explanation of the observed acceleration of the current universe when @xmath10 . note that a coupling to gravity in the quintessence models was also explored @xcite . the second approach called the cardassian expansion scenario has recently been proposed by freese and lewis @xcite as an alternative to dark energy in order to explain the current accelerated expansion of the universe . in this scenario the universe is flat and matter dominated but the standard frw dynamics is modified by the presence of an additional term @xmath11 such that @xmath12 , where @xmath13 is the hubble parameter ; and @xmath14 is the scale factor . however , let us note that this additional term can be interpreted as a phantom field modelled by the equation of state @xmath15 \rho$ ] , where @xmath16 . therefore for dust matter we obtain @xmath17 , and @xmath18 leads to the phantom field . the cardassian expansion with @xmath18 which can be interpreted as the phantom fluid effect . phantom fields lead to the super - accelerated expansion of the universe , i.e. @xmath19 . the simplest models describe this field in terms of minimally coupled real scalar field with the negative kinetic energy term @xmath20 @xcite . it is interesting that phantom fields are also present in string theories @xcite . and arise as a phenomenological description of quantum effects of particle production in terms of bulk viscosity @xcite . because the phantom fields violate the lorentz invariance condition most physicists believe that the phantoms open the doors on new physics @xcite . the investigation of the theoretical possibility to describe dark energy in terms of phantom field was the subject of many papers @xcite . the review on dark energy models were presented by copeland et al . @xcite . in the paper by dabrowski et al . @xcite it is considered the quantization of phantom field via the wheeler - dewitt equation in quantum cosmology . they showed that quantum effects give rise to avoiding the big - rip singularity . also some other basic problems in cosmology like the problem of direction of time can be solved . if we confront the phantom field model with the observation of snia data we obtain that it is the best candidate together with the @xmath21cdm model for the substantial form of dark energy @xcite . it is interesting to investigate new dynamical effects created in the presence of phantom fields . the main goal of this paper is to model the phantom fields in terms of scalar fields with a potential rather than in terms of the barotropic equation of state . for the latter case the dynamics is regular and can be represented on the two - dimensional phase space @xcite . usually phantom fields are modelled in terms of minimally coupled scalar fields the case of minimally coupled to gravity phantom fields was analyzed in the context of existence of periodic solutions @xcite . they demonstrated that the dynamics is trivial in a sense of nonexistence of periodic solutions . faraoni used the framework of phase space for investigating the dynamics of phantom cosmology , late - time attractors , and their existence for different shapes of potential @xcite . it was showed that dynamics of the flat frw universe can be reduced to the form of a two - dimensional dynamical system on a double sheeted phase space . it is a simple consequence of an algebraic equation for @xmath22 which can be expressed as a function of the hubble parameter @xmath23 and @xmath24 . in this case there is no place for chaotic behavior of trajectories in the phase space @xmath25 @xcite . the exit on the inflationary epoch and bounce was showed in the flat frw universe with two interacting phantom scalar fields @xcite . in the closed frw cosmological model with minimally coupled scalar field there appears transient chaos which has character of the scattering process @xcite . the scattering process takes place during the bounce a transition from a contracting to expanding universe . in this context the language of symbolic dynamics and topological entropy was used @xcite . the conformally coupled scalar field are also of interest . the significance of long - wavelength modes in the wkb approximation of a conformally coupled scalar field was analyzed in the inflationary scenario @xcite . while in this paper we concentrate on the conformal coupled scalar fields , it is worth to mention the works about the case of the generic ( non - minimal and non - conformal ) coupling between a phantom field and gravitation . faraoni @xcite discusses different motivations for choosing @xmath26 ( both positive and negative ) from the theoretical point of view and from observational constraints . the inflation and quintessence with the non - minimal coupling were studied in the context of the formulation of some necessary conditions for acceleration of the universe @xcite ( see also @xcite ) . faraoni also pointed out that the non - minimally coupling term different from the conformal coupling value ( @xmath27 ) spoils the equivalence principle of general relativity . it is interesting that some observational constraints on the non - minimal coupling ( possitive or negative ) can be found from cmb observations . tsujikawa and gumjjudpai @xcite studied this bounds for a special power - law potential . the dynamics with the exponential form of the potential of a scalar field was also studied @xcite . the two - field inflation models with negative non - minimal coupling and hybrid inflation are also interesting in the context of the large scale curvature perturbation @xcite . in turn the constraints on the ratio of the self - coupling and non - minimal coupling constant can be obtained during the inflation period @xcite . in this paper we ask what kind of dynamics can be expected from the frw model with phantom field . it is well known that the standard frw model reveals some complex dynamics . the detailed studies gave us a deeper understanding of dynamical complexity and chaos in cosmological models and resulted in conclusion that complex behavior depends on the choice of a time parameterization or a lapse function in general relativity @xcite . castagnino et al . @xcite showed that dynamics of the closed frw models with a conformally coupled massive scalar field is not chaotic if considered in the cosmological time . the same model was analyzed in the conformal time by calzetta and hasi @xcite who presented the existence of chaotic behavior of trajectories in the phase space . motter and letelier @xcite explained that this contradiction in the results is obtained because the system under consideration is non - integrable . therefore we can speak about complex dynamics in terms of nonintegrability rather than deterministic chaos . the significant feature is that nonintegrability is an invariant evidence of dynamical complexity in general relativity and cosmology @xcite . we can find many analogies between models with spontaneously symmetry breaking and the yang - mills systems . problems of chaotic behavior in the yang - mills cosmological models have been investigated by barrow and his collaborators . chaos from yang - mills can only occur in an anisotropic universe but it occurs for some arbitrarily small anisotropy ( see the study of bianchi i @xcite and other bianchi types @xcite ) . however , when a perfect fluid is added things change in an interesting way that contrast with the situation with a magnetic field @xcite . for the frw model with phantom it can be shown that there is a monotonous function along its trajectories and it is not possible to obtain the lyapunov exponents or construct the poincar sections . therefore it is useful to study the nonintegrability of the phantom system and set it in a much stronger form by proving that the system does not possess any additional and independent of hamiltonian first integrals , which are in the form of analytic or meromorphic functions . of course , it is not the evidence of sensitive dependence of solution on a small change of initial conditions . however , it is the possible evidence of complexity of dynamical behavior formulated in an invariant way @xcite . the notion of deterministic chaos is a controversial issue in the mixmaster models . the value of the numerically computed maximal lyapunov exponent for those systems depends on the time parameterization used @xcite . however , the existence of fractal structures in the phase space provides the coordinate independent signal of chaos in cosmology as it was shown by cornish and levin @xcite . in particular , they found that the bianchi ix has a form of chaotic scattering . the short scattering periods intermittent integrable motion and evolution of the system is chaotic . it is similar to a pin - ball machine . the main goal of this paper is to show that dynamics of phantom cosmology can be treated as a scattering process . only if the spontaneously symmetry breaking is admitted this process has chaotic character . evidences of dynamical behavior are studied in tools of symbolic dynamics , fractal dimension , and analytically the toda - brumer - duff test . we assume the model with frw geometry , i.e. , the line element has the form @xmath28 , \label{eq:1}\ ] ] where @xmath29 @xmath30 is the curvature index , @xmath31 and @xmath32 are comoving coordinates , @xmath33 stands for the cosmological time . it is also assumed that a source of gravity is the phantom scalar field @xmath24 with a generic coupling to gravity . the gravitational dynamics is described by the standard einstein - hilbert action @xmath34 where @xmath35 ; for simplicity and without lost of generality we assume @xmath36 . the action for the matter source is @xmath37 let us note that the formal sign of @xmath38 is opposite to that which describes the standard scalar field as a source of gravity , where @xmath39 is a scalar field potential . we assume @xmath40 and that conformal volume @xmath41 over the spatial 3-hypersurface is unity . @xmath42 is the coupling constant of the scalar field to the ricci scalar @xmath43 where a dot means the differentiation with respect to the cosmic time @xmath33 . if we have the minimally coupled scalar field then @xmath44 . we assume a non - minimal coupling of the scalar field @xmath45 . the dynamical equation for phantom cosmology in which the phantom field is modelled by the scalar field with an opposite sign of the kinetic term in action can be obtained from the variational principle @xmath46 . after dropping the full derivatives with respect to time we obtain the dynamical equation for phantom cosmology from variation @xmath47 as well as the dynamical equation for field from variation @xmath48 @xmath49 it can be shown that for any value of @xmath42 the phantom behaves like some perfect fluid with the effective energy @xmath50 and the pressure @xmath51 in the form which determines the equation of state factor @xmath52 + \xi \psi^{2}(2\dot{h}+3h^{2})}{-\frac{1}{2}\dot{\psi}^{2}+u(\psi ) - 3\xi h[h\psi^{2}+(\psi^{2})\dot{\ } ] } \equiv \frac{p_{\psi}}{\rho_{\psi}}. \label{eq:8}\ ] ] formula ( [ eq:8 ] ) differs from its counterpart for the standard scalar field @xcite by the presence of a negative sign in front of the term @xmath53 . in equation ( [ eq:8 ] ) the second derivative @xmath54 in the expression for the pressure can be eliminated and then we obtain @xmath55 of course such perfect fluid which mimics the phantom field satisfies the conservation equation @xmath56 we can see that complexity of a dynamical equation should manifest by complexity of @xmath57 . let us consider the frw quintessential dynamics with some effective energy density @xmath50 given in equation ( [ eq:8 ] ) . this dynamics can be reduced to the form like of a particle in a one - dimensional potential @xcite and the hamiltonian of the system is @xmath58 where a prime means the differentiation with respect to the conformal time @xmath59 . the trajectories of the system lie on the zero energy level for flat and vacuum models . note that if we additionally postulate the presence of radiation matter for which @xmath60 then it is equivalent to consider the hamiltonian on the level @xmath61 . of course the division on kinetic and potential parts has only a conventional character and we can always translate the term containing @xmath53 into a kinetic term . the dynamics of the model is governed by the equation of motion ( [ eq:7 ] ) , which is equivalent to the conservation condition ( [ eq:10 ] ) and the acceleration condition @xmath62 this equation admits the generalized friedmann first integral which assumes the following form @xmath63 if we postulate existence of radiation in model then left hand of this equation can be negative . let us consider now both cases of minimally and conformally coupled phantom fields . they can model the quintessence matter field in terms of matter satisfying the equation of state @xmath64 . for minimally coupled phantom fields ( @xmath44 ) the function of energy takes the form @xmath65 where @xmath66 , @xmath67 , @xmath68 , @xmath69 , @xmath70 is assumed . for conformally coupled phantom fields we put @xmath71 and rescale the field @xmath72 . then the energy function takes the following form for a simple mechanical system with a natural lagrangian function @xmath73 @xmath74 in contrast to the frw model with conformally coupled scalar field the kinetic energy form is positive definite like for classical mechanical systems . the general hamiltonian which represents the special case of a two coupled non - harmonic oscillators system is @xmath75 where @xmath76 , @xmath77 , @xmath78 , @xmath79 , and @xmath80 are constants . in order to study the integrability of dynamical systems we use painlev s approach . painlev s analysis gives necessary conditions for the integrability of dynamical systems and it is the most popular integrability detector . the recapitulation of painlev s analysis for system ( [ eq:15 ] ) was done by lakshmanan and sahadevan @xcite . they found that system ( [ eq:15 ] ) passes the painlev test and is integrable in the following four cases @xmath81 this result is in full agreement with conclusions concerning integrability of the two coupled quartic non - harmonic oscillator systems . therefore , phantom cosmology can be considered as coupled quartic non - harmonic oscillators . of course , painlev analysis gives necessary conditions for the integrability of dynamical systems . however , there exist whole classes of integrable systems which do not possess the painlev property . it means that the painlev approach gives over - restrictive conditions for integrability . it is obvious that for @xmath82 the frw phantom cosmology is integrable because of the possibility to separate of variables in the potential . then we have two decoupled quartic non - harmonic oscillators . the non - integrability of the non - flat frw model with the scalar field with the potential @xmath83 was investigated in an analytical way by ziglin @xcite . for a deeper analysis of integrability in terms of ziglin and morales - ruiz and ramis . it would be useful to compare hamiltonians for the cosmological model with phantom fields and the standard cosmological model with scalar fields . let us consider that @xmath84 for simplicity of presentation . then for both models we have hamiltonians @xmath85 and @xmath86 if we add a radiation component to the energy momentum tensor , which energy density scales like @xmath87 then both systems should be considered on the constant energy level @xmath88 or the constant @xmath89 can be absorbed by the new hamiltonian @xmath90 and @xmath91 is considered on the zero energy level . let us concentrate on the flat models to analyze the similarities to the yang - mills systems . while the standard cosmological model is described by the hamiltonian @xmath92 the phantom cosmological model is @xmath93 the crucial difference between system ( [ eq:18 ] ) and ( [ eq:19 ] ) lies in the definiteness of their kinetic energy forms . it is indefinite and has the lorentzian signature for the standard model , and it is positive definite for the phantom model . as a consequence we obtain that the configuration space for the standard model is @xmath94 whereas the condition @xmath95 determines the domain of the configuration space admissible for motion for the phantom model . note that if @xmath96 ( the model with the spontaneously symmetry breaking ) then this domain is bounded by four hyperbolas in every quarter of plane @xmath97 . the same situation can be obtained for the model without the spontaneously symmetry breaking ( @xmath98 ) and dark radiation ( @xmath99 ) . the flat phantom model with the spontaneously symmetry breaking is well known as the yang - mills systems which have been analyzed since the pioneering paper by savvidy @xcite . for this system the lyapunov exponents were found @xcite and the poincar sections were obtained @xcite . the spatially flat universes filled with the yang - mills fields exhibit chaotic oscillations of these fields @xcite . this system was also investigated by using the gaussian curvature criterion @xcite . let us apply this criterion to the non - flat phantom model . the potential function takes the following form @xmath100 according to this criterion the periodic and quasi - periodic orbits appear in the domains of the configuration space in which the gaussian curvature of a diagram of the potential function is positive . the line of zero curvature separates these domains from the instability regions where the curvature is negative . if the total energy of the system @xmath80 increases the system will be in a region of negative curvature for some initial conditions and the motion is chaotic . let us consider the dynamical system in an autonomous form in the form @xmath101 then we linearized around the special solution and we obtain equation @xmath102 where @xmath103 builds the jacobian and @xmath104 is the deviation vector connecting the points on two nearby trajectories corresponding the same value of parameter @xmath33 . in our case @xmath105 , @xmath106 , @xmath107 , @xmath108 then the local instability of nearby trajectories are determined by eigenvalues of the jacobian matrix , i.e. the effective potential @xmath109 . the eigenvalues are @xmath110 @xmath111 therefore the necessary condition for local instability is that at least one of the eigenvalues is positive @xcite . these positive values decide about the local instability of trajectories . note that the negative sign of the gaussian curvature of potential @xmath109 is the sufficient condition of local instability @xcite @xmath112 we can see that test ( [ eq:22 ] ) of negative curvature adopted in our case gives to @xmath113 . however it is not a sufficient condition for the chaos . it should be pointed out that this criterion has a purely local character in contrast to the lyapunov exponent . moreover the compactness of a region admissible for motion is required for chaos existence . it is worthwhile to mention that the toda criterion is only the measure of the local instability of nearby trajectories @xcite . therefore the presented analysis of the order - chaos transition should be combined with the poincar sections and other deeper indicators of the chaotic behavior . to make the numerical analysis we need to distinguish the the model without the spontaneously symmetry breaking ( @xmath98 ) and the model with the spontaneously symmetry breaking ( @xmath114 ) . in both cases the system is considered on the some energy level . we would like to stress out why we use the nonintegrability criterion instead of standard measure like the lyapunov exponents or poincar sections . the main reason is that in analogy to castagnino et al.s work @xcite we can find a monotonous function along a trajectory . this excludes the property of the recurrence the trajectories or topological transitivity in the standard wiggins chaos definition . then @xmath116 ( or @xmath117 ) is a monotonous function of the time parameter along the trajectories and trajectories escape to infinity for arbitrary initial conditions . in this case it is impossible to construct the poincar sections . it is obvious for any curvature and positive values of @xmath21 , @xmath118 , and @xmath119 . the phase space as well as the configuration space is unbounded in this case . in contrast to classical chaotic systems there is invariant compact chaotic set for this system . it does not mean the system is non - chaotic as it does not follow the wiggins definition of chaos . the essence of this phenomenon has different nature . the system is oversensitive with respect to small change of initial conditions like in chaotic scattering processes . the typical case of the model is a class of models without the spontaneously symmetry breaking @xmath120 in this case the domain admissible for motion is unbounded . for the classical mechanical systems with chaos there are chaotic sets of trajectories on a compact invariant submanifold . however it does not mean that the system with an unbounded region admissible for motion is nonchaotic because the essence of chaotic behavior is of different nature . the system can be oversensitive with respect to small changes of initial conditions the key ingredient of chaos . in this respect the system is similar to chaotic scattering processes @xcite . therefore it seems to be natural to use the methods of investigation of scattering processes . the class a model does not possess the property of sensitive dependence on initial conditions . moreover , there are no fractal structures of basin boundaries in the phase space . because the scattering of trajectories in the potential well is present , we deal with the nonchaotic scattering process . in more details it will be discussed in sec . [ sec:4 ] . the different chaotic behavior was found in the einstein - yang - mills ( eym ) color field in the flat bianchi i spacetime @xcite . it was considered the model with radiation and isotropic curvature in contrast to the mixmaster models . it was showed that the eym systems are also an example of chaotic scattering . because the lyapunov exponents are coordinate dependent they can not be used for invariant characterization of chaos . for this aim the methods of chaotic scattering are more suitable . they are extremely useful in this context because of noncompact phase space ( also the configuration space)the major obstacle of the standard analysis of chaos . let us note that the case of their model with flat spacetime is analogous to our second case with the cosmological constant . , @xmath121 , @xmath122 ( left ) and @xmath123 ( right ) , @xmath124 calculated from the hamiltonian constraint.,title="fig : " ] , @xmath121 , @xmath122 ( left ) and @xmath123 ( right ) , @xmath124 calculated from the hamiltonian constraint.,title="fig : " ] , @xmath121 , @xmath122 ( left ) and @xmath123 ( right ) , @xmath124 calculated from the hamiltonian constraint.,title="fig : " ] , @xmath121 , @xmath122 ( left ) and @xmath123 ( right ) , @xmath124 calculated from the hamiltonian constraint.,title="fig : " ] and @xmath125 ( upper ) ; @xmath126 , @xmath125 ( bottom ) . motion in narrow field between two hyperbolas is completely regular . complexity of behavior comes from motion near the origin of the configuration space.,title="fig : " ] and @xmath125 ( upper ) ; @xmath126 , @xmath125 ( bottom ) . motion in narrow field between two hyperbolas is completely regular . complexity of behavior comes from motion near the origin of the configuration space.,title="fig : " ] and @xmath127 . the empty region at the upper section comes from motion of a particle - universe along the @xmath128 ( @xmath126 ) axis.,title="fig : " ] and @xmath127 . the empty region at the upper section comes from motion of a particle - universe along the @xmath128 ( @xmath126 ) axis.,title="fig : " ] and @xmath125 . ] and @xmath125 for class i and landing at @xmath129 , @xmath130 , @xmath131 , @xmath132 ( grey ) or at @xmath133 , @xmath130 , @xmath131 , @xmath132 ( white ) , respectively . complexity of the fractal structure increases with the final state of the trajectories which means that trajectories spend more time in the region near the point @xmath134 in the configuration space . the solid areas in the center of figures come from the motion along the @xmath128-axis.,title="fig : " ] and @xmath125 for class i and landing at @xmath129 , @xmath130 , @xmath131 , @xmath132 ( grey ) or at @xmath133 , @xmath130 , @xmath131 , @xmath132 ( white ) , respectively . complexity of the fractal structure increases with the final state of the trajectories which means that trajectories spend more time in the region near the point @xmath134 in the configuration space . the solid areas in the center of figures come from the motion along the @xmath128-axis.,title="fig : " ] and @xmath125 for class i and landing at @xmath129 , @xmath130 , @xmath131 , @xmath132 ( grey ) or at @xmath133 , @xmath130 , @xmath131 , @xmath132 ( white ) , respectively . complexity of the fractal structure increases with the final state of the trajectories which means that trajectories spend more time in the region near the point @xmath134 in the configuration space . the solid areas in the center of figures come from the motion along the @xmath128-axis.,title="fig : " ] and @xmath125 for class i and landing at @xmath129 , @xmath130 , @xmath131 , @xmath132 ( grey ) or at @xmath133 , @xmath130 , @xmath131 , @xmath132 ( white ) , respectively . complexity of the fractal structure increases with the final state of the trajectories which means that trajectories spend more time in the region near the point @xmath134 in the configuration space . the solid areas in the center of figures come from the motion along the @xmath128-axis.,title="fig : " ] .,title="fig : " ] .,title="fig : " ] .,title="fig : " ] .,title="fig : " ] .,title="fig : " ] .,title="fig : " ] .,title="fig : " ] .,title="fig : " ] the idea of the description of a dark energy field in terms of a higgs field which creates inertial mass through the spontaneously symmetry breaking has been investigated lately @xcite . for @xmath114 we can distinguish two subclasses of the model with the spontaneously symmetry breaking for which chaotic behavior can be detected @xmath135 these are flat models with conformally coupled phantom fields for which trajectories have the property of topological transitivity in contrast to the case @xmath98 . in fig . [ fig:1 ] and [ fig:2 ] we present sample trajectories for both models and evolution for every trajectory is last for the same interval of time . from this simple picture we can initially conclude that the systems are sensitive on initial conditions . the first class of models is isomorphic with the well known yang - mills systems with the potential function @xmath136 . because the domain admissible for motion is bounded and its boundary has negative curvature , the property of recurrence of trajectories is present . we consider the both classes of systems on some distinguished energy level @xmath137 . the poincar sections for both classes are represented in figs . [ fig:3 ] , [ fig:4 ] , [ fig:5 ] . in fig . [ fig:3 ] it is shown the poincar section for the flat cosmological model with vanishing @xmath21 , @xmath118 , and @xmath138 ( class i ) . if we postulate additionally the existence of radiation matter in the model than we deal with the system on the constant non - zero energy level . in this case we obtain some chaotic distribution of points on the poincar section @xmath139 . in fig . [ fig:4 ] and [ fig:5 ] it is presented the flat cosmological model with the negative cosmological constant @xmath21 and @xmath138 ( class ii ) on the plane @xmath140 . in this case we can observe islands of stability . however , the trajectories wander to the non - physical region of @xmath141 . this allows many cycles to be considered if we continue the scale factor into negative values . but what it means physically is not clear . all these figures illustrate what we proved earlier , namely the standard chaos with recurrence of orbits and the property of sensitive dependence on initial conditions . we can not perform the analogous analysis for the class of models with @xmath142 because there is no chaos in wiggins standard sense . however they possess the property of complex behavior of trajectories similar to the non - chaotic scattering process . note that the evidence of chaos in terms of fractal basins , cantori or stochastic layers requires the recurrence of trajectories . similarly the poincar sections can be constructed from many cycles for a useful picture to emerge @xcite . in the case of @xmath138 , trajectories of the system return to the neighborhood of the origin time after time and we have the multiple scattering process on the potential walls ( boundaries of the domain admissible for motion ) . we can control and count how many times the trajectory gets closer to the origin than some fixed distance from the origin . with increasing this fix distance the number of controlled scattering events increases and a more complicated fractal structure arises . [ fig:6 ] , [ fig:7 ] and [ fig:8 ] illustrate fractal structures of the phase space of initial conditions for trajectories reaching some values @xmath143 and @xmath144 at some moment of time evolution . the fractal dimension calculated by counting cells which contain light and dark areas ( i.e. leading to two types of the final outcome ) is @xmath145 @xmath146 ; @xmath147 @xmath148 ; @xmath149 @xmath150 ; @xmath151 @xmath152 . the chosen method of counting cell enables us to order fractals with respect to the increasing complexity , i.e. for a nearly regular system the fractal dimension is close to one , in turn for a completely chaotic one is approaching two . one can also observe that complexity of dynamical behavior of trajectories grows more the longer trajectories come more often in the neighborhood of the origin of the coordinate system because the motion between branches of hyperbolas is regular ( see fig . [ fig:3 ] ) . [ fig:9 ] , [ fig:10 ] , [ fig:11 ] and [ fig:12 ] present the fractal structure of phase space of initial conditions for the category ii model . at enough small value of @xmath144 the system is regular and no chaotic behavior is present ( fig . [ fig:9 ] ) . with an increasing value of @xmath144 : @xmath153 @xmath154 ; @xmath155 @xmath156 , we observe the transition to chaos which cause to emerge the fractal structure of the space of initial conditions ( fig . [ fig:10][fig:12 ] ) . in fig . [ fig:13 ] we plot the phase space of initial conditions chosen at @xmath126 and @xmath125 calculated from the hamiltonian which lead to a given property in one cycle of evolution , i.e. to maximal expansion with a positive value of @xmath128 ( grey ) or negative value of @xmath128 ( white ) left image , and back to a final singularity again with a positive value of @xmath128 ( grey ) or a negative value of @xmath128 ( white ) right image . therefore if we consider evolution in a physical domain between initial and final singularities this system can not be chaotic . if we prolong its evolution to the non - physical domain @xmath141 then we obtain chaos . note that non - integrability indicators measure the true intrinsic complexity of the system in both cases @xcite . and @xmath127 and landing at @xmath157 ( dark grey ) and @xmath158 ( grey ) presented in left figure and @xmath159 ( dark grey ) and @xmath160 ( grey ) in right figure . the white poles correspond to the centers of stability in fig . [ fig:4].,title="fig : " ] and @xmath127 and landing at @xmath157 ( dark grey ) and @xmath158 ( grey ) presented in left figure and @xmath159 ( dark grey ) and @xmath160 ( grey ) in right figure . the white poles correspond to the centers of stability in fig . [ fig:4].,title="fig : " ] ( left ) and @xmath161 ( right ) . initial conditions leading to negative values of @xmath128 omitted for clarity.,title="fig : " ] ( left ) and @xmath161 ( right ) . initial conditions leading to negative values of @xmath128 omitted for clarity.,title="fig : " ] .,title="fig : " ] .,title="fig : " ] .,title="fig : " ] .,title="fig : " ] with @xmath125 for trajectories leading to @xmath162 ( or equivalent @xmath163 ) with @xmath164 ( grey ) and @xmath165 ( white ) the left image , and initial conditions of trajectories leading to @xmath126 with @xmath164 ( grey ) and @xmath165 ( white ) the right image . there is no chaos in one cycle of evolution.,title="fig : " ] with @xmath125 for trajectories leading to @xmath162 ( or equivalent @xmath163 ) with @xmath164 ( grey ) and @xmath165 ( white ) the left image , and initial conditions of trajectories leading to @xmath126 with @xmath164 ( grey ) and @xmath165 ( white ) the right image . there is no chaos in one cycle of evolution.,title="fig : " ] in this section we investigate behavior of a model without the spontaneously symmetry breaking . this is a model with @xmath166 and @xmath167 ( @xmath168 ) on different energy levels @xmath169 , @xmath170 and @xmath171 . in the analysis of this case we explore analogy to the classical system in which appears the chaotic scattering . from the equation of motion we obtain that for any initial condition trajectories escape to infinity ( @xmath172 @xmath173 as time goes to infinity ) . for energy @xmath174 the configuration space is unbounded and trajectories pass from one quadrant to another . on the other hand for @xmath175 motion is restricted to only this quadrant of the configuration space where the initial conditions are . in fig . [ fig:14 ] we present the analysis of dependence of scattering on the energy level . for this aim the initial condition in the configuration space are chosen on the line @xmath176 with @xmath177 under the conservation of the hamiltonian constraint . for @xmath175 trajectories go toward the origin of coordinate system and after some time escape to infinity passing through a line @xmath176 . on the @xmath178-axis we mark the initial value of @xmath14 and @xmath179-axis we mark final distance from the symmetry point with coordinates @xmath180 at the moment of intersection of line of initial conditions . in this figure there is no discontinuities , we do not also observe the fluctuation of distance @xmath181 . it enable us to conclude that we deal with the scattering process although the domain of interaction is not finite @xcite . so there is no chaotic scattering . the motion of the system is regular and there is no sensitivity of motion with respect to initial conditions . for a deeper confirmation of this statement we study numerically a larger number of initial conditions ( fig . [ fig:15 ] ) . in the configuration space the initial conditions are chosen as previously , i.e. @xmath176 and the initial conditions for velocities are parameterized in a natural way by an angle @xmath182 , i.e. @xmath183 , @xmath184 where @xmath185 . in fig . [ fig:15 ] ( a ) and ( b ) it is presented results of our analysis for @xmath186 and @xmath187 respectively . the grey area corresponds to initial conditions for which trajectories pass through line @xmath176 with @xmath188 . [ fig:15 ] ( c ) illustrates domains of initial conditions for @xmath174 for which trajectories outcomes reach different quadrants : i quadrant dark grey , ii quadrant grey , iii quadrant black , iv quadrant medium grey . the black line presents initial conditions taken from fig . [ fig:14 ] . we can not observe chaotic scattering in this case . in configuration space and @xmath189 calculated from the hamiltonian . on the @xmath178-axis we put initial position @xmath14 and on the @xmath179-axis we put a final distance from a symmetry point in the configuration space @xmath180 when a trajectory is escaping to infinity . for @xmath171 ( a ) and @xmath170 ( b ) all the trajectories escape to infinity ( @xmath190 ) without crossing @xmath126 or @xmath191 . for @xmath169 ( c ) some trajectories change a quarter after starting from i quarter : i solid line , ii dashed line , iii dotted line and iv dash - dot line.,title="fig : " ] + in configuration space and @xmath189 calculated from the hamiltonian . on the @xmath178-axis we put initial position @xmath14 and on the @xmath179-axis we put a final distance from a symmetry point in the configuration space @xmath180 when a trajectory is escaping to infinity . for @xmath171 ( a ) and @xmath170 ( b ) all the trajectories escape to infinity ( @xmath190 ) without crossing @xmath126 or @xmath191 . for @xmath169 ( c ) some trajectories change a quarter after starting from i quarter : i solid line , ii dashed line , iii dotted line and iv dash - dot line.,title="fig : " ] + in configuration space and @xmath189 calculated from the hamiltonian . on the @xmath178-axis we put initial position @xmath14 and on the @xmath179-axis we put a final distance from a symmetry point in the configuration space @xmath180 when a trajectory is escaping to infinity . for @xmath171 ( a ) and @xmath170 ( b ) all the trajectories escape to infinity ( @xmath190 ) without crossing @xmath126 or @xmath191 . for @xmath169 ( c ) some trajectories change a quarter after starting from i quarter : i solid line , ii dashed line , iii dotted line and iv dash - dot line.,title="fig : " ] and @xmath170 , respectively . the grey areas correspond to initial conditions for which trajectories pass through the line @xmath192 with @xmath193 . fig . ( c ) corresponds @xmath169 ( the configuration space is unbounded ) and illustrates domains of initial conditions for which trajectories outcomes reach different quadrants : i quadrant dark grey , ii quadrant grey , iii quadrant black , iv quadrant medium grey . the black lines represent initial conditions from fig . [ fig:14].,title="fig : " ] + and @xmath170 , respectively . the grey areas correspond to initial conditions for which trajectories pass through the line @xmath192 with @xmath193 . ( c ) corresponds @xmath169 ( the configuration space is unbounded ) and illustrates domains of initial conditions for which trajectories outcomes reach different quadrants : i quadrant dark grey , ii quadrant grey , iii quadrant black , iv quadrant medium grey . the black lines represent initial conditions from fig . [ fig:14].,title="fig : " ] + and @xmath170 , respectively . the grey areas correspond to initial conditions for which trajectories pass through the line @xmath192 with @xmath193 . ( c ) corresponds @xmath169 ( the configuration space is unbounded ) and illustrates domains of initial conditions for which trajectories outcomes reach different quadrants : i quadrant dark grey , ii quadrant grey , iii quadrant black , iv quadrant medium grey . the black lines represent initial conditions from fig . [ fig:14].,title="fig : " ] the current universe is in an accelerating phase of expansion that s why we check whether the scalar fields in the frw cosmology can explain this phenomenon . for this aim let s are consider acceleration equation which can be obtained from the raychaudhury equation @xmath194 in the special case of the phantom field minimally coupled to gravity @xmath44 and without radiation @xmath195 we have @xmath196 next , for zero energy level and @xmath197 from ( [ fint ] ) we have that @xmath198 and finally we receive the acceleration equation which independent of the form of the potential function @xmath199 in the case of conformally coupled phantom field @xmath71 we have @xmath200 and after the rescaling time and field variables : @xmath201 , @xmath202 and inserting the potential function ( [ eq:5 ] ) we have @xmath203 where prime denotes differentiation with respect to the conformal time and dot with respect to the cosmological time . we can also express equation of state parameter ( [ eq:8 ] ) in these new variables @xmath204 we clearly see that this is a good expression only if @xmath205 . for the physical region @xmath206 we can express the effective equation of state parameter as @xmath207 in fig . [ fig:16 ] we show the evolution of acceleration and evolution of coefficient of the equation of state @xmath208 for model class i ( flat model with @xmath209 ) with respect to the conformal time @xmath59 which is monotonous function of the cosmological time in the physical domain . note that in the case of @xmath210 the universe is decelerating and @xmath208 stays in the interval @xmath211 , with @xmath212 when @xmath213 and @xmath214 when @xmath191 . the analogous analysis made for the model class a without the spontaneously symmetry breaking ( presented in fig . [ fig:17 ] ) shows that the universe is always accelerating . in this case pressure @xmath215 is always negative . in the case of @xmath169 we can have positive value of @xmath208 and this means that density can be negative . moreover , during evolution there are crossings of the line @xmath216 and then @xmath217 goes to minus one . the amplitude of these oscillations depends on @xmath89 . ) with @xmath218 ( left panel ) and equation of state parameter @xmath219 from eq . ( [ eq:30 ] ) ( right panel ) for model class i with respect to conformal time for short period of time for trajectory from left panel of fig . [ fig:1 ] . shaded area denote unphysical regions of scale factor @xmath220 . in physical regions acceleration is always negative @xmath221 model decelerates and equation of state parameter never crosses @xmath222 barrier.,title="fig : " ] ) with @xmath218 ( left panel ) and equation of state parameter @xmath219 from eq . ( [ eq:30 ] ) ( right panel ) for model class i with respect to conformal time for short period of time for trajectory from left panel of fig . [ fig:1 ] . shaded area denote unphysical regions of scale factor @xmath220 . in physical regions acceleration is always negative @xmath221 model decelerates and equation of state parameter never crosses @xmath222 barrier.,title="fig : " ] ) ( left panel ) and the equation of the state parameter @xmath217 from eq . ( [ eq:31 ] ) ( right panel ) for model class a without spontaneously symmetry breaking for a sample trajectory starting from the same initial conditions in the configuration space ( @xmath223 , @xmath224 and @xmath225 calculated from the hamiltonian constrain ) but for different energy levels : @xmath171 ( top ) , @xmath170 ( middle ) and @xmath169 ( bottom ) . for every sample we have acceleration but the equation of the state parameter oscillates and aproaches @xmath226.,title="fig : " ] ) ( left panel ) and the equation of the state parameter @xmath217 from eq . ( [ eq:31 ] ) ( right panel ) for model class a without spontaneously symmetry breaking for a sample trajectory starting from the same initial conditions in the configuration space ( @xmath223 , @xmath224 and @xmath225 calculated from the hamiltonian constrain ) but for different energy levels : @xmath171 ( top ) , @xmath170 ( middle ) and @xmath169 ( bottom ) . for every sample we have acceleration but the equation of the state parameter oscillates and aproaches @xmath226.,title="fig : " ] ) ( left panel ) and the equation of the state parameter @xmath217 from eq . ( [ eq:31 ] ) ( right panel ) for model class a without spontaneously symmetry breaking for a sample trajectory starting from the same initial conditions in the configuration space ( @xmath223 , @xmath224 and @xmath225 calculated from the hamiltonian constrain ) but for different energy levels : @xmath171 ( top ) , @xmath170 ( middle ) and @xmath169 ( bottom ) . for every sample we have acceleration but the equation of the state parameter oscillates and aproaches @xmath226.,title="fig : " ] ) ( left panel ) and the equation of the state parameter @xmath217 from eq . ( [ eq:31 ] ) ( right panel ) for model class a without spontaneously symmetry breaking for a sample trajectory starting from the same initial conditions in the configuration space ( @xmath223 , @xmath224 and @xmath225 calculated from the hamiltonian constrain ) but for different energy levels : @xmath171 ( top ) , @xmath170 ( middle ) and @xmath169 ( bottom ) . for every sample we have acceleration but the equation of the state parameter oscillates and aproaches @xmath226.,title="fig : " ] ) ( left panel ) and the equation of the state parameter @xmath217 from eq . ( [ eq:31 ] ) ( right panel ) for model class a without spontaneously symmetry breaking for a sample trajectory starting from the same initial conditions in the configuration space ( @xmath223 , @xmath224 and @xmath225 calculated from the hamiltonian constrain ) but for different energy levels : @xmath171 ( top ) , @xmath170 ( middle ) and @xmath169 ( bottom ) . for every sample we have acceleration but the equation of the state parameter oscillates and aproaches @xmath226.,title="fig : " ] ) ( left panel ) and the equation of the state parameter @xmath217 from eq . ( [ eq:31 ] ) ( right panel ) for model class a without spontaneously symmetry breaking for a sample trajectory starting from the same initial conditions in the configuration space ( @xmath223 , @xmath224 and @xmath225 calculated from the hamiltonian constrain ) but for different energy levels : @xmath171 ( top ) , @xmath170 ( middle ) and @xmath169 ( bottom ) . for every sample we have acceleration but the equation of the state parameter oscillates and aproaches @xmath226.,title="fig : " ] in cosmology the mixmaster models are well known for their chaotic behavior @xcite . in these models chaos is caused by anisotropy of space . in this paper we considered the phantom cosmological models which are homogeneous and isotropic where the chaos is the consequence of the nonlinearity of the potential function for the scalar field . we found chaos is a generic feature of the phantom cosmology with the spontaneously symmetry breaking . it assumes the form of the multiple chaotic scattering process . in the case of the absence of symmetry breaking we did not find the sensitivity over initial conditions and the system behaves as the standard scattering process . there is no difference between the flat universes filled with minimal and conformally coupled scalar fields . as it was demonstrated by faraoni et al . @xcite there is no chaos for the former class of models because the solution corresponding @xmath227 are restricted to move in some two - dimensional submanifold of the phase space @xmath228 . there are no enough room for chaotic motion manifestation . the analogous result is valid for the other value of the coupling constant @xmath42 . note that we can observe chaotic behavior in some cases of the conformally coupled scalar field in the flat universes , when we consider the system on non - zero energy level . physically it means that the universe is filled with radiation matter . the reason for which the flat universes with scalar field are non - chaotic lies in lack of invariant measure of chaotic behavior in the general relativity @xcite . in general the standard chaos indicators like lyapunov exponents or kolmogorov - sinai entropy depends on time parameterization . in general relativity there is freedom of choice of the lapse function which defines reparameterization of time . the hamiltonian is given modulo the lapse function . therefore if we consider the hamiltonian frw system with scalar field on zero energy level it is possible that in the hamiltonian constraint @xmath14 i @xmath229 appear only in the combination @xmath230 . hence the motion of the system takes place in the 3-dimensional phase space @xmath231 on some 2-dimensional submanifold . for any value of coupling constant and general class of potentials there is no place for chaos ( even the models with the cosmological constant ) . the new picture appears if the system is described on a non - zero energy level . then the motion of the system is in the 3-dimensional submanifold of the 4-dimensional phase space . but in general case there is no possibility of choosing the lapse function in such a way that @xmath14 i @xmath229 contribute in the hamiltonian constraint by the hubble function . among the conformally coupled models there are in principal two types of behavior . first , chaotic scattering process takes place for case @xmath114 ( the spontaneously symmetry breaking ) . in this case trajectories possess the property of topological transitivity which guarantees their recurrence in the phase space and the scattering process takes place around the origin . it is similar to the chaos appeared in the yang - mills theory . exploring this analogy to this system we calculate a gaussian curvature of the potential function which measure local instability of nearby trajectories . we proof that this curvature is negative . because the configuration space is bounded in this case the mixing of trajectories is observed . second , in the opposite case of @xmath98 we characterize dynamics in the system in terms of symbolic dynamics . no fractal structure in the space of initial conditions was found and conclude that the scattering process has not the chaotic character . we also found that the universe filled with conformally coupled scalar field is accelerating without the positive cosmological constant . in principle , there are two sources of this acceleration . the acceleration is driven by negative energy density and coefficient of equation of state @xmath208 is greater than @xmath232 . alternatively , energy density is positive and the coefficient of equation of state can be smaller than @xmath232 . the latter there is crossing of @xmath233 . of course , for the explanation of snia data apart from the acceleration of the universe itself , the rate of acceleration is also crucial . the system with phantom scalar field coupled conformally to gravity in the frw universe can be treated as scattering process with chaotic and nonchaotic character . the generic feature of this class of systems with the spontaneously symmetry breaking is the chaotic scattering of trajectories . the paper was supported by the marie curie actions transfer of knowledge project cocos ( contract mtkd - ct-2004 - 517186 ) . we would like to thanks dr j. grzywaczewski for his hospitality during the staying in paris .

.3 cm quaternions are one of the simplest examples of a non - commutative associative algebra . when pure imaginary qua - ternions are used to represent 3d vectors , the adjoint action of the group of unit quaternions upon them provides a representation of all 3d rotations . because of this , quaternions have important applications in 3d euclidean computing @xcite . in symbolic geometric computing , when geometric entities and their transformations are represented by coordinate - free symbols , algebraic manipulations among the symbols often lead to rather simple and geometrically interpretable results @xcite . for example , using @xmath0 instead of @xmath1 as a vector variable , provides a typical coordinate - free representation of a 3d direction ( or point ) by a pure imaginary quaternion . under the quaternionic product , these variables generate a ring of non - commutative polynomials . it is natural to ask for symbolic algebraic algorithms to manipulate such _ quaternionic polynomials_. when the coefficients of the quaternionic polynomials are taken in the field of rational numbers @xmath2 , the quaternionic product , being associative and multilinear , must be a quotient of the @xmath2-tensor product modulo an ideal @xmath3 whose generators are some tensors . finding such generators and proving that they are sufficient to define the quaternionic product is the first task to finish . the first theorem in this paper provides very simple generators of @xmath3 , called the _ syzygy ideal _ of quaternionic polynomials . the next task is naturally on computing the grbner bases of @xmath3 and then the normal forms of quaternionic polynomials . this can always be done using the classical theory and algorithm on non - commutative grbner bases @xcite . for a fixed number @xmath4 of quaternionic variables , one can compute a grbner base of the syzygy ideal @xmath5 in the @xmath4 variables , and for every degree @xmath6 , the number of degree-@xmath6 polynomials in the computed grbner base is finite . after investigating the elements in the grbner bases for small values of @xmath4 and @xmath6 , we come up with some observations on the general form of the grbner base for arbitrary @xmath4 and @xmath6 , and of the normal forms of quaternionic polynomials with respect to the grbner base . the main part of this paper is the establishment of two theorems that provide explicitly ( 1 ) a reduced grbner base of the syzygy ideal @xmath5 of quaternionic polynomials in @xmath4 pure imaginary quaternionic variables , ( 2 ) the structure of the normal forms relative to the grbner base . the results are extremely simple . for example , each polynomial in the grbner base has either 2 or 4 terms ; the sequence of variables in a monomial of normal form always forms two intertwining non - descending subsequences in terms of the ordering of the variables ( figure [ normal ] and figure [ normal:2 ] of this paper ) . the proofs of the theorems are very long and are divided into many cases . we provide detailed proofs for several cases for illustration , but skip the rest by limitation of the size of this paper . the results in this paper are not only foundations for symbolic manipulations of quaternionic polynomials , but also helpful in exploring general results on the clifford product among vectors of arbitrary dimension @xcite , now that the dimension - three case is solved here as a by - product . let @xmath7 be a set of symbols representing quaternionic variables , and let @xmath8 be the corresponding quaternionic conjugates . a _ vector _ @xmath0 refers to a pure imaginary quaternion , _ i.e. _ , @xmath9 , and a _ scalar _ @xmath10 refers to a real quaternion , _ i.e. _ , @xmath11 . a _ monic quaternionic monomial _ in variables @xmath7 refers to the quaternionic product of a permutation of the elements in a multiset of the @xmath12 symbols @xmath13 . a _ quaternionic polynomial _ with rational coefficients is a linear combination of monic quaternionic monomials with coefficients in the rational numbers field @xmath2 . we always use juxtaposition of elements to represent their quaternionic product . for a monomial @xmath14 , its _ reversion _ is ^:=_i_k_i_2_i_1 . the quaternionic product is the quotient of the @xmath2-tensor product modulo some _ syzygy ideal_. since in symbolic manipulation , a scalar is characterized by its commutativity with everything , we first consider characterizing vectors and their quaternionic products by the syzygy ideal , then extend the results to general quaternionic variables . let @xmath15 be vector variables . they belong to a 3d real inner - product space with metric diag@xmath16 . as the standard basis @xmath17 can not be constructed from the given vector variables without using square - root and division operations , they are not allowed in manipulating the quaternionic polynomials of the @xmath18 . let @xmath19 . for a monic quaternionic monomial @xmath20 , the _ leading variable _ refers to @xmath21 , and the _ trailing variable _ refers to @xmath22 . the monomial is said to be _ non - descending _ if @xmath23 , and is said to be _ ascending _ if @xmath24 . the _ degree _ , or _ length _ , of the monomial is @xmath25 . the _ degree _ of a quaternionic polynomial is the biggest degree of its monomials disregarding the @xmath2-coefficients . the _ lexicographic ordering _ among monomials is always assumed . the _ leading term _ of a quaternionic polynomial is the term of highest order . the _ inner product _ of two vectors ( not necessarily vector variables ) @xmath26 is defined by : = ( _ i_j+_j_i)/2 . [ inner ] the result is a scalar , so it commutes with a third vector @xmath27 : @xmath28\v_k=\v_k[\v_i\v_j]$ ] . the _ vector product _ of two vectors @xmath26 is defined by _ i_j:=(_i_j-_j_i)/2 . the result is a vector , so its inner product with a third vector @xmath27 is a scalar . define : = ( _ i_j_k-_k_j_i)/2 . then @xmath29=[(\v_i\times \v_j)\v_k]$ ] is a scalar , so for a fourth vector @xmath30 , @xmath29\v_l=\v_l[\v_i\v_j\v_k]$ ] . define : = ( _ i_j_k_l+_l_k_j_i)/2 . [ thm:1 ] let @xmath15 be @xmath4 different symbols , where @xmath31 . define the product among them , denoted by juxtaposition of elements , as the @xmath2-tensor product modulo the two - sided ideal generated by the following tensors : -.15 cm : & .1 cm _ j_i_i ; + v3 : & .1 cm ( _ i_j+_j_i)_k-_k(_i_j+_j_i ) ; + v4 : & .1 cm ( _ i_j_k-_k_j_i)_l + & .8 cm -_l(_i_j_k-_k_j_i ) , [ eq0 ] for any @xmath32 in @xmath33 . denote by @xmath34 the @xmath2-algebra defined by the above product and generated by the @xmath18 . denote : = q(\{_i_l+_l_i , _ i_j_k-_k_j_i|ijk } ) . [ q : field ] then ( 1 ) the @xmath18 and @xmath35 span a 3d @xmath36-vector space @xmath37 ; + ( 2 ) if all the elements in ( [ q : field ] ) extending @xmath2 are real numbers , and the inner product of real vector space @xmath37 induced from ( [ inner ] ) is definite , then the product among the @xmath18 is the quaternionic product , _ i.e. _ , @xmath34 is the quaternionic polynomial ring generated by the @xmath18 as vector variables . the proof is composed of the following steps : _ step 1_. for all @xmath38 , _ i_i_j & = & _ j_i_i ; + ( _ i_j+_j_i)_k & = & _ k(_i_j+_j_i ) ; + ( _ i_j_k-_k_j_i)_l & = & _ l(_i_j_k-_k_j_i ) . [ eq1 ] the proof is easily established by verifying that v3 for @xmath39 , and v4 for @xmath40 or @xmath41 are in the ideal generated by elements in ( [ eq0 ] ) . so the center of the ring @xmath34 is the subring generated by the @xmath42 $ ] , @xmath28 $ ] , @xmath29 $ ] for all @xmath43 . obviously , the latter two equalities in ( [ eq1 ] ) can be written as the following _ shift invariance _ of @xmath29 $ ] and @xmath44 $ ] : = [ _ k_i_j ] , .5 cm [ _ i_j_k_l ] = [ _ l_i_j_k ] . [ eq2 ] _ step 2_. since @xmath45 , @xmath46 and @xmath47 for all @xmath43 are not in the ideal generated by elements v2 , v3 , v4 , @xmath42 $ ] , @xmath28 $ ] , @xmath29 $ ] for all @xmath43 are nonzero . consider the following _ cramer s rule _ : for all @xmath48 , _ l-[_i_j_l]_k+[_i_k_l]_j-_i[_j_k_l]=0 . [ cramer:1 ] by the first equality in ( [ eq2 ] ) , the left side of ( [ cramer:1 ] ) can be simplified to @xmath29\v_l-\v_l[\v_i\v_j\v_k]$ ] , which equals zero by the last equality in ( [ eq1 ] ) . so if @xmath49 , the @xmath18 span a 3d @xmath36-vector space . _ step 3_. the following relations can be verified using ( [ eq1 ] ) : ( _ i_j)_k & = & [ _ j_k]_i-[_i_k]_j ; + [ ( _ i_j)(_k_l ) ] & = & [ _ i_l][_j_k]-[_i_k][_j_l ] ; + ( _ i_j)(_k_l ) & = & [ _ j_k_l]_i-[_i_k_l]_j . [ eq3 ] notice that in the corresponding formulas for the vector algebra over @xmath50 , the signs on the right side of ( [ eq3 ] ) are opposite . in fact , the expressions in the three lines of ( [ eq3 ] ) are respectively equal to @xmath51 ; \hskip .2 cm \frac{\v_i\v_j\v_k\v_l-\v_k\v_l\v_i\v_j}{2}.\ ] ] if @xmath52 , it is easy to see that @xmath53 , and @xmath54 are not in the defining ideal of @xmath34 , so @xmath55 , and -.1 cm ( _ 1_2)_1_2 ] & -.2cm=&-.2 cm ( _ 1_2)(_1_2 ) + & -.2 cm = & -.2 cm ( _ 1_2_1_2+_2_1_2_1 - 2_1_1_2_2)/4 + & -.2cm&-.2 cm 0 . [ n2 ] then @xmath56 are linearly independent over @xmath36 . by ( [ eq1 ] ) and ( [ eq3 ] ) , the following _ cramer s rule _ can be verified : for all @xmath57 , -.2 cm [ ( _ i_j)_k_l]_m-[(_i_j)_k_m]_l + .1 cm + [ ( _ i_j)_l_m]_k -[_k_l_m](_i_j ) & -.2cm=&-.2 cm 0 . [ cramer:2 ] hence the @xmath18 and @xmath58 span a 3d @xmath36-vector space . _ step 4_. the inner product defined by ( [ inner ] ) can be extended bilinearly to all elements in @xmath37 . then @xmath59 is orthogonal to both @xmath60 and @xmath61 . furthermore , the inner products of the three elements with themselves separately are all nonzero . when the @xmath42 $ ] , @xmath28 $ ] , @xmath29 $ ] are all real numbers , let @xmath62| } , \hskip .24 cm \e_2=\v_1\times \v_2/\sqrt{|[(\v_1\times \v_2)(\v_1\times \v_2)]|},\ ] ] and let @xmath63 . then @xmath64 form an orthogonal basis of @xmath37 , and @xmath65=\pm 1 $ ] . furthermore , @xmath66=-\e_1 ^ 2\e_2 ^ 2 $ ] . so the metric of @xmath37 is either diag@xmath16 or diag@xmath67 . _ step 5_. it is easy to prove @xmath68 and @xmath69 . moreover , @xmath70 . so if @xmath71 for @xmath72 , then the product in @xmath37 is the quaternionic product . as a consequence , all polynomial identities among the @xmath18 in the ring @xmath34 are generated by relations of the form @xmath28\v_k=\v_k[\v_i\v_j]$ ] and @xmath29\v_l=\v_l[\v_i\v_j\v_k]$ ] , or equivalently , by relations of the form @xmath29=[\v_k\v_i\v_j]$ ] and @xmath44=[\v_l\v_i\v_j\v_k]$ ] . neither the assumption that the @xmath42 $ ] , @xmath28 $ ] and @xmath29 $ ] are real numbers , nor the assumption that the metric of @xmath37 is definite , can be represented by polynomial equalities of the @xmath18 . henceforth we refer to the ring @xmath34 defined in theorem [ thm:1 ] as the _ quaternionic polynomial ring _ generated by vector variables @xmath18 , and call the defining ideal of @xmath34 the _ syzygy ideal _ , and denote it by @xmath73 $ ] . let @xmath74 . define : = ( _ k+(-1)^k _ k)/2,.4 cm a(_k):=(_k-(-1)^k _ k)/2 . [ prop:1 ] @xmath75-[\v_k]\v_i$ ] is in @xmath73 $ ] . the commutation relation @xmath75=[\v_k]\v_i$ ] can also be written as the following shift - invariance : = [ _ k_i ] . the @xmath76 are in the 3d @xmath36-vector space spanned by the @xmath18 and @xmath58 . the following identities can be established by induction , although not very easy : for all @xmath77 , -.2 cm _ j_1_j_2_j_2l ] & -.2cm= & -.2cm_i=2 ^ 2l ( -1)^i [ _ j_1_j_i][_j_2_j_3_j_i_j_2l ] ; + a(_2l-1 ) & -.2cm= & -.2 cm _ ( 2l-2,1)_2l-1 [ _ 2l+1_(1 ) ] _ 2l-1_(2 ) ; + a(_2l ) & -.2cm= & -.2 cm _ ( 2l-2,2)_2l [ _ 2l_(1)]a(_2l_(2 ) ) ; + [ _ 2l+1 ] & -.2cm= & -.2 cm _ ( 2l-2,3)_2l+1 [ _ 2l+1_(1)][_2l+1_(2 ) ] , [ eq4 ] where ( i ) @xmath78 denotes that @xmath79 does not occur in the sequence , ( ii ) @xmath80 is a bipartition of the @xmath81 elements in the sequence @xmath82 into two subsequences @xmath83 and @xmath84 of length @xmath85 and @xmath86 respectively , ( iii ) in @xmath87 $ ] , the product of the @xmath85 elements in the subsequence is denoted by the same symbol @xmath83 , ( iv ) the summation @xmath88 is over all such bipartitions of @xmath82 , and the sign of permutation of the new sequence @xmath89 is assumed to be carried by the first factor @xmath87 $ ] of the addend . all the conclusions of the proposition follow ( [ eq4 ] ) and ( [ eq1 ] ) . alternatively , we can use existing results on clifford algebras to deduce ( [ eq4 ] ) directly . in the proof of theorem [ thm:1 ] , we have shown that the @xmath18 and @xmath58 span a 3d @xmath36-vector space . to show the difference between the quaternionic product and the clifford product among the vectors @xmath18 we need to reduce the field @xmath36 . [ lemma1 ] let @xmath34 be the @xmath2-algebra defined by ( [ eq0 ] ) . then for vectors @xmath90 , = -|_i_1_i_4 ] & [ _ i_1_i_5 ] & [ _ i_1_i_6 ] + [ _ i_2_i_4 ] & [ _ i_2_i_5 ] & [ _ i_2_i_6 ] + [ _ i_3_i_4 ] & [ _ i_3_i_5 ] & [ _ i_3_i_6 ] | . [ lem1:eqn ] using the second equality of ( [ eq3 ] ) in the reverse direction , we can write the right side of ( [ lem1:eqn ] ) after expanding the determinant as @xmath91[(\v_{i_2}\times\v_{i_3})(\v_{i_5}\times\v_{i_6})]\\ & + [ \v_{i_2}\v_{i_4}][(\v_{i_1}\times\v_{i_3})(\v_{i_5}\times\v_{i_6})]\\ & -[\v_{i_3}\v_{i_4}][(\v_{i_1}\times\v_{i_2})(\v_{i_5}\times\v_{i_6})]\\ = & -[\{[\v_{i_2}\v_{i_3}(\v_{i_5}\times\v_{i_6})]\v_{i_1 } -[\v_{i_1}\v_{i_3}(\v_{i_5}\times\v_{i_6})]\v_{i_2}\\ & \hfill + [ \v_{i_1}\v_{i_2}(\v_{i_5}\times\v_{i_6})]\v_{i_3}\}\v_{i_4 } ] . \ea\ ] ] by ( [ cramer:2 ] ) , the above expression can be simplified to @xmath92 @xmath93[(\v_{i_5}\times \v_{i_6})\v_{i_4}]$ ] , which equals the left side of ( [ lem1:eqn ] ) . let : = q(\{_i_l+_l_i , |1i , ln } ) . [ q : field2 ] then when @xmath52 , by ( [ n2 ] ) , @xmath56 span a 3d inner - product space over @xmath94 . when @xmath49 , let : = [ _ 1_2_3 ] . then @xmath95 , and @xmath29=\iota\{(\iota[\v_i\v_j\v_k])/(\iota\iota)\}\in \hat{\mathbb q}\iota$ ] . the center of @xmath34 as a @xmath94-algebra is @xmath96 . by multiplying ( [ cramer:1 ] ) with @xmath97 , we get that the @xmath18 span a 3d @xmath94-vector space @xmath98 . furthermore , from ( [ cramer:2 ] ) we get that the @xmath99 are in @xmath98 . the inner product between the @xmath18 can be extended bilinearly to an inner product in @xmath98 . in the @xmath94-clifford algebra @xmath100 over @xmath98 , by theorem 5.64 , theorem 5.65 and proposition 6.22 in @xcite , the two sides of the four equalities in ( [ eq4 ] ) are different expressions of the grade-0,1,2,3 part of @xmath101 , @xmath102 , @xmath101 , @xmath103 respectively . in @xmath100 , @xmath97 is not a scalar , but a grade-3 element . in the case of quaternions , the quaternionic product is a homomorphic image of the clifford product because of v2 , v3 , so ( [ eq4 ] ) is automatically satisfied . the right side of the third equality in ( [ eq4 ] ) when multiplied by @xmath97 , becomes a vector in @xmath98 , while the right side of the last equality when multiplied by @xmath97 , becomes an element in @xmath94 . these discussions lead to the following conclusion : in the @xmath2-algebra @xmath34 defined by ( [ eq0 ] ) , the @xmath18 and @xmath104 span a 3d @xmath94-vector space @xmath98 . for any @xmath105 , @xmath106 and @xmath107 are both in @xmath98 . _ in clifford algebra , by proposition 6.23 of @xcite , the outer product ( or exterior product ) of four vectors @xmath108 , @xmath30 equals one quarter of v4 . so when the field is extended to @xmath94 , v4 states that the dimension of the @xmath94-vector space spanned by the @xmath109 is at most three . from this aspect , v2 , v3 , v4 also generate the _ syzygy ideal _ of @xmath100 , if the latter is taken as a @xmath2-algebra . let @xmath110 be quaternionic variables , and let @xmath111 be their quaternionic conjugates . to simplify notation , we introduce @xmath112 to represent any of @xmath113 , and define @xmath114 . for a monomial @xmath115 , its _ conjugate _ is defined by : = _ i_k_i_2_i_1 . define the _ scalar part _ $ ] and the _ vector part _ @xmath117 of @xmath118 as following : : = ( + ) /2 , .4 cm a():=(-)/2 . then @xmath119 $ ] commutes with @xmath120 , and the quaternionic product among vectors @xmath121 is characterized by ( [ eq0 ] ) . from theorem [ thm:1 ] and proposition [ prop:1 ] we can deduce the following : [ thm:2 ] let @xmath7 and @xmath8 be @xmath12 different symbols , where @xmath31 . let @xmath112 be one of @xmath113 , and let @xmath122 . define the product among the @xmath12 symbols , denoted by juxtaposition of elements , as the @xmath2-tensor product modulo the two - sided ideal generated by the following tensors : -.15 cm : & -.15cm_i_i-_i_i ; + q1 : & -.15cm(_i+_i)_j - _ j(_i+_i ) ; + q2 : & -.15 cm _ i_i_j- _ j_i_i ; + q3 : & -.15cm(_i_j+_j_i)_k -_k(_i_j+_j_i ) ; + q4 : & -.15cm(_i_j_k+_k_j_i)_l + & .4 cm -_l(_i_j_k+_k_j_i ) , [ eq5 ] for any @xmath32 in @xmath33 . denote by @xmath123 the ring defined by the above product and generated by the @xmath112 , and denote -.24 cm : = q(\{_i+_i , _ i_j+_j_i , .8 cm + _ i_j_k+_k_j_i|ijk } ) . [ q : field:2 ] then ( 1 ) the @xmath124 and @xmath125 span a 3d @xmath126-vector space @xmath127 ; + ( 2 ) if all the elements in ( [ q : field:2 ] ) extending @xmath2 are real numbers , and the inner product of real vector space @xmath127 induced from ( [ inner ] ) for the @xmath128 is definite , then the product among the @xmath112 is the quaternionic product , with @xmath129 as the quaternionic conjugate of @xmath130 , _ i.e. _ , @xmath123 is the quaternionic polynomial ring generated by the quaternionic variables @xmath130 . similarly , we refer to the ring @xmath123 defined in theorem [ thm:2 ] as the _ quaternionic polynomial ring _ generated by quaternionic variables @xmath130 , and call the defining ideal of @xmath123 the _ syzygy ideal _ , and denote it by @xmath131 $ ] . [ prop:2 ] for all @xmath132 , let @xmath133 . then @xmath134-[\q]\p_j$ ] is in the syzygy ideal @xmath131 $ ] . @xmath135-[\q\p_j]$ ] is also in the syzygy ideal , and @xmath117 is in the 3d @xmath126-vector space @xmath127 . let @xmath136 be vector variables . for two quaternionic monomials @xmath137 in the variables , @xmath138 is said to be _ reduced _ with respect to @xmath139 , if @xmath138 is not a _ multiplier _ of @xmath139 , or @xmath139 is not a _ factor _ of @xmath138 , _ i.e. _ , there do not exist monomials @xmath140 , including rational numbers , such that @xmath141 . for two quaternionic polynomials @xmath142 and @xmath143 , @xmath142 is said to be _ reduced _ with respect to @xmath143 , if the leading term of @xmath142 is reduced with respect to that of @xmath143 . the term _ non - reduced _ " means the opposite . let @xmath144 be a set of quaternionic polynomials . a set of quaternionic polynomials @xmath145 is said to be a _ reduced grbner base _ of the ideal @xmath146 generated by the @xmath147 in the quaternionic polynomial ring , if ( 1 ) @xmath148 , ( 2 ) the leading term of any element in @xmath3 is a multiplier of the leading term of some @xmath149 , ( 3 ) the @xmath149 are pairwise reduced with respect to each other . the _ reduction _ of a polynomial @xmath142 with respect to a reduced grbner base @xmath150 is the repetitive procedure of dividing the first non - reduced term @xmath151 of @xmath142 by some @xmath149 whose leading term is a factor of @xmath151 , then updating @xmath142 by replacing @xmath151 with its remainder , until all terms of @xmath142 are reduced . the result is called the _ normal form _ of @xmath142 with respect to the grbner base . by the classical theory of non - commutative grbner base @xcite , a reduced grbner base always exists for the syzygy ideal of quaternionic polynomials . however , the grbner base can be computed by computers only for small values of @xmath4 ( number of variables ) and @xmath6 ( degree of polynomials in the grbner base ) . our goal is to find the general form of the grbner base for arbitrary @xmath4 which can even be a symbol . we first consider the multilinear case . in a _ multilinear monomial _ , every variable occurs at most once . multilinear polynomial _ is the sum of multilinear monomials with rational coefficients , where for every variable , its degree is the same in different monomials . the _ multilinear product _ is defined for two multilinear polynomials only when they do not have any common variable . multilinear addition _ is defined only for multilinear polynomials having the same set of variables . all multilinear polynomials together with the multilinear product and multilinear addition form an algebraic system , called the _ multilinear ring_. all monic multilinear monomials together with the multilinear product form another algebraic system , called the _ multilinear monoid_. an _ ideal _ of a multilinear ring or multilinear monoid is defined just as in the usual tensor - product case . the concepts of _ reduced grbner base _ , _ reduction _ and _ normal form _ have similar definitions , the theory and algorithm of non - commutative grbner base in @xcite are still valid . for example , for @xmath152 vector variables @xmath153 , the _ multilinear quaternionic polynomial ring _ @xmath154 $ ] is composed of all multilinear quaternionic polynomials whose variables are among the @xmath22 , including polynomials of degree 0 . the syzygy ideal @xmath155 @xmath156 $ ] of the ring is generated by elements of the form @xmath157\v_{i_c}-\v_{i_c}[\v_{i_a}\v_{i_b}]$ ] or @xmath158\v_{i_d}-\v_{i_d}[\v_{i_a}\v_{i_b}\v_{i_c}]$ ] for all @xmath159 . the grbner base of the syzygy ideal is denoted by @xmath160 $ ] . when the variables @xmath153 are replaced by another sequence @xmath161 , the syzygy ideal of @xmath162 $ ] , and the grbner base of the syzygy ideal are obtained from those of @xmath154 $ ] simply by a replacement of variables . from this aspect , what is important is the number @xmath81 of variables , not the specific sequence of variables . we introduce @xmath163 , @xmath164 , and @xmath165 to denote the multilinear quaternionic ring , its syzygy ideal , and the grbner base of the syzygy ideal , in @xmath81 _ unspecified vector variables_. notice that if @xmath165 is reduced , then it is a subset of a reduce grbner base of @xmath166 , as any polynomial of degree @xmath167 in @xmath166 is also in @xmath164 . [ main:1 ] let @xmath168 $ ] be the syzygy ideal of the multilinear ring @xmath169 $ ] of multilinear quaternionic polynomials in @xmath4 vector variables @xmath170 . \(1 ) [ grbner base ] the following are a reduced grbner base of @xmath168 $ ] : for all @xmath171 , @xmath172-[\v_{i_1}\v_{i_3}\v_{i_2}]$ ] , and @xmath173-[\v_{i_2}\v_{i_3}\v_{i_1}]$ ] ; @xmath174 -{[}\v_{i_2}\v\v_{i_1}\v_{i_3}]$ ] , where @xmath175 , and @xmath176 . -.63 cm \(2 ) [ normal form , see figure [ normal ] ] in a normal form , every term is up to coefficient of the form @xmath177 or @xmath178 , where + ( i ) @xmath179 , + ( ii ) @xmath180 is ascending , + ( iii ) every @xmath181 is an ascending monomial of length @xmath182 , + ( iv ) @xmath183 ( or @xmath184 if @xmath185 occurs ) is ascending , + ( v ) for every @xmath186 , if @xmath187 is the trailing variable of monomial @xmath181 , then @xmath188 . _ remark_. in a normal form , as shown in figure [ normal ] , although @xmath189 is lower than the trailing variable @xmath187 of @xmath181 , it is allowed that the order of @xmath189 be at anywhere from above @xmath190 up to below @xmath187 , and both @xmath191 and @xmath192 are possible for any variable @xmath193 in @xmath194 where @xmath195 . before proving the theorem , we introduce the following convenient term : let @xmath196 be the leading terms of polynomials @xmath197 respectively , if @xmath198 equals @xmath199 up to coefficient , then @xmath197 are said to have _ leading like terms_. .2 cm observe that in the sequence g3 , g@xmath81 for @xmath200 , the leading terms are pairwise reduced with respect to each other . the proof of the theorem is by induction , and is composed of many steps : _ step 1_. g3 , g@xmath81 for @xmath201 generate the ideal @xmath166 . the proof is the following . by proposition [ prop:1 ] , g3 , g@xmath81 are in the ideal . it is easy to show that for all @xmath32 , @xmath29-[\v_k\v_i\v_j]$ ] is reduced to zero by g3 , and @xmath44-[\v_l\v_i\v_j\v_k]$ ] is reduced to g4 up to sign by g3 . so all the generators of @xmath166 are in the ideal generated by g3 , g4 . _ step 2_. the theorem obviously holds for @xmath202 . for @xmath203 , in step 1 it is shown that for any @xmath204 , f= _ _ _ + _ _ _ + , [ form:1 ] where g3 stands for a polynomial of type g3 , and the @xmath205 s are rational numbers . notice that the leading variable of g3 is always @xmath206 or @xmath207 . assume that the leading term of @xmath142 is reduced with respect to g3 , g4 . further assume that before cancelling like terms on the right side of ( [ form:1 ] ) , the leading term is @xmath208 up to coefficient . then @xmath209 is not reduced with respect to g3 , g4 , so it must be cancelled by other leading terms of the addends on the right side of ( [ form:1 ] ) . the following observations on ( [ form:1 ] ) are obvious : \(i ) if @xmath209 comes from @xmath210 , _ i.e. _ , equals @xmath211 , then it is not the leading like term of either @xmath212 or @xmath213 . so @xmath209 can not come from @xmath210 . \(ii ) if @xmath209 comes from some @xmath214 , then it is not the leading like term of any other @xmath215 , so it has to be the leading like term of a unique @xmath216 . conversely , if @xmath209 comes from some @xmath216 , then it has to be the leading like term of a unique @xmath214 . based on the above two observations , we get that in @xmath208 , @xmath217 and @xmath218 . so either @xmath219 or @xmath220 . _ step 3_. for @xmath203 , we have the following conclusion : _ 4 3= _ i=1 ^ 3 _ i _ i + _ j=1 ^ 4 _ j _ j. [ prove:1 ] to prove this we consider only @xmath221 . if ( [ prove:1 ] ) holds for @xmath222 , then by interchanging subscripts 1,2 , the right side of ( [ prove:1 ] ) keeps the same form . @xmath223 the underlines give hint on the category of grbner base elements to be used in reduction . _ step 4_. by ( [ prove:1 ] ) , ( [ form:1 ] ) can be rewritten as f=_i<4 _ i _ i + _ j _ j _ j+.[step4:f ] by the analysis in step 2 , the leading term on the right side of ( [ step4:f ] ) before cancelling like terms remains the leading term after cancelling like terms , violating the assumption that the leading term of @xmath142 is reduced with respect to g3 , g4 . this proves that g3 , g4 form a grbner base of @xmath224 . _ step 5_. let @xmath225 be a term in the normal form of a degree-4 polynomial . by reduction with g3 , if @xmath226 then @xmath227 is mandatory . so the sequence of subscripts must be one of @xmath228 , @xmath229 , @xmath230 , and @xmath231 , where @xmath232 is a single subscript , each @xmath233 is an ascending sequence of subscripts , all the @xmath233 form a big ascending sequence of subscripts , and @xmath232 is smaller than the trailing subscript of @xmath233 . if @xmath234 , then it must be that @xmath235 , which is not reduced with respect to g4 . so @xmath236 . this proves the theorem for @xmath203 . _ step 6_. assume that the theorem holds for all @xmath237 , where @xmath238 . when @xmath239 , let @xmath240 and assume that its leading term is reduced with respect to g3 , g@xmath81 for @xmath81 up to @xmath241 . by induction hypothesis , the degree of @xmath142 must be @xmath241 . we need to prove that @xmath142 does not exist . since the generators of @xmath242 all have degree @xmath243 , it must be that f=___f_+ _ _ f _ _ , [ form:2 ] where the @xmath205 s are coefficients , and @xmath244 are of degree @xmath245 . we do reduction to @xmath246 with respect to the grbner base g3 , g@xmath81 for @xmath81 up to @xmath245 in the following special way : to every term @xmath247 of @xmath246 , let @xmath248 where @xmath249 is a monomial of degree @xmath250 and @xmath251 is the coefficient , do reduction to monomial @xmath249 by grbner base g3 , g@xmath81 for @xmath81 up to @xmath252 . after the reduction to every @xmath249 , @xmath253 , where @xmath254 , and @xmath255 . every term of @xmath256 after removing its trailing variable is in normal form . now do reduction to @xmath256 by g3 , g@xmath81 for @xmath81 up to @xmath245 . then @xmath257 , where @xmath258 are monomials of degree @xmath182 . the leading term of @xmath256 must be the leading term of a unique @xmath259 , because it can only be non - reduced with respect to such a polynomial . now substitute the above expressions of @xmath256 , @xmath246 into ( [ form:2 ] ) , and regroup the terms . we finally get f = s+t , s=_m < q__m _ _ m _ _ m m , t=__h _ _ , [ form:3 ] and where @xmath260 is a monomial of length @xmath261 , @xmath262 , with the following properties : \(a ) if the leading term @xmath263 of an addend @xmath264 is of the highest order among all leading terms in the addends of @xmath265 , then @xmath263 must be the leading term of @xmath265 , and has no leading like term in any other addend of @xmath265 . \(b ) furthermore , when @xmath266 and the leading term of @xmath267 is @xmath268 , then the degree-@xmath269 factor @xmath270 of @xmath263 is in normal form . _ step 7_. in ( [ form:3 ] ) , the leading term @xmath263 of @xmath265 comes from a unique @xmath264 of @xmath265 . we will establish the following conclusion : if @xmath263 has any leading like term in @xmath271 , then _ _ m m=_jq _ * _ _ j _ _ j j + _ _ w _ _ , [ establish ] where the @xmath272-summation is over all @xmath273 whose leading term has lower order than that of @xmath274 . once ( [ establish ] ) is established , then we can rewrite @xmath142 as @xmath275 , where = _ j < q__j _ _ j _ _ j j , = _ _ _ _ , [ form:4 ] such that either @xmath276 , or the leading term of @xmath277 does not have leading like term in @xmath278 . the two observations in step 2 are also valid in the current situation . the leading term of @xmath277 can only be cancelled by a term of @xmath278 , and conversely , the leading term of @xmath278 can only be cancelled by a term of @xmath277 . furthermore , the leading terms of @xmath277 and @xmath278 are both non - reduced with respect to g3 , g@xmath41 for @xmath41 up to @xmath245 . since the leading term of @xmath142 is reduced with respect to g3 , g@xmath41 for @xmath41 up to @xmath245 , it must have higher order than any term of @xmath277 and @xmath278 ; then it must be the leading term of g@xmath241 , violating the assumption that it is reduced with respect to g@xmath241 . this proves that g3 , g@xmath41 for @xmath41 up to @xmath241 is a grbner base once ( [ establish ] ) is true . _ step 8_. once it is proved that g3 , g@xmath41 for @xmath41 up to @xmath241 is a grbner base , then for a polynomial @xmath142 of degree @xmath241 , by reduction with respect to g3 , g@xmath41 for @xmath41 up to @xmath245 , we get that for any term @xmath279 in the result of reduction , where @xmath280 is a monomial of degree @xmath252 , the subscripts of @xmath281 and @xmath282 must be both in one of the following normal forms : @xmath228 , @xmath229 , @xmath230 , @xmath231 , @xmath283 . if the subscripts of @xmath284 are not in one of the normal forms , it is easy to show that ( 1 ) the subscripts of @xmath281 must be of the form @xmath229 , and @xmath285 ; ( 2 ) if @xmath286 , where @xmath287 is of length @xmath288 , then @xmath289 ; ( 3 ) @xmath290 . the only possibility is @xmath291 , the leading term of g@xmath241 . consider polynomial g@xmath241 . by reduction with respect to g3 , g@xmath41 for @xmath41 up to @xmath245 , we get g@xmath292 , where each term of @xmath143 has leading variable @xmath60 or @xmath61 . the second term @xmath293 and the terms of @xmath143 are each in a normal form described by the current theorem . hence , after reduction with respect to the g@xmath241 in the above form , each term of @xmath142 becomes normal . _ step 9_. we start to prove ( [ establish ] ) , which is the main part of the whole proof . let @xmath294 , and let the leading term of @xmath267 be @xmath295 , then up to scale @xmath296 is the leading term of @xmath265 , and monomial @xmath297 is in normal form , _ i.e. _ , its subscripts are in one of the forms @xmath298 and @xmath299 . let @xmath300 be the leading subscript and trailing subscript of @xmath233 respectively . then @xmath301 . let the middle part of @xmath233 after removing @xmath300 be @xmath302 . the tail part of the subscripts @xmath303 or @xmath304 @xmath305 must be either the subscripts @xmath306 from the leading term of g3 for @xmath307 , where @xmath308 and @xmath309 , or the subscripts @xmath310 from the leading term of g@xmath41 for @xmath311 , where @xmath312 is ascending . so subscripts @xmath313 @xmath314 where @xmath315 , belong to the leading term of @xmath316 , while subscripts @xmath317 @xmath318 where @xmath319 , belong to the leading term of @xmath320 @xmath321 , where @xmath41 is the length of @xmath322 plus 3 . we get _ _ m=_w_y_1_z_1_y_k-1_z_k-1_l_k_c_k . if @xmath260 is not in normal form , then by reduction with g3 , g@xmath41 for @xmath41 up to @xmath261 , the order of @xmath260 is decreased , and ( [ establish ] ) is trivially satisfied . below we assume that @xmath260 is normal . since @xmath263 has leading like term in @xmath271 , @xmath323 for @xmath315 , and @xmath324 for @xmath319 , must be non - reduced with respect to g3 , g@xmath41 for @xmath41 up to @xmath245 . it is easy to deduce @xmath325 , and when any of the two monomials is non - reduced with respect to g3 , the length of @xmath228 is less than 3 . so up to scale , t_s=_w_y_1_z_1_z _ w_y_1_z_1_y_2_z . only the head part and tail part of @xmath263 are non - reduced with respect to some g@xmath326 , g@xmath41 respectively , where @xmath327 are between 3 and @xmath245 . there are the following cases : \(i ) @xmath328 . the subscripts of @xmath263 are either @xmath329 or @xmath330 , where @xmath331 and @xmath332 . the first case is considered in step 3 and is conclusion ( [ prove:1 ] ) . the second case includes six different subscripts of @xmath263 : 51423 , 51432 , 52413 , 52431 , 53412 , 53421 . \(ii ) @xmath333 , @xmath334 . the subscripts of @xmath263 are either @xmath335 or @xmath336 , where @xmath331 and @xmath337 . ( ii.1 ) in the former case , the subscripts of @xmath263 are @xmath338 , where @xmath339 , and @xmath340 is an ascending sequence of length @xmath341 . ( ii.2 ) in the latter case , @xmath342 , and @xmath343 is a permutation of 1,2,3 . so @xmath263 has the following possible subscripts : @xmath344 , @xmath345 , and @xmath346 , where @xmath347 , and @xmath348 is an ascending sequence of length @xmath349 . \(iii ) @xmath350 , @xmath351 . the subscripts of @xmath263 are @xmath352 , where @xmath353 is an ascending sequence of length @xmath349 , @xmath354 but @xmath355 , and @xmath332 . obviously @xmath356 , but @xmath357 can take any value from 1 to @xmath245 . so @xmath263 has the following possible subscripts : @xmath358 , @xmath359 , @xmath360 , @xmath361 , and @xmath362 for @xmath363 . there are the following subcases , considering both the symmetries of the subscripts and the difference of the categories of grbner base elements to be used in reduction : ( iii.1 ) the subscripts of @xmath263 are @xmath364 , @xmath365 , or @xmath366 . ( iii.2 ) the subscripts of @xmath263 are @xmath367 , @xmath368 , or @xmath369 , where @xmath287 is an ascending sequence of length @xmath370 , and @xmath371 . ( iii.3 ) the subscripts of @xmath263 are @xmath372 . ( iii.4 ) the subscripts of @xmath263 are @xmath373 , where @xmath287 is an ascending sequence of length @xmath374 , and @xmath375 . ( iii.5 ) the subscripts of @xmath263 are @xmath376 , where @xmath287 is an ascending sequence of length @xmath370 , and @xmath371 . ( iii.6 ) the subscripts of @xmath263 are @xmath377 , where @xmath378 , @xmath287 is an ascending sequence of length @xmath379 , @xmath380 is an ascending sequence of length @xmath381 , and @xmath382 is an ascending sequence . \(iv ) @xmath350 , @xmath334 . the subscripts of @xmath263 are @xmath383 , where @xmath384 are two ascending sequences of length @xmath385 , @xmath386 but @xmath355 . so @xmath387 , and the subscripts of @xmath263 are @xmath388 , where @xmath287 is an ascending sequence of length @xmath389 , @xmath380 is an ascending sequence of length @xmath390 , and @xmath391 is an ascending sequence . there is no space to present the analyses of all the cases and subcases . we select to present the details for the following representative cases : ( i ) , ( ii.1 ) , ( iii.2 ) to ( iii.4 ) . _ step 10_. the polynomials on the left side of the following equalities are in @xmath392 : for monomials @xmath393 of degree @xmath394 respectively , \ { ( _ k+(-1)^k_k^)_j -_j(_k+(-1)^k_k^ ) & = & 0 , + _ j_k-_k_j -(-1)^j+k(_j^_k^-_k^_j^ ) & = & 0 . . [ reduction ] as long as @xmath395 , and by applying the above equalities to the leading term of a polynomial @xmath142 , the order of @xmath142 is decreased , this transformation of @xmath142 can always be replaced by reduction with respect to g3 , g@xmath326 for @xmath326 up to @xmath396 . let both @xmath397 and @xmath398 , where @xmath399 is a sequence of subscripts of length @xmath400 . the following order - reducing transformation is also induced by g3 , g@xmath326 for @xmath326 up to @xmath401 : _ i_3_a_i_1_i_2 & = & _ i_2(_a_i_1_i_3+(-1)^a _ i_3_i_1_a^ ) + & & -(-1)^a _ i_1_a^_i_3_i_2 . [ reduction:2 ] the proof of ( [ reduction:2 ] ) is simple : @xmath402 in later steps , ( [ reduction ] ) and ( [ reduction:2 ] ) are used directly in reduction . _ step 11_. consider case ( i ) of step 9 . let @xmath403 @xmath404 . then @xmath405 when the subscripts 1,2,3 undergo a permutation , the polynomial equalities used in the reduction remain in the same category : @xmath406 for @xmath407 ; @xmath408 for @xmath409 ; @xmath410 . so the above reduction is valid for all the six subcases of case ( i ) . _ step 12_. case ( ii.1 ) of step 9 . let @xmath411 , where @xmath339 , and @xmath340 is an ascending sequence of length @xmath341 . then @xmath412 _ step 13_. case ( iii.2 ) of step 9 . let @xmath413 @xmath414 , where @xmath287 is an ascending sequence of length @xmath370 , and @xmath371 . then @xmath415 in the reduction rules , @xmath416g4 stands for @xmath417g4 , and so for the rest . the reductions for the other two subcases of case ( iii.2 ) are similar . _ step 14_. case ( iii.3 ) of step 9 . let @xmath418 . then @xmath419 _ step 15_. case ( iii.4 ) of step 9 . let @xmath420 @xmath421 , where @xmath287 is an ascending sequence of length @xmath374 , and @xmath375 . then @xmath422 [ main:2 ] let @xmath423 $ ] be the syzygy ideal of the quaternionic polynomial ring @xmath424 $ ] in @xmath4 vector variables @xmath170 . @xmath426-[\v_{i_1}\v_{i_3}\v_{i_2 } ] , & { [ } \v_{i_3}\v_{i_1}\v_{i_2}]-[\v_{i_2}\v_{i_3}\v_{i_1}],\\ \v_{i_2}\v_{i_2}\v_{i_1}-\v_{i_1}\v_{i_2}\v_{i_2 } , & \v_{i_2}\v_{i_1}\v_{i_1}-\v_{i_1}\v_{i_1}\v_{i_2 } ; \\ \ea\ ] ] \(2 ) [ normal form , see figure [ normal:2 ] ] in a normal form , every term is up to coefficient of the form @xmath427 @xmath428 or @xmath429 , where + ( i ) @xmath179 , + ( ii ) @xmath180 is non - descending , + ( iii ) @xmath430 is non - descending , + ( iv ) every @xmath181 is a non - descending monomial of length @xmath431 , + ( v ) @xmath432 ( or @xmath433 if @xmath185 occurs ) is non - descending , + ( vi ) for every @xmath186 , @xmath434 , + ( vii ) for every @xmath186 , if the length of @xmath181 is nonzero , let @xmath187 be the trailing variable of @xmath181 , then @xmath435 . * _ sketch of the proof._*since the generators v2 , v3 , v4 of @xmath423 $ ] are homogeneous polynomials , if @xmath436 , @xmath437 $ ] , then by writing @xmath438 , where each @xmath439 is a homogeneous polynomial , and different @xmath143 s have different multisets of variables , we get that each @xmath440 $ ] . so we simply consider homogeneous polynomials of degree @xmath441 in @xmath442 vector variables , where @xmath443 . let the multiset of the variables in such polynomials be @xmath444 , whose number of elements is @xmath6 . the elements of @xmath444 are specifications of @xmath6 different vector variables @xmath445 . the definitions of multilinear product and multilinear addition of the multilinear ring for @xmath445 are valid for @xmath446 after minor revisions . let @xmath447 $ ] be the syzygy ideal of the multilinear ring @xmath448 $ ] of multilinear polynomials whose variables by counting multiplicity are in multiset @xmath446 . we only need to prove the theorem for @xmath447 $ ] . most steps of the proof of theorem [ main:1 ] are still valid , in particular , ( [ establish ] ) is valid , whose proof follows a similar classification as in step 9 . although there are more subtle cases to consider , the reductions for the new situations are much the same as for the cases of step 9 .

there is presently a great interest in the study of the radiative heat transfer between closely placed objects @xcite . the reason for that is the experimental verification of the prediction that the radiative heat transfer between two bodies can be greatly enhanced when they are brought sufficiently close to each other @xcite . this enhancement can lead to overcome the far - field limit set by the stefan - boltzmann law for black bodies and it takes place when the gap between the two objects is smaller than the thermal wavelength ( 9.6 @xmath0 m at room temperature ) . such an enhanced thermal radiation exchange originates from the contribution of evanescent waves that dominate the near - field regime and by now it has been observed in numerous experiments @xcite . these experiments have in turn triggered off the hope that near - field radiative heat transfer ( nfrht ) could lead to new applications in the context of thermal technologies such as thermophotovoltaics @xcite , heat - assisted magnetic recording @xcite , scanning thermal microscopy @xcite , nanolithography @xcite or thermal management @xcite , just to mention a few . the theoretical description of nfrht is usually done in the framework of fluctuational electrodynamics ( fe ) introduced by rytov in the 1950s @xcite . in this framework , it is assumed that the thermal radiation is generated by random , thermally activated electric currents in the interior of a material . these currents vanish on average , but their correlations are given by the fluctuation - dissipation theorem @xcite . thus , the technical problem involved in the description of the radiative heat transfer between two finite objects is to find the solution of the stochastic maxwell equations with random electric current as radiation sources . this problem can be quite challenging and an analytical solution can only be obtained in a few cases with simple geometries like , for instance , two parallel plates @xcite , two spheres @xcite or a sphere in front of a plate @xcite . in general , in order to solve this problem for complex geometries , which is usually necessary for a comparison with the experiments , one has to resort to numerical methods . in this respect , a lot of progress has been done in recent years and standard numerical methods in electromagnetism have already been combined with fe to describe nfrht between objects of arbitrary size and shape ( for a review see ref . those methods include , among others , the scattering matrix approach @xcite , the finite - difference time- and frequency - domain methods @xcite , the boundary element method @xcite , and volumen - integral - equation methods @xcite . recently francoeur and coworkers adapted the discrete dipole approximation ( dda ) to describe the nfrht between two optically isotropic objects of arbitrary shape @xcite . this new approach has been termed as thermal discrete dipole approximation ( tdda ) . it describes the nfrht between two bodies in the framework of fe by discretizing the objects in terms of point dipoles in the spirit of the dda method that is widely used for describing the scattering and absorption of light by small particles @xcite . the goal of this work is to generalize the tdda method to describe the radiative heat transfer and the thermal emission in magneto - optical ( mo ) systems . mo objects are of great interest in the context of nfrht . thus for instance , in a recent work by some of us we showed that the nfrht between two parallel plates made of doped semiconductors can be largely tuned by applying a static external magnetic field @xcite . doped semiconductors under a magnetic field present a sizable mo activity , that can be controlled by changing the magnitude and the direction of the field . more recently , it has been predicted that the nfrht between several mo particles , also made of doped semiconductors , can lead to the appearance of striking phenomena such a near - field thermal hall effect @xcite or the existence of a persistent directional heat current @xcite . motivated by these recent developments in the context of nfrht of mo systems , we present here a generalization of the tdda method to describe the radiative heat transfer between mo objects of arbitrary shape , something that is still missing in the literature . to be precise , we present a tdda method that is valid for optically anisotropic systems that can be described by an arbitrary electric permittivity tensor ( with @xmath1 ) . we also extend the tdda approach to describe the thermal emission of a finite object ( both mo and non - mo ) . moreover , we correct some basic errors of previous formulations of the tdda method and use our generalized approach to provide a rigorous demonstration of the kirchhoff law relating the emissivity and absorptivity of a finite mo object . finally , we also provide the correct expression for the radiative heat transfer between mo dipoles , which is the basis to study many - body effects in systems of non - reciprocal particles . our work is focused on the analysis of homogeneous mo objects with constant temperatures , but it can be straightforwardly generalized to more complicated situations including the description of nonhomogeneous materials or situations with complicated temperature profiles . our work provides a practical method to investigate many interesting effects in the context of the thermal emission and nfrht between mo systems . the rest of the paper is organized as follows . in section [ sec - dda ] we review the dda method for mo systems that we use as a starting point for our generalized tdda approach . section [ sec - dipoles ] is devoted to the description of the radiative heat transfer between mo dipoles . in section [ sec - tdda ] we describe in detail our generalized tdda approach for the description of the radiative heat transfer between two mo bodies . in section [ sec - te ] we show how the tdda can be used to describe the total thermal emission of an arbitrary finite object . in section [ sec - kirchhoff ] we address the issue of the kirchhoff law of thermal radiation of mo objects . in particular , we derive here the definition of the directional , polarization - dependent emissivity that has to be compared with the corresponding absorption cross section to prove kirchhoff s law . in section [ sec - results ] we present the numerical results obtained for the thermal radiation and radiative heat transfer between different objects of arbitrary size and shape . this section illustrates both the validity and the capabilities of the formalisms developed in the previous sections . we conclude the paper in section [ sec - conclusions ] with some additional remarks about our tdda approach and the main conclusions of our work . on the other hand , we have included two appendixes where we present some convergence tests and an explanation of the main differences between our tdda approach and previous formulations ( appendix a ) and an alternative derivation of the main result of section [ sec - tdda ] ( appendix b ) . our formulation of the tdda is based on a dda extension to describe the electromagnetic response of mo systems that has been recently put forward by one of us @xcite . to make this paper more self - contained , we present here a brief description of this method . let us consider a nonhomogeneous finite object characterized by a spatially - dependent dielectric permittivity tensor @xmath2 that is embedded in a homogeneous medium , which hereafter is assumed to be vacuum . in the absence of currents inside the object , the electric field is given by the solution of the volume - integral equation @xcite @xmath3 \mathbf e(\mathbf r^{\prime } ) d\mathbf r^{\prime } .\ ] ] here , @xmath4 is source electric field or the field in the absence of the object , @xmath5 is the magnitude of the vacuum wave vector , @xmath6 is the volume of the object , and @xmath7 is the vacuum dyadic green tensor given by @xcite @xmath8 , \end{aligned}\ ] ] where @xmath9 , @xmath10 , and @xmath11 denotes the exterior product . in the dda approach , the previous integral equation is solved by discretizing the volume @xmath6 as @xmath12 , where @xmath13 is the volumen of an homogeneous region where the electric field is assumed to be constant . thus , eq . ( [ eq - vie ] ) now reads @xmath14 \mathbf e(\mathbf r_n ) v_n .\ ] ] defining the dipole moments as @xmath15 \mathbf e(\mathbf r_n ) , \ ] ] we can rewrite eq . ( [ eq - vie - dda ] ) as @xmath16 where @xmath17 it can be shown that @xcite @xmath18 and @xmath19 here , @xmath20 is the so - called electrostatic depolarization dyadic that depends on the shape of the volume element @xmath13 @xcite . for the case of a parallelepiped of volume @xmath21 , @xmath20 adopts the form @xcite @xmath22_{ij } = \delta_{ij } \frac{2}{\pi } \arctan \left ( \frac{1}{l^2_{n , i } } \frac{v_n}{\sqrt{l^2_{n , x } + l^2_{n , y } + l^2_{n , z } } } \right ) . \nonumber\ ] ] thus , we can now rewrite eq . ( [ eq - vie - dda - p ] ) for the internal field , @xmath23 , as follows @xmath24 \right ] \mathbf e_n & = & \mathbf e_{0,n } + \nonumber \\ k^2_0 \sum_{m \neq n } \hat g_{nm } [ \hat \epsilon(\mathbf r_m ) - \hat 1 ] v_m \mathbf e_m , & & \end{aligned}\ ] ] where @xmath25 , @xmath26 and @xmath27 . the left hand side of eq . ( [ eq - en ] ) can be defined as the exciting field @xmath28 , i.e. , the field that excites the @xmath29-volumen element . now , defining the polarizability tensor of the @xmath29-volumen element , @xmath30 , as @xmath31 where @xmath32^{-1 } \right)\ ] ] is the quasistatic polarizability tensor , eq . ( [ eq - en ] ) can be rewritten as a set of coupled dipole equations for the exciting fields at each element @xmath33 it is worth stressing that this dda formulation includes automatically the so - called radiative corrections @xcite , which are related to the imaginary part of the green tensor , and it is thus fully consistent with the optical theorem . on the other hand , from the solution of eq . ( [ eq - exc ] ) , which constitutes a set of 3@xmath34 coupled linear equations for the exciting fields , one can get the dipole moments and the total internal fields as follows @xmath35^{-1 } \mathbf p_n .\end{aligned}\ ] ] let us conclude this section with a few useful remarks . first , for cubic volumen elements , the depolarization tensor is diagonal : @xmath36 . for volumen elements of spherical or cubic , optically isotropic materials , the polarizability tensor is diagonal : @xmath37 with @xmath38 finally , from the knowledge of the dipole moments and the internal fields , one can easily compute the different cross sections ( scattering , absorption , and extinction ) . in particular , the absorption cross section , which will play an important role later on , can be obtained as follows . assuming a plane - wave illumination , @xmath39 , the absorption cross section is given by @xcite @xmath40^{\ast } \right\ } .\ ] ] before presenting our generalized formulation of the tdda for arbitrary bodies , it is instructive to first discuss the radiative heat transfer between magneto - optical ( mo ) dipoles . this discussion will allow us to highlight the peculiarities of mo systems and to pinpoint some basic errors in both the existent formulation of the tdda @xcite and in the discussion of the radiative heat transfer between mo particles @xcite . the problem that we address in this section is the radiative heat transfer ( both in the near and in the far field ) between two mo particles that are small compared to their thermal wavelengths such that they can treated as point electrical dipoles . let us assume that these two dipoles ( or dipolar particles ) are located in positions @xmath41 and @xmath42 , they have arbitrary polarizability tensors @xmath43 and @xmath44 , and they are at temperatures @xmath45 and @xmath46 . to compute the net power exchanged between the two dipoles , we first compute the power dissipated in dipole 2 due to the emission from dipole 1 , @xmath47 , assuming that dipole 2 does not emit @xcite . this power is given by @xmath48 where @xmath49 is the local electric current density in the volume @xmath50 and @xmath51 is the local electric field at position @xmath52 . moreover , @xmath53 denotes the statistical average that takes into account the stochastic nature of the dipoles . let us recall that the thermal emission originates from the fluctuating part of the dipole moments . now , we can express @xmath54 and @xmath55 in terms of their fourier transforms @xmath56 using the fact that these two functions are real , one can easily show that @xmath57 since the fluctuation - dissipation theorem ( ftd ) will introduce a @xmath58-function of the type @xmath59 , we focus on the calculation of @xmath60 and in most cases we shall drop the argument @xmath61 to alleviate the notation . in order to determine the dipole moment and field , we need to solve the dda equations , see eq . ( [ eq - exc ] ) . the exciting field at the position of dipole 2 is given by @xmath62 to close this equation , we need the corresponding equation for @xmath63 , which reads @xmath64 notice that there is no source term in this case . introducing eq . ( [ eq - exc1 ] ) in eq . ( [ eq - exc2 ] ) , we arrive at @xmath65 where @xmath66^{-1 } .\ ] ] the source field @xmath67 , due to the emission of the fluctuating part of dipole 1 , @xmath68 , is given by @xmath69 . thus , @xmath70 where we have defined @xmath71 . now , using eqs . ( [ eq - pn ] ) and ( [ eq - en - pn ] ) , we can express the dipole moment @xmath72 and the internal field @xmath73 in terms of @xmath74 as @xmath75^{-1 } \hat \alpha_2 \mathbf e_{{\rm exc},2 } .\end{aligned}\ ] ] to compute @xmath76 it is convenient to use a matrix notation where column vectors like @xmath72 are understood as ( @xmath77 ) matrices and row vectors like @xmath78 are ( @xmath79 ) matrices . with this notation , we get rid of the dot product ( or scalar product ) in favor of matrix multiplications as follows @xmath80^{-1 } \hat \alpha_2 \right\ } . & & \end{aligned}\ ] ] now , using eqs . ( [ eq - alpha ] ) and ( [ eq - alpha0 ] ) it is easy to show that @xmath81^{-1 } \hat \alpha_2 = v_2 \left [ \hat \alpha_2 - \hat \alpha^{\dagger}_2 \left ( \hat l^{\dagger}_2/v + ik^3_0/(6\pi ) \hat 1 \right ) \hat \alpha_2 \right ] . \nonumber\ ] ] introducing this relation into eq . ( [ eq - mat ] ) and after several simple algebraic manipulations , we arrive at @xmath82 where @xmath83 notice that @xmath84 . now , making use of eq . ( [ eq - eexc2-p1 ] ) , we obtain @xmath85 the statistical average appearing in the previous equation can be determined with the help of the fluctuation - dissipation theorem ( fdt ) , which in its most general form reads @xcite @xmath86 \hat \chi_1 , \ ] ] where @xmath87^{-1}$ ] is the bose function . several remarks are pertinent at this stage . first , notice that the fdt of eq . ( [ eq - fdt - p1 ] ) involves @xmath88 , which contains two terms , see eq . ( [ eq - chi ] ) . the first one is the standard contribution , while the second one is related to the radiative correction and its contribution avoids the violation of the optical theorem @xcite . the origin of this term has been nicely explained by messina _ et al . _ @xcite and the result above is a generalization of the arguments of messina _ et al . _ to the mo case . it is worth stressing that the authors of ref . [ ] in the original formulation of the tdda ignored this term , which leads to severe errors in many situations . on the other hand , notice that in the first term in the expression of @xmath88 the combination @xmath89 appears , which for mo systems differs from @xmath90 . this latter combination was used in the work of ref . [ ] for the description of radiative heat transfer between anisotropic particles and therefore , that formulation is not valid for mo systems , as it was assumed in the work on the photon thermal hall effect of ref . [ ] . now , we can combine eqs . ( [ eq - fdt - p1 ] ) , ( [ eq - im2 ] ) , and ( [ eq - im1 ] ) to write the power dissipated in dipole 2 due to the emission from dipole 1 as follows @xmath91 k^4_0 \mbox{tr } \left\{\hat c_{21 } \hat \chi_1 \hat c^{\dagger}_{21 } \hat \chi_2 \right\ } .\ ] ] following the same line of reasoning , one can show that the power absorbed by the particle 1 due to the emission of particle 2 is given by @xmath92 k^4_0 \mbox{tr } \left\{\hat c_{12 } \hat \chi_2 \hat c^{\dagger}_{12 } \hat \chi_1 \right\ } , \ ] ] where @xmath93 can be obtained from @xmath94 by interchanging the indexes 1 and 2 . it is straightforward to show that @xmath95 thus , the net power exchange , @xmath96 , between the two particles is given by the landauer - like formula @xmath97 { \cal t}(\omega ) , \ ] ] where @xmath98 and @xmath99 is the frequency - dependent transmission function given by @xmath100 we emphasize that this result differs from that reported in refs . [ ] and it only coincides with those results for the case of non - mo particles . it is also important to remark that it can be shown that in the case of isotropic particles , it reduces to the result reported in the literature in which radiative corrections have been taken into account @xcite . in this section we present our generalized tdda for the description of the radiative heat transfer between mo objects of arbitrary shape and for arbitrary separations . in particular , we focus here on the case of two finite objects assumed to be at fixed temperatures @xmath45 and @xmath46 and do not consider the interaction with any thermal bath . in the spirit of the dda , we assume that these two bodies are described by a collection of @xmath101 ( body 1 ) and @xmath102 ( body 2 ) electrical point dipoles . each dipole is characterized by a volumen @xmath103 and a polarizability tensor @xmath104 , where @xmath105 indicates to which body the dipole belongs and @xmath106 if @xmath107 and @xmath108 if @xmath109 . the information about the individual dipole moments and the internal electrical field can be grouped into column supervectors ( denoted with a bar ) as follows @xmath110 @xmath111 the same notation will be used for the exciting and source electric fields . the calculation of the net radiative power exchanged by arbitrary objects is a straightforward generalization of the calculation for the case of two dipoles presented in the previous section and we shall follow the same strategy . thus , we first compute the power absorbed by object 2 due to the thermal emission of object 1 , @xmath47 , which is given by @xmath112 following the previous section , the goal is now to compute the exciting field @xmath113 . for this purpose , we use eq . ( [ eq - exc ] ) , which can be rewritten in a more compact form using the notation above @xmath114 here , @xmath115 with @xmath116 where @xmath117_{ij}$ ] is the vacuum dyadic green tensor connecting dipole @xmath118 in body @xmath119 with dipole @xmath120 in body @xmath121 and @xmath122 with @xmath105 . solving now for @xmath123 , we have @xmath124 where @xmath125^{-1 } = \\ & & \left ( \begin{array}{cc } \bar 1 - k^2_0 \delta \bar g_{11 } \bar \alpha_1 & - k^2_0 \delta \bar g_{12 } \bar \alpha_2 \\ -k^2_0 \delta \bar g_{21 } \bar \alpha_1 & \bar 1 - k^2_0 \delta \bar g_{22 } \bar \alpha_2 \end{array } \right)^{-1 } . \nonumber\end{aligned}\ ] ] the source field , @xmath126 , due to the emission of the fluctuating dipoles in body 1 , @xmath127 , is given by @xmath128 thus , @xmath129 from eq . ( [ eq - dbar ] ) it is easy to show that @xmath130^{-1 } \delta \bar g_{12 } \bar \alpha_2 \right]^{-1 } , \nonumber \\ \bar d_{21 } & = & k^2_0 \bar d_{22 } \delta \bar g_{21 } \bar \alpha_1 \left [ \bar 1 - k^2_0 \delta \bar g_{11 } \bar \alpha_1 \right]^{-1 } .\end{aligned}\ ] ] therefore , @xmath131 where @xmath132^{-1 } .\ ] ] now , we use eqs . ( [ eq - pn ] ) and ( [ eq - en - pn ] ) to express the dipole moment @xmath133 and the internal field @xmath134 in terms of @xmath135 as @xmath136 where @xmath137^{-1 } , \dots , [ \epsilon_0 v_{n_2,2}(\hat \epsilon_{n_2,2 } - \hat 1)]^{-1})$ ] . following the same steps as in the previous section , it is straightforward to show that @xmath138 where @xmath139 with @xmath105 . let us recall that @xmath140 is given by eq . ( [ eq - chi ] ) . again , the statistical average appearing in the previous equation can be computed with the fdt , which now reads @xmath141 \bar \chi_1 .\ ] ] now , we can combine the last two equations to write the power dissipated in body 2 due to the emission from body 1 as follows @xmath91 k^4_0 \mbox{tr } \left\{\bar c_{21 } \bar \chi_1 \bar c^{\dagger}_{21 } \bar \chi_2 \right\ } .\ ] ] analogously , one can show that the power absorbed by body 1 due to the emission of body 2 is given by @xmath92 k^4_0 \mbox{tr } \left\{\bar c_{12 } \bar \chi_2 \bar c^{\dagger}_{12 } \bar \chi_1 \right\ } , \ ] ] where @xmath142 can be obtained from @xmath143 by interchanging the indexes 1 and 2 . on the other hand , it can be shown that @xmath144 thus , the net power exchange , @xmath96 , between the two bodies is given by the landauer - like formula @xmath145 { \cal t}(\omega ) , \ ] ] where let us recall that @xmath98 and @xmath99 is the transmission function given by @xmath146 eqs . ( [ eq - pnet - bodies ] ) and ( [ eq - t - bodies ] ) are our central result for the description of the radiative heat transfer between anisotropic objects of arbitrary shape , which is valid in particular for mo systems . obviously , the result summarized in these equations reduces to that of eqs . ( [ eq - pnet - dip ] ) and ( [ eq - t - dip ] ) for the case of two dipoles . the result of eqs . ( [ eq - pnet - bodies ] ) and ( [ eq - t - bodies ] ) is valid that arbitrary separations between the two bodies , i.e. , it includes both the near - field and the far - field contributions . in the far - field , the formula can be simplified by neglecting the multiple scattering between the two bodies . this can be done by approximating the matrix @xmath143 by @xmath147^{-1 } \delta \bar g_{21 } \left[\bar 1 - k^2_0 \bar \alpha_1 \delta \bar g_{11 } \right]^{-1 } .\ ] ] let us conclude this section by saying that our derivation of the result for the radiative heat transfer between two finite objects is based on an intuitive division of the problem into two subproblems in which one object is assumed to be the emitter and the other one the receiver , and in each subproblem the receiver is assumed to not radiate . this is indeed the strategy followed by polder and van hove in their seminal paper in which they computed the radiative heat transfer between two parallel plates @xcite . anyway , one might wonder if this _ ad hoc _ division is fully justified in the problem addressed in this section . to demonstrate that this is indeed the case , we present in appendix [ sec - derivation ] an alternative derivation of the central result of this section starting from a different point of view in which both objects radiate simultaneously " and the net exchanged power is directly calculated . this alternative derivation detailed in appendix [ sec - derivation ] confirms the validity of the central result of this section summarized in eqs . ( [ eq - pnet - bodies ] ) and ( [ eq - t - bodies ] ) . the thermal emission of a finite object is another important issue that can be described with the help tdda , but it has not been done so far . the goal of this section , and the next one , is to fill this gap . we divide our discussion of the formulation of thermal radiation within tdda into two parts . the first one will be addressed in this section , where we present a convenient way to compute the total radiation power emitted by a finite object . the analysis of the angular and polarization dependence of the thermal emission will be carried out in the next section , where , in particular , we generalize kirchhoff s law of thermal radiation to the case of mo objects of arbitrary size and shape . the problem we want to address in this section is the calculation of the total radiation power emitted by a finite body at temperature @xmath148 . in this case , we model this body as a collection of @xmath34 point dipoles that interact with the electric field of a thermal bath at temperature @xmath148 . the total power emitted by the body , @xmath149 , must be equal to the total power absorbed from the bath when they are at the same temperature . thus , we can follow the previous section and write the emitted power as @xmath150 where we are using the same type of notation as in the previous section , i.e. , @xmath151 in this problem , the source field is the field of the bath , i.e. , @xmath152 , where we have used the notation of previous equation . thus , the exciting field @xmath153 is simply given by @xmath154 where @xmath155^{-1}$ ] . here , @xmath115 , where @xmath156_{ij}$ ] is the vacuum dyadic green tensor connecting dipole @xmath118 with dipole @xmath120 inside the body and @xmath157 . as in the previous section , it is straightforward to show that @xmath158 where @xmath159 . now , to compute the statistical average appearing in the previous equation , we make use of the fdt theorem for the field correlations of the bath , which reads @xcite @xmath160 \mbox{im } \ { \bar g \ } .\ ] ] now , we can combine the last two equations to write the emitted power as follows @xcite @xmath161 where @xmath162 is planck distribution function . in the previous equation one can identify @xmath163 as a quantity that , when divided by the geometrical cross section of the body , plays the role of an effective angular - averaged frequency - dependent emissivity . in the case of a single dipole , @xmath164 , which for a spherical , optically isotropic dipole ( @xmath165 ) reduces to @xmath166 $ ] , where @xmath167 is given by eq . ( [ eq - alpha - sphere ] ) . another fundamental question that we want to address in this work is the validity of kirchhoff s law for finite mo objects . this thermal radiation law states that the emissivity is equal to the absorptivity . while this law is well established in the case of extended systems , it is only very recently that a rigorous proof for arbitrary finite objects has been reported @xcite . in the case of a finite object , the absorptivity equals the absorption cross section . this proof was provided for reciprocal objects and our goal is to extend the analysis of the validity of this law to the case of mo ( non - reciprocal ) objects . in its most general form , kirchhoff s law involves the thermal emission in a given direction and for a given polarization . thus , our first task is to find a proper expression for the polarization - dependent directional emissivity . for this purpose , it is convenient to write the total emitted power in a way different than above , i.e. , in terms of poynting s vector as @xmath168 which describes the integrated flux across a differential section @xmath169 perpendicular to a radial unit vector @xmath170 , performed over a sphere of radius @xmath171 enclosing the emitting object . since this quantity does not depend on the actual choice of @xmath171 , for convenience , we will evaluate poynting s vector in the far field . now , we need to compute the statistical average of poynting s vector resulting from the thermal emission of an arbitrary body , i.e. , @xmath172 , where @xmath52 is the point of observation ( outside the object ) and @xmath51 and @xmath173 are the electric and magnetic field , respectively . as usual , we can express this average in terms of the fourier transforms of the fields as @xmath174 anticipating that the fdt theorem will introduce a @xmath58-function of the type @xmath59 , we shall focus on the calculation of @xmath175 and we shall not write explicitly the argument @xmath61 . it is worth noting that @xmath176 , in virtue of the superposition principle , can always be described in terms of two linearly polarized fields that are perpendicular to each other , both lying on a plane perpendicular to @xmath177 , as @xmath178 . in the spirit of the tdda , we assume that our body is described by a collection of @xmath34 point dipoles located at positions @xmath179 . using the notation of previous sections , the electric field generated by these dipoles is given by @xmath180 where we have defined the following row supervector of green tensors @xmath181 as usual , @xmath4 is the source field that originates from the fluctuating dipoles inside that body and is given by @xmath182 the total dipole moments @xmath183 can be obtained from the exciting field inside the body as @xmath184 , while @xmath153 fulfills the dda equation @xmath185 where @xmath186 . solving for @xmath153 , we obtain that @xmath187 from which it is straightforward to show that the total field is given by @xmath188 where @xmath189 in the far - field region , the magnetic field is simply related to the electric field via @xmath190 where @xmath191 . now , to select the contribution of a given polarization , we write the electric field as @xmath192 , where @xmath193 is a unit vector defining the polarization state and lies in a plane perpendicular to the radial direction @xmath177 . in our matrix notation , the field amplitude reads @xmath194 and we have @xmath195 which with the help of eq . ( [ eq - etotal ] ) can be written as @xmath196 now , we can make use of the fdt to write @xmath197 \bar \chi , \ ] ] which leads to the following expression for the radial component of poynting s vector for a polarization given by the unit vector @xmath193 @xmath198 let us stress that in the previous expression one must use the far - field expression of the green tensors , i.e. , @xmath199 where @xmath200 , @xmath201 , and @xmath202 . taking into account the standard definition of the total emitted power @xmath203 we can write @xmath204 equation ( [ eq - totempowerangpol ] ) can be written as @xmath205 where @xmath206 is the polarization - dependent directional emissivity given by @xmath207 let us stress that , as defined here , this emissivity has dimensions of area like the absorption cross section . on the other hand , it is worth noting that eq . ( [ eq - totemissionda ] ) permits establishing the following relationship between the total emissivity @xmath208 and the polarization - dependent directional emissivity @xmath206 @xmath209 where it is easy to verify that in the case of isotropic @xmath210 it reduces to @xmath211 . the quantity @xmath206 has to be compared with the absorption cross section for the same polarization and incident angle to verify kirchhoff s law . the expression for the absorption cross section is given by eq . ( [ eq - cabs1 ] ) , which can be written as @xmath212 where @xmath213 is a linearly polarized plane wave @xmath214 impinging as @xmath215 . a general demonstration of kirchhoff s law from eqs . ( [ eq - zeta - emissivity ] ) and ( [ eq - cabs2 ] ) is cumbersome . in what follows we will provide evidence of this fulfillment in the case of single dipoles whereas a numerical verification for finite objects will be provided in section [ sec - results ] . in the case of a single dipole we have that @xmath216 in eq . ( [ eq - zeta - emissivity ] ) and @xmath217 in eq . ( [ eq - cabs2 ] ) . then , using @xmath218 , the absorption cross section can be written as @xmath219 . on the other hand , if we assume that the dipole is located at the origin of the spherical coordinate system , the green tensor in far - field approximation is given by @xmath220 and eq . ( [ eq - zeta - emissivity ] ) becomes @xmath221 finally , for an isotropic system this leads to the well - known @xmath222 kirchhoff s result . in this section we present a series of results related to the different aspects of radiative heat transfer and thermal emission discussed in previous sections . our goal is to illustrate the capabilities of the tdda method put forward in this work , as well as to demonstrate its validity by comparing with established results in the literature . for these purposes , we shall consider below two materials . first , as an example of a non - mo material we choose sio@xmath223 , which is polar dielectric that has been amply studied in the context of nfrht . we use the dielectric function of sio@xmath223 reported in palik s book @xcite . as an example of a mo material we shall consider @xmath29-doped insb , a polar semiconductor , that when subjected to an external magnetic field becomes mo . for the sake of concreteness , we shall assume that magnetic field is applied in the @xmath224-direction , @xmath225 . in this case , the permittivity tensor of insb adopts the following form @xcite @xmath226 where @xmath227 } \right ) , \nonumber \\ \label{eq - epsilons } \epsilon_2(h ) & = & \frac{\epsilon_{\infty } \omega^2_p \omega_c}{\omega [ ( \omega + i \gamma)^2 - \omega^2_c ] } , \\ \epsilon_3 & = & \epsilon_{\infty } \left ( 1 + \frac{\omega^2_l - \omega^2_t}{\omega^2_t - \omega^2 - i \gamma \omega } - \frac{\omega^2_p}{\omega ( \omega + i \gamma ) } \right ) . \nonumber\end{aligned}\ ] ] here , @xmath228 is the high - frequency dielectric constant , @xmath229 is the longitudinal optical phonon frequency , @xmath230 is the transverse optical phonon frequency , @xmath231 defines the plasma frequency of free carriers of density @xmath29 and effective mass @xmath232 , @xmath233 is the phonon damping constant , and @xmath234 is the free - carrier damping constant . finally , the magnetic field enters in these expressions via the cyclotron frequency @xmath235 . let us point out that in this case the magneto - optics is induced by the magnetic field , which modifies the diagonal elements of the permittivity tensor and introduces off - diagonal terms . there are two contributions to the diagonal components , namely , optical phonons and free carriers . these latter ones are responsible for the magneto - optics at finite magnetic field . notice that we have neglected the contribution from intra - band transitions because we are interested in thermal properties at room temperature , where this contribution does not play any role . in all calculations below , we consider the particular case taken from ref . [ ] , where @xmath236 , @xmath237 rad / s , @xmath238 rad / s , @xmath239 rad / s , @xmath240 rad / s , @xmath241 @xmath242 , @xmath243 , and @xmath244 rad / s . as a reference , let us say that with these parameters @xmath245 rad / s for a field of 1 t. the explicit form for the dielectric tensor given by eq . ( [ perm - tensor - hz ] ) gives rise to a @xmath246 , see eq . ( [ eq - chi ] ) , having the form @xmath247 for simplicity , the results presented below are for homogeneous bodies , i.e. , with a spatially constant permittivity tensor , and for constant temperatures inside the bodies . moreover , all the results for finite systems were obtained by discretizing the objects in terms of cubes of equal size . the corresponding polarizability tensors of the cubes were computed with eqs . ( [ eq - alpha ] ) and ( [ eq - alpha0 ] ) with @xmath36 , which in the case of isotropic materials reduce to eq . ( [ eq - alpha - sphere ] ) . unless stated otherwise , the calculations of the radiative thermal conductance , power emission , etc . , were carried out at @xmath248 k. the issue that we want to address now is the description of the radiative heat transfer between two finite objects using the formalism detailed in section [ sec - tdda ] . to avoid known problems related to the so - called shape error , an error due to the description of objects that can not be exactly represented by cubical lattices @xcite , we consider the heat transfer between two identical cubes of side @xmath249 . in fig . [ fig - nfrht - comparison1 ] we show as solid lines the tdda results for the spectral radiative thermal conductance ( conductance per unit of energy ) as a function of the photon energy for two sio@xmath223 cubes of side @xmath250 @xmath0 m and various separations ranging from 100 nm to 5 @xmath0 m . to demonstrate the validity of our tdda approach , we compare our results with those obtained with the code scuff - em @xcite , which are shown in fig . [ fig - nfrht - comparison1 ] as open symbols . the scuff - em solver implements the fluctuating - surface - current ( fsc ) formulation of the heat transfer problem put forward in refs . [ ] in combination with the boundary element method ( bem ) . the fsc - bem combination used in scuff - em enables the description of the radiative heat transfer between homogeneous bodies of arbitrary shape and provide numerically exact results within the framework of fluctuational electrodynamics in the local approximation , i.e. , assuming that the dielectric function only depends on frequency . cubes of side @xmath250 @xmath0 m and various gaps , @xmath251 , at @xmath248 k. the solid lines correspond to the results computed with the tdda approach with 6859 dipoles per cube , while the open symbols correspond to the results obtained with scuff - em . ] as it can be seen in fig . [ fig - nfrht - comparison1 ] , our tdda method is able to reproduce the numerically exact results obtained with scuff - em . let us say that the tdda results shown here were obtained modeling each cube with 6859 cubic dipoles ( the analysis of the convergence of these results with the number of dipoles used in the simulations is provided in appendix [ sec - comparison ] ) . it is worth stressing that it becomes progressively more demanding to converge the tdda results upon reducing the gap between the cubes . thus for instance , in this example the difference between the tdda and the scuff - em results for the total conductance ( integrated over energy ) is of 2.5% for a gap of @xmath252 @xmath0 m , while it monotonically increases up to 6.2% for @xmath253 nm . the physical reason for this behavior is the fact that the nfrht in this case is dominated by surface phonon polaritons @xcite that have a penetration depth comparable to the gap size @xcite . this implies that the electric field inside the cubes varies on a length scale smaller than the gap and therefore , one needs an increasing number of dipoles to properly describe the radiative heat transfer as the gap diminishes . the important thing is that the error in the tdda calculations can be systematically reduced by refining the dipole grid , which can be done by increasing the number of dipoles or using adaptive meshes . let us stress that in this work we are not interested in presenting a detailed analysis of the convergence of our tdda approach , but rather in establishing its fundamental validity . we present another example of the comparison between our tdda results and those obtained with scuff - em for the spectral conductance of two sio@xmath223 cubes of varying size and gap @xmath254 @xmath0 m in fig . [ fig - nfrht - comparison2 ] . again , we find a very good agreement between both types of results , which becomes progressively worse ( for a fixed number of dipoles ) as the side of the cube increases , as expected . more importantly , we always find that this agreement can be systematically improved by increasing the number of dipoles in our cubic lattices ( for more details on the convergence of these results , see appendix [ sec - comparison ] ) . so in short , we conclude that our tdda method produces the correct results for the radiative heat transfer provided that a sufficiently large number of dipoles is employed in the discretization of the objects . cubes of different side values and separated by a gap of 0.5 @xmath0 m at @xmath248 k. the solid lines correspond to the results computed with the tdda approach with 6859 dipoles per cube , while the open symbols correspond to the results obtained with scuff - em . the error in the tdda calculations in the total conductance is of 3.1% for @xmath250 @xmath0 m and it increases monotonically up to 11.3% for @xmath255 @xmath0 m . ] @xmath0 m separated by a gap of 500 nm , at @xmath248 k , and for various values of the magnetic field , @xmath256 , applied along the @xmath224-direction . the inset shows the discretization geometry , where the number of dipoles per cube is 4913 ( each one has an edge of 59 nm ) . ] we now proceed to illustrate our approach in the case of mo objects . for this purpose , we consider the radiative heat transfer between two identical insb cubic particles of side @xmath257 @xmath0 m separated @xmath258 nm using 4913 cubic dipoles for each particle . the results for the spectral conductance for different values of the external magnetic field , @xmath256 , are shown in fig . [ fig - rht_2cubesinsb ] . in this case , the magnetic field is oriented as shown in the inset of this figure . as one can see , the spectral conductance in the absence of field ( @xmath259 ) is dominated by the contribution of two peaks that , as explained in ref . @xcite , are due to the contribution of surface plasmon polaritons ( low - energy peak ) and surface phonon polaritons ( high - energy peak ) . notice that the magnetic field induces a splitting of the low - energy peak , while it introduces a new peak at high energies that appears at the cyclotron frequency and therefore , its position is blue - shifted linearly with the external field . this behavior is very similar to what it was found in ref . @xcite for the case of two semi - infinite insb plates . these results will be discussed in more detail in a forthcoming publication and we show them here just to illustrate the capabilities of our tdda method . to conclude this subsection , we want to illustrate the validity of the far - field approximation of eq . ( [ eq - ffapprox ] ) . for this purpose , we show in fig . [ fig - ff - sio2 ] a comparison between the exact results and the far - field approximation for the room temperature spectral conductance for two identical cubes of sio@xmath223 of various sizes separated a distance of 20 @xmath0 m , which is larger than the thermal wavelength . the results were computed with 2197 dipoles in each cube , which was sufficient to converge the results with an accuracy of better than 1% . as it is clear from fig . [ fig - ff - sio2 ] , the approximation of eq . ( [ eq - ffapprox ] ) accurately reproduces the exact results . cubes with a gap of 20 @xmath0 m , @xmath248 k , and for various values of the cube side , @xmath249 . the solid lines correspond to the exact results and the dashed lines to the results computed with the far - field approximation of eq . ( [ eq - ffapprox ] ) . ] in this subsection we want to discuss the total thermal emission of a finite object . to test the validity of the formalism described in section [ sec - te ] we first consider the total thermal emission of a sphere of a non - mo material . a well - known result , first derived by kattawar and eisner @xcite , is that the emissivity of a non - mo sphere of arbitrary radius obtained within fluctuational electrodynamics is equal to the corresponding absorption efficiency , i.e. , to the absorption cross section normalized by the geometrical cross section . let us recall that we demonstrated this result in section [ sec - te ] for the case of a single spherical dipole . since the absorption efficiency can be computed exactly with the help of mie theory @xcite , the result of kattawar and eisner provides a very stringent test for our theory of thermal emission . we have computed both the emissivity and absorption efficiency using eqs . ( [ eq - emissivity ] ) and ( [ eq - cabs2 ] ) , respectively , for spheres of different sizes and materials and in all cases we have found that they are indeed identical . in fig . [ fig - emissivity - sio2 ] we show the tdda results for the total emissivity of a sio@xmath223 sphere for different radii ( solid lines ) as a function of the energy . let us clarify that we are showing the quantity defined in eq . ( [ eq - emissivity ] ) and normalized by the geometrical cross section to make it dimensionless . we also show the corresponding results for the absorption efficiency calculated with the exact mie theory @xcite . as one can see , there is a very good agreement between our tdda results and the exact results , which demonstrates the validity of our formalism . again , it is worth remarking that as one increases the size of the sphere , a larger number of dipoles is required to satisfactorily converge the results . to obtain the results of fig . [ fig - emissivity - sio2 ] we used about 12000 dipoles , which is enough to reproduce the exact results for spheres of radius up to approximately 10 @xmath0 m . in any case , it is important to emphasize there is no fundamental limitation to obtain the exact result within our tdda approach , although the practical problem to converge the results may become in some cases very difficult to overcome . sphere as a function of the energy for different values of its radius @xmath171 at @xmath248 k. the solid lines correspond to the results computed with tdda with @xmath260 dipoles , while the open symbols correspond to the results obtained with mie theory . ] for completeness , we show in fig . [ fig - total - emission - sio2 ] the total power emitted at room temperature by a sphere and a cube of sio@xmath223 of varying size . the results are normalized by the corresponding results for a black body ( stefan - boltzmann law ) . as one can see , for small spheres and cubes the emitted power is proportional to the volumen of the finite objects . this is a well - known result @xcite , which is due to the fact that in this regime the skin depth of this material at the relevant wavelengths is larger than the characteristic size of the object . this means in practice that the whole object contributes to the thermal emission . however , as the size increases , the emitted power becomes proportional the area of the object , which reflects the fact that the size becomes larger than the skin depth and then only the surface of the object significantly contributes to the thermal emission . notice also that in both cases ( spheres and cubes ) , when the characteristic size becomes on the order of a few microns , the emitted power becomes comparable to that of a black body of the same size . let us say that the thermal emission of a sio@xmath223 sphere was studied by krger _ et al . _ @xcite and our results agree with those reported in those references . sphere as a function of its radius @xmath171 at @xmath248 k. the power is normalized with the black body result , @xmath261 , where @xmath262 is the total area of the sphere . ( b ) the same as in panel ( a ) , but for a sio@xmath223 cube of side @xmath249 . ] the goal of this subsection is to illustrate the validity of kirchhoff s law in the case of finite mo objects . but before presenting the full numerical analysis , it is instructive to consider a mo dipole ( where we still consider mo activity is due to a magnetic field pointing along the @xmath224-direction ) since it is possible to provide an analytical solution for an arbitrary polarization vector @xmath193 . using the coordinate system depicted in fig . [ fig - coordinatesys ] the spectral , polarization - dependent directional emissivity for a single dipole with @xmath88 given by eq . ( [ chi - tensor - hz ] ) can be written for any polarization angle @xmath263 as [ see eq . ( [ eq - kirchhoff1dip ] ) ] @xmath264.\end{aligned}\ ] ] notice that there is no dependence on the azimuthal angle @xmath265 due to the symmetry introduced by the fact that @xmath266 . moreover , since we have to choose two orthogonal vectors , i.e. , @xmath263 and @xmath267 , the sum required to obtain the total emitted power is equal to @xmath268 $ ] , that , as expected , does not depend on the choice of @xmath263 . since the direction of the external magnetic field imposes a preferential direction , it is convenient to choose two values of @xmath263 so that one polarization vector has no component along the direction of the external magnetic field , @xmath269 rad ( let us call it @xmath270 ) , while the other is collinear with the magnetic field , @xmath271 rad ( let us call it @xmath272 ) . it is straightforward to see that @xmath273 , i.e. , independent of any incidence angle @xmath274 or @xmath265 , while @xmath275 $ ] presenting thus a @xmath276 dependence , varying from @xmath277 to @xmath278 . to illustrate the fulfillment of kirchhoff s law for a body of finite size , we consider an homogeneous insb sphere of radius @xmath279 @xmath0 m , using discretization unit cubes with an edge of @xmath280 nm ( @xmath281 2550 discretization elements ) . in fig . [ fig - kirchhofflaw](a ) we present spectra of the polarized emissivity and of the absorption cross section for a direction defined by @xmath282 rad and polarization angle by @xmath283 rad . two values of magnetic field , @xmath259 t and @xmath284 t , are shown . for both cases @xmath285 and @xmath286 ( for the same angles and polarization ) are identical . in order to analyze the same system as a function of the geometrical angles , we verify the @xmath287 dependence obtained for a single dipole in eq . ( [ eq - emissivmodipolesimple ] ) . this is clearly visible in fig . [ fig - kirchhofflaw](b ) , where we plot @xmath288 and @xmath289 for @xmath290 rad , @xmath291 rad and @xmath292 ev , as a function of @xmath263 for an applied field of @xmath284 t normalized to its value at @xmath259 t ( let us point out that , since the geometry is fully isotropic , for @xmath259 t it happens that @xmath293 and thus emissivity and absorption cross section are polarization insensitive ) . the next step in the verification is to compare the angular profile of the emissivity and of the absorption cross section for fixed polarizations . as it can be understood from eq . ( [ eq - emissivmodipolesimple ] ) , the most interesting choice is the @xmath272-polarization ( @xmath271 rad ) . in fig . [ fig - kirchhofflaw](c ) , we present a colormap of @xmath294 and @xmath289 ( for the same polarization ) for @xmath284 t normalized to @xmath259 t , as a function of @xmath265 and @xmath274 showing that they are independent of @xmath265 and that the dependence on @xmath274 is of the form of @xmath276 , as evidenced by the associated polar plot . the same plot , but for @xmath270-polarization is presented in fig . [ fig - kirchhofflaw](d ) demonstrating that for this configuration and polarization , the emissivity and absorption cross section is isotropic , within a numerical error smaller than 2% . let us conclude this subsection by saying that we have analyzed the polarization - dependent , directional emissivity and the corresponding absorption cross section for different sizes and shapes of insb particles under a magnetic field finding that in all cases these two quantities are identical . in this work we have focused on the description of the thermal radiation and radiative heat transfer of homogeneous objects with constant temperatures . however , all the basic formulas derived in this work are directly applicable to the case of inhomogeneous objects , with spatially dependent dielectric functions , and they can be straightforwardly generalized to consider the case of arbitrary temperature profiles in the interior of the objects . on the other hand , the tdda method presented here can easily be generalized in a number of additional ways to treat , for instance , the radiative heat transfer between surfaces and finite objects @xcite or between periodic systems @xcite . with respect to the improvement of the efficiency and accuracy of the tdda method presented here , there are a number of obvious improvements that one can implement without the need to modify the formulation detailed in this work . thus for instance , one can employ adaptive dipole lattices to better describe the non - uniform field profiles inside the objects @xcite , one can discretize the objects in terms of non - cubic dipoles with shapes better adapted to the geometries of the objects under study @xcite , or one can use a dda formulation based on the integration of the green s tensor ( igt ) in the spirit of ref . @xcite , rather than on the standard point - dipole interaction , as done in this work . in summary , we have presented in this work a new formulation of the tdda approach to describe the radiative heat between finite objects of arbitrary size and shape . this formulation corrects the deficiencies of previous tdda methods and allows us to describe the radiative heat transfer between mo objects and , more generally , between optically anisotropic objects with arbitrary permittivity tensors . we have shown how this tdda approach can also be used to describe the thermal emission of a finite object . moreover , we have provided very compact and transparent formulas for different physical quantities that can be directly employed in generalizations of the method presented here . in particular , we have corrected the existent formulas for the radiative heat transfer between non - reciprocal dipoles , which can be used to analyze different many - body effects in ensembles of mo particles . on the other hand , we have used our tdda approach to demonstrate kirchhoff s law relating the emission and absorption of non - reciprocal objects . our work opens the way for the rigorous description of different thermal radiation phenomena involving finite mo objects of arbitrary shape . we thank v. fernndez - hurtado for providing the numerical results obtained scuff - em shown in this paper , and j.j . senz and n. de sousa for stimulating discussions . this work was financially supported by the comunidad de madrid ( contract no . s2013/mit-2740 ) . j.c.c . also acknowledges financial support from the spanish mineco ( contract no . fis2014 - 53488-p ) and thanks the dfg and sfb 767 for sponsoring his stay at the university of konstanz as mercator fellow . m . also acknowledges funding from the spanish mineco ( contract no . mat2014 - 58860-p ) . in this appendix we present a brief analysis of the convergence of the results shown in section [ sec - results - rht ] as well as a comparison of our approach with previous formulations of the tdda @xcite . in fig . [ fig - convergence ] we show some convergence tests for the radiative heat transfer between two sio@xmath223 cubes of two different sizes . in this figure , together with the exact numerical results obtained with scuff - em , we show the tdda results obtained with different number of dipoles ( per cube ) . as one can see , upon increasing the number of dipoles , the tdda results first converge rapidly to approximately the correct result and then , the convergence improves slowly , but monotonically until a very satisfactory agreement with the scuff - em results is achieved . of course , as the size of the objects increases , the required number of dipoles increases accordingly and a proper convergence becomes much more demanding . the same occurs when the gap size is reduced ( not shown here ) . in that case , the electric field inside the sio@xmath223 particles varies more rapidly in space due to the smaller penetration depth of the surface phonon polaritons that dominate the nfrht in this polar dielectric @xcite . cubes of side @xmath250 @xmath0 m and a gap size of @xmath295 @xmath0 m . the different lines correspond to the results computed with the tdda approach with different number of dipoles per cube , as indicated in the legend . the open symbols correspond to the converged result obtained with scuff - em . ( b ) the same as in panel ( a ) , but for two sio@xmath223 cubes of side @xmath296 @xmath0 m . ] let us discuss now the relationship with the formalism presented in ref . we would like to focus on eqs . ( 3.16 ) and ( 3.22 ) in ref . @xcite and its consequences . those equations indicate that the thermally generated incident field is nil for all elemental dipoles in the emitter , and are only affected by the multiple scattering with the absorbing object ( or receiver ) . this is at variance with our eq . ( [ eq - excitingfield ] ) where all the dipoles in the emitter feel the thermally activated field of all the constituent dipoles but itself and so there is also multiple scattering inside the emitter . this multiple scattering in the emitting element accounts for the fact that the thermal wave activated deep inside the emitter has to first travel across it to reach the receiver . for small objects , neglecting that multiple scattering results in small deviations from the correct result , precisely due to such small size . however , for moderately large objects such a multiple scattering plays a more important role , especially in the frequency regions in which absorption is particularly large due to a resonant behavior . this is illustrated in fig . [ fig - differences ] where we compare our results for the radiative heat transfer between two sio@xmath223 cubes of side 2 @xmath0 m separated by a distance of @xmath295 @xmath0 m with those obtained ignoring the multiple scattering in the emitter . in particular , in this comparison we have made sure that both the results based on the approach of ref . @xcite and our results are properly converged , see fig . [ fig - differences ] and the corresponding caption . thus , ignoring this multiple scattering in the emitter prevents recovering the correct result and the error becomes progressively more severe as the size of the particles increases . on the other hand , we would like to recall that another important difference between our tdda approach and the formulations of refs . @xcite resides in the form of the fluctuation - dissipation theorem employed in those references . as one can see in eq . ( [ eq - fdt - p1 ] ) , in our case this theorem involves the quantity @xmath88 , which is given by eq . ( [ eq - chi ] ) . this quantity contains the so - called radiative corrections , whose absence , as in the expressions of refs . @xcite , leads to wrong results since the optical theorem is not fulfilled ( energy is not conserved ) . our result in eq . ( [ eq - chi ] ) is a generalization to the mo ( tensorial ) case , which is valid for non - reciprocal systems . cube of @xmath296 @xmath0 m at @xmath297 k and a gap distance of @xmath295 @xmath0 m , for a number of elemental dipoles ensuring that convergency is achieved . the solid lines are the results of eqs . ( 3.16 ) and ( 3.22 ) in ref . @xcite and the open circles are the converged result of our formalism . ] given the discrepancies found with the original formulation of tdda of ref . [ ] , it is reassuring to derive the central result of eq . ( [ eq - pnet - bodies ] ) in an alternative way . for this purpose , we have made use of the formalism put forward by messina _ et al . _ @xcite for the description of heat transfer in systems of multiple dipoles . this formalism was developed to study many - body effects in the heat transfer between dipoles , but as we show below , it can be adapted in the tdda spirit to describe the heat transfer between bodies of arbitrary size . as in section [ sec - tdda ] , we consider two bodies described by a collection of @xmath101 ( body 1 ) and @xmath102 ( body 2 ) electrical point dipoles . again , each dipole has a volumen @xmath103 and it is characterized by a polarizability tensor @xmath104 , where @xmath105 . in this case , rather than separating the problem into two problems of the emission of one body and the absorption of the other , as we did in section [ sec - tdda ] , we now consider that the fluctuating dipoles in both bodies are radiating at the same time . in this case , the total dipole moments can be grouped into column supervectors as follows @xmath298 where there are two contributions to the total dipole moments , one coming from the fluctuating dipoles , @xmath299 , and the other is an induced contribution arising from the interaction with the other dipoles . this second contribution is given by @xcite @xmath300 here , the indexes @xmath118 and @xmath120 refer to dipoles in both bodies . from this equation it is easy to show that the total dipole moments are given by @xmath301 here , we have used the notation of section [ sec - tdda ] . now the goal is compute the net power balance in , let us say , body 2 . this net power is given by @xmath302 the dipole moment @xmath133 can be obtained from eq . ( [ eq - t - alt ] ) : @xmath303 while @xmath134 can be related to @xmath133 via eq . ( [ eq - en - pn ] ) . then , with the usual algebraic manipulations , one can show that @xmath304 which with the help of the fdt theorem leads to @xmath305 \nonumber .\end{aligned}\ ] ] in thermal equilibrium ( @xmath306 ) , @xmath307 should vanish . therefore , they following relation must hold @xmath308 thus , @xmath307 can be rewritten as @xmath309 { \cal t}(\omega ) , \ ] ] where @xmath310 now , the remaining task is to show that the previous expression for the transmission reduces to eq . ( [ eq - t - bodies ] ) . for this purpose , one can use eq . ( [ eq - t - alt ] ) to show that @xmath311^{-1 } \bar \alpha_1 \delta \bar g_{12 } \right]^{-1 } , \nonumber \\ \bar t^{-1}_{21 } & = & k^2_0 \bar t^{-1}_{22 } \bar \alpha_2 \delta \bar g_{21 } \left [ \bar 1 - k^2_0 \bar \alpha_1 \delta \bar g_{11 } \right]^{-1 } .\end{aligned}\ ] ] from these equations it is straightforward to show that @xmath312 from which it is obvious that eq . ( [ eq - t - new ] ) reduces to eq . ( [ eq - t - bodies ] ) . this completes our alternative derivation of eq . ( [ eq - pnet - bodies ] ) , which confirms the validity of our result . w. a. challener , c. b. peng , a. v. itagi , d. karns , w. peng , y. y. peng , x. m. yang , x. b. zhu , n. j. gokemeijer , y. t. hsia , g. ju , r. e. rottmayer , m. a. seigler , and e. c. gage , nat . photon . * 3 * , 220 ( 2009 ) . b. c. stipe , t. c. strand , c. c. poon , h. balamane , t. d. boone , j. a. katine , j. l. li , v. rawat , h. nemoto , a. hirotsune , o. hellwig , r. ruiz , e. dobisz , d. s. kercher , n. robertson , t. r. albrecht , and b. d. terris , nat . photon . * 4 * , 484 ( 2010 ) .

their unusual physical and chemical properties , and their potential applications in a variety of nanotechnologies make carbon nanotubes ( cnts ) very interesting objects to technoscientists , @xcite despite possible health hazards . @xcite in particular , cnts have been proposed to fabricate several different integrated - circuits elements and electromagnetic devices , such as transmission lines , @xcite interconnects , @xcite and nanoantennas . @xcite the fabrications of a cnt - based amplitude modulator / demodulator @xcite and a fully integrated radio receiver @xcite have been reported . references demonstrate the potential of cnts as emitters of terahertz and infrared radiation . not surprisingly , the electromagnetic characteristics of cnt - based antennas have been examined in different frequency regimes ranging from the microwave to the visible . cnt morphology has been demonstrated to play a crucial role , as evinced by reported research on single - wall cnts ( swcnts ) , @xcite @xmath0-cm - long one - dimensional chains of electrically connected swcnts , @xcite planar periodic structures of swcnts,@xcite cnt bundles @xcite , and cnt arrays.@xcite continuing in that vein , here we report our work on the performance of multiwalled cnts ( mwcnts ) as antennas . a multiwalled cnt comprises @xmath1 concentric shells ( or tubes ) , each obtained by rolling a graphene sheet into a cylinder . the number of shells in a mwcnt can range from 2 to 200 , and the distance between consecutive shells from @xmath2 to @xmath3 ,@xcite which is close to the interlayer distance in graphite ( @xmath4 ) . the lattice structures of consecutive shells are generally uncorrelated with each other , and can even have different chiralities . in fact , several experiments on mwcnts have indicated that often the different shells have different periodicities.@xcite two consecutive shells of an mwcnt are called commensurate ( incommensurate ) if the ratio of their unit - cell lengths along the cnt axis is rational ( irrational ) , indicating the presence ( absence ) of translational symmetry . incommensurability affects the transport and optical properties of mwcnts . @xcite the critical issue when modeling the electromagnetic properties of an mwcnt is the intershell interaction leading to intershell electron tunneling or hopping . published data , although very variable , shows a strong dependence on the intrinsic symmetries of the shells , which dictates selection rules for the elements of the tunneling matrix , as determined by the conservation laws for energy and momentum . as may be expected , two incommensurate shells interact differently than two commensurate cells . @xcite for example , the fermi momentums of two incommensurate shells do not coincide within the first brillouin zone and therefore the intershell tunneling vanishes . @xcite an isolated mwcnt can be modeled in several different ways . @xcite considered an mwcnt as a set of coaxial , continuous , conducting cylinders accompanied by an appropriate kronig - penney - type potential in the radial direction . dyachkov and makaev @xcite as well as tunney and cooper @xcite assumed the intershell interaction to be so small that each shell could be considered to be in a perturbed eigenstate of an swcnt . a computer simulation with some input parameters extracted from experiment has also been reported.@xcite following refs . and , we decided to neglect the contribution of the intershell tunneling to the radiation characteristics of an mwcnt . indeed , though intershell interaction in defectless finite - length double - wall cnts shifts optical band gaps and distorts the density of electronic states , it does not subvert the intrinsic type of conductivity ( either metallic or semiconductor);@xcite therefore , it would influence the radiation characteristics only slightly . as the presence of localized defects can dramatically increase intershell conduction , @xcite we restrict ourselves here to mwcnts with defect - free shells by neglecting intershell conduction . analysis of the physical characteristics of cnt - based integrated circuits elements , such as antennas , must follow the general principles of electrodynamics and must account for the peculiar dispersion properties of electrons in cnts . a key element of the analysis is the formulation of the effective boundary conditions for the electromagnetic field on the cnt surface . @xcite these boundary conditions are dictated by the microscopic model of cnt used . hence , any attempt to describe an mwcnt with cross - sectional radius between @xmath5 and @xmath6 nm as a perfectly conducting rod @xcite can provide only rough estimates , and is inapplicable to mwcnts of small radius . we adopt a proper microscopic model for mwcnts here . a relevant boundary - value problem for an mwcnt of finite length is formulated in sec . [ model ] . since the radiation characteristics of an mwcnt are determined by its waveguiding properties , the dispersion equation for guided - wave propagation on an infinitely long mwcnt is derived in sec . [ guided ] . the boundary - value problem of sec . [ model ] is solved numerically by an integral - equation technique in sec . [ finite - length ] . this approach is well established in antenna theory @xcite and has been successfully applied to swcnt antennas @xcite and almost circular bundles of closely packed swcnts . @xcite section [ num ] contains numerical results for guided - wave parameters ( slow - wave and attenuation ) and the scattering properties of mwcnt in a wide frequency range from the terahertz to the visible regimes . sections [ discus ] presents an assessment of the potential of an isolated mwcnt function as an optical nanoantenna . concluding remarks are provided in sec . [ concl ] . an @xmath7 time - dependence is implicit , @xmath8 denotes the time , and @xmath9 is the angular frequency . let us model an mwcnt as a multishell structure comprising @xmath1 coaxial , infinitesimally thin cylinders in free space . each shell is achiral , i.e. , it has either a zigzag ( @xmath10 ) or an armchair ( @xmath11 ) configuration , with @xmath12 as an integer . therefore , our model is applicable to both semiconducting and metallic shells . the sequence of armchair and zigzag shells in the mwcnt is arbitrary . let the cylindrical axis of the chosen mwcnt be parallel to the @xmath13 axis of the cylindrical coordinate system @xmath14 . the centroid of the mwcnt is assumed to coincide with the origin of the coordinate system . let us enumerate shells in the mwcnt consecutively from @xmath15 to @xmath1 beginning from the innermost shell , so that their cross - sectional radii comply with the condition @xmath16 . the cross - sectional radius @xmath17 of the outermost shell is assumed to be much smaller than the free - space wavelength . the @xmath18-th shell , @xmath19 $ ] , possesses an axial conductivity per unit area denoted by @xmath20 . the transverse current density in every shell is neglected . intershell tunneling is also neglected , as mentioned in the previous section . as electromagnetic fields with azimuthal symmetry are easily excited in an mwcnt by a uniform external field , we decided to restrict ourselves to azimuthally symmetric fields . this restriction also holds for finite - length mwcnts in the long - wavelength regime ( sec . [ num1 ] ) . the electric hertz vector @xmath21 is then governed by the helmholtz equation @xmath22 where @xmath23 is the unit vector along the @xmath13 axis , @xmath24 the free - space wavenumber , and @xmath25 is speed of light in free space ( i.e. , vacuum ) . since @xmath26 depends only on @xmath27 and @xmath13 , the components of the electric and magnetic fields are as follows : @xmath28 @xmath29 on neglecting intershell tunneling , the two boundary conditions across the @xmath18-th shell are given by @xcite @xmath30 here @xmath31 is the axial current density on the surface @xmath32 , with @xmath33 the surface conductivity @xmath34 of the @xmath18-th isolated shell is available via quantum - transport calculations carried out in the tight - binding approximation as @xcite @xmath35 where @xmath36 is the electron charge , @xmath37 is the normalized planck constant , and @xmath38 is the axial projection of quasi - momentum of an electron . the integer @xmath39 $ ] labels the @xmath40-electron energy bands and the abbreviation @xmath41 restricts the variable @xmath38 to the first brillouin zone . the time constant @xmath42 ( being the relaxation time ) is assumed to be equal to the inverse of the relaxation frequency @xmath43 . the normalized matrix elements of the dipole transition between the conduction and valence bands , denoted by @xmath44 , are evaluated in the tight - binding approximation , after taking into account the transverse quantization of the charge carriers motion and the hexagonal structure of the graphene lattice . @xcite the first term on the right side of ( [ eq4 ] ) is the drude term corresponding to the intraband conductivity , while the second term describes the contribution of interband transitions between the valence and the conduction bands . the indexes @xmath25 and @xmath45 refer to the conduction and valence bands , respectively . the equilibrium fermi distribution functions @xmath46 \right\ } ^{-1}\ ] ] and @xmath47 \right\ } ^{-1}\ ] ] involve the temperature @xmath48 and the boltzmann constant @xmath49 . the electron energies @xmath50 and normalized matrix elements of the dipole transition @xmath44 for zigzag ( @xmath10 ) cnt are given by @xcite @xmath51 and @xmath52\,,\end{aligned}\ ] ] respectively , where @xmath53 ev is the overlap integral , @xcite @xmath54 , and @xmath55 nm is the interatomic distance in graphene . expressions for @xmath56 and @xmath57 for armchair ( @xmath11 ) cnts are available elsewhere . let us recall that intershell tunneling is negligibly small so that @xmath58 can be represented as a superposition of @xmath1 independent functions ; furthermore , we seek a solution of ( [ eq1 ] ) in the form of a guided wave . accordingly , @xmath59 where @xmath60 is the set of unknown coefficients and @xmath61 is the unknown guide wavenumber . with @xmath62 , the basis function @xmath63 is taken to satisfy the differential equation @xmath64 + \kappa^2 \phi_p(\rho ) = 0\ , , \quad p\in[1,n]\,,\ ] ] subject to the following boundary conditions at the surface @xmath32 : @xmath65 the appropriate solution of is @xmath66 where @xmath67 and @xmath68 are modified bessel functions of the zeroth order . whereas the boundary condition ( [ boundary4 ] ) is automatically satisfied by ( [ eq5 - 5 ] ) and ( [ eq5 - 6 ] ) , ( [ boundary3 ] ) remains to be satisfied . substitution of ( [ eq5 - 5 ] ) into ( [ boundary3 ] ) leads to a system of @xmath1 linear homogeneous equations with respect to the coefficients @xmath69 . a nontrivial solution of the system is provided by the dispersion equation @xmath70 the elements @xmath71 of the @xmath72 matrix @xmath73 are given by @xmath74 where @xmath75 is the kronecker delta . the dispersion equation ( [ eq12 ] ) has @xmath1 @xmath76-roots , each corresponding to a guided wave in the mwcnt . from each @xmath76-root , we can determine @xmath61 and the slow - wave coefficient @xmath77 . let us now consider the scattering of an electromagnetic wave incident on a mwcnt of finite length @xmath78 . the scattered field can be described by ( [ elfield1 ] ) and ( [ elfield2 ] ) with the electric hertz vector satisfying the helmholz equation ( [ eq1 ] ) , the usual radiation condition , @xcite and the boundary conditions ( [ boundary3 ] ) and ( [ boundary4 ] ) . the surface current density @xmath79 at @xmath80 is given by @xmath81 where @xmath82 is the @xmath13-component of the incident electric field . as the intershell tunneling through the two ends of the mwcnt is negligible , the current density @xmath79 satisfies the edge conditions @xmath83 thereby expressing the absence of concentrated charges on the ends . the boundary - value problem can effectively be solved by the integral - equation technique for the surface current density as follows . @xcite the potential @xmath84 must be linearly related to @xmath79 , @xmath19 $ ] , as @xmath85 where @xmath86 is the free - space green function for ( [ eq1 ] ) . setting @xmath87 in ( [ eq15a ] ) and making use of ( [ eqj ] ) , we arrive at the following system of @xmath1 integral equations with respect to unknown current densities : @xmath88 & & \label{eq18a } { \displaystyle\quad + \,c_q \exp ( ikz ) + d_q \exp ( - ikz ) } \\ & & \nonumber = \sum\limits_{p = 1}^{n } { \int\limits _ { - l / 2}^{l / 2 } { \left [ { \frac{2\pi i r_p } { \omega } g(z - { z}',r_p , r_q ) } \right . } } \\[5pt ] & & \qquad + \left . { \frac{i\delta _ { qp } } { 2k \sigma_p } \exp ( ik\vert z - { z}'\vert ) } \right ] j_p ( { z}')d{z}'\,,\quad q\in[1,n]\ , . \nonumber\end{aligned}\ ] ] here , @xmath89 and @xmath90 are unknown constants to be determined from the edge conditions ( [ eq14a ] ) . in the long - wavelength regime ( @xmath91 ) , the electromagnetic scattering properties of the mwcnt can be encapsulated in a polarizability tensor with only one nonzero component @xmath92 the integral on the right side of ( [ eq18a ] ) can be numerically handled by a quadrature formula . @xcite parenthetically , the foregoing formalism was recently applied to almost circular , closely packed bundles of finite - length parallel , identical , metallic swcnts . @xcite a finite - length mwcnt functioning as a receiving antenna can be characterized by the antenna efficiency @xcite @xmath93 where @xmath94 is the scattered power and @xmath95 is the power lost to ohmic dissipation . according to refs . and , surface waves in a cnt are strongly attenuated in the frequency regime of interband transitions . therefore , the surface current density @xmath96 in an swcnt of radius @xmath97 exposed to an incident electric field @xmath98 obeys ohm s law @xmath99 very well , @xmath100 being the surface conductivity . equation ( [ ohm0 ] ) can be considered as the first born approximation . @xcite application of the first born approximation ( [ ohm0 ] ) to an mwcnt allows us to express the axial surface current density @xmath79 on the surface of the @xmath18-th shell as @xmath101\,.\ ] ] this expression can be justified as follows . the electric field exciting the @xmath18-th shell is made of two components : ( i ) the electric field @xmath98 incident on the entire mwcnt and ( ii ) the sum of the electric fields radiated to the surface current densities @xmath102 , @xmath103 $ ] but @xmath104 . equation ( [ ohm ] ) is justified if the first component is much larger than the second component at @xmath105 , i.e. , the condition @xmath106 & & \qquad\quad z\in\left[-l/2,l/2\right]\,,\quad p\in\left[1,n\right]\ , , \label{condition}\end{aligned}\ ] ] holds true . substituting ( [ eq15a ] ) and ( [ ohm ] ) into ( [ condition ] ) , we arrive at the condition @xmath107 & & \nonumber { \displaystyle\quad \ll \left| \frac{c}{2}\int _ { - l / 2}^{l / 2 } { e^{(0)}_{z } ( { r_p , z}')\exp ( ik\vert z - { z}'\vert ) d{z}'}\right|}\ , , \\[5pt ] & & \qquad\quad z\in\left[-l/2,l/2\right]\,,\quad p\in\left[1,n\right]\ , , \label{condition1 } \end{aligned}\ ] ] for the applicability of the first born approximation . we found that condition ( [ condition1 ] ) holds true for mwcnts that are not too thick ( i.e. , @xmath17 is sufficiently small ) in the frequency regime wherein the surface conductivities @xmath20 , @xmath19 $ ] , are mostly determined by interband transitions the second term in ( [ eq4 ] ) . we estimate that mwcnts with outermost radius @xmath108 nm satisfy ( [ ohm ] ) at frequencies in the mid - infrared and the visible regimes . equation ( [ ohm ] ) contradicts the edge conditions ( [ eq14a ] ) . an analogous situation appears , for example , in the theory of diffraction by an aperture in an infinitesimally thin , perfectly conducting screen , wherein the exact solution must satisfy the meixner condition @xcite on the aperture edge but an approximate solution based on the huygens principle does not satisfy that condition . however , the error is strongly localized in the vicinity of the edge of the aperture and does not influence the scattered field in the far zone . @xcite for our problem , ( [ ohm ] ) may be interpreted as a version of the huygens principle for an mwcnt : the scattered field is generated by secondary current densities induced by the incident electric field on the surfaces of all shells . therefore , the applicability of ( [ ohm ] ) is unphysical only in the vicinity of the edges @xmath109 , but that should not affect the performance of the mwcnt as an antenna in the interband - transitions regime . when the mwcnt is electrically thin in cross - section ( @xmath110 ) , the further approximation @xmath111 is permissible . then the substitution of ( [ ohm ] ) in ( [ eq21a ] ) and ( [ eq21b ] ) yields @xmath112 @xmath113 where @xmath114 is the _ effective _ conductivity per unit length of an electrically thin mwcnt . the condition ( [ condition1 ] ) leads to the inequality @xmath115 ; consequently , @xmath116 as determined by ( [ eq21c ] ) and ( [ eq21d ] ) , @xmath117 and @xmath118 are the same as for a thin - wire resistive antenna whose conductivity per unit length is equal to @xmath119.@xcite thus , an electrically thin mwcnt which satisfies the condition ( [ condition1 ] ) may be modeled as a thin homogenous cylinder with conductivity per unit length determined from ( [ sigm ] ) . in order to present illustrative results , let us consider an mwcnt consisting of only zigzag shells . a large number of such mwcnts are possible . @xcite moreover , after a suitable modification of the effective boundary conditions ( [ boundary3 ] ) and ( [ boundary4]),@xcite the approach developed can be extended to mwcnts comprising shells with arbitrary chirality vectors . for definiteness , calculations were performed for two types of mwcnts , hereafter referred to as type a and type m , shown in fig . the @xmath18-th shell in an mwcnt of type a is in the @xmath120 configuration ; hence , two consecutive semiconducting shells are followed by a metallic shell . the @xmath18-th shell in an mwcnt of type m is in the @xmath121 confuguration ; hence , all @xmath1 shells are metallic . the radius of the @xmath18-th shell is given by @xmath122 9\sqrt{3}pb/(2\pi)\ , , & \quad{\rm type\,\ , m } \end{array}\right.\,.\ ] ] the frequency @xmath123 of the first interband transition for the @xmath18-th shell of the metallic type is determined by the energy difference between the lowest van hove singularity of the conducting band and the highest van hove singularity of the valence band . thus , an analysis of the dispersion equation ( [ dispers1 ] ) for zigzag shells yields @xmath124/(2\pi\hbar ) \\ & \simeq & \upsilon _ f / ( \pi r_p ) \ , , \end{aligned}\ ] ] where @xmath125 is the @xmath40-electron velocity at the fermi level . the frequency of the first interband transition for the @xmath18-th shell of the semiconducting type is determined by the minimal band - gap energy as @xmath126/3]/(2\pi\hbar)\\ & \simeq & \upsilon _ f / ( 3\pi r_p ) \,,\end{aligned}\ ] ] wherein the sign have to be chosen so that @xmath127/3 $ ] is an integer . thus , the frequency of the first interband transition is given by @xmath128 the foregoing formula has been confirmed by direct numerical simulation . @xcite the frequency regime below the interband - transition regime for an mwcnt is dictated by the condition @xmath129 where @xmath130 while the condition @xmath131 establishes the frequency regime for long - range guided - wave propagation . as a result , the frequency regime wherein long - range guided waves can produce geometric ( antenna ) resonances is as follows : @xmath132 at frequencies in the regime @xmath133 , the interband transitions contribute strongly to the surface conductivity of each shell such that the real and the imaginary parts of this quantity are approximately equal in magnitude . guided waves then get attenuated heavily . at frequencies in the regime @xmath134 , attenuation of guided waves is caused by fast electron relaxation in all shells . let us now examine guided - wave propagation at @xmath135 thz in an infinitely long mwcnt of type a consisting of @xmath136 shells . the relaxation time @xmath42 is taken to be vanishingly small . the shells numbered @xmath137 are metallic . all interband transitions for these shells occur at frequencies exceeding @xmath138 thz and therefore do not contribute to the effective conductivity ( per unit length ) of the mwcnt . the imaginary part of the surface conductivity of a metallic shell is positive - valued and exceeds the real part in magnitude . thus , the necessary condition for the long - range propagation of guided waves is satisfied . the semiconducting shells labelled @xmath139 have negligible surface conductivity , and therefore do not influence the scattering properties of the chosen mwcnt at @xmath140 thz . the surface conductivities of the semiconducting shells labelled @xmath141 are dictated mainly by the corresponding first interband transitions occurring at @xmath142 thz . at @xmath143 thz . the imaginary parts of the surface conductivities of these shells are negative and several times smaller than the surface conductivities of adjacent metallic shells . therefore , the surface current densities in these semiconducting shells are smaller and oppositely directed with respect to their counterparts in adjacent metallic shells . the real parts of the surface conductivities of these semiconducting shells are high enough to cause large ohmic losses . we considered only the two eigenmodes of guided - wave propagation in the chosen mwcnt with the smallest retardation , labeled as @xmath144 and @xmath145 . they correspond to the two roots of the dispersion equation ( [ eq12 ] ) , identified as @xmath146 and @xmath147 , with the smallest real parts : @xmath148 . of all eigenmodes , these two will mostly influence the scattering properties of the finite - length mwcnt , as discussed in sec . [ num1 ] . the radial dependences of @xmath149 and @xmath150 inside the mwcnt of type a for guided waves @xmath144 and @xmath145 are shown in fig . [ fig2](a ) . clearly , the axial component of the electric field is distributed over the entire cross section of the mwcnt . the azimuthal component of the magnetic field is discontinuous across each metallic shell , in accordance with the boundary condition ( [ boundary3 ] ) . the degree of discontinuity decreases as the shell number @xmath18 increases , in compliance with the generally decreasing magnitudes of @xmath151 in fig . [ fig2](b ) . outside the mwcnt , the radial distribution of the electric and magnetic fields is governed by the argument of the modified bessel function @xmath152 , whereby we conclude that the electromagnetic field is highly localized to the mwcnt . and @xmath150 , ( b ) @xmath151 on the surfaces of the shells , and ( c ) the axial component of the time - averaged poynting vector , for guided waves @xmath144 and @xmath145 , in an mwcnt of type a with @xmath136 shells at @xmath135 thz , in the limit @xmath153 . the discrete points in ( b ) , corresponding to different shells , are joined together only to aid the eye . , width=326 ] the axial surface current density @xmath151 , @xmath19 $ ] , is shown in fig . [ fig2](b ) for guided waves @xmath144 and @xmath145 . in opposition to the radial distributions of the axial electric field , the magnitude of the current density is maximal on the innermost shell , and then strongly decreases with the increase of the shell number @xmath18 . this behavior is in agreement with the @xmath154-dependence of the surface conductivity of a metallic swcnt of radius @xmath97 . @xcite figure [ fig2](c ) shows the radial distribution of the axial component of the time - averaged poynting vector @xmath155 for @xmath144 and @xmath145 , inside the mwcnt and in the vicinity of its outermost shell . the energy - flux density for @xmath144 is maximum near the surface of the innermost metallic shell and then slightly varies about some mean value between the 4@xmath156 and the 13@xmath156 shells . the energy - flux density for @xmath145 is mostly concentrated between the two innermost metallic shells and then decreases very rapidly as @xmath157 . clearly then , the two innermost metallic shells are dominant contributors to the retardation of both @xmath144 and @xmath145 . and ( b ) @xmath158 on @xmath1 for guided waves @xmath144 and @xmath145 in mwcnts of type a and m , when @xmath135 thz and @xmath159 s. discrete points are joined by lines to aid the eye . , width=326 ] the dependences of the real part of the slow - wave coefficient @xmath160 and the ratio @xmath161 of the guided waves @xmath144 and @xmath145 on the number @xmath1 of shells in mwcnts of types a and m are shown in fig . [ fig3 ] . the retardation coefficient @xmath162 for mwcnts of type m is higher than for mwcnts of type a of comparable @xmath17 , which is a significant observation in relation to the geometric resonances of finite - length mwcnts ( sec . [ num1 ] ) . for mwcnts of type m , @xmath163 increases as @xmath1 does . as @xmath1 increases , the additional outmost shell has a lower influence on @xmath163 . the increase of @xmath163 with increasing @xmath1 occurs until @xmath17 exceeds @xmath164 , per conditions ( [ eq14 ] ) and ( [ freq ] ) . figure [ fig3](a ) shows that , for guided waves @xmath144 and @xmath145 in mwcnts of type a , @xmath163 * increases with the addition of a metallic shell , but * decreases with the addition of a semiconducting shell . for an mwcnt with small @xmath1 , @xmath162 is thus determined mostly by the metallic shells . in contrast , for mwcnts with large @xmath1 , both @xmath162 and @xmath165 are strongly affected by the semiconducting shells of large radius . this occurs because the semiconducting shells labelled @xmath166 have their first interband transitions near the chosen frequency of @xmath140 thz , and , consequently , the real and imaginary parts of their surface conductivities are large in magnitude . an increase in the number of such shells greatly diminishes the parameters @xmath163 and @xmath167 of @xmath144 and @xmath145 in the mwcnts of type a. thus , _ interband transitions suppress guided - wave propagation in mwcnts_. let us now move on to the scattering properties of finite - length mwcnts in the terahertz and the far - infrared frequency regimes . here we focus only on the case when the incident electric field is parallel to the @xmath13 axis , thereby permitting us to investigate electromagnetic effects caused by the axial surface conductivities of the shells . of mwcnts of type a and m for different lengths @xmath78 and number of shells @xmath1 . the labels @xmath168 and @xmath169 denote the first geometric resonance @xmath170 in ( [ eq15 ] ) of the guided waves @xmath144 and @xmath145 , respectively . ( a ) @xmath171 s , ( b ) @xmath172 s.,width=326 ] figure [ fig4](a ) illustrates the frequency dependence of the imaginary part of the polarizability scalar @xmath173 in the long - wavelength regime ( @xmath174 ) for different lengths @xmath78 and shell numbers @xmath1 in mwcnts of type a and m. the labels @xmath168 and @xmath169 in this figure denote the first geometric resonance of the guided waves @xmath144 and @xmath145 , respectively . the geometric resonances occur at frequencies @xcite @xmath175 as we concluded from fig . [ fig3](a ) , the retardation coefficient @xmath162 for mwcnts of type m is higher than for mwcnts of type a of comparable @xmath17 . therefore the first geometric resonance ( @xmath176 ) for the mwcnts of type m appears in fig [ fig4](a ) at a higher frequency than for mwcnts of type a , both of the same length @xmath78 . the resonance frequencies for @xmath144 and @xmath145 depend on @xmath78 nonlinearly . as an example , in fig . [ fig4](a ) the resonance frequency of the mwcnt of type a increases by a factor of about @xmath177 ( from @xmath178 to @xmath179 thz ) while the length @xmath78 decreases by a factor of @xmath180 ( from @xmath181 to @xmath182 nm ) . that is reflected in ( [ eq15 ] ) by the strong dependence of @xmath163 on @xmath183 , when @xmath183 is close to @xmath184 . accordingly , the dependence of the frequency of the geometric resonance on @xmath78 is not explicit . the first resonance of the mwcnt of type a and length @xmath185 nm at @xmath179-thz frequency is not strong , because condition ( [ eq14 ] ) is not satisfied for this mwcnt , and the guided wave is strongly attenuated . we conclude that the resonance of an mwcnt antenna of type a or m near a given frequency are the most pronounced , if both ( [ eq14 ] ) and ( [ eq15 ] ) are satisfied . figure [ fig4](a ) contains the value of the antenna efficiency @xmath186 at the first geometric resonance for all mwcnts considered . this antenna efficiency depends both on the type and the length of the mwcnt . the antenna efficiency of an mwcnt is several time larger than that of an swcnt of the same length . our calculations make us conclude that the restrictions on the mwcnt dimensions given by ( [ eq14 ] ) and ( [ eq15 ] ) do not permit an increase in @xmath186 at the first geometric resonance . in contrast , the antenna efficiency of an almost circular bundle of closely packed swcnts can be increased up to unity , by increasing of number of metallic swcnts in the bundle . @xcite when conduction in an mwcnt is very diffuse , @xcite the relaxation time @xmath42 is close to that for graphite ( @xmath187 s ) . then the conditions ( [ eq14 ] ) and ( [ eq15 ] ) can be fulfilled only for mwcnts of type m , and that too with @xmath188 nm and @xmath189 nm . figure [ fig4](b ) shows the frequency dependence of @xmath190 of an mwcnt of type m with @xmath191 for @xmath192 s. the first geometric resonances of mwcnts of lengths @xmath193 nm and @xmath194 nm appear in the far - infrared ( @xmath195 thz ) and the mid - infrared ( @xmath196 thz ) regimes , respectively . thus , antenna resonances are pronounced and can be experimentally observed for short mwcnts with several shells only in the far - infrared and mid - infrared regimes . whereas secs . [ num0 ] and [ num1 ] address the frequency regime @xmath197 wherein interband transitions are not possible , we now proceed to the interband - transitions regime of sec . [ low - coupling ] , wherein the electromagnetic response properties of mwcnts are dominated by the interband transitions . this regime , delineated by the condition @xmath198 , can very well lie in the visible part of the electromagnetic spectrum . wang _ et al . _ have published the only known report on the experimental observation of antenna resonances of mwcnts in the visible regime.@xcite in their experiment , interference patterns of light scattered from an array of finite - length mwcnts with @xmath199 nm were observed . _ attributed the observed phenomenon to a _ length - matching _ antenna effect , which requires the spatial variations of the induced current density to satisfy the edge conditions ; furthermore , they assumed that a thick mwcnt has the same scattering response in the far zone as a perfect conducting cylinder with the same cross - sectional radius and length . however , hao and hanson @xcite subsequently showed that the model of an mwcnt as a single cylindrical shell with typical surface conductivity given , e.g. , by ( [ eq4 ] ) , can not describe the experimental observation recorded in ref . . the reason is the low surface conductivity of the mwcnt shells , as given by ( [ eq4 ] ) . in the model of ref . , however , the intershell electromagnetic coupling in the mwcnt was ignored , which is inappropriate if @xmath200 nm , as we pointed out in sec . [ low - coupling ] . the lacuna in the model of ref . can be removed by implementing the integral - equation technique of sec . [ inteq ] . therefore , we decided to examine the scattering properties of finite - length mwcnts of radius @xmath201 nm and @xmath202 nm in the visible regime , taking the intershell electromagnetic coupling into account ( but ignoring the intershell electron tunneling ) . as we discuss later in this subsection , our analysis confirms the absence of the length - matching antenna effect in the near - infrared and the visible regimes . let us first analyze the effective conductivity per unit length of an electrically thin mwcnt , as estimated by ( [ sigm ] ) . figure [ fig5 ] ( a ) presents the real part of @xmath203 as a function of @xmath1 for mwcnts in the near - infrared and the visible regimes . this figure shows that , for swcnts ( @xmath204 ) and mwcnts with @xmath205 $ ] , @xmath203 has strong resonances . the resonances weaken as @xmath1 increases and practically disappear for @xmath206 . this trend can be explained in the following way . in the near - infrared and the visible regimes , the surface conductivity @xmath207 of the @xmath18-th shell , obtained from ( [ eq4 ] ) , has many resonances corresponding to van hove singularities . as the resonances of different shells overlap , the weighted summation of the surface conductivities of all shells in ( [ sigm ] ) ensures that @xmath203 does not evince resonant behavior in the near - infrared and the visible regimes . the conductivity per unit length of an isolated shell of radius @xmath208 nm also has a large number of closely spaced resonances so that , instead of discrete lines , a band appears in its spectrum . as the surface conductivities of all mwcnt shells have a plasmon resonance @xcite in the ultraviolet regime at the free - space wavelength @xmath209 , @xmath210 for any mwcnt also has a resonance at this wavelength . is variable , and @xmath211 s. note that @xmath212 nm . ( b ) real and imaginary parts of @xmath213 of an mwcnt of type m with @xmath214 shells in the near - infrared and the visible regimes . , width=326 ] the real and imaginary parts of the effective per - unit - length conductivity @xmath203 for an mwcnt of type m with @xmath215 shells ( i.e. , @xmath216 nm ) are presented in fig . [ fig5](b ) . as this figure demonstrates , the real part of @xmath203 increases as the frequency increases and has a maximum near @xmath217 thz ( i.e. , @xmath218 nm ) , which is not a geometric - resonance frequency but , instead , is a plasmon - resonance frequency . @xcite in the near - infrared and the visible regimes , the condition @xmath219 holds ; therefore , the chosen mwcnt can not support surface - wave propagation , which occurs for a metallic wire with @xmath220 according to ref . . we also found that the electric field exciting a particular shell differs very slightly from the electric field incident on the mwcnt , when @xmath221 nm and the frequency lies in either the near - infrared or the visible regimes , thereby implying that the electromagnetic coupling between the shells is slight . the frequency dependence of the scattering power @xmath117 for such an mwcnt is the same as of @xmath222 , according to ( [ eq21c ] ) . let us now carry on to electrically thick mwcnts ( with @xmath223 nm or @xmath224 nm ) . for calculation of the electric current densities in their shells in the near - infrared and the visible regimes we used ( [ eq18a ] ) , with @xmath225 assumed to be independent of @xmath226 $ ] . such an approximation is sufficient to ascertain whether geometric resonances of azimuthally symmetrical modes exist in thick mwcnts or not . of course , @xmath213 can not be defined for electrically thick mwcnts . in the near - infrared and visible regimes , the surface conductivity of a shell of large radius ( @xmath227 nm ) , in accordance with ( [ eq4 ] ) , does not depend on that whether shell is metallic or semiconducting . therefore , though the plots in figs . [ fig6]-[fig8 ] were made for thick mwcnts of type m , all results presented therein are qualitatively true for thick mwcnts of type a also . versus free - space wavelength @xmath228 for mwcnts with @xmath229 ( solid line ) and @xmath230 shells ( dashed line ) and length @xmath231 nm . the scattered power for a perfectly conducting rod ( dotted line ) of cross - sectional radius 25 nm and length 350 nm was calculated by solving ( [ eq18a ] ) with @xmath232 and @xmath233.,width=326 ] in order to compare the electromagnetic responses of the chosen mwcnts and a perfectly conducting rod in the near - infrared and the visible regimes , we need to calculate the scattered power @xmath118 when the incident electric field is oriented parallel to the @xmath13 axis . when we calculated the scattered power @xmath118 for mwcnts of type m , length @xmath231 nm , and @xmath234 ( @xmath235 nm ) or @xmath236 ( @xmath237 nm ) , resonances did not show up for @xmath238 $ ] thz ( i.e. , @xmath239 $ ] nm ) in fig . [ fig6 ] . for comparison , the scattered power for a perfectly conducting rod of cross - sectional radius 25 nm and length 350 nm is also shown in fig . this nanorod antenna , in contrast to the mwcnts , has a set of resonances determined by the condition @xmath240 where @xmath76 is a correction factor that slightly exceeds unity and is a function of the ratio of the length and the wavelength as well as of the ratio of cross - sectional radius and the length . @xcite for the chosen nanorod antenna , the first resonance ( @xmath170 ) is characterized by @xmath241 and appears at @xmath242 thz ( i.e. , @xmath243 nm , which is confirmed by the dotted line fig . [ fig6 ] . the absence of antenna ( geometric ) resonances of the chosen mwcnts in the visible regime can be explained by the strong dissipation of electromagnetic energy in mwcnt shells and the small electromagnetic coupling between the shells , as discussed in sec . [ low - coupling ] . therefore , even an mwcnt with @xmath202 nm and comprising 140 shells can not support guided - wave propagation and , consequently , can not display the length - matching antenna effect . the same conclusion is also true in the visible regime for swcnts @xcite and planar arrays of finite - length swcnts . @xcite let us remark that a hypothetic multishell conductive structure with @xmath229 and @xmath244 nm can have an antenna resonance corresponding to @xmath176 in ( [ eq17 ] ) , provided the surface conductivity of every shell is five times that given by ( [ eq4 ] ) ; but such a structure has not been practically realized as yet . on the number @xmath1 of shells in an mwcnt of type m and length @xmath231 nm at frequency @xmath245 thz ( visible regime).,width=326 ] our approach , which is enhanced with respect to that of hao and hansen , @xcite can not explain the mwcnt antenna resonance reported by wang _ @xcite additional experiments are necessary to resolve the contradiction between interpretation of measurements proposed by wang _ et al . _ and the theoretical model of mwcnts developed here and based on ref . . the dependence of the antenna efficiency @xmath186 in the visible regime on the number @xmath1 of shells in the mwcnt is illustrated in fig . the antenna efficiency increases with the number of shells and tends to unity for thick mwcnts ; indeed , @xmath246 thz , we calculated @xmath247 for @xmath234 , but @xmath248 for @xmath236 . these days , thin metallic ( gold , silver , and aluminium ) wires of finite length are considered to be promising for application as optical nanoantennas . @xcite however , the fabrication of long , high - quality , thin , single - crystal wires of cross - sectional radius less than 5 nm and with perfect cylindrical form ( i.e. , without breaks , bends , deformations , etc . ) is a difficult technological problem . recently , cnts have been used as templates in order to promote the formation of high - quality single - crystal wires coated by perfect graphene cylinders . @xcite this is an exciting possibility for a composite nanoantenna comprising a solid metallic core covered by concentric carbon shells . clearly , such a structure is neither a metallic cylinder nor an isolated mwcnt . and ( b ) the ratio @xmath249 on the number @xmath1 of carbon shells for a surface - plasmon wave in an mwcnt with a gold core . the mwcnt is of type m , and its @xmath18-th shell has a @xmath250 zigzag configuration and radius @xmath251 , @xmath19 $ ] . the gold core has a cross - sectional radius @xmath252 nm . data for an isolated metal wire of cross - sectional radius @xmath253 is the same as for @xmath254 . discrete points are joined by lines to aid the eye . , width=326 ] surface - plasmon waves in the infrared and the visible regimes can propagate along a metal wire . @xcite surely , guided - wave propagation would be affected if the metallic wire were to be enclosed in an ensemble of concentric carbon shells . the theoretical approach of sec . [ guided ] can be applied to study the phenomenon of surface - plasmon waves as follows . suppose that the metal core is of cross - sectional radius @xmath255 , the radius of the innermost shell of an mwcnt . if @xmath253 is much less than the skin depth of the bulk metal and @xmath256 , then the metal core can be modeled as a solid cylinder with surface conductivity @xmath257 , @xcite where @xmath258 is the bulk volumetric conductivity of the metal . also , @xmath253 has to be higher than a critical value @xmath259 , which corresponds to the crystalline - noncrystalline transition in metals and separates the quasi - bulk behavior ( @xmath260 ) from the quasi - molecular behavior ( @xmath261 ) of a nanowire . then , ( [ boundary3])([current ] ) also hold at @xmath262 with @xmath263 , and the procedures of sec . [ guided ] are applicable . we considered the propagation of surface - plasmon waves in an mwcnt with a gold core . the bulk volumetric conductivity @xmath258 of gold was taken to follow the drude model with parameters given in ref . . the skin depth of gold in the visible regime is higher than @xmath264 nm . the value @xmath265 nm for gold was obtained with first - principles calculations in ref . . so we assumed that @xmath266 nm @xmath267 nm . figure [ fig8 ] presents the dependences of @xmath163 and @xmath268 of a surface - plasmon wave on the number @xmath1 of shells in mwcnt of type m with a gold core of cross - sectional radius @xmath269 nm . the @xmath18-th shell of the mwcnt has a @xmath250 zigzag configuration and radius @xmath251 , @xmath19 $ ] . in the near - infrared ( @xmath270 thz ) and the visible ( @xmath271 thz ) regimes , fig . [ fig8](a ) shows that the retardation coefficient @xmath163 depends only slightly on @xmath1 . according to fig . [ fig8](b ) , the value of @xmath272 is less than for an isolated metal wire ( @xmath254 ) and significantly decreases as @xmath1 increases . these trends can be explained in the following manner . the guided wave propagates partly inside the mwcnt and partly inside the gold core . the dissipation in the mwcnt shells is high because the real and the imaginary parts of the surface conductivities of the shells in the interband regime are of comparable magnitudes . this dissipation increases with frequency , in accordance with the frequency dependence of the effective per - unit - length conductivity of the mwcnt presented in fig . [ fig5](b ) . because of the weak electromagnetic coupling of cnt shells , the shells strongly absorb electromagnetic energy independently on each other which explains the strong dependence of @xmath268 on @xmath1 . furthermore , the weak coupling changes the radial electromagnetic field distribution of the surface - plasmon wave only slightly as @xmath1 increases , which explains the weak dependence of @xmath273 on @xmath1 . the decrease of the magnitude of @xmath268 with increasing @xmath1 increasing implies the enhancement of @xmath117 , the power lost to ohmic dissipation , and consequently the decreasing of the antenna efficiency @xmath186 of an mwcnt with a metal core . however , if @xmath1 is not too large ( @xmath274 or @xmath275 ) some worsening of antenna properties may be justified by other advantages that the metal - core mwcnt may confer . thus , an mwcnt can operate as an antenna in two different regimes . the first is the _ drude - conductivity regime _ , wherein the motion of conduction - band electrons is responsible for radiation properties . this regime has a distinct analogy with a classical radio - frequency wire antenna , as is clear from sec . [ low - coupling ] and [ num1 ] . the second regime is the _ interband - transitions regime _ , wherein quantum transitions of electrons between different energy states occur . this regime was considered in sec . [ num2 ] and does not have a classical analogy . the frequency @xmath184 separates the drude - conductivity regime @xmath197 from the interband - transitions regime @xmath198 . guided - wave propagation and geometric resonances of the guided waves are typical for macroscopic wire antennas . the guided waves have a quasi - transverse - electromagnetic structure and are characterized by low retardation and low attenuation . @xcite the existence of guided ( surface ) waves and geometric resonances also is typical for nanowire antennas in the drude - conductivity regime . @xcite but the guided wave has strong retardation and high attenuation , so that the frequency of a geometric resonance is not connected to the free - space wavelength but to a shorter effective wavelength that depends on the material properties . @xcite this general rule is also true for mwcnt antennas : fig . [ fig3 ] shows that guided waves have strong retardation and high attenuation , and fig . [ fig4 ] presents geometric resonances at @xmath276 demonstrating thereby the effective wavelength to be shorter than free - space wavelength . calculated data presented in fig . [ fig3](a ) indicate that the retardation coefficient increases when the number of shells increases . furthermore , the retardation coefficient is the highest for the @xmath144 mode in mwcnts of type m. that implies that a mwcnt with only metallic shells and operating in the @xmath144 mode offers attractive prospects for high antenna efficiencies in the terahertz regime . the frequency @xmath184 , separating the drude - conductivity regime from the interband - transitions regime , depends on the detailed electronic and geometric attributes of the mwcnt . according to ( [ freq ] ) , @xmath184 decreases as @xmath17 increases . as examples , for an mwcnt of type a , ( i ) @xmath277 nm and @xmath278 thz when @xmath279 , but ( ii ) @xmath280 nm and @xmath281 thz when @xmath282 . in the interband - transitions regime @xmath283 , guided - wave propagation and geometric resonances are absent for both swcnts and mwcnts , which cardinally distinguishes this regime from the drude - conductivity regime . as shown experimentally,@xcite the absorption and the scattering characteristics of an electrically thick mwcnt in the regime of optical transitions depend only slightly on the frequency . this conclusion , also borne out by the data in fig . [ fig6 ] , may be seem to be unexpected . in fact optical transitions are resonance processes and the surface conductivities of swcnts and mwcnt shells have resonances corresponding to the van hove singularities . but we found that the antenna parameters of electrically thin mwcnts are determined by the effective parameter @xmath203 defined in ( [ sigm ] ) . the overlapping of a large number of resonances of the surface conductivities of the different shells leads to a smooth frequency dependence of @xmath203 . such an effect is analogous to inhomogeneous broadening in an ensemble of all different harmonic oscillators . thus , within the framework of our model , it is impossible to interpret the observation scattering resonances of random arrays of mwcnts with average outermost radius @xmath5 nm reported in ref . . the same conclusion emerged from the model of hao and hansen . @xcite further experiments are required . antennas are objects that transform a near - field into a far - field and vice versa . the morphology of the near field of a nanoantenna possess a nanoscale , and is therefore determined by quantum size effects . therefore the coupling of a nanoantenna with its near field is stronger than that of a macroscopic antenna with its near field ; as a result , the transformation of the near field to the far field by a nanoantenna is more difficult , and the antenna efficiency of a thin - nanowire antenna is low . this property is inherent to different types of nanoantennas : swcnts @xcite , bundle of swcnts @xcite , metallic nanorods @xcite , and mwcnts as in fig . [ fig4 ] . the antenna efficiency can be enhanced by increasing the number of shells in an mwcnt ( fig . [ fig7 ] ) , the number of swcnts in a bundle ( fig . 7 in ref . ) , and the cross - sectional radius of a metallic rod antenna ( fig . 1 in ref . ) . we can thus conclude that the small value of antenna efficiency is the fundamental physical characteristic of nanoantennas . nevertheless , its antenna efficiency is tunable over a wide range . in order to achieve antenna efficiency close to unity , it is necessary to strongly suppress the influence of quantum size effects by ensuring that the cross - sectional radius is high . note that , since quantum size effects are not pronounced in gold nanowires of cross - sectional radius of several tens of nanometers , such nanowire - based antennas are expected to possess properties analogous to macroscopic antennas . the electromagnetic properties of mwcnts have only a slight frequency - dependence in the interband - transitions regime , per fig . [ fig6 ] . thus , and mwcnt can be considered to be a nanoantenna with sufficiently high @xmath284 ( @xmath285 in fig . [ fig7 ] ) and a wide operating - frequency range in the visible regime . such nanoantennas have properties similar to those of electrically small but macroscopic antennas in microwave regimes ( e.g. , short non - resonant dipoles ) . @xcite to conclude , we modeled the shells of an mwcnt as impedance sheets with axially directed surface conductivity , ignored intershell tunneling of electrons but incorporated intershell coupling , in an integral - equation approach . calculated data indicate that in a low - frequency regime called the drude - conductivity regime , wherein optical interband transitions do not occur , guided waves can propagate with low attenuation in an mwcnt which has metallic shells . in the same frequency regime , the axial polarizability of a finite - length mwcnt has a resonant behavior due to the antenna - length matching effect . however , the shells with surface conductivity due to interband transitions suppress guided - wave propagation . due of the high dissipation in such shells , mwcnts with outermost radius @xmath286 nm can not possess resonant properties in the visible regime . analysis of surface - plasmon - wave propagation in a mwcnt with a gold core shows that , in the near - infrared and the visible regimes , the shells behave effectively as lossy dielectric materials and suppress surface - wave propagation along the gold core . * the antenna efficiency @xmath186 of an mwcnt exceeds that of an swcnt but is less than that of an almost circular bundle of closely packed , metallic swcnts , provided that all three objects are of roughly the same outermost radius . therefore , swcnt - bundles are the most promising candidates for terahertz and mid - infrared antennas . * an mwcnt with at least 4 shells is recommended for application as a nanoantenna with a wide operating - frequency range in the visible regime . * filling the core of an mwcnt with a metal makes the mwcnt attractive as a nanoantenna , provided that the number of shells does not exceed 3 . the model developed in this paper can be applied for an mwcnt with achiral shells and negligibly small intershell tunneling . generalization of the boundary condition ( [ boundary3])-([eq4 ] ) is needed to take the chirality of shells and intershell tunneling into account . this research was partially supported by intas under projects 05 - 1000008 - 7801 and 06 - 1000013 - 9225 , international bureau bmbf ( germany ) under project blr 08/001 , and the belarus republican foundation for fundamental research and cnrs ( france ) under project f07f-013 . mvs thanks the world federation of scientists for a fellowship . al acknowledges the charles godfrey binder professorship endowment at the pennsylvania state university for partial support . s. a. maksimenko and g. ya . slepyan , nanoelectromagnetics of low - dimensional structures , in _ nanometer structures : theory , modeling , and simulation _ ( a. lakhtakia , ed . ) , ( spie press , bellingham , wa , usa , 2004 ) , pp . 145 - 206 . a. l. elias , j. a. rodriguez - manzo , m. r. mccartney , d. golberg , a. zamudio , s. e. baltazar , f. lopez - urias , e. munoz - sandoval , l. gu , c. c. tang , d. j. smith , y. bando , h. terrones , and m. terrones , nano lett . * 5 * , 467 ( 2005 ) .

this year marks the @xmath0 anniversary of two seminal papers by slonczewski @xcite and berger @xcite on the spin transfer torque ( stt ) phenomenon . this refers to the transfer of angular momentum from a spin - polarized charge current , or a pure spin current , to a local magnetization . while berger had already reported on stt - driven domain wall motion using 45 a of current in millimeter - wide films in 1985 @xcite , based on his predictions from the late 1970s @xcite , and slonczewski had developed the theory for spin - current - driven exchange coupling in magnetic tunnel junctions in 1989 @xcite , the two papers in 1996 predicted that a current flowing perpendicularly to the plane of a magnetic metal multilayer should be able to reorient the magnetization direction in one of the layers . these papers were published at a time when magnetic thin film technology had just become capable of beginning to implement these concepts in earnest , with giant magnetoresistance ( gmr ) @xcite multilayers being prepared for hard drive read - head production and the first high - quality magnetic tunnel junctions having been reported the year before @xcite . it nevertheless took a few more years before the first experimental demonstration of stt - driven magnetization precession was reported @xcite , followed by reports of stt - driven magnetic switching @xcite . in 1999 , hirsch revisited @xcite a different phenomenon , the so - called spin hall effect ( she ) , first described by dyakonov and perel in 1971 @xcite . the she can produce a pure spin current in a direction perpendicular to a charge current @xcite , which can in turn exert substantial stt on an adjacent magnetic layer . it is thus possible to use the she in a nonmagnetic metal to achieve , in principle , all the stt - related functions such as magnetic switching @xcite and driven magnetization precession @xcite that previously required a separate magnetic layer to spin - polarize the charge current . from a technological standpoint , the two most important consequences of stt are _ i _ ) the possibility of programming a magnetic memory , such as magnetoresistive random - access memory ( mram ) @xcite , using a small direct current through the bit , rather than larger currents through adjacent programming lines , and _ ii _ ) the possibility of creating nanoscopic and ultratunable microwave signal generators without the need for any semiconductor materials . our paper deals with the latter technology and reviews the fundamentals and state of the art of so - called spin transfer torque and spin hall effect nano - oscillators ( stnos and shnos ) . for more in - depth reviews of stt - based mram and stnos , see @xcite and @xcite . the term _ spintronics _ is often used to refer to electronics in which both the charge and the spin of the carriers in a metal or semiconductor are utilized ( for reviews of spintronics , see @xcite ) . however , this excludes the very active research currently being undertaken in magnetic insulators ; these can not transfer any charge current , but are excellent conductors of spin currents @xcite . a broader , more general and , we would argue , better definition of modern spintronics is any device and technology in which _ spin currents _ @xcite , with or without additional charge or heat currents , are directly employed and utilized for function . as spin is a vector in three - dimensional spin space , and the flow of the spin carriers takes place in three - dimensional real space , the resulting spin current is a @xmath1 tensor ( for illustrative purposes , the spin current may instead be pictured as a flow , in real space , of tiny magnetic moments , also in real space ) . the tensorial nature of the spin current where the direction of the flowing spins is , in the general case , independent of the real - space direction of the spin current provide for very rich physics and a tremendous potential for new device functionality . in the original formulation of stt , the spin current was assumed to be carried by a spin - polarized charge current , and a number of simplifying assumptions ( e.g. , device geometry and uniformity of magnetization ) were made such that the flow direction of the charge and spin currents were identical . most analysis and modelling of stnos still relies on such assumptions , which tend to be highly accurate in most cases . in this case , the spin - polarized current ( and hence the spin current ) arises directly from the spin polarization of the electrons at the fermi surface of the ferromagnetic metal . when an unpolarized current , say in cu , enters a ferromagnetic layer , it becomes spin - polarized in the so - called spin accumulation zone , which extends both into the cu and the ferromagnet in proportion to the spin diffusion lengths in each material @xcite . as a result , the current acquires spin polarization well inside the cu , reaches about half its full spin polarization at the nm / fm interface , and then , over a few nanometers in the fm , rapidly reaches the spin polarization of the bulk of the ferromagnet . the same processes govern how the spin polarization of the current decays as the current leaves the ferromagnet and again enters a nonmagnetic metal . as the spin diffusion length is short in ferromagnets , but can be several hundred nanometers even at room temperature in most ordinary metals , thin magnetic films are highly effective spin filters and the resulting spin - polarized currents can act over quite long distances . while spin filtering is a straightforward technique for generating spin - polarized currents , spin currents do not necessarily require any charge current to flow . in the case of the she , the charge and spin currents flow in perpendicular directions with respect to each other that is , the charge current is unpolarized and the spin current is free of net charge . while we do not discuss this further here , pure spin currents can also be created through magnetodynamic means , as in so - called spin batteries @xcite . the she can be viewed as a limiting case of the anomalous hall effect ( ahe ) @xcite , which was proposed by edwin hall in 1881 when he applied his earlier discovery of the ordinary hall effect ( he ) @xcite to ferromagnetic metals . as is well known , the he in nonmagnetic metals or semiconductors refers to the build - up of a so - called hall voltage across an electrical conductor , transverse both to the electric current in the conductor and to an external magnetic field perpendicular to the current . it is widely used for material characterization and for sensing magnetic fields . in ferromagnets , however , hall found that this effect could be ten times greater than in nonmagnetic metals . whereas the he arises from the magnetic field s lorentz force on the charge carriers , the ahe is a consequence of asymmetric spin - dependent scattering of the charge carriers , such that a much larger hall voltage can build up . the she is then a limiting case of this phenomenon in materials without any spontaneous magnetization but with substantial spin - orbit coupling . as there is no spin imbalance in a nonmagnetic material , the asymmetric spin - dependent scattering does not result in any hall voltage . what it does result in , however , is a net spin current transverse to the charge current and , in a steady state , the build - up of spin accumulation zones at the edges of the conductor , which can be directly observed by optical means @xcite and which is used in spin - current - driven devices . equipped with two different techniques for generating spin currents , we now briefly explain how spin currents can generate stt and control the magnetic states and the magnetodynamics of nanomagnets and magnetic thin films ( for an in - depth review of stt , see @xcite ) . since each electron carries angular momentum , a spin current can be thought of as a flow of angular momentum . when a spin - polarized current enters ( or reflects at ) a nonmagnetic ferromagnetic interface , the spin precession becomes incoherent over a very short distance . this results in a loss in its average spin angular momentum in the direction _ transverse to the local magnetization_. since the total angular momentum has to be conserved in a closed system , whenever the flow of angular momentum changes , a torque acts on the local magnetization . this torque is often referred to as the `` in - plane '' torque , where the plane is defined by the local magnetization and incoming spins . this torque may act gradually , as when spin - polarized electrons flow through a domain wall and have to rotate so as to follow the local magnetization , or very abruptly , as when a spin current enters a ferromagnet with its magnetization in a direction different to that of the spins in the spin current ; this is almost always the case in stnos and shnos . such an in - plane torque manifests itself as an energy - nonconserving force , similar to damping . furthermore , depending on the sign of the current , this in - plane torque is either parallel or antiparallel to the damping torque . when antiparallel to the damping torque , and for sufficient current magnitude , the current acts to destabilize the local magnetization , thus exciting large angle dynamics . there is also the possibility of an additional stt term that is oriented perpendicular to the plane defined by the local magnetization . this `` out - of - plane '' torque is often referred to as an `` effective field '' contribution , as it has the same symmetry as is typically used in field - driven dynamics . while the perpendicular contribution is small for all - metallic multilayers @xcite , it can be substantial in devices with tunneling barriers @xcite . it was not until sufficient advances in nanofabrication techniques had been achieved that magnetic devices with nanoscopic dimensions could be readily fabricated and the currents needed to induce magnetodynamics be decreased to the ma range . one of the first techniques relied on mechanical point - contacts , where a highly sharpened metallic tip is used to make electrical contact , typically with an area on the order of 10@xmath2 nm@xmath2 , to an extended multilayer film @xcite . soon after , lithographically defined contacts were employed in devices that are now termed nanocontacts , to distinguish them from their mechanical point - contact predecessors @xcite . the fabrication of nanocontacts involves opening a small electrical contact of either circular or elliptical @xcite cross - section through an insulating layer using either electron - beam lithography ( ebl ) or hole - colloid lithography ( hcl ) @xcite techniques . in both point - contact and nanocontact devices , it is only the current injection site that is confined to a nanoscopic region . however , in oscillators based on the nanopillar @xcite geometry , all of the magnetic layers are patterned into a wire with either circular or elliptical cross - section . fabrication is most often accomplished by a combination of ebl and ion milling . however , stencil @xcite and electrodeposition @xcite techniques have also been pursued . the nanopillar geometry ensures that all of the electric current passes through the entire multilayer stack , typically resulting in lower current densities ( < 10@xmath3 a / cm@xmath2 ) being needed to excite stt - induced magnetization switching or oscillations than in the case of nanocontact - based devices ( 10@xmath510@xmath6 a / cm@xmath2 ) . it should be noted that nanopillar fabrication is typically more involved and prone to deleterious edge effects that can increase oscillator linewidths and magnify sample - to - sample variations . due to the smaller magnetic volumes involved , nanopillar - based devices are also more sensitive to thermally induced fluctuations , which results in larger linewidths than in nanocontact devices @xcite . another important consideration is that , as all of the magnetic layers are patterned , significant stray fields are present in nanopillars and need to be considered in analyzing them . additionally , hybrid device architectures have also been pursued in which a nanopillar is fabricated on top of an extended film @xcite . such hybrid structures are useful if , for example , optical access to the extended magnetic layer is desired . generally speaking , the transport of spin current , and therefore of the associated stt , does not need to accompany a charge current . such nonlocal devices are engineered to promote different paths for the spin and charge currents . this geometry is particularly useful in tunnel - junction based devices where large voltages can cause electrical breakdown of the tunnel barrier . such devices typically require at least three terminals for both spin current generation and electrical detection . both stt - induced nonlocal switching @xcite and oscillations @xcite of the magnetization have been demonstrated . devices where the stt is provided by the she that is , where the charge and pure spin currents naturally flow in orthogonal directions , are prime examples of nonlocal oscillators @xcite . by and large , most material stacks utilized in stno devices are based on a ferromagnet / spacer / ferromagnet ( fm / s / fm ) trilayer . one fm is considered free and is most susceptible to the influence of stt . it is therefore advantageous for the free layer to be relatively thin and to have a low intrinsic gilbert damping . ni@xmath7fe@xmath8 , which goes by the trade name `` permalloy '' and is often referred to simply as nife , is the canonical choice for the free layer in the majority of stnos and shnos . in contrast , the other fm layer is fixed , either through pinning to an adjacent magnetic layer or by virtue of its intrinsically larger anisotropy or thickness . the primary role of the fixed layer is to spin - polarize the initially unpolarized electron current . the spacer layer may be metallic ( most typically cu ) , forming what is referred to as a pseudo spin valve ( psv ) , or insulating , as in the case of magnetic tunnel junctions ( mtjs ) . as the tunneling magnetoresistance ( tmr ) effect is much larger than the giant magnetoresistance ( gmr ) effect in all - metallic psvs , mtjs are becoming standard in most nanopillar devices @xcite . the use of mtjs in conventional nanocontacts is challenging , as the topmost fm layer would act to shunt the current . nevertheless , there has been some success in creating hybrid nanocontact mtjs , often termed `` sombrero '' structures , which aim to marry the low linewidths inherent in nanocontact devices with the large output powers found in mtjs @xcite . the materials that constitute the fm layers are diverse and can be broadly classified and distinguished by their equilibrium magnetization configurations , as described next . the vast majority of stno devices studied in the early stages were based on free and fixed fm layers with an intrinsic easy - plane anisotropy meaning that the magnetization prefers to lie in the plane of the film if no external magnetic field is applied @xcite . easy - plane materials , such as co and co@xmath9fe@xmath10 for the fixed layer and nife for the free layer , have most often been pursued . one drawback of utilizing materials with easy - plane anisotropy is that large ( 1 t ) fields are often required to overcome the demagnetization fields needed to pull the magnetization out - of - plane . owing to their improved thermal stability , scalability , and potential for low to zero field operation , materials with perpendicular magnetic anisotropy ( pma ) are extremely attractive . one of the first _ all perpendicular _ nanopillar stt devices combined a [ co / pt]@xmath11 multilayer as a fixed layer with a [ co / ni]@xmath11 multilayer free layer @xcite and was primarily devoted to stt - induced switching . later stnos with perpendicular free and fixed layers have shown zero - field operation @xcite . in these types of devices , the equilibrium magnetization direction of the fixed and free layers are orthogonal to each other . one of the first devices employed a perpendicular fixed layer based on a co / pt multilayer and a planar nife free layer @xcite . more recent devices have tended to deal with the opposite situation , with in - plane fixed layers using either co , co@xmath9fe@xmath10 , or even nife , and a perpendicular co / ni multilayer free layer @xcite . a tilted magnetization will have both in - plane and out - of - plane components ; its use in stnos is predicted to provide additional degrees of freedom to manipulate both static and dynamic states . relative to conventional in - plane or pma magnetic films , the use of tilted fixed layers can optimize microwave signal generation and enhance spin - torque efficiency while maintaining high output powers with low to zero field operation @xcite . to date , experimental work has primarily focused on the fabrication of tilted anisotropy exchange - coupled films by taking advantage of the competition between the pma of a co / pd or co / ni multilayer and the in - plane shape anisotropy of a nife of cofeb layer . @xcite . subsequent measurements have then focused on the fundamental spin wave excitations in such tilted anisotropy exchange springs @xcite . zero field operation in cofeb / mgo / cofeb mtjs with tilted cofeb free layers has been experimentally demonstrated @xcite . departing from the trilayer configurations discussed above , theory @xcite predicts that spin filtering by a single fm layer can generate a sizable spin accumulation at each interface . however , it is only when these interfaces are asymmetric that an imbalance of spin accumulation will lead to a significant stt . experiments on nanopillars with a single co layer showed that sufficient stt could be generated to produce spin - wave excitations @xcite and even switch the co magnetization @xcite . the latest work on single layer devices has been in the nanocontact geometry , where evidence of a periodic vortex / antivortex creation / annihilation mode has been found @xcite . stnos exhibit a diversity of fundamental spin wave modes that depend greatly on , for example , the device architecture , constituent fm materials , applied magnetic field strength , and applied magnetic field angle . here we summarize some of the most prevalent stt - induced excitations in both nanopillar and nanocontact stnos . the particular spin - wave mode generated in a nanopillar geometry depends most sensitively on the diameter or thickness of the free fm layer , as well as on the strength of the external magnetic field . frequently , the dimensions of the free layer are such that a magnetic vortex characterized by an in - plane curling of the magnetization with a 10-nm sized out - of - plane vortex core at the center is a stable equilibrium configuration @xcite . if so , then the stt will excite the fundamental gyration of the vortex core with frequencies typically less than 2 ghz @xcite . however , in a large enough external field , the vortex state will no longer be energetically favorable @xcite , the magnetization will become more or less uniform , and stt will typically excite highly elliptical clam - shell precession in the case of a free layer magnetized in - plane . similarly , if either the diameter or thickness of the free layer is sufficiently small that exchange interactions dictate a uniform magnetization , then such uniform precessions will also be generated . for nanocontact stnos where the equilibrium magnetizations of both the free and fixed layers prefer to lie in - plane , the type of spin wave excitation depends strongly on the external applied magnetic field angle . as described by both analytical @xcite and micromagnetic simulations @xcite , the initial experiments @xcite using in - plane applied magnetic fields excited a strongly nonlinear , self - localized solitonic bullet mode . however , when the free layer is magnetized perpendicularly to the film plane , as described by slonczewski @xcite , an exchange dominated propagating spin wave with a wave vector inversely proportional to the nanocontact radius is excited , as directly proven using brillouin light - scattering experiments @xcite . for intermediate oblique angles , the behavior becomes much more complex ; both simulations @xcite and experiments @xcite provide evidence of mode hoping @xcite between propagating and localized bullet modes @xcite . the use of elliptical nanocontacts has been useful in trying to disentangle the coupling mechanisms between these modes @xcite . special attention must be given in the nanocontact geometry to the critical role played by the current - induced oersted field . for typical drive currents and nanocontact dimensions , the oersted field is on the order of 0.1 t at the nanocontact edge ; this is a significant fraction of the external applied magnetic field and can not therefore be considered an insignificant perturbation . for example , the oersted field results in a localization of the propagating mode for applied field angles near the film plane , which in turn promotes mode coexistence with the solitonic bullet mode @xcite . furthermore , at large applied field angles , the oersted field induces an asymmetric fmr field landscape which results in asymmetric spin wave propagation @xcite . in short , in regions where the addition of the external field and the oersted field produce a local field maximum , spin waves can no longer propagate , resulting in beam - like spin - wave propagation towards ( and beyond ) the local field minimum on the other side of the nanocontact @xcite . finally , the oersted field also plays an important role in the formation of vortex oscillations in nanocontacts @xcite . in small external magnetic fields , the circulating oersted field favors a vortex state in the free layer via the zeeman interaction . this interaction in turn generates a confining potential for the vortex core . the action of stt is then to drive the vortex core away from the center of the nanocontact , whereas the natural damping tends to pull the core back in . in equilibrium , these two forces result in a self - sustained vortex gyration around the nanocontact . in stark contrast to nanopillars , where the vortex motion is physically constrained , these vortex orbits can even occur in regions outside of the nanocontact itself . in such cases , this results in a very large time - varying magnetoresistance signal and thus large output powers . as with vortices in nanopillars , the oscillation frequencies are usually less than 2 ghz and the emission spectrum usually contains several higher harmonics due to elliptical orbits . when driven past a critical velocity , the vortex core can flip polarity , leading to chaotic dynamics @xcite . significant changes in the fundamental magnetodynamical modes are found when the free layer exhibits pma , such as occurs in the oft - employed co / ni multilayers . when the applied field is perpendicular to the film plane , three distinct operational modes can be identified . at small applied magnetic fields , a small - angle precessional fmr - like mode is observed with relatively small power @xcite . as the magnitude of the drive current is increased ( decreased ) , there is a gradual increase ( decrease ) in mode power ( frequency ) consistent with an increasing precession angle . however , at larger applied fields , the oscillator behavior changes dramatically . as the applied field is increased , a dramatic drop in oscillator frequency and an increase in oscillator power are observed , consistent with the formation of a magnetic droplet soliton @xcite . the theoretical basis for the droplet soliton was initiated by ivanov and kosevich @xcite , who showed that the landau lifshitz equation can sustain a family of conservative magnon drop solitons in the absence of damping . while initially a theoretical curiosity , it was then shown that the balance between stt and damping underneath the nanocontact can provide the zero damping needed to sustain what are now called dissipative droplet solitons @xcite . finally , at very large external fields , the zeeman energy will dominate and the droplet will collapse @xcite . the influence of the spin current generated from she - induced spin accumulation on the ferromagnetic resonance of an adjacent magnetic layer has been initially shown on micrometer - sized permalloy discs @xcite . while the linewidth could be reduced by a factor of two , full compensation of the damping , and thus auto - oscillations , were not achieved in these micrometer - sized films , as the uniform application of the torque does not favor any auto - oscillation mode . shnos have thus been realized by nanopatterning of various parts of the bilayer to achieve the nonuniform stt , and can be categorized into four primary device geometries . in the three - terminal nanopillar geometry , the magnetic free layer is a nanosized element on top of an extended slab of the she - generating normal metal @xcite . the electrical read - out is carried out here through an mtj element including the oscillating free layer , which is how appreciable output powers can be achieved . another approach is to utilize a nanogap between two highly conductive electrodes , injecting the necessary high current densities into a nanosized region of an extended fm / pt bilayer @xcite . in the more recent nanoconstriction - based shnos , both the fm and the pt layer are patterned into a nanoscopic constriction @xcite . similar confinement of the she - induced spin current has been demonstrated in a 190 @xmath12 nanowire geometry @xcite . the benefits of the latter two shno device geometries over their stno counterparts include easier fabrication ( for example , they require the deposition of only two layers and fewer lithographic steps ) and direct optical access to the oscillating region . in contrast to stnos , which rely on the gmr or tmr effects , these shnos must however rely on the smaller anisotropic magnetoresistance ( amr ) effect in a single fm layer , and therefore typically exhibit lower output powers . the generation of a sufficient she - induced pure spin - current requires a material with a large spin - orbit coupling . to date , only results on shnos using pt and ta have been published . however , a few different fm materials have been studied in addition to nife , including co@xmath13fe@xmath13b@xmath14 @xcite , co / ni multilayers with pma @xcite , and the ferrimagnetic insulator yttrium iron garnet ( yig ) @xcite . for in - plane magnetized thin films , two different auto - oscillation modes have been observed in nanogap shnos . as described in @xcite , these devices have been studied for in - plane applied fields with microfocused brillouin light scattering and from _ ( i ) _ the sub - fmr frequency and _ ( ii ) _ the spatial size , which was smaller than the instrument resolution , the existence of a self - localized spin wave bullet was inferred @xcite . electrical measurements as a function of the temperature , however , showed the existence of the low - frequency and low - linewidth bullet mode only for cryogenic temperatures with a second high - frequency mode appearing at higher temperatures , and even being dominant at room temperature @xcite . micromagnetic simulations suggest that this high - frequency mode is generated within local field minima caused by the bullet mode which implies the simultaneous excitation of both @xcite . however , since the oersted field generated from the nonuniform current in the pt layer always counteracts the applied in - plane field , a similar field minimum in the nanogap is always present , allowing the high - frequency mode to form . using a 1 @xmath15 m large gap between the electrodes , and thus generating rather uniform current distributions , @xcite showed that the high - frequency mode most likely resembles a coherent excitation of the fmr mode . in the nanowire shnos , the existence of both bulk and edge modes has been identified @xcite while the linear high - frequency excitation shows an increasing threshold current with the strength of the applied field , the spin wave bullet has a relatively constant threshold current as long as the internal magnetization angle stays below a critical value @xcite . the output characteristics can be thus tuned between these two regimes via the application of a tilted magnetic field , where the electrically measured linewidth is lowest for high fields , since only the spin - wave bullet exists there @xcite . moreover , the interplay can be varied by the nanogap size , favoring the bullet mode for smaller sizes . when both modes are energized at the same time for intermediate field strengths , their linewidths are broadened due to mode coupling events @xcite . propagating spin waves with frequencies above fmr are important in , for example , mutual synchronization , and are predicted by micromagnetic simulation in nanogap shnos for out - of - plane fields @xcite . however , in this study , the current densities required are about twice as large as for the in - plane field modes , which inhibits the experimental observation of the propagating spin wave modes . nanoconstriction - based shnos show a mode that evolves from the thermal fmr , as shown by micro - focused bls measurements and electrical detection @xcite , and is therefore comparable to the high - frequency mode of nanogap shnos . however , with the constricted geometry around the oscillation center , the presence of the bullet is hindered and the single - mode excitation can be detected with lower linewidths . synchronization to external sources , as described later in regards to injection locking , or mutual phase locking of individual stos , as described here , are techniques often used to improve oscillator power and linewidth . there are several coupling mechanisms that have been investigated to synchronize stnos to each other , each with a varying degree of experimental success . synchronization of serially connected nanopillar stnos using their shared microwave electric current @xcite has been theoretically proposed , but to date has shown no clear success . another route to synchronize nanopillars is to utilize the relatively long - range dipolar stray microwave fields generated during precession . recent simulations show this to be a feasible solution for vortex - based oscillators @xcite . to date , however , the most successful experimental works on synchronizing stnos have occurred in the nanocontact geometry . in such devices , the nanocontacts are patterned on top of a shared free layer , which in turn provides the coupling medium . the first works to show that two stnos could be synchronized were published in 2005 @xcite it was soon shown experimentally @xcite , as well as investigated theoretically @xcite and micromagnetically @xcite , that the coupling mechanism is mediated by propagating spin waves . since these early experiments , progress has remained relatively slow . only in 2009 was it demonstrated that four vortex - based stnos could synchronize , this time by direct exchange interactions mediated by an antivortex . @xcite . then , in 2013 , the successful synchronization of three high - frequency stnos was shown for the first time @xcite . recent simulations have shown robust synchronization of an arbitrary number of nanocontact vortex oscillators mediated by the strong magnetodipolar interaction @xcite . the most recent work on synchronization has focused on the use of highly directional spin - wave beams . while it has been shown that the oersted field generated in the vicinity of the nc can dramatically alter the emission pattern of sws @xcite , its role in the synchronization behavior of multiple ncs has been largely ignored . synchronization is promoted ( impeded ) by the oersted field landscape when the nanocontacts are oriented vertically ( horizontally ) with respect to the in - plane component of the external field , due to the highly anisotropic spin - wave propagation . not only is robust synchronization between two oscillators observed for separations greater than 1000 nm , but synchronization of up to five oscillators has been observed in the vertical array geometry . furthermore , the synchronization must be considered driven , rather than mutual , in nature as the final frequency is enforced by the stno from which the sw beam originates @xcite . the precession frequency of the stnos is tunable over a large range , both through the direct current and the applied magnetic field . the precession frequency generally increases as the external magnetic field increases . its behavior is generally governed by the kittel equation , according to which the precession frequency increases as the square root of the applied field at low fields and linearly at high fields . theoretical works predict a very high frequency of 0.2 thz for the common ferromagnetic material ni@xmath16fe@xmath14 . @xcite in the very first work by kiselev _ et al . _ , @xcite a highest frequency of 25 ghz was demonstrated . the devices used were nanopillars of co / cu / co/ multilayer . a year later , rippard _ et al . _ @xcite demonstrated a frequency of 38 ghz in a nanocontact device based on a ni@xmath16fe@xmath14 free layer . the field tunability of frequency is 18.5 mhz / mt and the current tunability is about -200 to 400 mhz / ma for nanocontact stnos . @xcite bonetti _ et al . _ , @xcite reported a frequency as high as 46 ghz for high magnetic fields applied normally to the film plane . the stno frequency was limited by instrumentation . it was reported that it would be possible to achieve a frequency beyond 65 ghz , which would be potentially attractive for a range of millimeter - wave applications , such as short - range high - speed radio links ( ieee 802.15.3c ) near 60 ghz and vehicle radar for active cruise control at 77 ghz . @xcite an stno may be viewed as a time - dependent resistor with its resistance given by : @xmath17 where @xmath18 is the average resistance of the stno , @xmath19 is the magnetoresistance of the material stack , and @xmath20 is the precession frequency . for a gmr - based device , @xmath19 is low , and hence the power generated by the gmr - based stnos is also low , typically in the sub - nw range . @xcite higher powers have been achieved in magnetic tunnel junction based material stacks , for which @xmath19 is significantly larger . in magnetic tunnel junctions , powers of up to 3.6 @xmath15w have already been reported for spin - torque vortex oscillators with an feb free layer @xcite . however , the frequency of operation of these vortex oscillators is typically low . for uniform mode oscillation , the highest power achieved so far has been 2 @xmath15w , obtained from a sombrero - shaped nanocontact magnetic tunnel junction based stno with a perpendicular anisotropy free layer @xcite . for in - plane magnetized mtj - stnos , the highest power reported was 140 nw @xcite for a frequency around 5 ghz . while the frequency can be increased using the magnetic field , the output power usually drops with increasing field strength ; obtaining higher power at higher frequencies is thus a challenge . a novel method of achieving this involves frequency doubling in mtj - stnos @xcite . in that work , powers of up to 1.25 nw were demonstrated at a frequency of @xmath2110 ghz . modulation is a very important requirement for the communication applications of stnos . modulation studies of stnos are , however , surprisingly limited despite the issue s critical importance for communications applications . the first experimental demonstration of frequency modulation in an stno was reported in 2005 @xcite , with the demonstration that the addition of an alternating current to the direct current results in frequency - modulated spectral output , generating sidebands spaced at the modulation frequency . this phenomena is demonstrated in fig.[fig - mod1 ] ( b ) , where equally spaced multiple sidebands are produced upon the addition of an rf current ( 1 ma , 100 mhz ) . the device is shown in fig.[fig - mod1 ] ( a ) and corresponds to that of @xcite . as can be seen in this figure , the sidebands have unequal amplitudes and the carrier frequency shifts with the addition of the rf current . this phenomena was first reported in @xcite and was attributed to nonlinear frequency modulation . however , in 2010 , it was shown that the unequal sideband amplitudes result from a combination of nonlinear frequency and amplitude modulation ( nfam ) . it was shown that the combined nonlinear frequency and amplitude modulation model can accurately describe the experimental data without any adjustable parameters . the frequency of modulation was 40 mhz in @xcite and 100 mhz in @xcite . later , it was shown that a nanocontact - based stno can be successfully modulated to a modulation frequency of 3.2 ghz @xcite . the agreement of experiment with the nfam model is maintained until 1 ghz @xcite . it was also shown that modulation can be used to narrow the linewidth of stnos by as much as 85 % @xcite . this linewidth reduction is due to a general averaging effect of modulation on the nonlinearity of the stno signal and was explained on the basis of nonlinear auto - oscillator theory @xcite and nonlinear frequency modulation . in 2011 , modulation of a pair of nanocontact stnos was successfully demonstrated@xcite . in fig.[fig - mod2 ] , we reproduce the results from @xcite which show that the synchronized state of a pair of nanocontact stnos ( at 57 ma ) can be maintained while modulating . furthermore , the modulation behavior can be explained by assuming a single oscillator model , as shown in fig.[fig - mod2](b ) . this finding was very promising for applications involving a large array of synchronized stnos . many electronic circuits use injection - locked phase - locked loops ( plls ) as one of the building blocks of rf transceivers and so it is also crucial for applications to demonstrate the modulation of the injection - locked state of the stno . more recently , modulation of injection - locked mtj - stnos was demonstrated and the modulation - mediated nonresonant unlocking mechanism was described @xcite . interestingly , the unlocking phenomena sets a lower limit on the modulation frequency but no upper limit on the modulation frequency for injection - locked mtj - stnos . experimental studies of the modulation of vortex - based stnos are rather limited . the first digital modulation of vortex - based stnos was demonstrated in @xcite . using frequency shift keying ( fsk ) , it was shown that the oscillator can switch between two stabilized frequencies differing by @xmath22 in 20 ns . this performance is difficult to achieve in a voltage - controlled oscillator , and vortex - based stnos hence hold much promise for fsk modulation - based applications . more recently , analog frequency modulation has been demonstrated in a magnetic tunnel junction nanopillar in in - plane fields @xcite . in this work , the authors show modulation for modulation frequencies that are up to 10% of the oscillator frequency . in these experiments , however , the modulation period is comparable to the characteristic time of the vortex dynamics , and hence the frequency modulation data was modelled using `` deviation sensitivity , '' which differs significantly from the tunability measured in a quasistatic regime . when the modulation frequency of an stno is increased , the stno may not be perfectly modulated . a recent study has shown that the upper limit of the frequency of modulation in an mtj - stno is 200 mhz @xcite . in this work , the authors showed that the bandwidth of modulation is limited by the amplitude relaxation rate @xmath23 and that the power level of amplitude modulation strongly decreases beyond @xmath24 , where @xmath25 . this finding is in agreement with the nonlinear auto - oscillator theory of stnos @xcite . although the original nonlinear auto - oscillator theory does not include amplitude modulation , a recent theoretical study that includes an nfam scheme also suggests that the modulation bandwidth of stnos is determined by @xmath23@xcite . the miniature size of stnos and shnos compared to industry - standard oscillators inevitably makes their oscillation more susceptible to noise . the stability of any oscillator can be characterized in terms of its amplitude noise and phase noise , where the width of the spectral peak ( commonly referred to as the linewidth ) is usually governed by the phase noise as has also been found for stnos . the phase noise of shnos has not been as extensively studied as that of stnos , but one initial report @xcite provides a lowest linewidth on par with what has been achieved for stnos . a prerequisite for stable oscillation is that the oscillation is not intermittent . one type of intermittency is on off toggling , which has been observed for low drive currents @xcite . another type is mode - hopping between different distinct frequencies . stnos do in many cases show spectra with the presence of several modes represented by distinct frequencies , where time - domain measurements can often reveal a random mode - hopping between these frequencies . the mode - hopping rate can range from milliseconds down to nanoseconds , meaning that , depending on the bandwidth of the characterization setup , it might not always be observed . it should be noted that in some cases there is the possibility of actual coexistence ( in time ) , a clear indicator of which is the appearance of second - order intermodulation products @xcite . if the application is low bandwidth compared to the mode - hopping rate , the mode - hopping itself need not constitute too large a problem , since the fast fluctuations might be washed out . however , the mode - hopping has a large negative effect on the coherence of the oscillation . in a mode - hopping event from one frequency to another and then back again , the phase will be lost ( introducing phase noise ) . this becomes visible in the spectrum in the form of a dramatic linewidth increase @xcite . the occurrence of multiple simultaneously excitable modes is most evident in nanopillar stnos @xcite , where the modes emerge as various edge and center modes on account of its confined geometry . in nanocontact stnos , with their far - extending magnetic films , mode - hopping within the propagating mode has been experimentally observed only at very high drive currents ( close to the breakdown of the oscillation ) , @xcite while simultaneous excitability of the propagating and localized bullet mode can also induce mode - hopping @xcite . moreover , the propagating mode in nanocontact stnos generally exhibits discontinuous mode transitions as a function of the bias point @xcite . close to these transitions , the linewidths of the individual modes typically increase @xcite , which could be theoretically explained by rapid hopping between the modes @xcite , though this has not been experimentally confirmed yet . apart from control over the existence of multiple oscillation modes ( and the instability they introduce to each other ) , a second fundamental question is that of the stability of a single mode . this stability is primarily considered to be governed by thermal magnetic noise which is usually modelled as random fluctuations in the magnetic field . this has a straightforward implementation in the macrospin approximation . micromagnetically , it is usually implemented as fluctuations that are uncorrelated in both time and space @xcite . it has been noted @xcite , that no rigorous theory has been presented to explain the effects of spatial fluctuations on the spin - wave eigenmodes . another aspect of stability arises from the complexity of micromagnetic systems , where spatial inhomogeneity can be induced by locally varying fields even at zero temperature @xcite . such spatially varying fields may stem from the geometry through the fabricated shape of the free layer or the roughness in the nanometer - scale film thickness , or through local anisotropy fields in the metal grains . the noise properties have been found to be highly dependent on the strong amplitude frequency coupling @xcite which acts to transfer amplitude noise into frequency noise @xcite . as the magnetization is precessing around its steady - state orbit , fluctuations act to displace it along two orthogonal directions : in parallel with and perpendicularly to the orbit . parallel displacement introduces a single phase shift whereas a perpendicular perturbation ( fig . [ fig - phasenoise](a ) ) provides no initial phase difference . however , since a perpendicular perturbation places the magnetization away from the steady - state orbit at an oscillation amplitude different from the steady - state amplitude , the magnetization will momentarily take on another frequency associated with this new amplitude . during the time of precessional relaxation of the amplitude back to the steady - state orbit , the frequency will gradually approach the steady - state frequency , but before this frequency has been reached , the oscillator will have accumulated a phase shift as compared to the nominal frequency ( fig . [ fig - phasenoise](b ) ) . quantitatively , both theoretical calculations @xcite and experiments @xcite show that the main effect of perpendicular perturbations and of amplitude - frequency coupling is by far the dominating mechanism for the phase noise in stnos . in the framework of the general nonlinear auto - oscillator theory , it has also been demonstrated how , in the vicinity of the oscillation threshold , the linewidth increases due to spectral line asymmetry induced by several statistical solutions to the noise process @xcite . the high impact of the strong amplitude frequency coupling on the resulting phase noise means that the oscillator will be least sensitive to fluctuations when its amplitude frequency coupling is nonexistent , but can also be improved by achieving a higher amplitude relaxation rate @xcite . from a fundamental thermal noise perspective , the frequency stability should be increased by the use of materials with low thermal damping of the magnetization precession i.e . materials that exhibit a low value for the gilbert damping parameter . on the contrary , however , the amplitude relaxation rate is proportional to the gilbert damping , meaning that the amplitude perturbations will be damped out more slowly . since the amplitude fluctuations are the major source of frequency noise , a high value for the gilbert damping parameter is preferable . @xcite . several experimental studies report the presence of low - frequency @xmath26 frequency noise in both mtj nanopillar @xcite and gmr nanocontact stnos @xcite . from a characterization point of view , the high level of @xmath26 frequency noise is challenging in the sense that the standard phase noise measurement modes of spectrum analyzers are unable to lock onto the stno electric rf signal . the research community has been able to overcome the characterization problem through time - domain measurement of the waveform and subsequent digital signal processing @xcite , but from an application point of view , @xmath26 noise remains an issue . this @xmath26 frequency noise has been reported to increase in the presence of multiple oscillation modes @xcite , but is nonetheless of considerable strength in the apparently single - mode case . there is currently no theory of stnos that can account for the presence of low - frequency @xmath26 frequency noise in single - mode oscillations , but it has been suggested that this might be related to the existence of unexcited oscillation modes or @xmath26-type amplitude noise possibly introduced by microscopic material inhomogeneity @xcite . two major paths are currently discernible for stabilizing the generation of stno and shno frequencies by applying electrical signals to the device . the first is injection - locking , in which the stno @xcite or shno @xcite is locked to an injected rf signal at the same nominal frequency or at an integer or integer fraction multiple thereof . the locking can be achieved through direct electrical injection of the external rf signal ( superposed on the dc drive current ) or through an ac magnetic field from an adjacent microstrip line . one compelling system is that in which the stno / stho itself acts as the injection source , thereby acting to stabilize itself . such a system has been investigated @xcite analytically and , for the case of a vortex mtj stno , also numerically . with the correct choice of delay time for the reinjected signal , the linewidth for this vortex mtj stno case could be decreased by a factor of 4 . self - injection locking has recently been achieved also experimentally , showing a maximum decrease in linewidth by a factor of 2.5 @xcite . a second path is to use a phase - locked loop ( pll ) for the stno / shno . in a pll , the oscillator s rf signal is compared with a reference signal and the resulting phase difference is used to instantly rebias the oscillator , taking advantage of its frequency - tuning mechanism . pll - locking through an external reference signal has been experimentally achieved for stno vortex oscillations below 150 mhz @xcite and , very recently , for a mtj stno at 5 ghz @xcite . the limited number of reported studies is probably related to the lack of pll circuits with bandwidths high enough to track the stno signal . a unique feature of stnos and shnos is their ability to provide multiple functionalities , from microwave source to microwave detection and signal processing . this versatility opens up the possibility of employing stnos and shnos in a wide variety of applications . the most common applications include microwave sources , microwave detectors , noncoherent transceiver , magnonics , neuromorphic computing , magnetic field sensing , and magnetic field generation . these applications will be introduced in this section . moreover , the issues , challenges , and developments in stno- and shno - based applications will be discussed . the miniature size , very wide tunability , high integration level with cmos technology , and low power consumption of stnos make them good candidates for current - controlled oscillators in multiband multistandard radio systems @xcite . table i summarizes the performance of different stnos in the literature . as can be seen from table i , a single mtj stno @xcite can be used as a local oscillator ( lo ) covering the entire bandwidth of ultrawideband ( uwb ) applications . gmr stnos typically have higher and broader tunable frequency ranges . the operating frequency of gmr stnos has been experimentally observed up to 46 ghz and extrapolated to 65 ghz @xcite . it has been theoretically estimated that the operating frequency of stnos may approach 200 ghz if stno nanocontacts can be made much smaller @xcite . more recently , a very low gilbert damping and ultrafast spin precession with frequencies of up to 280 ghz has been demonstrated @xcite , which makes it possible to further extend the operating frequencies of stnos . the extremely high and broad operating frequency range of stnos also opens up the possibility of utilizing them in millimeter wave application for instance , in wideband data transmission systems . [ cols="<,<,<,<,<,<,<,<",options="header " , ] the phase noise of the mtj stno was measured by nanosc ab at about -65 dbc / hz at 1 mhz offset . the issues of spectrum impurity and the limited output power must be overcome before stnos enter widespread use as microwave sources . recently , significant progress has been achieved in understanding the fundamentals and in overcoming these two limitations of stnos @xcite . as can be seen in table i , a sombrero - shaped nanocontact geometry for mtjs has been developed @xcite that remedies the incompatibility of conventional nanocontact geometry with mtjs . this geometry enables the optimization of either the q factor ( narrower linewidth ) @xcite or the output power @xcite . specifically , a very narrow linewidth of 3.4 mhz @xcite and an output power of -26.2 dbm @xcite were obtained . these newly developed mtj stnos show significant improvement in either linewidth or output power in comparison with the conventional mtj stnos @xcite . in addition , recent research has shown that by using a tilted cofeb fl @xcite , the bias field can be removed , which opens up more possibilities for mtj stno - based applications . for gmr stnos , which typically have a narrower linewidth than mtj stnos , the output power has been enhanced by using cofemnsi ( cfms ) layers as magnetic layers @xcite . by further combining the cfms with nanocontact geometry , a large output power of -46 dbm and a narrow linewidth of 3 mhz can be achieved simultaneously at zero applied field @xcite . in addition , the phase noise of stnos has been reduced by using mutual synchronization and current modulation as an extrinsic tool @xcite . the limited output power has been improved by employing amplifiers dedicated to stnos @xcite . the achieved improvements of phase noise and output power bring stnos closer to being used as microwave sources . in addition to these issues , the device - to - device performance variation of stnos is significant . furthermore , the mode structure and temperature dependence of stnos need to be well understood so as to suppress the unwanted modes and to estimate the performance variation that results from changes in temperature . other issues that need to be taken into consideration in applications include the irreversible changes of stnos and thus reliability issues at high biasing current @xcite . the stt provides many useful phenomena as detailed in @xcite . one important phenomenon is the spin - torque diode effect discovered in mtj - based devices @xcite . the spin - torque diode effect can be described as follows : when a microwave signal with a frequency close to the natural fmr frequency of one of the mtj s electrodes is incident on the device , the induced oscillating tunnel current can excite magnetic precession via spin transfer . the resistance oscillation that results from this precession is combined with the oscillating current so as to produce a measurable dc voltage component across the tunnel junction @xcite . recently , an mtj microwave detector that does not require any external bias fields , and that possesses a sensitivity of about 75400 mv / mw has been demonstrated at room temperature @xcite . in addition , this microwave detector can provide very high sensitivity even at a low input microwave powers of 10 nw ( -50 dbm ) . its sensitivity is about one order of magnitude higher than that of a semiconductor diode , such as a schottky diode . the large sensitivity of mtj stnos makes them promising candidates as frequency - tunable resonant microwave detectors for a wide range of applications in telecommunications , radars , and smart networks @xcite . taking advantage of the fast switching speed of stnos , an stno - based wireless communication system with binary amplitude shift keying ( ask ) has been demonstrated in @xcite . in this system , the stno performs the ask modulation at the transmitter and a simple envelope detector recovers the data at the receiver . for this noncoherent communication system , the considerable phase noise of stnos is not an issue . an stno with 1 ns turn - on time ( or switching speed ) allows a communication system that can achieve a data rate as high as 1.48 gbps @xcite . considering also the spin - torque diode effect , stnos can be utilized as both modulators and demodulators in transceiver systems , as shown in fig . [ sto_based_transceiver ] . as modulators , stnos translate different currents representing the binary data `` 0 '' and `` 1 '' directly into signals at different frequencies . as demodulators , stnos enable the direct conversion of the received signals to high and low output voltages representing `` 0 '' and `` 1 '' . these stno - based transceivers may enable the development of low - complexity , low - cost , high data - rate , and noncoherent communication systems . stnos with a point contact ( fig . 2(a ) ) on an extended free layer allow spin - torque - driven emission as well as propagation of spin waves outward from the nanocontact @xcite . the spin wave emitted from an stno with a nanoscale electrical contact has been directly observed and can propagate for several micrometers @xcite . the emission of propagating spin waves is the basis of synchronizing arrays of stnos for microwave applications . moreover , stnos are very attractive as wave emitters for future use in magnonic devices . magnonics is an emerging and rapidly growing research field that connects magnetism , spintronics , and electronics @xcite . it investigates and utilizes magnetic phenomena connected with spin waves , in a broad sense . magnonics employs different devices to deal with spin waves . the key devices are spin - wave emitters , spin - wave manipulators ( modulators ) , and spin - wave detectors . stnos , as previously discussed , can play all the roles of these key devices . as a consequence , stnos have the potential to realize all the fundamental functions that are necessary for magnonics @xcite . neurons can be modelled as nonlinear oscillators that vary their rhythms based on incoming signals @xcite . to form a network of neurons coupled by synapses , the nonlinear oscillators need to be able to synchronize with each other in frequency or phase . the neural network can therefore be realized using the non - boolean nature of these nonlinear oscillators @xcite . stnos are able to act as nonlinear oscillators because they can synchronize when coupled via a mutual electrical or magnetic interaction . furthermore , it has been demonstrated that other spintronic devices including spin - torque memristors and spin - torque memories are very suitable in playing the roles of synapses and memory processes in mimicking neural networks @xcite . consequently , stnos , together with other spintronic devices , pave the way to full spintronic implementations of neural networks , including both polychronous wave computation and information storage @xcite . non - boolean computing based on stnos shows great potential in data - processing applications , particularly in neural networks . stno - based non - boolean computing is discussed in @xcite and its physical processes are reviewed in @xcite . particularly , the concept of a dual - pillar stno that decouples bias and read paths for non - boolean energy - efficient computing has been proposed , and a simulation framework has been presented in @xcite . the dual - pillar stno is composed of a gmr stno and an mtj stno . the gmr stno is employed for the bias path so as to achieve a low - resistance interface and hence to allow low bias - voltage , as well as high energy efficiency . the high - resistance mtj stno is used to sense a small current and then to produce an output signal with large amplitude . the magnetoresistive effect , which is necessary in the operation of stnos , is also widely used to translate detected magnetic field strength to resistance . specifically , gmr - based sensors have been commercialized for sensing magnetic fields which range in strength from the earth s field up to a couple of hundred oersted @xcite . in addition to this range , larger magnetic fields can also be measured by stnos , by virtue of measuring the change in their oscillation frequency caused by the difference between the bias field and the magnetic field being tested . this stno - based magnetic field sensor can provide high sensitivity up to 180 ghz / t @xcite . moreover , the output of the stno is the frequency or frequency change , rather than the signal amplitude , thus relaxing the requirement for the system to have a large signal - to - noise ratio . for this reason , stnos are very attractive for sensing large magnetic fields . the concept of employing stnos to generate localized ac magnetic fields for microwave - assisted magnetic recording ( mamr ) applications has been proposed in @xcite , theoretically analyzed in @xcite , and extensively studied in @xcite . toshiba recently demonstrated a new technology that uses the localized ac magnetic field to reverse the magnetization direction , allowing the excitement of a magnetization oscillation in only one specific magnetic layer of a multilayer magnetic recording system @xcite . this technology could enable the development of 3-d magnetic recording systems , where independent data are written and read from overlapping layers of a multilayer recording medium . the applications to which stnos are suited are not limited to the above suggestions . for instance , stnos have been found very useful as high - density massively parallel microwave signal processors and small phased array transmitters @xcite . it has also been demonstrated in @xcite that stnos can be employed to develop novel readers with high data transfer rates for hard disk drives ( hdds ) . other applications that would make full use of the nanoscale aspects and of the high - speed broad tunability of stnos need further exploration . with the recent emergence of the spin hall effect , a new type of spintronic oscillator , the shno , has been developed @xcite . this provides a new route for the development of microwave and magnonic devices @xcite . shnos exhibit several advantages over stnos , including easier nanofabrication , lower required dc current , direct optical access to the magnetodynamically active area , smaller radiation losses , and suppressed nonlinear damping process @xcite . however , the linewidth of shnos is on the same order as that of stnos , and their output power is currently lower than that of stnos . despite the issues with shnos , their advantages offer an opportunity to implement novel nanoscale microwave sources and emitters for wireless communications and magnonics applications @xcite . in addition to the need to improve stno and shno performance , stno electrical models , stno and shno integration into integrated circuits ( ic ) , and dedicated circuitry are also necessary for the development of applications based on stnos and shnos . electrical models of stnos and shnos are critical , since they can efficiently reduce the time , cost , and risk of errors involved in building prototypes of the entire system of the stno or shno with the ic . integration of the stno or shno with the ic is important , since it should enable better performance , miniaturization , and even single - chip solutions . dedicated circuitry is essential for stno / shno - based applications , since it can perform necessary functions such as biasing and amplification , and can extrinsically improve the performance of stnos or shnos . the issues , challenges , and development of the electrical models , integration , and dedicated circuitry are discussed in the following . the basic procedure for modeling stnos or shnos involves solving the landau - lifshitz - gilbert equation with a slonczewski spin - transfer term ( llgs ) . one possible approach to solving the llgs is based on micromagnetics @xcite . such a model presents the magnetization dynamics of stnos and shnos on a microscopic scale . it typically considers the anisotropy energy , the zeeman energy of the external field , the ferromagnetic exchange energy and the magnetostatic energy @xcite . however , the micromagnetic simulation typically takes days , which is not efficient for the development of stno / shno - based applications . additionally , the output dataset of this model can not be directly used in a circuit simulator . the complicated numeric computation involved in the micromagnetics - based model prevents it from being implemented in a hardware description language . a hardware description language is a specialized computer language used to describe both structural and behavioral aspects of designs including , but not limited to , electronic components , circuits , and systems @xcite . it allows the automatic analysis , simulation , and evaluation of designs , through the use of electronic design automation ( eda ) tools . the micromagnetics - based model is therefore not suited to evaluating stnos and shnos under different bias conditions together with other circuits , as would be required in stno / shno - based applications . an alternative approach to solving the llgs is based on the macrospin approximation , which assumes that only a spatially uniform magnetization precession is excited and that the spin - polarized current is uniform across the area of the free layer @xcite . a macrospin - based model is able to provide an analytical solution with acceptable accuracy , while allowing implementation in a hardware description language . as a result , a macrospin - based analytical model is needed to design and develop stno - based systems . several macrospin - based models of stnos @xcite have been developed and implemented in verilog - a , a commonly used hardware description language for analog systems . however , the models in @xcite require the aid of matlab to compute the effective magnetic field , which is then manually imported into the verilog - a model . consequently , they can not be considered fully implemented in verilog - a . moreover , these models @xcite use equations that are not fully validated . additionally , only one specific device has been used to verify the models , which is not sufficient . a comprehensive mtj stno model that can overcome the issues in @xcite , has been presented in @xcite . this further allows accurate replication of the phase noise and hence of the generated signal . moreover , this model can be used for various mtj stnos to estimate their performance together with the ics . nonetheless , this model is not suitable for simulations that involve momentary variations in the operating conditions , such as current or field modulation . to enable comprehensive analysis and evaluation of stnos or shnos together with ics , future models need to be divergence - free , while allowing all the necessary simulations and analyses . future models also need to include temperature dependence . to avoid the losses due to external components and connections , as well as to eliminate wave reflections , stnos and shnos need to be integrated with other technologies , such as silicon - cmos and gallium - arsenide ( gaas ) , which allow the circuits required for stnos and shnos to be implemented . additionally , integration is vital if the nanoscale of stnos and shnos is to be capitalized upon and to achieve compact stno / shno - based systems . the integration approach must be considered before implementing the circuits and developing the system , since it has a significant impact on the system topology and circuit design . the advantages and disadvantages of different possible integration approaches are discussed in @xcite . the wire - bonding approach has been successfully used in @xcite , mainly thanks to its independent preintegration characterization and optimization of the ic and stno . in @xcite , an mtj stno has been integrated with a dedicated cmos ic amplifier and an off - chip bias - tee using wire bonding . however , the use of the off - chip bias - tee limits both the integration level and the bandwidth of the system . the cmos ic proposed in @xcite includes an on - chip bias - tee and has been successfully integrated with an stno by bonding wires . the bonding wire has also been used as part of the bias - tee design , since it introduces extra high - frequency impedance into the ac decoupling path of the bias - tee . furthermore , the parasitics of the bonding wires in the signal path , as well as the on - chip bias - tee , determine the requirements of the input and output filters for achieving a flat frequency response over a large bandwidth . the wire - bonding approach is appropriate for the current level of maturity of stno technology , as it allows easy implementation , measurement , and replacement . monolithic and flip - chip - based approaches can ultimately result in a higher level of integration and better performance at high frequencies . the monolithic integration approach has been successfully used in hybrid cmos / mtj logic circuits ( fig . [ sto_ic_integration](a ) ) targeting the st - mram technology @xcite , which utilizes the same stt effect as that in stnos and shnos . for instance , a fully functional 64 mb st - mram built on a standard 90 nm process has been developed in @xcite . nevertheless , monolithic integration has not yet been implemented so as to handle the rf signals generated by stnos or shnos . in order to use monolithic integration , the top - layer surface roughness of the ic stacks needs to be very well controlled , as this layer acts as the cu buffer layer for fabricating stnos and shnos . the roughness of this layer has significant impact on the spatial variation in the multilayer structure , and hence on the dynamics of stnos and shnos @xcite . an rms roughness lower than @xmath27 is necessary to avoid unpredictable device performance and significant performance degradation @xcite . flip - chip - based integration is one of the principal methods of interconnecting ics @xcite . as the cost of flip - chip - based integration decreases @xcite , its application space broadens further . [ sto_ic_integration](b ) shows a possible integration between an stno / shno and a cmos ic using the flip - chip - based approach . compared to monolithic integration , flip - chip - based integration allows easier implementation and measurement and retains a low - loss interconnection @xcite . however , neither monolithic nor flip - chip - based integration will be suitable until the device - to - device performance has improved to the extent that no major adjustments ( such as to the topology ) of the circuitry are required for different devices . the stno / shno - based applications described above could be possibly achieved by integrating stnos or shnos into the existing system architectures and by employing the existing circuits implemented in other technologies @xcite . for instance , a system architecture similar to the simple noncoherent transceiver in @xcite has been used to develop an stno - based wireless communication system @xcite . an amplifier using the existing topology of a cascode differential pair with a resistive load has been implemented in @xcite to provide amplification for the stno . additionally , considering the device - to - device performance variation of stnos and shnos , dedicated circuitry may need to allow adjustable performance so as to avoid multiple designs for different stnos or shnos . for instance , the variable gain amplifier ( vga ) @xcite and the tunable lna @xcite , whose gain and bandwidth , respectively , can be tuned to some extent , may be suitable for providing adjustable gain and bandwidth for different stnos or shnos . however , several dedicated circuits need to be designed for stnos and shnos . the first important circuit is the bias - tee , which is necessary for applying the required dc bias current into stnos and shnos without disturbing the rf signals generated . current mirrors have been considered in @xcite for biasing stnos . however , the resistance ( biasing voltage ) of the stno changes as the angle of the magnetic field is varied , so the current mirror can not accurately copy the current from a current source to the stno under all circumstances . accordingly , an on - chip bias - tee that can provide accurate current for stnos and shnos is necessary . factors that also need to be taken into account during the design of the bias - tee include the maximum applied dc current , the applied integration approach , the resistance of stnos and shnos , the power supply voltage of ic technology , the clamp circuit for providing esd protection , and the parasitics . a dedicated sample bias - tee design is given in @xcite . this design considers the above factors and takes care of the trade - offs involved in selecting the number of turns , the separation distance between each turn , and the width of metalization needed for the inductor to act as the ac decoupling component . to implement the dedicated on - chip bias - tee , electromagnetic simulations may be required when custom inductors are necessary @xcite . the second important circuit for stnos and shnos is the amplifier , which needs to provide sufficient gain and bandwidth for a given application while featuring low noise and proper input impedance . for example , to use stnos or shnos as los , the amplifier should typically enhance the output power of the stnos and shnos to 0 dbm so as to drive an rf mixer to perform frequency conversion . to take full advantage of stnos or shnos in multiband multistandard radios , the bandwidth of the amplifier needs to cover multiple bands of interest . additionally , the amplifier must introduce as little noise as possible to the entire system . the input impedance of the amplifier needs to match the impedance of stnos or shnos for maximizing the transferred power . nevertheless , for stnos or shnos with a small resistance ( a few ohms ) , wideband impedance matching to such a resistance is not feasible . this is because any parasitic will introduce an impedance mismatch much larger than a few ohms . in this case , rather than maximizing the transferred power , an amplifier with large input impedance is preferable , since it can maximize the ac voltage delivered to the amplifier as well as the signal - to - noise ratio at the output @xcite . other important circuits include those that handle the phase noise of stnos and shnos , such as phase - locked loops ( plls ) and injection - lock ring oscillators ( ilros ) for frequency synthesis . this is due to the fact that the frequency fluctuation of stnos occurs every several nanoseconds or even few tens of nanoseconds @xcite . the fast frequency change requires a rapid lock time ( or settling time ) of at most a few nanoseconds , which is difficult to achieve in either plls or ilros . recently , the phase of an mtj stno , which generates a 5.12 ghz microwave signal , has been successfully locked by a pll to that of an 80 mhz reference clock , resulting in a significant increase in q from 800 to 5@xmath28 @xcite . this is the first demonstration of phase synchronization of stno output to a lower frequency external signal , and it may be a milestone toward practical applications of stnos . state - of - the - art stnos and shnos feature very wide tunability , high levels of integration with cmos , high operating frequencies , fast switching speeds , and miniature size . their unique blend of features suggests a wide range of applications , including microwave sources , microwave detectors , noncoherent transceivers , magnonics , neuromorphic computing , magnetic field sensing , and magnetic field generation . over the past decade , extensive research has taken place to understand and further exploit the nonlinear physical phenomena of stnos and shnos . in this paper , we have introduced the fundamentals , geometries , and functional properties of stnos and shnos . we have additionally reviewed the state of the art in stno and shno technologies . furthermore , we have presented the potential for these oscillators in their popular applications . however , challenges still remain on both the device and system level . these need to be overcome before the use of stnos and shnos can become widespread . this work is partially supported by the swedish research council ( vr ) , the swedish foundation for strategic research ( ssf ) , the knut and alice wallenberg foundation ( kaw ) , the european commission fp7-ict-2011-contract no . 317950 `` mosaic '' , and the european research council ( erc ) under the european community s seventh framework programme ( fp/2007 - 2013)/erc grant 307144 `` mustang '' . m. n. baibich , j. m. broto , a. fert , f. n. van dau , f. petroff , p. etienne , g. creuzet , a. friederich , and j. chazelas , `` giant magnetoresistance of ( 001)fe/(001)cr magnetic superlattices , '' _ phys . _ , vol . 61 , no . 21 , p. 2472 , 1988 . j. a. katine , f. j. albert , r. a. buhrman , e. b. myers , and d. c. ralph , `` current - driven magnetization reversal and spin - wave excitations in co @xmath29cu @xmath29co pillars , '' _ phys . _ , vol . 84 , pp . 31493152 , apr 2000 . j. wunderlich , b. kaestner , j. sinova , and t. jungwirth , `` experimental observation of the spin - hall effect in a two - dimensional spin - orbit coupled semiconductor system , '' _ phys . rev . _ , vol . 94 , no . 4 , pp . 14 , 2005 . i. m. miron , k. garello , g. gaudin , p .- j . zermatten , m. v. costache , s. auffret , s. bandiera , b. rodmacq , a. schuhl , and p. gambardella , `` perpendicular switching of a single ferromagnetic layer induced by in - plane current injection , '' _ nature _ , vol . 7359 , pp . 189193 , jul . 2011 . l. liu , c .- f . pai , d. c. ralph , and r. a. buhrman , `` magnetic oscillations driven by the spin hall effect in 3-terminal magnetic tunnel junction devices , '' _ phys 109 , no . 18 , p. 186602 , oct . 2012 . v. e. demidov , s. urazhdin , h. ulrichs , v. tiberkevich , a. slavin , d. baither , g. schmitz , and s. o. demokritov , `` magnetic nano - oscillator driven by pure spin current , '' _ nature mater . _ , vol . 11 , no . 12 , pp 10281031 , dec . 2012 . b. n. engel , j. kerman , b. butcher , r. w. dave , m. deherrera , m. durlam , g. grynkewich , j. janesky , s. v. pietambaram , n. d. rizzo , j. m. slaughter , k. smith , j. j. sun , and s. tehrani , `` a 4-mb toggle mram based on a novel bit and switching method , '' _ ieee trans . _ , vol . 41 , no . 1 , 132136 , jan . 2005 . r. l. stamps , s. breitkreutz , j. kerman , a. v. chumak , y. otani , g. e. w. bauer , j .- u . thiele , m. bowen , s. a. majetich , m. klui , i. l. prejbeanu , b. dieny , n. m. dempsey , and b. hillebrands , `` the 2014 magnetism roadmap , '' _ j. phys . _ , vol . 47 , no . 33 , p. 333001 , 2014 . m. zwierzycki , y. tserkovnyak , p. j. kelly , a. brataas , and g. e. w. bauer , `` first - principles study of magnetization relaxation enhancement and spin transfer in thin magnetic films , '' _ phys . b _ , vol . 71 , p. 064420 , feb 2005 . r. matsumoto , a. chanthbouala , j. grollier , v. cros , a. fert , k. nishimura , y. nagamine , h. maehara , k. tsunekawa , a. fukushima , and s. yuasa , `` spin - torque diode measurements of mgo - based magnetic tunnel junctions with asymmetric electrodes , '' _ appl . , vol . 4 , no . 6 , p. 063001 , 2011 . m. tsoi , a. g. jansen , j. bass , w. c. chiang , v. tsoi , and p. wyder , `` generation and detection of phase - coherent current - driven magnons in magnetic multilayers , '' _ nature _ , vol . 6791 , pp . 4648 , jul . 2000 . w. h. rippard , m. r. pufall , and t. j. silva , `` quantitative studies of spin - momentum - transfer - induced excitations in co / cu multilayer films using point - contact spectroscopy , '' _ appl . phys _ , vol . 82 , no . 8 , p. 1260 , 2003 . m. r. pufall , w. h. rippard , and t. j. silva , `` materials dependence of the spin - momentum transfer efficiency and critical current in ferromagnetic metal / cu multilayers , '' _ appl . _ , vol . 83 , no . 2 , pp . 323325 , 2003 . w. h. rippard , m. r. pufall , s. kaka , s. e. russek , and t. j. silva , `` direct - current induced dynamics in co@xmath9fe@xmath10/ni@xmath16fe@xmath14 point contacts , '' _ phys . _ , vol . 92 , p. 027201 , jan 2004 . r. k. dumas , s. r. sani , s. m. mohseni , e. iacocca , y. pogoryelov , p. k. muduli , s. chung , p. drrenfeld , and j. kerman , `` recent advances in nanocontact spin - torque oscillators , '' _ ieee trans . _ , vol . 50 , no . 6 , p. 4100107 , jun . 2014 . s. sani , j. persson , s. mohseni , y. pogoryelov , p. muduli , a. eklund , g. malm , m. kll , a. dmitriev , and j. kerman , `` mutually synchronized bottom - up multi - nanocontact spin - torque oscillators , '' _ nat . _ , vol . 4 , p. 2731 , nov . 2013 . s. i. kiselev , j. c. sankey , i. n. krivorotov , n. c. emley , r. j. schoelkopf , r. a. buhrman , and d. c. ralph , `` microwave oscillations of a nanomagnet driven by a spin - polarized current , '' _ nature _ , vol . 425 , no . 6956 , pp . 380383 , sep . 2003 . wegrowe , a. fabian , p. guittienne , x. hoffer , d. kelly , j ansermet , and e. olive , `` exchange torque and spin transfer between spin polarized current and ferromagnetic layers , '' _ appl . _ , vol . 80 , no . 37753777 , 2002 . f. abreu araujo , m. darques , k. a. zvezdin , a. v. khvalkovskiy , n. locatelli , k. bouzehouane , v. cros , and l. piraux , `` microwave signal emission in spin - torque vortex oscillators in metallic nanowires : experimental measurements and micromagnetic numerical study , '' _ phys . b _ , vol . 86 , p. 064424 , aug 2012 . f. abreu araujo , l. piraux , v. a. antohe , v. cros , and l. gence , `` single spin - torque vortex oscillator using combined bottom - up approach and e - beam lithography , '' _ appl . 102 , no . 22 , p. 222402 , 2013 . w. h. rippard , m. r. pufall , and s. e. russek , `` comparison of frequency , linewidth , and output power in measurements of spin - transfer nanocontact oscillators , '' _ phys . b _ , vol . 74 , p. 224409 , dec 2006 . j. z. sun , m. c. gaidis , e. j. osullivan , e. a. joseph , g. hu , d. w. abraham , j. j. nowak , p. l. trouilloud , y. lu , s. l. brown , d. c. worledge , and w. j. gallagher , `` a three - terminal spin - torque - driven magnetic switch , '' _ appl . _ , vol . 95 , no . 8 , p. 083506 , 2009 . l. xue , c. wang , y .- t . cui , l. liu , a. swander , j. z. sun , r. a. buhrman , and d. c. ralph , `` resonance measurement of nonlocal spin torque in a three - terminal magnetic device , '' _ phys . 108 , p. 147201 , apr 2012 . v. e. demidov , s. urazhdin , a. zholud , a. v. sadovnikov , a. n. slavin , and s. o. demokritov , `` spin - current nano - oscillator based on nonlocal spin injection , '' _ sci . rep . _ , vol . 5 , no . 8578 , pp . 15 , 2015 . v. e. demidov , s. urazhdin , h. ulrichs , v. tiberkevich , a. slavin , d. baither , g. schmitz , and s. o. demokritov , `` magnetic nano - oscillator driven by pure spin current , '' _ nature mater . _ , vol . 11 , pp . 10281031 , 2012 . z. duan , a. smith , l. yang , b. youngblood , j. lindner , v. e. demidov , s. o. demokritov , and i. n. krivorotov , `` nanowire spin torque oscillator driven by spin orbit torques , '' _ nat . commun . _ , vol . 5 , p. 5616 a. m. deac , a. fukushima , h. kubota , h. maehara , y. suzuki , s. yuasa , y. nagamine , k. tsunekawa , d. d. djayaprawira , and n. watanabe , `` bias - driven high - power microwave emission from mgo - based tunnel magnetoresistance devices , '' _ nature phys . _ , vol . 4 , no . 10 , 803809 , oct . h. maehara , h. kubota , y. suzuku , t. seki , k. nishimura , y. nagamine , k. tsunekawa , a. fukushima , a. m. deac , and s. yuasa , `` large emission power over 2 @xmath15w with high q factor obtained from nanocontact magnetic - tunnel - junction - based spin torque oscillator , '' _ appl . , vol . 6 , no . 11 , p. 113005 , 2013 . h. maehara , h. kubota , y. suzuki , t. seki , k. nishimura , y. nagamine , k. tsunekawa , a. fukushima , h. arai , t. taniguchi , h. imamura , k. ando , and s. yuasa , `` high q factor over 3000 due to out - of - plane precession in nano - contact spin - torque oscillator based on magnetic tunnel junctions , '' _ appl . 7 , no . 2 , p. 023003 , 2014 . s. mangin , d. ravelosona , j. a. katine , m. j. carey , b. d. terris , and e. e. fullerton , `` current - induced magnetization reversal in nanopillars with perpendicular anisotropy , '' _ nature mater . _ , vol . 5 , no . 3 , pp . 210215 , 2006 . d. houssameddine , u. ebels , b. delaet , b. rodmacq , i. firastrau , f. ponthenier , m. brunet , c. thirion , j .- p . michel , l. prejbeanu - buda , m .- c . cyrille , o. redon , and b. dieny , `` spin - torque oscillator using a perpendicular polarizer and a planar free layer , '' _ nature mater . _ , vol . 6 , no . 6 , pp . 447453 , jun 2007 . w. h. rippard , a. m. deac , m. r. pufall , j. m. shaw , m. w. keller , s. e. russek , g. e. w. bauer , and c. serpico , `` spin - transfer dynamics in spin valves with out - of - plane magnetized coni free layers , '' _ phys . rev . b _ , vol . 81 , p. 014426 , jan 2010 . s. m. mohseni , s. r. sani , j. persson , t. n. anh nguyen , s. chung , y. pogoryelov , and j. kerman , `` high frequency operation of a spin - torque oscillator at low field , '' _ phys . status solidi rapid res . _ , vol . 5 , no . 12 , pp . 432434 , 2011 . s. m. mohseni , s. r. sani , j. persson , t. n. a. nguyen , s. chung , y. pogoryelov , p. k. muduli , e. iacocca , a. eklund , r. k. dumas , s. bonetti , a. deac , m. a. hoefer , and j. kerman , `` spin torque - generated magnetic droplet solitons , '' _ science _ 6125 , pp . 12951298 , mar . s. m. mohseni , s. r. sani , r. k. dumas , j. persson , t. n. anh nguyen , s. chung , y. pogoryelov , p. k. muduli , e. iacocca , a. eklund , and j. kerman , `` magnetic droplet solitons in orthogonal nano - contact spin torque oscillators , '' _ physica b : condensed matter _ , 435 , pp . 84 87 , 2014 . s. chung , s. m. mohseni , s. r. sani , e. iacocca , r. k. dumas , t. n. anh nguyen , y. pogoryelov , p. k. muduli , a. eklund , m. hoefer , and j. kerman , `` spin transfer torque generated magnetic droplet solitons ( invited ) , '' _ j. appl 115 , no . 17 , p. 172612 , may 2014 . y. zhou , s. bonetti , c. l. zha , and j. kerman , `` zero - field precession and hysteretic threshold currents in a spin torque nano device with tilted polarizer , '' _ new j. phys . _ , vol . 11 , no . 10 , p. 103028 , 2009 . t. n. a. nguyen , y. fang , v. fallahi , n. benatmane , s. m. mohseni , r. k. dumas , and j. kerman , `` [ co / pd]-nife exchange springs with tunable magnetization tilt angle , '' _ appl . _ , vol . 98 , no . 17 , p. 172502 , 2011 . t. a. nguyen , n. benatmane , v. fallahi , y. fang , s. mohseni , r. dumas , and j. kerman , `` [ co / pd]4-co - pd - nife spring magnets with highly tunable and uniform magnetization tilt angles , '' _ j. magn . 324 , no . 3929 3932 , 2012 . s. chung , s. m. mohseni , v. fallahi , t. n. anh nguyen , n. benatmane , r. k. dumas , and j. kerman , `` tunable spin configuration in [ co / ni]-nife spring magnets , '' _ j. phys . _ , vol . 46 , no . 12 , p. 125004 , 2013 . s. tacchi , t. n. a. nguyen , g. carlotti , g. gubbiotti , m. madami , r. k. dumas , j. w. lau , j. kerman , a. rettori , and m. g. pini , `` spin wave excitations in exchange - coupled [ co / pd]-nife films with tunable tilting of the magnetization , '' _ phys . b _ , vol . 87 , p. 144426 , apr 2013 . s. tacchi , t. n. a. nguyen , g. gubbiotti , m. madami , g. carlotti , m. g. pini , a. rettori , v. fallahi , r. k. dumas , and j. kerman , `` [ co / pd]-cofeb exchange spring magnets with tunable gap of spin wave excitations , '' _ j. phys . _ , vol . 47 , no . 49 , p. 495004 , 2014 . w. skowroski , t. stobiecki , j. wrona , g. reiss , and s. van dijken , `` zero - field spin torque oscillator based on magnetic tunnel junctions with a tilted cofeb free layer , '' _ appl . , vol . 5 , no . 6 , p. 063005 , 2012 . s. sani , p. drrenfeld , s. mohseni , s. chung , and j. kerman , `` microwave signal generation in single - layer nano - contact spin torque oscillators , '' _ ieee trans . _ , vol . 49 , no . 7 , pp . 43314334 , july 2013 . v. s. pribiag , i. n. krivorotov , g. d. fuchs , p. m. braganca , o. ozatay , j. c. sankey , d. c. ralph , and r. a. buhrman , `` magnetic vortex oscillator driven by d.c . spin - polarized current , '' _ nature phys . _ , vol . 3 , no . 7 , pp . 498503 , jul . 2007 . v. sluka , a. kkay , a. m. deac , d. e. brgler , r. hertel , and c. m. schneider , `` spin - transfer torque induced vortex dynamics in fe / ag / fe nanopillars , '' _ j. phys . _ , vol . 44 , no . 38 , p. 384002 , 2011 . n. locatelli , v. v. naletov , j. grollier , g. de loubens , v. cros , c. deranlot , c. ulysse , g. faini , o. klein , and a. fert , `` dynamics of two coupled vortices in a spin valve nanopillar excited by spin transfer torque , '' _ appl . _ , vol . 98 , no . 6 , pp . , a. v. khvalkovskiy , j. grollier , a. dussaux , k. a. zvezdin , and v. cros , `` vortex oscillations induced by spin - polarized current in a magnetic nanopillar : analytical versus micromagnetic calculations , '' _ phys . b _ , vol . 80 , p. 140401 , oct 2009 . p. bortolotti , e. grimaldi , a. dussaux , j. grollier , v. cros , c. serpico , k. yakushiji , a. fukushima , h. kubota , r. matsumoto , and s. yuasa , `` parametric excitation of magnetic vortex gyrations in spin - torque nano - oscillators , '' _ phys . b _ , vol . 88 , p. 174417 , nov 2013 . e. grimaldi , a. dussaux , p. bortolotti , j. grollier , g. pillet , a. fukushima , h. kubota , k. yakushiji , s. yuasa , and v. cros , `` response to noise of a vortex based spin transfer nano - oscillator , '' _ phys . b _ , vol . 89 , p. 104404 , mar 2014 . r. lehndorff , d. e. brgler , s. gliga , r. hertel , p. grnberg , c. m. schneider , and z. celinski , `` magnetization dynamics in spin torque nano - oscillators : vortex state versus uniform state , '' _ phys . b _ , vol . 80 , p. 054412 , aug 2009 . g. consolo , b. azzerboni , g. gerhart , g. a. melkov , v. tiberkevich , and a. n. slavin , `` excitation of self - localized spin - wave bullets by spin - polarized current in in - plane magnetized magnetic nanocontacts : a micromagnetic study , '' _ phys . rev . b _ , vol . 76 , p. 144410 , oct 2007 . s. bonetti , v. tiberkevich , g. consolo , g. finocchio , p. muduli , f. mancoff , a. slavin , and j. kerman , `` experimental evidence of self - localized and propagating spin wave modes in obliquely magnetized current - driven nanocontacts , '' _ phys . 105 , p. 217204 , nov 2010 . m. madami , s. bonetti , g. consolo , s. tacchi , g. carlotti , g. gubbiotti , f. b. mancoff , m. a. yar , and j. kerman , `` direct observation of a propagating spin wave induced by spin - transfer torque , '' _ nature nanotech . _ , vol . 6 , p. 635 , august 2011 . s. bonetti , v. puliafito , g. consolo , v. s. tiberkevich , a. n. slavin , and j. kerman , `` power and linewidth of propagating and localized modes in nanocontact spin - torque oscillators , '' _ phys . b _ , vol . 85 , p. 174427 , may 2012 . n. c. krivorotov , i. n. andemley , j. c. sankey , s. i. kiselev , d. c. ralph , and r. a. buhrman , `` time - domain measurements of nanomagnet dynamics driven by spin - transfer torques , '' _ science _ , vol . 5707 , pp . 228231 , 2005 . r. k. dumas , e. iacocca , s. bonetti , s. r. sani , s. m. mohseni , a. eklund , j. persson , o. heinonen , and j. kerman , `` spin - wave - mode coexistence on the nanoscale : a consequence of the oersted - field - induced asymmetric energy landscape , '' _ phys . rev . lett . 110 , p. 257202 , jun 2013 . m. madami , e. iacocca , s. sani , g. gubbiotti , s. tacchi , r. k. dumas , j. kerman , and g. carlotti , `` propagating spin waves excited by spin - transfer torque : a combined electrical and optical study , '' _ phys . b _ , vol . 92 , p. 024403 , jul 2015 . q. mistral , m. van kampen , g. hrkac , j .- v . kim , t. devolder , p. crozat , c. chappert , l. lagae , and t. schrefl , `` current - driven vortex oscillations in metallic nanocontacts , '' _ phys . 100 , p. 257201 , jun 2008 . a. ruotolo , v. cros , b. georges , a. dussaux , j. grollier , c. deranlot , r. guillemet , k. bouzehouane , s. fusil , and a. fert , `` phase - locking of magnetic vortices mediated by antivortices , '' _ nature nanotech . _ , vol . 4 , no . 8 , 528532 , aug . m. manfrini , j .- v . kim , s. petit - watelot , w. van roy , l. lagae , c. chappert , and t. devolder , `` propagation of magnetic vortices using nanocontacts as tunable attractors , '' _ nature nanotech . _ , vol . 9 , no . 20 , pp . 121125 , feb 2014 , letter . s. petit - watelot , j .- v . kim , a. ruotolo , r. m. otxoa , k. bouzehouane , j. grollier , a. vansteenkiste , b. van de wiele , v. cros , and t. devolder , `` commensurability and chaos in magnetic vortex oscillations , '' _ nature phys . _ , vol . 8 , no . 9 , pp . 682687 , sep . 2012 . v. e. demidov , s. urazhdin , e. r. j. edwards , m. d. stiles , r. d. mcmichael , and s. o. demokritov , `` control of magnetic fluctuations by spin current , '' _ phys . 107 , p. 107204 , sep 2011 . m. ranjbar , p. drrenfeld , m. haidar , e. iacocca , m. balinskiy , t. q. le , m. fazlali , a. houshang , a. a. awad , r. k. dumas , and j. kerman , `` cofeb - based spin hall nano - oscillators , '' _ ieee magn . _ , vol . 5 , p. 3000504 , 2014 . m. collet , x. de milly , o. dallivy kelly , v. v. naletov , r. bernard , p. bortolotti , v. e. demidov , s. o. demokritov , j. l. prieto , m. mu noz , v. cros , a. anane , g. de loubens , and o. klein , `` generation of coherent spin - wave modes in yttrium iron garnet microdiscs by spin - orbit torque , '' _ arxiv:1504.01512 [ cond-mat.mtrl-sci]_ , 2015 . g. gerhart , e. bankowski , g. a. melkov , v. s. tiberkevich , and a. n. slavin , `` angular dependence of the microwave - generation threshold in a nanoscale spin - torque oscillator , '' _ phys . b _ , vol . 76 , no . 2 , p. 024437 2007 . f. abreu araujo , a. d. belanovsky , p. n. skirdkov , k. a. zvezdin , a. k. zvezdin , n. locatelli , r. lebrun , j. grollier , v. cros , g. de loubens , and o. klein , `` optimizing magnetodipolar interactions for synchronizing vortex based spin - torque nano - oscillators , '' _ phys . b _ , vol . 92 , p. 045419 , jul 2015 . s. kaka , m. r. pufall , w. h. rippard , t. j. silva , s. e. russek , and j. a. katine , `` mutual phase - locking of microwave spin torque nano - oscillators , '' _ nature _ , vol . 7057 , pp . 389392 , sep . 2005 . m. r. pufall , w. h. rippard , s. e. russek , s. kaka , and j. a. katine , `` electrical measurement of spin - wave interactions of proximate spin transfer nanooscillators , '' _ phys . _ , vol . 97 , p. 087206 , aug 2006 . t. kendziorczyk , s. o. demokritov , and t. kuhn , `` spin - wave - mediated mutual synchronization of spin - torque nano - oscillators : a micromagnetic study of multistable phase locking , '' _ phys . b _ , vol . 90 , p. 054414 , aug 2014 . a. houshang , p. drrenfeld , e. iacocca , s. r. sani , j. kerman , and r. k. dumas , `` spin wave beam driven synchronisation of multiple nanocontact spin torque oscillators , '' _ nature nanotech . _ , under review . m. hoefer , m. ablowitz , b. ilan , m. pufall , and t. silva , `` theory of magnetodynamics induced by spin torque in perpendicularly magnetized thin films , '' _ phys . _ , vol . 95 , no . 26 , p. 267206 s. bonetti , p. muduli , f. mancoff , and j. kerman , `` spin torque oscillator frequency versus magnetic field angle : the prospect of operation beyond 65 ghz , '' _ appl . _ , vol . 94 , no . 10 , p. 102507 , 2009 . k. muduli , y. pogoryelov , y. zhou , f. mancoff , and j. kerman , `` spin torque oscillators and rf currents - modulation , locking , and ringing , '' _ integr . 125 , no . 1 , pp . 147154 , 2011 . j. hasch , e. topak , r. schnabel , t. zwick , r. weigel , and c. waldschmidt , `` millimeter - wave technology for automotive radar sensors in the 77 ghz frequency band , '' _ ieee trans . theory tech . _ , vol . 60 , no . 3 , pp . 845860 , march 2012 . s. tsunegi , h. kubota , k. yakushiji , m. konoto , s. tamaru , a. fukushima , h. arai , h. imamura , e. grimaldi , r. lebrun , j. grollier , v. cros , and s. yuasa , `` high emission power and q factor in spin torque vortex oscillator consisting of feb free layer , '' _ appl . , vol . 7 , no . 6 , p. 063009 , 2014 . k. muduli , y. pogoryelov , s. bonetti , g. consolo , f. mancoff , and j. kerman , `` nonlinear frequency and amplitude modulation of a nanocontact - based spin - torque oscillator , '' _ phys . b _ , vol . 81 , no . 14 , p. 140408 k. muduli , y. pogoryelov , f. mancoff , and j. kerman , `` modulation of individual and mutually synchronized nanocontact - based spin torque oscillators , '' _ ieee trans . _ , vol . 47 , pp . 15751579 , jun . 2011 . p. drrenfeld , e. iacocca , j. kerman , and p. k. muduli , `` modulation - mediated unlocking of a parametrically phase - locked spin torque oscillator , '' _ appl . 105 , no . 25 , p. 252404 , 2014 . m. manfrini , t. devolder , j .- v . kim , p. crozat , n. zerounian , c. chappert , w. van roy , l. lagae , g. hrkac , and t. schrefl , `` agility of vortex - based nanocontact spin torque oscillators , '' _ appl . _ , vol . 95 , no . 19 , p. 192507 , nov . 2009 . m. quinsat , f. garcia - sanchez , a. s. jenkins , v. s. tiberkevich , a. n. slavin , l. d. buda - prejbeanu , a. zeltser , j. a. katine , b. dieny , m .- c . cyrille , and u. ebels , `` modulation bandwidth of spin torque oscillators under current modulation , '' _ appl . 105 , no . 15 , p. 152401 , 2014 . e. iacocca and j. kerman , `` analytical investigation of modulated spin - torque oscillators in the framework of coupled differential equations with variable coefficients , '' _ phys . b _ , vol . 85 , p. 184420 , may 2012 . d. houssameddine , u. ebels , b. dieny , k. garello , j .- p . michel , b. delaet , b. viala , m .- c . cyrille , j. a. katine , and d. mauri , `` temporal coherence of mgo based magnetic tunnel junction spin torque oscillators , '' _ phys . 102 , no . 25 , p. 257202 , jun . 2009 . s. bonetti , v. puliafito , g. consolo , v. tiberkevich , a. slavin , and j. kerman , `` power and linewidth of propagating and localized modes in nanocontact spin - torque oscillators , '' _ phys . b _ , vol . 85 , no . 17 , pp . 18 , may 2012 . a. eklund , s. sani , s. mohseni , j. persson , b. malm , and j. @xmath30kerman , `` triple mode - jumping in a spin torque oscillator , '' in _ noise and fluctuations ( icnf ) , 2013 22nd international conference on _ , june 2013 , pp . m. r. pufall , w. h. rippard , s. e. russek , and e. r. evarts , `` anisotropic frequency response of spin - torque oscillators with applied field polarity and direction , '' _ phys . b _ , vol . 86 , p. 094404 , sep 2012 . k. muduli , o. g. heinonen , and j. kerman , `` temperature dependence of linewidth in nanocontact based spin torque oscillators : effect of multiple oscillatory modes , '' _ phys . b _ , vol . 86 , p. 174408 , nov 2012 . j. c. sankey , i. n. krivorotov , s. i. kiselev , p. m. braganca , n. c. emley , r. a. buhrman , and d. c. ralph , `` mechanisms limiting the coherence time of spontaneous magnetic oscillations driven by dc spin - polarized currents , '' _ phys . b _ , vol . 72 , p. 224427 , dec 2005 . d. v. berkov , `` magnetization dynamics including thermal fluctuations : basic phenomenology , fast remagnetization processes and transitions over high - energy barriers , '' in _ handbook of magnetism and advanced magnetic materials _ , h. kronmller and s. parkin , eds.1em plus 0.5em minus 0.4emjohn wiley & sons , ltd , 2007 , vol . 3 . l. bianchini , s. cornelissen , j .- v . kim , t. devolder , w. van roy , l. lagae , and c. chappert , `` direct experimental measurement of phase - amplitude coupling in spin torque oscillators , '' _ appl . _ , vol . 97 , no . 3 , p. 032502 , 2010 . kim , v. tiberkevich , and a. n. slavin , `` generation linewidth of an auto - oscillator with a nonlinear frequency shift : spin - torque nano - oscillator , '' _ phys . 100 , p. 017207 , jan 2008 . b. georges , j. grollier , v. cros , a. fert , a. fukushima , h. kubota , k. yakushijin , s. yuasa , and k. ando , `` origin of the spectral linewidth in nonlinear spin - transfer oscillators based on mgo tunnel junctions , '' _ phys . b _ , vol . 80 , p. 060404 , aug 2009 . m. quinsat , d. gusakova , j. f. sierra , j. p. michel , d. houssameddine , b. delaet , m .- c . cyrille , u. ebels , b. dieny , l. d. buda - prejbeanu , j. a. katine , d. mauri , a. zeltser , m. prigent , j .- c . nallatamby , and r. sommet , `` amplitude and phase noise of magnetic tunnel junction oscillators , '' _ appl . _ , vol . 97 , no . 18 , p. 182507 , 2010 . j. f. sierra , m. quinsat , f. garcia - sanchez , u. ebels , i. joumard , a. s. jenkins , b. dieny , m .- c . cyrille , a. zeltser , and j. a. katine , `` influence of thermal fluctuations on the emission linewidth in mgo - based spin transfer oscillators , '' _ appl . _ , vol . 101 , no . 6 , p. 062407 , 2012 . kim , q. mistral , c. chappert , v. s. tiberkevich , and a. n. slavin , `` line shape distortion in a nonlinear auto - oscillator near generation threshold : application to spin - torque nano - oscillators , '' _ phys . 100 , p. 167201 , apr 2008 . o. j. lee , p. m. braganca , v. s. pribiag , d. c. ralph , and r. a. buhrman , `` quasilinear spin - torque nano - oscillator via enhanced negative feedback of power fluctuations , '' _ phys . b _ , vol . 88 , p. 224411 , dec 2013 . r. sharma , p. drrenfeld , e. iacocca , o. g. heinonen , j. kerman , and p. k. muduli , `` mode - hopping mechanism generating colored noise in a magnetic tunnel junction based spin torque oscillator , '' _ appl . 105 , no . 13 , p. 132404 , 2014 . a. eklund , s. bonetti , s. r. sani , s. m. mohseni , j. persson , s. chung , s. a. h. banuazizi , e. iacocca , m. @xmath31stling , j. @xmath30kerman , and b. g. malm , `` dependence of the colored frequency noise in spin torque oscillators on current and magnetic field , '' _ appl . 104 , no . 9 , p. 092405 , 2014 . w. h. rippard , m. r. pufall , s. kaka , t. j. silva , s. e. russek , and j. a. katine , `` injection locking and phase control of spin transfer nano - oscillators , '' _ phys . _ , vol . 95 , p. 067203 , aug 2005 . j. c. sankey , p. m. braganca , a. g. f. garcia , i. n. krivorotov , r. a. buhrman , and d. c. ralph , `` spin - transfer - driven ferromagnetic resonance of individual nanomagnets , '' _ phys . _ , vol . 96 , p. 227601 , jun 2006 . b. georges , j. grollier , m. darques , v. cros , c. deranlot , b. marcilhac , g. faini , and a. fert , `` coupling efficiency for phase locking of a spin transfer nano - oscillator to a microwave current , '' _ phys . _ , vol . 101 , p. 017201 , jul 2008 . a. dussaux , a. v. khvalkovskiy , j. grollier , v. cros , a. fukushima , m. konoto , h. kubota , k. yakushiji , s. yuasa , k. ando , and a. fert , `` phase locking of vortex based spin transfer oscillators to a microwave current , '' _ appl . _ , vol . 98 , no . 13 , p. 132506 , 2011 . r. lebrun , a. jenkins , a. dussaux , n. locatelli , s. tsunegi , e. grimaldi , h. kubota , p. bortolotti , k. yakushiji , j. grollier , a. fukushima , s. yuasa , and v. cros , `` understanding of phase noise squeezing under fractional synchronization of a nonlinear spin transfer vortex oscillator , '' _ phys . 115 , p. 017201 , jun 2015 . v. e. demidov , h. ulrichs , s. v. gurevich , s. o. demokritov , v. s. tiberkevich , a. n. slavin , a. zholud , and s. urazhdin , `` synchronization of spin hall nano - oscillators to external microwave signals , '' _ nat . _ , vol . 5 , p. 3179 , jan . 2014 . g. khalsa , m. d. stiles , and j. grollier , `` critical current and linewidth reduction in spin - torque nano - oscillators by delayed self - injection , '' _ appl . 106 , no . 24 , p. 242402 , 2015 . s. tsunegi , e. grimaldi , r. lebrun , h. kubota , a. s. jenkins , k. yakushiji , a. fukushima , p. bortolotti , j. grollier , s. yuasa , and v. cros , `` self - injection locking of a vortex spin torque oscillator by delayed feedback , '' _ arxiv:1509.05583v1 _ , 2015 . s. tamaru , k. yakushiji , b. wang , a. fukushima , and h. kubota , `` generation of highly stable 5 ghz microwave from a spin torque oscillator by phase locked loop referenced to a 80 mhz clock , '' in _ magnetics conference ( intermag ) , 2015 ieee _ , may 2015 , pp . 11 . p. villard , u. ebels , d. houssameddine , j. katine , d. mauri , b. delaet , p. vincent , m. c. cyrille , b. viala , j. p. michel , j. prouve , and f. badets , `` a ghz spintronic - based rf oscillator , '' _ ieee j. solid - state circuits _ , vol . 45 , no . 1 , pp . 214223 , 2010 . t. chen , p. drrenfeld , s. rodriguez , j. kerman , and a. rusu , `` a highly tunable microwave oscillator based on mtj sto technology , '' _ microw . _ , vol . 56 , no . 9 , pp . 20922095 , 2014 . k. muduli , o. g. heinonen , and j. kerman , `` bias dependence of perpendicular spin torque and of free - and fixed - layer eigenmodes in mgo - based nanopillars , '' _ phys . b _ , vol . 83 , no . 18 , p. 184410 , 2011 . s. mizukami , f. wu , a. sakuma , j. walowski , d. watanabe , t. kubota , x. zhang , h. naganuma , m. oogane , y. ando , and t. miyazaki , `` long - lived ultrafast spin precession in manganese alloys films with a large perpendicular magnetic anisotropy , '' _ phys . 106 , no . 11 , p. 117201 , 2011 . d. houssameddine , s. h. florez , j. a. katine , j. p. michel , u. ebels , o. mauri , d. ozatay , b. delaet , b. viala , l. folks , b. d. terris , and m. c. cyrille , `` spin transfer induced coherent microwave emission with large power from nanoscale mgo tunnel junctions , '' _ appl . _ , vol . 93 , no . 2 , p. 022505 , 2008 . k. mufuli , s. bonetti , y. zhou , j. @xmath30kerman , and o. g. heinonen , `` mgo - based spin torque oscillators with above 0.1 @xmath15w output power , '' in _ ieee international magnetics conference _ , 2009 . h. kubota , k. yakushiji , a. fukushima , s. tamaru , m. konoto , t. nozaki , s. ishibashi , t. saruya , , s. yuasa , t. taniguchi , h. arai , and h. imamura , `` spin - torque oscillator based on magnetic tunnel junction with a perpendicularly magnetized free layer and in - plane magnetized polarizer , '' _ appl . , vol . 6 , no . 10 , p. 103003 , 2013 . p. drrenfeld , s. m. mohseni , f. gerhard , m. ranjbar , f. qejvanaj , s. r. sani , s. chung , c. gould , l. w. molenkamp , and j. @xmath30kerman , `` low- current , narrow - linewidth microwave signal generation in single - layer - nimnsb based nano - contact spin torque oscillators , '' _ unpublished manuscript _ , 2015 . w. h. rippard , m. r. pufall , s. kaka , t. j. silva , and s. e. russek , `` current - driven microwave dynamics in magnetic point contacts as a function of applied field angle , '' _ phys . b _ , vol . 70 , no . 10 , p. 100406 , 2004 . q. mistral , j .- v . kim , t. devolder , p. crozat , c. chappert , j. a. katine , m. j. carey , and k. ito , `` current - driven microwave oscillations in current perpendicular - to - plane spin - valve nanopillars , '' _ appl . _ , vol . 88 , no . 19 , p. 192507 , 2006 . j. sinha , m. hayashi , y. k. takahashi , t. taniguchi , m. drapeko , s. mitani , and k. hono , `` large amplitude microwave emission and reduced nonlinear phase noise in co@xmath32fe(ge@xmath33ga@xmath33 ) heusler alloy based pseudo spin valve nanopillar , '' _ appl . _ , vol . 99 , no . 16 , p. 162508 , 2011 . t. seki , y. sakuraba , r. okura , and k. takanashi , `` high power radio frequency oscillation by spin transfer torque in a co@xmath34mnsi layer : experiment and macrospin simulation , '' _ j. appl 113 , no . 3 , p. 033907 , 2013 . t. seki , y. sakuraba , h. arai , m. ueda , r. okura , h. imamura , and k. takanashi , `` high power all - metal spin torque oscillator using full heusler co@xmath34(fe , mn)si , '' _ appl . 105 , no . 9 , p. 092406 , 2014 . t. yamamoto , t. seki , t. kubota , h. yako , and k. takanashi , `` zero - field spin torque oscillation in co@xmath34(fe , mn)si with a point contact geometry , '' _ appl . 106 , no . 9 , p. 092406 , 2015 . r. k. dumas , s. sani , s. m. mohseni , e. iacocca , y. pogoryelov , p. k. muduli , s. chung , p. drrenfeld , and j. kerman , `` recent advances in nanocontact spin - torque oscillators , '' _ ieee trans . _ , vol . 50 , no . 6 , 2014 . a. houshang , m. fazlali , s. r. sani , p. drrenfeld , e. iacocca , j. kerman , and r. k. dumas , `` effect of excitation fatigue on the synchronization of multiple nanocontact spin - torque oscillators , '' _ ieee magn . _ , vol . 5 , p. 3000404 , 2014 . a. a. tulapurkar , y. suzuki , a. fukushima , h. kubota , h. maehara , k. tsunekawa , d. d. djayaprawira , n. watanabe , and s. yuasa , `` spin - torque diode effect in magnetic tunnel junctions , '' _ nature _ , vol . 438 , pp . 339342 , 2005 . s. ishibashi , t. seki , t. nozaki , h. kubota , s. yakata , a. fukushima , s. yuasa , h. maehara , k. tsunekawa , d. d. djayaprawira , and y. suzuki , `` large diode sensitivity of cofeb / mgo / cofeb magnetic tunnel junctions , '' _ appl . , vol . 3 , no . 7 , , 2010 . c. wang , y .- t . cui , j. z. sun , j. a. katine , r. a. buhrman , and d. c. ralph , `` sensitivity of spin - torque diodes for frequency - tunable resonant microwave detection , '' _ j. appl . 106 , no . 5 , p. 053905 , 2009 . b. fang , m. carpentieri , x. hao , h. jiang , j. a. katine , i. n. krivorotov , b. ocker , j. langer , k. l. wang , b. zhang , b. azzerboni , p. k. amiri , g. finocchio , and z. zeng , `` giant spin - torque diode sensitivity at low input power in the absence of bias magnetic field , '' _ arxiv preprint arxiv:1410.4958 _ , 2014 . h. s. choi , s. y. kang , s. j. cho , i. y. oh , m. shin , h. park , c. jang , b. c. min , s. i. kim , s. y. park , and c. s. park , `` spin nano - oscillator - based wireless communication , '' _ sci . _ , vol . 4 , p. 5486 , 2014 . m. madami , s. bonetti , g. consolo , s. tacchi , g. carlotti , g. gubbiotti , f. b. mancoff , m. a. yar , and j. kerman , `` direct observation of a propagating spin wave induced by spin - transfer torque , '' _ nature nanotech . _ , vol . 6 , pp . 635638 , 2011 . m. pufall , w. rippard , g. csaba , d. nikonov , g. bourianoff , and w. porod , `` physical implementation of coherently - coupled oscillator networks , '' _ ieee j. explor . solid - state computat . devices circuits _ , vol . pp , no . 99 , pp . 18 , 2015 . p. m. braganca , b. a. gurney , b. a. wilson , j. a. katine , s. maat , and j. r. childress , `` nanoscale magnetic field detection using a spin torque oscillator , '' _ nanotechnology _ , vol . 21 , no . 23 , p. 235202 , 2010 . s. okamoto , n. kikuchi , m. furuta , o. kitakami , and t. shimatsu , `` microwave assisted magnetic recording technologies and related physics , '' _ j. phys . _ , vol . 48 , no . 35 , p. 353001 , 2015 . ( 2015 , july ) demonstration of a new technology for a multilayer magnetic recording to realize high - density hard disk drives . [ online ] . available : https://www.toshiba.co.jp/rdc/rd/detail_e/e1507_01.html r. sato , k. kudo , t. nagasawa , h. suto , and k. mizushima , `` simulations and experiments towards high - data - transfer - rate st - mram composed of a spin - torque oscillator , '' _ ieee trans . _ , vol . 48 , no . 5 , 2012 . i. firastrau , d. gusakova , d. houssameddine , u. ebels , m .- c . cyrille , b. delaet , b. dieny , o. redon , j .- c . toussaint , and l. d. buda - prejbeanu , `` modeling of the perpendicular polarizer - planar free layer spin torque oscillator : micromagnetic simulations , '' _ phys . rev . b _ , vol . 78 , no . 2 , p. 024437 , 2008 . t. chen , a. eklund , e. iaccoca , b. g. malm , s. rodriguez , j. kerman , and a. rusu , `` comprehensive and macrospin - based magnetic tunnel junction spin torque oscillator model - part i : analytical model of the mtj sto , '' _ ieee trans . on electron devices _ 62 , no . 3 , pp . 10371044 , 2015 . t. chen , a. eklund , e. iaccoca , b. g. malm , s. rodriguez , j. kerman , and a. rusu , `` comprehensive and macrospin - based magnetic tunnel junction spin torque oscillator model- part ii : verilog - a model implementation , '' _ ieee trans . on electron devices _ 62 , no . 3 , pp . 10451051 , 2015 . h. lim , s. ahn , m. kim , s. lee , and h. shin , `` a new circuit model for spin - torque oscillator including perpendicular torque of magnetic tunnel junction , '' _ adv . matter phys . 2013 , p. 169312 , 2013 . m. kim , h. lim , s. ahn , s. lee , and h. shin , `` advanced circuit - level model of magnetic tunnel junction - based spin - torque oscillator with perpendicular anisotropy field , '' _ j. semicond _ , vol . 13 , pp . 556561 , 2013 . t. chen , a. eklund , s. sani , s. rodriguez , b. g. malm , j. kerman , and a. rusu , `` integration of gmr - based spin torque oscillators and cmos circuitry , '' _ solid state electron . 111 , pp . 9199 , 2015 . ( 2010 , february ) `` fujitsu and university of toronto develop high - reliability read - method for spin - torque - transfer mram , next - generation non - volatile memory '' . [ online ] . available : http://www.fujitsu.com/global/about/resources/news/press-releases/2010/0210-03.html k. ikegami , h. noguchi , c. kamata , m. amano , k. abe , k. kushida , e. kitagawa , t. ochiai , n. shimomura , a. kawasumi , h. hara , j. ito , and s. fujita , `` a 4ns , 0.9 v write voltage embedded perpendicular stt - mram fabricated by mtj - last process , '' in _ proceedings of technical program-2014 international symposium on vlsi technology , systems and application ( vlsi - tsa)_.1em plus 0.5em minus 0.4emieee , 2014 , pp . n. d. rizzo , d. houssameddine , j. janesky , r. whig , f. b. mancoff , m. l. schneider , m. deherrera , j. j. sun , k. nagel , s. deshpande , h. j. chia , s. m. alam , t. andre , s. aggarwal , and j. m. slaughter , `` a fully functional 64 mb ddr3 st - mram built on 90 nm cmos technology , '' _ ieee trans . _ , vol . 49 , no . 7 , pp . 44414446 , 2013 . j. m. shaw , h. t. nembach , and t. j. silva , `` roughness induced magnetic inhomogeneity in co / ni multilayers : ferromagnetic resonance and switching properties in nanostructures , '' _ j. appl . 108 , no . 9 , p. 093922 , 2010 . j. wagner , r. wolf , n. joram , and f. ellinger , `` 0.59 ghz cmos variable gain amplifier with control linearization , low phase variations and self - matching , '' in _ european microwave integrated circuit conference ( eumic ) , 2014 9th _ , 2014 , pp . 100103 . tingsu chen ( s11 ) received an m.sc . degree in system - on - chip design from the kth royal institute of technology , sweden , 2011 . she is currently working toward a ph.d . degree at kth in the research area of modeling spin - torque oscillators and designing high - frequency circuits for spin - torque oscillator technology . randy k. dumas ( member ieee ) received his ph.d . in physics from the university of california , davis on the topic of reversal mechanisms in magnetic nanostructures in 2009 . from 20092012 , he was a postdoctoral research fellow at the university of gothenburg , where he is currently a permanent researcher . his research interests include the fabrication and characterization of magnetic nanostructures with a current focus on magnonics . anders eklund ( s13 ) received an m.sc . degree in engineering physics from kth royal institute of technology , sweden , in 2011 . he is currently working towards a ph.d . degree in physics at kth , experimentally investigating the frequency stability and mode structure of spin - torque oscillators by means of electrical characterization and synchrotron x - ray measurements . pranaba k. muduli ( member ieee ) received his ph.d . degree in physics from the humboldt university , berlin , germany in 2005 . he did postdoctoral research work at the university of north carolina , usa , the royal institute of technology , sweden , and the university of gothenburg , sweden . since july 2012 , he has been an assistant professor at the indian institute of technology delhi , india . he is also a researcher at the university of gothenburg , sweden . his current research interests include spintronics and nanomagnetism with an emphasis on spin - torque induced magnetization dynamics . afshin houshang ( s12 ) received his m.sc . in electroceramics engineering from shiraz university , shiraz , iran , 2010 . he is currently working towards a ph.d . degree in physics at the university of gothenburg in the area of synchronization phenomena in spin - torque and spin hall oscillators . ahmad a. awad ( member ieee ) received his ph.d . in physics from the autonomous university of madrid , spain , on high frequency magnetization and vortex dynamics in both magnetic and superconducting nanostructures in 2012 . since 2014 , he has been a postdoctoral research fellow at the university of gothenburg , sweden . his research interests include the characterization of magnetic nanostructures and spintronic devices with a focus on spin - torque induced magnetization dynamics . philipp drrenfeld ( s11 ) has a dipl .- ing . univ . degree ( 2010 ) from the university of wrzburg , germany . in 2015 , he received his ph.d . in physics from the university of gothenburg , sweden on the topic of spin - torque and spin hall nano - oscillators with single magnetic layers . his research interests include the nanofabrication and electrical characterization of spintronics and magnonics devices . b. gunnar malm ( m98 - sm10 ) was born in stockholm , sweden , in 1972 . he received an m.s . from uppsala university , sweden in 1997 and a phd in solid - state electronics from the royal institute of technology ( kth ) , stockholm in 2002 . he has been an associate professor at the school of ict , kth since 2011 . his recent work includes silicon photonics , silicon - carbide technology for extreme environments , and spintronics . he also serves on the kth sustainability council . ana rusu ( m92 ) received an m.sc . degree in electronics and telecommunications from the technical university of iasi , romania in 1983 , and a ph.d . degree in electronics from the technical university of cluj - napoca , romania , in 1998 . she has been with kth royal institute of technology , stockholm , sweden , since 2001 , where she is professor in electronic circuits for integrated systems . her research interests include low / ultralow - power high - performance cmos circuits and systems , sto - based systems , rf graphene , and high temperature sic circuits . johan @xmath30kerman ( m06 ) gained an ing . dipl . degree in 1994 from epfl , switzerland , an m.sc . in physics in 1996 from lth , sweden , and a ph.d . in materials physics in 2000 from kth royal institute of technology , stockholm . in 2008 , he was appointed full professor at the university of gothenburg and is a guest professor at kth royal institute of technology . he is also the founder of nanosc ab and nanosc instruments ab .

electromagnetic form factors ( emffs ) are the most fundamental observables that reveal the charge and magnetisation structures of the nucleon . a series of recent measurements of the emffs has renewed the understanding of the internal structure of the nucleon and has posed fundamental questions about its nonperturbative nature . the results of the ratio of the proton emffs , @xmath3 with the proton magnetic moment @xmath4 , obtained by measuring the transverse and longitudinal recoil proton polarisations @xcite , were found to decrease almost linearly as @xmath5 increases for @xmath6 . these results were in conflict with most previous measurements of the proton emffs from unpolarised electron - proton cross sections based on the rosenbluth separation method.these new experimental results have triggered subsequent theoretical and experimental works ( see , for example , recent reviews @xcite ) . this discrepancy is partially explained by the effects of two - photon exchange , which affect unpolarised electron - proton scattering at higher @xmath5 but have less influence on the polarisation measurements @xcite . moreover , the new experimental results of the proton emffs in a wider range of @xmath5 provided a whole new perspective on the internal quark - gluon structure of the nucleon . perturbative quantum chromodynamics ( pqcd ) with factorisation schemes @xcite predicts the different scalings of the dirac and pauli ffs , @xmath7 and @xmath8 : @xmath7 falls off as @xmath9 while @xmath8 decreases as @xmath10 , so that @xmath11 becomes flat at large @xmath5 . however , the experimental data show that the ratio @xmath11 increases with @xmath5 but @xmath12 becomes flat starting around @xmath13 . a similar discrepancy between the experimental data and pqcd was also found in the @xmath14 transition form factor @xcite even for higher @xmath5 . assuming isospin and charge symmetries , neglecting the strangeness in the nucleon , and using both the experimental data for the proton and neutron emffs , cates _ et al . _ @xcite have extracted the up and down emffs and have obtained remarkable results : the @xmath5 dependence of the up- and down - quark dirac ( @xmath15 ) and pauli ( @xmath16 ) form factors ( ffs ) are considerably different from each other . the down - quark dirac and pauli ffs are roughly proportional to @xmath9 but those of the up - quark fall off more gradually . moreover , while the ratios @xmath17 and @xmath18 ( @xmath19 is the anomalous magnetic moment ) are relatively constant above @xmath20 , they show a complicated behavior for lower @xmath5 regions . qattan and arrington @xcite elaborated the analysis of ref . @xcite , taking into account explicitly the effects of two - photon exchange and uncertainties on the proton form factor and the neutron magnetic ffs . they found that the ratio of the up - quark emffs ( @xmath21 ) has a roughly linear drop - off , while that of the down - quark emffs ( @xmath22 ) showed a completely different dependence on @xmath5 . as a result , the flavour - decomposed ffs behave in a different way from the proton emffs . it was also pointed out in ref . @xcite that theoretical descriptions of the flavour - decomposed ffs seem nontrivial @xcite . being motivated by these recent experimental works , we want to investigate the flavour structure of the nucleon emffs within the framework of the self - consistent su(2 ) and su(3 ) chiral quark - soliton models ( @xmath0qsms ) @xcite . the @xmath0qsm has described successfully various observables of the baryon octet and decuplet ( for reviews , see ) . in particular , the @xmath5 dependence of almost all form factors is well reproduced within the @xmath0qsm , so that the strange - quark emffs @xcite and the parity - violating ( pv ) asymmetries of polarised electron - proton scattering @xcite , which require nine different ffs ( six emffs and three axial - vector ffs ) with the same set of parameters , are in good agreement with experimental data . thus , it is worthwhile to examine the flavour structure of the nucleon emffs in detail . as mentioned , the nucleon emffs were already studied in the su(3 ) @xmath0qsm . however , praszaowicz _ et al . _ @xcite pointed out that the gell - mann - nishijima relation was not exactly fulfilled in the initial version of the @xmath0qsm and proposed the symmetry - conserving quantisation that makes the gell - mann - nishijima relation well satisfied . in this work , we present the results of the flavour - decomposed up- and down - quark emffs based on the su(3 ) @xmath0qsm with symmetry - conserving quantisation employed . we first show the dirac and pauli ffs of the nucleon and then examine the @xmath5 dependence of the up- and down - quark dirac and pauli ffs . the ratio of the flavour - decomposed dirac and pauli ffs will be discussed , compared with the recent experimental data @xcite . we also reexamine the results of the strange emffs , since there are new experimental data from pv polarised electron - nucleon scattering . in particular , the g0 collaboration recently measured the paritiy - violating asymmetries in the backward angle @xcite , which was first predicted in ref . @xcite . in addition to the flavour decomposed emffs of the nucleon , we also investigate the charge and magnetisation densities of the quark in a nucleon in the transverse plane . together with the new experimental data for the nucleon emffs , the nucleon gpds cast light on the concept of nucleon ffs @xcite . the present work is sketched as follows . in section 2 , we briefly review the general formalism of the emffs of the nucleon and its flavour decomposition and describe how to compute the emffs of the nucleon within the framework of the su(2 ) and su(3 ) @xmath0qsms . in the following sections we present the results and discuss their physical implications in the light of the recent experimental data : for sachs ffs in section 3 and for the dirac and pauli ffs in section 4 . in section 5 we also present the model results for the transverse charge and magnetic distributions of the quark inside both unpolarised and transversely polarised nucleons . the final section is devoted to the summary and the conclusions . the matrix element of a flavour vector current between the two nucleon states is expressed in terms of the flavour dirac and pauli ffs @xmath23 u_{n}(p,\,s ) , \label{eq : ff1 } \end{aligned}\ ] ] where @xmath24 represents the flavour vector current defined as @xmath25 @xmath0 denotes the flavour index , i.e. @xmath26 for the flavour decomposition . here , one has to bear in mind that @xmath27 is considered to be a unity flavour matrix . thus , the normalisation @xmath28 for @xmath29 applies only to the gell - mann matrices with @xmath30 and @xmath31 . the dirac spinor @xmath32 applies to the nucleon with mass @xmath33 , momentum @xmath34 and the third component of its spin @xmath35 . the square of the four momentum transfer is denoted by @xmath36 , with @xmath37 . the flavour dirac and pauli ffs can be combined to give the sachs ffs : @xmath38 in the breit frame , @xmath39 and @xmath40 are related to the time and space components of the flavour vector current , respectively : @xmath41 where @xmath42 are the pauli spin matrices . the @xmath43 is the corresponding spin state of the nucleon . in su(3 ) flavour the nucleon emffs are expressed in terms of the triplet and octet vector form factors @xmath44 while in flavour su(2 ) they are written as @xmath45 although the same notation is used for the form factors , it will always follow from the context which flavour case is being addressed . the matrix elements given in eq . ( [ eq : gm ] ) can be evaluated both in the su(2 ) and su(3 ) flavour @xmath0qsms . the model starts from the following low - energy effective partition function in euclidean space @xmath46 \cr & = & \int\mathcal{d}u \exp(-s_{\mathrm{eff}}[u])\ , , \label{eq : part } \end{aligned}\ ] ] where @xmath47 and @xmath48 denote the quark and pseudo - goldstone boson fields , respectively . after integrating over the quark fields , the effective chiral action @xmath49 is given by @xmath50 where @xmath51 represents the functional trace and @xmath52 the number of colours . the dirac @xmath53 operator , depending on the flavour space , is given by @xmath54 since , as isospin symmetry is assumed in this work , @xmath55 in su(2 ) and @xmath56 in su(3 ) , where @xmath57 the mass term @xmath58 containing the strange current quark mass @xmath2 will be treated as a perturbation . the single - quark hamiltonian @xmath59 is expressed as @xmath60 where @xmath61 stands for the chiral field for which we assume witten s embedding of the su(2 ) soliton into su(3 ) @xmath62 with the su(2 ) pion field @xmath63 as @xmath64 the integration over the pion field @xmath48 in eq . ( [ eq : part ] ) can be performed by the saddle - point approximation in the large @xmath52 limit due to the @xmath65 factor in eq . ( [ eq : echl ] ) . the su(2 ) pion field @xmath48 is written as the most symmetric hedgehog form @xmath66\ , , \label{eq : hedgehog}\ ] ] where @xmath67 is the radial profile function of the soliton . the @xmath0qsm nucleon state @xmath68 used in the computation of eqs . ( [ eq : ff1 ] ) and ( [ eq : gm ] ) is defined in terms of an ioffe - type current consisting of @xmath52 quarks : @xmath69 with the ioffe - type nucleon current @xmath70 defined as @xmath71 here , the matrix @xmath72 carries the hypercharge @xmath73 , isospin @xmath74 and spin @xmath75 quantum numbers of the baryon and the @xmath76 and @xmath77 denote the spin - flavour and colour indices , respectively . after minimizing the action in eq . ( [ eq : echl ] ) , we derive an equation of motion which is solved self - consistently with respect to the function @xmath67 in eq . ( [ eq : hedgehog ] ) . the corresponding unique solution @xmath78 is called the classical chiral soliton . the next step consists in quantising the classical soliton . this can be achieved by quantising the rotational and translational zero - modes of the soliton . the rotations and translations of the soliton are implemented by @xmath79 where @xmath80 denotes a time - dependent su(3 ) matrix , related to the orientation of the soliton in coordinate and flavour spaces , and @xmath81 stands for the time - dependent translation of the centre of mass of the soliton in coordinate space . the rotational velocity of the soliton @xmath82 is defined as @xmath83 treating @xmath82 and @xmath58 perturbatively with slowly rotating soliton and small @xmath58 considered , we find the collective hamiltonian , i.e , the hamiltonian in the collective coordinates of position of the centre of mass and the orientation of the soliton , which is given explicitly as @xmath84 in su(2 ) and as @xmath85 in su(3 ) . the @xmath86 is the classical mass of the state , the parameters @xmath87 are inertia parameters , @xmath73 is the hypercharge , @xmath88 is the pion - nucleon sigma term , the @xmath89s are the angular momentum operators and @xmath90 are the su(3 ) wigner @xmath91 functions . it is obvious that the strange quark in flavour su(3 ) leads to a more involved analysis , particularly to the symmetry breaking contributions . within the collective quantisation procedure the nucleon states given in eq . ( [ eq : ioffe ] ) will be mapped to collective rotational functions carrying the state quantum numbers . in flavour su(2 ) these functions are the eigenfunctions of the su(2 ) symmetrical hamiltonian , i.e. , the wigner @xmath91 functions given as @xmath92 in flavour su(3 ) the eigenfunctions of the su(3 ) symmetric part of the hamiltonian turn out to be the su(3 ) wigner @xmath91 functions @xmath93 on the contrary to the su(2 ) case , the nucleon state is no longer a pure octet state but is a mixed state with those in higher representations arising from flavour su(3 ) symmetry breaking , i.e. @xmath94 where @xmath95 and @xmath96 denote the mixing parameters . these parameters , as well as the @xmath97 , @xmath98 and @xmath99 in eq . ( [ eq : ham ] ) , may be found in refs . @xcite . a detailed formalism for the zero - mode quantisation can be found in refs . @xcite . in addiction , @xcite offers a detailed description as to how the form factors can be obtained numerically . we briefly summarise it here before we discuss the numerical results . the parameters existing in the model are the constituent quark mass @xmath100 , the current quark mass @xmath101 , the strange current quark mass @xmath102 , and the cutoff mass @xmath103 of the proper - time regularisation . however , not all of them are free parameters but can be fixed in the mesonic sector without any ambiguity . in fact , this is a merit of the @xmath0qsm in which mesons and baryons can be treated on an equal footing . for a given @xmath100 the regularisation cut - off parameter @xmath103 and the current quark mass @xmath101 in the lagrangian are fixed to the pion decay constant @xmath104 mev and the physical pion mass @xmath105 mev , respectively . the strange current quark mass is taken to be @xmath106 which approximately reproduces the kaon mass . though the constituent quark mass @xmath100 can be regarded as a free parameter , it is also more or less fixed . the experimental proton electric charge radius is best reproduced in the @xmath0qsm with the constituent quark mass @xmath107 mev . moreover , the value of 420 mev is known to yield the best fit to many baryonic observables @xcite . thus , all the numerical results in the present work are obtained with this value of @xmath100 . all the results presented in the following were computed completely within the model , in the same level of approximation , to keep consistency . in particular , magnetisation observables are presented not in terms of the physical nuclear magneton but , instead , in terms of the model nuclear magneton , i.e. defined as the model value for the nucleon mass , which , at this level of approximation used in this work , is @xmath108 we want to mention that the ratio between the model nuclear magneton and the physical one is the same as that between the value of @xmath33 in eq.([eq : m ] ) and the physical nucleon mass . to address the properties of the baryon octet implies immediately flavour structures of the su(3 ) baryons . however , it indicates simultaneously the question as to how accurate is the @xmath0qsm description of the strangeness content of the nucleon and its implications to the emff . such a question could easily be answered if one had precise experimental data on the strange emff . the present study may give some clues to the answer for that question in the light of the recent phenomenological data @xcite . the sachs em form factors @xcite are the most common form to encompass information about the electromagnetic structure of the nucleon . on the one hand , these form factors make it possible to express the cross section for elastic electron - proton scattering in the one - photon exchange approximation , without mixed terms ( @xmath109 ) in a form suitable for the separation of the electric and magnetic form factors . that is not the case when the cross section is expressed in terms of the dirac and pauli form factors ( [ eq : ff1 ] ) , where the mixed terms ( @xmath110 ) occur . even with the more recent polarisation transfer methods @xcite , the measured ratio between the longitudinal and transverse polarisation components is expressed in terms of the sachs form factors ratio @xmath111 . on the other hand , the sachs form factors have a merit that in the breit frame they may be apparently interpreted as the fourier transform of the charge and magnetisation distributions inside a nucleon . it comes from the fact that in the breit frame the proton does not exchange energy with the virtual photon with momentum @xmath112 . at a specific space - like @xmath113 invariant momentum transfer , the time and space components of the electromagnetic current , associated with the electric and magnetic form factors respectively , resemble the classical non - relativistic current density . hence the sachs em form factors are directly related to the charge and magnetisation distributions by the fourier transform . however , these relations are supposedly non - relativistic in nature due to the @xmath5 dependence of the breit - frame . both the preceding features of the sachs form factors are currently under scrutiny , as mentioned in introduction . discrepancies in the experimental results from the elastic @xmath114 cross section and polarisation transfer studies called for the inclusion of new aspects of elastic electron proton scattering , such as the two - photon exchange @xcite . the connection between form factors and densities , even apart from the non - relativistic limitation , has also been revised on general grounds @xcite . in the left panel of fig . [ fig:1 ] , the results of the ratio of the proton magnetic ff to the electric ff are depicted in comparison with the experimental data from the recoil polarisation experiments @xmath115 @xcite and the experiments with a polarised target @xmath116 @xcite . the su(2 ) results can describe the general tendency of the data very well , whereas those of su(3 ) seem slightly underestimated , as @xmath5 increase . the right panel of fig . [ fig:1 ] plots the results for the ratio @xmath117 , compared with the experimental data taken from @xmath118 @xcite and from @xmath119 @xcite and @xmath120 scatterings @xcite . we observe that the experimental data lie between the su(2 ) and su(3 ) results . the general tendency of the present results are in line with the experimental data : @xmath121 falls off slowly as @xmath5 increases , while @xmath117 increases systematically as a function of @xmath5 . as shown in the right panel of fig [ fig:1 ] , the su(3 ) results for the neutron are rather different from those in su(2 ) , the reason stemming , at least partially , from the strange quark contribution to the neutron electric ff . because of the embedding of the su(2 ) soliton into su(3 ) as shown in eq.([eq : embed ] ) , the contribution of the strange quark has the same asymptotic behavior of the nonstrange quarks . the effects due to different asymptotic tails were discussed in ref . @xcite in detail in the context of the strange vector ffs of the nucleon . thus , in a sense , a true answer may be found between the su(2 ) and the su(3 ) results . in order to decompose the proton emffs into the flavour ones , we need to compute the singlet vector form factors of the proton . then , we are able to express the flavour - decomposed emffs of the proton in terms of the singlet , triplet , and octet ffs of the proton : @xmath122 where we have suppressed the corresponding quark charge . the normalisations at @xmath123 for the proton obey @xmath124 , @xmath125 and @xmath126 . the flavour - decomposed magnetic moments are listed in table [ tab:1 ] in unit of the nuclear magneton . .the flavour - decomposed magnetic moments are defined as @xmath127 and are presented in unit of the model nuclear magneton @xmath128 . [ tab:1 ] [ cols="^,^,^,^",options="header " , ] we are now in a position to discuss the quark transverse charge densities inside both unpolarised and polarised nucleons . the traditional charge and magnetisation densities in the breit framework are defined ambiguously because of the lorentz contraction of the nucleon in its moving direction @xcite . to avoid this ambiguity one can define the quark charge densities in the transverse plane . then , they provide essential information on how the charges and magnetisations of the quarks are distributed in the transverse plane inside the nucleon . when the nucleon is unpolarised , the quark transverse charge density is defined as the two - dimensional fourier transform of the nucleon dirac ffs @xmath129 where @xmath130 denotes the impact parameter , i.e. the distance in the transverse plane to the place where the density is being probed , and @xmath131 is a cylindrical bessel function of order zero @xcite . note that the dirac ff at @xmath123 and the anomalous magnetic moment can be rederived from the transverse charge and magnetisation densities @xmath132 either for the nucleon or for each individual flavour . the anomalous magnetisation density in the transverse plane @xcite is defined as @xmath133 where @xmath134 is related to the pauli ff in the same way as @xmath135 to the dirac ff : @xmath136 .,title="fig:",scaledwidth=49.0% ] .,title="fig:",scaledwidth=49.0% ] .,title="fig:",scaledwidth=49.0% ] .,title="fig:",scaledwidth=49.0% ] in the upper panel of fig . [ fig : tcds ] , the transverse charge densities inside both a proton and a neutron are drawn . the results of the transverse charge density from the su(2 ) model is almost the same as that from the su(3 ) model for the proton . however , it is very interesting to observe that the tranverse charge density inside the neutron from the su(2 ) model is opposite to that of the su(3 ) one . as already found in figs . [ fig:1 ] and [ fig : f1f2 ] , the su(2 ) result is distinguished from the su(3 ) one , mainly due to the effects of the strange quark . these are in fact a surprising results , because the su(3 ) result interprets the inner structure of the neutron totally differently from the su(2 ) one : while the negative charge , which mainly come from the down and strange quarks inside a neutron , is centred on the neutron according to the su(3 ) @xmath0qsm , the su(2 ) model suggests that the positive one be located in the centre of the neutron . in ref . @xcite , the transverse charge density of the neutron was computed , based on the parametrisation of the experimental emffs , and was found to be negative in the centre of the neutron , which is in line with the present result from the su(3 ) model . to clarify this discrepancy between the su(2 ) and the su(3 ) models , it might be essential to know the strangeness content of the neutron . we will discuss later each contribution of a quark with different flavour to the transverse charge density inside the neutron more in detail . another interesting point in the transverse charge density inside a neutron is that it turns positive as @xmath130 increases . the reason will soon be clear when we discuss the flavour - decomposed transvese charge densities . the lower panel of fig . [ fig : tcds ] plots the transverse magnetisation densities inside both a proton and a neutron . the results from the su(2 ) model are similar to those from the su(3 ) model . as expected from their values of the anomalous magnetic moments , the transverse magnetisation densities inside a proton are positive but those inside a neutron turn out to be negative . we will soon observe that the up quark and the down quark are indeed magnetised oppositely each other , which explains the results of the transverse magnetisation densities inside a proton and a neutron . . ] in fig . [ fig : flvtcds ] , the transverse charge and magnetisation densities are depicted for each flavour , in the left panel and the right panel , respectively . the results of the transverse densities in fig . [ fig : flvtcds ] do not include the charges for each flavor . the charge densities for the up and the down quarks look similar to the proton one shown in fig . [ fig : tcds ] and the su(2 ) results generally larger in the center region but fall off faster than the su(3 ) ones . the strange quark case shows interesting features . while the charge density is found to be positive in the inner region , it becomes negative as @xmath130 increases . note that the down quark inside a nucleon is more magnetised than the up quark but was directed opposite to the up quark , which results in the negative larger value of the anomalous magnetic moment for the down quark than for the up quark ( see tab . [ tab:2 ] ) . the strange transverse magnetisation densities look very different from those for the up and down quarks : in the inner part of the nucleon , the strange quark is negatively magnetised . as @xmath130 increases , the strange magnetisation density turns positive . as a result , the strange anomalous magnetic moment turns out to be small but positive : @xmath137 ( see tab . [ tab:2 ] ) . as was discussed , the su(3 ) transverse charge density was very different from the su(2 ) one . we can understand the reason for it from the results of the flavour - decomposed transverse charge densities . the transverse charge densities inside a proton and a neutron can be respectively expressed in terms of the flavour - decomposed ones @xmath138 since the @xmath139 governs the transverse charge density inside a proton ( @xmath140 ) as shown in fig . [ fig : flvtcds ] , the @xmath141 has almost negligible effects on it . however , when it comes to the @xmath142 , @xmath143 and @xmath139 in eq . ( [ eq : pn ] ) are almost cancelled out each other , which results in a small amount of the negative density . in addition , @xmath141 contributes negatively to the @xmath142 , which finally leads to the negative value of the @xmath142 in the centred region , as shown in fig . [ fig : tcds ] . in the case of the su(2 ) model , the @xmath144 turns out to be larger than that from the su(3 ) model , so that the @xmath142 becomes positive but tiny . thus , the strange transverse charge density , though it is small , plays an essential role in explaining the negative value of the @xmath142 in the centre of the neutron within the framework of the @xmath0qsm . moreover , the strange transverse charge density turns positive around as @xmath130 increases . this explains partly the reason why the @xmath142 becomes negative at higher @xmath130 . axis . the lower panel depicts the corresponding transverse charge densities in the @xmath145 axis with @xmath146 fixed ( @xmath147 ) : the solid curve corresponds to the results of the transverse charge densities inside transversely polarised nucleons from the su(3 ) model , whereas the dashed curve to those from the su(2 ) model . the dotted and dash - dotted curves represent the su(2 ) and su(3 ) results for the transverse charge densities inside unpolarized nucleons , respectively . , title="fig:",scaledwidth=49.0% ] axis . the lower panel depicts the corresponding transverse charge densities in the @xmath145 axis with @xmath146 fixed ( @xmath147 ) : the solid curve corresponds to the results of the transverse charge densities inside transversely polarised nucleons from the su(3 ) model , whereas the dashed curve to those from the su(2 ) model . the dotted and dash - dotted curves represent the su(2 ) and su(3 ) results for the transverse charge densities inside unpolarized nucleons , respectively . , title="fig:",scaledwidth=49.0% ] axis . the lower panel depicts the corresponding transverse charge densities in the @xmath145 axis with @xmath146 fixed ( @xmath147 ) : the solid curve corresponds to the results of the transverse charge densities inside transversely polarised nucleons from the su(3 ) model , whereas the dashed curve to those from the su(2 ) model . the dotted and dash - dotted curves represent the su(2 ) and su(3 ) results for the transverse charge densities inside unpolarized nucleons , respectively . , title="fig:",scaledwidth=49.0% ] axis . the lower panel depicts the corresponding transverse charge densities in the @xmath145 axis with @xmath146 fixed ( @xmath147 ) : the solid curve corresponds to the results of the transverse charge densities inside transversely polarised nucleons from the su(3 ) model , whereas the dashed curve to those from the su(2 ) model . the dotted and dash - dotted curves represent the su(2 ) and su(3 ) results for the transverse charge densities inside unpolarized nucleons , respectively . , title="fig:",scaledwidth=49.0% ] when the nucleon is transversely polarised along the @xmath148 axis , which can be described by the transverse spin operator of the nucleon @xmath149 , the transverse charge density inside a transversely polarised nucleon is expressed @xcite as @xmath150 where @xmath151 is given in eq.([eq : rhom ] ) . the position vector @xmath152 from the centre of the nucleon in the transverse plane is denoted as @xmath153 . the @xmath148 axis is taken as the polarisation direction of the nucleon , i.e. @xmath154 . in the upper - left panel of fig . [ fig : tcdpol ] , we plot the transverse charge densities inside a transversely polarised proton . it is shown that the charge density for the transversely polarised proton is distorted in the negative @xmath145 direction . as discussed in refs . @xcite , the transverse polarisation of the nucleon in the @xmath148 axis induces the electric dipole moment along the negative @xmath145 direction , which is a well - known relativistic effect . in the case of the neutron , the situation is even more dramatic . as shown in the upper - right panel of fig . [ fig : tcds ] , the negative charge is located in the centre of the neutron , whereas the positive charge is surrounded . however , when the neutron is transversely polarised along the @xmath148 axis , the negative charge is shifted to the negative @xmath145 direction but the positive one is moved to the positive @xmath145 direction . this comes from the fact that the neutron anomalous magnetic moment is negative , which yields an induced electric dipole moment along the positive @xmath145 axis , as pointed out by ref . @xcite . it is very instructive to examine the transverse charge densities inside the transversely polarised nucleon for each flavour , since they reveal the inner structure of the nucleon more in detail . figure [ fig : flvtcdpol ] illustrates them . the up transverse charge density inside the transversely polarised nucleon , @xmath155 is shown to be shifted to the negative direction , while that for the down quark is more distorted upwards . it is natural , since the up and down quarks have positive and negative charges , respectively . however , it is remarkable to see that the down quark is influenced more strongly due to the transverse polarisation of the nucleon . the @xmath156 is even more interesting . as discussed previously , the strange anomalous magnetic moment is @xmath157 , which would induce the negative electric dipole moment along the negative @xmath158 . however , situation turns out to be more complicated . as shown in the right panel of fig . [ fig : flvtcdpol ] , the @xmath159 is shifted to the positive @xmath158 and turns negative starting from @xmath160 . in order to understand this surprising result , we need to reexamine the transverse magnetisation density for the strange quark , which has been presented in fig . [ fig : flvtcds ] . the strange magnetisation density is negative in the inner part of the nucleon and then it becomes positive from @xmath161 . thus , the electric dipole moment is correspondingly induced along the positive @xmath145 direction in the centred region , and then it becomes negative from @xmath162 , as drawn in the right panel of fig . [ fig : flvtcdpol ] . in the present work , we aimed at investigating the electromagnetic properties of the nucleon , based on the su(2 ) and su(3 ) chiral quark - soliton model with a symmetry - preserving quantisation employed . we considered the rotational @xmath1 corrections and the first - order @xmath2 corrections . we first presented the results of the ratio of the magnetic form factor to the electric form factor of the proton . it was shown that the results from the su(2 ) chiral quark - soliton model described the experimental data very well , whereas those of su(3 ) seemed slightly underestimated in higher @xmath5 . the general tendency of the present results were in greement with the experimental data . as for the neutron , the su(3 ) results turned out to be rather different from those in su(2 ) , which arose from the strange quark contribution to the neutron electric form factor . in particular , the neutron electric form factor is rather sensitive to the tail of the soliton . we then discussed that the up and the down electric form factors normalised by the dipole parametrisation were well reproduced in comparison with the data . as for the magnetic form factors , they deviate from the experimental data as @xmath5 increases but the general behaviour of the form factors are in line with the experimental data , which indicates that the @xmath5 dependence are well explained . we presented the prediction of the strange form factors normalised by the dipole form factor . the dirac and pauli form factors were predicted to be asymptotically proportional to @xmath9 and @xmath10 respectively in perturbative qcd . thus , @xmath163 and @xmath164 were measured experimentally to examine their @xmath5 dependence . we found that the present su(2 ) model explained well @xmath165 whereas the result from the su(3 ) model becomes underestimated in higher @xmath5 . both the su(2 ) and su(3 ) results for @xmath166 described the experimental data very well . on the other hand , the su(2 ) result for @xmath167 is in conflict with the data but that from the su(3 ) model is in agreement with the data except for the higher @xmath5 region . again this discrepancy can be understood by the sensitivity of the neutron electric form factor to the soliton tail . the results for the flavour - decomposed @xmath163 and @xmath164 were shown to be in general in good agreement with the experimental data . having performed the two - dimensional fourier transform of the nucleon electromagnetic form factors , we were able to produce the charge densities in the transverse plane inside a proton . as expected , both the su(2 ) and the su(3 ) transverse charge densities were positive in the proton . however , as for the neutron case , the result from the su(2 ) was opposite to that from the su(3 ) : the negative charge was located in the centre of the neutron while the positive one was distributed in outer part within the su(3 ) chiral quark - soliton model , it was other way around in the su(2 ) model . the explanation comes from the decomposed - flavour transverse charge densities in the su(3 ) model . in particular , the component of the strange quark played an essential role in spite of the smallness of its magnitude . since the up quark component mainly contributed to the transverse charge density inside a proton , the strange transverse charge density was almost negligible . on the other hand , the up and the down quark contributions were nearly cancelled out in such a way that the negative charge remained in the centre of the neutron with small magnitude . then the contribution of the strange quark came into play , so that the transverse charge densities inside a neutron finally became negative in the centre . when the proton was polarised along the positive @xmath148 direction , the corresponding transverse charge density was shifted to the negative @xmath145 direction , which indicated that the electric dipole moment was induced along the negative @xmath145 direction . it is just a well - known relativistic effect in electrodynamics . in the case of the neutron polarised along the @xmath148 axis , the negative charge was moved to the negative @xmath145 direction but the positive one was forced to the positive @xmath145 axis . it implies that the neutron anomalous magnetic moment is negative , which induces an electric dipole moment along the positive @xmath145 axis . we also decomposed the transverse charge densities inside the polarised nucleon for each flavour : the up transverse charge density for the nucleon transversely polarised along the positive @xmath148 axis was found to be shifted to the negative direction , while that of the down quark was more distorted upwards . since the up and down quarks have positive and negative charges , respectively , one can easily understand these features . however , the down quark was found to be affected more strongly due to the transverse polarisation of the nucleon . the strange charge density inside the transversely polarised nucleon was shifted to the positive @xmath158 and turned out to be negative in the outer region . this unexpected behavior of the strange charge density for the transversely polarised nucleon was explained in terms of the strange magnetisation density . since the transverse charge densities inside unpolarised and polarised nucleons pave the novel way for understanding the internal structure of the nucleon , it is interesting to investigate them for other baryons such as the @xmath168 isobar and hyperons . the transverse charge densities are directly connected to the generalized parton distributions of which the integrations over parton momentum fractions yield form factors and consequently the spatial distribution of partons in the transverse plane . moreover , the transverse charge densities for transition form factors provide a new aspect of understanding the inner structure of the baryons . for example , as ref . @xcite already studied , they exhibit explicitly multipole structures of the transitions in the transverse plane . thus , it is of great importance to examine the transverse charge densities for other baryons and for their transitions . corresponding investigations are under way . h .- ch.k . is grateful to r. woloshyn and p. navratil for their hospitality during his stay at triumf , where part of the present work was done . the work of h .- ch.k . was supported by basic science research program through the national research foundation of korea funded by the ministry of education , science and technology ( grant number : 2012r1a1a2001083 ) . 99 m. k. jones _ et al . _ [ jefferson lab hall a collaboration ] , phys . lett . * 84 * , 1398 ( 2000 ) [ nucl - ex/9910005 ] . o. gayou _ et al . _ [ jefferson lab hall a collaboration ] , phys . c * 64 * , 038202 ( 2001 ) . o. gayou _ et al . _ [ jefferson lab hall a collaboration ] , phys . lett . * 88 * , 092301 ( 2002 ) [ nucl - ex/0111010 ] . v. punjabi _ et al . _ [ jefferson lab hall a collaboration ] , phys . c * 71 * , 055202 ( 2005 ) [ erratum - ibid . c * 71 * , 069902 ( 2005 ) ] [ nucl - ex/0501018 ] . a. j. r. puckett , e. j. brash , m. k. jones , w. luo , m. meziane , l. pentchev , c. f. perdrisat and v. punjabi _ et al . _ , phys . rev . * 104 * , 242301 ( 2010 ) [ arxiv:1005.3419 [ nucl - ex ] ] . j. c. bernauer _ et al . _ [ a1 collaboration ] , phys . * 105 * , 242001 ( 2010 ) [ arxiv:1007.5076 [ nucl - ex ] ] . g. ron _ et al . _ [ jefferson lab hall a collaboration ] , phys . c * 84 * , 055204 ( 2011 ) [ arxiv:1103.5784 [ nucl - ex ] ] . x. zhan , k. allada , d. s. armstrong , j. arrington , w. bertozzi , w. boeglin , j. -p . chen and k. chirapatpimol _ et al . _ , phys . b * 705 * , 59 ( 2011 ) [ arxiv:1102.0318 [ nucl - ex ] ] . c. e. hyde - wright and k. de jager , ann . sci . * 54 * , 217 ( 2004 ) [ nucl - ex/0507001 ] . j. arrington , c. d. roberts and j. m. zanotti , j. phys . g * 34 * , s23 ( 2007 ) [ nucl - th/0611050 ] . c. f. perdrisat , v. punjabi and m. vanderhaeghen , prog . part . phys . * 59 * , 694 ( 2007 ) [ hep - ph/0612014 ] . m. vanderhaeghen and t. walcher , arxiv:1008.4225 [ hep - ph ] . j. arrington , k. de jager and c. f. perdrisat , j. phys . ser . * 299 * , 012002 ( 2011 ) [ arxiv:1102.2463 [ nucl - ex ] ] . p. a. m. guichon and m. vanderhaeghen , phys . lett . * 91 * , 142303 ( 2003 ) [ hep - ph/0306007 ] . g. blunden , w. melnitchouk and j. a. tjon , phys . * 91 * , 142304 ( 2003 ) [ nucl - th/0306076 ] . j. arrington , phys . c * 71 * , 015202 ( 2005 ) [ hep - ph/0408261 ] . j. arrington , w. melnitchouk and j. a. tjon , phys . c * 76 * , 035205 ( 2007 ) [ arxiv:0707.1861 [ nucl - ex ] ] . c. e. carlson and m. vanderhaeghen , ann . nucl . part . sci . * 57 * , 171 ( 2007 ) [ hep - ph/0701272 [ hep - ph ] ] . j. arrington , p. g. blunden and w. melnitchouk , prog . * 66 * , 782 ( 2011 ) [ arxiv:1105.0951 [ nucl - th ] ] . s. j. brodsky and g. r. farrar , phys . d * 11 * , 1309 ( 1975 ) . b. aubert _ et al . _ [ babar collaboration ] , phys . 052002 ( 2009 ) [ arxiv:0905.4778 [ hep - ex ] ] . s. uehara _ et al . _ [ belle collaboration ] , phys . d * 86 * , 092007 ( 2012 ) [ arxiv:1205.3249 [ hep - ex ] ] . g. d. cates , c. w. de jager , s. riordan and b. wojtsekhowski , phys . lett . * 106 * , 252003 ( 2011 ) [ arxiv:1103.1808 [ nucl - ex ] ] . i. a. qattan and j. arrington , phys . c * 86 * , 065210 ( 2012 ) [ arxiv:1209.0683 [ nucl - ex ] ] . g. eichmann , phys . d * 84 * , 014014 ( 2011 ) [ arxiv:1104.4505 [ hep - ph ] ] . m. rohrmoser , k. -s . choi and w. plessas , arxiv:1110.3665 [ hep - ph ] . i. c. cloet and g. a. miller , phys . c * 86 * , 015208 ( 2012 ) [ arxiv:1204.4422 [ nucl - th ] ] . j. o. gonzalez - hernandez , s. liuti , g. r. goldstein and k. kathuria , arxiv:1206.1876 [ hep - ph ] . d. diakonov , v. y. petrov and p. v. pobylitsa , nucl . b * 306 * , 809 ( 1988 ) . m. wakamatsu and h. yoshiki , nucl . phys . a * 524 * , 561 ( 1991 ) . a. blotz , d. diakonov , k. goeke , n. w. park , v. y. petrov and p. v. pobylitsa , nucl . a * 555 * , 765 ( 1993 ) . r. alkofer , h. reinhardt and h. weigel , phys . rept . * 265 * , 139 ( 1996 ) . c. v. christov , a. blotz , h. -ch . kim , p. pobylitsa , t. watabe , t. meissner , e. ruiz arriola and k. goeke , prog . part . phys . * 37 * , 91 ( 1996 ) [ hep - ph/9604441 ] . h. weigel , lect . notes phys . * 743 * , 1 ( 2008 ) . k. goeke , m. v. polyakov and m. vanderhaeghen , prog . phys . * 47 * , 401 ( 2001 ) [ hep - ph/0106012 ] . k. goeke , h. -ch . kim , a. silva and d. urbano , eur . j. a * 32 * , 393 ( 2007 ) [ hep - ph/0608262 ] . a. silva , h. -ch . kim , d. urbano and k. goeke , phys . d * 74 * , 054011 ( 2006 ) [ hep - ph/0601239 ] . h. -ch . kim , a. blotz , m. v. polyakov and k. goeke , phys . d * 53 * , 4013 ( 1996 ) [ hep - ph/9504363 ] . m. praszaowicz , t. watabe and k. goeke , nucl . a * 647 * , 49 ( 1999 ) [ hep - ph/9806431 ] . d. androic _ et al . _ [ g0 collaboration ] , phys . lett . * 104 * , 012001 ( 2010 ) [ arxiv:0909.5107 [ nucl - ex ] ] . x. -d . ji , j. phys . g * 24 * , 1181 ( 1998 ) [ hep - ph/9807358 ] . m. diehl , phys . * 388 * , 41 ( 2003 ) [ hep - ph/0307382 ] . a. v. belitsky and a. v. radyushkin , phys . rept . * 418 * , 1 ( 2005 ) [ hep - ph/0504030 ] . f. j. ernst , r. g. sachs , and k. c. wali , phys . 119 * , 1105 ( 1960 ) . r. g. sachs , phys . 126 * , 2256 ( 1962 ) . r. g. arnold , c. e. carlson , and f. gross , phys . c * 23 * , 363 ( 1981 ) . g. a. miller , phys . lett . * 99 * , 112001 ( 2007 ) [ arxiv:0705.2409 [ nucl - th ] ] . g. a. miller , ann . nucl . part . sci . * 60 * , 1 ( 2010 ) [ arxiv:1002.0355 [ nucl - th ] ] . s. venkat , j. arrington , g. a. miller and x. zhan , phys . c * 83 * , 015203 ( 2011 ) [ arxiv:1010.3629 [ nucl - th ] ] . b. d. milbrath _ et al . _ [ bates fpp collaboration ] , phys . rev . lett . * 80 * ( 1998 ) 452 [ erratum - ibid . * 82 * ( 1999 ) 2221 ] [ nucl - ex/9712006 ] . t. pospischil _ et al . _ [ a1 collaboration ] , eur . j. a * 12 * , 125 ( 2001 ) . g. maclachlan , a. aghalarian , a. ahmidouch , b. d. anderson , r. asaturian , o. baker , a. r. baldwin and d. barkhuff _ et al . _ , nucl . a * 764 * , 261 ( 2006 ) . m. paolone , s. p. malace , s. strauch , i. albayrak , j. arrington , b. l. berman , e. j. brash and w. j. briscoe _ et al . _ , phys . lett . * 105 * , 072001 ( 2010 ) [ arxiv:1002.2188 [ nucl - ex ] ] . a. j. r. puckett , e. j. brash , o. gayou , m. k. jones , l. pentchev , c. f. perdrisat , v. punjabi and k. a. aniol _ et al . _ , phys . c * 85 * , 045203 ( 2012 ) [ arxiv:1102.5737 [ nucl - ex ] ] . m. k. jones _ et al . _ [ resonance spin structure collaboration ] , phys . c * 74 * , 035201 ( 2006 ) [ nucl - ex/0606015 ] . c. b. crawford , a. sindile , t. akdogan , r. alarcon , w. bertozzi , e. booth , t. botto and j. calarco _ et al . _ , phys . lett . * 98 * , 052301 ( 2007 ) [ nucl - ex/0609007 ] . i. passchier , r. alarcon , t. s. bauer , d. boersma , j. f. j. van den brand , l. d. van buuren , h. j. bulten and m. ferro - luzzi _ et al . _ , phys . * 82 * , 4988 ( 1999 ) [ nucl - ex/9907012 ] . h. zhu _ et al . _ [ jefferson lab e93 - 026 collaboration ] , phys . * 87 * , 081801 ( 2001 ) [ nucl - ex/0105001 ] . g. warren _ [ jefferson lab e93 - 026 collaboration ] , phys . rev . * 92 * , 042301 ( 2004 ) [ nucl - ex/0308021 ] . e. geis _ et al . _ [ blast collaboration ] , phys . * 101 * , 042501 ( 2008 ) [ arxiv:0803.3827 [ nucl - ex ] ] . c. herberg , m. ostrick , h. g. andresen , j. r. m. annand , k. aulenbacher , j. becker , p. drescher and d. eyl _ et al . _ , eur . j. a * 5 * , 131 ( 1999 ) . d. i. glazier _ et al . _ [ a1 collaboration ] , eur . phys . j. a * 24 * , 101 ( 2005 ) [ nucl - ex/0410026 ] . b. plaster _ et al . _ [ jefferson lab e93 - 038 collaboration ] , phys . rev . c * 73 * , 025205 ( 2006 ) [ nucl - ex/0511025 ] . j. bermuth , p. merle , c. carasco , d. baumann , d. bohm , d. bosnar , m. ding and m. o. distler _ et al . _ , phys . b * 564 * , 199 ( 2003 ) [ nucl - ex/0303015 ] . a. silva , h. -ch . kim and k. goeke , phys . d * 65 * , 014016 ( 2002 ) [ erratum - ibid . d * 66 * , 039902 ( 2002 ) ] [ hep - ph/0107185 ] . supplemental material from @xcite at + http://link.aps.org/supplemental/10.1103/physrevc.86.065210

we present a method for the numerical calculation of solutions to general hyperbolic partial differential equations ( pdes ) over the sphere that combines several existing techniques in novel ways . our motivating interest is the analysis of gravitational radiation in a neighbourhood of infinity via friedrich s conformal field equations @xcite , where it is convenient to regard space - time as the product of a @xmath1 dimensional lorentzian manifold and a euclidean @xmath1-sphere . the main difficulty for the numerical solution of problems in this geometric setting stems from the fact that @xmath2 can not be covered by a single chart . hence the coordinate description of fields inevitably breaks down . spectral methods avoid most of the problems caused by this as fields on @xmath2 are expressed as a sum of functions that form a well defined bases on the sphere . assuming that the properties of the functions in the sum are well understood , it is possible to avoid working directly with coordinate expressions that suffer from coordinate singularities . as an example , we mention the recent work of @xcite in the context of kerr space - time shows that indeed issues such as coordinate singularities and instability are avoided in such an approach , leading to accurate evolutions of dynamical equations . as a consequence we use a spectral method in what follows . to implement our spectral method we choose to work with spin weighted spherical harmonics ( swshs ) and the associated @xmath3 and @xmath4 operators @xcite . swshs are a generalisation of scalar , vector and tensor harmonics on the sphere @xcite . swshs form an orthonormal basis for @xmath5 @xcite upon which the differential operators @xmath3 and @xmath4 act by raising and lowering spin - weight , respectively . in common with all spectral methods , this reduces the action of differential operators to algebraic manipulations . this is a property which we exploit to reduce pdes to systems of coupled odes . in order to apply swshs to a spectral evolution scheme it is necessary to decompose arbitrary fields on @xmath2 into a sum of swshs . many methods to do this have been proposed , since such decompositions are important for the analysis of data over the sphere , e.g.@xcite . we choose huffenberger and wandelt s method @xcite ( which is a modification of @xcite and @xcite ) for three reasons . firstly , the algorithm is theoretically exact if a minimum number of grid points are used . this is in contrast to several alternatives that are asymptotically exact in the limit of increasing numbers of grid points @xcite . secondly , the method can be applied simultaneously to functions of different spin weights . in our desired application we will be working with several functions each with different spin weights @xcite . as a consequence huffenberger and wandelt s method has reduced computational effort in comparison to a method which operates on each spin weighted function separately . thirdly , huffenberger and wandelt s method , via a clever mapping of the sphere into the @xmath1-torus allows for fast fourier transformations to be used , see also @xcite . this is in contrast to spectral methods adapted to the non - periodic coordinate on @xmath2 , e.g. @xcite . in order to calculate the values of swshs over @xmath2 huffenberger and wandelt used a formula relating swshs to reduced wigner @xmath6-functions evaluated at @xmath7 . they implemented the calculation of these @xmath6-functions via the three term linear recursive relations given by risbo and trapani and navaza @xcite . trapani and navaza s scheme is both faster and more accurate than risbo s but eventually becomes unstable . to cope with this use a non - linear scheme that is equivalent to the scheme proposed by przeau and reinecke @xcite . we build a hybrid linear / non - linear recursion that avoids the numerical problems that both linear and non - linear recursive schemes suffer from , see sections [ sec : algodelcalc ] and [ sec : wdelstab ] . a similar hybrid scheme has been proposed by luscombe and luban @xcite for the calculation of @xmath8 and @xmath9 symbols . to the best of the authors knowledge neither the use of a hybrid scheme for the calculation of reduced wigner @xmath6-functions at @xmath7 nor the use of this method in swsh decompositions of functions over @xmath2 has been used before . the @xmath0 maxwell equations on a deformed sphere , section [ sec : maxcnstrct ] , have non - constant coefficients . we will therefore need to perform a swsh decomposition of products of swshs . this requires the calculation of clebsch - gordan coefficients . to do this we use a hybrid linear / non - linear recursion for @xmath8 symbols originally presented by luscome and luban s method @xcite to schulten and gordan s linear scheme @xcite . as before , to the best of the author s knowledge this is the first time a hybrid linear / non - linear scheme has been used for the calculation of clebsch - gordan coefficients . an alternative to the above explicit decomposition is to use spectral transformations directly to perform decompositions of products of swshs @xcite ( the pseudo - spectral approach ) . we use this approach to check for accuracy . it is our goal to provide a self - contained presentation , with consistent conventions , that may be readily adapted to general non - linear hyperbolic pdes using the outlined spectral method . wherever possible we present results for both integer and half - integer spin . this will provide a foundation for future extension to the half - integer spin case . this paper is structured as follows : in [ sec : geom ] we provide the geometric background appropriate for formulation of problems later in the work , in particular we review the construction of frames adapted for use with the @xmath3-formalism . in [ sec : weightedquantities ] we explicitly show how an arbitrary smooth tensor field may be represented in terms of spin - weighted quantities on @xmath2 . in [ sec : introcgw3j ] we introduce and review the properties of swshs . we describe in detail how products of swshs may be decomposed and give details about the symmetry properties of clebsch - gordan coefficients ( wigner @xmath8-symbols ) appropriate for numerical use . in [ sec : specxform ] we briefly describe the spherical harmonic transformation of @xcite . in [ sec : cgw3jnumproc ] we describe how we compute clebsch - gordan coefficients ( wigner @xmath8-symbols ) numerically . in [ sec : errorstabilityspeed ] we demonstrate the property of spectral convergence for smooth test functions . in [ sec : wdelstab ] we demonstrate the instability of the recursive calculation of @xmath10 due to @xcite and discuss how our hybrid scheme avoids this . in [ sec : psandfscmp ] we contrast the pseudo - spectral and spectral approaches to the decomposition of products of swshs . in [ sec : advprob ] we construct the tensor advection equation in the @xmath3 formalism , thus showing how the standard ivp for the scalar and vector advection equation on @xmath2 may be formulated ; in [ sec : advnum ] we numerically solve this problem using our spectral method for test fields that lead to temporally periodic solutions . exploiting this periodicity by comparing solutions at integer multiples of one period ( i.e. stroboscopically ) yields a method for performing convergence tests . in [ sec : maxcnstrct ] we construct the ivp for the @xmath0 maxwell equation where the spatial geometry is conformally related to @xmath2 . in [ sec : nummax ] we numerically solve the problem , comparing spectral with pseudo - spectral methods . it is often useful to think of the manifold @xmath11as the submanifold of @xmath12 given by @xmath13 . the _ euler coordinates _ of @xmath14 can then be represented by @xmath15 where @xmath16 and @xmath17 . clearly , these coordinates break down at @xmath18 and @xmath19 . for later convenience , we introduce coordinates @xmath20 ( which are also referred to as euler coordinates ) by @xmath21 the set of complex unitary @xmath22-matrices with unit determinant @xmath23 endowed with the natural smooth manifold structure is diffeomorphic to @xmath14 @xmath24 since the latter is a lie group , we can use the map @xmath25 to endow @xmath14 with a lie group structure . in the following , it is therefore not always necessary to distinguish between @xmath14 with @xmath23 . any lie group is parallelizable , and a smooth global frame on @xmath23 can be constructed as follows . we define left and right translation maps @xmath26 on any lie group , the maps @xmath27 and @xmath28 are automorphisms for each element @xmath29 . now , choose a basis of the tangent space at the unit element @xmath30 ( i.e. a basis of the lie algebra ) @xmath31 and define , for @xmath32 , @xmath33 clearly , @xmath34 is a smooth global frame on @xmath11which is invariant under left translations while the frame @xmath35 is invariant under right translations . these fields have the following representation with respect to the euler parametrization @xmath36 one can show by direct lie group arguments ( or by using the coordinate representations of the fields ) that @xmath37=y_3,\quad[y_2,y_3]=y_1,\quad[y_3,y_1]=y_2,\ ] ] similarly for the right - invariant fields , and @xmath38=0,\quad\forall k , l=1,2,3.\ ] ] for later convenience , we define @xmath39 which , as we shall see later , are closely related to the @xmath3-operators defined in @xcite . we have @xmath40=2 { \mathrm{i}}y_3.\ ] ] the _ hopf map _ @xmath41 can be represented as @xmath42 here we again consider @xmath11as being embedded into @xmath12 by @xmath13 , and the manifold @xmath2is thought of being given by @xmath43 in @xmath44 . when we introduce standard polar coordinates on @xmath2 , namely @xmath45 then @xmath19 obtains the particularly simple representation @xmath46 in particular , it becomes obvious that the push - forward of @xmath47 to @xmath2along @xmath19 vanishes . indeed , @xmath14 is the principal bundle over @xmath48 with structure group @xmath49 generated by @xmath47 ( whose integral curves are closed ) and projection map @xmath19 ; this is the _ hopf bundle_. the hopf bundle can be identified with the bundle of orthonormal frames on @xmath48 with respect to any smooth riemannian metric . an explicit construction in terms of the coordinates above can be done as follows . let @xmath50 be an open subset of @xmath48 ; we assume that the poles @xmath51 are outside of @xmath50 so that the representation of the hopf map given by eq . ( [ eq : localhopf ] ) is well - defined and the euler coordinates cover @xmath52 . if we restrict to sufficiently small open subsets this is no loss of generality since for any sufficiently small choice of the open set @xmath50 we can always introduce the coordinates such that the poles are not in @xmath50 . let @xmath53 be a smooth orthonormal frame on @xmath50 and define the corresponding complex field @xmath54 we consider the action @xmath55 of @xmath56which is defined pointwise on @xmath50 , i.e. the group parameter @xmath57 is a smooth function on @xmath50 . any specification of the function @xmath58 therefore yields another smooth orthonormal frame on @xmath50 and can hence be interpreted as the smooth local section @xmath59 in the bundle of orthonormal frames , or equivalently , in the hopf bundle . doing this for every open subset @xmath50 of @xmath48 ( introducing coordinates so that the poles are not in @xmath50 as above ) , the full bundle of orthonormal frames can be recovered and can therefore be identified with the hopf bundle . at every point @xmath60 of @xmath50 , the fibre @xmath61 is the set of all orthonormal bases of @xmath62 . let @xmath63 be the dual coframe of @xmath64 and @xmath65 dual to @xmath66 . then , the above action of @xmath49 implies @xmath67 now let an arbitrary smooth tensor field @xmath68 of type @xmath69 for integers @xmath70 be given on @xmath50 , so that the function @xmath71 is defined by @xmath72 possibly after changing the order of the arguments of @xmath68 . in principle , @xmath73 is a function on @xmath74 . but under rotations of the frame , it gives rise to a unique function on @xmath52 which changes along the fiber according to the transformation of the frame . this function on @xmath14 is denoted by the same symbol @xmath73 for simplicity . in particular , its dependence on the fiber coordinate @xmath57 is given by @xmath75 where we consider @xmath76 as independent of @xmath57 ( because it is defined with respect to the reference frame @xmath66 ) . we get @xmath77 where @xmath78 is the _ spin - weight _ introduced in @xcite . hence , eq . ( [ eq : ethbracket ] ) becomes @xmath79(\nu)=-2 s \nu.\ ] ] in summary , every quantity @xmath73 on @xmath74 of spin - weight @xmath78 can be lifted to a smooth function @xmath80 on @xmath81 ( which we denote by the same symbol from now ) . vice versa , every such function on @xmath52 pulls back to a function with spin - weight @xmath78 on @xmath50 along a smooth section over @xmath50 . in the following , we will therefore often not distinguish between a function with spin - weight @xmath78 on @xmath2and the corresponding function on @xmath11 . in the case of the @xmath1-sphere with the standard round unit metric , we often consider the reference frame @xmath82 and choose @xmath50 as the set of all points on @xmath48 without the two poles @xmath51 . as the smooth local section , we choose @xmath83 . comparing this with the coordinate expressions above , we see that @xmath84 therefore , if @xmath73 is a function on @xmath50 with spin - weight @xmath78 and @xmath85 ( here , we exceptionally use two different symbols @xmath73 and @xmath85 ) the corresponding function on @xmath52 , then @xmath86 under all these conditions , it makes sense therefore to simplify the notation and write @xmath87 for a function on @xmath48 with spin - weight @xmath78 . in the same way , we obtain @xmath88 in application of the fourier - galerkin ( spectral ) method to the solution of pdes , products of spin - weighted spherical harmonics ( swsh ) will be encountered . this motivates the exploration of a convenient method of treating these product terms which will result in the appearance of the clebsch - gordan ( cg ) series , calculation of which will be facilitated by the relation of coefficients in this series to the wigner @xmath8-symbols . we proceed by first stating coordinate expressions for the well - known wigner @xmath89 matrices that form a basis for @xmath5 , which will allow for the use of commonly encountered identities from the treatment of angular momentum in quantum mechanics @xcite . the euler parametrization of a rotation can be written in terms of the euler coordinates @xmath90 , @xmath57 , @xmath91 introduced above . according to @xcite we have @xmath92 where @xmath93 ( or @xmath94 for spinorial quantities ) , @xmath95 , @xmath96 , @xmath97 . the quantity @xmath98 is the reduced wigner matrix element and satisfies @xmath99 together with the indicial symmetry @xmath100 . upon introduction of @xmath101 on @xmath2 as above we introduce , we assume this choice has been made henceforth , unless otherwise specified see [ sec : geom ] . ] the spin - weighted spherical harmonics as @xmath102 this fixes our convention to agree with @xcite . from eq . and eq . we immediately observe a useful property the swsh possess under complex conjugation : @xmath103 for later convenience , we compare ( see also eq . ( 4.15.122 ) ) of @xcite ) the algebraic action of the differential operators @xmath104 ( cf . and eq . ) and their explicit coordinatizations on @xmath105 : @xmath106\\\label{eq : ethraco } = & -\sqrt{(l - s)(l+s+1 ) } { } _ { s+1}y_{lm}(\vartheta,\varphi)\\\nonumber { \eth'}\,{}_sy_{lm}(\vartheta,\varphi ) = & ( \sin\vartheta)^{-s}\left(\partial_\vartheta + { \mathrm{i}}\csc\vartheta\partial_\varphi \right)\left[(\sin\vartheta)^s\vartheta\,{}_sy_{lm}(\vartheta,\varphi)\right]\\\label{eq : ethloco } = & \sqrt{(l+s)(l - s+1 ) } { } _ { s-1}y_{lm}(\vartheta,\varphi)\\ \delta_{\mathbb{s}^2}\,{}_sy_{lm}(\vartheta,\varphi ) = & \frac{1}{2}\left(\eth\eth'+\eth'\eth \right)\left[{}_sy_{lm}(\vartheta,\varphi)\right ] = \left(s^2-l(l+1)\right)\,{}_sy_{lm}(\vartheta,\varphi)\end{aligned}\ ] ] for later reference , we also restate the orthonormality relation : @xmath107 which is directly inherited from the properties of the @xmath89-matrices . observe that in eq . orthonormality holds for functions of the same spin - weight . we now describe a closed - sum decomposition for products such as : @xmath108 which together with the action of the @xmath104 operators in eq . and eq . will form the basis of our spectral scheme . the decomposition we seek is the so - called clebsch - gordan series which in bra - ket notation is given by @xcite : @xmath109 where @xmath110 . note that each clebsch - gordan coefficient in the series is real , i.e. @xmath111 . define : @xmath112 equation together with eq . thus provides us with the following decomposition of eq . : @xmath113 where @xmath114 . hence the product of two spin - weighted spherical harmonics may be decomposed into a finite linear combination of spin - weighted spherical harmonics with spin - weight equal to the sum of the original two spin - weights . spectral decomposition of evolution equations will also require the following identity : @xmath115 which may be obtained using eq . and eq . . in the interest of efficient numerical calculations utilising the relation of the clebsch - gordan coefficients to the wigner @xmath8-symbols is prudent due to the convenient symmetry properties the latter possess @xcite . we have @xmath116 the non - negative quantities @xmath117 are known as _ angular momentum numbers _ and may be integral or half - integral . the quantities @xmath118 are called the _ projective quantum numbers _ and are given by @xmath119 where @xmath120 . three further constraints are placed on the @xmath121 and @xmath122 : 1 . @xmath123 ; 2 . @xmath124 ; 3 . the _ triangle condition _ : @xmath125 where @xmath126 is any permutation of @xmath127 ; in the event these constraints fail to be satisfied the @xmath8-symbol is set to @xmath128 . the following symmetries will also be of use later 1 . invariance under permutation of columns @xmath129 2 . invariance under spatial inflection @xmath130 3 . _ regge symmetries _ @xmath131 from the symmetries of the @xmath8-symbols we recover the following symmetries on the @xmath132 of eq . that we will make use of later : @xmath133 when performing a decomposition such as in eq . numerically , it can be convenient to embed the above symmetries directly into the summation process . furthermore , there also exist efficient storage schemes for @xmath8-symbols , utilizing symmetries such as ( si - siii ) @xcite this allows for precalculation of all required @xmath8-symbols and evaluating eq . in this manner may be more efficient under certain circumstances . in this section we briefly describe the numerical implementation of huffenberger and wandelt s spectral algorithm @xcite that will allow for the decomposition of an integer spin - weighted function @xmath134 in terms of the swsh of eq . . numerical calculations must be limited to a finite grid , hence the decomposition must be truncated at a maximal harmonic ( band - limit ) @xmath135 . in terms of this band limit the algorithm has @xmath136 complexity which is achieved by exploiting a smooth periodic extension of the data to the @xmath1-torus so that existing fast fourier transform ( fft ) methods can be used . consider a function @xmath137 . by the peter - weyl theorem for compact groups @xcite we have @xmath138 fixing the fibre coordinate as @xmath139 and performing the map eq . we then define @xmath140 and arrive at the standard expansion @xmath141 henceforth , we work with the band - limited expression : @xmath142 where it is assumed that the function being decomposed may be completely expressed by a finite linear combination of the basis functions . we now describe the algorithm for evaluation of the forward transform @xmath143 . as a first step introduce the notation @xmath144 which allows for the rewriting of eq . as @xcite @xmath145 following from a factoring of rotations @xcite . the details of how the @xmath146 elements are calculated together with their symmetry properties are given in [ sec : algodelcalc ] . define the functional : @xmath147:=\int^{2\pi}_0\int^\pi_0 e^{-{\mathrm{i}}m\vartheta}e^{-{\mathrm{i}}n\varphi}{}_s f(\vartheta,\varphi ) \sin\vartheta \,d\vartheta d\varphi.\ ] ] equation together with eqs . ( [ eq : funcs2sphexpinf]-[eq : imnquadpr ] ) lead to @xmath148 we now wish to evaluate the expression for @xmath149 . this may be done exactly , by extension of the function @xmath150 to the @xmath1-torus @xmath151 , which will permit the application of the standard @xmath1d fourier - transform . although this requires computation of points outside the domain of interest , the corresponding increase in the speed of performing the calculation ( for large @xmath135 ) and favourable ( spectral)-convergence offered by this method compensates for the increased computational effort . define the extended function @xmath152 where @xmath153 now takes values in @xmath154 . clearly this does not change the value of @xmath149 because extension of the function leaves its value unchanged within the domain of integration as defined in eq . . the periodic extension is chosen by forming a linear combination of @xmath155 and examining symmetries using the defining relations of eq . and eq . . as the periodically extended function @xmath156 now possesses @xmath157 periodicity in both arguments , it may be written as the two dimensional , band - limited fourier sum @xmath158 substitution of this equation into eq . yields : @xmath159 \left[\int_{0}^{\pi}\exp\left({\mathrm{i}}(k - m')\vartheta\right)\sin\vartheta\,d\vartheta\right],\\ & = 2\pi\sum_{k =- l}^{l}f_{km } w(k - m ' ) , \label{eq : weightconvo } \end{aligned}\ ] ] where @xmath160 for @xmath161 is given by @xmath162 equation shows that @xmath163 is proportional to a discrete convolution in spectral space . by the convolution theorem , this implies that we may consider instead the inverse transform of @xmath160 which maps the function back to its spatial representation @xmath164 . performing point - wise multiplication with @xmath165 and transforming the result will yield @xmath163 . if the desired number of samples of the function @xmath166 over @xmath153 and @xmath167 on @xmath2 is to be @xmath168 and @xmath169 respectively , then for the number of samples for the extended function @xmath170 we take to be @xmath171 and @xmath169 . the spatial sampling intervals are given by @xmath172 and @xmath173 . note that in order to satisfy the nyquist condition , we must take @xmath174 and @xmath175 , where @xmath135 is the harmonic that the function @xmath176 is band - limited to . with the stated sampling , the quadrature weights may be written as @xmath177 upon performing a linear phase shift in eq . we finally arrive at the expression @xmath178 which may be evaluated using a @xmath1-dimensional fft . we note that if many transformations with the same band - limits are to be performed , the weights can be pre - calculated . overall , the complexity of the outlined algorithm is @xmath136 . two further linear improvements in execution speed are possible . the first reduces the total computation time of @xmath179 by a factor of @xmath1 . equation together with the symmetries of @xmath180 described in [ sec : algodelcalc ] allows for : @xmath181 where @xmath182 a second improvement ( also by a factor of @xmath1 ) is possible if the function @xmath183 being analysed is real . here the speedup is due to the fft , where real input results in hermitian output . we now describe the algorithm for evaluation of the backward ( inverse ) transform @xmath184 . the backward spherical harmonic transform maps the expansion coefficients @xmath185 , for @xmath186 , to a function on @xmath2 . because we are working with band - limited functions we can , at least in theory , perfectly reconstruct the original function . to this end , eq . must be evaluated . as the inverse transform does not contain integrals , issues of quadrature accuracy do not arise . define the functional : @xmath187:= { \mathrm{i}}^{s - n } \sum_{l=|s|}^l \sqrt{\frac{2l+1}{4\pi } } \delta^l_{(-m)s}\ , { } _ sa_{ln}\ , \delta^l_{(-m)n}.\ ] ] substitution of eq . together with eq . and eq . in eq . leads to @xmath188 evaluation of results in @xmath189 sampled on @xmath151 . as we require the function on @xmath2 we may truncate the output at @xmath190 discarding all data for @xmath191 . evaluating scales as @xmath136 , just as when performing the forward transformation . taking into account the symmetries of the @xmath180 matrices provided by eqs . ( [ eq : delindsym ] ) leads to an analogous halving of the number of operations required for the evaluation of eq . , as in the case of the forward transform . similarly , a further speedup is possible if the input data to the fft library is hermitian . in this section we follow @xcite and briefly outline an efficient recursive method for computing the @xmath180 that appear upon decomposition of the wigner @xmath6-matrices ( eq . ) when constructing the transformations in [ sec : algofwdxform ] and [ sec : algobwdxform ] . it can be seen directly from eq . that the elements @xmath192 have the following symmetries @xmath193 where the @xmath194 are combinatorial expressions purely dependent on the choice of indices @xmath195 and @xmath196 . suppose we require all possible @xmath180 up to a maximum @xmath135 . due to the symmetries of eqs . ( [ eq : delindsym ] ) , only a subset of all allowable indices need be calculated . for each choice of @xmath197 we restrict the indices for which @xmath180 is calculated to the set @xmath198 . we implement the _ trapani - navaza _ ( tn ) algorithm as follows : 1 . initialise @xmath199 ; 2 . iterate with @xmath200 3 . iterate with @xmath201 4 . iterate with @xmath202 5 . use symmetries ( i.e. eqs . ( [ eq : delindsym ] ) ) to find the remaining @xmath180 . note that the @xmath180 may be viewed as having a square pyramidal lattice structure with tn i - iv being a calculation of an octant subset tn v then allows for all values to be found : tn i corresponds to the apex @xmath203 ; tn ii corresponds to a tangent of values , a descent from the apex through the points @xmath204 ; tn iii corresponds to calculation of a right - angle , triangular lattice of surface values @xmath205 ; tn iv corresponds to calculation of interior points @xmath206 constrained to lie in the set @xmath198 ; tn v allows for recovery of all valid values of @xmath180 ( those outside the octant subset but within the square pyramid ) up to the chosen limit @xmath135 . one advantage of the tn algorithm , is that it is well suited to parallelization . performing steps tn i - iii and retaining the results then allows for the trivially parallelizable tn iv to be performed on multiple threads , as required . a disadvantage , however , is instability for large values of @xmath135 ( for our implementation @xmath207 ) . this issue , together with our proposal to correct for it without loss of efficiency will be described in [ sec : wdelstab ] . in the solution of pdes , product terms of swshs arise which may be decomposed as in eq . . two obvious paths are open to us : the first option is to implement a pseudo - spectral approach ; extracting the requisite @xmath208 by transforming two sets of appropriately seeded @xmath209 coefficient sets to functions , pointwise multiplying , then transforming back . we shall revisit this approach in [ sec : nummax ] . the second option , which we take , is direct computation of @xmath210 factors by calculation of clebsch - gordan coefficients ( or wigner-@xmath8-symbols by eq . ) . to this end we now describe an exact three - term linear recursive algorithm for calculating the wigner-@xmath8-symbols due to @xcite . the scheme we describe is numerically stable for ` small ' ( @xmath211 ) values of @xmath212 and @xmath213 . however , the possible occurrence of numerical loss of significance as well as overflow leads us to subsequently refine our approach . we apply the general conversion of a three term linear recursion relation into a hybrid recursion relation , given by luscombe and luban @xcite , to the recursion given in @xcite . our implementation extends the mentioned schemes to cover calculation of @xmath8-symbols of _ both _ integer and half - integer angular momentum numbers and projective quantum numbers . our aim is the simultaneous generation of @xmath214 for all @xmath215 where @xmath216 , @xmath217 , and where , in addition , @xmath218 is subject to the constraints discussed in [ sec : introcgw3j ] . consider eq . ; if @xmath219 each provide a valid @xmath8 symbol that are outside the range of validity are set to @xmath128 in the recursion relation . ] then the symbols may be connected via the following three - term linear recursion relation @xcite : @xmath220 where @xmath221^{1/2 } \left[(j_{2}+j_{3}+1)^{2}-j_{1}{}^{2}\right]^{1/2 } \left[j_{1}{}^{2}-m_{1}{}^{2}\right]^{1/2},\\\label{eq : w3jbcoeff } b(j_{1 } ) & : = -(2j_{1}+1 ) \left[j_{2}(j_{2}+1)m_{1}-j_{3}(j_{3}+1)m_{1}-j_{1}(j_{1}+1)(m_{3}-m_{2})\right].\end{aligned}\ ] ] the normalisation condition @xmath222 together with the phase convention @xmath223 allows for the determination of the family of @xmath8-symbols of eq . . the range of @xmath218 in eq . is divided into a ` classical ' and two complementary ` non - classical ' regions . a classical region is defined as the set of @xmath224 and @xmath225 for which it is possible to construct a vector diagram which physically corresponds to the coupling of angular momentum for more details , see @xcite . for our purposes it is sufficient to consider the non - classical region as values of @xmath226 that monotonically decrease as @xmath227 and @xmath228 . consider fig . [ fig : wig3jrecur ] where we illustrate the typical functional form of @xmath229 . in the classical region ( denoted @xmath230 ) the amplitude of @xmath229 oscillates about @xmath128 . within this region @xmath229 may evaluate to @xmath128 for specific choices of @xmath218 . the left boundary of the classical region is denoted by @xmath231 , the right boundary is denoted by @xmath232 . in the non - classical regions ( denoted @xmath233 ) @xmath226 monotonically decays to zero as the boundaries @xmath234 and @xmath235 are approached . -symbols as specified by eq . . parameters : @xmath236 @xmath237 @xmath238 @xmath239 @xmath240 . classical region denoted by @xmath230 , non - classical regions denoted by @xmath233 . ( red stars ) left boundary of classical region @xmath241 ; right boundary of classical region @xmath242 . there are 26 orders of magnitude difference between the largest and smallest values of @xmath229 in this family . computed using hybrid algorithm . ] in order to achieve numerical stability , the recursion relation of eq . that is used to generate the quantities @xmath229 must be performed in the direction of increasing @xmath226 . more explicitly , if the desired solution of a recurrence relation such as eq . , is monotonically decreasing ( as is the case if one iterates from classical to non - classical region ) then the other , linearly independent solution is monotonically increasing . thus numerical round - off in the calculation of a decaying solution of the recurrence relation triggers the growth of the unwanted , linearly independent , diverging solution @xcite . this discussion implies that instead , we should initialise eq . at the boundaries @xmath243 ( left - to - right recursion ) and @xmath244 ( right - to - left recursion ) and proceed toward the classical region @xmath230 . from eq . observe that @xmath245 and @xmath246 . the recursion relations at the boundaries , are given by @xmath247 note that the specification of _ one _ initial parameter at each boundary is sufficient to start the recursion ( eq . ) in each direction . as the initial choice of parameter for both the left and right recursions is arbitrary , the set of values generated by each recursion will be in error by factors @xmath248 and @xmath249 respectively ; explicitly @xmath250 where the recursion is terminated at a common , final @xmath218 value of @xmath251 . the left and right recursions must match values are achieved . this allows for the avoidance of singular terms in the classical region . ] at @xmath251 , which implies the constraint @xmath252 . we rescale the left recursion by @xmath253 and determine @xmath249 from the normalisation condition . finally , upon application of the phase condition , we have generated all valid @xmath229 as specified by . the algorithm just discussed suffers both from numerical loss of significance as well as overflow . this is due to the large variation in @xmath226 for particular parameter choices ; for example in the family of symbols shown in fig . [ fig : wig3jrecur ] there are 26 orders of magnitude difference between the largest and smallest values of @xmath226 . the loss of significance / overflow is often mitigated by rescaling of iterates in the recursion . alternatively , the use of a two - term nonlinear recursion may be employed @xcite . to this end , we instead work with ratios of successive @xmath229 terms . define the nonlinear left - to - right recursion @xmath254 and the nonlinear right - to - left recursion @xmath255 since @xmath256 and @xmath257 the initial values @xmath258 are known . the numerical advantage of making the transformations of eq . and eq . , is that @xmath259 and @xmath260 maintain values of order unity , throughout the recursion ( i.e. , at each iterate ) thus avoiding significance / overflow issues . a disadvantage introduced by the nonlinear scheme is that eq . and eq . are poorly defined for values of @xmath218 where @xmath261 . however , as this occurs only in the classical region @xmath230 , this motivates the consideration of a hybrid scheme a combination of nonlinear ( in the nonclassical regions @xmath233 ) and linear ( in the classical region @xmath230 ) . in order to split recursion schemes between the two methods , the location of the left @xmath231 and right @xmath232 boundaries must be known . the precise choice of @xmath231 and @xmath232 is not crucial ; the essential point is that nonlinear iteration is terminated near the boundary , such that no zero values of @xmath229 in the classical region are encountered . as @xmath226 monotonically decreases in the nonclassical regions and is known to be non - zero there , a simple algorithmic method of determining the aforementioned values is to begin iteration with eq . and eq . , comparing the magnitude of two consecutive iterates until a local maximum in @xmath226 is achieved . the coefficients as determined by the nonlinear scheme in the nonclassical region are hence computed using @xmath262 once the coefficients in the nonclassical regions are known , we may perform iteration using the three - term linear recurrence relation in the classical region using @xmath263 and @xmath264 as initial values in eq . for the left and right recursion respectively . the left and right hybrid recursion schemes must match at a common @xmath251 , upon appropriate rescaling , and normalisation ( eq . ) we finally apply the phase convention of eq . . this procedure allows for the determination of an entire family of @xmath229 as stated in eq . . if the coefficients are required at several stages of a calculation , it is generally more efficient to perform a recursive pre - computation , exploiting the symmetries ( si - siii ) ( see [ sec : introcgw3j ] ) in order to minimise memory usage . we implement this approach in our code where feasible , for details see @xcite . in order to check the outlined algorithms we perform comparisons with a symbolic calculation @xcite . symbolic calculations are slow . being of arbitrary accuracy they serve , however , as an excellent check . to verify our scheme all valid @xmath265 @xmath8-symbols were symbolically calculated up to a maximum entry of @xmath266 . figure [ fig : wig3jerrorfig ] shows the results of this procedure excellent accuracy ( machine precision ) is achieved . ( colour online ) an error comparison between the hybrid - recursive algorithm ( see [ sec : cgw3jnumproc ] ) and a direct symbolic calculation . in this calculation , random sets of valid wigner @xmath8-symbols were generated with possible parameters between @xmath267 and @xmath268 . upon generation of an invalid random combination of @xmath269 and @xmath270 , the parameters were discarded this process was iterated until @xmath271 distinct parameter sets were generated . the @xmath272 wigner @xmath8-symbols with largest error , according to the metric @xmath273 are shown , where @xmath274 and @xmath275 denote the value of a particular symbol as computed by the symbolic and recursive methods respectively ; the average of these symbols is shown in red . the @xmath276 symbols not shown , have a numerical error lower than that of the values displayed . in particular , we see that the numerical agreement between schemes is excellent ( with machine epsilon @xmath277 ) which implies an accurate implementation of the algorithm and stability for ` large ' parameter values . note that the hybrid - recursive approach can be adapted ( as demonstrated ) to compute symbols with half - integral angular and projective quantum numbers . , title="fig : " ] [ fig : wignererror_sub1 ] ( colour online ) an error comparison between the hybrid - recursive algorithm ( see [ sec : cgw3jnumproc ] ) and a direct symbolic calculation . in this calculation , random sets of valid wigner @xmath8-symbols were generated with possible parameters between @xmath267 and @xmath268 . upon generation of an invalid random combination of @xmath269 and @xmath270 , the parameters were discarded this process was iterated until @xmath271 distinct parameter sets were generated . the @xmath272 wigner @xmath8-symbols with largest error , according to the metric @xmath273 are shown , where @xmath274 and @xmath275 denote the value of a particular symbol as computed by the symbolic and recursive methods respectively ; the average of these symbols is shown in red . the @xmath276 symbols not shown , have a numerical error lower than that of the values displayed . in particular , we see that the numerical agreement between schemes is excellent ( with machine epsilon @xmath277 ) which implies an accurate implementation of the algorithm and stability for ` large ' parameter values . note that the hybrid - recursive approach can be adapted ( as demonstrated ) to compute symbols with half - integral angular and projective quantum numbers . , title="fig : " ] [ fig : wignererror_sub2 ] in this section we describe a simple method for checking the associated numerical error of our implementation of the spectral transformation outlined in [ sec : specxform ] . further , we check that well - known exponential decay properties of the magnitude of the representation of a function in spectral space are satisfied . although for the solution of a pde , band - limits @xmath135 that probe the border of stability of the tn recursive scheme ( see [ sec : algodelcalc ] ) may be difficult to achieve , we future - proof our work by proposing a simple linear / nonlinear hybrid scheme , in analogy to [ sec : cgw3jnumproc ] that restores stability . we then briefly contrast a pseudo - spectral approach with the full spectral scheme . as a preliminary check that the implementation of the spectral transformation is sound we proceed as follows . first the generation of a complex - valued function from a pseudo - random linear combination of the basis elements @xmath105 is made . the random weights of the aforementioned linear combination are denoted by @xmath278 . taking the inverse transformation of @xmath278 allows for the construction of the spatial representation of the function . performing the forward transformation in order to reconstruct the original random weights ( reconstructed weights denoted @xmath279 ) then allows for a comparison between @xmath278 and @xmath279 . as we work with floating - point arithmetic , we expect reconstruction to be within numerical tolerance . we summarise this procedure as @xmath280 by first separating the real and imaginary parts of each @xmath281 as @xmath282 } + { \mathrm{i}}\im{\left[{}_s\tilde{a}_{lm}\right]}\ ] ] we generate both @xmath283}$ ] and @xmath284}$ ] by sampling from the continuous uniform random distribution on the interval @xmath285 $ ] . although sampling random data as above provides us with a simple and robust diagnostic on our implementation , it is also informative to examine coefficient decay for specific test functions that are known to possess specific properties with respect to their spectral representation . we now inspect how the magnitude of coefficients of a function decay with increasing band - limit . in order to represent this decay in a convenient manner define the averaged coefficient : @xmath286 we now introduce the following smooth test functions : @xmath287 ^ 2\right),\\\label{eq : testcoeffdec4 } { } _ 0 k(\vartheta,\varphi)&:= \exp\left(-{}_0y_{1,-1}(\vartheta,\varphi)+{}_0y_{1,1}(\vartheta,\varphi ) \right),\\\label{eq : testcoeffdec5 } { } _ { -2 } q(\vartheta,\varphi ) & : = \frac{1}{256}{}_{-2}y_{25,-9}(\vartheta,\varphi)\exp\left(9.3\,{}_0y_{20,-2}(\vartheta,\varphi)-22.5\,{}_0y_{15,3}(\vartheta,\varphi ) \right),\\\label{eq : testcoeffdec6 } { } _ { -4 } r(\vartheta,\varphi ) & : = \frac{1}{4096 } { } _ { -2 } q^2(\vartheta,\varphi)\exp\left(-5\,{}_0y_{1,0}(\vartheta,\varphi ) \right),\end{aligned}\ ] ] the spatial representation of which is initially constructed numerically . performing the forward transformation yields the spectral representation , which together with eq . yields the averaged coefficient decay which is shown in fig . [ fig : coeffdecay01 ] . equation and eq . are clearly comprised of a finite linear combination of swsh and indeed this fact is reflected by their complete capture at the band - limits tested . as eq . is completely captured at bandlimits @xmath288 and @xmath289 , the only difference in the average magnitude of coefficients ( fig . [ fig : coeffdecay01 ] ) is solely due to the additional number of arithmetic operations that must be performed during the transformation . equation is a spin-@xmath1 function modulated in magnitude and phase by a smooth spin-@xmath128 function . in a similar manner eq . and eq . are smooth functions of spin weight @xmath290 and @xmath291 respectively . we expect that given smooth test functions the decay of the magnitude of the averaged coefficient @xmath210 should behave as @xmath292 @xmath293 for large @xmath197 @xcite . this behaviour demonstrated in fig.[fig : coeffdecay02 ] where we construct a linear fit on a semilog plot , is characterstic of the expected exponential convergence that smooth functions should display @xcite . in the spectral representation of the functions in eqs . ( [ eq : testcoeffdec1]-[eq : testcoeffdec4 ] ) . cyan ` @xmath294 ' : @xmath295 transformation of eq . at a band - limit of @xmath288 . green ` @xmath296 ' : @xmath295 transformation of eq . at a band - limit of @xmath289 . magenta ` @xmath297 ' : @xmath298 transformation of eq . at a band - limit of @xmath299 . blue ` @xmath300 ' : @xmath295 transformation of eq . at a band - limit of @xmath299 . black ` @xmath301 ' : @xmath302 transformation of eq . at a band - limit of @xmath303 . note : the difference between coefficient magnitudes for transformations of eq . at @xmath288 and @xmath289 is solely due to the additional number of arithmetic operations required when performing a transform at a higher bandlimit @xmath135 . b ) ( left to right ) blue ` @xmath300 ' : @xmath295 transformation of eq . at a band - limit of @xmath299 . cyan line : linear fit of @xmath304 , @xmath305 . green ` @xmath296 ' : @xmath306 transformation of eq . at a band - limit of @xmath289 . magenta line : linear fit of @xmath307 , @xmath305 . red ` @xmath294 ' : @xmath308 transformation of eq . at a band - limit of @xmath289 . black line : linear fit of @xmath309 , @xmath305 . see text for discussion . , title="fig : " ] [ fig : coeffdecay01 ] in the spectral representation of the functions in eqs . ( [ eq : testcoeffdec1]-[eq : testcoeffdec4 ] ) . cyan ` @xmath294 ' : @xmath295 transformation of eq . at a band - limit of @xmath288 . green ` @xmath296 ' : @xmath295 transformation of eq . at a band - limit of @xmath289 . magenta ` @xmath297 ' : @xmath298 transformation of eq . at a band - limit of @xmath299 . blue ` @xmath300 ' : @xmath295 transformation of eq . at a band - limit of @xmath299 . black ` @xmath301 ' : @xmath302 transformation of eq . at a band - limit of @xmath303 . note : the difference between coefficient magnitudes for transformations of eq . at @xmath288 and @xmath289 is solely due to the additional number of arithmetic operations required when performing a transform at a higher bandlimit @xmath135 . b ) ( left to right ) blue ` @xmath300 ' : @xmath295 transformation of eq . at a band - limit of @xmath299 . cyan line : linear fit of @xmath304 , @xmath305 . green ` @xmath296 ' : @xmath306 transformation of eq . at a band - limit of @xmath289 . magenta line : linear fit of @xmath307 , @xmath305 . red ` @xmath294 ' : @xmath308 transformation of eq . at a band - limit of @xmath289 . black line : linear fit of @xmath309 , @xmath305 . see text for discussion . , title="fig : " ] [ fig : coeffdecay02 ] [ fig : coeffdecay ] one aspect of the tn algorithm ( briefly mentioned in [ sec : algodelcalc ] ) is that when transformations with a high band - limit are required instability may arise . figure [ fig : wignerdel_sub1 ] shows that instability occurs at @xmath310 ; the situation is analogous to that when encountered whilst performing calculations of the @xmath8-symbols hence we proceed analogously . in order to ameliorate the issue , we construct a new nonlinear recursion relation . recall tn i - iv of [ sec : algodelcalc ] . fix @xmath197 and @xmath213 . observe that in the region @xmath311 we are performing recursion with a three term linear relation , furthermore , in this region we have values of @xmath180 that monotonically increase in magnitude when @xmath196 is decreased from @xmath197 . this leads to instability of the tn algorithm . in order to avoid this , we apply luscombe and luban s general method @xcite , for the construction of a hybrid linear / non - linear recursion to trapani and navaza s linear recursive scheme . to rewrite the three - term linear recursion relation stated in eq . as a two - term nonlinear recursion relation define the ratio @xmath312 . this results in @xmath313 to make use of this equation , we require a term @xmath180 within the range where nonlinear recursion is performed such that ratios may be converted to absolute values . while edge values with @xmath314 ( for fixed @xmath213 ) may be used to calculate explicit terms from ratios , for reasons of numerical accuracy , it is advantageous to instead perform three - term iteration from @xmath315 to the edge of the region for which values are computed by the nonlinear scheme ( see the dashed red arc in fig . [ fig : wignerdel_sub2 ] . hence we now write a new three term relation for a fixed @xmath196 connecting three different values of @xmath213 this is accomplished using the indicial symmetry given by the third of eq . @xmath316 begins to occur for @xmath317 . a ) @xmath318 under three - term linear recursion , with spurious values occurring in a region about @xmath319 . it is clear that these values are incorrect as @xmath320 . b ) hybrid scheme as discussed in text stability is maintained . dashed red arc indicates @xmath321 , dashed green line indicates edge of computational domain ( prior to application of symmetries ) for both recursion schemes . note : color scales are distinct in both figures . , title="fig : " ] [ fig : wignerdel_sub1 ] begins to occur for @xmath317 . a ) @xmath318 under three - term linear recursion , with spurious values occurring in a region about @xmath319 . it is clear that these values are incorrect as @xmath320 . b ) hybrid scheme as discussed in text stability is maintained . dashed red arc indicates @xmath321 , dashed green line indicates edge of computational domain ( prior to application of symmetries ) for both recursion schemes . note : color scales are distinct in both figures . , title="fig : " ] [ fig : wignerdel_sub2 ] [ fig : wigdelhyb ] consider two , smooth spin - weighted functions @xmath322 . both @xmath323 and @xmath324 may be decomposed according to eq . [ eq : funcs2sphtrunc ] , in terms of swshs of spin - weight @xmath325 and @xmath326 respectively , for some choice of band - limit @xmath135 . as we will see in [ sec : advnum ] product terms such as @xmath327 and hence products of the form @xmath328 will need to be expanded in terms of a linear combination of @xmath329 . we have already discussed how such expansions may be performed in a completely spectral scheme in [ sec : introcgw3j ] , however when formulating the solution of non - linear pdes in particular , multiple - sum terms can quickly become cumbersome without automatic code generation or some level of abstraction . an alternative to the above is the pseudo - spectral method . suppose we initially have the coefficients @xmath330 and @xmath331 representing the functions @xmath323 and @xmath324 sampled ( or intially seeded ) at some band - limit @xmath135 . the coefficients @xmath332 of the associated product @xmath333 can be calculated by performing the transformations : @xmath334 subsequently taking the pointwise product and transforming : @xmath335 we find an approximation to an expansion utilizing eq . directly . in order for the pseudo - spectral and spectral to coincide ( to numerical tolerance ) at some band - limit @xmath135 we have found that it is best to choose a band - limit for the pseudo - spectral scheme @xmath336 , then truncating the constructed coefficients at a band - limit of @xmath135 this is the so - called ` orszag 2/3 rule ' @xcite which aids in the suppression of spurious aliasing . we emphasize that this method also easily allows one to take into account the action of the @xmath104 operators on functions by embedding their action as multiplication ( see eq . and eq . ) in coefficient space , together with taking account of their spin raising and lowering properties when transformations are performed . let @xmath337 , and @xmath338 be the smooth time function given by @xmath339 for all @xmath340 and @xmath341 with non - vanishing gradient . for each @xmath341 , we consider the curve @xmath342 , @xmath343 , and the corresponding smooth tangent vector field @xmath344 ; in particular we have @xmath345 . because of this we can introduce coordinates @xmath346 on @xmath347 , where @xmath101 are standard polar coordinates on @xmath48 and where @xmath348 . let the level sets of @xmath349 be denoted by @xmath350 any such subset is a smooth embedded submanifold of @xmath347 . we make the same assumptions as in the last paragraph of section [ sec : weightedquantities ] in particular @xmath50 is a dense open subset of @xmath48 and we choose a smooth complex vector field ; in particular , @xmath213 should not be confused with the field in eq . ( [ eq : mmstar ] ) . we shall choose @xmath213 specifically in the two applications discussed below . notice also that since there is no metric defined on @xmath347 so far , we do also not assume a normalization for @xmath213 yet . ] @xmath213 on @xmath351 which is tangent to @xmath352 at each @xmath353 and which satisfies @xmath354 . let us suppose that @xmath355 is a smooth frame almost everywhere on @xmath347 . we let @xmath356 be the corresponding dual frame and hence deduce that @xmath357 . the following notion of spin - weight based on frame transformations of the form @xmath358 where @xmath57 can be an arbitrary smooth function on @xmath347 , is useful for the following discussion . any quantity @xmath359 on @xmath347 , which behaves likes @xmath360 under this transformation , is said to have spin - weight @xmath78 . for instance , the frame vector @xmath68 has spin - weight @xmath128 , @xmath213 has spin - weight @xmath361 and @xmath362 has spin - weight @xmath363 . of particular importance for the following discussions are commutators of the frame fields @xmath364\right>}.\ ] ] the assumptions above yield @xmath365 and all commutators can be computed from the following ones @xmath366 the functions @xmath367 , @xmath368 , @xmath369 , @xmath370 , @xmath371 and @xmath372 have the following transformation behavior under the frame transformations above : @xmath373 i.e. , these are functions have a well - defined spin - weight @xmath78 , while @xmath374 and hence do not have a spin - weight . notice that in particular , @xmath375 for @xmath376 , i.e. , for the reference frame . we want to start with the following _ advection problem_. let @xmath377 be a given smooth vector field on @xmath347 with @xmath378 . pick a smooth @xmath379-tensor field @xmath380 in a neighborhood of @xmath381 . we consider the initial value problem @xmath382 for an unknown @xmath379-tensor field @xmath383 on @xmath347 . since @xmath377 is smooth it generates a flow on @xmath347 which maps each point @xmath60 of @xmath347 to a point on the integral curve of @xmath377 through @xmath60 . due to the condition @xmath378 , it follows that @xmath377 has a non - vanishing ` time component ' in the direction of @xmath68 and , in general , a non - vanishing ` spatial component ' tangential to @xmath352 . solutions @xmath383 of are invariant under this flow generated by @xmath377 and are therefore advected in time along the ` spatial component ' of @xmath377 . we notice that in the case @xmath384 , reduces to the standard scalar advection equation . moreover , we observe that this initial value problem does not require the specification of a metric on @xmath347 . in order to bring eq . ( [ eq : advectionproblem ] ) into a form for which our formalism applies , we choose the frame @xmath385 above with @xmath348 and @xmath386 given by eq . ( [ eq : standardreferenceframe ] ) . hence the dual frame is @xmath387 with @xmath388 with this it follows that @xmath389 since the vector field @xmath377 has the property @xmath378 , there exists a smooth complex function @xmath390 so that @xmath391 notice that @xmath390 has spin - weight @xmath363 ( and hence @xmath392 spin - weight @xmath361 ) . for simplicity , we assume that @xmath390 is independent of @xmath353 . we restrict to the scalar and vectorial advection problems now , i.e. to @xmath384 and @xmath393 . in the scalar case , @xmath383 is a function with spin - weight zero hence eq . translates to @xmath394 using and for @xmath302 . we check easily that each term in this equation is of spin - weight @xmath128 . in general , consistency of the spin - weights of the terms in an equation is a good indication that it has been derived correctly . in the vectorial case , we decompose the vector @xmath383 as @xmath395 where @xmath396 is of spin - weight @xmath78 and where @xmath397 . projecting onto @xmath196 , @xmath398 and @xmath399 , and using , and and the fact that @xmath371 and @xmath372 vanish for the reference frame , we find @xmath400 using the relations and to write @xmath401 for any quantity @xmath402 of spin - weight @xmath78 , we find @xmath403 we realize that all terms are of a well - defined and consistent spin - weight and that all formally singular terms ( i.e. , those proportional to @xmath404 ) disappear ( as one expects ) . the first of eqs . is of the same form as and is decoupled from the other two . the third equation is the complex conjugate of the second one . in this section we solve and together with appropriate initial data by application of the fourier - galerkin method . we will choose advecting fields @xmath405 comprised of linear combinations of axial rotations . this choice is particularly amenable to analysis of numerical error and stability since initial data that is advected by such fields will undergo a time evolution that must periodically return to its initial state it is this periodic behaviour that we exploit for our numerical tests . recall that the generators of rotations in @xmath44 are given by : @xmath406 respectively . if we choose @xmath377 to be one of these generators and then decompose @xmath377 as in we find @xmath407 it is convenient to further normalize the advecting fields @xmath408 such that the final state of the fields being advected will again coincide with the choice of the initial field configuration after a temporal period @xmath409 . for general data eq . gives : @xmath410 for advecting fields and the fields being advected respectively . the time - dependence of the solution is carried by the expansion coefficients @xmath411 . equations ( [ eq : ethraco ] , [ eq : ethloco ] , [ eq : sylmorth ] , [ eq : spinnyspectroscope ] , [ eq : advectionproblemscalar ] , [ eq : advexpa ] ) lead to the spectral representation of the scalar advection problem : @xmath412 \left\{\vphantom{\sum^l_{l_a=1 } } \sqrt{l_b(l_b+1 ) } \mathcal{a}_l(-1,l_a , m_a;\,1,l_b , m - m_a ) \right.\\ & \left.\times{}_{0}n_{l_b , m - ma}(t ) \left({}_{-1}\xi_{l_a , m_a}-(-1)^{l_a+l_b+l}{}_{1}\xi_{l_a , m_a ) } \right ) \vphantom{\sum^l_{l_a=0}}\right\ } , \end{aligned}\ ] ] where @xmath413 $ ] is a boolean function equal to @xmath361 when the inequality is satisfied and @xmath128 when it is not . we introduce such functions to make explicit the distinct limits of summation of the terms in eq . . for the vector advection problem the spectral representation of the spin @xmath128 component of the system is again given by eq . . due to the condition @xmath414 and eq . we have @xmath415 , hence in order to completely specify the system , only the spin @xmath363 equation is required : @xmath416 \left\{\vphantom{\sum_{l_a=1}^l } \sqrt{l_b(l_b+1 ) } \left ( { } _ { -1}\xi_{l_a , m_a}\,{}_{-1}n_{l_b , m - m_a}(t ) \right . \right.\\ & \left.-{}_{-1}\xi_{l_b , m - m_a}\,{}_{-1}n_{l_a , m_a}(t)\right ) \mathcal{a}_l(-1,l_a , m_a;\,0,l_b , m - m_a)\\ & + \sqrt{(l_b-1)(l_b+2 ) } \left ( { } _ { -1}\xi_{l_b , m - m_a}{}_{1}n_{l_a , m - m_a}(t ) \right.\\ & \left . -{}_{1}\xi_{l_a , m - m_a}{}_{-1}n_{l_b , m - m_a}(t ) \right)\mathcal{a}_l(1,l_a , m_a;-2,l_b , m - m_a)\left[\delta_{l_b\geq 2}\right ] \left.\vphantom{\sum_{l_a=1}^l}\right\ } \end{aligned}\ ] ] figure [ fig : advectionsweepnst ] shows the results of numerical convergence tests performed using an explicit @xmath417th order runge - kutta ( rk ) method in time for a variety of time - steps @xmath418 and band - limits @xmath135 . we show the absolute value of the maximum difference in coefficient space @xmath419 between initial configurations of test fields and their final configurations upon advection for @xmath420 periods @xmath421 . excellent agreement with the expected @xmath417th order convergence is displayed . we find that in both scalar and vector cases the error associated with the temporal discretization dominates that of the spatial scheme this was verified by constructing a semilog plot of data generated for various values of @xmath135 vs. @xmath422 at fixed @xmath418 where we found a horizontal line . this is not unexpected as we are advecting smooth fields , by smooth fields and expect exponential convergence for the spatial sampling ( see [ sec : errorstabilityspeed ] ) whereas only @xmath417th order convergence in time is provided by the runge - kutta method . note , that here and in what follows when performing numerical expansions as in eqs . ( [ eq : clebschgordanseries],[eq : prodsoln ] ) all terms with @xmath423 are discarded . periods elapse . ( _ advection of scalar field _ ) : green ` @xmath424 ' : @xmath425 ; magenta ` @xmath426 ' : @xmath427 ; cyan ` @xmath294 ' : @xmath428 ; yellow ` @xmath297 ' : @xmath303 : advecting field corresponds to a @xmath429 axial rotation , initial data corresponds to eq . . ( _ advection of vector field _ ) : blue : @xmath427 , ` @xmath430 ' : @xmath431 component , ` @xmath301 ' : @xmath432 component ; red : @xmath428 , ` @xmath300 ' : @xmath431 component , ` @xmath296 ' : @xmath432 component ; black : @xmath303 , ` @xmath433 ' : @xmath431 component , ` @xmath434 ' : @xmath432 component : advecting field corresponds to the normalized sum of @xmath435 and @xmath436 axial rotations , initial data corresponds to a @xmath429 axial rotation . we find excellent agreement with the expected @xmath417th order convergence , prior to a saturation in convergence due to numerical roundoff error . ] as a second application , we study the initial value problem of the @xmath0-vacuum maxwell equations on @xmath347 , i.e. , the equations @xmath437 for the electromagnetic @xmath1-form @xmath438 on @xmath347 , where @xmath6 is the exterior derivative and @xmath300 is the hodge dual associated with a lorentzian metric @xmath439 on @xmath347 ; we assume the signature @xmath440 . in abstract index notation , these equations can also be written as @xmath441}=0,\quad \nabla^\mu f_{\mu\nu}=0.\ ] ] let @xmath442 be an orthonormal frame with respect to @xmath439 and choose the frame @xmath443 by @xmath444 and @xmath445 . this implies for example @xmath446 , @xmath447 and @xmath448 . let @xmath449 be the coframe dual to @xmath442 . if @xmath387 is the coframe dual to @xmath450 then @xmath451 since any smooth metric on @xmath48 is conformal to the standard metric on the unit sphere , a general smooth metric on @xmath347 is described by a smooth real is complex . its phase , however , would generate nothing but a rotation of the frame and would hence not contribute to the metric . ] strictly positive function @xmath452 of spin - weight zero so that @xmath453 where @xmath454 is given by eq . ( [ eq : standardreferenceframe ] ) . in the special case @xmath455 , one obtains the geometry of the standard round unit sphere . physically , our system can therefore be interpreted as vacuum electrodynamics in a universe of @xmath1-dimensional spatial spherical topology whose geometry is described by the function @xmath456 . notice that we allow @xmath456 to depend on time . as before , we assume @xmath348 . then , it follows @xmath457 in @xmath0 dimensions , the @xmath1-form @xmath438 has three independent components which we write as @xmath458 hence @xmath459 and @xmath460 can be interpreted as the two components of the electric field and @xmath461 as the component of the magnetic field . in fact , we can introduce a purely spatial `` electric '' one - form @xmath462 so that the one - form @xmath463 becomes @xmath464 for the following it is useful to define the three complex maxwell scalars as @xmath465 where @xmath466 has spin - weight @xmath78 . these fields satisfy the reality conditions @xmath467 we find that the electric one - form @xmath468 can be written as @xmath469 the first maxwell equation @xmath470 is equivalent to @xmath471 , and hence to @xmath472 this equation corresponds symbolically to the maxwell equation `` @xmath473 '' . using the same arguments as for the advection problem , we arrive at the following evolution equation for the magnetic field @xmath474 the second maxwell equation @xmath475 is equivalent to @xmath476 , and hence to @xmath477 this corresponds to three equations . the @xmath196-component is the constraint @xmath478 the @xmath398- and @xmath399-components yield evolution equations for the components of the electric field @xmath479 we see explicitly that all terms in the equations are non - singular and of consistent spin - weight . moreover , when we write the @xmath3-operator in a coordinate basis , we see that the evolution system is symmetric hyperbolic and hence gives rise to a well - posed initial value problem . it remains to show that the constraint @xmath480 propagates under the evolution . for this we derive the evolution equation for the constraint violation quantity @xmath481 given in eq . ( [ eq : constraint ] ) . we take the time derivative of eq . ( [ eq : constraint ] ) and use the evolution equations of @xmath482 , @xmath483 , @xmath484 . then it is straightforward to find that @xmath485 it follows that if the initial data satisfy the constraint , i.e. , @xmath480 at the initial time , then @xmath486 for all times and hence the constraints will be satisfied , up to machine precision , during the whole evolution . furthermore , as can easily be seen from the evolution equations , the reality conditions of eq . are preserved during the whole evolution provided they are fulfilled at the initial time . notice that @xmath487 is equivalent to @xmath488 . since @xmath468 is purely spatial this is equivalent to @xmath489 , where @xmath490 is the purely spatial exterior derivative on the initial hyper - surface . because @xmath48 is simply connected it follows that every solution @xmath468 of the constraint is of the form @xmath491 where @xmath492 is an arbitrary smooth scalar function of spin - weight zero on the initial hyper - surface . comparison of eq . with eq . yields @xmath493 as the constraint eq . is independent of @xmath494 the magnetic field may be prescribed freely , subject only to reality conditions . our goal is now the construction of numerical solutions by means of spectral decompositions of the dynamical equations governing the maxwell system . in order to check the numerical solutions thus constructed , we examine the preservation of the constraints associated with the system . in order to further ensure that our implementation is accurate and robust we compare our calculation to a pseudo - spectral implementation based on discussion of [ sec : psandfscmp ] . we proceed in two stages : we consider the case where @xmath2 is deformed at the initial time and examine how a simple choice of initial data evolves with this fixed geometry . we then allow for a time - dependent change of the geometry , again examining how our solution for the fields develops with time . as the method is entirely analogous to that of [ sec : advnum ] we do not explicitly state our expansions of eqs . ( [ eq : maxev0 ] , [ eq : maxev1 ] , [ eq : maxev2 ] , [ eq : inicondmapot ] , [ eq : constraint ] ) here . however it is worth pointing out that due to and we again have @xmath495 which implies that only a subset of the full dynamical system need be evolved , the rest may be extracted by these symmetries . we have also found it convenient to re - expand terms such as @xmath496 by defining an auxiliary function : @xmath497 rather than dealing with @xmath456 directly . we now test the following initial conditions : @xmath498 with all other values set to zero . this corresponds to a static deformation of @xmath2 , where at @xmath499 the deformation is chosen and fixed for all later times . we calculate the solution numerically making use of the spectral and pseudo - spectral methods together with the embedded rk@xmath500 m algorithm of @xcite . this last choice of integrator allows for local error estimation and hence adaptive control of step - size in time . in order to verify that our implementation is consistent we check that the constraints are satisfied this is done by ( pseudo)spectral decomposition of eq . in a similar manner to the preceding equations of this section . the results of this are shown for both spectral ( fig . [ fig : constr_sub1 ] ) and pseudo - spectral ( fig . [ fig : constr_sub2 ] ) implementations ; we find excellent agreement between both methods . ( colour online ) value of the constraints ( eq . ) associated with solutions of the maxwell system for static deformation of @xmath2 subject to the initial conditions of eq . using adaptive rk@xmath500 m as described in the text . red : @xmath427 , ` @xmath300 ' ; blue : @xmath428 , ` @xmath296 ' ; green : @xmath303 , ` @xmath301 ' ; magenta : @xmath288 , ` @xmath501 ' . observe that constraints are well - preserved during the entire course of the solution - with the spectral ( a ) and pseudo - spectral ( b ) methods providing comparable accuracy . at @xmath502 we find that we have approximately saturated the convergence in @xmath135 , that is , further increasing the band - limit will only marginally increase solution accuracy and a law of diminishing returns applies cf . [ fig : nrg ] . , title="fig : " ] [ fig : constr_sub1 ] ( colour online ) value of the constraints ( eq . ) associated with solutions of the maxwell system for static deformation of @xmath2 subject to the initial conditions of eq . using adaptive rk@xmath500 m as described in the text . red : @xmath427 , ` @xmath300 ' ; blue : @xmath428 , ` @xmath296 ' ; green : @xmath303 , ` @xmath301 ' ; magenta : @xmath288 , ` @xmath501 ' . observe that constraints are well - preserved during the entire course of the solution - with the spectral ( a ) and pseudo - spectral ( b ) methods providing comparable accuracy . at @xmath502 we find that we have approximately saturated the convergence in @xmath135 , that is , further increasing the band - limit will only marginally increase solution accuracy and a law of diminishing returns applies cf . [ fig : nrg ] . , title="fig : " ] [ fig : constr_sub2 ] checking other invariants of the system can give further insight into the performance of our numerical scheme . consider the energy as a function of time : @xmath503 due to eq . we can conclude that eq . is the integral of a positive definite quadratic form ( since @xmath504 for @xmath505 $ ] ) and hence @xmath506 for all @xmath353 . for time - independent @xmath456 , @xmath507 is a killing vector and hence by noether s theorem if @xmath508 is the initial time , then we have @xmath509 . that is , under a static deformation energy is conserved . in order to examine any deviations in @xmath510 that may occur due to numerical error , it is convenient to further define relative error in the energy via : @xmath511 which serves as a measure of the failure of energy conservation . we show the value of @xmath512 in fig . [ fig : nrg ] for the solution of the maxwell system with the initial conditions of eq . , again both spectral and pseudo - spectral methods perform with consistent accuracy affirming our intuition about energy conservation for the system in the case of static deformation . ( colour online ) the relative error in the energy @xmath513 as defined in eq . for numerical solution of the maxwell system subject to the initial conditions of eq . red : @xmath427 , ` @xmath300 ' ; blue : @xmath428 , ` @xmath296 ' ; green : @xmath303 , ` @xmath301 ' ; magenta : @xmath288 , ` @xmath501 ' . in both solutions , constructed via spectral ( a ) and pseudo - spectral ( b ) methods respectively , we find that oscillations in @xmath513 are observed for @xmath514 . we ascribe the aforementioned oscillations to the numerical scheme and not to any physical property of the maxwell system as for @xmath515 these spurious features disappear . it is important to note that even though oscillatory behaviour in @xmath513 is observed , it is bounded over the interval @xmath353 shown . , title="fig : " ] ( colour online ) the relative error in the energy @xmath513 as defined in eq . for numerical solution of the maxwell system subject to the initial conditions of eq . . red : @xmath427 , ` @xmath300 ' ; blue : @xmath428 , ` @xmath296 ' ; green : @xmath303 , ` @xmath301 ' ; magenta : @xmath288 , ` @xmath501 ' . in both solutions , constructed via spectral ( a ) and pseudo - spectral ( b ) methods respectively , we find that oscillations in @xmath513 are observed for @xmath514 . we ascribe the aforementioned oscillations to the numerical scheme and not to any physical property of the maxwell system as for @xmath515 these spurious features disappear . it is important to note that even though oscillatory behaviour in @xmath513 is observed , it is bounded over the interval @xmath353 shown . , title="fig : " ] we now allow a time - dependent deformation of @xmath2 using the pseudo - spectral method . the direction @xmath507 is no longer a killing vector and energy expression of eq . will no longer be conserved , however we still have @xmath506 for all @xmath353 . for initial conditions , we take : @xmath516 where @xmath517 governs the time - dependent deformation of @xmath2 over an interval @xmath518 $ ] . we show the solution at selected times , together with the geometric picture in fig.[fig : maxsoln ] . as @xmath456 is now time - dependent the constraint propagation equation suggests that numerical violations of the constraints grow with time . convergence of @xmath481 to a well - defined value with increasing band - limit @xmath135 is still expected and we find that a higher band - limit @xmath135 is required ( see fig.[fig : tdepconstr ] ) in constrast to the static ( no deformation ) case . finally we note that as anticipated , @xmath506 for @xmath518 $ ] ( see fig.[fig : tdepnrg ] ) . ( colour online ) maxwell @xmath0 solution , together with geometry as constructed by pseudo - spectral method at indicated times . parameters : @xmath303 , initial conditions and time - dependent @xmath456 as in eq . and eq . . in each subfigure we show the electric field ( red ) scaled by the dimensionless factor @xmath519 together with the magnetic scalar . nodes corresponding to the spatial sampling are indicated in black on the surface . note : color and axial scalings are distinct over figures . , title="fig : " ] ( colour online ) maxwell @xmath0 solution , together with geometry as constructed by pseudo - spectral method at indicated times . parameters : @xmath303 , initial conditions and time - dependent @xmath456 as in eq . and in each subfigure we show the electric field ( red ) scaled by the dimensionless factor @xmath519 together with the magnetic scalar . nodes corresponding to the spatial sampling are indicated in black on the surface . note : color and axial scalings are distinct over figures . , title="fig : " ] ( colour online ) maxwell @xmath0 solution , together with geometry as constructed by pseudo - spectral method at indicated times . parameters : @xmath303 , initial conditions and time - dependent @xmath456 as in eq . and in each subfigure we show the electric field ( red ) scaled by the dimensionless factor @xmath519 together with the magnetic scalar . nodes corresponding to the spatial sampling are indicated in black on the surface . note : color and axial scalings are distinct over figures . , title="fig : " ] ( colour online ) maxwell @xmath0 solution , together with geometry as constructed by pseudo - spectral method at indicated times . parameters : @xmath303 , initial conditions and time - dependent @xmath456 as in eq . and in each subfigure we show the electric field ( red ) scaled by the dimensionless factor @xmath519 together with the magnetic scalar . nodes corresponding to the spatial sampling are indicated in black on the surface . note : color and axial scalings are distinct over figures . , title="fig : " ] , ` @xmath300 ' ; blue : @xmath303 , ` @xmath296 ' ; green : @xmath288 , ` @xmath301 ' ; magenta : @xmath520 , ` @xmath501 ' , we find similar convergence behaviour of the contraints ( a ) to that of the time - independent deformation case , however we note that higher spatial resolution ( band - limit @xmath135 ) is required . in ( b ) we find that the relation @xmath521 is indeed satisfied . , title="fig : " ] [ fig : tdepconstr ] , ` @xmath300 ' ; blue : @xmath303 , ` @xmath296 ' ; green : @xmath288 , ` @xmath301 ' ; magenta : @xmath520 , ` @xmath501 ' , we find similar convergence behaviour of the contraints ( a ) to that of the time - independent deformation case , however we note that higher spatial resolution ( band - limit @xmath135 ) is required . in ( b ) we find that the relation @xmath521 is indeed satisfied . , title="fig : " ] [ fig : tdepnrg ] [ fig : tdepconstrnrg ] in this work we have presented a method for evolving tensorial equations on manifolds with spherical topology . it is based on the use of the spin - weighted spherical harmonics , a class of functions on the sphere which is closely related to irreducible representations of @xmath522 . we have demonstrated that our method exhibits the accuracy and rapid convergence to solutions that is expected from spectral methods . of course , this method is not limited to a single sphere . instead , our main application will be in systems defined on a manifold with spatial topology @xmath523 , where the tensor fields are split into various pieces intrinsic to the sphere factor and depending on a ` radial ' coordinate corresponding to the @xmath524 factor . this kind of topology occurs naturally in the description of the global structure of space - times in general relativity . this method can be further generalised to half - integer spin . this will allow us to solve spinorial equations , such as the dirac equation or weyl s equation for a ( massless ) neutrino on space - times with spheroidal components as discussed above . this research was partly funded by the marsden fund of the royal society of new zealand under contract number uoo0922 .

during the last decade , there has been a dramatic paradigm shift on the the definition of the gc systems . despite the formerly accepted idea of chemical homogeneity , the variations in the lighter elemental abundances in several gcs in our galaxy had been known for several decades . the lick - texas group is one of those who undertook the systematic study of lighter elemental abundances in several gcs in our galaxy since 1990 s ( e.g. * ? ? ? * ; * ? ? ? thanks to the advent of high performance multi - object high - resolution spectrographs mounted on the large aperture telescopes , it is now possible for us to look into the detailed substructure of elemental abundance distributions of the milky way gc systems ( e.g. * ? ? ? * ) . the decades - long lighter elemental variation issue in gc stars is now considered to be a generic feature of normal gcs in our galaxy , most likely engraved during the multi - phase normal gc formation ( e.g. * ? * ; * ? ? ? * ; * ? ? ? contrary to the normal gc system , one of the key features of the peculiar gcs , such as @xmath3 cen , is the spread or the distinctive substructure in the metallicity distributions ( e.g. * ? ? ? * ; * ? ? ? * ) , where the heavy elements must have been supplied by supernovae ( sne ) . to retain ejecta from energetic sne explosions , such peculiar gc systems must have been much more massive in the past and they are generally thought to be the remaining core of a disrupted dwarf galaxy and accreted to our galaxy later in time , expected from the hierarchical merging paradigm in the @xmath4cmd cosmological model ( e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ? the existence of these peculiar gcs have major implications in the context of near field cosmology ( e.g. * ? ? ? * ) ; are they one of the original building blocks to our galaxy , mitigating so - called the `` missing satellite problem ? '' @xcite ; how did they relate to the formation of the galactic halo and numerous streamers ? ( e.g. * ? ? ? these are examples of the outstanding problems that we have to challenge in the next decade . to measure metallicity of stars in gcs from high - resolution spectroscopy , the lte ( local thermodynamic equilibrium ) analysis is being widely used for the sake of convenience , where the final results critically depend on the stellar atmosphere models with a few appropriate input stellar parameters , such as effective temperature , surface gravity and turbulent velocity , and the oscillator strengths of the absorption lines . although simple , the derivation of stellar elemental abundances is not a trivial task even for nearby bright stars . the recent study of @xcite may highlight the current situation . they showed that the interferometric effective temperatures for nearby k giant stars do not agree with those from spectroscopic observations , suggesting a missing source of opacities in stellar atmosphere models . the situation would be even worse for fainter stars , such as those in gcs . another line of difficulty is that changes in surface gravity can mimic the chemical compositions in the regime of rgb stars in gcs ( for example , see * ? ? ? * ) , where h@xmath5 is the major source of the continuum opacity and the h@xmath5 population varies with the electron pressure , therefore , the surface gravity of rgb stars . the spread in the metallicity distribution of m22 has been a controversial topic for many years . the recent several studies of the cluster have found that m22 has a bimodal heavy elemental abundance distribution ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? the high - resolution spectroscopic elemental abundance measurements of rgb stars in the peculiar gc m22 by @xcite and @xcite showed a distinctive bimodal metallicity distribution . their results were based on the spectroscopic @xmath6 and @xmath7 , which require the excitation and ionization equilibria , @xmath8 and @xmath9 where @xmath10 and @xmath11 are iron abundances from and lines and @xmath12 is the excitation potential . it has been frequently suspected that the iron abundance from the line suffers from non lte ( nlte ) effects ( see , for example , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? since metal - poor stars have much weaker metal absorption in the ultraviolet , more non - local ultraviolet flux can penetrate from the deeper layers . this flux is vital in determining the ionization equilibrium of the atoms , resulting in deviations from lte @xcite . in this regard , the traditional spectroscopic surface gravity determination method may be in error , which led @xcite to propose to use the photometric gravity for elemental abundance study of rgb stars in gcs , where the bolometric correction could be the dominant source of uncertainty for the photometric gravity . in principle , the use of the [ fe / h]@xmath13 abundance as for the metallicity scale of rgb stars in gcs is most likely an appropriate approach since is by far the dominant species and , therefore , the number of atoms is unaffected by the nlte effect . in practice , however , the [ fe / h]@xmath13 abundances of rgb stars sensitively depends not only on the surface gravity and effective temperature but also on the metallicity of the input atmosphere model , which also affects the continuum opacity . we will show later that an iterative procedure is useful to reduce the error raised by the incorrect metallicity of the input atmosphere model . in their recent study , ( * ? ? ? * mu15 hereafter ) re - analyzed m22 rgb stars of @xcite using three different approaches : ( 1 ) spectroscopic @xmath6 and @xmath7 ( method 1 ) ; ( 2 ) spectroscopic @xmath6 and photometric @xmath7 ( method 2 ) ; and ( 3 ) photometric @xmath6 and @xmath7 ( method 3 ) . they confirmed a bimodal iron abundance distribution of m22 rgb stars by @xcite , when they relied on spectroscopic @xmath6 and @xmath7 ( method 1 ) . oddly enough , when they used photometric @xmath7 ( method 2 and 3 by mu15 ) , the allegedly well - established bimodal iron distribution of m22 disappeared in [ fe / h]@xmath13 . using the photometric gravity is most likely a correct approach but how can this be interpreted ? we will show later that it is likely that mu15 used incorrect surface gravity and the metallicity of input atmosphere models in their analysis and the separation in [ fe / h]@xmath13 can be brought out more fully if different methods to compute these parameters are used . in this paper , we revisit the internal metallicity distribution of m22 . we developed new methods to estimate the surface gravity and we found that there exists substantial metallicity difference between the @xmath0-@xmath1 and @xmath0-@xmath2 groups-@xmath1 ( calcium - weak ) and the @xmath0-@xmath2 ( calcium - strong ) groups are defined to be rgb stars with smaller or larger @xmath14 index values , respectively , at a given @xmath15 magnitude . they are equivalent to the @xmath2-process - poor ( @xmath0-@xmath1 ) and the @xmath2-process - rich ( @xmath0-@xmath2 ) groups classified by @xcite . ] in m22 @xcite . throughout this paper , metallicity refers to [ fe / h]@xmath13 , unless specified . before turning to the metallicity distribution of m22 , we would like to revisit the critical issues on the lte analysis ; surface gravity , effective temperature and the metallicity of the input atmosphere model . as mentioned above , @xcite suggested to use [ fe / h]@xmath13 derived from photometric gravity as metallicity of rgb stars . they pointed out that [ fe / h]@xmath13 is essentially independent of nlte effects , such as overionization by non - local uv flux since is the dominant species in gc rgb stars . in figure [ fig : ki03 ] , we show plots of @xmath16[fe / h ] against [ fe / h]@xmath17 , [ fe / h]@xmath13 , @xmath6 and @xmath7 for the six group 1 clusters and @xmath7 based on colors and absolute magnitudes . ] of @xcite . of particular concern is figure [ fig : ki03 ] ( a ) and ( b ) , where each gc is showing its own correlation between [ fe / h]@xmath17 versus @xmath16[fe / h ] ( or [ fe / h]@xmath13 versus @xmath16[fe / h ] ) . in figure [ fig : fit ] , we show plots of @xmath7 versus @xmath6 , @xmath6 versus @xmath18 , and @xmath7 versus @xmath18 for the six group 1 gcs . also shown are victoria - regina model isochrones for 12 gyr @xcite . to estimate the @xmath19 level of the model isochrones , we used our previous relation @xcite , @xmath20 } + 1.5 ) + ( 0.52 \pm 0.13).\ ] ] note that this metallicity - luminosity relation for rr lyrae variables gives @xmath21 = 18.54 @xmath22 0.13 mag for lmc . it also should be emphasized that the adopted age of the model isochrones does not affect our results presented in this work . in plots of @xmath7 versus @xmath6 and @xmath6 versus @xmath18 , the loci of the model isochrones are in excellent agreement with the observations . it should not be a surprise that , at a given temperature , the surface gravity of the metal - poor stars is lower and @xmath18 is smaller than those of the metal - rich stars . this implies that applying a single @xmath6 versus @xmath7 relation for groups of stars with heterogeneous metallicity may result in incorrect surface gravity estimates and , as a consequence , incorrect elemental abundances . this is especially true when one adopt [ fe / h]@xmath13 as metallicity of gc rgb stars . the line opacity does not vary with surface gravity since the almost all iron atoms are populated in the first ionized level . on the other hand , the h@xmath5 continuum opacity is sensitively dependent on the electron pressure and , therefore , surface gravity . as a consequence , in the regime of the linear part of the curve of growth , at fixed metallicity , the equivalent widths ( ews ) of lines increases with decreasing surface gravity . in the figure , the surface gravity dependency on the metallicity of the metal - poor rgb stars at a fixed temperature is given by @xmath23}}{\partial \log g } \approx 1.8\ { \rm dex}.\ ] ] in other words , a change in the surface gravity by 0.1 in the cgs unit corresponds to a change in metallicity by about 0.2 dex in the figure . on the other hand , a rather tight relation in @xmath7 versus @xmath18 can be found , suggesting that the @xmath18 magnitude can be a reddening- and distance - independent surface gravity indicator for rgb stars in gcs . ( see also , * ? ? ? * for the use of @xmath24 as a temperature indicator of heavily reddened metal - rich gc palomar 6 ) . but caution is advised that the observed @xmath18 magnitude can be vulnerable to the foreground differential reddening effect . as mentioned previously , @xcite suggested to use the photometric surface gravity . however , ironically , under the condition that lte is not valid , they concluded to use spectroscopic temperature from lines . we would like to discuss the utility of the spectroscopic effective temperature under the assumption that the conclusion made by @xcite is valid . in figure [ fig : log ] , we show plots of the @xmath10 against the excitation potential for ngc 6752-mg10 by @xcite . as they noted , the quality of their spectra is superb ; a resolving power of @xmath25 = 110,000 and s / n @xmath26 150 per pixel near 5140 . to perform a lte abundance analysis , we used the 2014 version of the stellar line analysis program moog @xcite and we interpolated @xmath27-enhanced kurucz atmosphere models with new opacity distribution functions using a fortran program kindly provided by dr . mcwilliam ( 2005 , private communication ) . using the weak lines , @xmath28 ) @xmath29 , we derived a spectroscopic temperature for this star by forcing the condition of the excitation equilibrium , i.e. @xmath30@xmath10/@xmath31 = 0 , and we obtained the effective temperature of 4275 k. note that our effective temperature for this star is slightly cooler than those from the line - by - line differential analysis with respect to ngc 6752-mg9 and ngc 6752-mg6 by @xcite , 4291 k and 4295 k , respectively . we derived the linear fits to the data , assuming @xmath10 @xmath32 slope@xmath33 , and we show them with thin dashed lines in the figure . in the parentheses in each panel , we also show the @xmath7 value in the cgs unit , the slope in the excitation potential versus the iron abundance , [ fe / h]@xmath17 and [ fe / h]@xmath13 . as can be seen , once the spectroscopic effective temperature is correctly determined , the assumption of the excitation equilibrium holds for rather wide range of the surface gravity , @xmath34 1.20 in this case . therefore , it can be concluded that the slightly incorrect input surface gravity does not significantly affect the spectroscopic effective temperature in our results presented here . it is well - known fact that the inferred metallicity of gc rgb stars from high resolution spectroscopy is critically dependent on the stellar parameters of the input atmosphere model . here , we show how [ fe / h]@xmath17 and [ fe / h]@xmath13 behave against the changes in the input stellar parameters . we also would like to demonstrate the importance of the iterative derivation of the metallicity . in figure [ fig : fehvspam ] , we show [ fe / h]@xmath17 and [ fe / h]@xmath13 of 9 rgb stars in ngc 6752 by @xcite . in each panel , the red crosses denote our spectroscopic [ fe / h]@xmath17 and [ fe / h]@xmath13 which were derived using the weak and lines measured by @xcite . it should be noted that our [ fe / h ] values are consistent with those of @xcite . using our spectroscopic @xmath6 , @xmath7 , and [ fe / h ] as reference grids , we examine how the changes in the stellar parameters of the input atmosphere models affect the resultant metallicity . we run moog using the kurucz atmosphere models , whose stellar parameters are different from the reference grids by @xmath16@xmath6 = @xmath22200 k , @xmath16[fe / h ] = @xmath221.0 , and @xmath35 = @xmath220.3 . we use the [ fe / h]@xmath13 abundances returned from moog as the reference metallicity and we calculate new atmosphere models , with which we run moog again . we iterate this process five times and we show our results in figure [ fig : fehvspam ] with blue solid lines . the figure shows that the inferred [ fe / h]@xmath17 abundances with sufficient numbers of iterations are only affected by the changes in the effective temperature . as shown , [ fe / h]@xmath17 abundances of individual stars against the changes in the metallicity and the surface gravity of the input model atmosphere converge to their spectroscopic [ fe / h]@xmath17 values within 2 or 3 iterations . however , the iterations with the effective temperature offsets of @xmath22200k fail to regain spectroscopic [ fe / h]@xmath17 , implying incorrect @xmath6 estimate results in irrecoverable deviations in the derived metallicity . the [ fe / h]@xmath13 behaves differently against the changes in input stellar parameters . the inferred [ fe / h]@xmath13 abundances with the effective temperature offsets shifted in the opposite direction of the changes in the [ fe / h]@xmath17 abundance . a rather simple explanation of the temperature effect in cool stars can be found in @xcite , for example . similar to [ fe / h]@xmath17 , the spectroscopic metallicity can be regained with the iterative derivation of the [ fe / h]@xmath13 abundances against the changes in the metallicity of the input atmosphere model but , however , the difference in [ fe / h]@xmath13 between the spectroscopic metallicity and that from the first iteration is much larger than that can be seen in [ fe / h]@xmath17 . as we have already discussed earlier , the [ fe / h]@xmath13 abundance sensitively depends on the surface gravity of the input atmosphere model . the figure shows that the spectroscopic metallicity can not be regained with incorrect surface gravity . we also performed the same procedures using the [ fe / h]@xmath17 abundances of individual stars as reference metallicity to calculate the input atmosphere model used in the next iteration and we obtained the same results shown above . figure [ fig : delfeh ] summarize our exercise . for both [ fe / h]@xmath17 and [ fe / h]@xmath13 , the inferred metallicity from the first iteration with offsets in the stellar parameters could be very different from those with correct stellar parameters . however , some discrepancies vanish after 2 or 3 iteration processes . our exercise demonstrates that , with the iterative procedure , the [ fe / h]@xmath17 abundance depends only on the effective temperature , while the [ fe / h]@xmath13 abundance depends both on the effective temperature and the surface gravity . it also shows the importance of having correct stellar parameters , especially for the lte analysis of [ fe / h]@xmath13 abundances , as we have discussed earlier . the vivid evidence of the bimodal metallicity distribution of rgb stars in m22 from narrow - band photometry and high - resolution spectroscopy can be found in @xcite , @xcite and @xcite . in our earlier study of the cluster , we extensively discussed that there are several observational lines of evidence that can not be easily explained without invoking a bimodal metallicity distribution between two groups of stars , namely the @xmath0-@xmath1 and the @xmath0-@xmath2 groups as shown in figure [ fig : m22cmd ] ( e.g. , see figures 9 , 13 21 of * ? ? ? * and references therein ) ; * the h&k absorption strengths rgb stars at a given @xmath15 magnitude in m22 from both narrow - band photometry @xcite and low - resolution spectroscopy @xcite show a bimodal distribution . * the infrared triplet by @xcite also show a bimodal distribution among rgb stars in m22 . * the @xmath36 versus @xmath15 cmd as shown in figure [ fig : m22cmd ] ( see also figure 19 of * ? ? ? * ) also requires a bimodal metallicity distribution in m22 rgb stars . the variation in the lighter elements only , such as cno , can not explain this distinct double @xmath36 rgb sequences of m22 . the differential foreground reddening effect can not reproduce the observed multi - color cmds accordingly @xcite . * the @xmath15 magnitude of the rgb bump , @xmath37 , of the @xmath0-@xmath2 group is significantly fainter than that of the @xmath0-@xmath1 group , which strongly suggests that the @xmath0-@xmath2 group is more metal - rich than the @xmath0-@xmath1 group is . the difference in the @xmath37 between the two groups can not be explained by the differential foreground reddening effect @xcite . * the slope of the @xmath0-@xmath2 rgb stars in the @xmath38 versus @xmath15 cmd is significantly larger than that of the @xmath0-@xmath1 rgb stars , indicative of the metal - rich nature of the @xmath0-@xmath2 group @xcite . * the cn - ch positive correlation superposed on two separate cn - ch anticorrelations @xcite can be expected naturally if m22 is composed of two groups of stars with heterogeneous metallicities @xcite . * finally , another piece of evidence can be found in the metallicity distribution of the blue horizontal branch ( bhb ) stars in m22 @xcite . in figure [ fig : hb ] , we show the metallicity distributions for m22 bhb stars . both the lte and the nlte treatments show the similar degree of metallicity spreads in [ fe / h]@xmath17 and [ fe / h]@xmath13(six stars for [ fe / h]@xmath17 and seven stars for [ fe / h]@xmath13 ) . as shown in table 3 of @xcite , their derived [ fe / h]@xmath13 value is less sensitive to the effective temperature and to the surface gravity , @xmath16[fe / h]@xmath13= @xmath220.06 dex and @xmath220.01 dex for @xmath16@xmath6 = @xmath22170 k and @xmath39 = @xmath220.20 respectively , and we are likely seeing the real metallicity spread in m22 bhb stars . as mentioned above , mu15 re - analyzed m22 rgb stars of @xcite using three different approaches . we show 17 rgb and asymptotic giant branch ( agb ) stars studied by mu15 in figure [ fig : m22cmd ] . among 12 rgb stars tagged by mu15 , six rgb stars belong to each of two rgb groups , according to our previous classification of rgb stars based on the @xmath14 index at a given @xmath15 magnitude . figure [ fig : pam ] shows a plot of the @xmath40 @xmath6 versus @xmath7 of spectroscopic target stars , where the filled symbols are for method 1 ( spectroscopic @xmath6 and @xmath7 ) and the open symbols are for method 2 ( spectroscopic @xmath6 and photometric @xmath7 ) of mu15 . also shown are victoria - regina model isochrones for 12 gyr with [ fe / h ] = @xmath411.40 , @xmath411.70 and @xmath412.00 dex @xcite . the most conspicuous feature of the figure is that the positions of stars from method 2 by mu15 are rather well aligned in a narrow strip between model isochrones with [ fe / h ] = @xmath411.40 and @xmath411.70 dex on the @xmath6 @xmath7 plane , which is most likely due to the adopted _ single _ @xmath6 @xmath7 relation by mu15 . it should be worth to point out that , because the cmd of mu15 has a single rgb sequence , their photometric effective temperature and surface gravity will also define a single isochrone . the broad band colors adopted by them are not sensitive to small [ fe / h ] differences . on the other hand , positions of stars from method 1 occupy rather wide ranges in surface gravity at given effective temperature . we performed a lte abundance analysis using ews of m22 rgb stars measured by mu15 to calculate the the mean iron abundance dependence on model atmospheres . in table [ tab : dep ] , we show our result for eight rgb stars in m22 with 0.5 @xmath42 1.5 , where @xmath43 is the photometric surface gravity in the cgs unit . as shown in the table , the change in the surface gravity by @xmath44 0.2 results in @xmath16[fe / h]@xmath13 @xmath45 0.1 dex at the fixed @xmath6 , in the sense that the [ fe / h]@xmath13 value increases with the surface gravity .. ] if we take this at face value , a single @xmath6 @xmath7 relation , which is valid only for mono - metallic stellar systems such as normal gcs in our galaxy , may be responsible for the narrow uni - modal [ fe / h]@xmath13 distribution of m22 rgb stars as claimed by mu15 as shown in their figure 2 . in figure [ fig : method2 ] , we show the metallicity distribution of m22 rgb stars using table 4 of mu15 ( method 2 ) . note again that we use 8 rgb stars with 0.5 @xmath42 1.5 only ( corresponding to 11.5 @xmath46 12.75 mag in figure [ fig : m22cmd ] ) to avoid the potential effect raised by very different surface gravity in the stellar atmosphere model calculations . according to our population classification scheme for m22 based on our @xmath14 index at a given @xmath15 magnitude , stars 61 , 71 , 200068 and 200076 belong to the @xmath0-@xmath1 group , while stars 51 , 88 , 200025 and 200101 belong to the @xmath0-@xmath2 group . figure [ fig : method2 ] ( a ) shows that each rgb group has different mean iron abundance both in [ fe / h]@xmath17 and [ fe / h]@xmath13 . it is very interesting to note that the differences in the mean iron abundances between the two groups are larger than a 2.5 @xmath47 level both in [ fe / h]@xmath17 and [ fe / h]@xmath13 , although the separation in the mean [ fe / h]@xmath13 values is as small as @xmath45 0.05 dex . in table [ tab : ks ] , we show our results ( mu15 m2 ) . figure [ fig : method2 ] ( b ) ( e ) show the cumulative metallicity distributions and the generalized histograms of the metallicity distributions for each group . as shown , the [ fe / h]@xmath17 distribution show two distinct peaks , while that for [ fe / h]@xmath13 shows a broad single peak in the overall metallicity distribution , similar to those obtained by mu15 . but it should be emphasized that the [ fe / h]@xmath13 distribution by mu15 is composed of two separate mono - metallic distributions ; only the mean [ fe / h]@xmath13 values from the method 2 by mu15 for each group of stars happen to be similar . we performed a student s @xmath48-test to see if the metallicity distributions of the two groups of stars are identical . we obtained that the significance levels to reject the hypothesis that the mean [ fe / h ] values of the @xmath0-@xmath1 and the @xmath0-@xmath2 groups are identical , are 1.93 % and 8.98 % for [ fe / h]@xmath17 and [ fe / h]@xmath13 , respectively . we also performed a randomization test . the significance levels to reject the hypothesis for being an identical metallicity distribution from bootstrap method are 0.00 % for both [ fe / h]@xmath17 and [ fe / h]@xmath13 , strongly suggest that the metallicity distributions for the @xmath0-@xmath1 and the @xmath0-@xmath2 groups by mu15 are not statistically identical . it should not be a surprise because @xcite already discussed many aspects of heterogeneous nature between the two groups as summarized in [ sec3 ] . figure [ fig : method2 ] ( f g ) shows @xmath16[fe / h ] (= [ fe / h]@xmath13 @xmath41 [ fe / h]@xmath17 ) against [ fe / h]@xmath17 and [ fe / h]@xmath13 and they are very intriguing . it should be reminded that we chose stars with a narrow range of @xmath15 magnitude for both groups to avoid the potential effect raised by the very different surface gravity . as shown in the figure , the discrepancies between [ fe / h]@xmath17 and [ fe / h]@xmath13 of the @xmath0-@xmath1 group rgb stars are preferentially much greater than those of the @xmath0-@xmath2 group , reaching as large as @xmath16[fe / h ] @xmath45 0.4 dex . for comparison , we show @xmath16[fe / h ] of m22 along with six group 1 gcs by @xcite in figure [ fig : ki03 ] , where the much greater @xmath16[fe / h ] values of the @xmath0-@xmath1 rgb stars in m22 can be clearly seen . if the two groups of stars have the same metallicity and the same surface gravity so that they suffer similar degree of nlte effect , one would expect to see the similar degree of @xmath16[fe / h ] for both groups , in sharp contrast to the results of method 2 by mu15 . we will show later that , with a mock peculiar gc , the incorrect surface gravity estimate by mu15 and the incorrect metallicity of input model atmospheres are responsible for the discrepancy in @xmath16[fe / h ] . another aspect needs to consider is the initial metallicity used during the atmosphere model calculations . in figure [ fig : mu15iter ] , we show plots for eight m22 rgb stars , similar to figure [ fig : fehvspam ] for ngc 6752 . in parentheses of each panel , we show the method for the metallicity derivation by mu15 ( m1 or m2 ) and the reference metallicity for the iterations ( i for [ fe / h]@xmath17 and ii for [ fe / h]@xmath13 ) . in the figure , the crosses denote the mu15 s [ fe / h]@xmath17 and [ fe / h]@xmath13 values from m1 and m2 methods and blue and red colors denote the @xmath0-@xmath1 and @xmath0-@xmath2rgb stars , respectively . using [ fe / h]@xmath17 and [ fe / h]@xmath13 abundances as trial metallicities and @xmath6 and @xmath7 from m1 and m2 methods by mu15 , we performed the iterative derivations of [ fe / h]@xmath17 and [ fe / h]@xmath13 for individual stars . as shown , [ fe / h]@xmath17 values do not vary significantly with the number of iterations with respect to the spectroscopic [ fe / h ] , meaning that the effective temperature adopted by mu15 is correct . on the other hand , the discrepancy in [ fe / h]@xmath13 for @xmath0-@xmath2 rgb stars from method 2 are preferentially larger , indicating that the surface gravities of the @xmath0-@xmath2 rgb stars adopted by mu15 is most likely underestimated . the figure also indicates that the separation in the [ fe / h]@xmath13 abundances between the @xmath0-@xmath1 and the @xmath0-@xmath2 groups becomes larger with the iteration processes . finally , comparisons of ews between the two groups may also help to elucidate the underlying metallicity distributions of the cluster . in figure [ fig : ew](a ) , we show the line - by - line ew difference between stars 88 ( @xmath0-@xmath2 ) and 200076 ( @xmath0-@xmath1 ) . both stars have similar visual magnitudes and colors , ( @xmath15 , @xmath49 ) = ( 12.54 , 0.89 ) for the star 88 and ( 12,39 , 0.90 ) for the star 200076 . therefore , if there is no significant foreground differential reddening and if the both stars have the same metallicity , they should have similar @xmath6 , @xmath7 and furthermore the similar ew strengths . as shown in the figure , the ews of the @xmath0-@xmath2 group star 88 are systematically larger than those of the @xmath0-@xmath1 group stars 200076 . a comparison between the mean ews of the four @xmath0-@xmath2 group stars ( 51 , 88 , 200025 and 200101 ) and the four @xmath0-@xmath1 group stars ( 61 , 71 , 200068 and 200076 ) show the same trend that the mean ews of the @xmath0-@xmath2 group are larger than those of the @xmath0-@xmath1 group , 12.0 @xmath22 0.4 m for lines and 1.8 @xmath22 0.4 m for lines . note that the eight stars above have similar visual magnitudes and colors and they should have similar @xmath6 , @xmath7 and [ fe / h ] if they belong to a single stellar population . a simple explanation why lines are less sensitive to changes in metallicity as follows . since both the fraction of atoms and the h@xmath5 continuum opacity of rgb stars depend on the electron pressure ( i.e. , metallicity or surface gravity ) and , furthermore , the two effects are expected to cancel out , the ews of lines grow with metallicity at fixed effective temperature . on the other hand , the atoms are the dominant species and only the electron pressure has an effect on the h@xmath5 continuum opacity , which has an opposite effect on the growth of ews with metallicity . therefore , the ews of lines grows at a slower rate . we suspect that the metallicity measurements by mu15 may be slightly incorrect and we devise new methods to derive metallicity of rgb stars in appropriate and consistent manners . in our previous study , @xcite showed that a combination of two normal gcs , m55 and ngc 6752 , can reproduce many aspects of peculiar photometric characteristics of m22 . in figure [ fig : m55n6752cmd ] , we show a composite cmds for m55 and ngc 6752 , which may highlight the importance of the choice of photometric passbands to distinguish multiple stellar populations in gcs . for ngc 6752 stars , we add offsets of 0.030 , @xmath410.010 , @xmath410.005 , and 0.700 mag in ( @xmath49 ) , @xmath36 , @xmath14 , , @xmath50 , and @xmath51 . ] and @xmath15 , respectively , in order to place ngc 6752 stars in the m55 colors and magnitude scale . in the figure , we also show rgb stars studied from high - resolution spectroscopy of the clusters @xcite and the victoria - regina model isochrones for 12 gyr with [ fe / h ] = @xmath411.84 and @xmath411.53 . despite the difference in metallicity , the rgb sequence of ngc 6752 is in excellent agreement with that of m55 in @xmath52 versus @xmath15 cmds , and it is difficult to discern different populations . in such a case , one can be easily misled to adopt a single @xmath6 @xmath7 relation for heterogeneous stellar populations . on the other hand , the distinct double rgb sequences can be clearly seen in @xmath36 versus @xmath15 and @xmath14 versus @xmath15 cmds , where the necessity for double @xmath6 @xmath7 relations is obvious . first , we derived the metallicity distributions of m55 and ngc 6752 by employing two separate @xmath6 @xmath7 relations , following the procedure recommended by @xcite . with the ew measurements by @xcite for and lines , we performed a lte abundance analysis . applying the color - temperature relation and the equation for the bolometric correction given by @xcite , we calculated the photometric effective temperature and the surface gravity using our own strmgren photometry . during our calculations , we adopted [ fe / h ] = @xmath411.95 and @xmath411.55 as the input metallicities for m55 and ngc 6752 , respectively @xcite , and we used the distance moduli and foreground reddening values by @xcite . during our analysis , we used weak lines only , @xmath28 ) @xmath29 , for both and in order to minimize the effect of the adopted micro - turbulent velocity on the derived metallicity . we show our results in table [ tab : mock ] and figure [ fig : photorelation ] . our [ fe / h]@xmath13 measurements for the clusters are in good agreement with those by @xcite , @xmath16[fe / h]@xmath13 = @xmath410.01 @xmath22 0.03 dex for m55 and @xmath16[fe / h]@xmath13 = 0.05 @xmath22 0.02 dex for ngc 6752 in the sense of current work minus those of @xcite . these small differences are thought to be mainly due to slightly different final color - temperature relations . @xcite used the @xmath53 color - temperature relations given by @xcite to derive the initial @xmath6 but they applied their own @xmath6 versus @xmath15 magnitude relation to derive their final adopted @xmath6 . in figure [ fig : photorelation ] ( b ) and ( c ) , we show @xmath16[fe / h ] (= [ fe / h]@xmath13 @xmath41 [ fe / h]@xmath17 ) against [ fe / h]@xmath17 and [ fe / h]@xmath13 , where the extent of @xmath16[fe / h ] for both clusters are in good agreement with those of six gcs studied by @xcite as shown in figure [ fig : ki03 ] . figure [ fig : photorelation ] ( d ) shows a plot of @xmath6 versus @xmath7 along with the victoria - regina model isochrones for 12 gyr with [ fe / h ] = @xmath411.84 and @xmath411.55 @xcite . at a given @xmath6 , the stars in the metal - poor cluster m55 have lower surface gravity , although the mean difference in @xmath7 between m55 and ngc 6752 is not as large as that can be inferred from model isochrones . using separate relations , the differences in the mean [ fe / h]@xmath17 and [ fe / h]@xmath13are 0.44 @xmath22 0.02 and 0.47 @xmath22 0.03 , respectively , and they are in excellent agreement . following the same procedure described earlier , we performed iterative derivations of [ fe / h]@xmath17 and [ fe / h]@xmath13 for the clusters . as shown in figure [ fig : photorelation ] and table [ tab : mock ] , [ fe / h]@xmath17 and [ fe / h]@xmath13 from iterative processes are in good agreement with those without the iterative process . we also make use of the ew measurements for the and lines for ngc 6752 rgb stars by @xcite . we show our results in figure [ fig : m55n6752comp2 ] and table [ tab : mock ] . note that the oscillator strengths for the individual lines are slightly different between those adopted by @xcite and @xcite . for our results presented in figure [ fig : m55n6752comp2 ] , we used the oscillator strengths by @xcite because they provided more lines . as summarized in table [ tab : mock ] , [ fe / h]@xmath17 and [ fe / h]@xmath13 abundances using @xmath54-values by @xcite and @xcite are slightly different , but the differences in the mean values are no larger than 0.04 dex . therefore , it can be said that the choice of the set of oscillator strengths does not affect our primary results presented here . the differences in [ fe / h]@xmath17 ( @xmath16[fe / h]@xmath17 = [ fe / h]@xmath17@xmath55 @xmath41 [ fe / h]@xmath17@xmath56 ) and [ fe / h]@xmath13(@xmath16[fe / h]@xmath13 = [ fe / h]@xmath13@xmath55 @xmath41 [ fe / h]@xmath13@xmath56 ) are 0.38 - 0.45 dex and 0.44 - 0.50 dex , respectively . assuming that our mock gc ( i.e. m55 + ngc 6752 ) is a mono - metallic gc with [ fe / h ] @xmath57 dex ( i.e. that of ngc 6752 ) , we calculated photometric effective temperatures and surface gravities of individual stars using the @xmath52 color - temperature relation and the equation for the bolometric correction given by @xcite . using the weak and lines only , @xmath28 ) @xmath58 , we derived the metallicity of individual stars in both clusters and we show our results in table [ tab : mock ] and figure [ fig : photorelation ] . the mean [ fe / h]@xmath17 abundance of m55 remains unchanged , while that of [ fe / h]@xmath13 increases almost 0.15 dex compared to the results from correct input stellar parameters for m55 presented in [ sec5.1 ] . the difference in the effective temperature between the two methods ( i.e. two separate relations versus a single relation for m55 and ngc 6752 ) is negligibly small , @xmath59 = 12 k , in the sense that the mean effective temperature of m55 rgb stars is slightly warmer when the correct @xmath52 color - temperature relation is used . as we have discussed earlier , the [ fe / h]@xmath17 abundance is relatively insensitive to changes in the surface gravity and the metallicity of the input model atmosphere since the effects due to change in the number of species and that in h@xmath5 continuum opacity are expected to cancel . however , the surface gravity becomes larger , @xmath39 @xmath45 0.1 , and the metallicity of the input model atmospheres is higher ( from @xmath411.95 to @xmath411.55 dex ) when a single relation with respect to ngc 6752 is used . both effects greatly enhance the h@xmath5 continuum opacity , while the fraction of to the total number of iron atoms is unaffected since is by far the dominant species . as a consequence , the mean [ fe / h]@xmath13 abundance of m55 appears to be enhanced at given ews . in figure [ fig : photorelation ] ( l ) , we show a plot of @xmath6 versus @xmath7 with a single relation , where rgb stars in both clusters are aligned well on a single locus . figure [ fig : photorelation ] ( j ) and ( k ) show @xmath16[fe / h ] against [ fe / h]@xmath17 and [ fe / h]@xmath13 showing large discrepancies in m55 rgb stars , reminiscent of the m22 @xmath0-@xmath1 stars from method 2 of mu15 as shown in figure [ fig : method2 ] . the metallicity distributions of [ fe / h]@xmath17 and [ fe / h]@xmath13 from a single @xmath6 @xmath7 relation shown in figure [ fig : m55n6752comp1 ] ( b ) and ( c ) are intriguing , since the two peaks in the [ fe / h]@xmath13 distribution become less conspicuous with a single relation with respect to ngc 6752 . our result with a mock gc strongly suggests that applying a single photometric relation in order to derive the effective temperatures and surface gravities of individual stars in peculiar gcs with heterogeneous metallicities and perhaps ages , such as m22 , may result in slightly incorrect metallicity scales and distributions . we also performed a traditional analysis using the spectroscopic effective temperatures and the surface gravities , which requires the excitation and ionization equilibria of iron abundances . our results are shown in figures [ fig : m55n6752comp1 ] ( i l ) and [ fig : m55n6752comp2 ] ( i l ) . as shown in table [ tab : mock ] , our mean spectroscopic [ fe / h ] values are in good agreement with those of @xcite and @xcite . the difference in metallicity between m55 and ngc 6752 is @xmath16[fe / h]@xmath13 = 0.32 0.42 dex , depending on the data sets . it should be mentioned that our spectroscopic [ fe / h ] value of ngc 6752 from @xcite is about 0.1 dex higher than that from @xcite , which is consistent with the iron abundances of [ fe / h ] = @xmath411.56 dex @xcite and @xmath411.65 dex @xcite for the cluster . the origin of this discrepancy of the mean metallicity of ngc 6752 is beyond the scope of this study and we decline to discuss this matter further . as shown in figure [ fig : fit ] , the model isochrones can provide a useful means to derive the stellar parameters . in figure [ fig:17gc ] , we show the similar plots for 17 gcs from the homogeneous elemental abundance study by @xcite . also shown are the victoria - regina isochrones for the age of 12 gyr and they appear to be in excellent agreement with observations . we devise a new strategy to derive evolutionary surface gravities of rgb stars in gcs under the assumption that the excitation equilibrium of lines is applicable in such stellar atmospheres and , furthermore , excitation equilibrium holds for rather wide range of the surface gravity . as shown in figure [ fig : fit ] and [ fig:17gc ] , at a given effective temperature , the surface gravity increases with metallicity , and , as a consequence , the metallicity , especially [ fe / h]@xmath13 , without the proper estimates of the surface gravity may not be correct . using the spectroscopic temperature and the [ fe / h]@xmath13 abundance from photometric method as initial input parameters , we interpolated the victoria - regina model isochrones to obtain the evolutionary surface gravity at the fixed effective temperature . then we derive the updated metallicity by running moog using the model atmosphere with the spectroscopic effective temperature and the evolutionary surface gravity in an iterative manner until the derived metallicity converged to within the internal measurement error between consecutive measurements , which usually requires 2 to 3 iterations . we show our new stellar parameters in figures [ fig : m55n6752comp1 ] ( p ) and [ fig : m55n6752comp2 ] ( p ) and metallicity distributions in figures [ fig : m55n6752comp1 ] ( n o ) and [ fig : m55n6752comp2 ] ( n o ) . the difference in metallicity between m55 and ngc 6752 becomes @xmath16[fe / h]@xmath13 = 0.44 0.56 dex and our results are shown in table [ tab : mock ] . we also note that using the photometric temperatures of individual stars does not change our results presented here . this approach should be reddening- and distance - independent and therefore it would be useful to derive surface gravity of gc stars with varying foreground reddening , such as rgb stars in m22 . as shown in figure [ fig : fit ] and [ fig:17gc ] , at a given metallicity , the @xmath15 magnitude differences from the hb , @xmath18 , of individual stars in gcs are well correlated with the surface gravities and the effective temperatures . similar to the previous approach , at a given @xmath18 and metallicity we determine the evolutionary surface gravity and effective temperature simultaneously by interpolating the victoria - regina model isochrones . for this purpose , we use our own photometry of the clusters and [ fe / h]@xmath13 derived from the photometric stellar parameters as an initial guess as have done previously . then we derive the updated metallicity by running moog using the model atmosphere with the evolutionary effective temperature and the surface gravity in an iterative manner until the derived metallicity converged to within the internal measurement error between consecutive measurements . we show our results in figures [ fig : m55n6752comp1 ] ( q t ) and [ fig : m55n6752comp2 ] ( q t ) . this approach provides similar results as those from the spectroscopic method and the method relying on the evolutionary surface gravity . the difference in metallicity between m55 and ngc 6752 becomes @xmath16[fe / h]@xmath13 = 0.44 0.52 dex as shown in table [ tab : mock ] . the merit of using @xmath18 is that it is also a reddening- and distance - independent parameter . however it can be vulnerable to the differential foreground reddening effect of the individual stars . it should be kept in mind that the main idea to deliver in our study is to demonstrate the importance of having appropriate stellar parameters for the lte abundance analysis in multiple stellar populations . following the same procedures as for m55 and ngc 6752 , we derive the iron abundances of the two groups of stars in m22 in four different manners . our results are consistent with the idea that the two groups of stars in m22 have different mean iron abundances as @xcite , @xcite and @xcite already showed . we derive the metallicity of m22 rgb stars based on the photometric effective temperature and surface gravity from our strmgren photometry of the cluster using the relations by @xcite . during our calculations , we adopted the apparent visual distance modulus of 13.60 mag , @xmath60 = 0.34 and [ fe / h ] = @xmath411.65 for m22 @xcite . as it was done before , we made use of the weak lines only , @xmath28 ) @xmath29 , for both and in order to minimize the effect of the adopted micro - turbulent velocity on the metallicity . in table [ tab : ks ] and figures [ fig : m22comb ] and [ fig : m22combdiff ] , we show our results . for non - differential analysis , the differences in the mean metallicity are @xmath16[fe / h]@xmath17 = 0.239 @xmath22 0.057 dex and @xmath16[fe / h]@xmath13 = 0.096 @xmath22 0.048 dex without iteration , and @xmath16[fe / h]@xmath17 = 0.233 @xmath22 0.048 dex and @xmath16[fe / h]@xmath13 = 0.108 @xmath22 0.052 dex after fifth iteration . note that our results are consistent with those from the method 2 by mu15 . in panels ( b ) and ( c ) of figures [ fig : m22comb ] and [ fig : m22combdiff ] , we show empirical distributions of the mean [ fe / h]@xmath17 and [ fe / h]@xmath13 for the two populations from the bootstrap method , strongly suggest that the metallicity distributions of the two groups of stars in m22 are different . as shown in table [ tab : ks ] , the significance levels to reject the hypothesis that the mean [ fe / h]@xmath17 values of the @xmath0-@xmath1 and the @xmath0-@xmath2 groups are identical are lower than 1% . however , those for [ fe / h]@xmath13 are rather large , @xmath45 12 % , for the non - differential analysis . it should not be confused that this rather large significance levels do not indicate that two groups of stars in m22 belong to the same population , but the lte analysis of the heterogeneous groups of star with a single @xmath6 relation may be in error . we also calculate the line - by - line differential iron abundances since the numbers of iron lines being measured by mu15 for individual stars are different . we selected the star 51 to be the reference star since its @xmath6 and @xmath7 are close to the average for the sample . also the stellar parameters for this star both from the photometric and spectroscopic methods by mu15 agree well as shown in figure [ fig : pam ] . for our differential abundance measurements , we did not adjust the stellar parameters of individual stars with respect to the reference star as have done by @xcite , for example , and we intended to calculate the proper metallicity offset differences among the sample stars with given stellar parameters . the differences in the mean metallicity from the differential analysis are @xmath16[fe / h]@xmath17 = 0.203 @xmath22 0.039 dex and @xmath16[fe / h]@xmath13 = 0.101 @xmath22 0.045 dex without iteration , and @xmath16[fe / h]@xmath17 = 0.220 @xmath22 0.043 dex and @xmath16[fe / h]@xmath13 = 0.129 @xmath22 0.051 dex with fifth iterations , consistent with those from non - differential analysis . as shown in table [ tab : ks ] , the separation of the mean [ fe / h]@xmath13 values between the @xmath0-@xmath1 and the @xmath0-@xmath2 groups is larger than 2.0 2.5 @xmath47 levels . we calculated the significance levels to reject the hypothesis that the mean [ fe / h ] values of the two groups of stars in m22 are identical . we obtained the significance levels lower than 1% for [ fe / h]@xmath17 and 7.5% for [ fe / h]@xmath13 , strongly suggesting that they are different . next , we derived the metallicity of individual stars based on the spectroscopic @xmath6 and @xmath7 . the differences in the mean metallicity between the two groups of stars are @xmath16[fe / h]@xmath17 = 0.203 @xmath22 0.050 dex and @xmath16[fe / h]@xmath13 = 0.204 @xmath22 0.052 dex for non - differential analysis and @xmath16[fe / h]@xmath17 = 0.194 @xmath22 0.044 dex and @xmath16[fe / h]@xmath13 = 0.228 @xmath22 0.053 dex for differential analysis , making the separation of the mean [ fe / h]@xmath13 values between the two groups larger than 3.9 @xmath47 to 4.3 @xmath47 levels . not surprisingly , our results are consistent with those obtained by @xcite , who relied on the traditional spectroscopic stellar parameters . as shown in table [ tab : ks ] , the significance levels to reject the hypothesis that the mean [ fe / h ] values of the two groups of stars in m22 are identical are very low , indicating that they are different . the metallicity based on the evolutionary stellar parameters also suggest that the metallicity distributions of each group of stars are indeed different . following the same procedure described in [ chap : fixedt ] , we obtained the differences in the mean metallicity of @xmath16[fe / h]@xmath17 = 0.224 @xmath22 0.061 dex and @xmath16[fe / h]@xmath13 = 0.168 @xmath22 0.066 dex for non - differential analysis and @xmath16[fe / h]@xmath17 = 0.191 @xmath22 0.043 dex and @xmath16[fe / h]@xmath13 = 0.172 @xmath22 0.060 dex for differential analysis . the mean [ fe / h]@xmath13 values of each group are different more than 2.5 @xmath47 to 2.9 @xmath47 levels and the low significance levels of being identical distributions also confirm that they are different . similar conclusion can be drawn when we use the evolutionary @xmath7 and @xmath6 at a given @xmath18 , where we used @xmath19 = 14.15 mag for m22 @xcite . following the same procedure described in [ chap : fixedvvhb ] , we obtained the differences in the mean metallicity of @xmath16[fe / h]@xmath17 = 0.216 @xmath22 0.047 dex and @xmath16[fe / h]@xmath13 = 0.128 @xmath22 0.062 dex for non - differential analysis and @xmath16[fe / h]@xmath17 = 0.181 @xmath22 0.034 dex and @xmath16[fe / h]@xmath13 = 0.132 @xmath22 0.056 dex for differential analysis . similar to the results shown above , the mean [ fe / h]@xmath13 values of each group are different more than 2.1 @xmath47 to 2.4 @xmath47 levels and the low significance levels of being identical distributions also confirm that they are different . as shown in figures [ fig : m22comb ] and [ fig : m22combdiff ] , it should be emphasized that the substructures not only in the [ fe / h]@xmath17 but also in the [ fe / h]@xmath13 distributions are notable , indicating that m22 contains multiple stellar populations with heterogeneous metallicities . the precision elemental abundance measurement of individual stars in ggs is not a trivial task , especially for a peculiar gc with multiple stellar populations with heterogeneous metallicity distributions . in the context of lte analysis , our demonstrations with a mock peculiar gc composed of the two normal gcs , ngc 6752 and m55 , showed that the internal absolute and the relative metallicity scales are vulnerable to the incorrect treatment of the input stellar parameters among multiple stellar populations with different metallicity . in particular , photometric surface gravity without taking care of proper metallicity effect can result in the [ fe / h]@xmath13 measurement error of as large as 0.1 0.2 dex . as we have discussed earlier , this is because the line opacity does not vary with surface gravity since the almost all iron atoms are populated in the first ionized level , while the h@xmath5 continuum opacity is sensitively dependent on the electron pressure and , therefore , surface gravity . as a consequence , changes in surface gravity can mimic the [ fe / h]@xmath13 abundance of rgb stars in gcs . in this regard , we developed methods independent of the traditional spectroscopic analysis approach , which is demanding excitation and ionization equilibria in and elements , to make use of the evolutionary surface gravity . the metallicity scales from these new approaches are in good agreement with those of previous studies by others @xcite . from our study of a mock peculiar gc , it is worth to mention three comments concerning the metallicity scale of the multiple stellar populations in a gc . first , the use of narrow band photometry , such as @xmath36 and @xmath14 which are sensitive to metallicity and less sensitive to interstellar reddening , is beneficial to discern small [ fe / h ] differences . second , our results show that our adaptive methods to estimate appropriate surface gravity would be essential in deriving the absolute and the relative [ fe / h]@xmath13 scale for the multiple stellar populations in peculiar gcs . third , it is very interesting to note that our new methods appear to provide similar metallicity scale as that from the traditional spectroscopic analysis , suggesting that , the metallicity scale from the widely used traditional spectroscopic approach which makes use of excitation and ionization equilibria in and elements , is valid , at least in the relative sense . contrary to the conclusion made by mu15 , our re - examination of m22 rgb stars showed that the peculiar gc m22 is composed of two groups of stars with heterogeneous metallicities , confirming our previous results and those of others @xcite , and the m22 saga will continue . this work has been supported by the center for galaxy evolution research ( grant no . 2010 - 0027910 ) and the basic science research program ( grant no . 2016r1a2b4014741 ) through the national research foundation of korea ( nrf ) funded by the korea government ( msip ) . lee thanks eugenio carretta for kindly providing ew measurements for ngc 6752 and m55 rgb stars and david yong and donghoh kim for their helpful discussions . he also appreciates the anonymous referee for critical reading and for providing helpful suggestions that greatly improved the manuscript . marino , a. f. , milone , a. p. , sneden , c. , bergemann , m. , kraft , r. p. , wallerstein , g. , cassisi , s. , aparicio , a. , asplund , m. , bedin , r. l. , hilker , m. , lind , k. , momany , y. , piotto , g. , roederer , i. u. , stetson , p. b. , & zoccali , m. 2012 , , 541 , 15 marino , a. f. , sneden , c. , kraft , r. p. , wallerstein , g. , norris , j. e. , da costa , g. , milone , a. p. , ivans , i. i. , gonzalez , g. , fulbright , j. p. , hilker , m. , piotto , g. , zoccali , m. , & stetson , p. b. 2011 , , 532 , a8 crrrrr [ fe / h]@xmath17 & 0.044 @xmath22 0.003 & @xmath410.044 @xmath22 0.003 & & @xmath410.007 @xmath22 0.002 & 0.000 @xmath22 0.001 + [ fe / h]@xmath13 & @xmath410.037 @xmath22 0.006 & 0.036 @xmath22 0.005 & & 0.090 @xmath22 0.004 & @xmath410.084 @xmath22 0.004 + crrrrrrrrrr mu15 m2 & 0.232 @xmath22 0.066 & 3.52 & 1.93 & 0.00 & & 0.053 @xmath22 0.022 & 2.43 & 8.98 & 0.00 + mu15 m2 ( 5@xmath61 iter . ) & 0.217 @xmath22 0.076 & 2.84 & 3.02 & 3.14 & & 0.090 @xmath22 0.041 & 2.20 & 7.24 & 5.57 + @xmath6 + @xmath7 ( a99 , no iter . ) & 0.239 @xmath22 0.057 & 4.20 & 0.76 & 0.00 & & 0.096 @xmath22 0.048 & 2.01 & 9.11 & 11.27 + @xmath6 + @xmath7 ( a99 , 5@xmath61 iter . ) & 0.233 @xmath22 0.048 & 4.24 & 0.72 & 0.00 & & 0.108 @xmath22 0.052 & 1.80 & 12.32 & 11.42 + spectroscopic & 0.203 @xmath22 0.050 & 3.48 & 1.32 & 0.00 & & 0.204 @xmath22 0.052 & 3.38 & 1.49 & 3.02 + fixed @xmath6 & 0.224 @xmath22 0.061 & 3.67 & 1.04 & 0.00 & & 0.168 @xmath22 0.066 & 2.54 & 5.44 & 2.85 + fixed @xmath18 & 0.216 @xmath22 0.047 & 4.57 & 0.52 & 0.00 & & 0.128 @xmath22 0.062 & 2.08 & 8.56 & 8.46 + & & & & & & & & & + + mu15 m2 & 0.213 @xmath22 0.064 & 3.30 & 1.92 & 0.00 & & 0.077 @xmath22 0.033 & 2.35 & 8.01 & 2.89 + mu15 m2 ( 5@xmath61 iter . ) & 0.205 @xmath22 0.069 & 2.96 & 2.60 & 2.71 & & 0.112 @xmath22 0.036 & 3.08 & 3.10 & 0.00 + @xmath6 + @xmath7 ( a99 , no iter . ) & 0.203 @xmath22 0.039 & 5.23 & 0.40 & 0.00 & & 0.101 @xmath22 0.045 & 2.25 & 6.66 & 5.55 + @xmath6 + @xmath7 ( a99 , 5@xmath61 iter . ) & 0.220 @xmath22 0.043 & 4.19 & 0.90 & 0.00 & & 0.129 @xmath22 0.051 & 2.19 & 7.33 & 5.74 + spectroscopic & 0.194 @xmath22 0.044 & 3.79 & 0.92 & 0.00 & & 0.228 @xmath22 0.053 & 3.74 & 1.09 & 0.00 + fixed @xmath6 & 0.191 @xmath22 0.043 & 4.43 & 0.45 & 0.00 & & 0.172 @xmath22 0.060 & 2.86 & 3.42 & 2.86 + fixed @xmath18 & 0.181 @xmath22 0.034 & 5.36 & 0.41 & 0.00 & & 0.132 @xmath22 0.056 & 2.36 & 5.67 & 5.61 + cccccccccc phot . & c09 ( no ) & @xmath412.087 @xmath22 0.018 & @xmath411.936 @xmath22 0.020 & & @xmath411.645 @xmath22 0.014 & @xmath411.471 @xmath22 0.015 & & 0.443 @xmath22 0.023 & 0.465 @xmath22 0.025 + & c09 ( 5@xmath61 ) & @xmath412.090 @xmath22 0.016 & @xmath411.927 @xmath22 0.028 & & @xmath411.642 @xmath22 0.014 & @xmath411.432 @xmath22 0.024 & & 0.448 @xmath22 0.021 & 0.496 @xmath22 0.037 + & y13 ( no ) & @xmath412.067 @xmath22 0.016 & @xmath411.896 @xmath22 0.021 & & @xmath411.687 @xmath22 0.005 & @xmath411.463 @xmath22 0.006 & & 0.379 @xmath22 0.017 & 0.443 @xmath22 0.022 + & y13 ( 5@xmath61 ) & @xmath412.073 @xmath22 0.014 & @xmath411.869 @xmath22 0.029 & & @xmath411.683 @xmath22 0.005 & @xmath411.414 @xmath22 0.010 & & 0.390 @xmath22 0.015 & 0.454 @xmath22 0.031 + & & & & & & & & & + phot . & c09 ( no ) & @xmath412.114 @xmath22 0.017 & @xmath411.789 @xmath22 0.022 & & @xmath411.645 @xmath22 0.014 & @xmath411.471 @xmath22 0.015 & & 0.469 @xmath22 0.022 & 0.318 @xmath22 0.027 + & c09 ( 5@xmath61 ) & @xmath412.110 @xmath22 0.016 & @xmath411.883 @xmath22 0.030 & & @xmath411.642 @xmath22 0.014 & @xmath411.432 @xmath22 0.024 & & 0.469 @xmath22 0.021 & 0.451 @xmath22 0.038 + & y13 ( no ) & @xmath412.093 @xmath22 0.016 & @xmath411.748 @xmath22 0.022 & & @xmath411.687 @xmath22 0.005 & @xmath411.463 @xmath22 0.006 & & 0.406 @xmath22 0.017 & 0.285 @xmath22 0.023 + & y13 ( 5@xmath61 ) & @xmath412.092 @xmath22 0.015 & @xmath411.825 @xmath22 0.031 & & @xmath411.683 @xmath22 0.005 & @xmath411.414 @xmath22 0.010 & & 0.409 @xmath22 0.016 & 0.411 @xmath22 0.032 + & & & & & & & & & + spec . & c09 & @xmath412.003 @xmath22 0.020 & @xmath412.002 @xmath22 0.019 & & @xmath411.583 @xmath22 0.027 & @xmath411.584 @xmath22 0.025 & & 0.420 @xmath22 0.033 & 0.418 @xmath22 0.031 + & y13 & @xmath411.952 @xmath22 0.014 & @xmath411.956 @xmath22 0.017 & & @xmath411.657 @xmath22 0.003 & @xmath411.635 @xmath22 0.004 & & 0.295 @xmath22 0.014 & 0.321 @xmath22 0.017 + & & & & & & & & & + fixed @xmath6 & c09 & @xmath412.001 @xmath22 0.019 & @xmath412.026 @xmath22 0.030 & & @xmath411.579 @xmath22 0.025 & @xmath411.455 @xmath22 0.025 & & 0.422 @xmath22 0.031 & 0.571 @xmath22 0.039 + & y13 & @xmath411.966 @xmath22 0.015 & @xmath411.890 @xmath22 0.028 & & @xmath411.655 @xmath22 0.003 & @xmath411.463 @xmath22 0.010 & & 0.311 @xmath22 0.015 & 0.426 @xmath22 0.030 + & & & & & & & & & + fixed @xmath18 & c09 & @xmath411.999 @xmath22 0.009 & @xmath412.006 @xmath22 0.027 & & @xmath411.651 @xmath22 0.011 & @xmath411.467 @xmath22 0.021 & & 0.348 @xmath22 0.014 & 0.539 @xmath22 0.034 + & y13 & @xmath412.005 @xmath22 0.008 & @xmath411.943 @xmath22 0.029 & & @xmath411.690 @xmath22 0.004 & @xmath411.465 @xmath22 0.010 & & 0.315 @xmath22 0.009 & 0.479 @xmath22 0.031 +

recent observational surveys ( e.g. , cosmos , goods , gems , sings , sdss , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) have allowed us for the first time to statistically explore issues of galaxy formation . one of the immediate challenges to disk formation theory that is presented by these observations comes in the form of evidence that there has been little change in the sizes of disk galaxies since @xmath2=1 , despite an expectation that disks should be growing in size over this time . observations tell us that there exist large disks by @xmath2=1 ( before the universe was even half of its present age ) , and suggesting that these disks must be assembled prior to this epoch ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? recent observations lend strong support to the existence of disks back to even higher redshifts @xcite , and for disk assembly at even higher redshifts in the form of clump - chain galaxies in the hubble ultra deep field @xcite . results from large surveys suggest that there has been only weak evolution in disk sizes since @xmath2=1 . there is no evidence that the size function of disks has evolved back to @xmath2=1 @xcite . @xcite found that the shape of the size function did not evolve with redshift , though the normalization ( or amplitude ) did . this can be interpreted in two ways . first , the normalization will vary if the number density of disks varies with time . however , observations suggest that the number density of disks is constant to @xmath2=1 @xcite . second , if it is assumed that galaxies dim with time ( e.g. , due to a passively evolving stellar population or decline in star formation rate ) , a 1 to 1.5 magnitude dimming since @xmath2=1 could explain the change in normalization , while keeping the size distribution of disks constant @xcite . hence , there is no immediate evidence for a change in the sizes of galaxy disks with time . studies of the evolution in the magnitude size relationship for disk galaxies back to @xmath2=1 have found similar results . importantly , these studies must use surface brightness evolution to interpret evolution within the magnitude size plane . this is not straightforward , as an increase in disk scale lengths at a fixed magnitude between @xmath2=1 and @xmath2=0 can be mimicked by a decrease in luminosity at a fixed size @xcite . selection effects are difficult to disentangle , and studies that require high and low @xmath2 samples to adhere to the same selection biases have found that large disks are consistent with no evolution in surface brightness @xcite . later studies concluded that this work was too restrictive , and that a careful treatment of completeness as a function of redshift supports surface brightness dimming over time @xcite . most studies conclude that pure size evolution at a fixed magnitude is ruled out @xcite . instead , the observed surface brightness evolution is best explained by luminosity dimming , with the amount of dimming dependent on galaxy size so that lower mass galaxies have undergone more dimming since @xmath2=1 than massive galaxies @xcite . however , some amount of size evolution ca nt be ruled out , and at least weak size evolution is favored @xcite . while magnitude is assumed to scale with the mass of a galaxy , it is obvious from the above discussion that magnitude at a given mass is not necessarily constant in time . the more fundamental relation is between size and stellar mass . magnitude at a given stellar mass is likely to increase back to @xmath2=1 due to the fact that there is an evolving star formation rate ( sfr ) mass relation @xcite . @xcite examined the stellar mass size relation for disk dominated galaxies back to @xmath2=1 , and found weak or no evolution . this result can be interpreted in two ways . first , given the change of @xmath81 magnitude of surface brightness dimming in the same population , the result is consistent with a passively evolving stellar population at a given mass , with no growth of galaxy disks . second , it could imply that galaxies are growing , but that they must grow in such a way as to evolve _ along _ the stellar mass size relation with time . on the other hand , galaxy disk formation theory predicts that the sizes of disks , both individual and as a population , should be growing since @xmath2=1 . in the standard picture , gas in a halo conserves its specific angular momentum ( equal to that of the dark matter ) as it cools to form a centrifugally supported disk that grows from the inside out @xcite . in the simplest model , in which the density profile of galaxies is modeled as a singular isothermal sphere ( sis ) , the radius of the resulting disk is proportional to the parent halo virial radius , which grows inversely with the hubble parameter , @xmath9 . for a concordance @xmath0cdm cosmology @xcite , this relation predicts that disks at @xmath2=1 should be nearly a factor of two smaller than their @xmath2=0 counterparts at the same circular velocity , @xmath10 @xcite . @xcite adopted more reasonable assumptions to update the simple theoretical model that suggests that galaxy disks should grow by nearly a factor of two since @xmath2=1 , primarily by adopting nfw density profiles rather than sis profiles . unlike the sis model , the concentrations of nfw profiles were lower in the past . combined with disk stability arguments @xcite , this model predicts a much weaker evolution in disk sizes ( 15%-20% ) at a fixed stellar mass , consistent with the @xcite results . these semi - analytic models ( sams ) do not follow the evolution of individual disk galaxies with time . rather , they examine an instantaneous population of disks . clearly , this is the type of information derived from observations , but it prevents an interpretation of the evolution in individual disk galaxies based on observations of the population as a whole . @xcite instead follow the evolution of individual disks in their sam , and demonstrate that disks tend to grow along the stellar mass size relation , with only weak evolution , consistent with the results of @xcite . the analytic models discussed above ( e.g. , * ? ? ? ? * ; * ? ? ? * ) have until recently been the only available theoretical tool with which to investigate the evolution of galactic disk sizes ( though see * ? ? ? simulations , with their ability to capture complex gas processes in mergers and subsequent sf , should be an ideal tool that allow a better trace of the distribution of stellar light , while sams must assume that light follows mass . however , simulations of disk galaxy formation in a cdm context have historically produced unrealistic disks that are too compact ( dense ) , too small overall , and rotating too fast at a given radius ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? this failure has been named the `` angular momentum catastrophe . '' this catastrophe has been largely associated with the overcooling problem , in which baryons cool too quickly at early times and become very dense and concentrated at the center of halos before merging . these halos then experience dynamical friction in subsequent mergers , and the resulting disks show the classic signs of the angular momentum catastrophe @xcite . thus , feedback mechanisms at early times have historically been invoked to prevent overcooling with moderate success @xcite but additional artificial exchange of angular momentum can still occur between baryons and dark matter in sph simulations with low numerical resolution @xcite . hence , to avoid the angular momentum catastrophe and produce realistic disks , simulations must have both a physically motivated feedback prescription , and very high numerical ( mass and force ) resolution @xcite . the simulations used in this study are part of a suite of very high numerical resolution disk galaxy simulations that incorporate a star formation ( sf ) and supernova ( sn ) feedback scheme that has been shown to overcome many of the past problems of disk galaxy simulations to successfully match a number of observed properties of galaxies ( as discussed further below , and shown by the results of this paper ) . in this paper , we use simulations of individual disk galaxies to follow their evolution in the magnitude size plane and the stellar mass size plane . we attempt to mimic the observations of these galaxies in a manner similar to the observed population , generating mock surface brightness images to measure disk scale lengths , avoiding the assumption that light must follow mass . we demonstrate that the simulated galaxies have properties similar to real galaxies . in 2 we describe our small sample of very high resolution simulated disk galaxies that span a representative range of masses , merger histories , and spin values . we show that our limited sample is in excellent agreement with the population of galaxies surveyed at both @xmath2=0 ( 3 ) and @xmath2=1 ( 4 ) . in 5 , we disentangle the evolution of individual galaxies from the evolution seen in the population as a whole . we conclude in 6 . these simulations were run with the n - body + smoothed particle hydrodynamics ( sph ) code gasoline @xcite in a fully cosmological @xmath0cdm context using wmap year 3 parameters ( @xmath11 , @xmath0=0.74 , @xmath12 , @xmath13=0.77 , n=0.96 ) . the galaxies were originally selected from either a 25mpc or 50mpc ( depending on their mass ) n - body simulation with uniform mass resolution throughout , and then resimulated at higher resolution with baryons using the volume renormalization technique @xcite . this technique allows for significantly higher resolution on the central galaxy while also capturing the effect of large scale torques that are thought to deliver angular momentum to the galaxy @xcite . the sf / sn scheme used in these simulations has been shown to effectively regulate star formation efficiency as a function of halo mass , resulting in a stellar mass metallicity relationship for the simulated galaxies that is in excellent agreement with observations both locally and at high @xmath2 @xcite . this regulation of star formation also leads to realistic trends in gas fractions , with our lowest mass galaxies being the most gas rich @xcite , reproducing the observed incidence rate of damped lyman @xmath14 systems ( qso - dlas ) at @xmath2 = 3 @xcite , and the column densities of both qso - dlas and grb - dlas at @xmath2 = 3 @xcite . of critical importance for the present study is the fact that these simulated disks maintain sufficient angular momentum to match the observed tully - fisher relationship @xcite and produce galaxies with realistic disk sizes , as shown below . the full details of our physically motivated sn feedback implementation were originally presented in @xcite . briefly , the sf prescription ensures that the sfr density is a function of gas density according to the observed slope of the kennicutt - schmidt law , and a sf efficiency parameter , @xmath15 , sets the normalization of this relation . each star particle represents a simple stellar population , born with a kroupa initial mass function @xcite . as massive stars go sn , energy and metals are deposited into the nearest neighbor gas particles . the sn feedback recipe calculates the radius affected , and turns off cooling in those affected neighboring gas particles until the end of the snowplow phase as described by the sedov - taylor solution @xcite . the amount of energy deposited amongst those neighbors is 0.4@xmath1610@xmath17 ergs , as was adopted in all of our previous work mentioned above . additionally , we include a uniform uv cosmic background following an updated model of @xcite . @xcite demonstrated that when force resolutions @xmath18100pc can be achieved , high density peaks that mimic sf regions in giant molecular clouds can be resolved with several hundreds of gas particles in @xmath810@xmath19 m@xmath7 clumps ( see also * ? ? ? * ; * ? ? ? this allows for the adoption of a realistic density threshold for sf ( 100 amu/@xmath20 , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) . to match the kennicutt - schmidt law , the high density sf threshold must be offset by a slightly higher value of @xmath15 = 0.1 . this prescription leads to enhanced gas outflows that remove low angular momentum gas from the central regions of the galaxy ( * ? ? ? * brook et al . , in prep . ) . at low masses , this creates a bulgeless disk with a linearly rising rotation curve , comparable to those observed ( oh et al . , in prep ; * ? ? ? * ; * ? ? ? the low mass galaxies used in this paper , drawn from a 25mpc volume , can achieve force resolutions @xmath8100pc . hence , these simulations adopt the more realistic feedback prescription with high density sf and @xmath15 = 0.1 . however , the more massive galaxies presented here are drawn from 50mpc volumes , making it computational expensive to achieve similar force resolutions in a reasonable time . hence , these simulated galaxies can not resolve the high density sf peaks easily , and a lower sf density threshold must be adopted , 0.1 amu/@xmath20 , with a lower @xmath15 = 0.05 . these latter values have been adopted in all of our previous work on mw mass galaxies @xcite , and are a compromise that allow realistic disks to form , but are inefficient at driving gas outflows . thus , low angular momentum material that might be lost from the central regions is maintained , helping lead to the creation of large bulges in the mw mass galaxies ( see also * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? additionally , the creation of these large bulges can be due to missing physics . in particular , agn feedback has not yet been added to these simulations , and is potentially a key mechanism to help create realistic bulges . due to our small sample size , it is difficult to quantify if our bulges are substantially different from observed bulges in the same galaxy mass range . the simulated galaxies used in this work were selected to span a range of spin values and merger histories , with the last major merger redshift as low at 0.8 . properties of these simulations are listed in table [ simsum ] . lcccccccccr h516 & 3.9@xmath1610@xmath21 & 2.6@xmath1610@xmath22 & 1.6@xmath1610@xmath23 & 3.3@xmath1610@xmath24 & 0.05 & 0.05 & 1.2 & 87 & 3.5@xmath1610@xmath25 + h799 & 2.2@xmath1610@xmath21 & 1.3@xmath1610@xmath22 & 1.6@xmath1610@xmath23 & 3.3@xmath1610@xmath24 & 0.04 & 0.05 & 3.0 & 87 & 1.9@xmath1610@xmath25 + h603 & 3.8@xmath1610@xmath26 & 3.1@xmath1610@xmath21 & 3.0@xmath1610@xmath27 & 6.3@xmath1610@xmath23 & 0.06 & 0.11 & 1.1 & 231 & 3.8@xmath1610@xmath25 + h986 & 2.1@xmath1610@xmath26 & 2.2@xmath1610@xmath21 & 3.0@xmath1610@xmath27 & 6.3@xmath1610@xmath23 & 0.04 & 0.07 & 0.8 & 231 & 2.4@xmath1610@xmath25 + h239 & 8.5@xmath1610@xmath26 & 7.8@xmath1610@xmath21 & 1.2@xmath1610@xmath25 & 2.1@xmath1610@xmath27 & 0.03 & 0.05 & 1.1 & 347 & 2.8@xmath1610@xmath25 + h258 & 8.0@xmath1610@xmath26 & 7.4@xmath1610@xmath21 & 1.2@xmath1610@xmath25 & 2.1@xmath1610@xmath27 & 0.04 & 0.07 & 0.8 & 347 & 2.8@xmath1610@xmath25 + h277 & 7.1@xmath1610@xmath26 & 6.9@xmath1610@xmath21 & 1.2@xmath1610@xmath25 & 2.1@xmath1610@xmath27 & 0.03 & 0.04 & 3.0 & 347 & 2.3@xmath1610@xmath25 + h285 & 8.7@xmath1610@xmath26 & 8.1@xmath1610@xmath21 & 1.2@xmath1610@xmath25 & 2.1@xmath1610@xmath27 & 0.02 & 0.05 & 1.9 & 347 & 3.0@xmath1610@xmath25 + all of our simulations are of field galaxies that do not undergo major mergers below @xmath280.5 , making them disk dominated at redshift zero . in order to compare our simulated disk scale lengths to observational results in as realistic a way as possible , we wish to fit the _ light _ profile rather than the underlying mass profile . we create artificial surface brightness images of our simulated galaxies using sunrise @xcite , a monte carlo radiative transfer program that produces a spectral energy distribution ( sed ) for each resolution element of the simulations . this is done by identifying the age and metallicity of each star particle , which are then convolved with the starburst99 stellar population synthesis models @xcite to produce an sed . sunrise assumes dust tracks with the metallicity of the gas particles , and performs ray tracing from each star particle to compute the observed sed including absorption and scattering . from this , we generate mock images in chosen filter bands . figure [ fig1 ] shows artificial edge - on and face - on @xmath1 band images , including dust reprocessing , for two different mass galaxies . most of the observational surveys that have investigated the evolution of disk sizes worked in the rest frame @xmath1 band across redshifts . to compare to these results as a function of redshift , we have generated face on @xmath29 band surface brightness images at @xmath2=0.5 and @xmath2=1 that include the effects of surface brightness dimming , and using rest frame face on @xmath1 band surface brightness images for the @xmath2=0 galaxies . two of our disk galaxies are undergoing a major , disruptive merger at @xmath2=1 . for these galaxies , we created face on images for the most massive progenitor at a time just prior to the merger ( @xmath2=1.25 ) while the disk was still rather undisturbed . the face on sunrise images were fit with a sersic bulge component and an exponential disk component using the publicly available 2d surface brightness fitting code galfit @xcite . additionally , bars exist in both h603 and h986 at @xmath2=1 that were fit by a third component , reducing the b / d ratio of these galaxies at this step . the galfit results were checked against a 1d , radially averaged fit generated from the same images , and found to be in good agreement . to mimic observations , the fits were required to be a good match down to a limiting surface brightness value of @xmath826 mag / arcsec@xmath30 ( hence the importance of using surface brightness dimmed images at @xmath2=1 ) . example fits are shown in fig . [ sb ] for the same galaxies shown in fig . [ fig1 ] , with @xmath31 band fits shown simply for a comparison of the @xmath1 band results to those at a longer wavelength . we verified that using redshifted , surface brightness dimmed sunrise images in the @xmath29 band at higher z yielded similar disk scale lengths as the rest frame @xmath1 band . as can be seen in table [ simsum ] , a couple of these galaxies have undergone major mergers since @xmath2=1 , while some galaxies have had a very quiescent history . we note that at @xmath2=1 , two of our galaxies ( h239 and h285 ) had no component that could be well fit by an exponential . this may be due to mergers that occur in these galaxies near this time . galaxy h239 has the most active merger history of our simulated galaxies , with continual mergers ( both major and minor ) until @xmath280.5 . galaxy h285 begins to accrete a satellite with 1/7 of the stellar mass of the main halo at @xmath2=1.25 . however , the core of this satellite does not merge with the main halo until @xmath280.8 . in both cases , we searched in 250 myr periods from 0.75 @xmath32 1.25 to identify a step with a possible exponential disk , but none could be found . hence , those steps have been excluded from the high redshift analysis below . @xcite demonstrated that measuring disk scale lengths based on light rather than a kinematic decomposition could dramatically increase the resulting disk - to - total , d / t , ratio of their simulated galaxies , making measurement technique a potentially important reconciler between observations and simulations . in fig . [ ml ] we plot the @xmath2=0 surface mass density as a function of radius for both the stars ( solid line ) and the gas ( dashed line ) in the simulated galaxies . also shown as the dotted line is the exponential fit derived in the @xmath29 band for these simulations ( listed in table [ simsum ] ) . in a future paper ( mccleary et al . , in prep ) , we will explore in detail the differences between mass / kinematic results and light results ( e.g. , scale lengths , bulge - to - disk ratios , the role of dust and inclination , etc . ) . here , however , we note that an initial comparison of mass versus light exponential scale lengths demonstrates a less dramatic difference in our simulations than in @xcite . as seen in fig . [ ml ] , the @xmath29 band scale length is generally a good match to the stellar surface mass density . for a quantitative comparison , scale lengths were derived for the stellar mass over the same region that the disk dominates the light profile . the two methods were found to agree to within 10% for 75% of the galaxies . variation in the techniques appears to be mostly attributable to varying mass - to - light ratios ( mccleary et al . , in prep ) . the bulge / disk decompositions were performed on the dust extinguished sunrise results . however , disk scale lengths and central surface brightnesses were found to be identical in both the face on sunrise images with and without dust reprocessing , while the central bulge component was dust extinguished . thus , using the dust extinguished versus non dust extinguished decomposition has no effect on the results for the disk analysis presented below . it is generally expected that dust should have some effect on both the central surface brightness and the scale length of the disk component , even for face on orientations @xcite . we speculate that our more massive galaxies have little / no effect from dust due to the fact that they are also more gas poor than observed galaxies ( because they use up gas forming too many stars , discussed in more detail in section [ comp ] ) . our low mass galaxies are gas rich , in agreement with observed galaxies in this mass range @xcite , but metal poor @xcite , making dust negligible . it is important to note that the following analysis purposely focuses on the evolution of the _ disk _ alone . that is , the discussion below of how properties evolve with time neglects the bulge component of the galaxy . this is done for three reasons . first , as discussed above , our mw - mass galaxies tend to have larger bulges than observed . this is a problem that has historically plagued mw - mass simulated disks ( e.g. , * ? ? ? * ; * ? ? ? * ) , and is likely due to the inability to correctly model sf and feedback at the necessary high resolutions to drive loss of low angular momentum gas , and due to missing physics such as agn feedback . hence , while the disks of our simulated galaxies appear to be in good agreement with observed disks ( as presented below ) , the growth of the bulge in these massive galaxies is probably not modeled correctly . second , observations of the evolution of disk galaxies in the magnitude size plane have either a ) selected galaxies that can be fit with a small pure sersic index , @xmath33 , so that they are disk dominated , or b ) done a bulge / disk decomposition . in either case , the results we wish to compare to are concerned with disks only . finally , our goal here is to examine the growth of _ disks _ using these simulations , not bulges or spheroidal components . `` observable '' properties of the simulated galaxies are listed in table [ obssum ] . total magnitudes in each band are derived from sunrise , with each galaxy being integrated out to roughly 50 comoving kpc ( i.e. , essentially all of the flux from the simulated galaxy is contained ) . table [ obssum ] also lists the @xmath2=0 stellar mass derived for the entire galaxy ( disk + spheroid ) based on sunrise colors using the k_sdss_bell routine in the kcorrect package @xcite . a comparison to table [ simsum ] shows that the simulated stellar masses are generally @xmath840% larger than those derived from the photometric results . a more detailed comparison will be presented in mccleary et al . , ( in prep ) . in table [ obssum ] we list the bulge - to - disk ratios ( b / d ) despite the caveat mentioned above regarding large bulges in our higher mass galaxies . disk scale lengths in the @xmath29 band at @xmath2=0 are given in table [ obssum ] , while the @xmath1 band scale lengths used in the following plots are given in table [ evol ] . as seen in observational studies @xcite , the @xmath1 band scale lengths tend to be longer than the @xmath29 band scale lengths , at least for the six most massive galaxies ( with the two dwarf galaxies exhibiting different behavior ) . it has been debated whether this trend for disk scale lengths to be longer at shorter wavelengths is due to dust effects or age gradients @xcite . as noted above , dust plays no role in our face on _ disk _ fits , meaning that this trend in the simulations is due entirely to age and metallicity gradients . lcccccccc h516 & -16.8 & -15.9 & 0.08 & 0.14 & 0.53 & 3.4@xmath1610@xmath22 & 1.0 & 51 + h799 & -16.2 & -15.6 & 0.00 & 0.00 & 0.37 & 1.1@xmath1610@xmath22 & 0.7 & 43 + h603 & -21.3 & -20.5 & 1.25 & 0.72 & 0.52 & 2.2@xmath1610@xmath21 & 4.0 & 143 + h986 & -21.1 & -20.3 & 0.63 & 0.38 & 0.45 & 1.3@xmath1610@xmath21 & 2.8 & 137 + h239 & -22.5 & -21.8 & 0.35 & 0.16 & 0.41 & 4.2@xmath1610@xmath21 & 2.5 & 246 + h258 & -22.4 & -21.5 & 0.86 & 0.45 & 0.50 & 5.3@xmath1610@xmath21 & 5.3 & 204 + h277 & -22.1 & -21.2 & 0.63 & 0.42 & 0.52 & 4.3@xmath1610@xmath21 & 2.3 & 250 + h285 & -22.3 & -21.3 & 1.33 & 0.78 & 0.55 & 5.6@xmath1610@xmath21 & 4.5 & 203 + size , magnitude ( or mass ) , and rotational velocity make up a fundamental plane for disk galaxies @xcite . * see their figure 5 ) demonstrated that these simulated galaxies lie on the observed magnitude velocity relation for disk galaxies ( also known as the tully - fisher relation ) . in this paper we derive disk scale lengths in order to extend this analysis , and show that our galaxies are also a good match to the observed magnitude size and velocity size relations as well . all of our galaxies have sizes and velocities ( table [ obssum ] ) that agree well with the observed velocity size trend @xcite . two of the more massive galaxies have slightly high velocities for their size ( h239 and h277 ) , but still lie within the 2@xmath34 scatter of observed galaxies . the @xmath1 band disk magnitudes , m@xmath35 , and scale lengths , h@xmath35 , of our simulated galaxies at @xmath2=0 are shown in fig . [ fig : z0 ] . the simulated galaxies ( red circles ) are compared to three observational samples , @xcite and @xcite at the massive end , and @xcite at the low mass end . the @xcite data have adopted a dust correction based on @xcite , while the @xcite data are dust corrected based on inclination , following equations 1 , 2 , and 7 listed in @xcite . again , possibly due to low gas ( and therefore dust ) content , the six more massive galaxies shown here are free of dust effects . as such , they should be compared to dust corrected observational data . at the low mass end , however , @xcite makes no internal dust corrections , as the low metallicities of the dwarf galaxies in her sample are expected to lead to little effect from dust . clearly , the disk sizes of the simulated galaxies are in good agreement with observed disk sizes . we use the @xmath1 band scale lengths of the observational samples in fig . [ fig : z0 ] , which can be compared to the @xmath29 band results at @xmath2=1 in the next section since the @xmath1 band is redshifted into the @xmath29 band at @xmath2=1 . however , the magnitude size ( and stellar mass size ) relation has been measured in multiple bands for a large sample of sdss galaxies @xcite . the sdss results are consistent with the results plotted in fig . [ fig : z0 ] , though we purposely select observational samples that have done a bulge / disk decomposition and fit a pure exponential to the disk surface brightness fit , in order to isolate the evolution of the disk and eliminate any contamination from a central spheroid . as discussed in the introduction , historically it has been a challenge for cosmological disk galaxy simulations to match the observed fundamental relations for galaxy disks ( e.g. , * ? ? ? the failure to reproduce observed trends is the result of the angular momentum catastrophe , exacerbated by the fact that previous studies followed mass / kinematic results rather than light profiles . our success at matching the observed disk scaling relations is a result of the fact that our disk baryons maintain angular momentum through time , due to an increased resolution and a physically motivated feedback mechanism ( see * ? ? ? * for a review of how each of these processes contribute to the formation of realistic disks ) . a number of works have attempted to isolate the separate role of resolution and feedback ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) . in n - body + sph simulations , dark matter particles are typically an order of magnitude more massive than the gas particles , leading to an exchange of kinetic energy in two body interactions that kinematically heats the disk , randomizing velocity vectors , and potentially turning a disk component into a spheroid @xcite . this effect can be reduced at higher mass resolutions that lower the discrepancy in particle masses . @xcite used controlled ( non - cosmological ) n - body+sph simulations to evaluate the amount of angular momentum that can be lost purely due to resolution . they concluded that 10@xmath19 particles within the virial radius are necessary for a disk galaxy to maintain roughly 90% of its original angular momentum . yet even if resolution can be increased to the point of maximum angular momentum retention , disks will still suffer dramatic angular momentum loss compared to observed disk galaxies if energy feedback is neglected . feedback prevents rapid , early cooling of gas particles ( the overcooling problem discussed in the introduction ; see references therein ) . heating and expansion of the gas creates a hot reservoir , allowing the gas to cool at later times after the era of rapid mergers , and preventing angular momentum loss via dynamical friction in mergers ( e.g. , * ? ? ? additionally , by preventing overcooling , feedback prevents gas from rapidly turning into collisionless star particles . without feedback , halos of all masses are equally efficient at converting gas into stars @xcite , producing galaxies that are too gas poor compared to @xmath2 = 0 disk galaxies . because feedback regulates star formation , it allows for gas reservoirs to develop that allow disks to survive to the present day @xcite . the existence of large , thin disks at the present day thus requires feedback . not all sn feedback schemes lead to disk galaxies that satisfy observed constraints such as the tully - fisher relation , size velocity or size luminosity relations , or the stellar mass metallicity relation . because the nearest neighbor gas particles surrounding sne are dense and cold , a simple energy deposition will quickly be radiated away and not affect the simulation ( e.g. , * ? ? ? * ; * ? ? ? two main schemes have been adopted to overcome this problem . in the first , a multiphase model of the ism is implemented with a hot gas reservoir inside each gas particle @xcite , overcoming the problem of thermal energy being spread over the entire particle mass , and preventing the hot gas particles from being artificially influenced by their cold gas nearest neighbors . in the second , cooling is turned off in the gas particles near a sn explosion in order to mimic the sub - resolution adiabatic expansion of the sn @xcite . the `` blastwave '' scheme adopted here also turns off cooling in nearest neighbor particles , but attempts to model this based on what is known about actual sne , determining the radius of each sn remnant based on the analytic blastwave solution for a sn remnant @xcite , and cooling is only turned off for those particles within the blast radius . because many sne typically contribute feedback within a dense star forming region , the thermal energy from all of these sne can combine to create a larger blast radius . the differences in the resulting disk between a simple energy deposition ( `` thermal '' feedback ) and the blastwave model have been examined in @xcite and @xcite . as examined by these previous works , the adoption of the `` blastwave '' feedback model , combined with high resolution ( all simulations presented here have more than 10@xmath19 particles within the virial radius at @xmath2 = 0 , see table [ simsum ] ) , overcome past problems with cosmological disk galaxy simulations and allow for the present study . having established that these simulated disk galaxies match observed disk scaling relations at @xmath2=0 , we now use them to investigate the degeneracy in the magnitude size evolution with time . first , we establish that the simulations also match the available observational data at @xmath2=1 . the evolution of the magnitude size relation for these simulated galaxies is shown in fig . [ fig : magsize ] . the top panel shows the simulation results at @xmath2=0 , 0.5 , and 1 . it is evident that the magnitude size relation for these galaxies is evolving in time , dimming in surface brightness since @xmath2=1 . despite the fact that these are individual galaxies being followed in time , the snapshot of the population of disks at each redshift is consistent with the observations . this is evident in the bottom two panels . the middle panel compares the simulated disk scale lengths at @xmath2=1 to the decomposed @xmath29 band disk scale lengths of galaxies at 0.9 @xmath36 1.2 from @xcite and miller et al.(_in prep _ ) . no dust corrections are applied to the observational @xmath2=1 data , as these galaxies are expected to be low metallicity where corrections are uncertain and often ignored @xcite . any realistic dust corrections are likely to be tiny , and not affect the results . the bottom panel is a reproduction of fig . [ fig : z0 ] , shown again for easy comparison of the results across redshifts . while observational biases allow for a direct comparison only at the massive end at @xmath2=1 , the disk sizes of the simulated galaxies are in good agreement with observed disk sizes , as a function of redshift . we note that observational data that used a single sersic component fit ( e.g. , * ? ? ? * ) are also in good agreement with the @xmath2=1 data presented here ( and lie in the same magnitude range ) , despite the lack of a bulge / disk decomposition . for clarity , we show only the decomposed data in fig . [ fig : magsize ] . the top panel of fig . [ fig : magsize ] shows that there has been apparent dimming of the surface brightness of the simulated population since @xmath2=1 . [ deltamu ] quantifies this evolution , showing the change in the surface brightness , @xmath37 , of the simulated galaxy disks . we have defined @xmath37 such that @xmath38 data points in fig . [ deltamu ] at @xmath2=0 , 0.5 and 1 are for the mean @xmath37 for the population at that time , and error bars reflect the standard deviation of the sample . because our two dwarf galaxies are too faint to be generally observable at @xmath2=1 , fig . [ deltamu ] shows the evolution of these two galaxies separately from the higher mass galaxies . however , the evolution is roughly similar in both mass ranges , with about 1.5 magnitudes in surface brightness dimming between @xmath2=1 and @xmath2=0 . a dimming of 1.5 magnitudes is consistent with observational studies that have examined the surface brightness evolution of disk dominated galaxies ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? lcccccccc h516 & 22.7 & 1.5 & -15.7 & 2.0 & 0.95 & 1.0 & 4.3@xmath1610@xmath39 & 1.69 + h799 & 22.1 & 0.7 & -15.6 & 0.0 & 0.6 & 1.3 & 1.8@xmath1610@xmath39 & 1.32 + h603 & 22.5 & 2.8 & -19.8 & 1.0 & 4.2 & 2.4 & 1.3@xmath1610@xmath21 & 2.78 + h986 & 21.2 & 1.0 & -19.9 & 0.2 & 3.0 & 1.9 & 1.0@xmath1610@xmath21 & 3.22 + h239 & 19.3 & n / a & -21.5 & n / a & 3.2 & n / a & 9.0@xmath1610@xmath6 & 10.17 + h258 & 21.3 & 1.6 & -21.1 & -0.2 & 6.3 & 2.7 & 3.2@xmath1610@xmath21 & 2.53 + h277 & 19.7 & 0.6 & -20.8 & 0.4 & 2.4 & 1.4 & 1.9@xmath1610@xmath21 & 1.95 + h285 & 21.4 & n / a & -20.6 & n / a & 4.6 & n / a & 2.0@xmath1610@xmath21 & 3.07 + the previous section examined the population of simulated disks in terms of surface brightness evolution . we now decompose the surface brightness trends into luminosity and size evolution , and demonstrate how individual galaxies evolve in each property with time . table [ evol ] quantifies the changes in the disk @xmath1 band central surface brightness ( @xmath40 , as opposed to @xmath41 plotted in fig . [ deltamu ] ) , size , and magnitude for each of our simulated galaxies . it is widely expected that galaxies should undergo dimming since @xmath2=1 , due to declining star formation rates . the key question is how much galaxies dim . it is already apparent from fig . [ fig : magsize ] that our more massive galaxies are undergoing growth in size between @xmath2=1 and @xmath2=0 , meaning that the change in surface brightness of @xmath81.5 magnitudes over this time can not be entirely due to luminosity dimming at a fixed size . the changes in size and magnitude that are quantified in table [ evol ] are also visualized in fig . [ dmagsize ] . now it can be seen that there is in fact a dependency on mass in the evolution of the galaxies . our more massive galaxies are growing in size and generally undergoing only a small amount of dimming in total magnitude ; some are even getting brighter with time . meanwhile , there is little change in the scale length of our lowest mass galaxies , but their evolution in luminosity is erratic ( for this sample of two galaxies ) due to their bursting star formation histories . the luminosity evolution of the low mass dwarf galaxies is tied up with their particular star formation histories . while h799 is quiescent between @xmath2=1 and @xmath2=0.5 so that it dims , it undergoes another burst of sf at low @xmath2 , increasing its magnitude . on the other hand , h516 stays relatively quiescent between @xmath2=1 and @xmath2=0 , so that it undergoes significant luminosity dimming ( of 2.0 magnitudes ) . this large amount of dimming may at first seem to be at odds with the fact that low mass galaxies are generally undergoing more sf today than they were in the past ( this is discussed further in section [ starmass ] , and is one form of galaxy `` downsizing '' ) . however , this extreme dimming for low mass galaxies has been seen in observational results as well @xcite . @xcite concluded that small galaxies ( those with half light radii @xmath18 2 kpc ) have undergone roughly 2.5 magnitudes of dimming since @xmath2=1 . given the bursty sf nature of the two low mass dwarfs we have in our sample , it is not possible to compare statistically to the observations , but there is nothing to suggest that our simulated dwarfs are inconsistent with the @xcite results . at the higher mass end , @xcite find @xmath81.5 mag of dimming since @xmath2=1 for galaxies with half light radii @xmath42 3 kpc ( disk scale lengths @xmath42 1.8 kpc ) , if the strictest interpretation is made that all surface brightness evolution is in luminosity rather than size . clearly , this is not true for our high mass galaxies , which are growing with time . if galaxies are growing with time , then comparing galaxies that are a given size ( e.g. , 3 kpc ) at @xmath2=0 with galaxies that are of similar size at @xmath2=1 means that similar galaxies are _ not _ being compared . those @xmath2=0 galaxies will have been smaller at @xmath2=1 , and it is unknown from our simulation results what size galaxies with scale lengths larger than @xmath82.5 kpc at @xmath2=1 will have evolved to by @xmath2=0 . in summary , if galaxies are growing since @xmath2=1 , this will mimic a larger change in luminosity on the magnitude size relation than actually occurs @xcite . our simulated disk galaxies _ as a population _ appear to match the observational results for the evolution in the size - magnitude plane very well ( fig . [ fig : magsize ] ) . yet [ dmagsize ] demonstrates what the observations can not witness : individual galaxies are growing with time , so that the evolution in the magnitude size plane can not be due to dimming alone . disk scale length evolution for each galaxy is plotted in fig . [ evolsize ] , normalized to their size at @xmath2=1 . these values are also quantified in table [ evol ] . only a few of these galaxies show negligible size evolution , with the two lowest mass galaxies undergoing the smallest size changes . the remaining galaxies show clear growth , and about half evolve by the amount predicted by the simple sis model , ( nearly a factor of two back to @xmath2=1 ) , or more . the dotted line in fig . [ evolsize ] shows the growth predicted for the simple sis model @xcite , while the dashed line shows the growth determined by @xcite using 16 galaxies at @xmath2=1 . however , it is not necessarily expected that any given galaxy should follow the @xmath9 analytic growth predicted by the sis model . @xcite demonstrated that more sophisticated models adopting nfw profiles could bring the observations and theory closer in line . the prediction that sizes scale with @xmath9 is based on the assumption that @xmath43 where @xmath10 is the circular velocity of the halo and @xmath44 is the radius at which the mean density is equal to 200 times the critical density . then @xmath45 where @xmath46 is the dimensionless spin parameter @xcite . if the disk size follows @xmath9 , both @xmath46 and @xmath10 must stay roughly constant with time . while @xmath10 does stay roughly constant back to @xmath2=1 in these galaxies , @xmath46 does not . it is known that @xmath46 may vary for individual halos due to mergers or smooth accretion @xcite . the previous work of analytic models do not require @xmath46 to stay constant , as they simply use a snapshot in time and the spin values of halos at that snapshot . that is , they study a population of galaxies at a given time , and make no attempt to follow an individual galaxy with time as we do here . hence , it is not surprising that some of the simulated galaxies presented here deviate from growing precisely as @xmath9 . a better question , if we want to compare our disk sizes to analytic models , is to ask if the scale lengths measured based on our light profiles are similar to those predicted by analytic models . for an nfw density distribution , the disk scale length , r@xmath47 is @xmath48 where @xmath49 measures the halo concentration , and @xmath50 is nfw the scale radius . @xmath51 uses @xmath52 to determine the energy of the nfw halo compared to the sis model , and @xmath53 is a factor that accounts for the gravitational effect of the disk ( adiabatic contraction ) . @xmath54 and @xmath55 are the fraction of disk angular momentum and mass , respectively , to the total halo . for sizes to scale as @xmath9 , it is assumed that @xmath56 = 1 . this is likely to not be true , though , as demonstrated by @xcite . we have determined the values of @xmath51 , @xmath57 , @xmath58 , and @xmath59 for each of our galaxy halos at @xmath2=0 , directly from the simulation results . we compared the @xmath60 expected from these quantities , as derived by equation [ eq:4 ] , to the @xmath29 band scale lengths listed in table [ obssum ] . the results agree to within 10% for only the two lowest mass galaxies . for the more massive galaxies , the predicted nfw @xmath60 is consistently shorter than that found for the @xmath29 band , and for four of the six massive galaxies , the measured @xmath29 band results are a factor of 2 - 4 larger than the nfw prediction . the main source of this discrepancy appears to be in the value obtained for the spin value , @xmath46 , for the entire halo . if the spin of the gas , @xmath61 , within the halo is used instead , the discrepancy shrinks so that six of the eight galaxies have nfw @xmath60 and @xmath29 band results within 10% of each other . as listed in table [ simsum ] , the values of @xmath46 are generally lower than @xmath61 for all halos except the two lowest mass galaxies . this trend for @xmath61 to be larger than @xmath46 has been seen previously in simulations @xcite , and has been suggested as one reason why disks may have a smaller fraction of low angular momentum than their dark matter halos @xcite . recently , @xcite used a similar simulation run with gasoline to show that gas that enters the virial radius of a halo and cools toward the disk is torqued by the hot halo gas , so that the angular momentum of the disk gas becomes aligned with the hot halo . this is true even for gas that is initially counter - rotating with the disk , and hence is a powerful method to remove negative and low angular momentum material from the disk . the lack of low angular momentum material for the disk gas will lead to larger values of @xmath61 compared to @xmath46 for the total halo . the disk stars , whose light we trace in the results of this paper , form from this cold gas with larger spin values , and thus @xmath61 will be a better predictor for the nfw @xmath60 than @xmath46 . the result that the disk baryons are lacking a low angular momentum component compared to the dm is not trivial . this result needs to be examined in detail . however , this requires a full evaluation of the history of angular momentum in these halos , which is beyond the scope of the current paper . we reserve such a study for future work . as discussed above , numerous observational studies have found that there has been a decline in the surface brightness of disk galaxies , by a magnitude or more , since @xmath2=1 . due to the fact that there is little evidence for a change in the size function of disk galaxies over this time period ( e.g. , * ? ? ? * ) , the observations can only be interpreted as an upper limit to luminosity evolution . that is , if disks do not change in size , then all of the evolution must be in luminosity . however , due to the degeneracy in the magnitude size plane , the option of possible growth is left open . @xcite investigated this disk growth . after showing that they , too , agreed with @xmath81 magnitude of surface brightness dimming back to @xmath2=1 , they then used galaxy colors to derive stellar masses and considered the stellar mass size relation for their galaxies . they found that this relation , unlike the magnitude size relation , showed little or no evolution back to to @xmath2=1 . disk growth may still occur , but must occur in such a way that galaxies remain on the same stellar mass size relation with time . they concluded that this was evidence for weak inside - out growth of galaxy disks , as did additional later studies @xcite . the right panel of fig . [ dmagsize ] shows the evolution in the stellar mass size plane for our simulated galaxies . in agreement with previous results @xcite , these galaxies are growing along a path that keeps them on the same stellar mass size relation with redshift . the stellar mass shown in the right panel of fig . [ dmagsize ] is for the disk only . to separate the disk growth from total stellar growth of these galaxies ( i.e. , from the bulge and halo spheroidal stellar components ) , a kinematic decomposition was done to identify disk stars at @xmath2=0 . to identify disk particles , the galaxies are first aligned so that the disk angular momentum vector lies along the _ z_-axis . j@xmath62/j@xmath63 is calculated for each star particle in the galaxy , where j@xmath62 is the angular momentum in the _ x - y _ plane , and j@xmath63 is the momentum that a particle would have in a circular orbit with the same orbital energy . disk stars are identified as those having near circular orbits , so that j@xmath62/j@xmath63 @xmath42 0.8 . 0.2 , which matches the eccentricities observed in the milky way disk @xcite . ] these @xmath2=0 disk stars were then searched for in their most massive progenitor at @xmath2=0.5 and @xmath2=1 to find the mass of these stars that had formed at each step , with the results plotted in fig . [ dmagsize ] . if galaxy disks are evolving approximately along the same stellar mass size relation with time , then the surface mass density of disks should show little evolution back to @xmath2=1 @xcite . that is , a galaxy with stellar mass of 10@xmath21 m@xmath7 at @xmath2=1 is roughly the same size ( though maybe just slightly smaller ) as a galaxy at @xmath2=0 with 10@xmath21 m@xmath7 , yielding roughly the same surface densities . [ barden ] tentatively confirms this result , showing that the surface mass density of our population of disks evolves little with time , with surface mass density , @xmath64 defined as @xmath65 the red points with error bars , connected by the solid black line , show the mean and standard deviation for the whole sample at @xmath2 = 0 , 0.5 , and 1 . the colored lines are the results for individual simulated disks with time . we note , however , that the dashed line shows the prediction for the surface mass density evolution for the simple sis model @xcite . the standard deviation within our small sample is large , so that the simulated galaxies are fully consistent with the predicted growth . while our sample is clearly too small to derive statistical results , the change in the mean surface density of the population between @xmath2=1 and @xmath2=0 is only 0.1 dex , half of the evolution predicted by the sis model . we note that an updated model by @xcite that incorporates the evolution of nfw halos derived in n - body simulations predicts only @xmath80.2 dex change in the stellar surface mass density back to @xmath2=1 . this small evolution in surface mass density is due to evolution along a stellar mass size relation that changes little with time . due to the fact that the stellar mass size relation shows little change with time , examining the evolution of galaxy disks at a fixed _ stellar mass _ is similar to examining the evolution at a fixed _ size_. however , because galaxies are growing with time , a galaxy that is , e.g. , 3 kpc in size and 10@xmath21 m@xmath7 in stellar mass at @xmath2=1 has increased in mass and size by @xmath2=0 and another galaxy has moved from a lower mass and smaller size into the same bin . observations , and the results presented here , tell us that these two galaxies of same size at @xmath2=1 and @xmath2=0 have a different luminosity : a galaxy at @xmath2=1 with similar stellar mass and size to a @xmath2=0 galaxy will be brighter . note that this is _ not _ equivalent to saying that an individual galaxy was brighter in the past . instead , galaxies at fixed stellar mass were brighter in the past . because they are the same stellar mass , the brighter @xmath2=1 galaxy must then be forming more stars . two of our galaxies can be compared in this way . galaxy h603 has a stellar disk mass of 1.28@xmath1610@xmath21 m@xmath7 at @xmath2=0 , while h258 has a stellar disk mass of 1.25@xmath1610@xmath21 m@xmath7 at @xmath2=1 . the sfr within the last 100myr for h258 at @xmath2=1 is 3.1 m@xmath7/yr , while it is 0.6 m@xmath7/yr for h603 at @xmath2=0 . the lower sfr for h603 is due to the fact that it resides in a less massive halo than h258 ( see table [ simsum ] ) . even at @xmath2=1 , h603 had a lower sfr than h258 , of 2.1 m@xmath7/yr . this is because these galaxies follow a sfr stellar mass relation ( which also leads to their reproduction of the observed stellar mass metallicity relation , * ? ? ? [ fig9 ] shows the sfr stellar mass and specific sfr ( ssfr ) stellar mass relations for our simulated disks . the left panels of fig . [ fig9 ] shows the ssfr ( top ) and sfr within the last 100 myr ( bottom ) versus stellar disk mass . these left panels confirm the values listed in the paragraph above . the dwarf galaxies are included in this plot for full disclosure , but lie outside of the mass range observable at high @xmath2 , unlike the higher mass galaxies . two trends among the higher mass galaxies are noteworthy . first , the sfr of any individual galaxy tends to decrease slightly with time , but second , the entire population as a whole can be observed to shift to lower sfrs and ssfrs @xcite . the decline in sfr and ssfr in has been observed extensively in populations of galaxies @xcite , and is one form of `` downsizing '' in galaxy evolution . the right panels of fig . [ fig9 ] also show sfr and ssfr for the simulated galaxy disks , but now as a function of total halo mass . this is included to demonstrate that although , e.g. , h603 at @xmath2=0 and h258 at @xmath2=1 have similar disk stellar masses , this does not correspond to similar virial masses for those redshifts ( colors yellow and purple in the figures , respectively ) . the deeper potential well of the more massive galaxy ( h258 ) leads to a higher sfr.=1.25 due to the fact that it is undergoing a merger at @xmath2=1 . galaxy h258 is 50% more massive than h603 at @xmath2=1 . ] in conclusion , attempting to compare galaxies that are a similar stellar mass at @xmath2=1 and @xmath2=0 ( and by the tight stellar mass size relation , similar size ) leads to inherently comparing galaxies of two different virial masses . this then leads to the observed difference in luminosity for galaxies of a similar size at @xmath2=1 and @xmath2=0 , caused by the sfr stellar mass relation , and the corresponding stellar mass halo mass relation @xcite . in fig . [ fig10 ] we reproduce some of the simulated scaling relations , this time with observed scaling relations directly overplotted for comparison . the left panel of fig . [ fig10 ] shows the low redshift relations for sdss galaxies ( solid line , with 2@xmath34 scatter in dotted lines ) from @xcite . it is worth noting that @xcite performed bulge / disk decompositions with the sdss data , so that the sizes plotted are fits to purely exponential disk scale lengths ( converted from the half light radius presented in their paper ) . @xcite also use higher @xmath2 data from the deep2 survey to measure the change in size since @xmath2=0 , finding @xmath80.1 dex of size evolution at a fixed mass back to @xmath2=1 . in the mass range of overlap ( m@xmath4 @xmath42 10@xmath6 m@xmath7 ) , our simulated galaxies are entirely consistent with observed galaxies within the scatter . the central panel of fig . [ fig10 ] reproduces the stellar mass sfr relation , now with observational data from @xcite plotted for @xmath2=0 ( solid line ) and @xmath2=1 ( dashed line ) , corrected from a salpeter imf to a kroupa imf . at both redshifts , the minimum mass observed is roughly 10@xmath6 m@xmath7 . the right panel shows the same data , but now for the ssfr . the data point with error bars in the lower right corner of the right panel provides an estimate of the typical scatter in the @xcite data . while our galaxies are certainly consistent with the data within the scatter , and the general slope at each @xmath2 is consistent with the observations , there is overall a tendency for the simulated galaxies to be more massive than the observations . we believe the discrepancy is in stellar mass rather than sfr , because these simulated galaxies have both @xmath2=1 and @xmath2=0 disk sfrs in good agreement with observed galaxies in the same halo mass range @xcite . we note that the @xcite data is for total stellar mass , while our simulated results are for the disk only ( bulge excluded ) . however , including the bulge mass will only exacerbate the discrepancy . for the massive galaxies , the discrepancy is probably due to the fact that these simulations are known to form too many stars @xcite . we can learn something about _ when _ this overproduction of stars occurs in the simulations by also examining the stellar mass ssfr plot ( right panel of fig . [ fig10 ] ) . while within the observed scatter , is appears that the ssfrs of the simulated disks are consistently lower than observational results at both @xmath2=1 and @xmath2=0 @xcite . these plots demonstrate that these simulated galaxies must be forming too many stars at early times , at redshifts higher than plotted here ( @xmath66 1 ) . because the sfrs of the galaxies are in good agreement with observed galaxies in the same halo mass range at both @xmath2=1 and @xmath2=0 @xcite , the fact that the ssfrs are lower than observed values requires that the stellar mass of these disks be too large by @xmath2=1 . the story is slightly different for the two dwarf disk simulations . these two dwarf galaxies have been published previously in @xcite , where it was demonstrated that their properties are a good match to observed dwarf disk galaxies , including stellar mass @xcite . however , the observed ssfrs of low mass galaxies is on average higher than that of massive galaxies @xcite , indicating that low mass galaxies have been very inefficient in the past at turning gas into stars . clearly , that is not the case with our two simulated dwarf disk galaxies , which have too low ssfrs compared to most observed dwarf galaxies due to the fact that they form the bulk of their stars prior to @xmath2=1 . hence , while approximately the correct mass in stars is formed , the star formation is biased toward @xmath66 1 . we have verified that these two simulations , when run with stronger sn feedback , have burstier sf histories with their sf spread out more evenly down to the present time . this leads to approximately the same mass in stars formed by @xmath2=0 , but with a more uniform sf history . this means that the stronger feedback dwarf runs have higher ssfrs than currently presented here , and ssfr values larger than the massive galaxy simulations shown here . future work will include these new dwarf disks , which show similar stellar masses , rotational velocities , and sizes to those presented in this paper . thus , in all of the simulated galaxies the star formation history is biased toward too much star formation at high @xmath2 . by @xmath2=0 , though , the dwarf galaxies have formed approximately the observed amount of stars for their halo mass , while the more massive galaxies have overproduced the total amount of stars . it is suspected that this overproduction of stars in the massive galaxy disks may be the result of the adopted low sf density threshold , which allows for most of the disk to be capable of sf at any given time , unlike the high density threshold which limits sf only to rare density peaks . however , as discussed in section 2.1 , it is currently computationally expensive to generate the more massive galaxies with the force resolution required to resolve high density peaks where sf occurs , and no simulations of sufficiently high resolution are currently available to test this hypothesis . the adopted sf threshold is a compromise that allows stars to form at radii observed in real galaxy disks , and we have demonstrated that it leads to scale lengths in good agreement with observed galaxies . however , we note that even if our massive galaxies are forming too many stars , it is by roughly a factor of 3 at most , and at early times . this means that the more massive galaxies may shift up to 0.5 dex to lower stellar masses in the stellar mass size plane , but will still grow along the stellar mass size relation , so that our conclusions are unaffected . likewise , the @xmath1 band magnitudes we present here are unlikely to be affected , as the bulk of overproduced stars must occur at @xmath66 1 . since @xmath1 band light is dominated by young stars , and our low @xmath2 sfrs are in good agreement with observed galaxies in the same halo mass range , the magnitude size relations presented in figs . [ fig : z0 ] and [ fig : magsize ] are unlikely to change . hence , the conclusions presented here about the origin of the evolution in the size luminosity relation remain valid . for the first time , a sample of very high resolution galaxy disks simulated within a fully cosmological context have been used to determine the evolution of the galaxy disk size magnitude relation . unlike previous simulations , we determine disk scale lengths and magnitudes by fitting the stellar light profile rather than decomposing our simulations based on kinematics ( and hence mass ) . we generate artificial surface brightness images using sunrise @xcite , in order to derive disk properties using methods used by observers . exponential disk scale lengths were derived using galfit @xcite with redshifted , surface brightness dimmed @xmath29 band images at @xmath2=1 and 0.5 , and rest frame @xmath1 band images at @xmath2=0 . this allows for a direct comparison between simulated and observational results in the size magnitude plane . [ fig : z0 ] demonstrates that these galaxies overcome past inabilities for cosmological disk simulations to retain angular momentum . artificial loss of angular momentum is minimized due to high resolution and a physically motivated sn feedback recipe , and hence we produce disks with realistic sizes compared to observed disks . fig . [ fig : magsize ] shows that the sizes are in agreement with observed disk sizes back to @xmath2=1 . this success , and our previous successes in matching the observed stellar mass metallicity relation with time @xcite , and the tully - fisher relationship @xcite , indicate that these simulated disks reproduce the fundamental scaling relations for disk galaxies . having established this , we can now use these simulated disks to investigate and interpret observational findings . we use the simulated disk properties to investigate the evolution in the magnitude size relation with time . observations have generally concluded that there has been 1 to 1.5 magnitudes of surface brightness dimming since @xmath2=1 @xcite . we find similar evolution for our galaxies ( see fig . [ deltamu ] ) . combined with a lack of evidence for an evolving size function for galaxy disks @xcite , and little to no growth in galaxy disks sizes back to @xmath2=1 @xcite , the observed surface brightness evolution can only be interpreted as an upper limit to luminosity evolution if there has been no change in galaxy disk sizes . we have shown that we 1 ) are in good agreement with the observed magnitude size relation with time and the evolution in surface brightness , but 2 ) a number of the simulated disks are clearly undergoing a large change in size ( figs . [ dmagsize ] and [ evolsize ] ) . our ability to follow individual galaxies with time , unlike the observations , allows us to interpret the evolution in the magnitude size plane . we find that the evolution is dependent on mass , with our massive galaxies undergoing larger changes in size than magnitude . our dwarf galaxies undergo the least change in size , though there is no immediate theoretical expectation that this should be the case . recent results @xcite show that the rate of growth of halos has little dependence on mass , and thus low mass galaxies should grow just as much in size as their more massive counterparts . we conclude that a larger sample is needed to study if low mass galaxies truly grow less than more massive galaxies . on the other hand , the evolution in magnitude for dwarf disk galaxies is dependent on their bursty sf history , but an individual dwarf can undergo significant dimming of at least 2 magnitudes . a dichotomy of evolution with mass has been seen in observations , with low mass galaxies exhibiting significantly more dimming since @xmath2=1 than more massive galaxies ( e.g. , * ? ? ? * ; * ? ? ? the halo properties for these galaxies ( @xmath67 ) predict a scale length that is generally shorter than the scale lengths derived from the simulated light profiles . fig . [ ml ] demonstrates that the light and mass distributions yield scale lengths in good agreement , so the discrepancy between @xmath60 predicted for nfw halos and from our light profiles can not be due to a discrepancy between light and mass profiles . rather , using @xmath61 , the spin of the gas in the halo , brings the predicted and measured scale lengths into decent agreement . thus , the angular momentum distribution of the gas from which the stars are forming has been modified from that of the dark matter ( see also * ? ? ? * ; * ? ? ? @xcite have demonstrated that accreted gas is torqued by the hot halo after entering the virial radius , potentially preventing counter - rotating and low angular momentum gas from reaching the disk . a full examination of how this process occurs is beyond the scope of this paper and left for future work . having determined that these galaxies are growing in size with time , we can investigate this growth in stellar mass . like previous theoretical models @xcite , we find that these galaxies grow in such a way as to stay along approximately the same stellar mass size relation with time . hence , at a fixed stellar mass , there is little change back to @xmath2=1 in the size of galaxies . due to the fact that our larger galaxies have grown since @xmath2=1 , we do not have the ability to directly compare results at a fixed size . in particular , we do not have a population of simulated disks at @xmath2=1 with scale length greater than 3 kpc , though these larger disks do exist observationally . presumably we might generate these larger disks if we simulate higher mass halos . however , it is not clear from this study how these larger disks would evolve to @xmath2=0 . observationally , the number density of disk galaxies shows no evidence for change back to @xmath2=1 @xcite . in conjunction with the apparent lack of evolution in the size function of disks , this fact has been interpreted to mean that either galaxies are done growing by @xmath2=1 , or that the rate of destruction of galaxies at given size must be equally matched by the rate of growth of galaxies into that size . the fact that our galaxies are growing , while still matching the size - magnitude evolution , suggests that galaxies are moving into and out of a given bin in size during this time interval . it is beyond the scope of this work to answer what happens to these larger galaxies , though we may speculate . there are two possibilities that arise due to mergers ; perhaps the larger disks undergo mergers and become early type galaxies @xcite , or perhaps mergers cause them to become bulge dominated disks . as the majority of size - magnitude evolution studies have focused on pure disks or disks with small bulges , this would remove them from the sample being considered @xcite . finally , we demonstrate that the simulated galaxy disks follow sfr mass and ssfr mass relations similar to observations . while the growth of dark matter halos is nearly self - similar with mass , observations demonstrate that some process must break this self - similarity in the baryons ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) . in the mass range of galaxy halos presented here , sn feedback is the process that regulates sf as a function of halo mass , and likely drives gas outflows that vary as a function of mass ( though outflow rates are left for future work ) . the regulation of sf as a function of mass leads to the reproduction of the observed stellar mass metallicity relationship @xcite , and the sfr mass relation shown here that is responsible for the reproduction of the size luminosity relation back to @xmath2=1 . a galaxy at a fixed m@xmath4 at @xmath2=1 will have a higher sfr than a @xmath2=0 counterpart , and thus a higher luminosity as well . in conjunction with the weak evolution of the stellar mass size relation , this tells us that the difference in luminosity observed at a fixed size between @xmath2=1 and @xmath2=0 is due to the fact that the @xmath2=1 galaxy has a higher sfr and luminosity . we would like to thank s. miller and r. ellis for use of their @xmath2=1 disk scale length data . ab acknowledges support from the sherman fairchild foundation . as was supported by the gordon & betty moore foundation . fg and tq were funded by nsf ast-0908499 . fg acknowledges support from a theodore dunham grant , hst go-1125 , nsf grant ast-0607819 and nasa atp nnx08ag84 g . pj acknowledges support from the w. m. keck foundation . simulations were run at tacc , arsc , and nas .

the critical behavior of time - independent systems , especially in terms of the spectral statistics and the eigenstate statistics , has attracted great attention . on the spectrum side , hofstadter s butterfly spectrum of the harper model has been a paradigm for critical spectral statistics , representing a multifractal spectrum @xcite of a system exactly on the metal - insulator transition point . on the eigenstate side , mainly through studies in time - independent models , such as the power - law random banded matrix ( prbm ) model @xcite and the standard anderson tight - binding model ( tbm ) @xcite , it has been well - established that for a system on a metal - insulator transition point or the anderson transition point @xcite , its eigenstates show clear fractal features . this background of understanding the critical behavior of time - independent systems motivated our interest in the critical behavior of periodically driven systems . below we first introduce recent related studies of critical floquet spectra , and then briefly describe the motivation and the results of this work . it is well known that the floquet ( quasi - energy ) spectrum of a delta - kicked version of the harper model also displays hofstadter s butterfly patterns @xcite . interestingly , though the kicked harper model ( khm ) can be classically chaotic , its spectrum , due to its fractal nature , does not follow the bohigas - giannoni - schmit conjecture @xcite at all . this makes the khm not only a fruitful model for gaining new insights into the issue of quantum - classical correspondence in classically chaotic systems , but also an intriguing model to study critical spectral statistics . indeed , for quite a long time , studies of fractal floquet spectra were largely restricted to the khm and its variants @xcite . in a proposal to experimentally realize hofstadter s butterfly floquet spectrum in cold - atom laboratories , wang and gong @xcite recently demonstrated that hofstadter s butterfly floquet spectrum can be synthesized by use of a double - kicked cold - atom rotor system @xcite under a quantum resonance condition . @xcite then showed that the butterfly floquet spectrum of the cold - atom system studied in refs . @xcite is equivalent to that of the standard khm if and only if one system parameter takes irrational values . in addition to motivating a cold - atom realization of critical floquet spectra of periodically driven systems , refs . @xcite seem to have offered a general strategy for synthesizing critical floquet spectra in driven systems . using an approach extended from refs . @xcite , recently wang and gong @xcite showed that the floquet spectra of a class of driven su(2 ) systems also display butterfly patterns and multifractal properties that are characteristics of highly critical spectra . this establishes a completely different class of critical driven systems without a connection with the khm context . interestingly , the driven su(2 ) model in ref . @xcite can be understood as a simple extension of the well - known kicked top model ( ktm ) @xcite in the quantum chaos literature . because the ktm has just been experimentally realized in a cold @xmath0cs system @xcite , it can be expected that a critical driven su(2 ) system may also be experimentally realized using the collective spin of a @xmath0cs atomic ensemble . an alternative experimental realization may be based on a driven two - mode bose - einstein condensate @xcite , which represents a strongly self - interacting driven system . given the above - mentioned class of driven quantum systems with critical floquet spectra , it becomes necessary to study the behavior of the associated floquet eigenstates . theoretically speaking , because driven su(2 ) systems always have a finite number of floquet eigenstates , the eigenstate analysis becomes much easier than in the khm , with the latter necessarily involving an infinite number of eigenstates for a fractal floquet spectrum . a careful investigation of the floquet eigenstates over the entire spectrum will help to better understand the critical behavior in time - dependent systems in general . experimentally speaking , information about the eigenstate statistics may be more directly accessible to measurements than a fractal spectrum . to analyze the critical behavior of the floquet eigenstates in driven su(2 ) systems , we adopt the same approach as in previous studies of time - independent systems . that is , we shall numerically examine the fluctuations of the eigenstates @xcite . the eigenstate fluctuations can be characterized by a set of inverse participation ratios ( ipr ) : @xmath1 where @xmath2 is the index of the eigenstates , @xmath3 represents one eigenstate under investigation , and @xmath4 are the basis states . for convenience we focus on the ipr @xmath5 ( i.e. , @xmath6 ) . by analogy to critical eigenstate behavior in time - independent systems , we expect that @xmath5 scales anomalously with the hilbert space dimension @xmath7 as @xmath8 where @xmath9 is a fractal dimension of a particular eigenstate @xmath10 . but is there also a unique fractal dimension @xmath11 for the average behavior of all the floquet eigenstates , for example , via the slope of the averaged @xmath12 , denoted @xmath13 , versus @xmath14 ? to that end , we shall examine if , as the system gets closer to the thermodynamic limit ( @xmath15 ) , the distribution of @xmath16 shows signs of a scale - invariant form @xcite . in other words , whether the distribution function of @xmath12 , denoted @xmath17 $ ] , only shifts as @xmath7 varies . certainly , the system under our study has only a finite size @xmath7 . in time - independent anderson - transition studies using the prbm or the tbm , it was conjectured that the variance of @xmath12 , denoted @xmath18 , scales with @xmath7 as @xmath19 with @xmath20 , @xmath21 , and @xmath22 being three adjustable parameters @xcite . for a @xmath23-dimensional system on the anderson transition point , it was shown that @xmath22 is related to @xmath11 by @xmath24 where @xmath25 equals @xmath26 or @xmath27 depending upon whether or not the system has time - reversal symmetry @xcite . as one main task of this work , we shall examine if these results for time - independent systems still hold for critical floquet eigenstates . furthermore , we hope to see how the criticality of the eigenstates of unitary operators differs from the criticality of the eigenstates of self - adjoint operators . results along this direction will also be relevant to recent investigations on the unitary anderson model " @xcite , the thue - morse sequence generating multifractal eigenstates of the quantum baker s map @xcite , the one - parameter model of quantum maps showing multifractal eigenstates @xcite , as well as recent experimental and theoretical studies of anderson transition in kicked - rotor systems @xcite . we now briefly summarize the main findings of this work . for the driven su(2 ) systems studied here , we consider two different parameter regimes : in one regime the floquet spectra display clear butterfly patterns , and in the other regime , the butterfly patterns of the floquet spectra have dissolved due to increased strength of the driving fields . for both regimes , we find that @xmath28 $ ] is not as smooth as observed in the tbm or prbm , indicating some non - universal features in dynamical systems . the @xmath28 $ ] for cases with dissolved butterfly patterns is however smoother . for either regime , it is found that the ensemble average @xmath29 does scale linearly with @xmath14 , with the slope of the @xmath30 vs @xmath14 curve clearly defining the fractal dimension @xmath11 for all the eigenstates . we also find it possible to fit the variance of @xmath31 by eq . , but with the exponent @xmath22 given by @xmath32 instead ( with @xmath33 ) ; i.e. , a factor of _ two _ is missing from the denominator as compared with eq . ( 4 ) for time - independent critical systems . to further understand this difference , we propose a random matrix model , which we call power - law random banded unitary matrix " ( prbum ) model . by tuning the parameters of the prbum , the @xmath11 value associated with the prbum can be varied . more interestingly , we observe that the variance @xmath18 of the prbum also follows eq . ( [ var_goe ] ) , with the exponent @xmath22 again given by eq . ( [ dq_coe ] ) . this suggests that our findings about the floquet eigenstate statistics based on driven su(2 ) systems do reflect some general aspects of critical floquet eigenstates . this paper is organized as follows . in sec . [ sec3 ] , we present detailed results of the eigenstates statistics in our driven su(2 ) models , with or without time - reversal symmetry . in sec . [ sec4 ] , we introduce the prbum to represent a class of critical floquet operators , discuss the statistics of the eigenstates of prbum , and then compare the associated results with those found in actual dynamical systems . in sec . [ sec5 ] , we study the eigenstate statistics of the standard kicked top model @xcite , which represents a classically chaotic , but non - critical , driven system . section [ sec6 ] concludes this work . the focus in ref . @xcite is on the fractal spectral statistics . here , using the same model we study the statistics of the floquet eigenstates . the first floquet operator under study is given by @xmath34 where @xmath35 are angular momentum operators satisfying the su(2 ) algebra and @xmath36 is the conserved total angular momentum quantum number that defines a @xmath37-dimensional hilbert space . readers can refer to ref . @xcite for detailed descriptions and motivations of this model . this model is also called as a double - kicked top model " ( dktm ) in ref . @xcite . eigenstates of the @xmath38 operator are denoted as @xmath39 , with @xmath40 . states @xmath41 will be chosen as our representation for eigenstate analysis . to analyze the floquet eigenstates , it is necessary to express the floquet operator in symmetric basis states , a procedure that block - diagonalizes the floquet matrix . on the one hand , this will simplify our analysis ; on the other hand , this is necessary for the sake of comparison between an actual dynamical system and the prbum model proposed below . the dktm floquet operator @xmath42 in eq . ( [ floquet_coe ] ) has a parity unitary symmetry @xmath43 where @xmath44 . this symmetry can be used to block diagonalize the @xmath42 matrix into two disconnected sub - matrices associated with either odd - parity or even - parity subspaces . without loss of generality we only present below results for the @xmath36-dimensional odd - parity subspace . besides the parity symmetry , @xmath42 also has a time - reversal anti - unitary symmetry @xmath45 , with @xmath46 where @xmath47 is the complex conjugation operator . to explore the implication of this time - reversal symmetry for the eigenstate statistics , we shall also consider a variant of @xmath42 , i.e. , @xmath48 evidently , @xmath49 differs from @xmath42 only in the last factor , i.e. , @xmath50 in @xmath42 is replaced by @xmath51 . because of this difference , we call @xmath42 in eq . ( [ floquet_coe ] ) the @xmath52 model and call @xmath49 the @xmath53 model . it is easy to check that the latter does not have the parity symmetry or the time - reversal symmetry . for the @xmath53 model , which can not be reduced to any block - diagonal form , we examine the eigenstates of the full floquet matrix . for both cases we define a dimensionless system parameter , @xmath54 this choice of @xmath55 being @xmath56 times the golden mean is to ensure that the resulting floquet eigenstate statistics is indeed representative of driven systems with fractal floquet spectra . as detailed below , we consider _ two _ different regimes for the product @xmath57 . in the first regime defined by @xmath58 , the floquet spectra show clear butterfly patterns ; in the second regime defined by @xmath59 , the butterfly spectra have dissolved , with fractal dimensions of the spectra increased @xcite . this is a time - reversal symmetric system . because dyson s circular ensemble of random unitary matrices @xcite with time - reversal symmetry is called circular - orthogonal - ensemble " ( coe ) , we regard the @xmath52 model as an example of critical coe statistics . figure [ jx - jxdistalphaj1.0](a ) shows the distributions of the _ logarithm _ of the ipr @xmath5 , denoted @xmath28 $ ] , for different @xmath36 . it is seen that the distribution function @xmath28 $ ] is not as smooth as that observed in early anderson - transition studies @xcite . nevertheless , it is clear that as @xmath36 increases , the left tail of @xmath28 $ ] systematically shifts to the left direction associated with more negative @xmath12 . the profile of @xmath28 $ ] , though somewhat changes as @xmath36 increases , does maintain its main features as @xmath36 increases . due to these features that are similar to early findings for the critical eigenstates in time - independent systems , it can be expected that the average of @xmath12 will show a scaling behavior with @xmath60 . as shown in fig . [ jx - jxdistalphaj1.0](b ) , this is indeed the case . therein , @xmath30 , obtained by averaging @xmath12 over all eigenstates ( in the odd - parity subspace ) , displays an excellent linear behavior with @xmath61 . from the slope of the fitting line in fig . 1(b ) , we are able to obtain the fractal dimension @xmath62 . the distribution profile @xmath28 $ ] in fig . 1(a ) is seen to display rich features , with significant fluctuations and multiple notable peaks . qualitatively , this reflects that our system is an actual dynamical system and hence the underlying rich dynamics will manifest itself through some non - universal statistical features . related to this observation we also note that in our calculations , all the floquet eigenstates are treated equally and all of them are used for averaging . this is in contrast to the common procedure in analyzing time - independent critical systems , where only those energy eigenstates in a certain small energy window around zero eigenvalue are included to examine the distribution of @xmath12 @xcite . the justification for including all floquet states in our analysis is as follows : the quasi - energy spectra lie on a unit circle and hence all states with different eigenphases on the unit circle should be treated on equal footing . to double check this understanding , we have also taken windows of different widths centered around _ zero _ value of the eigenphase and then calculate the distribution of @xmath31 . no improvement in the smoothness of @xmath63 is found . rather , we obtained similar distribution of @xmath12 with clear fluctuations . it is also tempting to connect the non - universal features of @xmath63 with the phase space structures of the underlying classical limit . however , such a perspective , which calls for a good understanding of quantum - classical correspondence in critical systems , is unlikely to succeed because the classical limit of our dynamical model is completely chaotic @xcite . in fig . [ jx - jxdistalphaj1.0](c ) , we plot @xmath64 $ ] vs @xmath60 ( filled circles ) , where @xmath65 is the variance of @xmath16 and @xmath20 is a fitting parameter , whose value is found by fitting our data points with the empirical formula given in eq . ( [ var_goe ] ) . as seen in fig . [ jx - jxdistalphaj1.0](c ) , the fitting is reasonably good , yielding that @xmath66 $ ] scales as @xmath67 , with @xmath68 [ @xmath11 is already determined by the fitting in fig . [ jx - jxdistalphaj1.0](b ) ] , @xmath69 , and @xmath70 . despite obvious fluctuations around the fitting curve , the result in fig . [ jx - jxdistalphaj1.0](c ) suggests that the tool borrowed from traditional anderson - transition studies for time - independent systems can be still useful here . furthermore ( probably more interestingly ) , the fitting in fig . [ jx - jxdistalphaj1.0](c ) also unexpectedly reveals a big difference from what can be expected from eq . ( [ dq_goe ] ) with @xmath33 and @xmath71 : here @xmath68 instead of @xmath72 . therefore , an intriguing difference between time - independent critical systems and periodically driven critical systems is observed here . for the @xmath52 model , with @xmath73 and @xmath74 in the representation of odd - parity basis states defined in the text . the size of the floquet matrix ensemble is important for numerical simulation . in order to construct the necessary ensemble , we consider a range of @xmath75 , i.e. , @xmath76 , yielding respectively @xmath77 and @xmath78 matrices for the different values @xmath36 . ( b ) @xmath79 , the mean value of @xmath12 averaged over all floquet eigenstates , as a function of @xmath60 . the slope of the fitting line gives @xmath80 . ( c ) logarithm of @xmath66 $ ] as a function of @xmath60 , where @xmath36 is the dimension of the odd - parity hilbert subspace . filled circles are our numerical results for the @xmath52 model , and the solid line is the fitting of the numerical results using the empirical formula given in eq . ( [ var_goe ] ) with @xmath81 , @xmath82 and @xmath83 . the plotted variables here and in all other figures are dimensionless.,width=283,height=264 ] as mentioned above , for this parameter regime the butterfly patterns in the floquet spectra have dissolved almost completely . we present the associated eigenstate statistics in fig . [ jx - jxdistalphaj10.0 ] . in fig . [ jx - jxdistalphaj10.0](a ) , we show the distribution profile of @xmath12 for different @xmath36 . in contrast to the previous case shown in fig . 1(a ) , @xmath84 $ ] is now much smoother ( essentially only one peak is left ) . from the same panel , we also see a systematic left - shift of the distribution function as @xmath36 increases . this systematic left - shift leads to an evident linear behavior of the average value of @xmath12 as a function of @xmath61 , as shown in fig . [ jx - jxdistalphaj10.0](b ) . the slope of the fitting line in fig . [ jx - jxdistalphaj10.0](b ) gives the fractal dimension @xmath85 . comparing this result with that in fig . 1(b ) , one sees that though the fractal dimension of the floquet spectra increases due to increasing @xmath57 @xcite , the fractal dimension of the associated eigenstates may decrease . in fig . [ jx - jxdistalphaj10.0](c ) , we examine the variance of @xmath12 as a function of @xmath60 ( again , for the odd - parity subspace ) . same as in fig . 1(c ) , we fit our results with the empirical formula given in eq . ( [ var_goe ] ) . the fitting in fig . [ jx - jxdistalphaj10.0](c ) is better than that in fig . 1(c ) , consistent with the fact that the distribution of @xmath12 is quite smooth here . the fitting in fig . [ jx - jxdistalphaj10.0](c ) gives @xmath86 , and @xmath83 , where the value of @xmath11 is found in fig . [ jx - jxdistalphaj10.0](b ) . the finding that @xmath22 is not equal to @xmath72 but @xmath11 again strengthens our early observation from fig . 1 . for the @xmath52 model with @xmath59 is presented for the odd - parity subspace . other parameters are the same as in fig . [ jx - jxdistalphaj1.0 ] . ( b ) @xmath87 is plotted as a function of @xmath60 . the slope of the fitting line gives @xmath88 . ( c ) logarithm of @xmath66 $ ] as a function of @xmath60 , where @xmath36 is the dimension of the odd - parity hilbert subspace . filled circles are our numerical results for the @xmath52 model , and the solid line is the fitting of the numerical results using the empirical formula given in eq . ( [ var_goe ] ) , with @xmath89 , @xmath90 and @xmath83.,width=283,height=264 ] for the @xmath53 model , with @xmath91 and @xmath92 . the numbers in the brackets are the size of the floquet matrix ensemble . the dimension of the hilbert space is given by @xmath93 . ( b ) @xmath79 , the ensemble mean value of @xmath12 averaged over all floquet eigenstates , as a function of @xmath14 . the slope of the curve of @xmath79 vs @xmath14 gives @xmath94 . ( c ) logarithm of @xmath95 $ ] vs @xmath96 . filled circles are numerical results for the @xmath53 model , and the solid line is the fitting of the numerical results using the empirical formula given in eq . ( [ var_goe ] ) , with @xmath97 , @xmath98 , and @xmath99.,width=302,height=283 ] to verify if our findings above are general , we now turn to the @xmath53 model [ eq . ( [ floquet_cue ] ) ] . due to the lack of time - reversal symmetry here , this case can be regarded as an example of critical circular - unitary - ensemble " ( cue ) statistics . all the eigenstates of the floquet operator @xmath49 will be considered . for this regime where the butterfly patterns of the floquet spectra can be clearly seen , fig . [ jx - jydistalphaj1.0](a ) displays the distribution of @xmath12 for different hilbert space dimension @xmath93 . analogous to the previous case with time - reversal symmetry , @xmath28 $ ] displays interesting fluctuations . as @xmath7 increases , @xmath28 $ ] undergoes changes in its profile , shifts its left tail , but also maintains many features . in fig . [ jx - jydistalphaj1.0](b ) we obtain again a nice linear scaling behavior of @xmath79 with @xmath14 . from the slope of the linear scaling , we obtain the fractal dimension @xmath100 . this @xmath11 value is different from that for the @xmath52 model with the same values of @xmath57(note that the spectral statistics for the @xmath53 model also differs from that for the @xmath52 model @xcite ) . same as in fig . [ jx - jxdistalphaj1.0](c ) , in fig . [ jx - jydistalphaj1.0](c ) we study the variance of @xmath12 [ now denoted @xmath18 ] as a function of @xmath96 , using the fitting formula given in eq . ( [ var_goe ] ) . the fitting , though with clear fluctuations , yields that @xmath95 $ ] scales as @xmath102 , with @xmath103 , @xmath104 , and @xmath105 [ @xmath11 value obtained from fig . [ jx - jydistalphaj1.0](b ) ] . remarkably , though eq . ( 4 ) with @xmath33 and @xmath106 ( because of the lack of time - reversal symmetry ) predicts @xmath107 , here we have @xmath105 instead . the important common feature shared by the @xmath53 model and the @xmath52 model is thus the missing of a factor of 2 in the numerically obtained @xmath22 value as compared with the empirical formula for time - independent critical systems . this interesting finding also supports the use of eq . ( [ var_goe ] ) as a tool for understanding floquet eigenstate statistics . our numerical observations here will be further strengthened by a random matrix model . for the @xmath53 model with @xmath59 is presented for different hilbert space dimension @xmath93 . other parameters are the same as in fig . [ jx - jydistalphaj1.0 ] . ( b ) @xmath79 is plotted as a function of @xmath96 . the slope of this linear curve gives @xmath108 . ( c ) logarithm of @xmath95 $ ] vs @xmath96 . filled circles are our numerical results for the @xmath53 model , and the solid line is the fitting of the numerical results using the empirical formula given in eq . ( [ var_goe ] ) , with @xmath109 , @xmath110 and @xmath99.,width=283,height=264 ] just like the @xmath52 model , in this regime the butterfly patterns of the floquet spectra have dissolved . the statistical properties of the floquet eigenstates are shown in fig . [ jx - jydistalphaj10.0 ] . in fig . [ jx - jydistalphaj10.0](a ) , the distributions of @xmath12 is seen to be much smoother than those seen in fig . this is somewhat expected from our early findings in the @xmath52 model . figure [ jx - jydistalphaj10.0](b ) shows a linear scaling of @xmath13 vs @xmath96 , with its slope giving @xmath108 . in fig . [ jx - jydistalphaj10.0](c ) , we study the variance of @xmath12 as a function of @xmath96 , as compared with the empirical formula given in eq . ( [ var_goe ] ) : the fitting with the empirical formula is excellent , yielding @xmath111 , and @xmath99 , where the value of @xmath11 is determined in fig . [ jx - jydistalphaj10.0](b ) . once again , here we find @xmath99 instead of @xmath112 [ as suggested by eq . ( 4 ) with @xmath106 ] . in studies of time - independent critical systems , the prbm model at criticality @xcite has proved to be fruitful . the prbm is an ensemble of random hermitian matrices whose matrix elements @xmath114 are independently distributed gaussian random numbers with mean @xmath115 and the variance satisfying @xmath116^{-1}. \label{varh_ij}\ ] ] the case @xmath117 represents the critical point and @xmath118 is a parameter characterizing the ensemble . a straightforward interpretation of this model is that it describes a one - dimensional sample with random long - range hopping , with the hopping amplitude decaying as @xmath119 . motivated by our results above for critical floquet states , we aim to propose a class of random unitary matrices , whose floquet eigenstate statistics can show some general aspects of critical statistics and can be used to shed some light on actual dynamical systems . our natural starting point for generating such random unitary matrices are the hermitian prbm . to generate a random unitary matrix from a hermitian matrix in the prbm ensemble , we employ the algorithm by mezzadri , whose original motivation is to generate cue random matrices @xcite from general gaussian random matrices . for the sake of completeness , we have presented a description of mezzadri s algorithm in appendix [ app1 ] . for our purpose , that is , to generate a critical random unitary matrix , we first set the starting point of mezzadri s algorithm as a prbm ensemble at the critical point ( @xmath120 ) . we then generate an ensemble of random unitary matrices ( denoted @xmath121 ) of the cue class . significantly , because of the use of prbm as the input for mezzadri s algorithm , we find that the variance of the matrix elements @xmath122 thus obtained also satisfies a power - law , i.e. , @xmath123^{-1}. \label{var_unitary}\ ] ] here the parameter @xmath124 is a common prefactor of the matrix elements , which can be determined by the unitary condition . the parameters @xmath125 and @xmath126 in eq . ( [ var_unitary ] ) depend on the parameters @xmath127 and @xmath128 of the prbm used . as three computational examples , panels ( d)-(f ) of fig . [ var_prbum ] present the dependence of @xmath129 $ ] upon @xmath130 , for three ensembles of random unitary matrices we generated , with sizes @xmath131 , and @xmath132 . if the scaling of @xmath133 with @xmath134 is indeed a power law as described by eq . ( [ var_unitary ] ) , then one should see a linear dependence of @xmath129 $ ] in @xmath130 . this is indeed the case in figs . [ var_prbum](d)-(f ) . note that the deviations in figs . [ var_prbum](d)-(f ) from the fitting straight lines at very large values of @xmath130 are due to two trivial reasons . first , for very large @xmath134 , the value of @xmath133 is vanishingly small and hence @xmath135 $ ] becomes extremely large , thus yielding large fluctuations . second and more importantly , for a fixed matrix size , if @xmath134 is very large , then the available number of matrix elements become insufficient for good statistics . indeed , as the matrix size increases from @xmath136 to @xmath137 , it is seen from figs . [ var_prbum](d)-(f ) that the validity window of the linear fitting gradually extends to larger values of @xmath130 . on @xmath130 is plotted , where @xmath138 represents a matrix element at the @xmath139th row and @xmath140th column . here the prbm ensemble as the input is set at the critical point @xmath120 with the parameter @xmath141 . panels ( a)-(c ) are for prbum - coe , with the dimension @xmath142 and @xmath132 , respectively . the fitting function of eq . ( [ var_unitary ] ) ( solid lines ) gives @xmath143 , @xmath144 , and @xmath145 . panels ( d)-(f ) are for prbum - cue , with the dimension @xmath146 and @xmath132 , respectively . the fitting function of eq . ( [ var_unitary ] ) ( solid lines ) yields @xmath147 , @xmath148 , and @xmath149.,width=283,height=238 ] the random unitary matrices generated in the above manner , with their matrix elements satisfying the power - law scaling of eq . ( [ var_unitary ] ) , are defined as power - law random banded unitary matrix " of the cue type ( prbum - cue ) . as detailed in appendix a , one can then generate prbum of the coe type ( prbum - coe ) via @xmath150 . as shown in panels ( a)-(c ) of fig . [ var_prbum ] , the variance of the matrix elements of prbum - coe also obeys eq . ( [ var_unitary ] ) , with different values of @xmath125 and @xmath126 . to check whether the prbum - coe and prbum - cue ensembles show critical statistics , we analyzed their eigenstates , especially in terms of the distribution and the scaling of @xmath12 . it is found that as we tune the parameter @xmath128 of the prbm used in the algorithm , the resulting fractal dimensions @xmath11 can be also tuned continuously . for example , the @xmath11 value of prbum can be made close to that of our driven su(2 ) models . in particular , at @xmath141 , we obtain @xmath151 for prbum - coe and @xmath152 for prbum - cue , yielding @xmath153 and @xmath154 , respectively . these two @xmath11 values are quite close to the @xmath11 values of the @xmath52 and @xmath53 models found in fig . 1 and fig . 3 . below we describe these findings in detail . obtained for prbum - coe , with the matrix dimension @xmath155 and @xmath156 . the numbers in the brackets give the size of the ensemble . ( b ) same as in fig . [ jx - jxdistalphaj1.0](b ) and fig . 2(b ) , yielding @xmath157 . ( c ) same as in fig . [ jx - jxdistalphaj1.0](c ) and fig . 2(c ) , but with @xmath36 replaced by @xmath7 . the fitting curve gives @xmath158 , @xmath159 , and @xmath68.,width=283,height=264 ] this random unitary matrix ensemble is intended to model a critical floquet operator with time - reversal symmetry . the results for prbum - coe generated from prbm with @xmath160 are shown in fig . [ coe_dist ] . in fig . [ coe_dist](a ) , we show the distributions of @xmath12 for different values of the matrix dimension @xmath7 ( which is the counterpart of @xmath36 in the @xmath52 model ) , with all the eigenstates of the prbum - coe ensemble considered . in contrast to the @xmath52 dynamical model with a small @xmath57 [ see fig . 1(a ) ] , @xmath84 $ ] here displays very smooth behavior . figure [ coe_dist](b ) depicts a nice linear relation between @xmath161 and @xmath14 . the slope of the straight line in fig . [ coe_dist](b ) gives the fractal dimension @xmath157 , a value close to that in the @xmath52 model with @xmath162 . as in fig . [ jx - jxdistalphaj1.0](c ) , fig . [ coe_dist](c ) shows the fitting of the variance of @xmath12 with @xmath7 , using eq . ( [ var_goe ] ) . interestingly , the values of the fitting parameters are found to be @xmath163 , @xmath164 , both are similar to those determined in fig . 1(c ) . more interestingly , this fitting shows that @xmath95 $ ] scales as @xmath102 , with @xmath68 . this supports our finding in fig . [ jx - jxdistalphaj1.0](c ) and fig . we have also studied other cases of prbum - coe using other prbm as the input of mezzadri s algorithm . for example , we find that if the parameter @xmath128 is set at @xmath165 , then the @xmath11 of the prbum - coe ensemble is around 0.24 , which is close to the @xmath11 value previously found in the @xmath52 model with @xmath166 . these results clearly support our use of prbum - coe to illuminate the critical eigenstate statistics in the @xmath52 model . this ensemble aims to model a critical floquet operator without time - reversal symmetry . all eigenstates of an ensemble of prbum - cue matrices are used for our statistical analysis . for @xmath160 , fig . [ cue_dist](a ) displays @xmath28 $ ] versus @xmath12 , showing again a smooth dependence . figure [ cue_dist](b ) shows the corresponding @xmath79 versus @xmath167 , which yields the fractal dimension @xmath168 . in fig . [ cue_dist](c ) , we fit the dependence of @xmath169 $ ] in @xmath96 , yielding @xmath95 \sim n^{-\gamma}$ ] , with @xmath105 [ instead of @xmath170 predicted by eq . this also confirms our early observations in the @xmath53 model . the values of the fitting parameters are found to be @xmath171 and @xmath172 , which are close to what we found in fig . [ jx - jydistalphaj1.0](c ) . we have also checked that if we perform analogous calculations for @xmath173 , then the @xmath11 value for the prbum - cue ensemble will be close to that found in fig . 4(b ) . given these results , we are led to the conclusion that prbum as proposed above do share some general aspects with periodically driven systems having critical eigenstate statistics . obtained for prbum - cue , with the matrix dimension @xmath174 and @xmath175 . the numbers in the brackets give the size of the ensemble . ( b ) same as in fig . [ jx - jydistalphaj1.0](b ) and fig . 4(b ) , yielding @xmath168 . ( c ) same as in fig . [ jx - jydistalphaj1.0](c ) and fig . 4(c ) , the fitting gives @xmath176 , @xmath172 , and @xmath105.,width=283,height=264 ] for the standard classically chaotic kicked top model , for @xmath177 and @xmath92 . the numbers in the brackets are the size of the floquet matrix ensemble . in constructing the ensembles we have considered a range of @xmath75 , i.e. , @xmath178 . ( b ) distributions of @xmath12 for the standard dyson s coe matrices , with the same matrix dimension as in the kicked top model and the same ensemble size . ( c ) analogous to fig . [ jx - jxdistalphaj1.0](b ) and fig . 7(b ) , the scaling behavior of @xmath79 vs @xmath61 is shown . open circles are numerical results for the kicked - top model , crosses are numerical results associated with coe random matrices , and the solid curve represents the theoretical prediction from the random matrix theory . the scaling shows that @xmath179 in the standard kicked top model , which is dramatically different from our observations made from the double - kicked top model.,width=302,height=264 ] finally , as a numerical control " experiment , we study the floquet eigenstate statistics of the standard kicked top model . this will help appreciate the difference between a normal driven system and a critical driven system , both of which can have a chaotic classical limit . consider then the following floquet operator for the standard kicked top model @xcite , @xmath180 which is just the last two factors of eq . ( [ floquet_coe ] ) , with the same parity symmetry and time - reversal symmetry as the @xmath52 model . in addition , we set the parameter @xmath181 at the same value as given in eq . ( [ parameter1 ] ) . we construct a statistical ensemble by considering a range of @xmath75 , i.e. @xmath178 ( with chaotic classical limits ) . we carry out the floquet eigenstate statistics in the odd - parity subspace , whose dimension is @xmath36 . because the classical limit is found to be chaotic , we compare the statistics with that associated with dyson s coe matrices in random matrix theory ( rmt ) . figure [ haake_top](a ) and ( b ) compare @xmath28 $ ] associated with @xmath182 with that obtained from coe matrices , for different @xmath36 . the difference between the actual dynamical system and the coe can hardly be seen . figure [ haake_top](c ) depicts @xmath183 as a function of @xmath61 , with the results of the kicked top ( open circles ) almost on top of those of coe matrices ( crosses ) . the solid line in fig . [ haake_top](c ) represents the theoretical curve for @xmath184 obtained from rmt , i.e. , @xmath185 . the agreement between numerical coe results , analytical rmt result , and the kicked top system as a classically chaotic dynamical system is almost perfect . from the curve shown in fig . [ haake_top](c ) , it is clear that @xmath11 here is unity and as such the system does not show critical behavior . this non - critical behavior indicates that the floquet states of the kicked top model are essentially random states , a feature fundamentally different from our double - kicked top system that has a butterfly spectrum and critical statistics in the floquet eigenstates . it is also interesting to note that in fig . [ haake_top](a ) and ( b ) , as @xmath36 increases , @xmath28 $ ] becomes narrower and develops higher peaks . this is an indication that , unlike the critical cases studied above , @xmath63 for the standard kicked top model approaches a dirac - delta type singular function with zero width ( i.e. @xmath186 ) as @xmath36 increases . in this numerical study we have examined the statistics of the floquet eigenstates of a recently proposed double - kicked top model with multifractal floquet spectra . following the methodologies used in studies of anderson transition in time - independent systems , we have shown that the floquet eigenstates associated with multifractal floquet spectra also display critical behavior . in particular , we focus on the distribution of @xmath12 and examine how the quantity @xmath13 averaged over all states scales with the dimension of the hilbert space @xmath7 . it is shown that @xmath187 scales linearly with @xmath14 , with the slope of this linear scaling giving the fractal dimension @xmath11 of the floquet eigenstates . the values of @xmath11 are found to be far from unity ( as a comparison , we showed that similar analysis for a standard kicked top with a chaotic classical limit yields @xmath179 ) , constituting strong evidence that the floquet eigenstates are fractal and hence lying between localized and delocalized states . though we have worked on @xmath5 only , we note that similar analysis can be done for @xmath188 defined in eq . ( [ pq ] ) . one may then define a generalized fractal dimension @xmath189 and further establish the multifractal nature of the floquet eigenstates . the variance of @xmath12 , denoted @xmath18 for a hilbert space of dimension @xmath7 , is also examined . in anderson - transition studies with prbm , @xmath18 is known to scale as @xmath102 with @xmath190 for one - dimensional systems , where @xmath191 for a system with ( without ) time - reversal symmetry . by contrast , in our critical driven system , @xmath18 is seen to scale similarly , but with @xmath192 . this reflects an interesting difference between time - dependent systems and time - independent systems . indeed , eigenstates of prbm are to model those of critical hermitian operators , whereas floquet eigenstates of a critical driven system should be understood in terms of critical unitary operators . to justify this understanding , we have introduced a random unitary matrix ensemble called prbum , with the variance of the matrix elements of the unitary matrices following a power - law distribution . we show that the eigenstates of prbum share many critical statistical features with the double - kicked top model . most important , the variance of @xmath12 of prbum does scale as @xmath193 , which is the same as in the double - kicked top model as a critical driven system . we hence anticipate that this scaling property of the variance of @xmath31 may be general in critical driven systems . these results complement the spectral results in ref . @xcite and should motivate further mathematical and theoretical studies in critical driven systems . j.w . acknowledges support from national natural science foundation of china ( grant no.10975115 ) , and j.g . is supported by the nus start - up fund ( grant no . r-144 - 050 - 193 - 101/133 ) and the nus yia " ( grant no . r-144 - 000 - 195 - 101 ) , both from the national university of singapore . this is a simple and numerically stable algorithm to generate the cue matrices from an ensemble of complex random matrices @xmath194 , whose elements are gaussian distributed random numbers with mean zero and variance _ unity_. in particular , applying the gram - schmmidt ortho - normalization method to the columns of an arbitrary complex matrix @xmath195 , one can factorize @xmath195 as : @xmath196 where @xmath197 is a unitary matrix and @xmath198 is an invertible upper - triangular matrix . one can easily prove that the above factorization is not unique . because of this non - uniqueness , the random unitary matrices @xmath199 are not distributed with haar measure @xcite , i.e. , the @xmath199 matrices are not uniformly distributed over the space of random unitary matrices . fortunately , this factorization can still be made unique by imposing a constraint on the @xmath198 matrices . by some group theoretical arguments , it was shown @xcite that if one finds a factorization such that the elements of main diagonal of @xmath198 become real and strictly positive , then @xmath199 matrices would be distributed with haar measure and hence form cue . following these results , the major steps of mezzadri s algorithm are the following . first , we start with an @xmath200 complex gaussian random matrix @xmath195 . second , we factorize @xmath195 by any standard @xmath201decomposition routine such that @xmath202 . third , we create a diagonal matrix @xmath203 where @xmath204 are the diagonal elements of @xmath198 . as a final step , we define @xmath205 and @xmath206 . by construction , the diagonal elements of @xmath207 are always real and strictly positive , and as such @xmath208 would be distributed with haar measure and can be used to form the desired cue . the symmetric coe matrices can be constructed from the cue matrices in a very simple manner . in particular , let @xmath121 be a member of the cue generated above , then it can be shown that @xmath150 will be a member of coe . for the generation of prbum advocated in this work , we propose to replace @xmath195 in the first step by a member in the prbm ensemble that models anderson transition . though there is no mathematical theory for our procedure , the uniformly distributed eigenphases ( not shown here ) of our prbum ensemble thus generated suggest its uniform distribution . see also some related papers : r. lima and d. shepelyansky , phys . * 67 * , 1377 ( 1991 ) ; r. artuso _ et . al _ , _ ibid _ * 69 * , 3302 ( 1992 ) ; r. ketzmerick , k. kruse , and t. geisel , _ ibid _ * 80 * , 137 ( 1998 ) ; i. i. satija , phys . e * 66 * , 015202 ( 2002 ) ; j. b. gong and p. brumer , phys . rev . lett . 97 , 240602 ( 2006 ) . q. zhang , p. hnggi , and j. b. gong , , 053607 ( 2008 ) ; new journal of physics , * 10 * , 073008 ( 2008 ) . q. xie and w. hai , eur . j. d * 33 * , 265 ( 2005 ) ; m.p . strzys , e.m . graefe , and h.j . korsch , new j. phys . * 10 * , 013204 ( 2008 ) ; j. b. gong , l. molina - morales , and p. hnggi , , 133002 ( 2009 ) .

[ [ general - context . ] ] general context . + + + + + + + + + + + + + + + + linear boundary problems are a crucial concern of applied mathematics @xcite . one might thus expect a rich _ algebraic theory _ with _ symbolic algorithms _ , so as to support the exact solution and manipulation of a suitable class of boundary problems . alas , this is not the case yet . there may be two reasons for this . one is that the classical algebraic theory of ( nonlinear ) differential equations , the _ differential algebra _ built up by ritt @xcite and kolchin @xcite , does not lend itself easily to boundary conditions : the elements of a differential ring or field are not functions but abstract objects that can not be evaluated at a boundary point . the other reason is of a pragmatic nature . if a differential equation can be solved at all , one tends to delegate boundary conditions to _ adhoc postprocessing _ steps that would adapt the integration constants / functions of the `` general solution '' . while this may be viable for linear ordinary differential equations ( lodes ) , the notion of general solution is much less useful in the case of partial differential equations ( lpdes ) : for example , there is not much point in solving the @xmath0 per se , but there are useful representations of those solutions that satisfy dirichlet boundary conditions on the unit disc ( poisson kernel ) . so even though we will be dealing only with lodes in this paper , the larger context of lpdes should be kept in mind . in the admittedly modest case of _ boundary problems for lodes _ , an algebraic theory was set up in @xcite . the decisive step was to expand the well - known structure of a differential ring , complementing its derivation by a compatible baxter operator ( integral operator ) . this solves at once two problems : it provides the algebraic structure needed for expressing green s operators ( solution operators for boundary problems ) , and it yields an evaluation ( a multiplicative functional ) . the notion of evaluation is the key for imposing boundary conditions on the otherwise abstract objects of a differential ring ; one supplies as many evaluation as needed ( in addition to the one coming from the baxter operator ) . in @xcite , the basic theory of @xcite was refined and generalized . moreover , the authors have introduced a multiplicative structure on boundary problems that will also be crucial for our buildup of the heaviside calculus . the multiplication of boundary problems corresponds to the composition of their green s operators in reverse order ; see equation . the framework of boundary problems of @xcite was _ implemented _ several times : the first implementation was coded in mathematica / th@xmath1orem@xmath2 as an external package for boundary problems with constant coefficients . this version was superseded by a new implementation as an internal th@xmath1orem@xmath2 functor for generic integro - differential algebras in @xcite . now in the third generation , the most recent implementation is the maple package _ intdiffop _ @xcite , which provides additional support for singular boundary problems ( all previous packages being restricted to regular boundary problems ) . [ [ heaviside - calculus . ] ] heaviside calculus . + + + + + + + + + + + + + + + + + + + our main concern in this paper is to build a bridge from the framework of symbolic boundary problems to the classical heaviside calculus @xcite . since o. heaviside s main idea , subsequently made rigorous by j. mikusiski , was to treat the differential operator as a _ symbolic multiplier _ , it is perhaps natural to reflect on possible connections to contemporary symbolic methods for differential equations . but why to symbolic boundary problems ? the answer lies in heaviside s so - called _ fundamental formula _ , the algebraic analog of a well - known relation for the laplace transform . if @xmath3 is the `` symbolic multiplier '' representing the differential operator and @xmath4 is a suitable function , then one has @xmath5 . the point is that @xmath3 , which is supposed to be invertible in this calculus , must somehow `` remember '' the integration constant that is lost in differentiation . we will come back to this point in more detail ( section [ sec : methorious - functions ] ) . at this point it suffices to say that an appropriate action of @xmath3 on functions involves an evaluation operator . it is used in the heaviside - mikusinski calculus for incorporating the initial values associated to a differential equation , thus yielding at once the solution of the whole initial value problem . it is thus natural staying in the frame of lodes to ask if one can build up a more general calculus that would allow to incorporate multiple boundary values , given by several evaluation operators . the basic ideas of heaviside and mikusiski have been vastly extended , specifically by l. berg @xcite and i. dimovski @xcite . the latter has also broached a question closely related to the one raised above , namely the setup of _ nonlocal convolutions _ @xcite and _ custom - tailored convolutions _ for boundary problems @xcite . as far as we know , however , there are no direct generalizations of the fundamental formula from one to more evaluation operators acting on univariate functions . [ [ basic - setup . ] ] basic setup . + + + + + + + + + + + + our own approach is different in many respects . it does not qualify as an _ operational calculus _ in dimovski s understanding @xcite , for whom its crucial feature is that `` operators '' and `` operands '' are merged in a single data structure , whereas algebraic analysis keeps the module of operand functions separate from the ring of operators . in this sense , we follow the line of algebraic analysis since we distinguish the ring of methorious operators ( section [ sec : localization ] ) from the module of methorious functions ( section [ sec : methorious - functions ] ) . moreover , our approach is genuinly algebraic while the hallmark of the heaviside - mikusiski tradition is an ingenious mix of algebra and analysis . nevertheless , we believe that our setup is close in spirit to the original heaviside - mikusiski setup , enjoying the following attractive features : * its _ basic philosophy _ follows closely in heaviside s footsteps , making the differential operator invertible by `` remembering '' suitable boundary data . * it provides an algebraic structure that accommodates boundary problems for an _ arbitrary number of evaluation points _ as well as _ nonlocal conditions_. * all boundary problems are covered _ uniformly _ , so one need not set up custom - tailored multiplications ( convolutions ) for each type of boundary conditions one wants to consider . * it illuminates the passage from one to several evaluations algebraically : localization takes place in a _ noncommutative ring_. in contrast , convolution algebras are commutative by definition @xcite . * the construction is _ generic in the coefficient algebra _ ( it works for the class of umbral integro - differential algebras ) . from the viewpoint of analysis , however , the class of umbral coefficient algebras is rather limited compared to mikusiski s setup ( continuous or even @xmath6 functions on the positive half - axis ) . while this might be relaxed by suitable limit considerations , this is not in the interest of the present paper , where we want to focus on the algebraic aspects . our goal in this paper is to build a _ first bridge _ between the algebraic theory of boundary problems @xcite and heaviside s tactic of using `` symbolic fractions '' ( localization in a suitable convolution ring ) for integrating differential equations with initial / boundary data . in fact , it gives a new justification ( see after proposition [ prop : greensop - localization ] ) for a notational device initiated in @xcite : since then we have written @xmath7 for the green s operator associated to a boundary problem @xmath8 , originally chosen in view of the anti - isomorphism mentioned above . the work in this paper may also be seen as an answer to an ( implicit ) question originating from @xcite . the setting there was restricted to @xmath9 $ ] coefficients and treated via ore algebras . the object of central interest was the _ integro - differential weyl algebra _ , which has shown to have two important quotient algebras : the ring of localized differential operators @xmath10[x]$ ] and the usual ring of integro - differential operators @xmath9[{\partial } , { { \textstyle \varint}}]$ ] . the former has @xmath11 as a two - sided inverse , but no action on @xmath9 $ ] ; the latter has the action , but @xmath12 is only a right inverse of @xmath11 . now the question arises : can one build up a structure , with an action on @xmath9 $ ] or other coefficient algebras , such that the derivation has a two - sided inverse ? we show in this paper that the answer is affirmative , provided the derivation is enhanced in a way similar to heaviside s symbolic multiplier @xmath3 . as always in algebra , localizing a ring sheds new light on its structure , especially in the noncommutative setting ( here the theory of fraction rings is somewhat more delicate due to the ore condition , see section [ sec : localization ] ) . we hope this will also be the case for our present construction . but one must bear in mind that the subject has only been touched and various issues remain in a preliminary and unsatisfactory state ( see the conclusion ) . nevertheless , a wealth of new relations has been uncovered , and we are confident that they will allow interesting generalizations and refinements . [ [ structure - of - the - paper . ] ] structure of the paper . + + + + + + + + + + + + + + + + + + + + + + + we start out by summarizing the basic theory of symbolic boundary problems , giving special emphasis to their monoid structure ( section [ sec : monoid - bp ] ) . our construction is based on a certain kind of boundary conditions , which we have called `` umbral '' because of certain relations to the umbral calculus ( section [ sec : umbral - bc ] ) . after this preparation , we tackle the task of localization for an integro - differential algebra with an umbral character set ( section [ sec : localization ] ) . for the resulting ring of fractions , we construct a module of functions on which it acts naturally ( section [ sec : methorious - functions ] ) . we conclude with some remarks about possible extensions and generalizations ( section [ sec : conclusion ] ) . [ [ notation . ] ] notation . + + + + + + + + + all rings are with unit but not necessarily commutative . a domain is a ( commutative or noncommutative ) ring without zero divisors . the zero vector space of any dimension will be denoted by @xmath13 . we use the notation @xmath14 for indicating that a vector space @xmath15 is a subspace of a vector space @xmath16 . our basic setting is that of @xcite . we review the main results here for making our present treatment more self - contained and for introducing various pieces of notation and terminology in their proper places . we start from the notion of _ integro - differential algebra _ , a natural generalization of differential algebras that permits the algebraic formulation of linear boundary problems . [ def : intdiffalg ] we call @xmath17 an _ integro - differential algebra _ over @xmath18 if @xmath19 is a commutative @xmath18-algebra with @xmath18-linear operations @xmath11 and @xmath12 such that the three axioms @xmath20 are satisfied , where @xmath21 is the usual shorthand notation for @xmath11 . this definition differs from definition 4 in @xcite as it uses as the _ differential baxter axiom _ rather than the variant @xmath22 used earlier . in fact , one checks immediately that for commutative @xmath19 both axioms are equivalent ( assuming the other axioms ) . the advantage of is that it is more symmetric and that it also sufficient for the noncommutative case , where must be supplemented by the corresponding dual axiom . nevertheless , _ we will from now on tacitly assume that all integro - differential algebras are commutative , as in @xcite . _ incidentally we note the following _ convention on precedence _ : the scope of an integral sign @xmath12 covers all factors to its right , unless otherwise specified . therefore the above term @xmath23 is to be parsed as @xmath24 . this helps avoiding the proliferation of parentheses in nested integrals . the differential baxter axiom is in general stronger than the following _ pure baxter axiom _ @xmath25 which is the defining axiom for the so - called rota - baxter algebras @xcite . ( in fact , one may relax the conditions on @xmath19 to include noncommutative algebras over arbitrary commutative rings with one . moreover , one may add a so - called weight term for incorporating the discrete setting where @xmath12 is , for example , the operator of partial summation . in that case , one must of course also adapt to account for the weight . remarkably , the differential baxter axiom is the same with or without weight . ) every integro - differential algebra comes with a multiplicative projector , namely the _ evaluation _ @xmath26 . in fact , the strong baxter axiom is equivalent to the weak one combined with the multiplicativity of @xmath27 . the presence of the character @xmath27 leads to the direct sum decomposition @xmath28 where @xmath29 is the usual _ ring of constants _ while @xmath30 is called the _ ideal of initialized functions _ ( one checks immediately that the multiplicativity of @xmath27 is equivalent to @xmath31 being an ideal ) . so @xmath27 is the projector onto @xmath32 along @xmath31 . moreover , both @xmath11 and @xmath12 are @xmath32-linear rather than just @xmath18-linear ( for @xmath11 this is of course trivial , but for @xmath12 this is again equivalent to the multiplicativity of @xmath27 ) . in this paper , we want to restrict ourselves to boundary problems for lodes . for reflecting this property in the ( integro ) differential structure , in @xcite we have called a differential algebra @xmath33 ordinary if @xmath34 . _ we will henceforth assume that all integro - differential algebras are ordinary in this sense . _ restricting ourselves to ordinary integro - differential algebras has a number of pleasant implications . first of all , the evaluation is now a multiplicative linear operator @xmath35 , meaning a _ character _ ( in the sense `` multiplicative linear functional on an algebra '' ) . another consequencence , which will be of some importance later , is that the polynomials behave as usual . for any integro - differential algebra @xmath19 , we set @xmath36 . now the pure baxer axiom ensures that @xmath37 , and so on . it turns out that the elements of @xmath9 $ ] are `` really '' polynomials , and they satisfy the usual differential equations ( see before eq . ( 6 ) and also eq . ( 8) in @xcite for more details ) . [ prop : pol - ring ] let @xmath19 be an integro - differential algebra over @xmath18 , and let @xmath9 $ ] be the subalgebra generated by @xmath38 . then @xmath9 $ ] is isomorphic to the univariate polynomial ring over @xmath18 . if @xmath19 is an ordinary integro - differential algebra , we have moreover @xmath39,\ ] ] and @xmath9 \le { \mathcal{f}}$ ] is an integro - differential subalgebra . [ ex : standard ] the _ standard example _ of an ( ordinary ) integro - differential algebra is @xmath40 with @xmath41 in the usual sense and @xmath42 . this integro - differential algebra contains many important integro - differential subalgebras , for example the analytic functions @xmath43 , the exponential polynomials , and of course the polynomial ring @xmath44 $ ] . an important integro - differential subalgebra of @xmath43 is formed by the holonomic power series @xcite . given a differential algebra @xmath33 , one may form the ring of differential operators @xmath45}$ ] . it is therefore natural to expect a similar ring of _ integro - differential operators _ @xmath46}$ ] , which contains differential operators ( needed for specifying differential equations ) as well as integral operators ( needed for specifying green s operators ) . the evaluation @xmath27 can be used for writing a condition like @xmath47 in the form @xmath48 . but for boundary problems one usually needs more than one evaluation ( see example [ ex : second - order ] below ) . for algebraizing general boundary conditions , we start from a _ character set _ @xmath49 , meaning each @xmath50 is a multiplicative linear functional just as @xmath27 is . in fact , we will always assume @xmath51 . based on the characters of @xmath49 , we can from arbitrary _ stieltjes conditions _ ( * ? ? ? 14 ) ; their normal form is @xmath52 with @xmath53 and @xmath54 almost all zero . the summands with @xmath12 make up the so - called global part , those without the local part . the _ order _ of a boundary condition is the largest @xmath55 such that @xmath56 in the normal form . in the standard setting of example [ ex : standard ] , a typical stieltjes condition like @xmath57 is encoded as @xmath58 with @xmath59 , where we have written @xmath60 for the standard character @xmath61 . henceforth we shall always consider example [ ex : standard ] with the character set @xmath62 . for each character set @xmath49 one may then build up the corresponding _ operator ring _ @xmath63}$ ] . the details are given in @xcite ; at this point it suffices to note that the operator ring is given by the quotient of a free algebra modulo a certain ideal of relations like the leibniz rule , and also as well as . the resulting operator ring decomposes nicely into a direct sum ; see ( * ? ? ? 17 ) for a proof . [ prop : intdiffop - direct - sum ] for any integro - differential algebra @xmath17 and character set @xmath49 , we have the decomposition @xmath63 } = { { \mathcal{f}}[{\partial}]}{\dotplus}{{\mathcal{f}}[{{\textstyle \varint}}]}{\dotplus}(\phi)$ ] . here @xmath64 is the two - sided ideal generated by @xmath49 in the operator ring @xmath65}$ ] . we refer to its elements as _ boundary operators _ since they may also be described in terms of stieltjes conditions : one checks first that the collection of stieltjes conditions can be characterized as the right ideal @xmath66 . it turns out that @xmath64 is then the left @xmath19-module generated by @xmath67 , so every boundary operator can be written as @xmath68 for some functions @xmath69 and stieljes conditions @xmath70 . in particular , every stieltjes condition is also a boundary operator . the decomposition @xmath63 } = { { \mathcal{f}}[{\partial}]}{\dotplus}{{\mathcal{f}}[{{\textstyle \varint}}]}{\dotplus}(\phi)$ ] reflects the basic needs for an _ algebraic formulation of boundary problems _ : we need @xmath45}$ ] for the differential equation , @xmath64 for the boundary conditions , and @xmath71}$ ] for the solution ( green s operators ) . let us see how these ingredients look like for the simplest possible boundary problem in the standard setting of example [ ex : standard ] . [ ex : second - order ] the boundary problem @xmath72 can be encoded in the standard setting @xmath40 by the differential operator @xmath73}$ ] and the boundary conditions @xmath74 . here we can choose @xmath75 or any character set containing that . the green s operator can be written as @xmath76}$ ] , where we use the standard abbreviations @xmath77 and @xmath78 . note that @xmath79 is the integral from @xmath80 to @xmath81 , and @xmath82 the integral from @xmath81 to @xmath83 . in the @xmath6 setting , @xmath79 and @xmath82 are adjoint operators . the algebraic description in terms of the two functionals @xmath84 and @xmath85 contains some arbitrariness since we can clearly form any linear combination of the two functionals without changing the solution operator @xmath86 . in general , a _ boundary space _ is a finite - dimensional subspace of @xmath87 generated by stieltjes conditions . so the boundary space of example [ ex : second - order ] is @xmath88 $ ] . the lattice of all boundary spaces will be denoted by @xmath89 . we will usually restrict the coefficient functions of the differential operators to a differential subalgebra @xmath90 so as to ensure solutions in the ambient algebra @xmath19 . this means we require that @xmath91 be _ saturated _ for @xmath19 in the sense of ( * ? ? ? _ whenever an integro - differential algebra @xmath17 is specified , it is understood that a saturated coefficient algebra @xmath91 is set aside . _ [ def : boundary - problem ] a _ boundary problem _ is a pair @xmath92 consisting of a monic differential operator @xmath93 $ ] and a boundary space @xmath94 . so the boundary problem in example [ ex : second - order ] is given by @xmath95)$ ] . conversely , we can think of a boundary problem @xmath92 as finding a solution @xmath96 such that @xmath97 for an arbitrary forcing function @xmath98 . of course it suffices that @xmath99 ranges over any @xmath18-basis of @xmath100 ; this is how the boundary conditions are normally given in the first place . the _ traditional formulation _ in terms of @xmath101 and @xmath4 is more intuitive , but it conceals the fact that we are really working in the operator ring @xmath63}$ ] . clearly every differential operator @xmath102 has a certain order written as @xmath103 , which as in example [ ex : second - order ] usually coincides with the number of given boundary conditions ( the dimension of @xmath100 ) . in fact , this a necessary but in general not sufficient condition for the boundary problem to be _ regular _ in the sense that there is a unique solution @xmath96 for every given forcing function @xmath98 . in turns out that this is equivalent to the following formulation in terms of the relevant spaces . [ def : regular - bp ] a boundary problem @xmath92 is called _ regular _ if @xmath104 , and _ singular _ otherwise . here we have written @xmath105 for the _ orthogonal _ , meaning the space of all @xmath98 such that @xmath106 for all @xmath107 . in other words , @xmath105 is the space of admissible functions : those that satisfy the given boundary conditions . the orthogonal , together with the analogous notion @xmath108 for subspaces @xmath109 , establishes a galois connection between @xmath19 and its dual @xmath87 . this is similar to the situation in algebraic geometry with its galois connection between affine varieties and radical ideals . see @xcite for more details . the above definition of regularity is not suitable for algorithmic purposes . however , it turns out to be equivalent to the following explicit _ regularity test _ 6.1 ) , which is well - known in the special case of local boundary conditions @xcite . [ lem : regularity - test ] let @xmath92 be a boundary problem over @xmath19 and choose bases @xmath110 for @xmath100 and @xmath111 for @xmath112 . then @xmath92 is regular iff @xmath113 and the _ evaluation matrix _ @xmath114 \in k^{n \times n}$ ] is regular . regular boundary problems are exactly those that have a _ green s operator _ in the classical sense ( see @xcite on generalized green s operators for singular boundary problems ) . in our algebraic setting , green s operators can be defined within the operator ring @xmath63}$ ] . [ def : greens - operator ] let @xmath92 be a regular boundary problem over an integro - differential algebra @xmath19 . then @xmath115}$ ] is called its _ green s operator _ if @xmath116 and @xmath117 . the two conditions for green s operators are just that @xmath118 satisfies the differential equation @xmath119 as well as all boundary conditions @xmath58 for @xmath120 . since we have assumed a coefficient algebra @xmath91 saturated for @xmath19 , one can prove ( * ? ? ? 26 ) that _ every regular boundary problem _ has a green s operator in @xmath63}$ ] . in a leap of faith , we write @xmath121 just as in @xcite . the relation to actual inverses will become clear in proposition [ prop : greensop - localization ] . apart from the localization , there is another good reason for the notation @xmath122 . it turns out that boundary problems can be multiplied in such a way that @xmath123 is satisfied . for that purpose , we have defined the multiplication of boundary problems in the fashion of a semi - direct product by @xmath124 it is easy to check that one obtains a _ monoid _ @xmath125 \ltimes k\phi}$ ] in this way , with @xmath126 as the neutral element . furthermore , it turns out that the regular boundary problems form a submonoid @xmath125_{\phi}}\subset { { \mathcal{e}}[{\partial } ] \ltimes k\phi}$ ] , and means that @xmath127 is an anti - isomorphism from @xmath125_{\phi}}$ ] to the multiplicative monoid of green s operators . for a proof of these facts , we refer to ( * ? ? ? the monoid algebra @xmath128_{\phi}}$ ] will be called the _ ring of boundary problems _ ( we suppress the qualification `` regular '' since we will not investigate the singular case in the frame of this paper ) . its elements will supply the numerators in the localization to be constructed below ( section [ sec : localization ] ) . at this point we should also note that a regular boundary problem @xmath129 need not be _ well - posed _ in the sense that its green s operator @xmath86 is continuous in the standard setting of example [ ex : standard ] . for example , consider the regular boundary problem @xmath130 or @xmath131 in traditional notation . one checks immediately that @xmath132}$ ] is its green s operator , meaning the general solution is given by @xmath133 clearly , this green s operator @xmath86 is not continuous , at least not if one endows @xmath134 with the usual topology induced by @xmath135 . following hadamard , a well - posed boundary problem @xcite must be regular ( meaning the solution @xmath101 exists and is unique for each given @xmath4 ) as well as stable ( meaning the solution @xmath101 depends continuously on @xmath4 ) . in the example above , the source of the instability is clear we have imposed a second - order condition for a first - order differential equation . going back to the algebraic setting then , we call a boundary problem @xmath129 well - posed if it is regular and @xmath100 can be generated by stieltjes conditions whose order is smaller than that of @xmath102 ; otherwise @xmath92 is called _ ill - posed_. as one easily checks , the well - posed boundary problems form a submonoid of @xmath125_{\phi}}$ ] . the green s operator @xmath86 of a regular boundary problem @xmath92 of order @xmath136 factors naturally as @xmath137 where @xmath138 is the so - called _ fundamental right inverse _ of @xmath102 , defined as the green s operator of the initial value problem @xmath139 $ ] , and @xmath140 is the projector onto @xmath112 along @xmath105 . the fundamental right inverse @xmath138 exists since initial value problems are always solvable ( * ? ? ? * prop . we may thus think of as a two step process : using @xmath138 one solves first the initial value problem ( which is usually much easier ) , then one incorporates the boundary conditions via the projector @xmath140 . the submonoid relation @xmath125_{\phi}}\subset { { \mathcal{e}}[{\partial } ] \ltimes k\phi}$ ] means that multiplying regular problems leads to regular problems again . it is interesting to note that the converse is also true , provided the order condition of lemma [ lem : regularity - test ] holds . [ lem : reg - factors ] let @xmath141 be boundary problems over an integro - differential algebra @xmath19 with @xmath142 and @xmath143 . then @xmath144 and @xmath145 are regular whenever @xmath146 is . let us write @xmath147 and @xmath148 . choose fundamental systems @xmath149 for @xmath150 and @xmath151 for @xmath152 . take @xmath18-bases @xmath153 of @xmath154 and @xmath155 of @xmath156 . then @xmath157 is a fundamental system for @xmath158 , and a @xmath18-basis of @xmath159 is given by @xmath160 , @xmath155 . the latter fact follows since the sum @xmath161 in is always direct ( * ? ? ? * prop . 3.2 ) . by lemma [ lem : regularity - test ] the regularity of the boundary problem @xmath162 means that its evaluation matrix @xmath163 is regular . but this is only possible if both diagonal blocks @xmath164 and @xmath165 are regular , and these are just the evaluation matrices of @xmath144 and @xmath166 . for practical applications , one is not so much interested in multiplying boundary problems thus increasing their order as to factor them into lower - order problems . interestingly , factorization will also be instrumental in the localization process ( lemma [ lem : regularization ] ) . we repeat here as the main result the so - called _ factorization theorem _ [ thm : fact - bp ] given a regular boundary problem @xmath167_{\phi}}$ ] , every factorization @xmath168 of the differential operator lifts to a factorization @xmath169 with @xmath141 regular and @xmath170 . we note that the right factor may be chosen to be an arbitrary subspace of @xmath100 as long as regularity holds for @xmath166 ; in particular one can always choose initial conditions for @xmath171 to ensure regularity . in contrast , @xmath172 is determined by the choice of @xmath150 and @xmath152 alone . this is the content of the following uniqueness result ( * ? ? ? 31 ) , called the _ division lemma _ for boundary problems ( see below for the notion of subproblem ) . [ lem : division ] given a regular boundary problem @xmath167_{\phi}}$ ] and any factorization @xmath168 of the differential operator , there is a unique boundary problem @xmath144 such that for any regular subproblem @xmath173 we have the lifted factorization @xmath174 . the _ notion of subproblem _ was defined ( under the name `` right factor '' , which is avoided here due to the ambiguity in cases like example [ ex : embed - single - cond ] ) for regular boundary problems in ( * ? ? ? 29 ) , but it can be extended naturally to all boundary problems . hence a subproblem of a boundary problem @xmath167 \ltimes k\phi}$ ] is any @xmath175 \ltimes k\phi}$ ] with @xmath152 a right divisor of @xmath102 and @xmath171 a subspace of @xmath100 . we denote this relation by @xmath176 . we conclude this section with a few remarks on _ algorithmic issues_. while at the present stage of our work we do not aim at a computational realization of the localization , we see this nevertheless as a mid - term goal . in fact , most of the results in this paper are algorithmic , with two exceptions : ( 1 ) the ore condition is needed in @xmath177 $ ] for monic operators ; this is not supported by customary packages but appears to be adaptable ( see the remarks after proposition [ prop : left - ext ] ) . ( 2 ) one would need an algorithm for determining the monomials required in definition [ def : umbral - characters ] . once these two gaps are filled , one should be able to compute in the localization since all the constructions reviewed in this section are themselves algorithmic @xcite : * the _ operator ring _ @xmath63}$ ] is defined as a quotient by a certain ideal of relations . this ideal is described by a noncommutative grbner basis for an ideal with infinitely many generators ( on another view : a noetherian and confluent rewrite system ) . * if the _ ground algebra _ @xmath19 and the _ character set _ @xmath49 are computable , the normal forms of @xmath63}$ ] are computable as well . typically , @xmath19 is an algorithmic fragment of @xmath134 like the exponential polynomials @xcite , while @xmath49 consists of finitely many point evaluations @xmath178 . * the _ green s operator _ of a regular boundary problem ( and hence the solutions for arbitrary forcing functions up to quadratures ) can be computed as long as one has a fundamental system for the underlying homogenous differential equation . for the latter , one can in principle rely on the vast body of results from differential galois theory @xcite . * the _ multiplication and factorization _ of boundary problems can be computed as long as the same is true for the constituent differential operators . in this way one can often solve higher - order boundary problems by decomposing them into smaller factors such as in a recent application to actuarial mathematics @xcite . for the algorithmic theory of factoring linear differential operators , we refer to @xcite . as detailed in the introduction , the algorithms addressed above are _ implemented _ in mathematica and maple packages . we have seen that for a given integro - differential algebra @xmath19 and character set @xmath49 , the natural choice of boundary conditions is the _ stieltjes conditions _ , defined in @xcite as the right ideal @xmath67 of @xmath63}$ ] and characterized by their normal forms . the choice of @xmath67 is vindicated by theorem [ thm : fact - bp ] , which asserts that stieltjes conditions are sufficient for describing arbitrary factor problems ( and they are clearly also sufficient for multiplying boundary problems ) . in fact , stieltjes conditions with nonzero global part appear in the left factor problem even if the original boundary problem has only local conditions ; see e.g. example 33 in @xcite . for the purpose of localization we must isolate a subclass of stieltjes conditions . first of all , we must distinguish carefully between the zero condition @xmath179 and degenerate boundary conditions that only _ act _ as zero : we call a stieltjes condition @xmath99 _ degenerate _ if @xmath180 for all @xmath181 . in the standard setting of example [ ex : standard ] , there are plenty of degenerate boundary conditions . if @xmath4 is a bump function supported , say , on an interval disjoint from @xmath182 $ ] , the stieltjes condition @xmath183 is clearly degenerate . in contrast , the integro - differential subalgebra of analytic functions @xmath43 does not have degenerate global conditions of this form as we shall see below ( example [ ex : complete ] ) . the subclass of stieltjes conditions to be chosen must be so as to ensure a sufficient supply of regular boundary problems . in particular , we shall find ourselves in the situation of embedding a singular boundary problem @xmath92 into a surrounding regular problem ( lemma [ lem : regularization ] ) . hence we must enlarge @xmath184}$ ] by a suitable differential operator @xmath185}$ ] and the boundary space @xmath100 by _ new boundary conditions _ @xmath186 in such a manner that the resulting evaluation matrix is regular ( lemma [ lem : regularity - test ] ) . the key for solving this problem is found in the fortunate fact every integro - differential algebra @xmath19 contains the polynomial ring ( proposition [ prop : pol - ring ] ) , so the _ initial value problem _ @xmath187)\ ] ] is a natural choice for extending the given boundary problem @xmath129 . then the monomials @xmath188 are a fundamental system by proposition [ prop : pol - ring ] , and the corresponding evaluation matrix is @xmath189 . but this is not enough for embedding singular problems since the evaluation matrix of the surrounding problem involves combining the _ given _ boundary conditions @xmath120 with the monomials of the fundamental system . indeed , the crucial step in the proof of lemma [ lem : regularization ] needs that @xmath190 for _ some _ monomial @xmath191 $ ] . this is the motivation for the following definition . [ def : umbral - characters ] a stieltjes condition @xmath192}$ ] is called _ umbral _ if @xmath190 for some monomial @xmath193 $ ] . furthermore , we call @xmath49 an _ umbral character set _ if every nondegenerate stieltjes condition is umbral . the reason for our terminology is that there is an interesting link between stieljes conditions @xmath192}$ ] with @xmath190 and the _ umbral calculus_. indeed , every umbral condition defines a nontrivial shift - invariant operator ( proposition [ prop : umbral - cond ] ) . this is clear for the local parts @xmath194 but needs some justification for global terms @xmath195 with arbitrary @xmath196 $ ] . the point is that one may apply the following special case of the well - known `` antiderivative leibniz rule '' . ( in the general setting of analytic functions one writes @xmath197 as an infinite series of iterated integrals of @xmath4 times iterated derivatives of @xmath198 . here we have @xmath199 so that only finitely many derivatives are nonzero and the series terminates . ) [ lem : int - part - pol ] in any integro - differential algebra @xmath17 , we have the formula @xmath200 for all @xmath98 . here @xmath201 is defined by @xmath202 and @xmath203 . this is essentially the special case @xmath204 of equation ( 14 ) in @xcite , but the proof is short enough to present here . we use induction over @xmath136 . the base case @xmath205 is clear , so assume for @xmath206 ; we prove it for @xmath207 . since @xmath208 we have @xmath209 by the pure baxter axiom . using the induction hypothesis , the right summand becomes @xmath210 after an index transformation . adding the extra term @xmath211 means extending the summation to @xmath212 , which yields for @xmath207 . [ lem : umbral - global - cond ] let @xmath213 be a global condition in @xmath63}$ ] . then we have @xmath214 as a functional @xmath9 \rightarrow k$ ] , where @xmath215 \rightarrow k[x]\ ] ] is a shift - invariant operator with expansion coefficients @xmath216 . by a well - known result of the umbral calculus ( 2.1.7 ) , incidentally also presented in ( * ? ? ? 92 ) , the operator @xmath217 is shift - invariant . for seeing that @xmath218 , we apply lemma [ lem : int - part - pol ] to obtain @xmath219 . the _ operator expansion _ allows us to associate a shift invariant operator with a stieltjes condition of the special form @xmath213 . but this association generalizes immediately to arbitrary stieltjes conditions since the local conditions are unproblematic . [ prop : umbral - cond ] let @xmath99 be a stieltjes condition in @xmath63}$ ] . then there is an _ associated shift - invariant operator _ @xmath220 \rightarrow k[x]\ ] ] with coefficients @xmath221 such that @xmath222 . clearly , the associated operator @xmath217 is nonzero iff @xmath99 is an umbral condition . let @xmath223 \rightarrow k[x]$ ] be the shift operator @xmath224 with @xmath225 . using the normal form we can write the given boundary condition as @xmath226 where @xmath227 is clearly shift - invariant while @xmath228 is the shift - invariant operator corresponding to @xmath229 according to lemma [ lem : umbral - global - cond ] . then the terms @xmath230 in the second sum are shift - invariant , hence we have @xmath231 with the shift - invariant operator @xmath232 and the formula @xmath233 for its expansion coefficients follows from the general result referred to in the proof of lemma [ lem : umbral - global - cond ] . umbral boundary conditions appear to be _ abundant _ , at least in the cases most crucial to us , especially in the @xmath234 setting ( containing many important integro - differential subalgebras notably the analytic and holonomic functions as well as the exponential polynomials ) . [ prop : cinf - weierstrass ] in the standard setting of example [ ex : standard ] , the point evaluations form an umbral character set . consider an arbitrary nondegenerate stieltjes condition @xmath235 so that there is a function @xmath96 with @xmath236 . let @xmath237 be the maximum of @xmath238 for all @xmath239 with @xmath240 or @xmath241 . we consider now the banach space @xmath242 on the compact interval @xmath243 $ ] . its norm @xmath244 is given by @xmath245 for all @xmath246 . there is a little known generalization of the weierstrass approximation theorem due to nachbin @xcite , which asserts ( in a simple special case ) that @xmath44 $ ] is dense in @xmath242 . hence we may choose a polynomial sequence @xmath247 that converges to @xmath248 in the @xmath249 topology . one checks immediately that @xmath250 is a continous functional with respect to the @xmath249 norm . in detail , one has @xmath251 with operator - norm bound @xmath252 therefore we have @xmath253 . this is impossible if @xmath99 vanishes on all of @xmath44 $ ] . hence there is some @xmath254 $ ] with @xmath255 . clearly there exists a smallest monomial @xmath256 in @xmath257 with @xmath190 . not every character set is umbral , though . it is natural to introduce the following _ necessary conditions _ for a character set @xmath49 to be umbral ( we write @xmath258 for the canonical value @xmath259 of a character @xmath50 ) : 1 . the character set @xmath49 must clearly be _ separative _ in the sense that @xmath260 implies @xmath261 . otherwise @xmath262 is a nonzero stieltjes condition with @xmath263 ) = 0 $ ] . 2 . every character @xmath50 must be _ complete _ in the sense that every global condition of the form @xmath213 is umbral whenever it is nondegenerate . this may also be expressed as @xmath264 \rightarrow f = 0 $ ] . here orthogonality refers to the nondegenerate bilinear form @xmath265 . if this bilinear form is positive definite with @xmath266 or @xmath267 , we have a pre - hilbert space @xmath268 . following the terminology of ( * v.24 ) , completeness of @xmath239 then means that any @xmath239-orthonormal basis of @xmath9 $ ] is complete in @xmath268 . these conditions are most likely not sufficient . at the moment we can not give a counterexample for corroborating this claim . but it is intuitively clear that the completeness properties associated with each @xmath239 separately can not prevent linear dependencies between the actions of distinct global conditions @xmath195 . it remains an interesting task to formulate stronger conditions on @xmath49 that ensure an umbral character set in a natural way . [ ex : separative ] as an example of a _ nonseparative character set _ , consider the exponential polynomials @xmath269}$ ] with the nonstandard character @xmath239 defined by @xmath270 and @xmath271 for all @xmath136 , effectively mixing evaluation at @xmath272 for monomials with evaluation at @xmath273 for the exponential . choosing @xmath266 for simplicity , let @xmath27 be the honest evaluation @xmath274 . then clearly @xmath239 and @xmath27 coincide on @xmath9 $ ] while they are in fact distinct characters on @xmath269}$ ] since @xmath275 . this shows that separativity is necessary but not sufficient . nevertheless , it ensures that all _ local boundary conditions _ are indeed umbral . [ prop : local - bc - umbral ] let @xmath49 be a separative character set for an integro - differential algebra @xmath17 . then every local boundary condition is umbral . using , every local boundary condition can be written in the form @xmath276 assuming @xmath263 ) = 0 $ ] , we show that all coefficients @xmath277 are zero . collecting them in the vector @xmath278 we obtain @xmath279 linear homogeneous equations for the unkowns @xmath277 by applying @xmath99 to the monomials @xmath280 . the matrix of the corresponding system @xmath281 is given by @xmath282 with blocks @xmath283^{n \times s } \ ; .\ ] ] since @xmath49 is separative , the values @xmath284 are mutually distinct . then we may apply the subsequent lemma [ lem : det ] to obtain @xmath285 , hence @xmath286 has the unique solution @xmath287 . [ lem : det ] the determinant of @xmath288^{n \times n}$ ] , with @xmath279 and blocks , is given by @xmath289 where @xmath290 is the @xmath291 vandermonde determinant and @xmath292 denotes the superfactorial . the determinant is a special case of ( * ? ? ? 1.1 ) or ( * ? ? ? 20 ) , apart from the linear factor @xmath293 in @xmath294-th row of @xmath295 . thus we have @xmath296 which is . let us now turn to _ completeness_. in the analysis setting , every @xmath239 is complete since we know already the stronger result that every stieltjes condition is umbral ( proposition [ prop : cinf - weierstrass ] ) . it is nevertheless instructive to have a closer look at this case . [ ex : complete ] consider the point evaluation @xmath297 in the _ standard setting _ of example [ ex : standard ] . here we get an inner product @xmath298 which can be extended to all continuous functions @xmath299 on @xmath300 $ ] and so gives rise to the pre - hilbert hausdorff space @xmath301 , { \langle \cdot | \cdot \rangle}_\phi)$ ] . it is well - known that @xmath302 is a complete sequence in this space ( * ? ? ? we may assume @xmath303 by a scale transformation . by the usual gram - schmidt process , the monomials @xmath256 can be transformed to an orthonormal basis @xmath304 of @xmath305 $ ] , which consists in this case of the legendre polynomials @xmath306 in every pre - hilbert hausdorff space @xmath307 , an orthonormal basis @xmath308 has the following well - known property ( * ? ? ? v.2.5 ) : if @xmath309 is nonzero , we have @xmath310 for some @xmath311 . applying this to @xmath312 \supset c^\infty[0,1]$ ] and fixing any @xmath313 \subset c^\infty({\mathbb{r}})$ ] , we see that @xmath239 is complete in the sense defined above . for if @xmath314 for all @xmath311 , then @xmath4 vanishes on @xmath315 $ ] and hence @xmath195 is degenerate . the above argument shows in fact that every global condition @xmath316 over @xmath134 is umbral or degenerate . working over @xmath43 , we can exclude the degenerate case . indeed , the identity theorem of complex analysis ensures that @xmath317 whenever @xmath4 vanishes on @xmath315 $ ] . this is in stark contrast to the @xmath234 case , which has plenty of degenerate global conditions @xmath195 as we have observed at the beginning of section [ sec : umbral - bc ] . it becomes clear from the above example that completeness is really an _ analytic property _ of some sort . it is therefore not surprising that one can construct `` purely algebraic '' examples lacking this property . [ ex : incomplete ] to give an example of an _ incomplete character _ on @xmath318}$ ] with @xmath266 , we proceed similar to example [ ex : separative ] . define @xmath319 for @xmath320 and @xmath321 , @xmath322 . then clearly @xmath323 is nondegenerate since @xmath324 . but we have @xmath325 for all @xmath311 . as noted above , separativity and completeness are most likely not strong enough to ensure an umbral character set . note that umbrality of @xmath49 implies that two stieltjes conditions in @xmath63}$ ] are _ linearly independent _ on @xmath9 $ ] whenever they are linearly independent on @xmath19 . if one of the two conditions is local and the other global , it is reasonable to expect this property to follow from completeness . this expections is fulfilled . [ lem : global - not - local ] let @xmath17 be an integro - differential algebra with a complete character @xmath239 . then a nondegenerate global condition @xmath195 never coincides on @xmath9 $ ] with any local condition based on @xmath239 . assume @xmath213 coincides on @xmath9 $ ] with @xmath326 , where @xmath3 is chosen minimal . then the umbral expansion breaks off at @xmath327 , and we have @xmath328 for @xmath329 . for @xmath330 , we get @xmath331 for all @xmath332 , so the condition @xmath333 is degenerate since @xmath239 is complete . for showing that this can not happen , it suffices to prove that all @xmath334 are nondegenerate whenever @xmath195 is . we use induction on @xmath332 . the base case @xmath335 is trivial , so assume the claim for fixed @xmath294 . by the induction hypothesis we can choose @xmath336 with @xmath337 . using with @xmath198 in place of @xmath4 and @xmath338 in place of @xmath198 , we have @xmath339 so our choice of @xmath198 entails @xmath340 . if the first summand on the right - hand side is different from @xmath83 , then @xmath341 witnesses to @xmath342 being nondegenerate , and the induction is complete . otherwise @xmath343 must be nonzero , and we may use the special case @xmath344 of the pure baxter axiom to derive @xmath345 and again the induction is complete . let us start by reviewing the general setting for localization in a noncommutative unital ring @xmath237 , following @xcite . as in the commutative case , the denominator set @xmath346 ( the elements that should become invertible ) must clearly form a _ multiplicative set _ @xmath346 , meaning a submonoid of @xmath347 . clearly we must stipulate @xmath348 , otherwise the localization is the zero ring . if @xmath237 is a domain , one might want to take @xmath349 . while this is always possible in the commutative setting , one needs an additional condition if @xmath237 is not commutative . indeed , let us strive for a localization @xmath350 on the left , meaning all elements have the form @xmath351 with @xmath352 and @xmath353 . since @xmath354 this must also be possible for the reverse product so that @xmath355 for some @xmath356 and @xmath357 . multiplying out , we get the necessary condition @xmath358 known as the _ left ore condition _ ; the set @xmath346 is then called _ left permutable_. if @xmath237 is a domain , this condition is actually sufficient for guaranteeing the existence of a unique localization @xmath359 , which can be constructed essentially as in the commutative case . however , if @xmath237 has _ zero divisors _ , in general one does not get an embedding @xmath360 . in this case , the _ extension _ @xmath361 is a ring homomorphism that is not injective ( and of course not surjective ) . its kernel contains at least those @xmath362 that yield @xmath363 for some @xmath352 since this implies @xmath364 and hence @xmath365 , due to @xmath366 being invertible in @xmath350 . in the classical localization @xmath350 , the kernel should be optimal in the sense that it contains no other elements than these necessary ones . [ def : localization ] let @xmath237 be an arbitrary ring with @xmath367 . then @xmath361 is called a left _ ring of fractions _ if 1 . all elements @xmath366 with @xmath352 are invertible in @xmath350 , 2 . every element of @xmath350 has the form @xmath368 for some @xmath352 , @xmath369 , 3 . and the kernel of @xmath370 is given by @xmath371 . the ring homomorphism @xmath370 is called the _ extension_. the missing condition is now easy to establish . assume @xmath352 is a right zero divisor so that @xmath372 for some @xmath362 . then also @xmath373 and hence @xmath365 since @xmath366 is invertible in @xmath350 . but this implies @xmath374 for some @xmath356 by item ( c ) of definition [ def : localization ] . accordingly , one calls a set @xmath346 with the property @xmath375 _ left reversible_. together with the left ore condition , this turns out to be sufficient for the existence of a left ring of fractions . [ thm : ore - localization ] let @xmath237 be an arbitrary ring . then for any @xmath376 , the left ring of fractions @xmath350 exists iff @xmath346 is multiplicative , left permutable and left reversible . see ( * ? ? 10.6 ) . the setting would become much nicer when the extension @xmath377 is injective so that we can regard @xmath378 as an _ embedding_. unfortunately , the localization of @xmath63}$ ] that we will work out in the sequel is not of this type . in fact , one can easily show that the injectivity of @xmath370 is equivalent to having only regular elements in @xmath346 . following @xcite , we call an element @xmath352 regular if it is both left and right regular , where left regular means @xmath372 implies @xmath379 for all @xmath362 while right regular means @xmath380 implies @xmath379 for all @xmath362 . as @xmath63}$ ] contains plenty of zero divisors , it is hard to achieve a regular denominator set @xmath346 . of course there are analogous definitions for the _ right ring of fractions _ @xmath381 , but in general the existence of @xmath350 does not imply the existence of @xmath381 or vice versa , and even when both exist they need not be isomorphic . in fact , we shall be dealing with a case that is left permutable ( lemma [ lem : prob - left - permutable ] ) but not right permutable ( proposition [ prop : not - right - permutable ] ) . the localization that we shall construct is based on the monoid @xmath125_{\phi}}$ ] of regular boundary problems over an integro - differential algebra @xmath17 with umbral character set @xmath49 . the ring @xmath237 to be localized is the _ ring of boundary problems _ @xmath128_{\phi}}$ ] , which is a monoid algebra . for such cases , the above setting can be somewhat simplified . a monoid @xmath346 is called _ left permutable _ if it satisfies the ore condition with respect to itself , meaning @xmath382 for all @xmath383 . in the absence of addition , the analog of condition says that for all @xmath384 with @xmath385 there must exist @xmath356 with @xmath386 ; in this case we call @xmath346 _ left reversible _ ( as a monoid ) . we call @xmath346 a _ left ore monoid _ if @xmath346 is both left permutable and left reversible ; cf . also ( * ? ? ? the following lemma is a special case of ( * ? ? ? * lem . 6.6 ) but we include its proof since it dispenses with the technical machinery used there . it tells us that we can _ transfer left permutability _ from @xmath346 to the monoid algebra @xmath387 . [ lem : per - from - mon ] if @xmath346 is a left permutable monoid , then @xmath346 is a left permutable subset of @xmath387 . similarly , if @xmath346 is a left reversible monoid , then @xmath346 is a left reversible subset of @xmath387 . given @xmath388 , @xmath389 and @xmath352 , we have to find @xmath390 , @xmath391 and @xmath356 such that @xmath392 using the ore condition in @xmath346 , we can successively find @xmath393 and @xmath394 such that @xmath395 is fulfilled . multiplying these equations by @xmath396 , @xmath397 , @xmath398 , @xmath399 on the left yields the system @xmath400 if we set @xmath401 and @xmath402 with coefficients @xmath403 . summing these equations gives the desired ore condition . the proof of the second statement follows immediately by induction on the number of terms . [ cor : ore - from - mon ] if @xmath346 is a left ore monoid , the left ring of fractions @xmath404 exists . immediate from lemma [ lem : per - from - mon ] and theorem [ thm : ore - localization ] . we shall now prove that @xmath125_{\phi}}$ ] is a left ore monoid . the multiplication of boundary problems is realized by the semi - direct product of the multiplicative monoid @xmath177 $ ] acting on the additive monoid of boundary spaces . projecting onto the first factor , it is then clear that the monic differential operators of @xmath177 $ ] must form a left ore monoid . a coefficient algebra with this property shall be called _ left extensible_. [ ex : gcd - diffop ] left permutability requires _ some _ common left multiple for given differential operators @xmath405 $ ] . hence it may seem tempting to restrict ourselves to those coefficient algebras @xmath406 that even have a _ least common left multiple _ @xmath407 . but this setting is not suitable for our purposes : as far as we know , there are only two natural examples : the rational functions @xmath408 allow an adaption of the euclidean algorithm @xcite , and of course one may always take @xmath409 . the first case yields a differential field , which excludes the existence of an integro - differential structure @xcite . the second case restricts us to differential operators with constant coefficients , which is excessively restrive . fortunately it turns out that we do not need least common left multiples since left extensible coefficient algebras are easy to come by , as the next lemma shows . in particular , any integro - differential algebra @xmath19 allows @xmath410 $ ] so that @xmath177 = a_1(k)$ ] is the _ weyl algebra_. as another example , consider the standard setting of example [ ex : standard ] , where for @xmath91 one may take the larger ring of _ analytic functions _ @xmath43 . [ prop : left - ext ] any left noetherian differential domain @xmath406 is left extensible . if @xmath91 is a left noetherian domain , it satisfies left permutability by ( * ? ? ? 5.4 ) and is therefore a left ore domain . but @xmath177 $ ] is a special case of a skew - polynomial algebra and therefore inherits the property of being a left ore domain by ( * ? ? ? but this is not enough since we need the _ monic _ differential operators of @xmath177 $ ] to form a left ore monoid . hence assume @xmath411 for some monic differential operators @xmath412 $ ] . since @xmath177 $ ] is a domain , @xmath413 implies @xmath414 because @xmath415 is not possible . hence the monic differential operators from @xmath177 $ ] form a left reversible monoid . but this monoid is indeed a left ore monoid : according to ( * ? ? ? 1.5.1 ) , the set of all monic polynomials in the skew polynomial ring @xmath177 $ ] is left permutable whenever @xmath91 is a left noetherian domain . let us remark that the generalization from _ least _ common left multiples to common left multiples is crucial even for the basic case @xmath410 $ ] . for example when @xmath416 and @xmath417 , the least common left multiple of @xmath150 and @xmath152 in @xmath418 $ ] is @xmath419 and so gives @xmath420 when taken in @xmath177 $ ] . this shows that the least common left multiple of two monic operators in @xmath177 $ ] need not be monic . the conventional computation tools for the weyl algebra are therefore not directly applicable for computing monic common left multiples in @xmath177 $ ] , but some recent methods @xcite can be adapted to this purpose [ a. bostan , private communication ] . as explained at the beginning of section [ sec : umbral - bc ] , the main tool for ensuring the left ore condition in @xmath125_{\phi}}$ ] is the embedding of singular boundary problems into regular ones . this will be achieved in the regularization lemma [ lem : regularization ] , by successively embedding the stieltjes conditions generating a given boundary space . hence the crucial step is to embed a _ single stieltjes condition _ , which we require to be umbral so that we can build up a polynomial fundamental system . [ lem : one - cond - high ] let @xmath192}$ ] be an umbral stieltjes condition over a given integro - differential algebra @xmath19 . then there is a @xmath332 such that @xmath421)$ ] is a regular boundary problem since @xmath99 is umbral , there is a minimal mononomial @xmath422 with @xmath423 . clearly , @xmath424 is a fundamental system for @xmath425 . using the boundary conditions @xmath426 , we obtain the @xmath427 evaluation matrix @xmath428 where the off - diagonal bottom entries vanish by the assumption on @xmath422 . note that @xmath294 in lemma [ lem : one - cond - high ] can also be _ zero _ , for example if @xmath429 . in that case , it is not necessary to add any initial conditions since the first - order problem @xmath430)$ ] is already regular . [ ex : embed - single - cond ] as a more or less typical case consider @xmath431 . here the minimal mononomial is @xmath81 , which leads back to our standard example [ ex : second - order ] , the regular boundary problem @xmath432 ) = ( { \partial}^2 , [ { \text{\scshape\texttt e}}_0 , { \text{\scshape\texttt e}}_1])$ ] . as mentioned earlier , we can not split off the first - order subproblem @xmath433)$ ] since it is singular ( this is an example of a _ subproblem that is not a right factor _ ) : if it were not , we would know from the division lemma [ lem : division ] that the corresponding unique left factor is @xmath434)$ ] , as shown in ( * ? ? ? 28 ) and also below ( example [ ex : ore - quad ] ) . but of course multiplying out just gives the degenerate problem @xmath435 ) \cdot ( { \partial } , [ { \text{\scshape\texttt e}}_1-{\text{\scshape\texttt e}}_0 ] = ( { \partial}^2 , [ { \text{\scshape\texttt e}}_1-{\text{\scshape\texttt e}}_0])\ ] ] and not the desired regularized problem @xmath95)$ ] . based on the case of a single boundary condition , we can now embed an arbitrary boundary problem into a regular one provided we work with an umbral character set . this is the subject of the following _ regularization lemma_. [ lem : regularization ] let @xmath49 be an umbral character set for an integro - differential algebra @xmath19 . then for an arbitrary boundary problem @xmath436 \ltimes k\phi}$ ] there is a regular boundary problem @xmath437_{\phi}}$ ] that has @xmath92 as a subproblem . let @xmath167 \ltimes k\phi}$ ] be an arbitrary but fixed boundary problem with @xmath102 a differential operator of order @xmath320 and @xmath438 \le { \mathcal{f}}^*$ ] of dimension @xmath439 . we write @xmath440 for the space of initial conditions @xmath441 $ ] and @xmath442 for the partial boundary space @xmath443 $ ] with the convention that @xmath444 . we will now prove that for each @xmath445 there is a regular boundary problem @xmath446 that has @xmath447 as a subproblem . taking @xmath448 , the theorem follows . we use induction on @xmath294 . the base case follows by setting @xmath449 . for the induction step , assume @xmath450 is a regular boundary problem that has @xmath451 as a subproblem . we have to construct a regular problem @xmath446 that has @xmath452 as a subproblem . letting @xmath453 be the green s operator of @xmath450 , assume first that @xmath454 is degenerate . since @xmath454 vanishes on @xmath19 , we obtain @xmath455^\perp } \qquad\text{and hence}\qquad [ \beta_k ] \le \tilde{\mathcal{a}},\ ] ] so we may set @xmath456 in that case . now assume @xmath454 is nondegenerate and hence umbral . lemma [ lem : one - cond - high ] yields a regular problem @xmath457)$ ] . we define the boundary problem @xmath458 + \tilde{\mathcal{a}})\\ & = ( \tilde{t } \tilde{s } , [ { \text{\scshape\texttt e}}\tilde{s } , \dots , { \text{\scshape\texttt e}}{\partial}^{r-1 } \tilde{s } , \beta_k ] + \tilde{\mathcal{a } } ) , \end{aligned}\ ] ] where the last equality follows since the conditions @xmath459 and @xmath460 are equivalent for @xmath461 . the boundary problem @xmath446 is clearly regular since it is the product of two regular problems , and it has @xmath447 as a subproblem because @xmath462 by the induction hypothesis . we are now ready to prove _ left permutability _ for the monoid @xmath125_{\phi}}$ ] by merging the given factors into a singular problem that is subsequently embedded into a regular problem by virtue of lemma [ lem : regularization ] . [ lem : prob - left - permutable ] let @xmath49 be an umbral character set for an integro - differential algebra @xmath19 with left extensible coefficient algebra @xmath91 . then @xmath125_{\phi}}$ ] is a left permutable monoid . given @xmath463_{\phi}}$ ] , we must find @xmath464_{\phi}}$ ] such that @xmath465 . since @xmath91 is left extensible , proposition [ prop : left - ext ] yields a common left multiple @xmath102 and cofactors @xmath466 and @xmath467 such that @xmath468 . now set @xmath469 . by lemma [ lem : regularization ] there is a regular boundary problem @xmath446 that has @xmath92 as a subproblem . but then the boundary problem @xmath446 has @xmath144 and @xmath166 as regular subproblems , and lemma [ lem : division ] yields regular boundary problems @xmath470 and @xmath471 such that @xmath472 as claimed . lemma [ lem : prob - left - permutable ] is also true if one replaces @xmath125_{\phi}}$ ] by the monoid of well - posed boundary problems defined before eq . this can be seen readily by inspecting the above proof ( and the proofs of lemma [ lem : division ] and [ lem : regularization ] ) . let us give a simple example of a nontrivial _ ore quadruple _ @xmath144 , @xmath166 , @xmath473 , @xmath474 that will also serve a good purpose later on . [ ex : ore - quad ] in the standard setting of example [ ex : standard ] , consider the two simplest first - order problems on @xmath315 $ ] , namely @xmath475 ) \qquad\text{and}\qquad ( t_2 , { \mathcal{b}}_2 ) = ( { \partial } , [ { \text{\scshape\texttt e}}_1]).\ ] ] in that case , we have of course @xmath476 , and we apply lemma [ lem : regularization ] to the boundary problem @xmath477)$ ] . we end up with @xmath435 ) \cdot ( { \partial } , [ { \text{\scshape\texttt e}}_0 ] ) = ( { \partial}^2 , [ { \text{\scshape\texttt e}}_0 , { \text{\scshape\texttt e}}_1]),\ ] ] which is already regular . the corresponding cofactors are then @xmath478 ) \qquad\text{and also}\qquad ( \tilde{t}_2 , \tilde{{\mathcal{b}}}_2 ) = ( { \partial } , [ { { \textstyle \varint}}_0 ^ 1])\ ] ] since clearly @xmath479 ) = ( { \partial}^2 , [ { { \textstyle \varint}}_0 ^ 1 { \partial } , { \text{\scshape\texttt e}}_1 ] ) = ( { \partial}^2 , [ { \text{\scshape\texttt e}}_0 , { \text{\scshape\texttt e}}_1])$ ] . as mentioned before , @xmath125_{\phi}}$ ] is left permutable but _ not right permutable_. this means there are boundary problems that do not have a common right factor . actually , more is true : even if we start from two distinct problems with the same differential operator , _ any _ common right multiple comes from a singular factor . [ prop : not - right - permutable ] let @xmath49 be an arbitrary character set for an integro - differential algebra @xmath19 with coefficient algebra @xmath91 . assume @xmath480_{\phi}}$ ] have a common right multiple @xmath481 for some right factors @xmath482 \ltimes k\phi}$ ] . then both @xmath483 and @xmath484 are singular whenever @xmath485 . assume @xmath485 . projecting onto the boundary spaces , we have @xmath486 . if @xmath487 or @xmath488 , we have @xmath489 , so in fact @xmath490 . choosing @xmath491 we have @xmath492 but @xmath493 . hence we can write @xmath494 for some @xmath495 and nonzero @xmath496 . but then @xmath497 implies that @xmath484 is singular . by symmetry , we see that @xmath484 is singular as well . having established that @xmath125_{\phi}}$ ] is left permutable , the only thing missing for the localization is left reversibility , which is very easy in our case . hence we obtain the desired _ left ring of fractions_. [ thm : localization ] let @xmath49 be an umbral character set for the integro - differential algebra @xmath17 with left extensible coefficient algebra @xmath91 . then there exists a left ring of fractions @xmath498_{\phi}^\star}$ ] of the ring of boundary problems @xmath128_{\phi}}$ ] with denominator set @xmath125_{\phi}}$ ] . by corollary [ cor : ore - from - mon ] it suffices to check that @xmath125_{\phi}}$ ] is an ore monoid . since left permutability of @xmath125_{\phi}}$ ] has been shown in lemma [ lem : prob - left - permutable ] , it remains to show that @xmath125_{\phi}}$ ] is a left reversible monoid . so assume we have @xmath499 for some regular problems @xmath500 . by left extensibility of @xmath91 we have @xmath501 for some monic @xmath502 $ ] of order @xmath136 . let @xmath503 $ ] be the corresponding space of initial conditions so that @xmath504 is a regular problem . then we obtain the regular product problem @xmath505 where the two parenthesized factors have the same differential operator by the choice of @xmath506 . since @xmath92 is a regular right factor of this problem , the division lemma [ lem : division ] implies that @xmath507 . we conclude that the monic differential operators of @xmath177 $ ] are indeed left reversible . the fraction ring @xmath498_{\phi}^\star}$ ] shall be called the _ ring of methorious operators _ ( from the greek word meaning `` making up a boundary '' ) . note that it exists in particular in the smooth setting : if one chooses @xmath40 , every character is umbral ( proposition [ prop : cinf - weierstrass ] ) , and any noetherian domain can be used for the coefficient algebra ( proposition [ prop : left - ext ] ) , in particular @xmath410 $ ] or @xmath508 . in the sequel , we shall confine ourselves to the latter setting . as already observed after theorem [ thm : ore - localization ] , we must expect that @xmath509_{\phi}}\rightarrow { k{\mathcal{e}}[{\partial}]_{\phi}^\star}$ ] _ is not an embedding_. for example , we have @xmath510 ) - ( { \partial } , [ { \text{\scshape\texttt e}}_1 ] ) \in { \operatorname{ker}(\epsilon)}$ ] in the standard setting of @xmath40 . for checking this , we use the characterization of @xmath511 given in definition [ def : localization ] : it suffices to find a regular problem that annihilates @xmath512 from the left . indeed , we have @xmath434 ) n = 0 $ ] since we know from example [ ex : ore - quad ] that @xmath435 ) ( { \partial } , [ { \text{\scshape\texttt e}}_0 ] ) = ( { \partial}^2 , [ { \text{\scshape\texttt e}}_0 , { \text{\scshape\texttt e}}_1 ] ) = ( { \partial } , [ { { \textstyle \varint}}_0 ^ 1 ] ) ( { \partial } , [ { \text{\scshape\texttt e}}_1 ] ) .\ ] ] at the moment we do not know @xmath511 in explicit form . however , we have the following conjecture . [ conj : extension - kernel ] let @xmath49 be an umbral character set for an integro - differential algebra @xmath19 with left extensible coefficient algebra @xmath91 , and let @xmath509_{\phi}}\rightarrow { k{\mathcal{e}}[{\partial}]_{\phi}^\star}$ ] be the extension into the ring of methorious operators . then we have @xmath513 iff @xmath514 , where @xmath515 is the green s operator of @xmath516 . the previous example is a case in point . the _ intuitive reason _ for our conjecture is this : generically , a linear combination of green s operators has a finite - dimensional cokernel ( meaning its image is annihilated by just finitely many functionals in the case of a single operator these functionals are the given boundary conditions ) . such an operator is in some sense `` almost invertible '' . but if a linear combination degenerates into an element of @xmath64 , its image becomes one - dimensional , and we can not expect to invert such an `` operator '' . extending an operator ring is much more useful if its elements may still be viewed as _ operating _ on some presumably extended domain of `` functions '' . as explained in the introduction , the operational calculus of mikusiski avoids this problem by merging `` operators '' and `` operands '' in the mikusiski field . in contrast , we shall follow the algebraic analysis approach of keeping operators and operands in separate structures . to this end we construct a suitable _ module of fractions _ that extends a given integro - differential algebra @xmath17 . in analogy to the mikusiski approach we would have to start from a ring @xmath237 of integral operators acting on @xmath19 , construct its localization @xmath517 and let it act on the corresponding localization @xmath518 . however , this does not work for the following reason : the natural candidate for @xmath237 would be the @xmath18-algebra of green s operators for regular boundary problems ( since this is the only ring over which we have sufficient control ) . but this monoid / ring is dual to the regular boundary problems , as we know from . since @xmath125_{\phi}}$ ] is a left ore monoid but not a right ore monoid , @xmath237 is a right but not a left ore ring . hence we can only create a _ right _ ring of fractions @xmath517 and correspondigly only a _ right _ module of fractions that should extend @xmath19 . but the given integro - differential algebra @xmath19 only has a natural _ left _ action of @xmath519}$ ] . in fact , it is not quite true that one can not construct a left module of fractions from a given left module @xmath520 over a right ore ring @xmath237 . if @xmath521 is the right ring of fractions , one may of course construct the _ usual scalar extension _ . one may then refer to @xmath523 as a left module of fractions ( * ? ? ? but the problem with this construction is that we do not know anything about its structure . for example , there is no useful characterization of the kernel of @xmath524 as in the ore construction of theorem [ thm : fraction - module ] below . we must therefore take the _ differential operators _ as our starting point , and this is why we have constructed the localization @xmath498_{\phi}^\star}$ ] of the ring of boundary problems @xmath525_{\phi}}$ ] . one might be tempted to take just the simpler ring @xmath177 $ ] instead of the unwieldy monoid ring @xmath525_{\phi}}$ ] . but this would be too simplistic : in that case one gets a two - sided inverse @xmath526 of the differential operator , no new functions are generated , and of course we can not tackle boundary problems in such a setting . it is essential to work with the richer structure @xmath525_{\phi}}$ ] , for which we will have to set up a suitable left action on @xmath19 . but first let us briefly review the general setting for the _ localization of modules_. as in theorem [ thm : ore - localization ] , we go back to the setting of general rings @xmath237 . in this case we can construct the module of fractions in pretty much the same way as the ring of fractions ( * ? ? ? ii.3.3 ) . [ thm : fraction - module ] let @xmath520 be a left @xmath237-module , and let @xmath376 be a multiplicative , right permutable and right reversible denominator set @xmath376 . then there exists a left @xmath350-module @xmath527 , and the kernel of the extension @xmath528 consists of those @xmath529 for which there exists an @xmath352 with @xmath530 . now let @xmath17 be an integro - differential algebra and @xmath63}$ ] its ring of operators . for setting up a suitable left action of @xmath125_{\phi}}$ ] on @xmath19 , we recall the _ fundamental formula _ @xcite of the mikusiski calculus @xmath531 where @xmath3 denotes the inverse of the standard convolution operator @xmath532 , defined by @xmath533 , and @xmath534 is simply the multiplicative identity generated by the construction of the mikusiski field . of course one may write @xmath535 but we keep the explicit notation for intuitive reasons since we can interpret as the distributional derivative of a function @xmath536 that is continuous except for a jump at @xmath80 . in that case , @xmath534 is the dirac distribution concentrated at @xmath80 . at any rate , the intuitive content of is that @xmath3 is a kind of enhanced differential operator that memorizes the function value at @xmath80 that is otherwise lost . taking this clue , we define the action of @xmath125_{\phi}}$ ] on @xmath19 by @xmath537 where @xmath140 is the projector onto @xmath112 along @xmath105 . in the special case of @xmath476 , this recovers if we think of @xmath538)$ ] as some kind of algebraic representation of the dirac distribution @xmath534 . the definition has the consequence that we must extend @xmath19 so that it contains the new elements @xmath92 , prior to the extension effected by the localization . this is somewhat _ unsatisfying _ but appears to be inevitable in view of the facts pointed out at the opening of the section . perhaps in the future one can find a more powerful localization that generates at once the extension objects needed in the fundamental formula . at any rate , we must now also define the action of @xmath125_{\phi}}$ ] on the new elements , but this is clearly @xmath539 where the boundary problem on the right side is the usual product in the problem monoid @xmath125_{\phi}}$ ] . this will ensure that the action is well - defined . the _ intuitive meaning _ of the ideal element @xmath540 is that it records various integration constants that were lost under differentation . when the element is created by , it encodes all the boundary values @xmath164 for @xmath120 , and the function @xmath4 is confined to the kernel of @xmath102 . further action by a boundary problem @xmath541 according to leads to a new differential operator @xmath542 while leaving @xmath4 intact . hence we see that the function inside an ideal element is always in the kernel of the associated differential operator . we may therefore restrict the ideal elements @xmath540 to those with @xmath543 . after the action of , the new boundary space is @xmath544 , so the old boundary conditions @xmath107 are retained ( the ones for which we have boundary values ) . the new boundary conditions @xmath545 for @xmath546 give zero on @xmath4 since actually @xmath543 . in some sense , we have added redundant boundary data . we must therefore _ regulate redundancy _ in the ideal elements . this can be achieved by declaring that @xmath540 should be the same as @xmath547 for any @xmath548_{\phi}}$ ] with green s operator @xmath453 . this can be understood as follows . the boundary conditions in @xmath547 are @xmath546 and @xmath549 with @xmath120 . the former yield zero on @xmath550 since @xmath453 is the green s operator of @xmath541 , the latter give back @xmath551 just as for the ideal element @xmath552 . this leads to the following definition . let @xmath553 be the subspace of @xmath554_{\phi}}$ ] generated by the ideal elements @xmath555 with @xmath543 . furthermore , let @xmath556 be the subspace of @xmath553 generated by the elements @xmath557 then we define the _ module of methorious functions _ to be @xmath558 . let us now check that @xmath559 is indeed a module . [ prop : prob - action ] let @xmath17 be an integro - differential algebra with character set @xmath49 . the definitions and induce a monoid action of @xmath125_{\phi}}$ ] on @xmath559 such that it becomes a @xmath525_{\phi}}$]-module . any monoid @xmath307 acting on a @xmath18-algebra @xmath79 extends to an action of the monoid ring @xmath560 $ ] that makes @xmath79 into a @xmath560$]-module . hence it suffices to verify the statement about the monoid action . first of all we must check that does not depend on the representative @xmath561 , meaning it maps the subspace @xmath556 into itself . this follows immediately from the associativity of the multiplication in @xmath125_{\phi}}$ ] . clearly we have @xmath562 and @xmath563 , so the unit element is respected . associativity is immediate in the case of , so it remains to check that @xmath564 and @xmath565 are equal . the former is @xmath566 where @xmath140 is the projector onto @xmath112 along @xmath105 and @xmath567 is correspondingly the projector onto @xmath568 along @xmath569 . if @xmath86 and @xmath453 are the green s operators of @xmath92 and @xmath541 , respectively , one can easily check that the projector associated with the composite problem @xmath570 is @xmath571 , meaning it projects onto @xmath572 along @xmath573 . hence we obtain for the other side of the prospective equality @xmath574 where the first and the last term is identical to the corresponding terms on the left hand side while the middle terms are equal since their difference is in @xmath556 . since @xmath559 is a @xmath525_{\phi}}$]-module , we obtain the @xmath498_{\phi}^\star}$]-module @xmath575 of _ methorious hyperfunctions _ by localization via theorem [ thm : fraction - module ] . we can not expect the extension @xmath576 to be injective since its kernel @xmath577_{\phi}}\colon ( t , { \mathcal{b } } ) \cdot \phi = 0 \}\ ] ] contains elements like @xmath578 ) - ( { \partial } , [ { \text{\scshape\texttt e}}_0])$ ] , which is annihilated upon multiplying with @xmath434)$ ] from the left ( see the example before conjecture [ conj : extension - kernel ] ) . but fortunately no elements of @xmath19 are lost . [ prop : hypfun - extension ] let @xmath17 be an integro - differential algebra with character set @xmath49 . then we have an embedding @xmath579 . assume @xmath580 for some @xmath98 and @xmath581_{\phi}}$ ] . then @xmath582 implies that @xmath543 and @xmath583 . but the former means that @xmath584 and the latter that @xmath585 . since @xmath92 is a regular problem , we have a direct sum @xmath586 and so @xmath317 . in the module of methorious hyperfunctions , we can finally justify our earlier notation @xmath587 for the _ green s operator _ of a regular boundary problem @xmath167_{\phi}}$ ] . to avoid confusion with @xmath588_{\phi}^\star}$ ] we refrain from the reciprocal notation for green s operators in the scope of the following theorem . but the result of the theorem is of course that it anyway does not matter how we interpret @xmath589 since it amounts to the same . [ prop : greensop - localization ] we have @xmath590 for all @xmath591 . moreover , if @xmath543 then @xmath592 . we have @xmath593 since @xmath594 and @xmath595 . multiplying by @xmath596_{\phi}^\star}$ ] yields the first result claimed . now assume @xmath543 . we obtain @xmath597 since @xmath598 for @xmath584 . again we multiply by @xmath599 to gain the result . we conclude with an example that hints at possible applications of our noncommutative mikusiski calculus . the classical mikusiski calculus has only one _ fundamental formula _ since boundary values ( or rather : initial values ) are only processed at @xmath80 . in contrast , there are plenty of fundamental formulae in the noncommutative mikusiski calculus . [ ex : fundamental - formulae ] let us write @xmath600 and @xmath601 as abbreviations for the problems @xmath602 ) \in { k{\mathcal{e}}[{\partial}]_{\phi}^\star}$ ] and @xmath602 ) \in { { \mathcal{f}}_{\phi}^\star}$ ] , respectively . then we have @xmath603 by the definition of the action . so we have algebraic representations for all the dirac distributions . but there are other methorious functions that do not have any distributional counterpart . for example , let us consider @xmath604 ) \in { k{\mathcal{e}}[{\partial}]_{\phi}^\star}$ ] and @xmath605 ) \in { { \mathcal{f}}_{\phi}^\star}$ ] . this yields the fundamental formula @xmath606 so @xmath370 is a kind of `` smeared out '' distribution that keeps the mean value of a given function @xmath4 . [ ex : inhom - bp ] finally let us see how one solves _ inhomogeneous boundary problems _ in the noncommutative mikusiski calculus . we use the standard setting of @xmath40 and consider the problem @xmath607 for given boundary values @xmath608 . we know how to compute the green s operator @xmath86 of the homogeneous problem @xmath609)$ ] as mentioned at the end of section [ sec : monoid - bp ] . in this case the projector @xmath140 onto @xmath610 $ ] along @xmath611^\perp}$ ] is given by @xmath612 . hence we have @xmath613 ) \ , u & = u '' + \big ( u(0 ) \ , ( 1-x ) + u(1 ) \ , x \big ) \ , ( { \partial}^2 , [ { \text{\scshape\texttt e}}_0 , { \text{\scshape\texttt e}}_1])\\ & = f + \big ( a \ , ( 1-x ) + b \ , x \big ) ( { \partial}^2 , [ { \text{\scshape\texttt e}}_0 , { \text{\scshape\texttt e}}_1]),\\[0.5ex ] u & = gf + a \ , ( 1-x ) + b \ , x \end{aligned}\ ] ] by proposition [ prop : greensop - localization ] . as already indicated at various points above , our construction has several _ loose ends_. to begin with , it would be preferrable to localize in a subring ( or even all ) of @xmath63}$ ] rather than the monoid ring @xmath525_{\phi}}$ ] . this would have the advantage that one has a natural action on the underlying integro - differential algebra @xmath17 , and the somewhat artificial action on the module of methorious functions @xmath559 would be unnecessary . regarding the latter , we have already remarked in section [ sec : methorious - functions ] that the current two - stage process of creating the localization @xmath575 is unsatisfactory : in mikusiski s setup , all the `` ideal elements '' like @xmath3 and @xmath534 are an immediate result of the localization while we have to supply them offhand , prior to localization . we would like to find a better formulation in the future that will avoid this kind of inadequacy . one possible path towards such an improved localization is suggested by the prominent appearance of _ singular problems _ in the proof of lemma [ lem : prob - left - permutable ] . it may be worthwhile to expand the ring @xmath525_{\phi}}$ ] to include ( all or some ) singular boundary problems . of course this will also increase the kernel of the corresponding extension ( we can not expect to invert singular problems ) , but perhaps the resulting ring of fractions is more natural . in particular , it may be possible to interpret the sum of two ( singular ) boundary problems in some useful way , in contrast to the formal sums of @xmath525_{\phi}}$ ] . the results of @xcite will be useful for the work in this direction . staying with the current construction , there are some obvious open questions : fist of all , the _ kernel of the extension _ @xmath370 into the methorious operators should be determined , possibly by proving cojecture [ conj : extension - kernel ] . likewise , the kernel [ eq : kernel - hypfun ] of the extension @xmath614 into the methorious hyperfunctions is to be computed . finally , we would like to draw attention to the interesting link between integro - differential algebras and the umbral calculus ( section [ sec : umbral - bc ] ) , which deserves to be studied in more depth . there is also an intriguing relation @xcite between the umbral calculus and the rota - baxter algebras ( both of these being favourite topic of g .- rota ) , which might benefit from the new perspective afforded by boundary problems in integro - differential algebras ( since the latter are special cases of differential rota - baxter algebras and hence of plain rota - baxter algebras ) . as a more mundane goal , it would also be important to find a better characterization of umbral character sets that strengthens the separativity and completeness conditions given before example [ ex : separative ] . first of all we would like to acknowledge georg regensburger , who has worked with the first author on an earlier ( much more specialized and mathematically flawed ) version of localization . we thank him also for his insight about monic common left multiples and the corresponding reference ( last paragraph of proposition [ prop : left - ext ] ) . we express our gratitude also to a. bostan for his pertinent remarks on left common multiples in the weyl algebra ( see the remark after the proof of proposition [ prop : left - ext ] ) . furthermore , we are grateful for illuminating discussions with i. dimovski and m. spiridonova .

very soon , experiments at the large hadron collider ( lhc ) will begin direct exploration of physics at the tev scale . strong theoretical arguments suggest that this physics will include new particles and forces not present in the standard model ( sm ) . several theoretically motivated extensions of the standard model at the tev scale have been proposed . after new physics discovery at the lhc , the main task of the experiments will be to determine which of the proposed models , if any , is correct . unfortunately , there exists a broad and well - motivated class of sm extensions for which this task would be highly non - trivial . in these models , the new tev - scale particles carry a new conserved quantum number , not carried by the sm states . the lightest of the new particles is therefore stable . furthermore , the stable particle interacts weakly , providing a very attractive `` weakly interacting massive particle '' ( wimp ) candidate for dark matter with relic abundance naturally in the observed range . models of this class include the minimal supersymmetric standard model ( mssm ) and a variety of other supersymmetric models with conserved r parity , little higgs models with t parity ( lht ) , and models with universal extra dimensions ( ued ) with kaluza - klein ( kk ) parity . all these models have a common signature at a hadron collider : pair - production of new states is followed by their prompt decay into visible sm states and the lightest new particle , which escapes the detector without interactions leading to a `` missing transverse energy '' signature . if this universal signature is observed at the lhc , how does one determine which of these models is realized ? one crucial difference between the mssm and models such as lht or ued is the correlation between spins of the new particles and their gauge charges . in all these models , all ( or many of ) the new states at the tev scale can be paired up with the known sm particles , with particles in the same pair carrying identical gauge charges . however , while in the lht and ued models the two members of the pair have the same spin , in the mssm and other supersymmetric models their spins differ by 1/2 . thus , measuring the spin of the observed new particles provides a way to discriminate among models . experimental determination of the spin of a heavy unstable particle with one or more invisible daughter(s ) in hadron collider environment is a difficult task . one possibile approach , which recently received considerable attention in the literature @xcite , is to use angular correlations between the observable particles emitted in subsequent steps of a cascade decay , which are sensitive to intermediate particle spins . this strategy is promising , but its success depends on the availability of long cascade decay chains , which may or may not occur depending on the details of the new physics spectrum . it is worth thinking about other possible strategies for spin determination . in this paper , we explore the possibility of using 3-body decays of heavy new particles to determine their spin . the most interesting example is the 3-body decay of the mssm gluino into a quark - antiquark pair and a weak gaugino , q+|q+ . in a large part of the mssm parameter space , this decay has a large branching ratio : this occurs whenever all squarks are heavier than the gluino . , which first arises at one - loop level and generically has a partial width comparable to the tree - level decay . the gluino decay patterns in this parameter region have been analyzed in detail in ref . ] under the same condition , gluino pair - production dominates susy signal at the lhc . we will argue that the invariant mass distribution of the jets produced in reaction contains non - trivial information about the gluino spin , and can be used to distinguish this process from , for example , its ued counterpart , @xmath3 . it is important to note that the jet invariant mass distribution we study depends not just on the spin of the decaying particle , but also on the helicity structure of the couplings which appear in the decay , as well as on the masses of the decaying particle , the invisible daughter , and the off - shell particles mediating the decay . if all these parameters were measured independently , the jet invariant mass distribution would unambiguously determine the spin . however , independent determination of many of the relevant parameters will be very difficult or impossible at the lhc . in this situation , proving the spin-1/2 nature of the decaying particle requires demonstrating that the experimentally observed curve can not be fitted with any of the curves predicted by models with other spin assignments , independently of the values of the unknown parameters . this considerably complicates our task . still , interesting information can be extraced . for example , we will show that , even if complete ignorance of the decaying and intermediate particle masses is assumed , the jet invariant mass distribution allows one to distinguish between the decay in the mssm and its ued counterpart ( assuming the couplings specified by each model ) at the lhc . the paper is organized as follows . after setting up our notation and reviewing the basics of three - body kinematics in section 2 , we present a simple toy model showing how dijet invariant mass distributions from three - body decays can be used to probe the nature of the decaying particle and its couplings in section 3 . section 4 discusses using this observable for mssm / ued discrimination , and contains the main results of the paper . section 5 contains the conclusions . appendix a contains the polarization analysis of the decay @xmath3 in ued , which sheds some light on the main features of the dijet invariant mass distribution in this case . appendix b contains a brief review of the kullback - leibler distance , a statistical measure used in our analysis . we are interested in three - body decays of the type a q + |q + b , where @xmath4 and @xmath5 are tev - scale particles . the main focus of this paper will be on the case when @xmath4 is the gluino of the mssm or the kk gluon of the ued model , and @xmath5 is a neutalino or chargino of the mssm or a kk electroweak gauge boson of the ued ; however the discussion in this section applies more generally . we assume that @xmath6 and @xmath7 are massless , and denote their four - momenta by @xmath8 and @xmath9 respectively . to describe the kinematics in lorentz - invariant terms , we introduce the `` mandelstam variables '' , s & = & ( p_1+p_2)^2 = ( p_a - p_b)^2 , u & = & ( p_1+p_b)^2 = ( p_a - p_2)^2 , t & = & ( p_2+p_b)^2 = ( p_a - p_1)^2 , of which only two are independent since s+t+u = m_a^2 + m_b^2 . the allowed ranges for the mandelstam variables are determined by energy and momentum conservation ; in particular , 0s ( m_a - m_b)^2 . we will assume that @xmath10 can not be reconstructed , either because @xmath5 is unobservable or is unstable with all decays containing unobservable daughters . moreover , since the parton center - of - mass frame is unknown , no information is available about the motion of particle @xmath4 in the lab frame . due to these limitations , the analysis should use observables that can be reconstructed purely by measuring the jet four - momenta , and are independent of the velocity of @xmath4 in the lab frame . the only such observable is @xmath11 , and the object of interest to us is the distribution @xmath12 . this is given by = _ e_b - p_b^e_b+p_b ||m|^2 , where e_b & = & , p_b & = & , and @xmath13 is the invariant matrix element for the decay , with the bar denoting the usual summation over the final state spins and other quantum numbers and averaging over the polarization and other quantum numbers of @xmath4 . if it is produced in a polarized state . in the examples of this paper , production is dominated by strong interactions and @xmath4 will always be produced unpolarized . ] the quantity @xmath14 can be expressed in terms of the variables ; substitutions t m_a^2 - , u+m_b^2 should be made in @xmath14 before performing the integral in eq . . notice that eq . is valid in the rest frame of the particle @xmath4 ; however , since @xmath11 is lorentz - invariant , its lorentz transformation is a trivial overall rescaling by time dilation , and the shape of the distribution is unaffected . the strategy we will pursue is to use this shape to extract information about the decay matrix element @xmath13 , which is in turn determined by the spins and couplings of the particles @xmath4 and @xmath5 . to separate the effects of non - trivial structure of the decay matrix element > from those due merely to kinematics , it will be useful to compare the dijet invariant mass distributions predicted by various theories to the `` pure phase space '' distribution , obtained by setting the matrix element to a constant value . from eq . , the phase space distribution is given by = . this distribution . ] is shown by a solid black line in fig . [ fig : toy ] . notice that the phase space distribution has an endpoint at @xmath15 , with the asymptotic behavior given by ~(s-)^1/2 as the endpoint is approached . and @xmath16.[fig : toy ] , title="fig:",width=226 ] ( 5,0)@xmath11 ( -185,100)@xmath17 to illustrate how the chiral structure of the couplings involved in the decay can be determined from the dijet invariant mass distribution , consider a situation when the particles @xmath4 and @xmath5 are real scalars . introduce a massive dirac fermion @xmath18 of mass @xmath19 , and consider the following two models : model 1 defined by _ 1= y_a a | p_l q + y_b b | p_r q + h.c . and model 2 defined by _ 2= y_a a | p_l q + y_b b | p_l q + h.c . the matrix element for the decay in model 1 is given by _ spin|m_1|^2= 2 y_a^2 y_b^2 ( m_*^2 s ) ( + ) , while in model 2 it is given by _ spin|m_2|^2= 2 y_a^2 y_b^2 ( ( m_a^2+m_b^2 ) t u -m_a^2 m_b^2 ) ( + ) ^2 . the dijet invariant mass distributions in the two models are shown by the blue / dashed line ( model 1 ) and red / dot - dashed line ( model 2 ) in fig . [ fig : toy ] . their strikingly different shapes are due to the angular momentum conservation law and to the different helicity structure of the couplings . to understand this , consider this decay in the @xmath4 rest frame . in this frame , @xmath20 when @xmath21 , the quark and the antiquark travel in the same direction , as illustrated in fig . [ fig : am ] . since @xmath4 and @xmath5 have zero spin , the sum of the quark and antiquark helicities must vanish for this kinematics . in model 1 , the quark and the antiquark have the same helicity , and the decay is forbidden for @xmath21 ; in model 2 , it is allowed . in contrast , when @xmath15 , the particle @xmath5 is at rest , and the quark and the antiquark travel in the opposite directions . by angular momentum conservation , their helicities must be equal . in model 1 , this is the case , and the distribution approaches that of pure phase space in the limit @xmath22 . in model 2 , this kinematics is forbidden , the matrix element vanishes at the endpoint , and the distribution behaves as @xmath23 . rest frame for @xmath21 and @xmath15 in the two toy models of section 3 . [ fig : am],title="fig:",width=566 ] ( -440,60)@xmath24 ( -440,10)@xmath25 ( -360,100)model 1 : same helicities ( -160,100)model 2 : opposite helicities in this section , we will show that measuring the shape of the dijet invariant mass distribution arising from a three - body decay of a heavy colored particle may allow to determine whether the decaying particle is the gluino of the mssm or the kk gluon of the ued model . we will begin by comparing the analytic predictions for the shapes of the two distributions at leading order . we will then present a parton - level monte carlo study which demonstrates that the discriminating power of this analysis persists after the main experimental complications ( such as the combinatioric background , finite energy resolution of the detector , and cuts imposed to suppress sm backgrounds ) are taken into account . , title="fig:",width=453 ] ( -250,55)@xmath6 ( -205,55)@xmath7 ( -320,0)@xmath26 ( -200,0)@xmath27 ( -260,20)@xmath28 ( -80,55)@xmath6 ( -35,55)@xmath7 ( -155,0)@xmath26 ( -30,0)@xmath27 ( -90,2)@xmath28 we consider the mssm in the region of the parameter space where all squarks are heavier than the gluino , forbidding the two - body decays @xmath29 . in this situation , gluino decays through three - body channels . we study the channel ( p_a ) q(p_1)+|q(p_2)+_1 ^ 0(p_b ) , where @xmath6 and @xmath7 are light ( 1st and 2nd generation ) quarks , and @xmath30 is the lightest neutralino which we assume to be the lsp . ( note that many of our results would continue to hold if @xmath30 is replaced with a heavier neutralino or a chargino . the only extra complication in these cases would be a possible additional contribution to the combinatoric background from the subsequent cascade decay of these particles . ) the leading - order feynman diagrams for the process are shown in fig . [ fig : mssm_fd ] ; the vertices entering these diagrams are well known ( see for example ref . the spin - summed and averaged matrix element - squared has the form ( up to an overall normalization constant ) _ spin|m_mssm|^2= |c_l|^2 f(s , t , u;m_l * ) + |c_r|^2 f(s , t , u;m_r * ) , where f(s , t , u ; m ) = + + 2 . here @xmath0 , @xmath1 , @xmath31 and @xmath32 are the masses of the gluino , the neutralino , the squarks @xmath33 and @xmath34 , respectively . in order to keep the analysis general , we will not assume any relationships ( such as msugra contraints ) among these parameters , and will always work in terms of weak - scale masses . we also define c_l & = & t^3_q n_12 - t_w ( t^3_q - q_q ) n_11 , c_r & = & t_w q_q n_11 , where @xmath35 , and @xmath36 is the neutralino mixing matrix is real . it is always possible to redefine the neutralino fields to achieve this . however one should keep in mind that the neutralino eigenmasses may be negative with this choice . ] in the basis @xmath37 . we have neglected the mixing between the left - handed and right - handed squarks , which is expected to be small in the mssm . since up and down type quarks are experimentally indistinguishable , the dijet invariant mass distribution @xmath12 should include both the contributions of up - type and down - type squarks . , title="fig:",width=453 ] ( -250,55)@xmath6 ( -205,55)@xmath7 ( -320,0)@xmath38 ( -200,0)@xmath39 ( -260,2)@xmath40 ( -80,55)@xmath6 ( -35,55)@xmath7 ( -155,0)@xmath38 ( -30,0)@xmath39 ( -95,2)@xmath40 the counterpart of the decay in the universal extra dimensions ( ued ) model is the decay g^1(p_a ) q(p_1)+|q(p_2)+b^1(p_b ) , where @xmath41 and @xmath39 are the first - level kaluza - klein ( kk ) excitations of the gluon and the hypercharge gauge boson , respectively . we ignore the mixing between @xmath39 and the kk mode of the @xmath42 field , which is small provided that the radius of the extra dimension is small , @xmath43 , and assume that the @xmath39 is the ltp . as in the mssm case , the decay is expected to have a substantial branching fraction when all kk quarks @xmath44 and @xmath45 are heavier than the kk gluon . note that in the original ued model @xcite , the kk modes of all sm states were predicted to be closely degenerate in mass around @xmath46 ; it was however later understood @xcite that kinetic terms localized on the boundaries of the extra dimension can produce large mass splittings in the kk spectrum . since such kinetic terms are consistent with all symmetries of the theory , we will assume that they are indeed present , and treat the masses of the @xmath41 , @xmath39 , @xmath44 and @xmath45 fields as free parameters . the leading - order feynman diagrams for the decay are shown in fig . [ fig : ued_fd ] . ( we ignored the contribution of the diagrams mediated by @xmath47 with @xmath48 , which are suppressed by the larger masses of the higher kk modes . ) the relevant couplings have the form & & g_3 g^1_+ & & g_1 b^1 _ , where @xmath49 and @xmath50 are the hypercharges . the spin - summed and averaged matrix element - squared has the form ( up to an overall normalization constant ) _ spin|m_ued|^2= y_l^2 g(s , t , u;m_l * ) + y_r^2 g(s , t , u;m_r * ) , where @xmath31 and @xmath32 are the masses of the left- and right - handed quark kk modes @xmath45 and @xmath44 , and g(s , t , u;m)= + + 2 , with h_1(s , t , u ) & = & 4(t u - m_a^2 m_b^2 ) + ( 2 s ( m_a^2+m_b^2 ) + t u -m_a^2 m_b^2 ) , h_2(s , t , u ) & = & 4s(m_a^2+m_b^2 ) - ( 2 s ( m_a^2+m_b^2)+t u - m_a^2 m_b^2 ) . and @xmath16.[fig : dgdsall],title="fig:",width=226 ] ( 5,0)@xmath11 ( -185,100)@xmath17 armed with the expressions and , it is straightforward to obtain the dijet invariant mass distributions for gluino and kk gluon decays and compare them . for example , the two distributions for a particular choice of parameters , along with the pure phase space distribution , are shown in fig . [ fig : dgdsall ] . while not as strikingly different as the two toy models of section [ sec : toy ] , the curves predicted by the mssm and the ued are clearly distinct . ( the suppression of the ued distribution compared to phase space around @xmath21 and @xmath51 can be easily understood using angular momentum conservation , as explained in appendix [ app : pol ] . ) in this section , we will perform a simplified analysis of the discriminating power of these distributions , ignoring experimental complications such as cuts , finite energy resolution , combinatoric and sm backgrounds , and systematic errors . we will return to include some of these complications in the following section . the distrubution in each model depends on a number of parameters , including the mass of the mother particle @xmath0 , the mass of the invisible daughter @xmath1 , and the masses of intermediate particles : @xmath52 in the mssm case and @xmath53 in the ued case . we assume that the partners of the up - type quarks of the first two generations and the down - type quarks for all three generations are degenerate , and do not include the diagrams with intermediate stops ( or kk tops ) since they produce tops in the final state . furthermore , since the yukawa couplings for the first two generations are small , it is safe to assume that @xmath54 in the mssm and @xmath55 in ued . since an overall rescaling of all masses does not affect the shape of the distribution , we need four dimensionless parameters to specify the mass spectrum in each model ; we use the particle masses in units of @xmath0 . experimentally , these four parameters may be very difficult to obtain independently . a direct measurement of the masses of squarks / kk quarks may well be impossible , since these particles may be too heavy to be produced on - shell . also , while it is easy to measure @xmath56 ( one can use the endpoint of the dijet invariant mass distribution or other simple observables such as the effective mass @xcite or its variations @xcite ) , it is much more difficult to measure @xmath0 and @xmath1 individually @xcite , which would be required in order to obtain @xmath57 . in this study , we will conservatively assume no prior knowledge of any of these parameters . ( of course , if some independent information about them is available , for example the overall mass scale is constrained by production cross section considerations , this information can be folded into our analysis , increasing its discriminating power . ) in addition to the unknown masses , the matrix elements in the mssm depend on the neutralino mixing matrix elements , @xmath58 and @xmath59 , although only the ratio @xmath60 affects the shape of the distribution . again , this parameter is difficult to measure at the lhc , and we will assume that it is unknown ; fortunately , the effect of varying it is quite small . to quantify the discriminating power of the proposed observable , we use the following procedure . we assume that the experimental data is described by the mssm curve with a particular set of parameters . we then ask , how many events ( assuming statistical errors only ) would be required to rule out the ued as an explanation of this distribution ? to answer this question , we scan over 50000 points in the ued parameter space : m_b / m_a & = & ( 0 0.5 ) , m(q_l^1)/m_a=(1.05 3.0),m(d_r^1)/m_a & = & ( 1.05 3.0 ) , m(u_r^1)/m_a=(1.05 3.0 ) . for each point in the scan , we compute the kullback - leibler ( kl ) distance ( see appendix [ app : kl ] ) between the ued distribution with the parameters at that point , and the `` experimental '' distribution . we then find the `` best - fit ued '' point , which is the point that gives the smallest kl distance among the scanned sample . finally , we compute the number of events required to rule out the best - fit ued point at a desired confidence level . decays . [ fig : scan],title="fig:",width=37 ] decays . [ fig : scan],title="fig:",width=264 ] the results of this analysis are shown in fig . [ fig : scan ] . the mssm parameters used to generate the `` data '' are : @xmath61 , @xmath62 , @xmath63 the parameters @xmath64 and @xmath65 were then scanned between @xmath66 and @xmath67 , and for each point in the scan the procedure described in the previous paragraph was performed . [ fig : scan ] shows the number of events required to rule out the ued interpretation of the signal at the 99.9% c.l . ( in the language of appendix [ app : kl ] , this corresponds to @xmath68 . ) in a typical point in the model parameter space , about 6000 events are required . for comparison , the pair - production cross section for a 1 tev gluino at the lhc is about 600 fb , corresponding to 12000 gluinos / year at the initial design luminosity of 10 fb@xmath69/year . the number of events useful for the measurement studied here depends on the branching ratio of the decay . since this branching ratio is generically of order one , we expect @xmath70 useful events / year at the initial stages of the lhc running . thus , at least under the highly idealized conditions of this simplified analysis , this method of model discrimination is quite promising in a wide range of reasonable model parameters . the mssm and ued distributions look more alike when the squark - gluino mass splitting is small ; as a result , more events are required in that region . as the squarks become heavier , however , the distributions quickly converge to their asymptotic shapes , and the discriminating power is approximately independent of the squark masses in most of the scanned region . of the `` true '' model.,title="fig:",width=264 ] ( -5,-10)@xmath71 ( -230,120)@xmath72 we checked that the conclusions of this analysis are approximately independent of the value of @xmath60 used to generate the `` data '' . they do , however , depend sensitively on the ratio @xmath57 : as @xmath57 grows , the mssm and ued distributions become more and more alike . this is illustrated in fig . [ fig : mbma ] , which shows the number of events needed to rule out the `` wrong '' model ( assumed to be ued ) at the 99.9% c.l . , as a function of @xmath57 of the `` true '' model ( assumed to be the mssm with @xmath73 and @xmath74 ) . the ued scan parameters are the same as in eq . , except that we vary @xmath75 in this case . it is clear that the discriminating power of the dijet invariant mass distribution falls rapidly ( approximately exponentially ) with growing @xmath57 . this can be understood as follows . the main feature of the invariant mass distributions that allows for model discrimination is the presence of the sharp dip at @xmath21 in the ued case . according to the goldstone boson equivalence theorem , if the daughter particle @xmath5 in the ued case is highly boosted , the decays into its longitudinal component will dominate . the particle @xmath5 is highly boosted in the vicintiy of @xmath21 , provided that the mass ratio @xmath57 is small ; as @xmath57 grows , the boost becomes less pronounced and the decays into the longitudinal component of @xmath5 are less dominant . this is illustrated in fig . [ fig : polratio ] , which compares the ratio of partial decay rates into the longitudinal and transverse modes of @xmath5 for @xmath16 and @xmath76 . however , it is exactly the decays into the longitudinal mode of @xmath5 that are mainly responsible for the characteristic dip at @xmath21 ; this feature is far less pronounced for the decays into transverse modes . this means that as @xmath57 is increased , the dip gradually disappears , and the discriminating power of our observable fades away . ( so that the @xmath5 particle is actually highly boosted ) . in figure [ fig : polratio ] we can see the ratio of the decay distribution into the longitudinal modes to the decay modes into the transverse modes of @xmath5 for two different ratios @xmath77 . for @xmath78 we can clearly see that the longitudinal modes dominates for low @xmath11 , while for @xmath79 the decay into transverse modes is enhanced . however , it is exactly the decays into the longitudinal mode that are responsible for the characteristic dip of the distribution around @xmath21 which lets us distinguish the ued case from the mssm case . this means that the characteristic dip will disappear if we increase @xmath77 , and will our discriminating power too . into the longitudinal component of @xmath5 to the decay distributions into the transverse components of @xmath5 for @xmath16(solid ) and @xmath76(dashed ) . for low @xmath57 the daughter particle is highly boosted at @xmath21 and will mainly be longitudinally polarized . as @xmath1 increases , the transeverse polarization becomes more important.,width=264 ] given the large number of simplifying assumptions made in the analysis of the previous section , a skeptical reader may well wonder how meaningful the results presented above are . in this section , we will repeat the analysis in a more realistic setting : effects of experimental cuts and combinatoric background will be included . we will also bin the distributions , to approximate the effects of finite jet energy resolution . since this analysis involves generating large samples of monte carlo ( mc ) events for each model , we were not able to perform a scan over the model parameter space , as we did in the previous section . instead , we will present a test case , comparing the mssm distribution for a single point in the mssm parameter space with the distribution generated by the `` best - fit '' ued model for that point . the chosen mssm point has the following parameters : @xmath80 tev , @xmath81 gev , @xmath82 tev . the corresponding `` best - fit '' ued point , found by the procedure described in the previous section , has the following parameters : @xmath83 tev , @xmath84 gev , @xmath85 tev . ( note that the value of @xmath56 , which can be determined independently , is the same for these two points . ) using madgraph / madevent v4.1 @xcite event generator , we have simulated a statistically significant sample ( about 20000 ) of parton - level monte carlo events for each model in @xmath86 collisions at @xmath87 tev . the simulated processes are pp qq|q|q ^0_1 ^0_1 in the mssm , and its counterpart , pp qq|q|q b^1 b^1 , in ued . with the chosen model parameters , the dominant contribution to the processes and comes from pair - production of @xmath88 , followed by the three - body decay , which is of primary interest to us . in the mc simulation , we did not demand that the @xmath88 be on - shell ; the full tree - level matrix elements for the @xmath89 reactions and were simulated , so that the subdominant contributions with off - shell @xmath88 are included . we imposed the following set of cuts on the generated events : _ i4.0 ; r(i , j)0.4 ; p_t , i100 gev ; 100 gev , where @xmath90 , @xmath91 label the four ( anti)quarks in each event . the first three cuts are standard for all lhc analyses , reflecting the finite detector coverage , separation required to define jets , and the need to suppress the large qcd background of soft jets . the @xmath92 cut is common to all searches for models where new physics events are characterized by large missing transverse energy , such as the mssm and ued models under consideration . detailed studies have shown that this cut is quite effective in suppressing the sm backgrounds , including both the physical background , @xmath93 , @xmath94 , and a variety of instrumental backgrounds ( see , for example , the cms study @xcite ) . while we have not performed an independent analysis of the sm backgrounds , based on previous work we expect that , with a sufficiently restrictive @xmath92 cut , one will be able to obtain a large sample of new physics events with no significant sm contamination . ( green / light - gray ) , and its ued counterpart @xmath95 ( blue / dark - gray ) , including realistic experimental cuts and the combinatoric background ( monte carlo simulation ) . right panel : theoretical dijet invariant mass distributions > from a single gluino / kk gluon decay with the same model parameters and no experimental cuts . [ fig : mg],width=302 ] ( -60,-105)@xmath11(gev@xmath96 ) ( -220,90)@xmath36 ( green / light - gray ) , and its ued counterpart @xmath95 ( blue / dark - gray ) , including realistic experimental cuts and the combinatoric background ( monte carlo simulation ) . right panel : theoretical dijet invariant mass distributions > from a single gluino / kk gluon decay with the same model parameters and no experimental cuts . [ fig : mg],width=264,height=241 ] ( -10,-105)@xmath11 ( -215,90)@xmath97 the dijet invariant mass distruibutions obtained from the mssm and ued mc samples are shown in fig . [ fig : mg ] . the distributions are normalized to have the same total number of events , since the overall normalization is subject to large systematic uncertainties and we do not use any normalization information in our study . note that for each mc event , we include all @xmath98 possible jet pairings ; 4 out of these correspond to combining jets that do _ not _ come from the same decay , and thus do not follow the theoretical distributions computed above . in fig . [ fig : mg ] , we selected can be measured independently . ] the jet pairs with @xmath99 ; all pairs with larger values of @xmath11 arise > from the wrong jet pairings . however , some of the wrong jet pairs do have @xmath11 in the selected range , forming a combinatoric background to the distribution we want to study . nevertheless , it is clear from fig . [ fig : mg ] that even after realistic cuts and the combinatoric background are included , the distributions in the two models retain their essential shape difference expected from the simplified theoretical analysis of the previous section . assuming that the experimental data is described by the mssm histogram and ignoring systematic uncertainties , we find ( using the standard @xmath100 test ) that about 750 events would be required to rule out the ued curve at the 99.9% c.l . note that this number is smaller than those obtained in the previous section , indicating that the performed cuts actually enhance the difference between the mssm and ued distributions . on the other hand , the actual discriminating power of the analysis is likely to be somewhat lower than this estimate , since the systematic uncertainty in the cut efficiencies was not taken into account here . our parton - level analysis does not explicitly take into account the smearing effect due to the finite jet energy and direction resolution of a real detector . the hadronic calorimeter energy resolution for a jet of energy e can be approximated by 0.05 + ( ) ^0.5 , and is in the @xmath101% range for the jets that pass the cuts . we can crudely estimate @xmath102 to be of order @xmath103 , evaluated at @xmath104 . the fractional uncertainty of the measurement of @xmath11 in our analysis is then roughly between 10% ( for points with @xmath105 ) and 30% ( for points with low @xmath11 ) . the bin size used in fig . [ fig : mg ] is of the order of this uncertainty for large @xmath11 , and larger for small @xmath11 , so we expect that the smearing introduced by binning in our analysis provides a reasonable , if crude , description of the expected smearing due to finite jet energy resolution . a more detailed investigation of this effect , and other potential detector effects , would be required to fully understand the applicability of the proposed method in a realistic experimental situation . in this paper , we have investigated how the dijet invariant mass distributions > from three - body decays of a color - octed tev - scale new particle , such as the gluino of the mssm and the kk gluon of the ued model , can be used to determine the nature of this particle . the production cross section for the color - octet state at the lhc is expected to be large , and the branching ratio for the three - body decays is significant whenever all squarks / kk quarks are heavier than the gluino / kk gluon . if this is the case , the dijet invariant mass distribution can be determined accurately at the lhc . the main complication of the analysis is that the distributions in the two models we considered depend on a number of parameters in addition to the spin of the decaying particle . however , even allowing for complete ignorance of these parameters , we found the dijet invariant mass to be a very promising tool for model discrimination . the simplified analysis of this paper did not take into account a number of potentially important effects . since the particles involved are colored , the qcd loop corrections to the decay amplitudes are expected to be significant , and may modify the tree - level distributions we studied . also , our analysis is performed at the parton level and does not include detector effects . while we expect that many systematic effects would cancel out since the analysis relies only on the shapes of the distributions and is insensitive to the overall normalization , a better understanding of the systematics is required . we believe that the promising conclusions of this preliminary analysis motivate a more detailed study of these issues . 1.0 cm we are grateful to matt reece and itay yavin for useful discussions . this research is supported by the nsf grant phy-0355005 . is also supported in part by the doe oji grant de - fg02 - 01er41206 . the main feature of the invariant mass distribution of the ued case , which makes it distingishable from susy , is the dip at @xmath21 . this feature can be understood by analyzing the decay amplitudes of the individual polarization components of the mother and daughter particles and considering conservation of angular momentum . as shown with the two toy models in section [ sec : toy ] , conservation of angular momentum can lead to suppression of the invariant mass distributions with respect to the pure phase space distribution at @xmath21 , as well as at @xmath15 . the couplings in the ued case have the same chiral structure as the second toy model of section [ sec : toy ] , with the quark and antiquark having opposite helicities . the added complication in the ued case is that the mother and daughter particles are massive spin one particles . we use @xmath106 and @xmath107 to denote the projections of the @xmath4 and @xmath5 spins on the direction of the momentum @xmath8 of the quark @xmath6 . these operators have eigenvalues @xmath108 ; the corresponding eigenstates have polarization vectors @xmath109 , @xmath110 , and @xmath111 . the transitions among these eigestates are described by a @xmath112 matrix of decay amplitudes . using the ued lagrangian , we have evaluated these amplitudes and obtained the dijet invariant mass distribution corresponding to each entry . in the intermediate state . the diagrams with @xmath113 lead to distributions that are identical , up to a parity reflection , to the ones presented here . ] these distributions , divided out by the pure phase space distribution , are plotted in fig . [ fig : pol ] . at @xmath21 the spin projections of the quark - antiquark pair sums up to zero , and the final state has no angular momentum ( see the right panel of fig . [ fig : am ] ) . therefore the polarizations of @xmath4 and @xmath5 must be the same . this will result in a suppression of all non - diagonal components in the transition matrix at @xmath21 , resulting in a dip there . at @xmath15 , however , the spin projections of the quark - antiquark pair add up to @xmath114 ( see fig . [ fig : am ] ) . thus the only allowed decays at @xmath115 are the longitudinal component of @xmath4 to @xmath116 and @xmath117 to the longitudinal component of @xmath5 . both features at the ends of the distribution can be nicely observed in fig . [ fig : pol ] . and @xmath118 in arbitrary units . the polarization vectors are along the momentum @xmath8 of the outgoing quark @xmath6 . notice that at @xmath21 only the diagonal elements are unsuppressed due to angular momentum conservation , resulting in a dip of the distribution . [ fig : pol],title="fig:",width=529 ] ( -420,290)@xmath119 ( -420,205)@xmath120 ( -420,120)@xmath121 ( -340,330)@xmath122 ( -200,330)@xmath123 ( -70,330)@xmath124 the main feature of the invariant mass distribution of the ued case ( the dip at @xmath21 ) can be understood by looking at the decay of the individual polarization components of the mother and daughter particles and considering conservation of angular momentum . as shown with our toy models in section [ sec : toy ] , conservation of angular momentum can in some kinematic cases lead to the suppression of the invariant mass distributions compared to the pure phase space distribution ( by pure phase space we mean the invariant mass distribution with the matrix element set one , thus considering only pure kinematic effects and the propagator of the intermediate particle ) . figure [ fig : pol ] shows the invariant mass distributions of these individual decay modes divided by the phase space distribution . the ued case has the same chirality assignments as our second toy model in fig . [ fig : am ] , with the obvious added complication that the mother and daughter particles are massive spin one particles . apply now the argument of fig . [ fig : am ] along the direction of the quark @xmath125 . at @xmath21 the spin projection of the quark - antiquark pair sums up to zero , therefore the polarization of @xmath5 must be the same as the one of @xmath4 . this will result in a suppression ( compared to pure phase space ) of all other components at @xmath21 . at @xmath126 however , the quark - antiquark pair has spin projection @xmath127 . thus the only allowed decays at @xmath115 are the longitudinal component of @xmath4 to @xmath128 and @xmath129 to the longitudinal polarization of @xmath5 . both features at the ends of the distribution can be nicely observed in [ fig : pol ] . a convenient measure to quantify how much two continuous probability distributions differ from each other is the _ kullback - leibler distance_. ( for a recent application in the collider phenomenology context , see ref . @xcite . ) in this appendix , we will briefly review this measure . suppose that the data sample consists of @xmath36 events distributed according to the theoretical prediction of model @xmath130 . consider a second model , @xmath131 , which predicts a distribution different from @xmath130 . we can quantify the discriminating power of our data sample by the ratio of conditional probabilities for @xmath131 and @xmath130 to be true , given the data : = . this equation can be rewritten using bayes theorem : @xmath132 where @xmath133 and @xmath134 are the priors probabilities for @xmath131 and @xmath130 to be true before the experiment at hand is conducted . ( in this paper , we assumed that the mssm and ued are _ a priori _ equally likely , so we set @xmath135 . ) suppose that each event @xmath136 ( @xmath137 ) is characterized by a single variable @xmath138 ( in our case , the dijet invariant mass ) . since the @xmath36 events are independent , we have @xmath139 for large @xmath36 we can approximate @xmath140 and use the normalization condition @xmath141 to obtain @xmath142 where the _ kullback - leibler distance _ ( also called _ relative entropy _ ) is defined as @xmath143 it follows that the number of events needed to constrain the probability of model @xmath131 being true , relative to the probability of @xmath130 being true , to be less than @xmath144 , is given by @xmath145 this number provides a convenient and physically meaningful measure of how different the @xmath131 and @xmath130 distributions are . two properties of the kullback - leibler distance are worth mentioning in our context . first , while this is not manifest from its definition , the kl distance is non - negative , and zero if and only if @xmath146 . second , it is invariant under transformations @xmath147 : for example , it does not matter whether we consider the jet invariant mass distribution in terms of @xmath11 or @xmath148 . a. j. barr , phys . b * 596 * , 205 ( 2004 ) [ arxiv : hep - ph/0405052 ] . j. m. smillie and b. r. webber , jhep * 0510 * , 069 ( 2005 ) [ arxiv : hep - ph/0507170 ] . a. datta , k. kong and k. t. matchev , phys . rev . d * 72 * , 096006 ( 2005 ) [ erratum - ibid . d * 72 * , 119901 ( 2005 ) ] [ arxiv : hep - ph/0509246 ] . c. athanasiou , c. g. lester , j. m. smillie and b. r. webber , jhep * 0608 * , 055 ( 2006 ) [ arxiv : hep - ph/0605286 ] . l. t. wang and i. yavin , jhep * 0704 * , 032 ( 2007 ) [ arxiv : hep - ph/0605296 ] . c. kilic , l. t. wang and i. yavin , arxiv : hep - ph/0703085 . a. alves , o. eboli and t. plehn , phys . d * 74 * , 095010 ( 2006 ) [ arxiv : hep - ph/0605067 ] . m. toharia and j. d. wells , jhep * 0602 * , 015 ( 2006 ) [ arxiv : hep - ph/0503175 ] . h. e. haber and g. l. kane , phys . rept . * 117 * , 75 ( 1985 ) . m. perelstein and c. spethmann , jhep * 0704 * , 070 ( 2007 ) [ arxiv : hep - ph/0702038 ] . t. appelquist , h. c. cheng and b. a. dobrescu , phys . rev . d * 64 * , 035002 ( 2001 ) [ arxiv : hep - ph/0012100 ] . m. carena , t. m. p. tait and c. e. m. wagner , acta phys . b * 33 * , 2355 ( 2002 ) [ arxiv : hep - ph/0207056 ] . i. hinchliffe , f. e. paige , m. d. shapiro , j. soderqvist and w. yao , phys . d * 55 * , 5520 ( 1997 ) [ arxiv : hep - ph/9610544 ] . c. g. lester and d. j. summers , phys . b * 463 * , 99 ( 1999 ) [ arxiv : hep - ph/9906349 ] ; a. barr , c. lester and p. stephens , j. phys . g * 29 * , 2343 ( 2003 ) [ arxiv : hep - ph/0304226 ] . p. meade and m. reece , phys . d * 74 * , 015010 ( 2006 ) [ arxiv : hep - ph/0601124 ] . can be downloaded from http://madgraph.phys.ucl.ac.be/ + for a discussion of the madgraph project , see f. maltoni and t. stelzer , jhep * 0302 * , 027 ( 2003 ) [ arxiv : hep - ph/0208156 ] . , cern - lhcc-2006 - 021 , pp .

non - gaussianity of the cosmic microwave background ( cmb ) is progressively becoming a crucial probe of the inflationary dynamics , since models beyond single field slow - roll inflation predict large distinct non - gaussianities above the current limit of measurement ( lyth et al . 2003 , 2006 ; bartolo et al . 2004 ; dvali et al . 2004 ; chen 2005 ; seery & lidsey 2005 ; chen & szapudi 2006 ; seery & hidalgo 2006 ; ling & wu 2008 ; senatore et al . 2008 , 2009 ; chen et al . moreover , up - to - date a number of plausible alternative early universe models also predict skewed primordial fluctuations , and are in principle distinguishable from other scenarios through the shape dependence of high - order correlation functions ( alishahiha et al . 2004 ; arkani - hamed et al . 2004 ; creminelli & senatore 2007 ; brandenberger 2008 ; buchbinder , khoury & ovrut 2008 ; lehners & steinhardt 2008 ; mcallister & silverstein 2008 ) . an even small degree of primordial non - gaussianity would indicate a quite different structure formation scenario from the concordance cosmological model in which density perturbations are assumed to be a gaussian random field , and alter significantly the statistics of voids ( kamionkowski et al . 2008 ; song & lee 2008 , for example ) . imprints of primordial non - gaussianity can manifest in the statistical properties of the lyman-@xmath0 forest qso spectra at intermediate redshift ( viel et al . 2008 ) , in the large - scale distribution of neutral hydrogen ( pillepich et al . 2007 ) , in the reionization history of the universe ( crociani et al . 2008 ) , and in the abundance , clustering and biasing of dark matter halos ( kang et al . 2007 ; carbone et al . 2008 ; dalal et al . 2008 ; desjacques et al . 2008 ; grossi et al . 2008 ; loverde et al . 2008 ; matarrese & verde 2008 ; mcdonald & seljak 2008 ; pillepich et al . 2008 ; seljak 2008 ; slosar et al . 2008 ; jeong & komatsu 2009 ) . topology of the large scale structure is also very sensitive to the initial conditions of the matter density field , and potentially represents a strong and independent test of deviations from the gaussian hypothesis ( park et al . 1998 , 2001 , 2005 ; gott et al . 2008a , b ; hikage et al . 2008b ) . hence , there is significant interest in quantifying the gaussianity of the cosmic background radiation , and a multitude of non - gaussian estimators have been applied so far . in particular , data from the _ _ w__ilkinson _ _ m__icrowave _ _ a__nisotropy _ _ p__robe ( wmap ) mission has been used to constrain primordial non - gaussianity . recent measurements of the one- and three - point probability functions , the bispectrum and trispectrum , the genus statistic and the other minkowski functionals ( cabella et al . 2005 ; hinshaw et al . 2007 , 2008 ; creminelli et al . 2007 ; gott et al . 2007 ; spergel et al . 2007 ; curto et al . 2008 ; hikage et al . 2008a ; komatsu et al . 2008 ; vielva & sanz 2008 ; pietrobon et al . 2009a ) from the first three or five years of wmap data are all consistent with gaussianity . however , a considerable number of anomalies like asymmetries , intrinsic alignments in the data or the presence of unusual cold spots have also been reported ( cruz et al . 2006 ; tojeiro et al . 2006 ; chiang et al . 2007 ; naselsky et al . 2007 ; rth et al . 2007 , 2009 ; vielva et al . 2007 ; hansen et al . 2008 ; dickinson et al . 2009 ; diego et al . 2009 ; li et al . 2009 ; pietrobon et al . 2009b ; rossmanith et al . 2009 ) , as well as some claims of non - gaussianity ( jeong & smoot 2007 ; yadav & wandelt 2008 ) , which , if confirmed , would rule out a large class of inflationary models . in this paper we investigate the gaussian hypothesis using the clustering statistics of pixels that lie above or below a threshold . for a gaussian field , the scale dependence of this statistic , and its dependence on threshold , have been predicted ( jensen & szalay 1986 ) . it turns out that a careful accounting for the effects of noise is crucial to performing this test . kashlinsky et al . ( 2001 ) and hernndez - monteagudo et al . ( 2004 ) presented an analysis of the problem when the noise is spatially homogeneous . in the case of wmap , the noise is spatially inhomogeneous ; the present work shows how to incorporate this complexity into the analysis . the layout of the paper is as follows . in section [ noise_model ] we outline the theoretical framework for the two - point statistics above a temperature threshold , and show how to incorporate the complexity of inhomogeneous noise into the two - point formalism . some details on errorbar calculations are left in appendix a. in section [ wmap5 analysis ] we test our theory against wmap five - year data . we study and quantify the inhomogeneous noise properties and show that the abundance and clustering of pixels ( both scale and dependence on threshold temperature ) appears to be inconsistent with the gaussian prediction if one ignores the fact that the noise is spatially inhomogeneous . we compare the number density of pixels above threshold and the one- and two - point statistics measurements with the theory , allowing for inhomogeneous noise . while this improves the agreement between measurements and predictions , discrepancies remain . although this may be an indication of primordial non - gaussianity , we discuss a variety of other possibilities , among which the effect of smoothing the map . we inspect all the wmap five - year channels , but we present in the main text results only for the w1 differencing assembly ( da ) , and leave in appendix b results for other da s . a final section summarizes our findings , and highlights ongoing and future studies . denote the observed value in a pixel by @xmath1 , which is the sum of the true signal @xmath2 plus noise @xmath3 , both of which have mean zero . we consider a model in which the signal is homogeneous and may have spatial correlations whereas the noise , which is independent of the signal , may be inhomogeneous and have spatial correlations . by this we mean that the rms value of the noise @xmath4 may fluctuate from pixel to pixel and these fluctuations may be correlated , and that the actual value of the noise in one pixel could depend of that in another . let @xmath5 denote the observed one - point distribution of @xmath6 , @xmath7 the distribution of @xmath2 , @xmath8 the distribution of the rms value of the noise in a pixel , and @xmath9 the distribution of the noise when the rms value of the noise is @xmath4 . note that @xmath10 . our convention is to use capital letters for average quantities , and lower case letters for actual ( varying ) quantities . later on in the paper we will also drop the ( understood ) subscript @xmath3 of the rms of the noise , for clarity of notation . since the distribution of the noise @xmath11 is independent of the signal @xmath2 , @xmath12 where @xmath13 is the dirac delta and @xmath14 the variance of @xmath6 is : @xmath15 where we have used @xmath16 to denote the rms value of the signal , and @xmath17 to denote the variance of the noise upon averaging over all pixels . it is also straightforward to show that @xmath18 , as expected . the fraction of pixels above some threshold @xmath19 is @xmath20 if @xmath21 denotes the mean value of @xmath6 in such pixels , then @xmath22 two point statistics may be computed similarly . for two pixels separated by the angular distance @xmath23 , @xmath24 where @xmath25 notice that if we integrate over all pixels then @xmath26 where @xmath27 and @xmath28 this is because the signal and noise are uncorrelated , and so @xmath29 for all @xmath30 . clearly , if the noise is not spatially correlated , then @xmath31 and @xmath32 . if the rms values of the noise are perfectly correlated , meaning @xmath33 , then it is convenient to write @xmath34 and replace @xmath35 with a delta function centered on @xmath36 . one then averages over the distribution of @xmath4 , namely : @xmath37 if , in addition , one replaces @xmath8 with a delta function centered on @xmath38 , then @xmath39 , and the homogeneous case is recovered . the noise distributions @xmath8 and @xmath40 are measured directly from the wmap five - year data . the two - point unweighted correlation function @xmath41 estimates the excess of probability of finding a pair at a distance @xmath23 , compared to a random catalog ( e.g. jensen & szalay , 1986 ) . similarly , one can define the two - point weighted correlation function @xmath42 , which differs from @xmath41 only because each member of the pair is now weighted by some ( average ) mark . in this study , the mark is given by @xmath43 . two - point correlation functions of regions above some threshold are readily obtained by integration above the threshold level . within the formalism previously outlined , the two point correlation function of pixels above some threshold @xmath19 is @xmath44 and the associated weighted function is @xmath45 where @xmath46 , @xmath47 and @xmath48 have been previously defined . to gain intuition , it is helpful to define @xmath49 and @xmath50 in terms of which @xmath51 \label{xiintuit}\end{aligned}\ ] ] and @xmath52 . \label{wiintuit}\end{aligned}\ ] ] this shows that @xmath53 and @xmath54 can be thought of as weighted averages over the values at fixed @xmath55 and @xmath56 . this re - writing shows clearly that the key quantity of interest are @xmath57 , which is the convolution of the signal and noise distributions , and @xmath58 . this rewriting also allows one to study two limiting cases . if the rms values of the noise are perfectly correlated , meaning @xmath59 , then : @xmath60 and @xmath61 if , in addition , one replaces @xmath8 with a delta function centered on @xmath38 , then @xmath62 and @xmath63 we are particularly interested in the case where the signal @xmath64 is bivariate gaussian with @xmath65 , and @xmath66 and the noise @xmath9 is gaussian with rms @xmath4 . then @xmath67 with @xmath68 , @xmath69 , and @xmath70 @xmath71 @xmath72 where @xmath73 is the covariance matrix of the temperature field , @xmath74 the power spectrum of the noise map , @xmath75 the _ wmap _ window function and @xmath76 the additional smoothing due to finite pixel size , mask influence and an optional additional gaussian beam smoothing . if the noise is spatially uncorrelated , then clearly @xmath77 and therefore @xmath78 . in the approximation where @xmath33 , rms noise varies spatially on scales much larger than those of interest , then @xmath79 . the `` standard '' approximation , rms noise independent of position , has @xmath80 and we recover the hernndez - monteagudo et al . ( 2004 ) formula . uncertainties in the correlation functions above threshold are estimated from the optimal variance limit , containing cosmic variance , instrumental noise ( knox 1995 ; hernndez - monteagudo et al . 2004 ) , and finite binsize effects . see appendix a for more details . the data from the five years of the wmap mission are available online at _ _ _ a__rchive for _ _ m__icrowave _ _ b__ackground _ _ da__ta ( lambda ) website ( http://lambda.gsfc.nasa.gov/ ) , nasa s cmb thematic data center . for the purposes of this study , we use the wmap five - year full resolution `` foreground - reduced '' coadded sky maps ( produced by performing a weighted pixel - by - pixel mean of the five single year maps , see hinshaw et al . 2008 ) for the w ( the 94 ghz channel ) , v ( the 61 ghz channel ) and q ( the 41 ghz channel ) differencing assemblies . we show results for the w1 channel in the main text , and leave in appendix b major results for all the other individual da s . the data consist of four fields for each pixel : 1 . the thermodynamic temperature in mk , 2 . the q polarization temperature in mk , 3 . the u polarization temperature in mk , and 4 . the effective number of observations , @xmath81 . the maps are provided in the healpix scheme ( grski et al . 1999 ) at a resolution of @xmath82 , giving a total of 3145728 pixels separated on average by @xmath83 . the maps have been cleaned of galactic foreground emission using external templates , as explained in gold et al . it is indeed necessary to mask out regions of strong foreground emission . of the different masks provided from the lambda legacy archive , which allow for the selective exclusion of portions of the sky at different flux levels , we chose the most conservative , kq75 essentially an extension and improvement of the standard kp0 cut in the 3-yr data release . this choice removes about 28% of the pixels . point sources are masked based on a combination of external catalog data and wmap - detected sources ( wright et al . note that point sources are the largest astrophysical contaminant to the temperature power spectrum . our theory predictions ( section [ noise_model ] ) depend on the tt power spectrum , on the shape of the wmap beam , and on the noise properties . we used the beam and window transfer functions from the five - year wmap data , which have been significantly improved with the 5yr data release ; models of the instrument gain and beam response are now accurate to better than @xmath84 , and errors in the five - year beam transfer functions are reduced by a factor of @xmath85 as compared to the three - year analysis ( hill et al . the wmap5 tt power spectrum we use comes from a weighted combination of 153 individual cross - power spectra , and has been improved by using a gibbs - based maximum likelihood estimate for @xmath86 ( dunkley et al . 2008 ) and a pseudo @xmath87 estimate for higher @xmath88 ( nolta et al . the pixel - pixel covariance matrix is essentially diagonal ( hinshaw et al . 2003 , 2008 ) , so the rms noise in pixel @xmath89 is @xmath90 , where the noise per observation , @xmath91 , is provided for each coadded channel . this value for @xmath91 is obtained by averaging the values of the five possible year - by - year difference combinations ( jarosik et al . 2007 ; hinshaw et al . 2008 ) . the data reduction process should be free of contamination induced by foregrounds , and should be insensitive to monopole and dipole moments . at @xmath82 , the generic data pipeline involves the following main steps : 1 . dipole and monopole removal outside kq75 ; 2 . selection of pixels above a temperature threshold ; 3 . healpix coordinate and pixel - noise assignment . for the five - year coadded wmap data ( w1 channel ) , after application of the kq75 mask , in mollweide projection . the scale is logarithmic , in mk units i.e. @xmath92 is shown.,scaledwidth=45.0% ] for the wmap5 w1 coadded channel , after application of the kq75 mask.,scaledwidth=45.0% ] the distribution of the rms noise values in a pixel at @xmath82 is shown in mollweide projection in figure [ psigma_map ] , and as histogram in figure [ psigma_hist ] , for the wmap5 w1 channel . the distribution is reasonably broad , indicating that the noise is inhomogeneous . this is a direct consequence of the fact that not all pixels were observed the same number of times . the evident symmetrical pattern present in figure [ psigma_map ] is due to the scanning strategy of the wmap satellite . at @xmath82 resolution for four different choices of @xmath23 as indicated in the panels , from the wmap5 w1 coadded channel . [ psigma12],scaledwidth=49.0% ] in particular , equations ( [ xiintuit ] ) and ( [ wiintuit ] ) show that the clustering of pixels separated by @xmath23 depends on the joint distribution of @xmath93 given @xmath23 . figure [ psigma12 ] shows an example of how the joint distribution @xmath40 varies as a function of separation @xmath23 for the resolution @xmath82 . the four panels were constructed by randomly selecting @xmath94 pixels , and for each selected value of @xmath95 , looking at the corresponding @xmath56 distributions at four different angular distances . note that the scatter around the one - to - one line increases with increasing @xmath23 . this is not surprising ; nearby pixels have similar @xmath81 whereas more widely separated ones do not ( figure [ psigma_map ] ) . figure [ sigma12fits ] shows some examples of the corresponding conditional distributions @xmath96 . when inserted into equations ( [ xiintuit ] ) and ( [ wiintuit ] ) , such fits allow a prediction of how the clustering of pixels above a certain threshold temperature should depend on threshold if the signal is gaussian and the noise is inhomogeneous . the histogram in figure [ pdobs ] shows the distribution of temperature in the data , for the w1 coadded channel . for homogeneous noise , the predicted distribution would be given by convolving the expected signal ( in this case gaussian with variance determined by smoothing the best - fit power spectrum on the scale of a pixel ) with that of the noise ( gaussian with average rms @xmath38 ) . for inhomogeneous noise , the predicted distribution ( i.e. equation [ pd ] ) is given by convolving the expected gaussian signal with the noise of rms @xmath97 , and then weighting by the distribution of @xmath97 shown in figure [ psigma_hist ] . the dotted line in figure [ pdobs ] shows a gaussian with the same rms as the data . this fit has shorter tails than the data . while this may be indicating that the signal is non - gaussian ( in fact , the sample skewness of the temperature distribution is @xmath98 mk@xmath99 , and the sample excess kurtosis 0.12698 mk@xmath100 ) , the solid curve in figure [ pdobs ] shows the predicted distribution when inhomogeneous noise effects are included : it provides substantially better agreement with the measurements . nevertheless , the one - point distribution itself can not rule out a priori the presence of a non - gaussian signal , for reasons made explicit in fang & pando ( 1997 ) ; hence it is not a suitable statistics to investigate departures from gaussianity . this is also relevant for the detection of non - gaussianity claimed by jeong & smoot ( 2007 ) , where the one - point distribution was used . we also note that their equation ( 9 ) is missing one integration over the distribution of the noise ( see again our equation [ pd ] for comparison ) . for the wmap5 w1 channel ( histogram ) . solid line shows the expected distribution ( equation [ pd ] ) given the distribution of the noise . dotted line is a gaussian with the same rms as the data . [ pdobs],scaledwidth=45.0% ] , for the wmap5 w1 channel . points are measurements from the data . dotted line is the theoretical prediction for homogeneous noise , and solid line includes inhomogeneity . different temperature thresholds are considered , as explained in the main text.,scaledwidth=49.0% ] the number density of pixels above ( below ) a given temperature threshold is simply obtained by multiplying the fraction of pixels above ( below ) the threshold @xmath101 ( eq . [ fdt ] ) by the total number density after masking ( i.e. the total number of pixels after masking over the `` available '' or `` unmasked '' area ) . if the noise were homogeneous , then this prediction is rather straightforward : equation ( [ fdt ] ) simply reduces to @xmath102 , where @xmath103 . in practice , due to the inhomogeneity of the noise , one needs to integrate ( [ fdt ] ) numerically , using the distributions @xmath104 measured from the data ( figure [ psigma_hist ] ) . the solid lines in figure [ number_density_512 ] are the results of this integration , when accounting for inhomogeneous noise . dotted lines are the theoretical predictions which assume homogeneous noise . points in figure [ number_density_512 ] show measurements from the wmap5 w1 channel , for pixels above or below a threshold , where @xmath105 , respectively . errorbars are from a poisson analysis . while the solid lines slightly improve the fits , significant departures still remain , especially at higher thresholds . we have measured the correlation function of pixels above and below a threshold temperature from the wmap five - year coadded maps , for a range of thresholds . in each case , we measured the signal in two ways : one in which all pixels are treated equally , and another in which the pixels are weighted by their temperature . we adopt here the standard estimator @xmath106 for the correlation function ( jensen & szalay 1986 ) , where @xmath107 and @xmath108 are the number of data and random pairs , respectively . in particular , the number of random pairs is computed by distributing random points on a unit sphere , and then by applying the same procedures ( i.e. kq75 masking and healpix coordinate assignment ) as for the data pipeline . for the weighted correlation estimates , the number of data pairs , @xmath107 , is simply replaced by its temperature weighted counterpart , @xmath109 . we also tested the landy & szalay ( 1993 ) and the hamilton ( 1993 ) estimators , and found no significant differences in the calculations . figures [ xihi ] and [ wihi ] show an example of the results for pixels with @xmath110 , @xmath111 , @xmath112 , @xmath113 , @xmath114 and @xmath115 , respectively . in all panels , symbols show the measurements of hot and cold pixels at the fixed threshold ; errors are from a poisson analysis . dotted curves in all the left panels show the results based on homogeneous noise , solid curves in all the right panels account for the fact that the noise is inhomogeneous ( equations [ xitheta ] and [ wtheta ] ) . shaded areas are derived as explained in appendix a ( see also hernndez - monteagudo et al . 2004 ) , and represent the @xmath116 optimal variance errors . accounting for inhomogeneity significantly improves the fit to the data . however , discrepancies between the observed correlation functions and the theoretical predictions still remain , especially for pixels at higher thresholds , or at small angular scales . in particular , we note the interesting fact that hot and cold pixels cluster differently within the same temperature thresholds . this feature is also present in all the other wmap5 channels ( see appendix b ) , but we find that our theory is always in good agreement with the clustering of cold pixels in the q and v frequency bands , and only in slight disagreement with the hot patch clustering , especially for pixels at higher thresholds . moreover , the fact that the wmap5 data appear to be systematically stronger ( weaker ) at small ( large ) angular scales than the theoretical predictions , even at lower temperature thresholds ( figures [ xihi ] and [ wihi ] ) , is particularly true in the w channel ( see figure [ cf_512_nu_0.25_0.75_app ] for comparison ) . we suggest some possible interpretations in the discussion section and in appendix b , and present a more detailed investigation of this effect using realistic non - gaussian mock simulations in a forthcoming paper . while the detected trend may be a unique signature of primordial non - gaussianity ( i.e. a primordial non - gaussianity would in fact enhance the clustering of the cold pixels and reduce that of the hot ones ) , we study the effect of smoothing the map next , and provide some other plausible explanations for these anomalies in the final section . we have investigated the effect of smoothing the wmap five - year maps with a gaussian beam . in particular , we have tried different smoothing scales , corresponding to a fwhm of @xmath117 , @xmath118 , @xmath119 and @xmath120 , respectively . we have repeated the same analysis as for the unsmoothed maps , and characterized the number density and the clustering statistics of pixels above threshold . figure [ number_density_512_smooth ] summarizes the results for the number density ( w1 channel ) , after smoothing with different gaussian beams . the number of pixels above or below the considered temperature threshold is reduced ( a function of the smoothing scale ) , and measurements from the data and theoretical predictions appear now in agreement within about @xmath116 at the threshold levels considered . finally , we have characterized the clustering statistics and showed in figure [ clustering_smooth_512 ] the case of inhomogeneous noise , for an arbitrary choice of the threshold ( @xmath121 ) . the gaussian smoothing is included in our theory via equations ( [ ctheta_eq ] ) and ( [ ctheta_noise ] ) , in addition to the change of the noise variance distribution . after smoothing the map with a gaussian beam , the pixel noise may be no longer independent among different pixels ( i.e. @xmath122 ) . however , accounting for this contribution ( equation [ n1n2_cov ] ) resulted in no significant difference , even at small angular scales . in fact , the average variance of the noise in the w1 channel is 0.0213 mk@xmath123 , which is about 0.32@xmath124 of the typical variance of the corresponding cosmological signal . since the zero - level amplitude of the noise is already very low with respect to that of the signal , the mean amplitude of the correlated noise will be also very low . in addition to that , note that the wmap mission has been designed to have uncorrelated pixel noise , and if such correlations arise when smoothing the maps , their amplitude is expected to drop quickly with the smoothing scale . when the smoothing scale is @xmath120 , we find that the tension between data and theory is alleviated . the observed distribution of temperatures in wmap5 pixels is slightly non - gaussian ( figure [ pdobs ] ) . this departure from gaussianity is not unexpected if the intrinsic signal is gaussian but the noise distribution is inhomogeneous white noise ( figures [ psigma_map ] and [ psigma_hist ] ) . in fact , when accounting for inhomogeneous noise ( section [ noise_model ] ) , we find that the one - point statistics is substantially in better agreement with the measurements . however , the predicted dependence of clustering on pixel height is not in good agreement with the data , even after accounting for the inhomogeneity of the noise ( figures [ xihi ] , [ wihi ] ) . although we found that an appropriate smoothing scale would be able to alleviate the discrepancy between data and measurements ( figures [ number_density_512_smooth ] and [ clustering_smooth_512 ] ) , other possible reasons for these discrepancies are : + _ masking procedure_. in our analysis we adopted the kq75 mask , which allows for selective exclusions of bright portions of the sky ( @xmath125 pixel cut ) . although this is rather a drastic cut and the mask was significantly improved with the wmap five - year data release ( hill et al . 2008 ) , it may not completely account for all the galactic foreground effects . however , cutting out all the galactic plane ( @xmath126 strip ) and repeating our analysis resulted in no significant change . edge effects due to pixels which lie very close to the mask could also affect our analysis . + _ cold spot contamination_. inoue & silk ( 2007 ) suggest that the presence of low density regions in the southern galactic cap could account for our anomalies , but restricting our analysis to @xmath127 , the northern galactic cap , resulted in no significant change . + _ contamination by point sources_. lopez - caniego et al . ( 2007 ) detected 98 new sources ( i.e. @xmath128 ) not present in the wmap three - year catalog . at higher frequencies wmap estimates neglect the deviations of the point spread function from a gaussian shape , and carry out a blind search for sources . with the wmap five - year data release , the mask for point sources has been significantly improved ( wright et al . 2008 ) , and other studies ( gonzalez - nuevo et al . 2008 ; chen & wright 2008 ) updated the point source catalog . however , massardi et al . ( 2008 ) recently provided a new version of the catalog based on wmap5 data and found new point sources ( 484 sources detected ) , carrying a complementary blind and non - blind approach . this may be important to our analysis , since even a very low level contamination in the maps can produce spurious non - gaussianities . in fact , we find some discrepancies at small angular scales ( around 20 ) especially in the w band , and their origin may be associated with extragalactic dust emission ( one type of point - source contamination ) , which peaks at high frequencies . in a forthcoming study , we will be addressing the effect of contamination induced by point sources in more depth , using an updated version of the source catalog and mock simulations . + _ foreground subtraction contamination_. uncertainties in the external foreground template model used for the foreground subtractions ( gold et al . 2008 ) may introduce anomalies at the percentage level . the template itself has noise , which may be correlated at small separations . this is a delicate issue , since galactic foregrounds are non - gaussian and anisotropic , and even low level contamination in the maps can produce detectable non - gaussianities ( park et al . 2002 ; naselsky et al . 2005 ; kim , naselsky & christensen 2008 ) . on the other side , it is also worth noticing that recently vio & andreani ( 2008 ) showed that the benefits of using more sophisticated methods for foreground cleaning , such as the harmonic internal linear combination , are overestimated . we are also addressing the foreground subtraction systematics in a forthcoming study . + _ wmap beam , window function , absolute calibration and pixel - noise uncertainties_. window functions were computed from the symmetrized beam profiles following the hermite method in page et al . a typical wmap window function has an uncertainty of @xmath129 ( hill et al . 2008 ) , and these uncertainties add in quadrature in the cosmological analysis . absolute calibration uncertainties in the five - year wmap data are estimated to be @xmath130 ( hinshaw et al . recent studies ( groeneboom et al . 2009 ; kathrine wehus et al . 2009 ) have also pointed out that the noise levels of these maps are underestimated , and that there are some problems with the standard wmap transfer functions as well . it may be that these facts play an important rule in our study ( see also colombo et al . + _ secondary anisotropies and phase transitions in the early universe_. spurious non - gaussianities could arise from secondary anisotropies , such as gravitational lensing , sunyaev - zeldovich effect or sachs - wolfe effects ( babich & pierpaoli 2008 ; carbone et al . phase transitions in the early universe may also introduce a new source of non - gaussianity ( silvestri & trodden 2008 ) . all these effects may be difficult to disentangle from a pure primordial non - gaussian signal , even after a clear detection of primordial non - gaussianity . _ real non - gaussian signatures at small scales_. a better understanding of all the previous points is necessary before we can claim that the signal we see is due to primordial non - gaussianity . however , we found an interesting difference in the clustering of hot and cold pixels within the same temperature threshold level ; this fact may be a unique signature of primordial non - gaussianity . we present a detailed investigation of this feature in a forthcoming study . we finally note that , as this work was being refereed , hou et al . ( 2009 ) presented a frequentist analysis of the correlation functions of the local extrema , and also found inconsistency with gaussian simulations plus differences in the clustering of hot and cold peaks although comparison with their work is not direct since they considered bigger angular scales . while seeking for primordial non - gaussianity is at the moment a new frontier in cosmology , ongoing efforts are currently devoted to the characterization of non - gaussian confusion effects , to reliable theoretical predictions of non - gaussianity from models ( see for example boyle & steinhardt 2008 ; fergusson & shellard 2008 ; munshi & heavens 2009 ) , till the extraction of information from data ( i.e. raeth et al . 2008 ; gong et al . 2009 ) or the search for observational signatures of primordial non - gaussianity imprinted in the large - scale structure of the universe . even a small degree of primordial non - gaussianity can be a crucial probe of the inflationary dynamics or alternative universe models , hence studies of non - gaussianity may eventually become a powerful and solid probe of ultra - high energy physics and inflation . we note that our model for the effects of inhomogeneous noise may be useful in other studies ( see for instance yu & lu 2008 ) . extending our formalism for inhomogeneous noise to peak rather than pixel statistics ( for example heavens & sheth 1999 ; heavens & gupta 2001 ) is more complicated ; this will complement numerical monte - carlo analyses of this problem ( larson & wandelt 2004 ; tojeiro et al . 2006 ; ayaita et al . 2009 ; hou et al . 2009 ) , and is the subject of work in progress . we thank an anonymous referee for helpful comments and suggestions . gr thanks fernando atrio - barandela , pravabati chingangbam , jacek guzik , mike jarvis and licia verde for many useful discussions . gr is grateful to the organizers of the 24th texas symposium on relativistic astrophysics for their kind hospitality in vancouver , during which time he had interesting conversations with eiichiro komatsu and the final stage of this work was completed . gr would also like to thank kitpc and in particular robert brandenberger for hospitality in beijing during the 2009 workshop `` connecting fundamental physics with observations '' . cbp acknowledges the support of the korea science and engineering foundation ( kosef ) through the astrophysical research center for the structure and evolution of the cosmos ( arcsec ) . we acknowledge the use of the _ _ l__egacy _ _ a__rchive for _ _ m__icrowave _ _ b__ackground _ _ d__ata _ _ a__nalysis ( _ lambda _ ) , support for which is provided by the national aeronautics and space administration ( nasa ) office of space science . some of the results in this paper have been derived using the _ healpix _ package ( grski et al . this work was partly supported by nasa grant nng05gk81 g , and partly by the project of knowledge innovation program ( pkip ) of chinese academy of sciences , grant no . kjcx2.yw.w10 . 99 alishahiha , m. , silverstein , e. , & tong , d. 2004 , phys . d , 70 , 123505 arkani - hamed , n. , creminelli , p. , mukohyama , s. , & zaldarriaga , m. 2004 , journal of cosmology and astro - particle physics , 4 , 1 ayaita , y. , weber , m. , & wetterich , c. 2009 , arxiv:0905.3324 babich , d. , & pierpaoli , e. 2008 , phys . rev . d , 77 , 123011 bartolo , n. , komatsu , e. , matarrese , s. , & riotto , a. 2004 , phys . rep . , 402 , 103 bond , j. r. , & efstathiou , g. 1987 , mnras , 226 , 655 boyle , l. , & steinhardt , p. j. 2008 , arxiv:0810.2787 brandenberger , r. h. 2008 , arxiv:0808.0746 buchbinder , e. i. , khoury , j. , & ovrut , b. a. 2008 , physical review letters , 100 , 171302 cabella , p. , liguori , m. , hansen , f. k. , marinucci , d. , matarrese , s. , moscardini , l. , & vittorio , n. 2005 , mnras , 358 , 684 carbone , c. , verde , l. , & matarrese , s. 2008 , apj , 684 , l1 chen , g. , & szapudi , i. 2006 , apj , 647 , l87 chen , x. 2005 , phys . rev . d , 72 , 123518 chen , x. , & wright , e. l. 2008 , arxiv:0809.4025 chen , x. , hu , b. , huang , m .- x . , shiu , g. , & wang , y. 2009 , arxiv:0905.3494 chiang , l .- y . , naselsky , p. d. , & coles , p. 2007 , apj , 664 , 8 colombo , l. p. l. , pierpaoli , e. , & pritchard , j. r. 2008 , arxiv:0811.2622 creminelli , p. , senatore , l. , zaldarriaga , m. , & tegmark , m. 2007 , journal of cosmology and astro - particle physics , 3 , 5 creminelli , p. , & senatore , l. 2007 , journal of cosmology and astro - particle physics , 11 , 10 crociani , d. , moscardini , l. , viel , m. , & matarrese , s. 2008 , arxiv:0809.3909 cruz , m. , tucci , m.,martnez - gonzlez , e. , & vielva , p. 2006 , mnras , 369 , 57 curto , a. , martinez - gonzalez , e. , mukherjee , p. , barreiro , r. b. , hansen , f. k. , liguori , m. , & matarrese , s. 2008 , arxiv:0807.0231 dalal , n. , dor , o. , huterer , d. , & shirokov , a. 2008 , phys . rev . d , 77 , 123514 desjacques , v. , seljak , u. , & iliev , i. t. 2008 , arxiv:0811.2748 dickinson , c. , et al . 2009 , arxiv:0903.4311 diego , j. m , cruz , m. , gonzalez - nuevo , j. , maris , m. , ascasibar , y. , & burigana , c. 2009 , arxiv:0901.4344 dunkley , j. , et al . 2008 , arxiv:0803.0586 dvali , g. , gruzinov , a. , & zaldarriaga , m. 2004 , phys . d , 69 , 083505 fang , l .- z . , & pando , j. 1997 , arxiv : astro - ph/9701228 fergusson , j. r. , & shellard , e. p. s. 2008 , arxiv:0812.3413 grski , k. m. et al . , 1999 , in proc . mpa / eso conf . `` evolution of large - scale structure : from recombination to garching '' gold , b. , et al . 2008 , arxiv:0803.0715 gong , y. , wang , x. , zheng , z. , & chen , x. 2009 , arxiv:0904.4257 gonzlez - nuevo , j. , massardi , m. , argeso , f. , herranz , d. , toffolatti , l. , sanz , j. l. , lpez - caniego , m. , & de zotti , g. 2008 , mnras , 384 , 711 gott , j. r. , colley , w. n. , park , c .- g . , park , c. , & mugnolo , c. 2007 , mnras , 377 , 1668 gott , j. r. i. , et al . 2008 , apj , 675 , 16 gott , j. r. i. , choi , y .- y . , park , c. , kim , j. , & for the sdss collaboration 2008 , arxiv:0812.1406 groeneboom , n. e. , eriksen , h. k. , gorski , k. , huey , g. , jewell , j. , & wandelt , b. 2009 , arxiv:0904.2554 grossi , m. , branchini , e. , dolag , k. , matarrese , s. , & moscardini , l. 2008 , mnras , 390 , 438 hamilton , a. j. s. 1993 , evolution of galaxies and their environment , 15 hansen , f. k. , banday , a. j. , gorski , k. m. , eriksen , h. k. , & lilje , p. b. 2008 , arxiv:0812.3795 heavens , a. f. , & gupta , s. 2001 , mnras , 324 , 960 heavens , a. f. , & sheth , r. k. 1999 , mnras , 310 , 1062 hernndez - monteagudo , c. , kashlinsky , a. , & atrio - barandela , f . 2004 , a@xmath131a , 413 , 833 hikage , c. , coles , p. , grossi , m. , moscardini , l. , dolag , k. , branchini , e. , & matarrese , s. 2008 , mnras , 385 , 1613 hikage , c. , matsubara , t. , coles , p. , liguori , m. , hansen , f. k. , & matarrese , s. 2008 , mnras , 389 , 1439 hill , r. s. , et al . 2008 , arxiv:0803.0570 hinshaw , g. , et al . 2003 , apjs , 148 , 135 hinshaw , g. , et al . 2007 , apjs , 170 , 288 hinshaw , g. , et al . 2008 , arxiv:0803.0732 hou , z. , banday , a. j. , & gorski , k. m. 2009 , arxiv:0903.4446 inoue , k. t. , & silk , j. 2007 , apj , 664 , 650 jarosik , n. , et al . 2007 , apjs , 170 , 263 jensen , l. g. , & szalay , a. s. 1986 , apj , 305 , l5 jeong , d. , & komatsu , e. 2009 , arxiv:0904.0497 jeong , e. , & smoot , g. f. 2007 , arxiv:0710.2371 kamionkowski , m. , verde , l. , & jimenez , r. 2008 , arxiv:0809.0506 kang , x. , norberg , p. , & silk , j. 2007 , mnras , 376 , 343 kashlinsky , a. , hernndez - monteagudo , c. , & atrio - barandela , f. 2001 , apj , 557 , l1 kathrine wehus , i. , ackerman , l. , eriksen , h. k. , & groeneboom , n. e. 2009 , arxiv:0904.3998 kim , j. , naselsky , p. , & christensen , p. r. 2008 , arxiv:0810.4008 knox , l. 1995 , phys . d , 52 , 4307 komatsu , e. , et al . 2008 , arxiv:0803.0547 landy , s. d. , & szalay , a. s. 1993 , apj , 412 , 64 larson , d. l. , & wandelt , b. d. 2004 , apj , 613 , l85 lehners , j .- l . , & steinhardt , p. j. 2008 , phys . d , 77 , 063533 li , t .- , liu , h. , song , l .- m . , xiong , s .- l . , & nie , j .- y . 2009 , arxiv:0905.0075 ling , y. , & wu , j .- p . 2008 , arxiv:0809.3398 lpez - caniego , m. , gonzlez - nuevo , j. , herranz , d. , massardi , m. , sanz , j. l. , de zotti , g. , toffolatti , l. , & argeso , f. 2007 , apjs , 170 , 108 lo verde , m. , miller , a. , shandera , s. , & verde , l. 2008 , journal of cosmology and astro - particle physics , 4 , 14 lyth , d. h. , ungarelli , c. , & wands , d. 2003 , phys . d , 67 , 023503 lyth , d. h. , & riotto , a. 2006 , physical review letters , 97 , 121301 massardi , m. , lpez - caniego , m. , gonzlez - nuevo , j. , herranz , d. , de zotti , g. , & sanz , j. l. 2008 , arxiv:0810.2338 matarrese , s. , & verde , l. 2008 , apj , 677 , l77 mcallister , l. , & silverstein , e. 2008 , general relativity and gravitation , 40 , 565 mcdonald , p. , & seljak , u. 2008 , arxiv:0810.0323 munshi , d. , & heavens , a. 2009 , arxiv:0904.4478 naselsky , p. , chiang , l .- y . , olesen , p. , & novikov , i. 2005 , phys . d , 72 , 063512 naselsky , p. d. , christensen , p. r. , coles , p. , verkhodanov , o. , novikov , d. , & kim , j. 2007 , arxiv:0712.1118 nolta , m. r. , et al . 2008 , arxiv:0803.0593 page , l. , et al . 2003 , apjs , 148 , 39 park , c. , colley , w. n. , gott , j. r. i. , ratra , b. , spergel , d. n. , & sugiyama , n. 1998 , apj , 506 , 473 park , c. , gott , j. r. i. , & choi , y. j. 2001 , apj , 553 , 33 park , c .- g . , park , c. , & ratra , b. 2002 , apj , 568 , 9 park , c. , et al . 2005 , apj , 633 , 11 pietrobon , d. , cabella , p. , balbi , a. , de gasperis , g. , & vittorio , n. 2009 , mnras , 682 pietrobon , d. , cabella , p. , balbi , a. , crittenden , r. , de gasperis , g. , & vittorio , n. 2009 , arxiv:0905.3702 pillepich , a. , porciani , c. , & matarrese , s. 2007 , apj , 662 , 1 pillepich , a. , porciani , c. , & hahn , o. 2008 , arxiv:0811.4176 rth , c. , schuecker , p. , & banday , a. j. 2007 , mnras , 380 , 466 rth , c. , morfill , g. e. , rossmanith , g. , banday , a. j. , & grski , k. m. 2009 , physical review letters , 102 , 131301 raeth , c. , morfill , g. , rossmanith , g. , banday , a. j. , & gorski , k. m. 2008 , arxiv:0810.3805 rossmanith , g. , raeth , c. , banday , a. j. , & morfill , g. 2009 , arxiv:0905.2854 seery , d. , & lidsey , j. e. 2005 , journal of cosmology and astro - particle physics , 9 , 11 seery , d. , & hidalgo , j. c. 2006 , journal of cosmology and astro - particle physics , 7 , 8 seljak , u. 2008 , arxiv:0807.1770 senatore , l. , tassev , s. , & zaldarriaga , m. 2008 , arxiv:0812.3658 senatore , l. , smith , k. m. , & zaldarriaga , m. 2009 , arxiv:0905.3746 slosar , a. , hirata , c. , seljak , u. , ho , s. , & padmanabhan , n. 2008 , journal of cosmology and astro - particle physics , 8 , 31 silvestri , a. , & trodden , m. 2008 , arxiv:0811.2176 song , h. , & lee , j. 2008 , arxiv:0811.1339 spergel , d. n. , et al . 2007 , apjs , 170 , 377 tojeiro , r. , castro , p. g. , heavens , a. f. , & gupta , s. 2006 , mnras , 365 , 265 viel , m. , branchini , e. , dolag , k. , grossi , m. , matarrese , s. , & moscardini , l. 2008 , arxiv:0811.2223 vielva , p. , wiaux , y. , martnez - gonzlez , e. , & vandergheynst , p. 2007 , mnras , 381 , 932 vielva , p. , & sanz , j. l. 2008 , arxiv:0812.1756 vio , r. , & andreani , p. 2008 , arxiv:0811.4277 yadav , a. p. s. , & wandelt , b. d. 2008 , physical review letters , 100 , 181301 yu , b. , & lu , t. 2008 , phys . rev . d , 78 , 063008 wright , e. l. , et al . 2008 , arxiv:0803.0577 we estimate the uncertainties in the correlation functions above threshold from the optimal variance limit , containing cosmic variance , instrumental noise , and finite binsize effects . in more detail , the ultimate accuracy with which the cmb power spectrum can be determined at each @xmath88 is given by ( knox 1995 ) : @xmath132 \equiv \delta c_{\rm l}\ , \label{opt_var}\ ] ] where @xmath133 is the number of pixels , @xmath38 the average pixel noise , @xmath134 the wmap window function , and @xmath135 the fraction of the sky covered by the experiment . the uncertainty in the angular correlation function for narrow bins in @xmath23 is then ( hernndez - monteagudo et al . 2004 ) : @xmath136 ^ 2\big\}^{1/2 } \label{delta_c_theta}\end{aligned}\ ] ] where @xmath137 and @xmath76 is the additional smoothing due to finite pixel size , mask influence and an optional gaussian beam . however , in practice @xmath138 is not measured at a point , but smeared out over a region of size @xmath139 . in the limit of large @xmath88 and small @xmath23 , @xmath140 $ ] ( bond & efstathiou 1987 ) . if the bin size is not infinitesimal , one must replace : @xmath141 & \rightarrow & { 2 \over ( t_{\rm max}^2 -t_{\rm min}^2 ) } \int_{t_{\rm min}}^{t_{\rm max } } j_0 [ ( l+1/2)t]~t~{\rm d}t \nonumber \\ & = & { 2[t_{\rm max } j_{1}(k t_{\rm max } ) - t_{\rm min } j_1 ( k t_{\rm min } ) ] \over k ( t_{\rm max}^2 -t_{\rm min}^2 ) } \label{j0}\end{aligned}\ ] ] where @xmath142 and @xmath143 . this expression should then substitute @xmath144 $ ] in ( [ delta_c_theta ] ) , to give @xmath145 . the uncertainties in the correlation function above threshold are finally derived from : @xmath146 we have inspected all the wmap five - year channels ( w band at 94 ghz ; v band at 61 ghz ; q band at 41 ghz ) , and performed the same analysis described in the main text for each single differencing assembly ( da ) . for this study , it is ideal to consider individual da s rather than their combinations , as wmap beams and noise properties are well - defined within each single da . figure [ psigma_map_app ] shows the distributions of the rms noise values for all the da s considered ; note the characteristic bi - modality of these histograms , present in all the channels . figure [ pdobs_app ] shows the distributions of temperatures in the data at different frequencies . dotted lines in the figure are gaussians with the same rms as data , solid curves show the predicted distributions when inhomogeneous noise effects are included . departures from the gaussian fits are more significant in the w frequency range . figure [ number_density_512_app ] shows the number density of pixels above ( below ) different temperature thresholds . dotted lines are theoretical predictions with homogeneous noise , solid lines include inhomogeneity . points are measurements from all the wmap5 da s . finally , figure [ cf_512_nu_0.25_0.75_app ] highlights some examples of the unweighted correlation functions from the wmap5 pixel - pixel temperature fields , at @xmath82 . two pixel - thresholds are considered ( @xmath110 and @xmath147 ) , as indicated in the panels . solid curves show the predictions associated with gaussian signal plus inhomogeneous noise . points are measurements of the clustering of hot and cold pixels at corresponding temperature thresholds , and shaded areas are the @xmath148 optimal variance errors . essentially , we find that the difference in clustering between hot and cold pixels is still present in all the other da s . however , our theory is always in good agreement with the clustering of cold pixels , and in slight disagreement with that of the hot ones , both in the q and in the v channels . therefore the 94 ghz frequency ( w band ) seems to be the most discrepant one , with respect to our theoretical predictions . in particular , we checked our small - scale theory expectations against wmap mock measurements from simulations with identical beam and noise properties , and found full consistency . hence , we suggest that those small - scale discrepancies may be due to extragalactic dust emission ( one type of point - source contamination ) , which has a peak at high frequencies and is not accounted in our analysis .

in this paper we present an attempt to formalize what may be called a _ string field theory ( sft ) _ for ( closed ) topological strings with hodge property . from the very first days of string theory it was considered as a kind of generalization of the perturbative expansion of the quantum field theory in the ( functional ) integral representation . the space of graphs with @xmath0 loops with metrics on edges ( schwinger proper times ) was generalized to moduli space of riemann surfaces . indeed , the latter space really looks like a principle @xmath1 bundle over the former space near the points of maximal degeneracy ( i.e. , where maximal number of handles are pinched ) . a natural question is whether there are special string theories that degenerate exactly to quantum field theories ( may be , of the special kind ) . would it happen such theories should enjoy both finiteness of string theory and ( functional ) integral description of quantum field theory . one of the first attempts to construct a theory of this type was done by zwiebach in @xcite . he divided the moduli space into two regions : the internal piece and the boundary . he observed that surfaces representing the boundary region may be constructed from those representing the internal piece by gluing them with the help of cylinders ( with flat metric ) . therefore , he proposed to take integrals over the moduli spaces in two steps : first , to take an integral over the internal pieces , such that this would produce vertices , and then take an integral along metrics on cylinders , that would exactly reproduce integral along the schwinger parameters on graphs in qft prescription . in this approach , he came with the infinite number of vertices of different internal genera and with different number of external legs . however , he observed that such vertices satisfy quadratic relations that where a quantum version of some infinity - structure . at that time community of theoretical physicists seemed not to be impressed by the lagrangian with infinite number of ( almost uncomputable ) vertices . the next attempt was done by witten @xcite . he assumed that in topological string theories there may be a limit in the space of two - dimensional theories such that the measure of integration goes to the vicinity of the points of maximal degeneration . in the type b theories such limit seems to be the large volume limit of the target space ; this motivated witten s chern - simons - like representation for the topological string theory . this approach was further developed by bershadsky , cecotti , ooguri , and vafa in @xcite . we note that the tropical limit of gromov - witten theory @xcite ( type a topological strings ) seems to realize the same qft degeneration of string theory . indeed , the tropical limit of a riemann surface mapped to a toric variety is represented by the graph mapped to the moment map domain . in the development of topological string theory it became clear that the proper object is not just a measure on the moduli space of complex structures of riemann surfaces , but rather a differential form on this space . in original formulation these differential forms were assigned to the tensor algebra of cohomology of some complex ; such objects are called gromov - witten invariants . we say that gromov - witten invariants are qft - like if the differential forms of non - zero degree have support only in a vicinity of the points of maximal degeneration . we generalized the definition of gromov - witten invariants in @xcite by lifting it from the cohomology of a complex to the full complex . such generalization involved enlargement of the moduli space from deligne - mumford space to kimura - stasheff - voronov space @xcite , and we called it zwiebach invariants ( in fact , some pieces of this construction appeared earlier in @xcite and @xcite ) . the complex of states involved in the definition of zwiebach invariants is a bicomplex due to the action of the second differential . the second differential represents the substitution of a special vector field corresponding to the constant rotation of the phase of the local coordinate at a marked point into differential forms on the kimura - stasheff - voronov space . once we have some zwiebach invariants , it is possible to produce new zwiebach invariants by contraction of an acyclic hodge sub - bicomplex . in fact , it is one of the main properties of zwiebach invariants . consider a sub - bicomplex , where these two differentials act freely . we call it hodge contractible bicomplex . the operation of contraction of a hodge contractible bicomplex turns zwiebach invariants into induced zwiebach invariants on the coset with respect to contactible bicomplex . induced zwiebach invariants are differential forms whose support is a union of the support of the initial zwiebach invariants and small neighbourhoods of the points of maximal degeneration . this procedure is a generalization from intervals to cylinders of the procedure of induction of @xmath2-structures , see e. g. @xcite . this way we can obtain qft - like gromov - witten invariants . we just should start with zwiebach invariants that have ( in some suitable sense ) no support inside the kimura - stasheff - voronov spaces . in fact , it is even enough to consider a weaker condition , motivated by applications . that is , usually people consider the integrals of gromov - witten invarians only over the tautological classes in the moduli space of curves . so , we call a set of zwiebach invariants vertex - like if the integral over the kimura - stasheff - voronov spaces of any their non - zero component multipled by the pullback ( from the deligne - mumford space ) of any tautological class vanishes . consider vertex - like zwiebach invariants . assume we contract a hodge contractible bicomplex down to cohomology . we obtain differential forms on the deligne - mumford space , such that the integral of the product of any such form of non - zero degree with any tautological class vanishes the interior of the moduli space . integrals of such forms over the moduli spaces turn out to be sums over graphs ( corresponding to degenerate riemann surfaces ) . they resemble feynman diagramms , and generation function for the integral over moduli spaces resemble diagrammatic expansion of perturbative quantum field theory . in this paper , we do nt construct examples of vertex - like zwiebach invariants ( we are going to do this explicitly in a future publication , as well as the corresponding theory for the spaces introduced in @xcite ) . rather we conjecture that they exist and study the consequences of this assumption . we call the emerging construction the _ hodge field theory _ , and now we will explain it in some detail . first of all , degree zero parts of vertex - like zwiebach invariants induce the structure of homotopy cyclic hodge algebra on the target complex @xcite . we remind that a cyclic hodge algebra is just a hodge dgbv - algebra with one additional axiom ( @xmath3-axiom , see below ) . in fact , this structure is interesting by itself , without any reference to zwiebach invariants . it has first appeared in the paper of barannikov and kontsevich @xcite ; it captures the properties of polyvector fields on calabi - yau manifolds . more examples of dgbv algebras are studied in @xcite and @xcite . it is possible to understand the structure of dgbv - algebra as a natural generalization of the algebraic structure studied in @xcite . in the hodge field theory construction we consider only a particular case , where we obtain axioms of of a cyclic hodge algebra itself , not up to homotopy . we are aware of the fact that demanding existance of vertex - like zwiebach invariants simultaneously with vanishing homotopy piece of cyclic hodge algebra conditions may be too restrictive , and while considering only those relations that lead to axioms of cyclic hodge algebra may be too weak , however we proceed . in the hodge field theory construction we define graph expressions for the analogues of gromov - witten invariants multiplied by tautological classes using _ only cyclic hodge algebra data_. we call them hodge field theory correlators . the corresponding action of the hodge field theory is written down explicitely in section [ action ] . our main result is the proof that the hodge field theory correlators satisfy all universal equations that follow from relations among tautological classes in cohomology . the first result of this kind is due to barannikov and kontsevich . they have noticed that there is a solution of the wdvv equation that is associated to a dgbv - algebra ( this solution is the critical value of the bcov action @xcite , see ( * ? ? ? * appendix ) and ( * ? ? ? * appendix ) ) . later , we reproved this in @xcite . then , in @xcite we proved some other low - genera universal equations . here we generalize all these result and put all calculations done before in a proper framework . in particular , the main problem for us was to define a graph expression in tensors of a cyclic hodge algebra that corresponds to the full gromov - witten potential with descendants . the first steps were done in @xcite , where we introduced the definition of descendants at one point in hodge field theory ( mostly for combinatorial reasons ) . but then we observed that it is a part of a natural definition of potential with descendants in cyclic hodge algebras that appears as a special case of degeneration of vertex - like zwiebach invariants multiplied by tautological classes . in this paper , we present and study this construction . we prove in a completely algebraic way that hodge field theory correlators satisfy the same equations as a gromov - witten potential : string , dilaton , and the whole system of pdes coming from tautological relations in the cohomology of the moduli space of curves ( see also @xcite for some preliminary results ) . in what follows we will not only present the proof but also will do our best relating algebraic definitions and statements on hodge field theory to analoguous constructions and theorems in the theory of zwiebach invariants . in section [ sec - gw ] we remind all necessary facts about the axiomatic gromov - witten theory . in section [ sec - z ] we define zwiebach invariants and explain the motivation to consider the sums over graphs in cyclic hodge algebras . in section [ sec - defpot ] we define cyclic hodge algebras and the corresponding descendant potential . in section [ sec - proper ] we state the main properties of the descendant potential in cyclic hodge algebras , and the rest of the paper is devoted to the proofs . a.l . was supported by the russian federal agency of atomic energy and by the grants intas-03 - 51 - 6346 , nsh-8065.2006.2 , nwo - rfbr-047.011.2004.026 ( rfbr-05 - 02 - 89000-nwo - a ) , and rfbr-07 - 02 - 01161-a . s.s . was supported by the grant snsf-200021 - 115907/1 . s.s . is grateful to the participants of the moduli spaces program at the mittag - leffler institute ( djursholm , sweden ) for the fruitful discussions of the preliminary versions of the results of this paper . the remarks of c. faber , o. tommasi , and d. zvonkine were especially helpful . i.s . was supported by the grant rfbr-06 - 01 - 00037 . in this section we remind what is gromov - witten theory and explain its basic properties that we are going to reproduce in hodge field theory construction . let us fix a finite dimensional vector space @xmath4 over @xmath5 together with the choice of a homogeneous basis @xmath6 and a non - degenerate scalar product @xmath7 on it . let @xmath8 be a distinguished even element of the basis . consider the moduli spaces of curves @xmath9 . on each @xmath9 we take a differential form @xmath10 of mixed degree with values in @xmath11 . the whole system of forms @xmath12 is called _ gromov - witten invariants _ , if it satisfies the axioms @xcite : 1 . there are two actions of the symmetric group @xmath13 on @xmath10 . first , we can relabel the marked points on curves in @xmath9 ; second , we can interchange the factors in the tensor product @xmath11 . we require that @xmath10 is equivariant with respect to these two actions of @xmath13 . in other words , one can think that each copy of @xmath4 in the tensor product is assigned to a specific marked point on curves in @xmath9 . the forms must be closed , @xmath14 . 3 . [ pull - back ] consider the mapping @xmath15 forgetting the last marked point . then the correspondence between @xmath10 and @xmath16 is given by the formula @xmath17 the meaning of the right hand side is the following . we want to turn a @xmath18-valued form into a @xmath11-valued one . so , we take the copy of @xmath4 corresponding to the last marked point and contract it with the vector @xmath8 using the scalar product . [ factor ] consider an irreducible boundary divisor in @xmath9 , whose generic point is represented by a two - component curve . it is the image of a natural mapping @xmath19 , where @xmath20 and @xmath21 . we require that @xmath22 here on the right hand side we contract with a scalar product the two copies of @xmath4 that correspond to the node . + in the same way , consider the divisor of genus @xmath23 curves with one self - intersection . it is the image of a natural mapping @xmath24 . in this case , we require that @xmath25 as before , we contract two copies of @xmath4 corresponding to the node . 5 . we also assume that @xmath26 . let us associate to each @xmath27 the set of formal variables @xmath28 , @xmath29 . by @xmath30 denote the formal power series in these variables defined as @xmath31 the first sum is taken over @xmath32 for @xmath33 , @xmath34 for @xmath35 , and @xmath36 for @xmath37 . on the right hand side , we contract each copy of @xmath4 with the factor of the tensor product associated to the same marked point . the formal power series @xmath38 is called gromov - witten potential associated to the system of gromov - witten invariants @xmath12 . the coefficients of @xmath30 , @xmath39 , are called correlators and denoted by @xmath40 vectors @xmath41 are called primary fields . the main properties of gw potentials come from geometry of the moduli space of curves . first , one can prove that coefficients of @xmath42 satisfy string and dilaton equations : @xmath43 the string equation is a corollary of the fact that @xmath44 ; here @xmath15 is the projection forgetting the last marked point and @xmath45 is the divisor in @xmath46 whose generic point is represented by a two - component curve with one node such that one component has genus @xmath47 and contains exactly two marked points , the @xmath48-th and the @xmath49-th ones . it is assumed that @xmath50 . the dilaton equation is a corollary of the fact that , in the same notations , @xmath51 . of course , we assume that @xmath52 . second , any relation in the cohomology of @xmath9 among natural @xmath53-@xmath54-strata gives a relation for the correlators . let us explain this in more detail . the moduli space of curves @xmath9 @xcite has a natural stratification by the topological type of stable curves . we can combine natural strata with @xmath53-classes at marked points and at nodes and @xmath54-classes on the moduli spaces of irreducible components . these objects are called @xmath53-@xmath54-strata . a convenient way to describe a @xmath53-@xmath54-stratum in @xmath9 is the language of stable dual graphs . take a generic curve in the stratum . to each irreducible component we associate a vertex marked by its genus . to each node we associate an edge connecting the corresponding vertices ( or a loop , if it is a double point of an irreducible curve ) . if there is a marked point on a component , then we add a leaf at the corresponding vertex , and we label leaves in the same way as marked points . if we multiply a stratum by some @xmath53-classes , then we just mark the corresponding leaves or half - edges ( in the case when we add @xmath53-classes at nodes ) by the corresponding powers of @xmath53 . also we mark each vertex by the @xmath54-class associated to it . let us remark that by @xmath54-classes on @xmath9 we mean the classes @xmath55 where @xmath56 is the projections forgetting the last @xmath57 marked points . it is just another additive basis in the ring generated by the ordinary @xmath54-classes ( @xmath58 , @xmath59 , in our notations ) . the basic properties of these classes are stated in @xcite . using the properties of gw invariants , one can express the integral of @xmath10 over a @xmath53-@xmath54-stratum @xmath60 in terms of correlators . consider a special case , when @xmath60 is represented by a two - vertex graph with no @xmath53- and @xmath54-classes . then , according to axiom [ factor ] , the integral of @xmath10 is the product of integrals of @xmath61 and @xmath62 over the moduli spaces corresponding to the vertices , contracted by the scalar product : @xmath63 here we assume that the genus of one component of a generic curve in @xmath60 is @xmath64 and @xmath65 marked points with labels @xmath66 , @xmath67 , are on this component . the other component has genus @xmath68 and @xmath69 marked points with labels @xmath70 , @xmath71 , lie on it . of course , @xmath72 , @xmath73 . now consider a special case , when @xmath60 is represented by a one - vertex graph with @xmath53- and @xmath54-classes . let us assign a vector in the basis of @xmath4 to each leaf ( to each marked point ) . then , according to axiom [ pull - back ] the integral @xmath74 is equal to @xmath75 combining these two special cases one can obtain an expression in correlators that corresponds to an arbitrary @xmath53-@xmath54-stratum . suppose that we have a linear combination @xmath76 of @xmath53-@xmath54-strata that is equal to @xmath47 in the cohomology of @xmath9 ( a tautological relation ) . since @xmath14 , the integral of @xmath77 over @xmath76 is equal to zero , for an arbitrary choice of primary fields . this gives an equation for correlators . usually , one consider also the pull - backs of @xmath76 to @xmath78 , @xmath79 , multiplied by arbitrary monomials of @xmath53-classes . of course , they are also represented as vanishing linear combinations of @xmath53-@xmath54-strata . this gives a system of pdes for the formal power series @xmath30 , @xmath39 . for the detailed description of the correspondence between tautological relations and universal pdes for gw potentials see , e. g. , @xcite or @xcite . there are @xmath80 basic tautological relations known at the moment : wdvv , getzler , belorousski - pandharipande , and topological recursion relations in @xmath81 , @xmath82 , @xmath83 , @xmath84 , @xmath85 @xcite . in gromov - witten theory ( and also in topological string theory ) the gromov - witten invariants is usually a structure on the cohomology of a target manifold ( the space @xmath4 ) of on the cohomology of a complex of some other gometric origin . we have introduced the notion of zwiebach invariants in @xcite in order to formalize in a convenient way what physicists mean by topological conformal quantum field theory at the level of a complex rather than at the level of the cohomology . the very general principles of homological algebra imply that algebraic stuctures on the cohomology are often induced by some fundamental structures on a full complex ( the standard example is the induction of the infinity - structures from differential graded algebraic structures ) . such induction usually can be represented as a sum over trees with vertices corresponding to fundamental operations and edges corresponding to the homotopy that contracts the complex to its cohomology . we would like to stress that gromov - witten invariants also can be considered as an induced structure on the cohomology of a complex . in this case , the fundamental structure on the whole complex is determined by zwiebach invariants . we are able to associate some structure on a bicomplex with a special compactification of the moduli space curves ( kimura - stasheff - voronov compactification ) . so , complexes are replaced by bicomplexes , where the second differential reflects the rotation of attached cylinders ( or circles ) . this is an appearance of the string nature of the problem . as an induced structure we indeed obtain a gromov - witten - type theory that , under some additional assumptions , can be presented in terms of a sum over graphs . below we explain the whole construction , following @xcite and with some additional details . we remind the construction of the kimura - stasheff - voronov compactification @xmath86 of the moduli space of curves of genus @xmath0 with @xmath87 marked point . it is a real blow - up of @xmath9 ; we just remember the relative angles at double points . we can also choose an angle of the tangent vector at each marked point ; this way we get the principal @xmath88-bundle over @xmath86 . we denote the total space of this bundle by @xmath89 . there are also the standard mappings between different spaces @xmath89 . first , one can consider the projection @xmath90 forgetting the last marked point . suppose that under the projection we have to contract a sphere that contains the points @xmath91 , @xmath92 , and a node . denote the natural coordinates on the circles corresponding to @xmath91 and a node on a curve in @xmath93 by @xmath94 and @xmath95 . let @xmath96 be a coordinate on the circle corresponding to @xmath91 in @xmath89 . then @xmath97 under the projection @xmath98 . in the same way , if we contract a sphere that contains two nodes and @xmath92 , then @xmath99 , where @xmath100 and @xmath101 are the coordinates on the circles corresponding to the two nodes of a curve in @xmath93 and @xmath102 is a coordinate on the circle at the resulting node in @xmath89 in the same way , when we consider the mappings @xmath103 representing the natural boundary components of @xmath89 , we have @xmath104 , where @xmath105 and @xmath106 are the coordinates on the circles corresponding the points that are glued by @xmath107 into the node and @xmath95 is the coordinate on the circle at the node . for the mapping @xmath108 we also have @xmath109 with the same notations . let us fix a hodge bicomplex @xmath110 with two differentials denoted by @xmath111 and @xmath112 and with an even scalar product @xmath113 invariant under the differentials : @xmath114 the hodge property means that @xmath115 where @xmath116 and @xmath4 is orthogonal to @xmath117 . below we consider the action of @xmath111 and @xmath118 on @xmath119 . we denote by @xmath120 and @xmath121 the action of @xmath111 and @xmath118 respectively on the @xmath122-th component of the tensor product . on each @xmath89 we take a differential form @xmath123 of the mixed degree with values in @xmath11 . the whole system of forms @xmath124 is called _ zwiebach invariants _ , if it satisfies the axioms : 1 . @xmath10 is @xmath13-equivariant . 2 . @xmath125 , @xmath126 . @xmath127 for all @xmath128 ( we denote by @xmath129 the substitution of the vector field generating the action on @xmath89 of the @xmath122-th copy of @xmath130 ) ; @xmath123 is invariant under the action of @xmath88 ; 4 . @xmath131 , where @xmath90 is the mapping forgetting the last marked point . @xmath132 , where @xmath133 represents the boundary component . in the same way , @xmath134 for the mapping @xmath135 @xmath136 , @xmath137 and @xmath138 are the coordinates on the circles at the corresponding points . zwiebach invariants on the bicomplex with zero differentials determine gromov - witten invariants . indeed , in this case the factorization property implies that @xmath124 is lifted from the blowdown of kimura - stasheff - voronov spaces , i.e. it is determined by a set of forms on deligne - mumford spaces . then it is easy to check that this system of forms satisfied all axioms of gromov - witten invariants . induced zwiebach invariants are obtained by the contraction of @xmath117 . we denote by @xmath139 the contraction operator . this means that @xmath140 , @xmath141 is the projection to @xmath117 along @xmath4 , @xmath142 , and @xmath143 . we construct an induced zwiebach form @xmath144 on a homotopy equivalent modification @xmath145 of the space @xmath89 . at each boundary component @xmath146 we glue the cylinder @xmath147 $ ] such that @xmath146 in @xmath89 is identified with @xmath148 in the cylinder . so , we have the mappings @xmath149\to\tilde{\mathcal{s}}_{g , n}$ ] and @xmath150\to\tilde{\mathcal{s}}_{g , n}$ ] representing the boundary components with glued cylinders . we take a form @xmath123 , restrict it to @xmath11 , and extend it to the glued cylinder by the rule that @xmath151 \right)\ ] ] in the first case and @xmath152 \right)\ ] ] in the second case , where @xmath153 $ ] is the bivector obtained from the operator @xmath154 , @xmath155 is the coordinate on @xmath156 $ ] . this determines @xmath144 completely . now it is a straightforward calculation to check that the forms @xmath144 are @xmath157-closed and satisfy the factorization property when restricted to the strata @xmath158 . the induced zwiebach invariants determine gromov - witten invariants . the correlators of the corresponding gromov - witten potential are given by the integrals over the fundamental cycles of @xmath159 ( we just forget the circles at marked points in @xmath145 ) of the forms @xmath160 . in fact , the fundamental class of @xmath159 is represented as a sum over all irreducible boundary strata in @xmath9 . indeed , a boundary stratum @xmath146 in @xmath9 has real codimension equal to the doubled number of the nodes of its generic curve . but then we add in @xmath159 a real two - dimensional cylinder for each node . a simple explicit calculation allows to express the integral over the component of the fundamental cycle of @xmath159 corresponding to @xmath146 . it splits into the integrals of the initial zwiebach invariants ( multiplied by @xmath53-classes ) over the moduli spaces corresponding to the irreducible components of curves in @xmath146 ; they are contracted with the bivectors @xmath161 $ ] ( obtained from the operator @xmath162 via the scalar product ) corresponding to the nodes according to the topology of curves in @xmath146 . so , we represent the correlators of the induced gromov - witten theory as sums over graphs . then one can observe that @xmath163 determines a multiplication on @xmath110 . topology of the spaces @xmath164 and @xmath165 implies that the whole algebraic structure that we obtain on @xmath110 is the structure of cyclic hodge algebra up to @xmath111-homotopy , see @xcite . let us assume that the initial system of zwiebach invariants is simple enough , i. e. , it induces the explicit structure of cyclic hodge algebra on @xmath110 and only the integrals of the zero - degree parts of the initial zwiebach invariants ( multiplied by @xmath53-classes ) are non - vanishing on fundamental cycles . in this case , the induced gromov - witten potential can be described in very simple algebraic terms . it is the motivation of the definition of the hodge field theory construction given in the next section . in this section , we describe in a very formal algebraic way the sum over graphs obtained as an expression for the gromov - witten potential induced from zwiebach invariants in the previous section . in this section , we recall the definition of cyclic hodge dgbv - algebras @xcite ( cyclic hodge algebras , for short ) . a supercommutative associative @xmath5-algebra @xmath110 with unit is called cyclic hodge algebra , if there are two odd linear operators @xmath166 and an even linear function @xmath167 called integral . they must satisfy the following axioms : 1 . @xmath168 is a bicomplex : @xmath169 2 . @xmath170 , where @xmath116 and @xmath117 is represented as a direct sum of subspaces of dimension @xmath171 generated by @xmath172 for some vectors @xmath173 , i. e. @xmath174 ( hodge decomposition ) ; 3 . @xmath111 is an operator of the first order , it satisfies the leibniz rule : @xmath175 ( here and below we denote by @xmath176 the parity of @xmath177 ) ; 4 . @xmath118 is an operator of the second order , it satisfies the @xmath178-term relation : @xmath179 5 . @xmath118 satisfies the property called @xmath3-axiom : @xmath180 ( here @xmath181 and @xmath182 are the operators of multiplication by @xmath183 and @xmath184 respectively , @xmath185 means supertrace ) . define an operator @xmath186 related to the particular choice of hodge decomposition . we put @xmath140 , and on each subspace @xmath187 we define @xmath139 as @xmath188 we see that @xmath189=0 $ ] ; @xmath190 $ ] is the projection to @xmath117 along @xmath4 ; @xmath191 is the projection to @xmath4 along @xmath117 . consider the integral @xmath167 . we require that @xmath192 these properties imply that @xmath193 , @xmath194 , and @xmath195 . we can define a scalar product on @xmath110 as @xmath196 we suppose that this scalar product is non - degenerate . using the scalar product we may turn any operator @xmath197 into the bivector that we denote by @xmath198 $ ] . here we explain a way to encode some tensor expressions over an arbitrary vector space in terms of graphs . consider an arbitrary graph ( we allow graphs to have leaves and we require vertices to be at least of degree @xmath199 , the definition of graph that we use can be found in @xcite ) . we associate a symmetric @xmath87-form to each internal vertex of degree @xmath87 , a symmetric bivector to each egde , and a vector to each leaf . then we can substitute the tensor product of all vectors in leaves and bivectors in edges into the product of @xmath87-forms in vertices , distributing the components of tensors in the same way as the corresponding edges and leaves are attached to vertices in the graph . this way we get a number . let us study an example : @xmath200 we assign a @xmath201-form @xmath202 to the left vertex of this graph and a @xmath199-form @xmath203 to the right vertex . then the number that we get from this graph is @xmath204 . note that vectors , bivectors and @xmath87-forms used in this construction can depend on some variables . then what we get is not a number , but a function . consider a cyclic hodge algebra @xmath110 . there are some standard tensors over @xmath110 , which we associate to elements of graphs below . here we introduce the notations for these tensors . we always assign the form @xmath205 to a vertex of degree @xmath87 . there is a collection of bivectors that will be assigned below to edges : @xmath161 $ ] , @xmath206 $ ] , @xmath207 $ ] , @xmath208 $ ] , @xmath209 $ ] , @xmath210 $ ] , and @xmath211 $ ] . in pictures , edges with these bivectors will be denoted by @xmath212 respectively . note that an empty edge corresponding to the bivector @xmath207 $ ] can usually be contracted ( if it is not a loop ) . the vectors that we will put at leaves depend on some variables . let @xmath213 be a homogeneous basis of @xmath4 . in particular , we assume that @xmath8 is the unit of @xmath110 . to each vector @xmath27 we associate formal variables @xmath28 , @xmath36 , of the same parity as @xmath27 . then we will put at a leaf one of the vectors @xmath214 , @xmath36 , and we will mark such leaf by the number @xmath87 . in our picture , an empty leaf is the same as the leaf marked by @xmath47 . there is a subtlety related to the fact that @xmath110 is a @xmath215-graded space . in order to give an honest definition we must do the following . suppose we consider a graph of genus @xmath0 . we can choose @xmath0 edges in such a way that the graph being cut at these edge turns into a tree . to each of these edges we have already assigned a bivector @xmath198 $ ] for some operator @xmath216 . now we have to put the bivector @xmath217 $ ] instead of the bivector @xmath198 $ ] , where @xmath218 is an operator defined by the formula @xmath219 . in particular , consider the following graph ( this is also an example to the notations given above ) : @xmath220 an empty loop corresponds to the bivector @xmath207 $ ] . an empty leaf corresponds to the vector @xmath221 . a trivalent vertex corresponds to the @xmath199-form given by the formula @xmath222 . if we ignore this remark , then what we get is just the trace of the operator @xmath223 . but using this remark we get the supertrace of this operator . in fact , this subtlety will play no role in this paper . it affects only some signs in calculations and all these signs will be hidden in lemmas shared from @xcite . so , one can just ignore this remark . we are going to define the potential using correlators . let @xmath224 be the sum over graphs of genus @xmath0 with @xmath87 leaves marked by @xmath225 , @xmath226 , where @xmath227 are vectors in @xmath110 , and @xmath228 are just formal symbols . the index of each internal vertex of these graphs is @xmath229 ; we associate to it the symmetric form . there are two possible types of edges : edges marked by @xmath161 $ ] ( thick black dots in pictures , `` heavy edges '' in the text ) and edges marked by @xmath207 $ ] ( empty edges ) . since an empty edge connecting two different vertices can be contracted , we assume that all empty edges are loops . consider a vertex of such graph . let us describe all possible half - edges adjusted to this vertex . there are @xmath230 , @xmath39 , half - edges coming from @xmath0 empty loops ; @xmath231 half - edges coming from heavy edges of graph , and @xmath57 leaves marked @xmath232 . then we say that the type of this vertex is @xmath233 . we denote the type of a vertex @xmath234 by @xmath235 . consider a graph @xmath236 in the sum determining the correlator @xmath224 we associate to @xmath236 a number : we contract according to the graph structure all tensors corresponding to its vertices , edges , and leaves ( for leaves , we take vectors @xmath227 ) . let us denote this number by @xmath237 . also we weight each graph by a coefficient which is the product of two combinatorial constants . the first factor is equal to @xmath238 here @xmath239 is the order of the automorphism group of the labeled graph @xmath236 , @xmath240 is the set of internal vertices of @xmath236 . in other words , we can label each vertex @xmath234 by @xmath241 , delete all empty loops , and then we get a graph with the order of the automorphism group equal to @xmath242 . the second factor is equal to @xmath243 the integrals used in this formula can be calculated with the help of the witten - kontsevich theorem @xcite . so , the whole contribution of the graph @xmath236 to the correlator is equal to @xmath244 . one can check that the non - trivial contribution to the correlator @xmath245 is given only by graphs that have exactly @xmath246 heavy edges . the geometric meaning here is very clear . the number @xmath237 comes from the integral of the induced gromov - witten invariants of degree zero , while the coefficient @xmath247 is exactly the combinatorial interpretation of the intersection number of @xmath248 with the stratum whose dual graph is obtained from @xmath236 by the procedure described after the definition of @xmath249 . we fix a cyclic hodge algebra and consider the formal power series @xmath250 defined as @xmath251 abusing notations , we allow to mark the leaves by @xmath252 , @xmath253 , or @xmath183 ; all this variants are possible and denote the same . for example , consider the trivial cyclic hodge algebra : @xmath254 , @xmath255 , @xmath256 . then @xmath257 , and the correlator @xmath258 consists just of one graph with one vertex , @xmath0 empty loops , and @xmath87 leaves marked by @xmath259 . the explicit value of the coefficient of this graph is , by definition , @xmath260 so , in the case of trivial cyclic hodge algebra we obtain exactly the gromov - witten potential of the point ( i. e. , @xmath261 ) that we denote below by @xmath262 . abusing notation , we use the same symbol @xmath263 for the correlators in gw theory and in hodge field theory . we hope that it does not lead to a confusion . for instance , @xmath258 in the trivial example above is the correlator of the trivial hodge field theory , while @xmath264 is the correlator of the trivial gw theory . in this section , we prove that the potential satisfies the same string and dilaton equations as gw potentials . if @xmath265 , we have : @xmath266 consider a graph @xmath236 contributing to the correlator on the left hand side of the string equation . the special leaf that we are going to remove is marked by @xmath267 and is attached to a vertex @xmath234 of genus @xmath268 ( i. e. , with @xmath268 attached light loops ) with @xmath269 more attached leaves labeled by indices in @xmath270 , @xmath271 , and @xmath272 attached half - edges coming from heavy edges and loops . let us remove the leaf @xmath267 and change the label of one of the leaves attached to the same vertex from @xmath273 to @xmath274 . this way we obtain a graph @xmath275 contributing to the @xmath276-th summand of the right hand side of . we take the sum of these graphs over @xmath277 . of course , we skip the summands where @xmath278 . note that this sum is not empty ( if @xmath236 gives a non - zero contribution to the left hand side of ) . indeed , if it is empty , this means that @xmath278 for all @xmath277 . therefore , since we expect that the contribution to @xmath279 of the vertex @xmath234 on the left hand side is nonzero , it follows that @xmath280 and @xmath281 . so , there are three possible local pictures : @xmath282 the first picture can be replaced with the bivector @xmath283 $ ] , which is equal to zero . therefore , @xmath237 is also equal to zero . in the second case , we also get @xmath47 since @xmath284 . the third picture is possible only when it is the whole graph , and this is in contradiction with the assumption that @xmath265 . note also that @xmath285 and @xmath286 for all @xmath276 . indeed , we have just removed the leaf with the unit of the algebra , so this ca nt change anything in the contraction of tensors . therefore , @xmath287 . also both the leaf @xmath267 and the vertex @xmath234 are the fixed points of any automorphism of @xmath236 . the same is for the vertex corresponding to @xmath234 in @xmath275 . therefore , the automorphism groups are isomorphic for both graphs . since we make no changes for empty loops , it follows that @xmath286 . let us prove that @xmath288 . indeed , the vertices of @xmath236 and @xmath275 are in a natural one - to - one correspondence . moreover , the local pictures for all of them except for @xmath234 and its image in @xmath275 are the same . therefore , the corresponding intersection numbers contributing to @xmath279 and @xmath289 are the same . the unique difference appeares when we take the intersection numbers corresponding to @xmath234 and its images in @xmath275 , @xmath277 . but then we can apply the string equation of the gw theory of the point , and we see that @xmath290 ( here @xmath291 is an arbitrary on - to - one mapping ) . this implies that @xmath288 . so , we have @xmath292 . in order to complete the proof of , we should just notice that when we write down this expression for all graphs contributing to the left hand side of , we use each graph contributing to the right hand side of exactly once . if @xmath293 , we have : @xmath294 consider a graph @xmath236 contributing to the correlator on the left hand side of . the special leaf that we are going to remove is marked by @xmath295 and is attached to a vertex @xmath234 of genus @xmath268 ( i. e. , with @xmath268 attached light loops ) with @xmath269 more attached leaves labeled by indices in @xmath270 , @xmath271 , and @xmath272 attached half - edges coming from heavy edges and loops . let us remove the leaf @xmath295 . we obtain a graph @xmath296 contributing to the right hand side of . let us prove this . indeed , if we remove a leaf and do nt get a proper graph , it follows that we have a trivalent vertex . since the contribution of this vertex to @xmath279 should be non - zero , it follows that the unique possible local picture is @xmath297 but this picture is the whole graph , and it is in contradiction with the condition @xmath293 . the same argument as in the proof of the string equation shows that @xmath298 and @xmath299 . also , the contribution to @xmath279 and @xmath300 of all vertices except for the changed one is the same . the change of the intersection number corresponding to the vertex @xmath234 is captured by the dilaton equation of the trivial gw theory : @xmath301 ( again , @xmath291 is an arbitrary on - to - one mapping ) . this implies that @xmath302 , and , therefore , @xmath303 let us write down the last equation for all graphs @xmath236 contributing to the left hand side . observe that any graph @xmath296 contributing to the right hand side occurs @xmath304 times , since the leaf @xmath295 could be attached to any its vertex . therefore , any graph @xmath296 contributing to the right hand side of appears in these equations with the coefficient @xmath305 this completes the proof . as we have explained in section [ rel - for - corr ] , any linear relation @xmath76 among @xmath53-@xmath54-strata in the cohomology of the moduli space of curves gives rise to a family of universal relations for the correlators of a gromov - witten theory . [ thm - tautolog ] the system of universal relations coming from a tautological relation in the cohomology of the moduli space of curves holds for the correlators @xmath306 of cyclic hodge algebra . note that some special cases of this theorem were proved in @xcite . our argument below is a natural generalization of the technique introduced in these papers . also we are able now to give an explanation why we have managed to perform all our calculations there , see remark [ last - remark ] . let us give here a brief account of the proof of this theorem . first , the definition of correlators of the hodge field theory can be extended to the intersection with an arbitrary tautological class @xmath307 of degree @xmath308 in the space @xmath9 , not only the a monomial in @xmath53-classes . in that case , we do the following . again , we consider the sum over all graphs @xmath236 with @xmath309 heavy edges , and the number @xmath237 is defined as above . instead of the coefficient @xmath247 we use the intersection number of @xmath307 with the stratum whose dual graph is obtained from @xmath236 by the procedure described right after the definition of @xmath249 . namely , a vertex with @xmath0 loops is replaced by a vertex marked by @xmath0 . this definition is very natural from the point of view of zwiebach invariants . however , we know from gromov - witten theory that this extension of the notion of correlator is unnecessary . indeed , all integrals with arbitrary tautological classes can be expressed in terms of the integrals with only @xmath53-classes via some universal formulas . the main question is whether these universal formulas also work in hodge field theory . actually , the main result that more or less immediately proves the theorem is the positive answer to this question . the rest of the paper is devoted to the proof of main theorem , and here we would like to overview it here . in section 6 , we study the structure of graphs that can appear in formulas for the correlators of hodge field theory . we prove that if @xmath310 and there is at least one heavy edge in @xmath236 , then all vertices have genus @xmath311 , i. e. , there is at most one empty loop at any vertex . this basically means that in calculations we ll have to deal only with genera @xmath47 and @xmath312 . also this allows us to write down the action of a hodge field theory . in section 7 , we prove the main technical result ( main lemma ) . informally , it states that @xmath313 when we apply these two operators to the correlators of a hodge field theory . in order to prove it , we look at a small piece ( consisting just of one heavy edge and one or two vertices that are attached to it ) in one of the graphs of a correlator . of course , in the correlator we can vary this small piece in an arbitrary way , such that the rest of the graph remains the same . so , when we consider the sum of all these small pieces , it is also a correlator of the hodge field theory . thus we reduce the proof to a special case of the whole statement . but since the genus of a vertex is @xmath311 , it appear now to be a low - genera statement that can be done by a straightforward calculation . in section 8 , we present the proof of the main theorem . consider a @xmath53-@xmath54-stratum @xmath307 whose stable dual graph has @xmath59 edges . there is a universal expression of the integral over @xmath307 coming from gromov - witten theory . it includes @xmath122 entries of the scalar product restricted to @xmath4 . in terms of graphs , it means that we are to introduce new edges with the bivector @xmath206 $ ] on them , and there are @xmath122 such edges in our expression . a direct corollary of the main lemma is that we can always replace @xmath206 $ ] by @xmath207-\psi [ g_-g_+]-[g_-g_+]\psi$ ] . in sections 8.2 , 8.3 , and 8.4 , we show that when we replace @xmath206 $ ] by @xmath207-\psi [ g_-g_+]-[g_-g_+]\psi$ ] at all edges corresponding to the scalar product restricted to @xmath4 , we obtain a new expression for the integral over @xmath307 that again contains only heavy edges and empty loops , as any ordinary correlator . the main problem now is to understand the combinatorial coefficient of a graph @xmath236 obtained this way . since we have a sum over graphs with heavy edges and empty loops , it is natural to identify again these graphs with the corresponding strata in the moduli space of curves . then we can calculate the intersection index of the stratum corresponding to a graph @xmath236 and the initial class @xmath307 . roughly speaking , the main thing that we have to do is to decide about each node in @xmath307 ( represented initially by @xmath206 $ ] ) , whether we have this node in stratum corresponding to @xmath236 . if yes , then we have an excessive intersection ( so , we must put @xmath314 on one of the half - edges of the corresponding edge ) , and we keep this edge in @xmath236 ( so , we replace @xmath206 $ ] with @xmath315 $ ] or @xmath316\psi$ ] ) . if no , then we do nt have this edge in @xmath236 , so we contract @xmath206 $ ] , i. e. , replace it with @xmath207 $ ] . so , the procedure that we used to get rid of the scalar product is the same as the procedure of the intersection of @xmath307 with strata of the complementary dimension . this means ( section 8.5 ) that the universal formula coming from gromov - witten theory is equivalent to the natural formula for the `` correlator with @xmath307 '' coming from zwiebach theory . the tautological relation is a sum of classes equal to zero . so , while the universal formula coming from gromov - witten theory gives ( in the case of a vanishing class ) a non - trivial expression in correlators , the natural formula coming from zwiebach theory gives identically zero . this proves our theorem , see section 8.6 . in this section , we recall several useful lemmas shared in @xcite . in particular , these lemmas give some strong restrictions on graphs that can give a non - zero contirbution to the correlators defined above . [ tillmann ] @xcite the following vectors and bivectors are equal to zero : @xmath317 also let us remind another lemma in @xcite that is very useful in calculations . [ g - jump ] @xcite for any vectors @xmath318 , @xmath319 , @xmath320 both lemmas are just simple corollaries of the axioms of cyclic hodge algebra . consider a graph studied in section [ subsec - correlators ] . it can have leaves , empty and heavy loops , and heavy edges . consider a vertex of such graph . let us assume that there are @xmath321 empty loops , @xmath322 heavy loops , @xmath323 heavy edges going to the other vertices of the graph , and @xmath324 leaves attached to this vertex : @xmath325 this picture can be considered as an @xmath326 form . let us denote it by @xmath327 . [ lemma - emptyedges ] if @xmath328 and @xmath329 , then @xmath330 . in other words , if there are at least two empty loops at a vertex , then there should not be any heavy loops or edges attached to this vertex . otherwise the contribution of the whole graph vanishes . this implies in the definition of correlators one should consider only graphs of one of the following two types : 1 . one - vertex graphs with no heavy edges ( loops ) . 2 . arbitrary graphs with at most one empty loop at each vertex . the contribution of all other graphs vanishes . this corollary dramatically simplifies all our calculations with graphs given below . also we can write down now the action of the hodge field theory . let @xmath331 , @xmath332 $ ] be the `` dimension zero '' part of the potential of the hodge field theory , namely , @xmath333 the first sum is taken over @xmath36 such that @xmath293 . so , it is exactly the generating function for the vertices of our graph expressions . then the action of the hodge field theory is equal to @xmath334 if we put @xmath335 for @xmath34 , then we immediately obtain the bcov - type action discussed in ( * ? ? ? * appendix ) and ( * ? ? ? * appendix ) . the similar actions were also studied in @xcite and @xcite . we consider the form @xmath327 and we assume that @xmath328 . first , let us study the case when @xmath336 . in this case , our @xmath326-form can be represented as a contraction via the bivector @xmath207 $ ] of two forms , @xmath337 and @xmath338 . let us prove that the last one is equal to zero . indeed , this two - form can be represented as @xmath339 , where the two - form @xmath340 is represented by the picture @xmath341 according to lemma [ tillmann ] , @xmath342 . therefore , @xmath343 and the whole form @xmath327 is also equal to zero . now consider the case when @xmath344 . in this case , our @xmath326-form can be represented as a contraction via the bivector @xmath207 $ ] of two forms , @xmath345 and @xmath346 . let us prove that the last one is equal to zero . indeed , @xmath347 here the first equality is definition of @xmath346 , the second one is just an equivalent redrawing , the third equality is application of lemma [ g - jump ] , the fourth one is again an equivalent redrawing . the last picture contains the bivector which is equal to zero according to lemma [ tillmann ] . therefore , the whole picture is equal to zero , and @xmath348 . so , the whole form @xmath327 is equal to zero also in this case . this proves the lemma . the main technical tool that we use in the proof of theorem [ thm - tautolog ] is the lemma that we prove in this section . [ main - lemma ] for any @xmath349 , @xmath350 , @xmath351 a simple corollary of this lemma is the following : [ cor - main - l ] for any @xmath352 , @xmath353 , @xmath354 in other words , we can state informally that @xmath355 . the proof of the lemma can be reduced to a small number of special cases . we consider correlators whose graphs have only one heavy edge . [ [ section ] ] the first case is the following : let @xmath356 . we prove that for any @xmath349 , @xmath357 first , we see that according to the definition of the correlator , the left hand side of equation is the sum over graphs with two vertices and with @xmath161 $ ] on the unique edge that connects the vertices . for each @xmath358 we can consider the corresponding distribution of leaves between the vertices ( to be precise , let us assume that @xmath359 ) . then the coefficient of such graph is @xmath360 , and we take the sum over all possible positions of @xmath111 at the leaves . using the leibniz rule for @xmath111 and the property that @xmath361=-g_-$ ] , we see that this sum is equal to the sum over graphs with two vertices and with @xmath362 $ ] on the unique edge that connects the vertices . for each @xmath358 , @xmath363 , we consider the corresponding distribution of leaves between the vertices . then the coefficient of such graph is still @xmath360 , and the underlying tensor expression can be written ( after we multiply the whole sum by @xmath364 ) as @xmath365 let us recall that the @xmath178-term relation for @xmath118 implies that @xmath366 using this , we can rewrite the whole sum over graphs as @xmath367\\ -\sum_{i\not=1}\left[\left(v_1,\prod_{j\not=1,i}v_j\cdot g_-(v_i)\right)\cdot\right . \\ \left.\sum_{i'\sqcup j ' \sqcup\{i\}=\{2,\dots , n\ } } \left(|j'|-1\right)\langle \tau_{a_1 } \tau_0 \prod_{j\in i ' } \tau_{a_j } \rangle_0 \langle \tau_{a_i } \tau_0 \prod_{j\in j ' } \tau_{a_j } \rangle_0\right].\end{gathered}\ ] ] using that @xmath368 equation , and the fact that @xmath369 we can rewrite expression as @xmath370 the last formula coincides by definition with the right hand side of equation multiplied by @xmath364 . this proves the first special case . [ [ section-1 ] ] the second case is in genus @xmath312 . let @xmath371 . we prove that for any @xmath349 , @xmath372 according to the definition of the correlator , the left hand side of equation is the sum over graphs of two possible types . the first type include graphs with two vertices and two edges . the first edge is heavy and connects the vertices ; the second edge is an empty loop attached to the first vertex . for each @xmath358 , @xmath373 , we can consider the corresponding distribution of leaves between the vertices ( we assume that leaves with indices in @xmath374 are at the first edge ) . then the coefficient of such graph is @xmath375 . the second type include graphs with one vertex and one heavy loop . all leaves are attached to this vertex , and the coefficient of such graph is @xmath376 . for both types of graphs , we take the sum over all possible positions of @xmath111 at the leaves . using the leibniz rule for @xmath111 and the property that @xmath361=-g_-$ ] , we get the same graphs as before , but there is no @xmath111 , and instead of @xmath161 $ ] we have @xmath377 $ ] on the corresponding edge . using lemma [ g - jump ] , we move @xmath378 in graphs of the first type to the leaves marked by indices in @xmath218 . using the @xmath3-axiom and lemma [ g - jump ] , we move @xmath378 in graphs of the second type to all leaves . this way we get graphs of the same type in both cases . we get graphs with one vertex , one empty loop attached to it , all leaves are also attached to this vertex , and there is @xmath378 on one of the leaves . one can easily check that the coefficient of the graphs with @xmath378 at the @xmath48-th leaf is equal to @xmath379 it is exactly the unique graph contributing to the @xmath48-th summand of the right hand side of equation , and the coefficient is right . this proves that special case . [ [ section-2 ] ] consider @xmath37 . let @xmath380 . in this case , the statement that for any @xmath349 , @xmath381 is immediately reduced to @xmath382 ; it is a simple corollary of lemma [ tillmann ] . consider the left hand side of equation . as usual , using the leibniz rule for @xmath111 and the property that @xmath361=-g_-$ ] , we can remove all @xmath111 , but then we must change one of @xmath161 $ ] on edges to @xmath377 $ ] . let us cut out the peaces of graphs that includes this edges with @xmath377 $ ] , all empty loops , leaves and halves of heavy edges attached to the ends of this special edge . since we consider the sum over all possible graphs contributing to correlators , these small pieces can be gathered into groups according to the type of the rest of the initial graph . each group forms exactly one of the special cases studied above . so , we know that @xmath378 should jump either to one of the leaves or to one of the heavy edges attached to the ends of its edge . in the first case , we get exactly the graphs in the right hand side of equation ; in the second case , we get zero . one can easily check that we get the right coefficients for the graphs in the right hand side of equation . this proves the lemma . consider the expression in correlators corresponding to a @xmath53-@xmath54-stratum as it is described in section [ integrals - strata ] . to each vertex of the corresponding stable dual graph we assign the sum of graphs that forms correlator in the sense of section [ subsec - correlators ] . the leaves of these graphs corresponding to the edges of the stable dual graph ( nodes ) are connected in these pictures by edges with @xmath206 $ ] ( the restriction of the scalar product to @xmath4 ) . we call the edges with @xmath206 $ ] `` white edges '' and mark them in pictures by thick white points , see . the axioms of cyclic hodge algebra imply a system of linear equations for the graphs of this type . in particular , it has appeared that playing with this linear equations we can always get rid of white edges in the sum of pictures corresponding to a stable dual graph , see @xcite . however , previously it was just an experimental fact . now we can show how it works in general . the numerous examples of the correspondence between stable dual graphs and graphs expressions in cyclic hodge algebras and also of the linear relations implied by the axioms of cyclic hodge algebra are given in @xcite . below , we explain how one can represent the expression in correlators corresponding to a @xmath53-@xmath54-stratum in terms of graphs with only empty and heavy edges and with no white edges . the unique tool that we need is lemmas [ main - lemma ] and [ cor - main - l ] proved above . consider a stable dual graph with two vertices and one edge connecting them : @xmath383 the corresponding expression in correlators is @xmath384 ( here we denote by @xmath385 an arbitrary choice of @xmath386 ) . it is convenient for us to rewrite this expression as @xmath387^{\alpha_1\alpha_2}\cdot \langle\tau_{b_0}(x_{\alpha_2})\prod_{i=1}^{n_2}\tau_{b_i}(e_\bullet ) \prod_{i=1}^{l_2}\tau_{v_i}(e_1)\rangle_{g_1},\end{gathered}\ ] ] where @xmath388 is the basis of the whole @xmath110 . using the fact that @xmath389 and applying lemma [ cor - main - l ] , we obtain @xmath390^{\alpha_1\alpha_2}\cdot \langle\tau_{b_0}(x_{\alpha_2})\prod_{i=1}^{n_2}\tau_{b_i}(e_\bullet ) \prod_{i=1}^{l_2}\tau_{v_i}(e_1)\rangle_{g_1 } \\ = \langle\tau_{a_0}(x_{\alpha_1})\prod_{i=1}^{n_1}\tau_{a_i}(e_\bullet ) \prod_{i=1}^{l_1}\tau_{u_i}(e_1)\rangle_{g_1}\cdot \\ [ id]^{\alpha_1\alpha_2}\cdot \langle\tau_{b_0}(x_{\alpha_2})\prod_{i=1}^{n_2}\tau_{b_i}(e_\bullet ) \prod_{i=1}^{l_2}\tau_{v_i}(e_1)\rangle_{g_1 } \\ - \langle\tau_{a_0 + 1}(x_{\alpha_1})\prod_{i=1}^{n_1}\tau_{a_i}(e_\bullet ) \prod_{i=1}^{l_1}\tau_{u_i}(e_1)\rangle_{g_1}\cdot \\ [ g_-g_+]^{\alpha_1\alpha_2}\cdot \langle\tau_{b_0}(x_{\alpha_2})\prod_{i=1}^{n_2}\tau_{b_i}(e_\bullet ) \prod_{i=1}^{l_2}\tau_{v_i}(e_1)\rangle_{g_1 } \\ - \langle\tau_{a_0}(x_{\alpha_1})\prod_{i=1}^{n_1}\tau_{a_i}(e_\bullet ) \prod_{i=1}^{l_1}\tau_{u_i}(e_1)\rangle_{g_1}\cdot \\ [ g_-g_+]^{\alpha_1\alpha_2}\cdot \langle\tau_{b_0 + 1}(x_{\alpha_2})\prod_{i=1}^{n_2}\tau_{b_i}(e_\bullet ) \prod_{i=1}^{l_2}\tau_{v_i}(e_1)\rangle_{g_1}\end{gathered}\ ] ] in all three summands of the right hand side we still have two correlators , whose leaves corresponding to the nodes are connected by some special edges . but now the connecting edge is either marked by @xmath207 $ ] ( an empty edge ) or by @xmath161 $ ] ( an ordinary heavy edge ) . so , this way we get rid of the white edge in this case . informally , in terms of pictures , we can describe equation as @xmath391 when we put @xmath53 , we mean that we add one more @xmath53-class at the node at the corresponding branch of the curve . dashed circles denote correlators . we perform the same calculation as above , but now we explain it in terms of informal pictures from the very beginning . so , the first step is the same as above : @xmath392 then we apply lemma [ main - lemma ] to each of the summands in the right hand side : @xmath393 @xmath394 and @xmath395 we take the sum of these three expressions , and we see that all pictures where we have edges with @xmath211 $ ] and @xmath210 $ ] are cancelled . so , we get an expression for the sum of graphs representing the initial stratum in terms of graphs with only empty and heavy edges . the general argument is exactly the same as in the second example . in fact , this gives a procedure how to write an expression in graphs with only empty and heavy edges ( and no white edges ) starting from a stable dual graph . let us describe this procedure . take a stable dual graph corresponding to a @xmath53-@xmath54-stratum of dimension @xmath122 in @xmath9 . first , we are to decorate it a little bit . for each edge , we either leave it untouched , or substitute it with an arrow ( in two possible ways ) . at the pointing end of the arrow , we increase the number of @xmath53-classes by @xmath312 . each of these graphs we weight with the inversed order of its automorphism group ( automorphisms must preserve all decorations ) multiplied by @xmath396 , where @xmath397 is the number of arrows . consider a decorated dual graph . to each its vertex we associate the corresponding correlator of cyclic hodge algebra ( we add new leaves in order to represent @xmath54-classes ) . then we connect the leaves corresponding to the nodes either by empty edges ( if the corresponding edge of the decorated graph is untouched ) or by heavy edges ( if the corresponding edge of the dual graph is decorated by an arrow ) . we can simplify the resulting graphs obtained in the previous subsection . first , we can contract empty edges ( as much as it is possible ; it is forbidden to contract loops ) . second , we can remove leaves added for the needs of @xmath54-classes . indeed , each such leaf is equipped with a unit of @xmath110 , so it does nt affect the contraction of tensors corresponding to a graph . moreover , when we remove all leaves corresponding to @xmath54-classes , we still have graph with at least trivalent vertices . otherwise , this graph is equal to zero , c. f. arguments in the proofs of string and dilaton equations . so , we obtain final graphs that have the same number of heavy edges as the dimension of the initial @xmath53-@xmath54-stratum , the same number of leaves as the initial dual graph , and some number of empty loops , at most one at each vertex . the exceptional case is when @xmath398 ; in this case we obtain only one graph , with one vertex , @xmath87 leaves , and @xmath0 empty loops . in the first case , let us turn a graph like this into a stable dual graph . just replace its vertices with no empty loops by vertices of genus zero , vertices with empty loops by vertices of genus one , heavy edges are edges , and leaves are leaves . there are no @xmath53- or @xmath54- classes . it is obvious that the codimension of the stratum corresponding to this dual graph is @xmath122 . indeed , in this case it is just the number of nodes . so , to each @xmath53-@xmath54-stratum @xmath399 of dimension @xmath400 we associate a linear combination @xmath401 of strata of codimension @xmath122 with no @xmath53- or @xmath54-classes , whose curves have irreducible components of genus @xmath47 and @xmath312 only . we prove it two steps . first , consider a one - vertex stable dual graph with no edges ( just a correlator ) . in this case , the intersection number @xmath403 is just by definition @xmath404 , where @xmath405 is the cyclic hodge algebra graph that turns into @xmath406 via the procedure described above . then , consider a stable dual graph with one edge . it is the intersection of the one - vertex stable dual graph with an irreducible component of the boundary . for a given @xmath406 , this component of the boundary either intersects it transversaly , or we have an excessive intersection . in the first case , the corresponding node is not represented in @xmath406 . this means that in @xmath405 it should be an empty edge . in the second case , this node is one of the nodes of @xmath406 , so it should be a heavy edge of @xmath405 . also , it is an excessive intersection , so we are to add the sum of @xmath53-classes with the negative sign at the marked points ( half - edges ) corresponding to the node , see ( * ? ? ? * appendix ) . if @xmath398 , this means that each of the vertices of the initial stable dual graph also has dimension @xmath47 , and the corresponding correlator of cyclic hodge algebra is represented by one one - vertex graph with no heavy edges . also this means that each edge of the initial stable dual graph is replaced in the algorithm above by an empty edge . so , we can think that we just work with the correlators of the gromov - witten theory of the point . in this case proposition becomes obvious . consider the system of subalgebras @xmath408 of the cohomological tautological algebras of @xmath9 generated by strata with no @xmath53- or @xmath54-classes and with irreducible curves of genus @xmath47 and @xmath312 only . let @xmath76 be a linear combination of @xmath53-@xmath54-strata of dimension @xmath122 in @xmath9 . then the expression in correlators of cyclic hodge algebras corresponding to @xmath76 is equaivalent to a sum of some graphs with coefficients equal to the intersection of @xmath76 with classes in @xmath409 . so , if the class of @xmath76 is equal to zero , then the corresponding equation ( and also the whole system of equations that we described in section [ rel - for - corr ] ) for correlators of cyclic hodge algebra is valid . theorem is proved . evidently , @xmath410 is a module over @xmath411 . also it is obvious that @xmath410 is closed under pull - backs and push - forwards via the forgetful morphisms . this explains why it was enough to make only one check in the simplest case in order to get the system of equations in @xcite ( cf . an argument in the last section in @xcite ) . from the point of view of the theory of zwiebach invariants , both propositions look very natural . indeed , we try to give a graph expression for the integral of an induced gromov - witten form multiplied by a tautological class @xmath399 . since we know that we are able to integrate only degree zero parts of induced gromov - witten invariants , we should just take the sum over all graphs that correspond to the strata of complimentary dimension in @xmath410 . the coefficients are to be the intersection numbers of these strata with @xmath399 . on the other hand , we know that in any gromov - witten theory it is enough to fix the integrals of gromov - witten invariants multiplied by @xmath53-classes . then the integrals of gromov - witten invariants multiplied by arbitrary tautological classes are expressed by universal formulas . we can try to use these universal formulas also in hodge field theory . they are exactly our expressions with white edges . so , we have two different natural ways to express in terms of graphs the integrals of induced gromov - witten invariants multiplied by tautological classes . propositions [ prop1 ] and [ prop2 ] state that these two different expressions coinside . a. losev , hodge strings and elements of k. saito s theory of primitive form , topological field theory , primitive forms and related topics ( kyoto , 1996 ) , 305335 , progr . , 160 , birkhaeuser boston , boston , ma , 1998 .

as has been shown recently in refs . @xcite , an antiproton implanted in a heavy nucleus serves as an attractor for surrounding nucleons that can lead to a sizable increase of the central nucleon density . this effect is caused by the strong attractive scalar and vector potentials acting on the antiproton , as follows from the @xmath1-parity transformation of nuclear potentials @xcite . correspondingly , the antiproton also creates an attractive potential acting on nucleons . this leads to the concentration of nucleons around the antiproton and , as result , to a considerable increase of the nucleon density . within the relativistic mean field ( rmf ) model , the @xmath1-parity transformed nuclear optical potential is about @xmath2 mev at the normal nuclear matter density @xmath3 @xmath4 , while a phenomenological value of an antiproton optical potential is limited within the range of @xmath5 mev @xcite . therefore , in order to fit the empirical optical potential , the antiproton coupling constants with @xmath6- , @xmath7- and @xmath8-meson fields should be reduced with respect to the values given by the @xmath1-parity transformation . the rmf calculations with reduced coupling constants @xcite still show quite strong compressional effects for light and medium nuclei . an important question , which arises here is whether the compression process is fast enough to develop before the @xmath9-annihilation . the total @xmath10-annihilation cross section in vacuum can be parameterized at low relative velocities @xmath11 as @xmath12 where @xmath13 mb and @xmath14 mb@xmath15c @xcite . using these numbers we can estimate the life time of an antiproton inside the nuclear matter at normal density : @xmath16 this is , of course , a very short time in nuclear scale . however , as argued in ref . @xcite , this time can become much longer , up to 20 fm / c , for deeply bound antiprotons due to the phase space suppression factors . therefore , the compression effects can , in - principle , show up in @xmath9-nuclear interactions . in the present work , we apply a dynamical transport model in order to study the formation and decay of the compressed @xmath9-nuclear system . our calculations are based on the giessen boltzmann - uehling - uhlenbeck ( gibuu ) model @xcite , which has been recently supplemented by the relativistic mean fields @xcite . apart from collision terms , the gibuu model solves the coupled ( through the mean fields ) vlasov equations for nucleon and antiproton phase space distribution functions . as well known @xcite , the vlasov equation provides a semiclassical limit of the time - dependent hartree - fock calculations . thus , the compressional effects found in refs . @xcite should also be reproduced as a static solution of the coupled vlasov equations . it will be demonstrated that the compression process is characterized by the time scale which is comparable with the @xmath9 life time in nuclear environment . thus , @xmath9 has , indeed , a chance to annihilate inside the compressed nucleus . we will show , that the @xmath9-annihilation in a compressed nucleus should lead to the collective expansion of the residual nuclear system . the appearance of the high - energy tails in the kinetic energy spectra of the emitted nucleons is predicted . the distributions in the total invariant mass of produced mesons reveal a noticeable shift toward lower invariant masses , when the annihilation takes place inside the compressed nucleus . the annihilation of slow antiprotons inside heavy nuclei was , first , proposed by rafelski @xcite as a unique opportunity to study nuclear matter in unusual conditions . later , cahay et al . @xcite studied the @xmath9 annihilation inside nuclei within an intranuclear cascade model . in ref . @xcite , antiproton annihilation events into pions at the center of @xmath17ca and @xmath18ag nuclei were simulated . the mean field effects were , however , completely neglected in @xcite . in sect . ii , we describe the theoretical model applied in calculations . sect . iii contains the results of the time evolution study for the compression and explosion dynamics . in sect . iv , we propose several observable signals sensitive to the @xmath9-annihilation in the compressed nucleus . the summary and outlook are given in sect . in calculations , we apply the gibuu model developed in giessen university . for the detailed description and related references , we refer the reader to the web page @xcite , where the new version of the model is presented . below , we mostly describe the new features implemented in the present work . below we consider a system composed of an antinucleon interacting with baryons . this system is described by the rmf lagrangian of the following form @xcite : @xmath19 \psi_j \nonumber \\ & + & { 1 \over 2}\partial_\mu\sigma\partial^\mu\sigma - u(\sigma ) - { 1 \over 4 } f_{\mu\nu } f^{\mu\nu } + { 1 \over 2 } m_\omega^2 \omega_\mu\omega^\mu~ , \label{lagr}\end{aligned}\ ] ] where @xmath20 are the baryon ( @xmath21 ) and antinucleon ( @xmath22 ) fields , respectively ; @xmath6 is the isoscalar - scalar meson field ( @xmath23 , @xmath24 ) ; @xmath25 is the isoscalar - vector meson field ( @xmath26 , @xmath27 ) ; and @xmath28 . here @xmath29 and @xmath30 denotes , respectively , the isospin 1/2 and 3/2 nonstrange baryonic resonances , and @xmath31 stands for the @xmath32 baryons explicitly propagated in the gibuu model @xcite . in the case of the spin 3/2 , 5/2 and 7/2 baryonic resonances , their fields @xmath20 carry also one or more vector indices , which are dropped in eq.([lagr ] ) and below for brevity . when appropriate , the covariant summation is assumed over these indices . for simplicity , the isovector and electromagnetic terms are disregarded in ( [ lagr ] ) . the selfinteractions of the @xmath6-field are included in ( [ lagr ] ) via the term @xmath33 in order to avoid an unrealistically high compressibility coefficient of the nuclear matter @xcite : @xmath34 some comments are in order to gain more insight into eq . ( [ lagr ] ) . following ref . @xcite , the antinucleon field @xmath35 in the lagrangian density ( [ lagr ] ) is represented in terms of wave functions of _ physical antinucleons_. these wave functions can be obtained by the @xmath1-parity transformation acting on the wave functions of the dirac sea nucleons ( see ref . @xcite for details ) , which appear in the relativistic description of the nucleon @xcite . by applying the same transformation , the nonlinear rmf lagrangian of refs . @xcite ( neglecting terms responsible for the baryon - antibaryon annihilation ) can be expressed as ( [ lagr ] ) with the following relations between coupling constants : @xmath36 the relations ( [ gpar_relation ] ) are satisfied if the physical system would be exactly symmetric with respect to the @xmath1-parity transformation . however , this is not necessary to be true in a many - body system @xcite . the reason is that the concept of the @xmath1-parity symmetry is strictly applicable on the level of the elementary processes only . however , the rmf lagrangian ( [ lagr ] ) is dealing with the effective interactions , which are usually tuned to describe the bulk properties of the nuclear medium and/or the properties of some selected nuclei . due to the many - body effects , such as the pauli blocking or mixed scalar - vector terms in the scattering amplitudes , these effective interactions may not obey the exact @xmath1-parity symmetry anymore . to take into account possible deviations from the @xmath1-parity symmetry , we introduce an overall scaling of the antinucleon - meson coupling constants with respect to the values given by ( [ gpar_relation ] ) ( see ref . @xcite ) : @xmath37 where @xmath38 is a scaling factor . throughout the paper , we consider two options for the scaling factor of the antinucleon - meson coupling constants : @xmath39 , motivated by the @xmath1-parity , and @xmath40 , which is in a better agreement with the empirical @xmath9a optical potential . for other baryonic fields we put in the present work , for simplicity , the same coupling constants as for the nucleon : @xmath41 , @xmath42 . all calculations have been performed emloying the nl3 parameterization @xcite of the rmf model . this parameterization provides quite reasonable nuclear matter properties : the binding energy 16.299 mev / nucleon , the compressibility coefficient @xmath43 mev and the nucleon effective mass @xmath44 at @xmath45 . moreover , the nl3 parameterization reproduces the ground state properties of spherical and deformed nuclei very well @xcite . the dirac equations of motion for baryons have the following form : @xmath46 where @xmath47 is the effective ( dirac ) mass . within the mean field approximation the @xmath6- and @xmath7-fields are treated classically . they satisfy the ( nonlinear ) klein - gordon - like equations with the source terms due to coupling to baryons and an antinucleon : @xmath48 where @xmath49 is the partial scalar density and @xmath50 is the partial baryon current . equation ( [ kgeqomega ] ) has to be supplemented by the four - transversality condition @xmath51 instead of solving the dirac equations ( [ diraceqb ] ) , we will describe the baryons and antinucleon dynamics by the coupled set of the semiclassical kinetic equations @xcite : @xmath52 f_j(x,{\bf p^\star } ) = i_j[f_b , f_m]~ , \label{kineq}\ ] ] where @xmath53 ; @xmath54 ; @xmath55 ; and @xmath56 is the distribution function ( df ) in a six - dimensional phase space @xmath57 with @xmath58 being the spatial components of the kinetic four - momentum @xmath59 the baryons and antinucleon are assumed to be on the respective effective mass shells : @xmath60 the l.h.s . of eq . ( [ kineq ] ) describes the propagation of the @xmath61-th type particles in the classical @xmath6- and @xmath7-fields . the r.h.s . ( [ kineq ] ) is a collision integral , which represents the ( in)elastic two - body collisions with corresponding vacuum cross sections as well as the resonance decays . the complete description of the collision integral structure , in - particular , the differential elementary cross sections included into the gibuu model can be found in @xcite and in refs . therein . the in - medium modification of the baryon - baryon and baryon - meson cross sections is neglected in the present work . we will apply the full kinetic equations , including collision terms , only to describe the _ post - annihilation _ evolution of a system . by this reason , the antiproton df is excluded from the collision integral . instead , we enforce @xmath9 to annihilate into mesons at some preselected time ( see sect . iie ) . thus , in the present work the collision integral includes the nucleon , @xmath62 and higher baryon resonances up to the mass of 2 gev , which can be excited in the meson - baryon and baryon - baryon collisions . a possible hyperon formation in the processes @xmath63 and @xmath64 is included too . the `` valence mesons '' @xmath65 are explicitly taken into account . they are assumed to propagate freely between collisions , i.e. we neglect the mean field potentials acting on these mesons . the scalar density and the baryon current of the @xmath61-th type baryons are expressed in terms of df as follows : @xmath66 where @xmath67 is the spin - isospin degeneracy factor ( @xmath68 etc . ) . one can show @xcite that the kinetic equations ( [ kineq ] ) with the @xmath6- and @xmath7-fields evolving according to eqs . ( [ kgeqsigma]),([kgeqomega ] ) lead to the continuity equations @xmath69 and the energy - momentum conservation @xmath70 where the energy - momentum tensor is written as @xmath71 here we have also included possible contributions of the `` valence '' mesons @xmath72 , which can be produced at the annihilation . it is assumed that @xmath73 for the valence mesons . in order to solve eq . ( [ kineq ] ) numerically , df is represented by the set of point - like test particles : @xmath74 where @xmath75 is the number of physical particles of the type @xmath61 and @xmath76 is the number of test particles per physical particle ( the same for all types @xmath61 ) . the test particle positions @xmath77 and kinetic momenta @xmath78 are evolving in time according to the following equations : @xmath79 with @xmath53 and @xmath54 . it is easy to check that df ( [ tp ] ) with @xmath77 and @xmath80 satisfying eqs . ( [ rdot]),([pstardot ] ) gives a formal solution of the vlasov equation in the case when the collision integral in ( [ kineq ] ) is equal to zero . equations ( [ rdot]),([pstardot ] ) are equivalent to the hamiltonian equations of motion for the test particle positions @xmath77 and canonical momenta @xmath81 : @xmath82 where @xmath83 is the single - particle energy ( see @xcite ) . however , it is more convenient to propagate in time the test particle kinetic momenta rather than the canonical ones , since then eq . ( [ kgeqsigma ] ) for the @xmath6-field decouples from eq . ( [ kgeqomega ] ) for the @xmath7-field . when the collision integral in eq . ( [ kineq ] ) is taken into account , the test particles are propagated between the two - body collisions using eqs . ( [ rdot]),([pstardot ] ) . all calculations have been performed in the parallel ensemble mode . in this mode , the two - body collisions are permitted between the test particles belonging to the same parallel ensemble only , while the mean field is averaged over @xmath76 parallel ensembles of the test particles propagated simultaneously ( see eq . ( [ tp ] ) ) . therefore , a single parallel ensemble can be considered as a physical event . in actual calculations , we have neglected the time derivatives of the meson fields in eqs . ( [ kgeqsigma]),([kgeqomega ] ) . however , the spatial derivatives were treated without any simplifying assumptions . the reason for such a strategy is that we are dealing with nuclear systems which have large density gradients , but evolving slowly , as compared with the spatial and temporal scales involved in the mesonic equations of motion . indeed , including the temporal gradients would lead to the frequent oscillations of the mesonic fields with a period of less than @xmath84 , where @xmath85 is the meson mass this gives the period of 2.5 fm / c ( 1.5 fm / c ) for the @xmath6- ( @xmath7- ) field . by taking into account the finite wave lenghts of these oscillations would further reduce the periods . the treatment of such oscillations would strongly complicate the numerical calculations , in - particular , due to the classical meson field radiation . on the other hand , the characteristic periods of the oscillations are significantly smaller than the characteristic compression times ( @xmath0 fm / c , see sect . iii a below ) . therefore , one can approximately average - out the mesonic fields with respect to these oscillations , that is actually assumed in our model . the @xmath6- and @xmath7-fields are , therefore , calculated from the equations @xmath86 within the same approximation , the energy - momentum tensor has the following form : @xmath87 the factor @xmath88 in eq . ( [ twotd ] ) reflects the fact , that due to the omission of the time derivatives of the meson fields in the lagrangian density , only the first term in the r.h.s . contributes to the three - momentum density @xmath89 ( @xmath90 ) . although eqs . ( [ eqsigma]),([eqomega ] ) are not covariant , they provide a better description of the nuclear surface than pure local fields @xcite . this improves the stability of a nuclear ground state and is more appropriate for studying nuclear response to external hadronic and electromagnetic probes . equations ( [ eqsigma]),([eqomega ] ) have been solved numerically by applying the alternating direction implicit iterative method of douglas described in ref . @xcite . due to the scalar density dependence on the effective mass ( see eqs.([mstar]),([rhosj ] ) ) , additional iterations are needed to solve eq.([eqsigma ] ) . in other words , the scalar density has to be computed selfconsistently . to evaluate the meson fields , we used a uniform grid in coordinate space with steps @xmath91 . for the systems @xmath9@xmath92o , @xmath9@xmath17ca and @xmath9@xmath93pb considered below , the grid covered a cubic volume with the side of 10 , 20 and 30 fm , respectively , centered at the center - of - mass ( c.m . ) of a @xmath9a system . by numerical reasons , the @xmath94-functions in coordinate space , introduced in eq . ( [ tp ] ) , have been replaced by the gaussians of the width @xmath95 : @xmath96 the width of the gaussian and the grid step sizes are pure numerical parameters which should resolve the coordinate space nonuniformities of the system . in our case the characteristic space scale is given by the radius of the smallest considered nucleus @xmath92o , i.e. @xmath97 fm . on the other hand , in order to have smooth density distributions , the number @xmath76 of test particles per physical particle ( see eq . ( [ tp ] ) ) should be correlated to the width of the gaussian as @xmath98 . this puts a restriction on too small width due to cpu time increase . as an optimum choice , we fixed in the present work @xmath99 fm . we have realised , however , that there is a rather moderate tendency of increasing maximum compression ( see discussion in sect . iii a below ) with decreasing gaussian width . the number of test particles per nucleon was set to @xmath100 in the most of calculations . the equations of motion ( [ rdot]),([pstardot ] ) have been solved by applying the second - order in time predictor - corrector method @xcite with the time step of 0.1 fm / c . this value is small enough to resolve the time scale of a few fm / c for the compression processes ( see figs . [ fig : rho_u_vs_z ] and [ fig : rho_vs_t ] below ) . we have checked , that taking smaller time step does not influence the results . the full numerical scheme conserves the total energy with the accuracy of about 5% of the initial total binding energy of the @xmath9a system . the nucleons were distributed in coordinate space according to the woods - saxon density profile . the momenta of nucleons were sampled according to the local fermi distribution . the initial antiproton df was chosen as a gaussian wave packet in coordinate and momentum space @xcite located at the center of a nucleus ( @xmath101 ) : @xmath102 where @xmath103 is the width in coordinate space . equation ( [ cohstate ] ) implies , that the antiproton is at rest . the width of the initial antiproton distribution in momentum space is @xmath104 , which follows from the uncertainty relation . if not mentioned explicitly , the calculation is done with the choice @xmath105 fm . this value agrees with results of the static rmf calculations of ref . @xcite . as for nucleons , the antiproton df ( [ cohstate ] ) is projected onto test particles according to eq . ( [ tp ] ) with the @xmath94-functions in coordinate space replaced by gaussians . to avoid misunderstanding , we note that one should distinguish the width @xmath103 of the physical antiproton spatial distribution in eq . ( [ cohstate ] ) and the width of the test particle gaussian . after the initialization , the system of nucleons and antiproton was propagated in time according to eqs . ( [ rdot]),([pstardot ] ) . the meson fields have been calculated by eqs . ( [ eqsigma]),([eqomega ] ) with the source terms given by the scalar densities ( [ rhosj ] ) and the baryon currents ( [ jbj ] ) . in such a way , the evolution of the system toward compressed state has been followed . in this calculation , the collision term in the r.h.s . of the kinetic equation ( [ kineq ] ) has been set to zero , i.e. we considered a pure mean - field vlasov dynamics . this was done to see most clearly the role of the mean fields . an introduction of the @xmath106 and @xmath107 elastic collisions would mainly lead to a dissipation of the collective energy into heat . as pointed out in ref . @xcite , this effect is rather small and , therefore , can not change significantly the compression dynamics . the reason is , that the elastic collisions are not frequent on the time scale of compression ( see figs . [ fig : rho_u_vs_z ] , [ fig : rho_vs_z ] and [ fig : rho_vs_t ] ) . indeed , the mean time @xmath108 between nucleon - nucleon collisions can be estimated as @xmath109 , where @xmath110 mb is the elastic nucleon - nucleon cross section ( c.f . @xcite ) and @xmath111 is the fermi velocity . this gives @xmath112 fm / c for the nucleon density @xmath113 . the pauli blocking effect will further increase @xmath108 . a similar estimate can also be done for @xmath106 elastic collisions . at certain time moment @xmath114 , which is an external parameter to our model , we simulated the annihilation of an antiproton . this implies that annihilation occurs instantaneously , as a single quantum mechanical transition , in distinction to description of this process via the collision term in a kinetic equation . in the last case , the antiproton distribution function would gradually disappear on the way to the compressed state . the purpose of the present work is to look at the strongest possible effect of the nuclear compression on observables . therefore , we let the compressed system to be formed , and simulate the sudden annihilation afterwards . the ambiguity in the in - medium annihilation cross sections is taken into account by varying the parameter @xmath114 . in the actual calculations , the annihilation was simulated as follows : for each antiproton test particle , the closest in coordinate space nucleon test particle was chosen to be the annihilation partner . at large enough values of the total in - medium c.m . energy @xmath115 of the annihilating @xmath116 pair ( see below ) , the annihilation event of the test particle pair into mesons was simulated using the quark model @xcite , which has been already implemented in the gibuu model @xcite earlier . a quark and an antiquark with the same flavour are assumed to annihilate and transfer their total four - momentum to the remaining ( anti)quarks . the remaining four ( anti)quarks form two orthogonal @xmath117 jets with equal energies in the c.m . frame . the jets were hadronized via the lund string fragmentation model @xcite in the jetset version included into the pythia 6.225 program package . the applied annihilation model corresponds to the r2 type diagram in classification of ref . @xcite , i.e. to the quark rearrangement with one @xmath118 annihilation vertex . in this sense , the model has some similarity with the two - meson doorway models of refs . @xcite . to illustrate how the model works we have performed simulations of the @xmath119 annihilation . [ fig : pimul_ppbar ] shows the pion multiplicity distribution for the @xmath119 annihilation at rest in vacuum compared to the data compilation from refs . the calculated distribution is somewhat shifted to smaller pion multiplicities with respect to the data : the calculated average pion multiplicity @xmath120 compared with the experimental value of @xmath121 . we would like to remark , that non - vanishing contribution of the @xmath122 channel in calculations is completely due to the final states with other particles : @xmath123 ( 78% ) , @xmath124 ( 14% ) , @xmath125 ( 6% ) and @xmath126 + photons ( 2% before @xmath127 decay ) . in calculations , we took into account @xmath127 decays into @xmath128 or into final states with pions and disregarded photons afterwards . however , it is not clear to us how photons were counted in the data ( see also ref . @xcite ) . fig . [ fig : pimom_ppbar ] shows the calculated charged pion momentum distributions in the c.m . frame of the annihilating @xmath119 pair at rest in vacuum . from the partial contributions of the channels with various pion multiplicities we observe , as expected , that the hard(soft ) part of the total momentum distribution is populated mainly by the low(high ) pion multiplicity events . the experimental data are described reasonably well , except for the momenta @xmath129 gev / c , where the calculations significantly overestimate the data . we believe that the accuracy of the model in describing the data in figs . [ fig : pimul_ppbar ] and [ fig : pimom_ppbar ] is sufficient for the exploratory studies of global observables in the present work . certainly , the improvement of the annihilation model is needed to perform more detailed study of the mesonic final states in the annihilation . below we concentrate more on the in - medium effects on the annihilation . due to the mean field , the invariant energy of the annihilating @xmath116 pair can be substantially below the vacuum threshold value of @xmath130 . this makes the direct application of the jetset model for the @xmath9-annihilation in nuclei physically and numerically problematic . to overcome this difficulty , we introduced the corrected invariant energy as follows @xcite : @xmath131 where @xmath132 . the quantity @xmath133 is a vacuum analog of the total in - medium invariant c.m . energy @xmath115 with @xmath134 . provided that @xmath135 , we have used @xmath133 in the jetset simulation in order to produce the mesonic final states . this lower limit of @xmath115 is due to the fact that the jetset model does not generate enough direct @xmath136 and @xmath137 annihilation final states . in order to take into account the in - medium effects , in - particular , to ensure the correct in - medium threshold condition @xmath138 , where @xmath139 are the vacuum masses of the produced mesons , the annihilation event was accepted with the probability @xmath140 where @xmath141 is the invariant phase space volume , @xmath142 are the four - momenta of the produced mesons satisfying the vacuum mass shell conditions @xmath143 , @xmath144 . finally , the three - momenta of the produced mesons in the c.m . frame of the annihilating @xmath116 pair were multiplied by the common factor adjusted to get the correct in - medium total c.m . energy @xmath115 . in - fact , the way we simulate the in - medium effects eq.([p ] ) implies using the vacuum matrix elements of the annihilation channels , which are given by the jetset model , while taking into account the in - medium effects in the phase space factors only . similar procedures have been applied earlier in refs . @xcite . at @xmath145 , the final @xmath136 or @xmath137 channel was chosen by monte - carlo according to the probability ratio @xmath146 where @xmath147 is the ratio of the @xmath136 and @xmath137 final state probabilities for the @xmath119 annihilation at rest ( see table vi in ref . @xcite ) . for the zero total charge @xmath148 of the annihilating @xmath116 pair , the charge states of the outgoing pions were also determined from the data compilation of ref . since for @xmath149 the data are absent , the charges of the @xmath137 final states were determined by assuming that the @xmath150 and @xmath151 final channels have equal probabilities . the momenta of the outgoing pions were distributed microcanonically according to the available two- or three - body phase space . after the annihilation is simulated , the residual nucleons and produced mesons were propagated in time according to the full kinetic equations ( [ kineq ] ) , including both the baryonic mean fields and collision integrals . this takes into account the entropy production caused by the two - body collisions at the expansion stage . moreover , important processes of the meson rescattering and absorption , e.g. @xmath152 or @xmath153 , @xmath154 are included in the collision integral . these processes influence the observed particle spectra . as demonstrated in refs . @xcite by static rmf calculations , a deeply - bound antiproton - nucleus system can be significantly compressed as compared with a normal nucleus . now we want to study the real dynamics of such a system starting from the unperturbed nuclear ground state at @xmath155 . fig . [ fig : rho_u_vs_z ] ( top panels ) shows the nucleon and antiproton density profiles calculated at different times along the axis @xmath156 drawn through the center of the @xmath9@xmath17ca system . [ fig : rho_u_vs_z ] ( bottom panels ) also shows the nucleon and antiproton potentials @xmath157 , @xmath158 , along the same axis . left and right panels present results for @xmath40 and @xmath159 , respectively . we see that the initial configuration is unstable and the system starts to shrink . both nucleon and antiproton central densities grow quite fast , reaching their maxima within several fm / c . in the course of the compression process , the nucleon potential becomes deeper in the case of the reduced antiproton coupling constants ( @xmath40 ) and does not , practically , change in the @xmath1-parity motivated case ( @xmath159 ) . the antiproton potential deepens quite strongly with time for the both sets of the antiproton coupling constants . in the case of @xmath40 , the first maximum of the central nucleon density ( @xmath160 @xmath4 ) is reached at @xmath161 fm / c . at later time the system rebounds and oscillates approaching gradually a static configuration with the nucleon density @xmath162 @xmath4 at the center . since the annihilation is switched off in this calculation , the compressed configuration may exist , in principle , infinitely long time . however , due to numerical reasons , the stability is destroyed by a gradual test particle escape from a box in the coordinate space , where the mean field is computed . nevertheless , the numerical accuracy is good enough to trace the stable system up to at least @xmath163 fm / c . in the case of @xmath159 , the compression process is much faster than in the case of @xmath40 . already at @xmath164 fm / c we observe the first maximum of the central nucleon density with @xmath165 @xmath4 . a smaller value @xmath166 @xmath4 is reached asymptotically after some oscillations . in fig . [ fig : rho_vs_z ] we compare time evolution of the nucleon density distribution along the central axis @xmath156 for the light ( @xmath9@xmath92o ) and heavy ( @xmath9@xmath93pb ) systems . for @xmath9@xmath92o , the bell - like shape of the density distribution is reached quite fast . however , in the case of @xmath9@xmath93pb , we observe a quickly growing peak in the center , while peripheral nucleons still do not react on the compression . this leads to the delayed shape equilibration via a complicated compression - decompression cycle . [ fig : rho_vs_t ] shows the time evolution of the central nucleon density for the three systems : @xmath9@xmath92o , @xmath9@xmath17ca and @xmath9@xmath93pb . the case without annihilation is shown by the dotted lines . for a comparison , we also present the central density time evolution in the respective ground state nuclei without an antiproton inside ( dashed lines ) . we see , that at long enough times of the order of several tens fm / c , the static compressed configuration is indeed reached . the small oscillations of the central density in the compressed system visible at @xmath167 fm / c are approximately of the same amplitude as the oscillations of the respective ground state . thus , the reason for these small amplitude oscillations is the fermionic ground state instability due to the classical treatment of particles @xcite . in principle , this instability can be removed by either employing the pauli potential as in refs . @xcite or by adding a friction force to the hamiltonian equations of motions for the test particles @xcite . such modifications are beyond the scope of the present work . we expect that they would not essentially modify the density profiles of the compressed configurations , which are in an overall agreement with the previous static hartree calculations of refs . @xcite . one can also notice ( see lower panels of fig . [ fig : rho_vs_t ] ) a peculiar feature of the @xmath9@xmath93pb system : the dip in the central nucleon density at @xmath168 fm / c . this is mostly a consequence of the delayed shape equilibration mentioned above in discussing fig . [ fig : rho_vs_z ] . the especially strong density drop for @xmath40 is partly caused by the symmetry loss due to the finite number of test particles . the dip is absent in lighter systems , since the shape equilibration for them is much faster ( @xmath169 fm / c , see upper panels in figs . [ fig : rho_u_vs_z ] and [ fig : rho_vs_z ] ) . since the compression time is of primary importance , we have also studied the sensitivity of our results to the width @xmath103 of the initial antiproton df ( [ cohstate ] ) . we have found , that for a larger ( smaller ) width the compression time becomes somewhat longer ( shorter ) . in - particular , for the @xmath9@xmath92o system at @xmath159 the time needed to reach the first density maximum is 5 fm / c ( 2.5 fm / c ) for @xmath170 fm ( @xmath171 fm ) . for the same system at @xmath40 the first density maximum is reached at 10 fm / c ( 3 fm / c ) for @xmath170 fm ( @xmath171 fm ) . the value of the maximum density reached in the compression process is practically insensitive to @xmath103 . we have to admit also , that there is some numerical uncertainty in our calculations due to the choice of the width @xmath95 of the test particle gaussian and the grid step size . e.g. in the calculation with @xmath172 fm for the lightest system @xmath9@xmath92o the maximum and saturation densities are @xmath173 % higher than in calculation with @xmath99 fm . setting the smaller grid step is not feasible for technical reasons . overall , this numerical uncertainty is comparable to the one due to different choices of the scaling factor @xmath174 in our calculations . next , we study the dynamics of a residual nucleus after sudden annihilation of an antiproton . the annihilation was simulated as described in sect . iie . for each considered @xmath9a system and the scaling factor @xmath174 , three different annihilation times @xmath114 have been chosen : they correspond to ( i ) the early ( @xmath175 ) annihilation from a non - compressed ground - state nucleus , ( ii ) the annihilation at the time moment when the first maximum of the central density is reached , and ( iii ) the late annihilation from an asymptotic compressed configuration . one can see from fig . [ fig : rho_vs_t ] , that the annihilation from a non - compressed ground state nucleus ( thin solid lines ) does not lead to significant expansion of the residual nuclear system . the central nucleon density stays always below but close to @xmath45 in this case . in the case of annihilation from the compressed configurations ( thick solid and dash - dotted lines ) , we observe that , for the light systems @xmath9@xmath92o and @xmath9@xmath17ca , the central nucleon density decreases sharply after annihilation and reaches values well below @xmath45 . this is a clear indication of the collective expansion of a system from the initially compressed state . on the other hand , for the heavy @xmath9@xmath93pb system , the expansion is not very pronounced at any choice of the annihilation time . it is interesting , that if the annihilation is switched on at the first density maximum ( thick solid lines ) , then after an abrupt falling down the central density stays for some time @xmath176 fm / c close to @xmath45 before decreasing further . this is explained by an inertial compression : after annihilation , the periphery of a residual nucleus still continues to move to the center during some time until rebound . in the case of the @xmath9@xmath17ca system , this is demonstrated in figs . [ fig : vrad_ca40_xip03 ] and [ fig : vrad_ca40_xip1 ] , where we show the baryon density and the radial collective velocity , @xmath177 at several times as a function of the radius @xmath178 . the collective velocity has been determined as @xmath179 at @xmath180 fm , the evolution of the radial collective velocity field is noticeably influenced by the fermionic ground state instability discussed above . nevertheless , one can still observe the inertial compression . indeed , the radial collective velocity at @xmath181 fm / c for @xmath40 and at @xmath182 fm / c for @xmath159 reveals the fast outward motion at the center , while the peripheral nucleons still continue to move to the center . this explains plateaus in the central density evolution near @xmath183 fm / c in fig . [ fig : rho_vs_t ] . at @xmath184 fm / c the whole system starts to expand . this is reflected in the monotonically increasing radial collective velocity with radius . the especially strong rise of @xmath185 at large @xmath178 is due to emission of fast particles . at later times @xmath186 fm / c the expansion is replaced by the inward motion of the matter in the central zone . however , the fast particles are still continuing to escape from the dense region . we expect that in reality the system will break - up into fragments before the inward motion will start ( see discussion sect . it is presently well established ( see @xcite and refs . therein ) that nuclear matter at low densities ( @xmath187 ) becomes unstable with respect to small density perturbations , so called spinodal instability . however , in order these density perturbations to develop into nuclear fragments , the system must stay long enough time @xmath188 fm / c in the spinodal region . one can see from fig . [ fig : rho_vs_t ] , that the light systems @xmath9@xmath92o and @xmath9@xmath17ca spend a long time in this region . therefore , the residual nuclear systems can undergo a multifragment break - up , if the annihilation happens in the compressed configurations . in other words , the multifragment break - up of nuclei after the @xmath9-annihilation may serve as a signal of the compression prior the annihilation . in table [ tab : fragmenting_source ] we collect the estimated parameters of fragmenting sources for the @xmath9@xmath92o and @xmath9@xmath17ca systems . the case of @xmath9-annihilation from the state of maximum central density is considered here ( see thick solid lines in fig . [ fig : rho_vs_t ] ) . the sources have been determined by selecting nucleons in the space region where the baryon density is larger than @xmath189 . they are characterized by the neutron ( @xmath190 ) and proton ( @xmath191 ) numbers , the collective kinetic energy per nucleon ( @xmath192 ) and the residual excitation energy per nucleon ( @xmath193 ) . these parameters are defined as @xmath194 where @xmath195 and @xmath196 are the neutron and proton densities , respectively ; @xmath197 ; and @xmath198 is the ground state energy per nucleon of a nucleus with neutron number @xmath190 and proton number @xmath191 computed within our model . the collective kinetic energy ( [ e_kin^coll ] ) is calculated neglecting pressure effects . due to the initially isospin - symmetric nuclei and the neglect of the isovector mesons in the lagrangian density ( [ lagr ] ) , we obtained in all cases @xmath199 in the source . it turned out also , that the coulomb interaction optionally included in some of the calculations does not change this result . thus , only the charge numbers @xmath191 are given in table [ tab : fragmenting_source ] . the time moment for determination of the source parameters has been chosen at @xmath200 fm / c , when the central nucleon density is about @xmath201 , i.e. inside the spinodal region . the calculated residual excitation energy is typically @xmath202 mev / nucleon that corresponds to the temperatures @xmath203 mev . according to the statistical multifragmentation model @xcite , in this energy domain the multifragment break - up is the dominating decay channel of residual nuclei . the collective kinetic energy is @xmath204 mev / nucleon in the case of @xmath9@xmath92o . this is well above the coulomb energy of the source , which is only about 0.4 mev / nucleon for @xmath205 . unfortunately , such a source is too small to experience the real multifragment break - up , rather a fermi break - up into small clusters @xcite . in the case of larger sources , produced in @xmath9@xmath17ca annihilation the collective kinetic energy is considerably smaller , @xmath206 mev / nucleon , but still significant with respect to the total coulomb energy @xmath207 mev / nucleon for @xmath208 . thus , we expect some signs of collective expansion to be visible in kinetic energy spectra of produced fragments . let us now consider other observable effects . [ fig : ekin_spectra ] shows the c.m . kinetic energy spectra of the nucleons emitted from the @xmath9@xmath92o , @xmath9@xmath17ca and @xmath9@xmath93pb systems after the @xmath9-annihilation . in order to separate emitted nucleons from the bound nucleons of a residual nucleus , we used a simple criterion : only those nucleons , both protons and neutrons , were included in the spectra which are separated by at least 3 fm from the other test particles of a given parallel ensemble at @xmath163 fm / c . one can see , that nucleons with the kinetic energy @xmath209 , where @xmath210 mev is the fermi energy of the nuclear matter at @xmath45 , are abundantly emitted . such nucleons are knocked - out from the nucleus by the mesons produced after the annihilation @xcite . in table [ tab : slopes ] we list the slope parameters @xmath211 of the nucleon kinetic energy spectra obtained by the maxwell - boltzmann fit @xmath212 in the region of @xmath213 mev . we would like to mention , that the authors of ref . @xcite report the slope temperature of about 60 mev for the kinetic energy spectrum of the emitted protons in the case of @xmath9@xmath17ca system , which is not so far from our results for the annihilation in the ground - state nucleus at @xmath175 . we want to emphasize that the kinetic energy spectra of nucleons emitted after the @xmath9-annihilation from the compressed @xmath9@xmath92o , and @xmath9@xmath17ca systems are significantly harder than the spectra of nucleons from the annihilation at @xmath175 . this can be explained by two effects . first , the collective expansion of the outer shell will increase the slope temperature , typically , by several mev ( see table [ tab : fragmenting_source ] and figs . [ fig : vrad_ca40_xip03 ] , [ fig : vrad_ca40_xip1 ] ) . second , just after the annihilation the nucleon potential at the center of a nucleus grows suddenly by @xmath214 mev . this creates an additional push for the fast nucleons emitted from the nucleus . although the hardening effect is most pronounced for the lightest system @xmath9@xmath92o , it is also quite visible for @xmath9@xmath17ca . for the heaviest system @xmath9@xmath93pb , we observe almost identical high energy tails of the nucleon spectra for the different annihilation times . the reason is that the collective expansion is practically absent in this system ( see fig . [ fig : rho_vs_t ] ) . also , the yield of fast nucleons is reduced by their subsequent rescatterings in the residual nucleus . the meson production from the @xmath9-annihilation inside nucleus is influenced both by the mean field via the potentials of the annihilating pair and by the final state interactions ( fsi ) , i.e. the two - body collisions and resonance decays . it is instructive to disentangle the contributions of the mean field effects from the rescattering and absorption effects . to this aim , we have performed additional calculations by subsequently switching off the fsi and the mean field . corresponding results are shown in figs . [ fig : pimul_pbaro16],[fig : pimom_pbaro16 ] and [ fig : minv_spectra_pbaro16 ] . figure [ fig : pimul_pbaro16 ] shows pion multiplicity distributions for the @xmath9@xmath92o system . in the case of reduced antiproton coupling constants ( @xmath40 ) , the mean field alone does not produce any noticeable modification of the pion multiplicity distribution . on the other hand , fsi leads to a rather substantial shift toward smaller @xmath215 . for the case of @xmath159 , we observe a strong pion multiplicity reduction due to smaller @xmath115 for the @xmath116 annihilation , while the fsi effects are much weaker . one can also see a quite significant compressional effect for @xmath159 , which is only very weak for @xmath40 . in fig . [ fig : pimom_pbaro16 ] , we present charged pion momentum spectra in the c.m . frame of the @xmath9@xmath92o system . fsi strongly modifies these spectra , mostly due to the @xmath152 processes , which effectively decelerate pions . as we have already observed earlier in fig . [ fig : ekin_spectra ] , emitted nucleons gain energy , correspondingly . the effect of the baryonic mean field on the pion momentum spectrum is relatively moderate : we observe some depletion of the high - momentum tail , which is more pronounced in the case of @xmath159 . the compressional effect is visible in the reduction of the pion yield in the momentum range @xmath216 gev / c . this can be understood from figs . [ fig : pimom_ppbar ] and [ fig : pimul_pbaro16 ] : in vacuum , the events with @xmath217 and @xmath218 contribute substantially to this momentum range , while the probability of such events is substantially suppressed in a compressed nucleus due to the reduced annihilation phase space @xcite . the pion momentum spectrum has clearly the two components : the slow pions , which have undergone rescatterings via the @xmath30-resonance excitation ( @xmath219 less than about 300 mev / c ) , and the high energy pions , which were emitted from the system almost without secondary interactions . the similar result has been obtained in earlier intranuclear cascade calculations @xcite . following ref . @xcite , we have also fitted the low energy part of the pion spectrum ( @xmath220 mev , @xmath221 ) by the maxwell - boltzmann distribution @xmath222 this fit has produced the following slope temperatures of the charged pion momentum spectrum for the @xmath9@xmath92o system : @xmath223 mev and 44 mev for @xmath40 ( @xmath224 mev and 36 mev for @xmath159 ) in the case of early annihilation ( @xmath175 ) and annihilation at the time of maximum compression , respectively . the extracted @xmath225 is smaller than the slope temperature of @xmath226 mev of the low energy pions for the @xmath9@xmath17ca system reported in ref . however , in the calculation without mean field , we obtain @xmath227 mev , which is in a good agreement with the result of ref . therefore , the difference between our results and those of ref . @xcite is caused by the mean field acting on the annihilating pair in medium . moreover , we see the softening of the pion spectrum in the case of larger antinucleon couplings due to smaller @xmath115 of the annihilating pair . figure [ fig : minv_spectra_pbaro16 ] presents the distributions of the annihilation events in the total invariant mass of produced mesons from the @xmath9@xmath92o system . the invariant mass is defined as @xmath228 where @xmath229 is the sum of four - momenta of the mesons produced in a given annihilation event . the calculations were done for the case of @xmath175 . in the absence of fsi , @xmath230 should be equal to the invariant energy @xmath115 of the annihilating @xmath9n pair . indeed , without fsi and without mean field , as expected , we get a quite sharp peak at @xmath130 only slightly smeared out due to the fermi motion of nucleons and momentum spread of the initial antiproton df ( [ cohstate ] ) . the baryonic mean field leads to the shift of a peak position toward smaller @xmath230 and to some broadening of the distribution . the broadening is due to the spatial spread of the initial antiproton df ( [ cohstate ] ) , which results in different mean - field potentials acting on different annihilating @xmath9n test particle pairs . additionally , the fsi leads to a very strong broadening of the invariant mass spectrum due to the deceleration and absorption of the annihilation mesons . this is clearly seen in fig . [ fig : minv_spectra_pbaro16 ] for the case when the rmf was switched off ( dotted line ) . nevertheless , the full calculation ( solid line ) shows quite strong softening of the @xmath230 distribution due to the mean - field effects . obviously , this effect is stronger for the case of @xmath159 as compared with @xmath40 . finally , in fig . [ fig : minv_spectra ] , we systematically study how the meson invariant mass spectra are affected by the nuclear compression effects . the results are shown for the different @xmath9-nucleus systems . due to the strong reduction of the nucleon effective mass with the scalar density , the meson invariant mass spectra become softer when the annihilation happens in the compressed configurations , as compared with the annihilation in the normal state at @xmath175 . the effect is , again , more pronounced for the light systems @xmath9@xmath92o and @xmath9@xmath17ca . in the @xmath9@xmath92o system the shift is almost 500 mev even in the case of @xmath40 . we believe that results presented in figs . [ fig : ekin_spectra],[fig : pimul_pbaro16],[fig : pimom_pbaro16 ] and [ fig : minv_spectra ] constitute the set of observables sensitive to the compressional effects in nuclei induced by an antiproton . we have performed dynamical modeling of possible compression effects in nuclei due to the presence of an antiproton . the semiclassical transport gibuu model @xcite incorporating the relativistic mean fields for the baryons and antibaryons has been employed in calculations . the model reproduces reasonably well the earlier static calculations of bound @xmath9-nuclear systems @xcite . in this work , we did not consider the stopping process of an incident antiproton in a target nucleus . this is a rather complicated problem due to the unknown in - medium cross sections of the @xmath9-scattering and annihilation . this problem will be addressed in a forthcoming paper . instead , we have assumed , that the antiproton has penetrated to the center of the nucleus , stopped there due to an inelastic collision , and then get captured to the lowest energy state . such events should be very rare , with a probability of the order of @xmath231 for the central collisions @xcite . as proposed in ref . @xcite , formation of bound antiproton - nucleus states can be triggered by the emission of fast nucleons , pions or kaons . we have shown , that during the time interval of @xmath0 fm / c after creation of the initial state the central density of the target nucleus grows up to the values of @xmath232 depending on somewhat uncertain values of the antiproton coupling constants . we expect that the life time of strongly bound antiprotons can be long enough to observe this cold compression effect . detailed kinetic simulations of the post - annihilation evolution of residual nuclei have been carried out at different assumptions on the annihilation time . it is shown , that the @xmath9-annihilation in compressed light systems , like @xmath9@xmath92o and @xmath9@xmath17ca , leads to the pronounced collective expansion of the residual nucleus , which may result in the multifragment break - up . another clear signature of the nuclear compression is the hardening of the kinetic energy spectra of emitted nucleons . on the other hand , the invariant mass distribution of produced mesons is shifted to smaller invariant masses due to the in - medium reduction of the nucleon effective mass at high scalar density . similar phenomena are expected also for the case of @xmath233-nucleus bound states which can be produced via the @xmath234 reaction on nuclei . another interesting possibility is that the compressed zone of the nucleus might undergo a deconfinement phase transition . then one can expect formation of a quark - antiquark droplet with a non - zero baryon number and relatively low temperature @xcite . our main assumption in the present study was that the annihilation takes place in the central region of a nucleus . the experimental selection of the central annihilation events is a difficult problem . no clear trigger condition for such events has been invented so far . one suggestion is that the central annihilation events , in average , will be characterised by isotropic emission of secondary particles and high fragment multiplicity @xcite . however , further theoretical and experimental efforts are needed to develop a good trigger condition for the central annihilation . despite of these difficulties , we propose to study the above predictions in antiproton - nucleus reactions at the future facility for antiproton and ion research ( fair ) at gsi ( darmstadt ) . the support by the frankfurt center for scientific computing is gratefully aknowledged . the authors are grateful to a.s . botvina , th.j . brvenich , i.a . pshenichnov , j. ritman and h. stcker for stimulating discussions . this work has been partially supported by the dfg grant 436 rus 113/711/0 - 2 ( germany ) and the grant ns-3004.2008.2 ( russia ) . .[tab : fragmenting_source ] fragmenting source parameters for the different annihilating systems and values of the scaling factor @xmath174 of the antiproton coupling constant . @xmath114 denotes the annihilation time moment . @xmath236 is the time moment when the source parameters have been determined . @xmath237 is the central nucleon density . @xmath191 , @xmath192 and @xmath193 are the charge number , collective kinetic energy per nucleon and residual excitation energy per nucleon , respectively ( see eqs . ( [ nz]),([e_kin^coll ] ) and ( [ e_res^coll ] ) ) . [ cols="^,^,^,^,^,^,^,^,^ " , ]

in 2007 rob spekkens proposed a toy theory @xcite with the aim of showing that many of the characteristic features of quantum mechanics could result from a restriction on our knowledge of the state of an essentially classical system . the theory describes a simple type of system which mimics many of the features of a quantum qubit . the success of the toy theory in replicating characteristic quantum behaviour is , in one sense , quite puzzling , since the mathematical structures employed by the two theories are quite different . quantum mechanics represents states of systems by vectors in a hilbert space , while processes undergone by systems are represented by linear maps . in contrast , the toy theory represents states by subsets and processes by relations . this ` incomparability ' means that the mathematical origins of the similarities ( and differences ) between the two theories are not easy to pinpoint . in this paper we consider the toy theory from a new perspective - by looking at the ` algebra ' of how the processes of the theory combine . the mathematical structure formed by the composition of processes in any physical theory is a _ symmetric monoidal category _ @xcite , with the objects of the category representing systems , and the morphisms representing processes undergone by these systems - we term this the _ process category _ of the theory . work initiated by abramsky and coecke @xcite , and continued by many other authors @xcite has investigated the structure of the process category of quantum mechanics . mathematical structures of this category have been identified which correspond directly to key physical features of the theory . in principle , any theory whose processes form a category with these mathematical features should exhibit these quantum - like physical features . it will be gratifying then to note that the process category of the toy theory shares many of these features , thus ` accounting for ' the similarities which this theory has with quantum mechanics . the main aim of this paper is to characterise the process category of the toy theory . this turns out to be less straightforward than one might have hoped ! in the case of quantum mechanics it is straightforward to see that the states of a system correspond exactly with the vectors of the hilbert space describing that system , and that processes correspond exactly with linear maps ; thus we can immediately conclude that the process category of quantum mechanics is * hilb * , whose objects are hilbert spaces and whose morphisms are linear maps . such a quick and easy statement of the process category for the toy theory is not possible . this is due largely to the way that the valid states and processes of the theory are defined : this is via an inductive procedure where the valid states and transformations for a collection of @xmath0 systems must be known before those for a collection of @xmath1 systems can be deduced . there is also a certain degree of ambiguity in the way in which the theory was originally stated , and , since the toy theory was never fully constructed ( until now ) , its consistency was not even guaranteed . another feature which was not addressed in the theory is _ compositional closure _ _ , that is : _ if we compose two valid processes , either in parallel or sequentially ( when types match ) , do we again obtain a valid process ? _ obviously , this is a natural operational requirement , and as indicated above , this notion of compositional closure is the operational cornerstone to modelling physical processes in a symmetric monoidal category . in this paper , it is exactly the compositional closure assumption which allows us to formulate the toy theory in terms of no more than a few generators , with clear operational meanings . _ _ _ _ _ the structure of the paper is as follows . we begin with a very brief summary of the toy bit theory , and a discussion of the ambiguities in its definition . we then define a categories * spek * and * mspek * which we claim is the process category for the toy theory . this is a sub - category of * rel * , whose objects are sets and whose morphisms are relations . the next section is devoted to showing the general form of the relations which constitute the morphisms of * mspek*. we then go on to argue that these relations are exactly those which describe the processes of the toy theory , thus demonstrating that * mspek * is the process category of the toy theory . finally we note some of the key categorical features of * mspek * and link these to the characteristically quantum behaviour exhibited by the toy bit theory . the category * spek * was first proposed by us in @xcite , and we provide some explicit pointers to useful information therein throughout this text . for full details of spekkens s toy bit theory , the reader is referred to the original paper @xcite . here we provide a very brief summary of the key points . there is just one type of elementary system in the theory , which can exist in one of four states . we will denote the set of these four states , the _ ontic state space _ by @xmath2 . alternatively we can depict it graphically as : @xmath3 the ontic states are to be distinguished from the _ epistemic _ states , which describe the extent of our knowledge about which ontic state the system occupies . for example we might know that the system is either in ontic state 1 or 2 . we would depict such an epistemic state in the following fashion : @xmath4 the epistemic state is intended to be the analogue of the quantum state . note that an epistemic state is simply a subset of the ontic state space . given a composite system of @xmath1 elementary systems , its ontic state space is simply the cartesian product of the ontic state spaces of the composite systems , @xmath5 . thus such a system has @xmath6 ontic states . the key premise of the theory is that our knowledge of the ontic state is restricted in a specific way so that only certain epistemic states are allowed . we refer the reader to the original paper for the full statement of this epistemic restriction , which spekkens refers to as the _ knowledge balance principle_. it suffices here to say that , given an @xmath1-component system , an epistemic state is a @xmath7 or @xmath8 element subset of the ontic state space , the @xmath9 case being the situation of maximal knowledge and the @xmath8 case being the situation of total ignorance . for a single system this means that there are six epistemic states of maximal knowledge : @xmath10 and one of non - maximal knowledge : @xmath11 a consequence of the knowledge balance principle is that any measurement on a system inevitably results in a probabilistic disturbance of that system s ontic state . a measurement in the toy theory essentially corresponds to asking which of a collection of subsets of the ontic state space contains the ontic state . an example would be a measurement on a single system corresponding to the question `` is the ontic state in the subset @xmath12 or the subset @xmath13 ? '' . if the initial epistemic state of the system was the subset @xmath14 then answering this question would allow us to pin down the ontic state precisely , to 1 if we get the first answer or 3 if we get the second , in violation of the knowledge balance principle . thus we hypothesise that if , for example , we get the first answer to our question , then the ontic state undergoes a random disturbance , either remaining in state 1 or moving into state 2 , with equal probability . the epistemic state following the measurement then would be @xmath12 . the effect of such a probabilistic disturbance on the _ epistemic _ state is best modelled by viewing it as a _ relation _ on the ontic state space , defined by @xmath15 , @xmath16 , @xmath17 , @xmath18 , where the notation denotes that the element 1 in the domain relates to both the elements 1 and 2 in the codomain , the element 2 in the domain , while the elements 3 and 4 in the domain relate to no elements in the codomain . below we will omit specification of those elements in the domain that relate to no elements in the codomain . thus in general the transformations of the theory are described by _ relations_. in his original paper spekkens derives the states and transformations allowed in the theory , apparently appealing only to the following three principles : 1 . the epistemic state of any system must satisfy the knowledge balance principle globally i.e. it should be a @xmath7 or @xmath8 element subset of the ontic state space . 2 . when considering a composite system , the ` marginal ' epistemic state of any subsystem should also satisfy the knowledge balance principle . by the marginal epistemic state we mean the following . suppose the whole system has @xmath1 components , and we are interested in an @xmath19-component subsystem . the epistemic state of the whole system will be some set of @xmath1-tuples : to get the marginal state on the @xmath19-component system we simply delete the @xmath20 entries from each tuple which correspond to the subsystems which are not of interest to us . 3 . applying a valid transformation to a valid state should result in a valid state . the second principle is used , for example , to rule this out as an epistemic state for a two - component system : @xmath21 the third principle is used extensively . it is invoked to show that the transformations on a single element system constitute permutations on the set of ontic states , since any other function on @xmath22 would lead to some valid epistemic state being transformed into an invalid state . another illuminating example of the third principle in action is the elimination of this state as a valid epistemic state : @xmath23 this state would be allowed by the first and second principles , but upon making a certain measurement on one of the systems , the resulting measurement disturbance would transform it into this state : @xmath24 which clearly fails to satisfy the first principle . the mode of definition of the theory , via the three principles stated earlier , raises some interesting issues . firstly , this approach seems to be necessarily iterative . compare it , for example , to how we would describe the form of valid states in quantum mechanics - in one line we can say that they are the normalised vectors of the system s state space . secondly , do these rules actually uniquely define the toy theory ? there does seem to be a problem with the third rule . when we use it to rule out the state in diagram [ ruledoutby3 ] , our argument is that we already know what the valid measurement disturbance transformations on a single elementary system are , and when we apply one of them to this state we obtain a state which clearly violates the knowledge balance principle . however , it is not clear that we could not have made the alternative choice - that this state should be valid , and therefore that the transformation which we had previously thought was valid could no longer be considered as such . it seems that considerations other than the three rules above come into deciding which should be the valid states of the theory , but it is nowhere clearly stated exactly what they are . the second point touched slightly on our final issue : is the theory as spekkens presents it consistent ? he derives valid states and transformations for systems with up to three elementary components . however , can we be sure that these states and transformations , when combined in more complex situations involving four or more elementary systems , wo nt yield a state which clearly violates the knowledge balance principle ? currently there seems to be no such proof of consistency . in fact , in the process of re - expressing the theory in categorical terms , we will develop such a proof . we briefly review the key structural features of the process categories of quantum - like theories . a physicist - friendly tutorial of the category theoretic preliminaries is in @xcite . a survey on the recent applications of these to quantum theory and quantum information is in @xcite . [ smcdef ] a _ symmetric monoidal category _ @xmath25 is a category equipped with the following extra structure : a bifunctor @xmath26 ; a _ unit object _ @xmath27 ; and four natural isomorphisms , left and right unit : @xmath28 , @xmath29 , associative : @xmath30 and commutative : @xmath31 . furthermore these objects and natural isomorphisms obey a series of _ coherence conditions _ @xcite . as argued in @xcite , the process category of any physical theory is a smc . the bifunctor @xmath32 is interpreted as adjoining two systems to make a larger compound system . * hilb * becomes a smc with the tensor product as the bifunctor , and @xmath33 as the unit object . * rel * is a smc with the cartesian product as the bifunctor and the single element set as the unit object . there is a very useful graphical language for describing smcs , due to joyal and street @xcite , which we will make extensive use of . it traces back to penrose s earlier diagrammatic notation for abstract tensors @xcite . in this language we represent a morphism @xmath34 by a box : @xmath35 @xmath36 , the composition of morphisms @xmath37 and @xmath38 is depicted as : @xmath39 the identity morphism @xmath40 is actually just written as a straight line this makes sense if you imagine composing it with another morphism . turning to the symmetric monoidal structure , a morphism @xmath41 is depicted : @xmath42 and if @xmath34 and @xmath43 then @xmath44 is depicted as : @xmath45 the identity object @xmath27 is not actually depicted in the graphical language . morphisms @xmath46 and @xmath47 are written as : @xmath48 the associativity and left and right unit natural isomorphisms are also implicit in the language . the symmetry natural isomorphism is depicted as : @xmath49 in fact the graphical language is more than just a useful tool ; it enables one to derive all equational statements that follow from the axioms of a smc : two morphisms in a symmetric monoidal category can be shown to be equal using the axioms of a smc iff the diagrams corresponding to these morphisms in the graphical language are isomorphic , where by diagram isomorphism we mean that the boxes and wires of the first are in bijective correspondence with the second , preserving the connections between boxes and wires . the categories described by the graphical language are in fact _ strict smcs , that is , those for which @xmath50 , @xmath51 and @xmath52 are equalities . mac lane s _ strictification theorem @xcite , which establishes an equivalence of any smc with a strict one , enables one to apply the diagrammatic notation to any smc . _ _ we will often refer to morphisms of type @xmath46 as _ states _ , since in a process category they represent the preparation of a system @xmath53 in a given state . in * hilb * such morphisms are in bijection with the vectors of the hilbert space @xmath53 ; in * rel * they correspond to subsets of the set @xmath53 . the process categories of _ quantum - like _ theories possess a range of additional structures . a _ dagger category _ is a category equipped with a contravariant involutive identity - on - objects functor @xmath54 . dagger symmetric monoidal category _ ( @xmath55-smc ) is a symmetric monoidal category with a dagger functor such that : @xmath56 , @xmath57 , @xmath58 , @xmath59 , and @xmath60 . * hilb * is a @xmath55-smc with the _ adjoint _ playing the role of the dagger functor . * rel * is a @xmath55-smc with _ relational converse _ as the dagger functor . [ csdef ] in a smc @xmath61 a _ compact structure _ on an object @xmath53 is a tuple @xmath62 where @xmath63 is a dual object to @xmath53 which may or may not be equal to @xmath53 , and @xmath64 and @xmath65 satisfy the conditions : @xmath66 and the dual diagram for @xmath67 . a _ compact closed _ category @xmath61 is a smc in which all @xmath68 have compact structures . * hilb * is a compact closed category , where for each object @xmath53 the _ bell - state _ ket @xmath69 ( with @xmath70 a basis for the hilbert space @xmath53 ) familiar from quantum mechanics provides the morphism @xmath64 , and the corresponding bra acts as @xmath65 . perhaps unsurprisingly , given this example , abramsky and coecke in @xcite showed that this structure underlies the capacity of any quantum - like theory to exhibit information processing protocols such as _ teleportation _ and _ entanglement swapping_. furthermore abramsky has shown @xcite that any theory whose process category is compact closed will obey a generalised version of the no - cloning theorem . for the compact structures of interest in this paper @xmath71 is always equal to @xmath53 see proposition [ inducedcsprop ] below . hence , from here on we wo nt distinguish these two objects . this simplifies the graphical language , which can be extended by introducing special elements to represent the morphisms of the compact structure @xmath64 and @xmath65 : @xmath72 equation [ cccomdiag ] and its dual are then depicted as : @xmath73 this extension of the graphical language now renders it completely equivalent to the axioms of a compact closed category ; for a detailed discussion we refer to @xcite . two morphisms in a compact closed category can be shown to be equal using the axioms of compact closure iff the diagrams corresponding to these morphisms in the graphical language are isotopic . compact closure is of particular importance to us since in any compact closed category there will be _ map - state duality _ : a bijection between the hom - sets @xmath74 and @xmath75 ( in fact both these will further be in bijection with @xmath76 ) : @xmath77 and @xmath78 that this is a bijection follows from equation [ yanking ] . if we have a morphism with larger composite domain and codomain the number of hom - sets in bijection increases dramatically . for example the morphisms of the hom - sets @xmath79 and @xmath80 are in bijection : explicitly the conversion between morphisms from the two sets can be depicted as : @xmath81 clearly manoeuvres like this can convert any ` input ' line into an ` output ' line , by using the unit and co - unit morphisms to ` bend lines around ' . [ decdef ] a _ _ diagram equivalence class _ _ ( dec ) is a set of morphisms in a compact closed category which can be inter - converted by composition with the units and co - units of the factors of their domains and codomains . the final structure of interest is the _ basis structure _ , which can be seen as the ` dagger - variant ' of carboni and walters s frobenius algebras @xcite . this structure is discussed at length in @xcite . [ bsdef ] in a @xmath55-smc a _ basis structure _ @xmath82 on an object @xmath53 is a commutative isometric dagger frobenius comonoid @xmath83 . for more details on this definition see section 4 of @xcite where basis structures are referred to as ` observable structures ' . we represent the morphisms @xmath84 and @xmath85 graphically as : @xmath86 basis structures are so named because in * hilb * they are in bijection with orthonormal bases , via the correspondence : @xmath87 where @xmath88 is an orthonormal basis for @xmath89 . this is proved in @xcite . [ sepbsprop ] two basis structures @xmath90 and @xmath91 induce a third basis structure @xmath92 , with : @xmath93 [ inducedcsprop ] any basis structure induces a self - dual dagger compact structure , with @xmath94 and @xmath95 . basis structures are identified in @xcite as a key structure underlying a variety of quantum - like features , for example the existence of incompatible observables , and information processing tasks such as the quantum fourier transform . in the section we give the definitions of two key categories . the first definition is a convenient stepping stone to the second : [ defspek ] the category * spek * is a subcategory of * rel * , defined inductively , as follows : * the objects of * spek * are the single - element set @xmath96 , the four element set @xmath97 , and its @xmath1-fold cartesian products @xmath5 . * the morphisms of * spek * are all those relations generated by composition , cartesian product and relational converse from the following generating relations : 1 . all permutations @xmath98 of the four element set , represented diagrammatically by : @xmath99 there are 24 such permutations and they form a group , @xmath100 . 2 . a relation @xmath101 defined by : @xmath102 represented diagrammatically by : @xmath103 3 . a relation @xmath104 defined by @xmath105 and represented diagrammatically by : @xmath106 4 . the relevant unit , associativity and symmetry natural isomorphisms . @xcite[spek : basisstruc ] @xmath107 is a basis structure in * frel * and hence also in * spek*. this category turns out to be the process category for the fragment of the toy theory containing only epistemic states of maximal knowledge . in particular , as discussed in detail in @xcite , the interaction of the basis structure @xmath107 and the permutations of @xmath108 results in three basis structures , analogous to the @xmath109- , the @xmath110- and the @xmath111-bases of a qubit . furthermore , as also discussed in @xcite4 , * spek * also contains operations which in quantum theory correspond with projection operators , while these are not included in the toy theory . indeed , compact closure of * spek * implies _ map - state duality _ _ : operations and bipartite states are in bijective correspondence . _ _ _ the process category for the full toy theory has a similar definition : the category * mspek * is a sub - category of * rel * , with the same objects as * spek*. its morphisms are all those relations generated by composition , cartesian product and relational converse from the generators of * spek * plus an additional generator : @xmath112 by construction , both * spek * and * mspek * inherit dagger symmetric monoidal structure from * rel * , with cartesian product being the monoidal product , and relational converse acting as the dagger functor . * spek * and * mspek * are both compact closed . by propositions [ spek : basisstruc ] and [ sepbsprop ] it follows that each object in these categories has a basis structure , and hence by proposition [ inducedcsprop ] they both are compact closed . the primary task of the remainder of this paper is to demonstrate that * mspek * is indeed the process category of spekkens s toy theory . we first make some preliminary observations . a _ * spek * diagram _ is any valid diagram in the graphical language introduced in section [ seccategories ] which can be formed by linking together the diagrams of the * spek * generators , as described in definition [ defspek ] . there is clearly a bijection between the possible compositions of * spek * generators , and * spek * diagrams . a * spek * diagram with @xmath19 inputs and @xmath1 outputs represents a morphism of type @xmath113 : a relation between sets @xmath114 and @xmath5 . the number of relations between two finite sets @xmath53 and @xmath115 is clearly finite itself : it is the power set of @xmath116 . thus the hom - set * frel*(a , b ) is finite . since @xmath117 we can be sure that the hom - sets of * spek * are finite . on the other hand , there is clearly an infinite number of * spek * diagrams which have @xmath19 inputs and @xmath1 inputs - we can add more and more internal loops to the diagrams . thus many diagrams represent the same morphism . however the morphisms of * spek * are , by definition , all those relations resulting from arbitrary compositions of the generating relations , i.e. any relation that corresponds to one of the infinity of * spek * diagrams . hence any proof about the form of the morphisms in * spek * is going to have to be a result about the relations corresponding to each possible * spek * diagram , even though in general many diagrams correspond to a single morphism . if we know the relation corresponding to one diagram in one of * spek * s diagram equivalence classes ( recall definition [ decdef ] ) , then it is straightforward to determine the relations corresponding to all of the other diagrams . [ bendlinelemma ] given a * spek * diagram and corresponding relation : @xmath118 then the relation corresponding to the following diagram : @xmath119 is given by , for all @xmath120 : @xmath121 where @xmath122 is the subset of @xmath123 which is related by @xmath124 to @xmath125 . every diagram equivalence class in * spek * has at least one diagram of type @xmath126 , representing a state , where we make every external line an output . relations of this type can be viewed as subsets of the set @xmath5 and it will be convenient for us to concentrate on characterising these morphisms . via lemma [ bendlinelemma ] any results on the general form of states will translate into results on the general form of all morphisms . in what follows we will therefore make no distinction between the inputs and outputs of a * spek*-diagram : a diagram with @xmath19 inputs and @xmath1 outputs will simply be referred to as a @xmath127-legged diagram . our proof will involve building up * spek * diagrams by connecting together the generating morphisms . here we show what various diagram manipulations mean in concrete terms for the corresponding relations . henceforth , remembering that we only have to consider states , we will assume that the relation corresponding to any @xmath1-legged diagram is of type @xmath126 . first we introduce some terminology . the _ composite _ of an @xmath19-tuple @xmath128 and an @xmath1-tuple @xmath129 is the @xmath127-tuple @xmath130 from @xmath131 . by the @xmath132__-remnant _ _ of an @xmath1-tuple we mean the @xmath133-tuple obtained by deleting its @xmath134 component . by the @xmath135__-remnant _ _ of an @xmath1-tuple ( where @xmath136 ) we mean the @xmath137-tuple obtained by deleting the @xmath138 component of its @xmath132-remnant ( or equivalently , deleting the @xmath139 component of its @xmath140-remnant ) . _ [ relationtree ] consider linking two diagrams , the first representing the relation @xmath141 the second representing the relation @xmath142 via a permutation @xmath143 , to form a new diagram as shown : @xmath144 the relation corresponding to this diagram is given by @xmath145 or , in less formal language , for every pair of a tuple from @xmath124 and a tuple from @xmath146 obeying the condition @xmath147 , we form composite of the @xmath134 remnant of the tuple from @xmath124 , and the @xmath138 remnant of the tuple from @xmath146 . _ _ [ relationloop ] given a diagram representing the relation @xmath141 , consider forming a new diagram by linking the @xmath134 and @xmath138 legs of the original diagram via a permutation @xmath143 . @xmath148 the relation corresponding to this diagram is given by : @xmath149 or , in less formal language , we take the @xmath135-remnant of every tuple for which @xmath150 . _ _ [ relationcapping ] consider linking two diagrams , the first representing the relation @xmath151 the second representing the relation @xmath152 via a permutation @xmath143 , to form a new diagram as shown : @xmath153 the relation corresponding to this diagram is given by : @xmath154 or , in less formal language , we take the @xmath134 remnant of every tuple for which @xmath155 . _ it is convenient to single out a particular sub - group of the @xmath108 permutation sub - group . this consists of the four permutations which do nt mix between the sets @xmath12 and @xmath13 : ( 1)(2)(3)(4 ) ( the identity ) , ( 12)(3)(4 ) , ( 1)(2)(34 ) and ( 12)(34 ) . we term these the _ phased _ permutations . all other permutations are termed _ unphased_. we single out this sub - group because the relations corresponding to * spek * diagrams generated from @xmath156 , @xmath157 , and the four phased permutations have a significantly simpler general form . we term such diagrams _ phased diagrams_. all other diagrams are termed _ unphased diagrams_. a morphism which corresponds to a phased diagram is termed a _ phased morphism _ , all other morphisms being _ unphased morphisms_. it is straightforward to see that any unphased * spek * diagram can be viewed as a collection of phased sub - diagrams linked together via unphased permutations . we refer to these sub - diagrams as _ zones_. furthermore , note that any permutation in @xmath108 can be written as a product of phased permutations and the unphased permutation ( 1)(3)(24 ) ( which we denote by @xmath158 ) . since a single phased permutation constitutes a phased zone with two legs , we can in fact view an unphased diagram as a collection of phased zones linked together by the permutation @xmath158 : @xmath159 here @xmath160 represents a phased sub - diagram and a square box represents an unphased permutations . note that such a diagram is not necessarily planar , i.e. it may involve crossing wires . we distinguish between _ external zones _ which have external legs ( e.g. @xmath161 , @xmath162 and @xmath163 in diagram [ generalunphasedeq ] ) , and _ internal zones _ all of whose legs are connected to other legs within the diagram ( e.g. @xmath164 and @xmath165 ) . to the external legs we associate an enumeration , such that legs from the same external zone appear consecutively . in the first stage of the proof we determine the general form of the relation corresponding to a _ phased diagram_. this stage itself splits into two phases : first we determine the general form of the morphisms of a new category * halfspek * , and then we show how to use this result to prove our main result in * spek*. in the second stage we draw on the results of the first to determine the general form of the relation corresponding to _ any _ * spek * diagram . we build up to the full theorem via a simplified case . for this we need a new category . [ defhalfspek ] the category * halfspek * is a subcategory of * frel*. it is defined inductively , as follows : * the objects of * halfspek * are the single - element set @xmath96 , the two element set @xmath166 , and its @xmath1-fold cartesian products @xmath167 . * the morphisms of * halfspek * are all those relations generated by composition , cartesian product and relational converse from the following generating relations : 1 . all permutations @xmath168 of the two element set . there are 2 such permutations , the identity and the operation @xmath169 which swaps the elements of @xmath170 . together they form the group @xmath171 . a relation @xmath172 defined by : @xmath173 3 . a relation @xmath174 @xcite @xmath175 is a basis structure in * frel * and hence also in * halfspek*. _ the existence of this basis structure came as a surprise to the authors . naively one might think ( as many working in the area of categorical quantum mechanics initially did ) that on a set @xmath110 in * frel * there is a single basis structure with @xmath84 given by @xmath176 for all @xmath177 . the ` basis vectors ' ( or copyable points in the language of @xcite ) are then the elements of this set . but this is not the case . there are many ` non - well - pointed ' basis structures such as @xmath175 for which the number of copyable points is less than the number of elements of the set . in related work , pavlovic has classified all basis structures in * frel * and evans et al . have identified the pairs of complementary basis structures ( in the sense of @xcite ) among these @xcite . _ next we determine the general form of the relations which constitute the morphisms of * halfspek * , to which the considerations made at the beginning of section [ aboutdiagsandstates ] also apply . we say that an element of @xmath167 has _ odd _ parity if it has an odd number of ` 1 ' elements , and that it has _ even _ parity if it has an even number of ` 1 ' elements . we will use @xmath143 to represent a particular parity , odd or even , and @xmath178 will represent the opposite parity . whether an odd - parity @xmath1-tuple has an odd or even number of ` 0 ' elements clearly depends on whether @xmath1 itself is odd or even . we could have chosen either 0 or 1 to play the role of labelling the parity ; we have chosen 1 since it will turn out to be more convenient later on . _ [ halfspekmorphisms ] the relation in * halfspek * corresponding to an @xmath1-legged * halfspek*-diagram is a subset of @xmath167 , consisting of all @xmath179 @xmath1-tuples of a certain parity , which depending on the diagram may either be even or odd : if the product of all the permutations appearing in the diagram is the identity , then parity is even , and if it is @xmath169 , then the parity is odd . _ we prove this result by induction on the number of generators @xmath180 required to construct the diagram . remember that we need only consider those diagrams whose corresponding relations are states . there is just one possible base case ( @xmath181 ) , a diagram composed purely of the generator @xmath182 for which @xmath183 : the corresponding state consists of the single 1-tuple ( 0 ) , which is indeed the unique 1-tuple of even parity . now consider a diagram @xmath184 built from @xmath180 generators with a corresponding state @xmath185 consisting all @xmath179 @xmath1-tuples of parity @xmath143 . it is easily seen that composing @xmath184 with either @xmath186 , @xmath187 or @xmath188 respectively yields a diagram whose corresponding state consists of all @xmath189 @xmath0-tuples , all @xmath190 @xmath191-tuples , and all @xmath189 @xmath0-tuples of parity @xmath143 ; and that composing with @xmath169 yields all @xmath179 @xmath1-tuples of parity @xmath178 . finally consider producing a disconnected diagram by laying the @xmath182 diagram alongside @xmath184 : it is easily seen that the corresponding state consists of @xmath190 @xmath191-tuples of parity @xmath143 . we want to apply our results on * halfspek * to the category of real interest , * spek*. to do this we first need to digress to discuss some structural features of relations . the category * rel * has another symmetric monoidal structure , namely the _ disjoint union _ or _ direct sum _ _ , denoted by @xmath192 . concretely , if we can partition a set @xmath53 into @xmath19 subsets @xmath193 , then we have @xmath194 , and recalling that a relation @xmath195 is a subset of @xmath116 , we can decompose @xmath124 into @xmath196 components of the form @xmath197 , such that @xmath198 . the relations @xmath197 are termed the _ components _ of @xmath124 _ with respect to partitions _ @xmath194 , @xmath199 . in category theoretic terms , this is _ biproduct _ _ , and there is a _ distributive law _ with respect to the cartesian product : @xmath200 for sets @xmath201 and relations @xmath202 . for @xmath194 , @xmath199 , @xmath203 and @xmath204 , and relations @xmath195 , @xmath205 and @xmath206 , we then have : @xmath207 for @xmath208 by distributivity we have : @xmath209 if for @xmath210 and @xmath211 the only non - empty components are @xmath212 and @xmath213 we call it _ parallel_. given parallel relations @xmath214 , @xmath215 and @xmath216 with respect to the partition @xmath208 , the relations : @xmath217 are all also easily seen to be parallel with respect to the same partition of @xmath53 . _ _ _ _ _ _ _ _ _ _ we can use these insights to make a connection between * halfspek * and * spek*. the generators of the phased morphisms of * spek * , i.e. @xmath218 , @xmath219 and the phased permutations on @xmath22 , are all parallel with respect to the following partition of @xmath220 . we conclude that all phased morphisms of * spek * are also parallel with respect to this partition . we refer to the two components of a phased * spek * morphism as its @xmath12-component and @xmath13-component . [ halfspekgenarespekgen ] the @xmath12-components of the generators of the phased morphisms of * spek * are simply the generators of * halfspek * with the elements of @xmath221 re - labelled according to @xmath222 , @xmath223 . similarly the @xmath13-components of the generators of the phased morphisms of * spek * are simply the generators of * halfspek * with the elements of @xmath221 re - labelled according to @xmath224 , @xmath225 . [ spekdis2halfspekd ] a state @xmath226 corresponding to a phased * spek * diagram @xmath184 is equal to the union of two states @xmath227 and @xmath228 . @xmath229 and @xmath230 are obtained by the following procedure . form a * halfspek * diagram @xmath231 by replacing every occurence of @xmath218 and @xmath219 in @xmath184 with @xmath232 and @xmath233 , and replacing every occurence of a permutation with its @xmath12 component , re - labelled as a @xmath234 permutation as described in proposition [ halfspekgenarespekgen ] . form a second * halfspek * diagram @xmath235 in the obvious analogous fashion using @xmath13 components of permutations . @xmath229 and @xmath230 are the states corresponding to @xmath231 and @xmath235 , once again under the re - labelling described in proposition [ halfspekgenarespekgen ] . note that @xmath231 and @xmath235 will appear identical as graphs , both to each other and to @xmath184 , but the labels on some of their permutations will differ . from proposition [ spekdis2halfspekd ] and theorem [ halfspekmorphisms ] now follows : [ phasedmorphisms ] a phased morphism in * spek * of type @xmath126 is a subset of @xmath5 , consisting of @xmath9 @xmath1-tuples , divided into two classes of equal number : * the first class consists of tuples of 1s and 2s , all of either odd or even parity . * the second class consists of tuples of 3s and 4s , again all of either odd or even parity . note that we are adopting the convention that tuples of the first class have odd parity if they have an odd number of 2s , even parity if they have an even number of 2s . tuples of the first class have odd parity if they have an odd number of 4s , even parity if they have an even number of 4s . recall from section [ proofstructuresec ] that any unphased diagram can be viewed as a collection of phased zones linked together by the permutation @xmath158 , ( 1)(3)(24 ) . we also enumerated the external legs . [ externalnonphasedth ] the relation in * spek * corresponding to an @xmath1-legged * spek*-diagram with @xmath19 zones , none of which is internal , is a subset @xmath185 of @xmath5 with the following form : 1 . it contains @xmath9 @xmath1-tuples ; each entry corresponding to an external leg . all those entries corresponding to a given zone of the diagram are termed a _ zone _ of the tuple ( whether we are referring to a zone of a diagram , or zone of a tuple should be clear from the context ) . each zone of the tuple has a well - defined _ type _ ( components either all 1 or 2 , or all 3 or 4 ) and _ parity _ ( as defined for phased relations ) . 3 . a _ block _ @xmath115 is a subset of @xmath185 such that the @xmath134 zone of all @xmath1-tuples in @xmath115 has the same parity and type . the sequence of types and parities of each zone is called the _ signature _ of the block . the @xmath9 tuples of @xmath185 are partitioned into @xmath236 equally sized blocks each with a unique signature . each of the @xmath236 blocks has a different type signature - these exhaust all possible type signatures . the parity signature of a block is the following simple function of its type signature : @xmath237 here @xmath238 and @xmath239 are boolean variables representing the parity and type of the @xmath134 zone . we adopt the convention that an odd parity is represented by 0 and an even parity by 1 , whilst the type @xmath12 is represented by 0 and the type @xmath13 by 1 . the set adj(@xmath240 ) consists of the zones directly adjacent to the @xmath134 zone . @xmath241 denotes the parity of the type @xmath239 tuples in the relation corresponding to the @xmath134 zone seen as an independent phased diagram ( recall theorem [ phasedmorphisms ] ) . [ internalnonphasedth ] the relation in * spek * corresponding to an @xmath1-legged * spek*-diagram with @xmath19 zones , of which @xmath242 are external is either : * a subset , @xmath185 of @xmath5 which satisfies conditions ( 1 ) and ( 2 ) of theorem [ externalnonphasedth ] and which is partitioned into @xmath243 blocks . the signatures of these blocks are determined as follows : 1 . begin with the state corresponding to the diagram obtained by adding an external leg to every internal zone ( i.e. we will have @xmath236 blocks , each with @xmath19 zones , exhausting all possible type signatures ) . 2 . eliminate all blocks whose type signatures do not satisfy the following constraints , one for each internal zone : @xmath244 where @xmath240 is the label of the internal zone , and @xmath245 labels its adjacent zones . 3 . finally from each block delete the zones corresponding to internal zones . * _ or _ it is equal to the empty set , @xmath246 . this second possibility occurs iff the constraints in equation [ typeconstrainteq ] are inconsistent . a simple counting argument shows that within each block , every tuple with the correct type and parity signature occurs . theorems [ externalnonphasedth ] and [ internalnonphasedth ] thus completely characterise the state corresponding to any * spek*-diagram . the input data is the shape of the diagram , which determines _ adj(i ) _ , and the ` intrinsic parities ' @xmath241 of each zone . futhermore , note from theorem [ phasedmorphisms ] that , as we would expect , the general form of phased morphisms is a special case of the form described above , with @xmath247 . _ we now give an example of theorem [ externalnonphasedth ] in action . consider the following schematic**spek**-diagram ( circles simply denote zones ) , which has three zones , all external ( @xmath248 ) and five external legs ( @xmath249 ) : @xmath250 where the labels by each zone denote the intrinsic parities of that zone ( i.e. the parity of the type-12 tuples and the parity of the type-34 tuples , recall theorem [ phasedmorphisms ] ) . we conclude from theorem [ externalnonphasedth ] that the state corresponding to this diagram will consist of @xmath251 5-tuples each with 3 zones , and that these will be partitioned into @xmath252 equally sized blocks . every combination of types appears exactly once amongst these blocks , and the parity signatures are easily determined from equation [ typeparityeq ] as follows . first we note , from the intrinsic parities that @xmath253 , @xmath254 and @xmath255 . we then see that equation [ typeparityeq ] here is essentially three equations : @xmath256 from this we can establish the signatures of the eight blocks : @xmath257 the first zone will have two elements , the second zone one element and the third zone two elements . within each block every possible collection of tuples consistent with the signature will appear meaning that each block will consist of four tuples . _ _ we go on to give an example of theorem [ internalnonphasedth ] . consider the following diagram : it is identical to the diagram in the previous example except that the second zone has been internalised . @xmath258 this diagram has two external zones @xmath259 and four external legs @xmath260 ; thus we expect the corresponding state to consist of @xmath261 4-tuples , each with 2 zones , partitioned into @xmath262 blocks . we now determine the signatures of the blocks applying theorem [ internalnonphasedth ] . according to step ( 1 ) we begin with the blocks from the previous example . step ( 2 ) requires that we eliminate all blocks whose types do not satisfy the constraint @xmath263 . this leaves : @xmath264 finally we delete the second zone from each block , leaving : @xmath265 _ the proofs of theorems [ externalnonphasedth ] and [ internalnonphasedth ] proceed as inductions over the process of building up a diagram by linking together phased zones via the permutation @xmath158 . in the course of this process it is clearly possible for an external zone to become an internal zone , as its last external leg is linked to some other zone - we refer to this as _ internalising _ a zone . it turns out that internalisation of a zone complicates the inductive proof . to get around this we do the induction in two stages . in the first stage we build up a diagram identical to the one we are aiming for , _ except _ that every zone that should be internal is given a single external leg in the following fashion . suppose we need to link together two zones via a permutation , and this will result in the internalisation of the left hand zone : @xmath266 we instead link the left hand zone to the permutation via a @xmath218 morphism . the @xmath218 morphism becomes part of the original zone , and provides it with an external leg : @xmath267 in the second stage we cap off all these extra external legs with the @xmath219 morphism . since @xmath218 and @xmath219 constitute a basis structure the result is the diagram which we are aiming for : @xmath268 we now prove theorem [ externalnonphasedth ] . the proof uses an induction over the process of building up a diagram . this building up can be split into two phases : firstly we connect all of the zones which will appear in the final diagram into a ` tree - like ' structure with no closed loops , secondly we close up any loops necessary to yield the desired diagram . henceforth we will represent the signature of a tuple with @xmath19 zones as @xmath269 where @xmath238 is the parity of the @xmath134 zone , and @xmath239 is its type . again , if @xmath143 is a parity , @xmath178 indicates the opposite parity , and likewise if @xmath270 is a type , @xmath271 represents the other type . [ unphasedtreelemma ] consider an @xmath1-legged non - phased diagram @xmath161 with @xmath19 zones . suppose the corresponding state @xmath272 satisfies the conditions in theorem [ externalnonphasedth ] . now consider linking the @xmath134 leg of @xmath161 to the @xmath138 leg of an @xmath273-legged phased diagram @xmath162 ( with corresponding state @xmath274 ) , via @xmath158 , to create an @xmath275-legged diagram @xmath164 with @xmath276 external zones . we will assume that the @xmath134 leg of @xmath161 lies within its @xmath277 external zone . the state @xmath278 corresponding to @xmath164 also satisfies the conditions in theorem [ externalnonphasedth ] . by the 1-@xmath134-remnants of @xmath279 we mean the @xmath134-remnants of those tuples in @xmath279 with a 1 in the @xmath134 position . we define the 2- , 3- , and 4-@xmath134-remnants similarly . by proposition [ relationtree ] the elements of @xmath280 comprise all the possible composites of the @xmath281-@xmath134-remnants of @xmath279 and the @xmath282-@xmath138-remnants of @xmath274 , where @xmath283 . it is clear that the zone structure of the tuples of @xmath279 is inherited by these composites , and that the @xmath282-@xmath138-remnants of @xmath274 constitute an additional zone within the composites - we conventionally consider this to be the @xmath284 zone . thus the tuples of @xmath280 satisfy condition ( 2 ) . consider a block @xmath285 , with signature @xmath286 , in which for definiteness we assume that @xmath287 ( the argument runs entirely analogously if @xmath288 ) . the composites of the 1-@xmath134-remnants of @xmath115 and the @xmath289-@xmath138-remnants of @xmath274 all have the same signature , @xmath290 , and constitute a block @xmath291 . likewise the composites of the 2-@xmath134-remnants of @xmath115 and the @xmath292-@xmath138-remnants of @xmath274 constitute a block @xmath293 of signature @xmath294 @xmath295 . thus each ` parent ' block in @xmath279 gives rise to two ` progeny ' blocks in @xmath280 . by hypothesis , each block @xmath296 has a unique type signature , thus the progeny blocks derived from different parent blocks are distinct . thus @xmath280 is partitioned into @xmath297 blocks , thus satisfying condition ( 3 ) . it is also clear that if all possible type signatures are represented by the @xmath236 blocks of @xmath279 then this is also true for the @xmath297 blocks of @xmath280 , and thus that @xmath280 satisfies condition ( 1 ) . note that @xmath115 consists of @xmath298 @xmath1-tuples , and will have @xmath299 1-@xmath134-remnants and a similar number of 2-@xmath134-remnants , all of which will be distinct . similarly @xmath274 will have @xmath300 @xmath289-@xmath134-remnants and a similar number of 4-@xmath134-remnants , again all distinct . thus , both @xmath301 and @xmath302 will consist of @xmath303 tuples . this holds for all blocks @xmath285 , of which there are @xmath236 . thus in total @xmath280 consists of @xmath304 tuples , and so satisfies condition ( 4 ) . finally we turn to condition ( 5 ) . recall from above that a parent block of signature @xmath305 @xmath306 yields two progeny blocks , of signatures @xmath307 @xmath308 and @xmath309 . note that those progeny blocks for which the @xmath277 zone and its new adjacent zone have different types exhibit a parity flip on the @xmath277 zone , relative to the parent block , and on the new adjacent zone , relative to its ` intrinsic parity ' . no such flip occurs if the zones have the same type . note that the term @xmath310 is equal to 0 if @xmath311 and 1 if @xmath312 . if condition ( 2 ) holds for @xmath279 then the correct parity signature for the blocks of @xmath280 can be obtained simply by adding the term @xmath313 to the @xmath314 and @xmath315 equations ( equation [ typeparityeq ] ) . thus we conclude that condition ( 5 ) will hold for @xmath280 as well . [ unphasedlooplemma ] consider an @xmath1-legged diagram @xmath184 with @xmath19 external zones . suppose the corresponding state @xmath185 satisfies the conditions in theorem [ externalnonphasedth ] . now consider forming a new @xmath137-legged diagram @xmath316 , with corresponding state @xmath317 , by linking the @xmath134 leg of @xmath184 ( in the @xmath277 zone of @xmath184 ) , to the @xmath138 leg ( in the @xmath318 zone ) , via @xmath158 . @xmath317 also satisfies the conditions in theorem [ externalnonphasedth ] . by the @xmath319-@xmath320__-remnants _ _ of a set of tuples we mean the @xmath320-remnants of all those tuples with @xmath281 in the @xmath134 position and @xmath321 in the @xmath138 position . from proposition [ relationloop ] , @xmath317 consists of the @xmath322-@xmath320-remnants of @xmath185 . the zone structure of @xmath185 is clearly inherited by these remnants , thus the tuples of @xmath317 satisfy condition ( 2 ) . consider a block @xmath323 with signature @xmath324 @xmath325 . it is straightforward to see that only one quarter of the tuples in this block have @xmath320-remnants which are @xmath322-@xmath320-remnants . all the @xmath322-@xmath320-remnants have the same signature : @xmath326 if @xmath327 and @xmath328 @xmath329 if @xmath330 . thus each ` parent ' block in @xmath185 gives rise to one ` progeny ' block in @xmath317 , with one quarter as many tuples . the type signatures of the progeny blocks are identical to those of the parent blocks ; by hypothesis each parent block had a different type signature and so the progeny blocks deriving from different parent blocks are all distinct . thus @xmath317 is partitioned into @xmath236 blocks , each containing @xmath331 tuples , and so satisfies conditions ( 1 ) and ( 3 ) . since the type signatures of progeny and parent blocks are identical , if @xmath185 satisfies condition ( 4 ) , so will @xmath317 . we now turn to condition ( 5 ) . closing a loop between the @xmath277 and @xmath318 zones means that they now become adjacent to each other . in the previous paragraph we saw that a progeny block for which the @xmath277 and @xmath318 zones have different types exhibits a parity flip on both these zones . using similar reasoning as in the previous lemma we conclude that , if condition ( 2 ) holds for @xmath185 , the correct parity signature for the blocks of @xmath317 can be obtained simply by adding the term @xmath332 to the @xmath314 and @xmath333 equations ( equation [ typeparityeq ] ) . thus we conclude that condition ( 5 ) will hold for @xmath317 as well . we can now prove * theorem [ externalnonphasedth ] * : by induction . the base case is a diagram consisting of a single zone , and it is clear from theorem [ phasedmorphisms ] that this satisfies all the conditions . any other diagram is built up via two inductive steps : linking a new phased zone onto the existing diagram and closing up internal loops within the diagram . lemmas [ unphasedtreelemma ] and [ unphasedlooplemma ] respectively show that if a diagram satisfied the conditions prior to either of these steps , the resulting new diagram will also satisfy the conditions . we now address the issue of internalising zones . recall that this step involves capping off external legs with the @xmath219 relation . throughout this section @xmath184 will denote a diagram with no internal zones , and @xmath316 will denote the diagram obtained by internalising some of @xmath184 s zones . the corresponding states will be @xmath185 and @xmath317 . suppose we obtain @xmath316 by capping the @xmath277 external leg of @xmath184 with the @xmath219 morphism . from lemma [ relationcapping ] we conclude that @xmath317 consists of the 1-@xmath277- and 3-@xmath277-remnants of @xmath185 . suppose that in going from @xmath184 to @xmath316 we internalised the @xmath134 zone of @xmath184 . from the proposition above we deduce that each block @xmath323 for which @xmath334 will give rise to one progeny block @xmath335 with the same number of tuples as @xmath115 , while all those blocks for which @xmath336 will give rise to no progeny blocks . from this we can conclude that those blocks which do give rise to progeny blocks satisfy a constraint on their type signatures , derived from setting @xmath334 in equation [ typeparityeq ] . depending on the form of @xmath241 , and whether the number of zones adjacent to the @xmath134 is odd or even , this constraint takes one of four forms : @xmath337 @xmath338 we describe the constraints in ( [ type0eq ] ) as _ type-0 constraints _ , and those in ( [ type1eq ] ) as _ type-1 constraints_. suppose @xmath184 has @xmath1 external legs and @xmath19 external zones . suppose that in going to @xmath316 we internalise @xmath339 of its zones . _ each internalisation gives rise to a corresponding constraint . _ there are now two possibilities : 1 . not all of the constraints are consistent . in this case none of the blocks in @xmath185 satisfy all of the constraints , and none of them will give rise to progeny blocks . thus @xmath340 . all @xmath339 constraints are consistent , and of these @xmath341 are linearly independent ( this essentially means that @xmath342 of the constraints can be derived from the remaining @xmath341 ) . each independent constraint reduces the number of blocks which can give rise to progeny by one half . thus only @xmath343 of the blocks in @xmath185 give rise to progeny blocks in @xmath317 , and @xmath317 can have at most @xmath343 blocks - this maximum is attained if all of the progeny blocks are distinct . [ duplicationlemma ] the following are equivalent : 1 . the constraints are consistent and there are @xmath341 linearly independent constraints . the @xmath343 blocks in @xmath185 which can give rise to progeny blocks in @xmath317 are partitioned into @xmath344 sets , each consisting of @xmath345 blocks which all yield identical progeny blocks . thus in total there are @xmath344 distinct progeny blocks . for brevity we will describe this as @xmath346-_fold duplication _ of progeny blocks . the proof of this lemma requires a number of preliminary definitions . the _ iz - set _ is the set of zones which are internalised in going from @xmath184 to @xmath316 . the _ non - internalised adjacent zones ( niazs ) _ of an internalised zone are all the zones adjacent to it which are not themselves members of the iz - set adjacency closure set ( acs ) _ is a subset of the iz - set with the minimal number of elements such that the disjoint union of the niazs of each element contains each niaz an even number of times . _ consider the following seven zone diagram ( external legs are suppressed for clarity ) . the filled - in zones are those which we internalise . @xmath347 the iz - set is @xmath348 . the niazs for 1 are @xmath349 for 3 are @xmath350 , for 6 are @xmath351 and for 7 are @xmath352 . zones 3 , 6 and 7 together constitute an acs . zone 1 is not part of any acs . _ given a set @xmath146 of zones in @xmath184 , and a block @xmath323 , the _ @xmath146-mirror of @xmath115 _ , @xmath353 is the block with the same type signature as @xmath115 except on the zones in @xmath146 , where the types are opposite . [ dupmirrorprop ] suppose @xmath184 has an acs @xmath124 . now , so long as the blocks @xmath354 ( i.e. a block and its @xmath124-mirror ) both yield progeny blocks in @xmath317 , these progeny blocks will be identical . conversely , if any two blocks @xmath355 yield identical progeny blocks in @xmath317 , they must be mirrored with respect to some acs in @xmath184 . for two blocks in @xmath185 to give identical progeny blocks in @xmath317 they must have identical type and parity on every zone which is not internalised . note from equation [ typeparityeq ] that two blocks in @xmath185 which are type mirrored on a single zone will otherwise differ only in parities on all the zones adjacent to this zone . now consider two blocks which are type - mirrored on a set of zones @xmath124 : in the case where @xmath124 constitutes an acs all the parity flips predicted by equation [ typeparityeq ] cancel one another out on the zones which will still be visible in @xmath317 . [ mirrorconstraintprop ] suppose @xmath184 has an acs @xmath124 whose member zones have corresponding constraints which satisfy the following condition : those zones with an odd number of adjacent zones within the iz - set have type-1 constraints , while those zones with an even number of adjacent zones within the iz - set have type-0 constraints . then if a block @xmath323 satisfies the constraints and gives rise to progeny blocks in @xmath317 , so does its @xmath124-mirror . conversely , if a block @xmath323 and its mirror with respect to some acs @xmath124 in @xmath184 both give rise to progeny blocks , the constraints corresponding to the zones of @xmath124 must satisfy the condition above . if a constraint contains an even number of terms relating to zones from a set @xmath124 then given a block @xmath323 either ( i ) both @xmath115 and @xmath356 satisfy the constraint ( ii ) neither @xmath115 nor @xmath356 satisfy the constraint . conversely , if both @xmath115 and @xmath356 satisfy a constraint , it must contain an even number of terms relating to zones from @xmath124 . [ constraintlindepprop ] suppose @xmath184 has an acs @xmath124 whose member zones have corresponding constraints which satisfy the condition in the previous proposition . then the constraints together form a linearly dependent set . the converse is also true . if @xmath124 is an acs and the condition on constraints is satisfied then each term appears in the constraints an even number of times altogether . summing all the constraints together then results in all the terms cancelling out , yielding the single equation 0=0 . this is a necessary and sufficient condition for the constraints to be linearly dependent . we can now prove lemma [ duplicationlemma ] . from propositions [ dupmirrorprop ] and [ mirrorconstraintprop ] we conclude that for @xmath1-fold duplication to take place the iz - set must contain @xmath1 acss and that the constraints must satisfy the condition in proposition [ mirrorconstraintprop ] . the converse is also clearly true . every linearly dependent set amongst the constraints corresponding to the internalised zones reduces the total number of linearly independent constraints by one . from proposition [ constraintlindepprop ] we conclude that for there to be @xmath1 linearly dependent sets the iz - set must contain @xmath1 acss and that the constraints must satisfy the condition in proposition [ mirrorconstraintprop ] . the converse is also clearly true . thus we conclude that both statements in the lemma are equivalent to a third statement : the iz - set contains @xmath342 acss , and the constraints corresponding to the internalised zones satisfy the condition of proposition [ mirrorconstraintprop ] . given a diagram @xmath184 without internal zones which satisfies the conditions in theorem [ externalnonphasedth ] , a diagram @xmath316 with internal zones formed by capping off external legs of @xmath184 with @xmath219 morphisms will satisfy the conditions in theorem [ internalnonphasedth ] . * theorem [ internalnonphasedth ] * follows as a straightforward corollary . _ [ mspeknumbers ] all * mspek * morphisms of type @xmath126 are subsets of @xmath5 containing @xmath357 or @xmath8 @xmath1-tuples . _ any * mspek * diagram @xmath316 can be obtained from a * spek * diagram @xmath184 simply by capping one or more legs of @xmath184 with the morphism @xmath358 . suppose we obtain @xmath316 by capping a single external leg of @xmath184 ( the @xmath134 , say ) with @xmath358 . the state @xmath317 corresponding to @xmath316 consists of the @xmath134-remnants of @xmath185 , the state corresponding to @xmath184 . suppose @xmath184 has @xmath1 external legs : since @xmath184 is a * spek*-diagram @xmath185 consists of @xmath9 tuples . then , unless some of the @xmath134-remnants of @xmath185 are identical , @xmath317 will also consist of @xmath9 tuples , despite only having @xmath359 external legs . furthermore it is clear that either all the @xmath134-remnants of @xmath185 are distinct , or @xmath185 is partitioned into pairs , the elements of which yield identical @xmath134-remnants , meaning that the addition of each @xmath358 cap either halves the number of tuples or leaves it unchanged . we know that the epistemic states of the toy theory are subsets of the sets @xmath5 , and that the transformations on these states are relations between these sets . thus we can see immediately that the toy theory s process category must be some sub - category of * frel * , restricted to the objects @xmath5 . furthermore we know that it can not be the full sub - category restricted to these objects , since some subsets of @xmath5 clearly violate the knowledge balance principle . we will show now that ( a strong candidate for ) the process category for the toy theory in its entirety is * mspek * , while if we restrict the toy theory to states of maximal knowledge ( consistent with the knowledge balance principle ) , the process category is * spek*. the morphisms of the process category of the toy theory are closed under composition , cartesian product and relational converse . there is no feature of the toy theory which would put any restrictions on which operations could be composed , so we expect the states and transformations to be closed under composition . since the cartesian product is used by the toy theory to represent composite systems we also expect the states and transformations to be closed under cartesian product . every epistemic state corresponds to an outcome for at least one measurement ( measurements correspond to asking as many questions as possible from canonical sets , epistemic states correspond to the answers ) . recalling the discussion of measurement in section 2 , we see that given a state @xmath226 the disturbance resulting from the corresponding measurement outcome can be decomposed as @xmath360 , where @xmath361 is the relational converse of @xmath185 . thus we expect the relational converse of each state also to feature in the physical category of the theory . the toy theory state corresponding to the subset @xmath362 along with its relational converse are then easily seen to constitute a compact structure on @xmath22 . we thus have map - state duality , and it is straightforward then to show that if states are closed under relational converse , so is any morphism in the physical category . note that this point sharpens our discussion about the consistency of the toy theory , in section 2 . if the states and transformations which spekkens has derived for up to three systems , under the operations of composition , cartesian product and relational converse , yield states which violate the knowledge balance principle , then the theory as presented is inconsistent . all of the generating morphisms of * mspek * are states or transformations of the toy theory , or can be derived from them by composition , cartesian product or relational converse . the only generator for which this is less than obvious is @xmath218 . this is formed by composing spekkens s ghz - like state ( see section v of @xcite ) with the relational converse of the state @xmath363 defined in the proof above . all of the states and transformations derived by spekkens in his original paper @xcite are morphisms of * mspek*. when we restrict to states of maximal knowledge all of the states and transformations are morphisms of * spek*. by inspection of @xcite . [ spekisclosure ] * mspek * is the minimal closure under composition , cartesian product and relational converse of the states and transformations described in @xcite . * spek * is the minimal closure under these operations of the states of @xcite corresponding to maximal knowledge and the transformations which preserve them . [ spekmspekarekbp ] all states @xmath364 of * mspek * and * spek * satisfy the knowledge balance principle on the system corresponding to @xmath5 viewed as one complete system . all those of * spek * satisfy the principle maximally . recall that the knowledge balance principle requires that we can know the answer to at most half of a canonical question set . a system with @xmath1 elementary components has @xmath8 ontic states . a canonical set for such a system consists of @xmath365 questions , each answer to a question halving the number of possibilities for the ontic state . thus , we know the answer to @xmath19 such questions ( @xmath366 ) , iff our epistemic state is a subset of @xmath5 with @xmath367 elements . we conclude from theorem [ mspeknumbers ] that all states of * mspek * satisfy the knowledge balance principle on the system as a whole . we conclude that all states of * spek * correspond to the maximum knowledge about the system as a whole consistent with the knowledge balance principle . [ subsysarekbptoo ] all states @xmath364 of * mspek * and * spek * satisfy the knowledge balance principle on every subsystem of the system corresponding to @xmath5 . given an epistemic state @xmath226 of a composite system with @xmath1 elementary components , the ` marginal ' state on some subsystem is obtained from @xmath185 by deleting from the tuples of @xmath185 the components corresponding to the elementary systems which are not part of the subsystem of interest . suppose this epistemic state corresponds to a * spek * or * mspek * diagram , @xmath184 . the elementary systems which are not part of the subsystem correspond to a certain collection of external legs of @xmath184 , and , by lemma [ relationcapping ] , if we cap these with the * mspek * generator @xmath358 , the effect on the state @xmath185 is exactly as just described . composing a * spek * or * mspek * morphism with @xmath358 yields some morphism of * mspek * , which by proposition [ spekmspekarekbp ] satisfies the knowledge balance principle . from corollary [ spekisclosure ] and propositions [ spekmspekarekbp ] and [ subsysarekbptoo ] we reach two key conclusions : * the states and transformations derived by spekkens in @xcite for systems of up to three components are all consistent with the knowledge balance principle . * the process category of the toy theory must , at least , contain all of the morphisms of * mspek*. the second conclusion begs the question , could * mspek * be a strict sub - category of the process category of the toy theory i.e. could the toy theory contain operations not contained in * mspek * ? it is difficult to answer this question , since , as discussed at the end of section 2 it is not clear what the rigorous definition of the toy theory is , or whether there is an unambiguous way to extend it beyond three systems . certainly , * mspek * is the process category of a theory which coincides with spekkens s theory up to the case of three qubits , and whose states and transformations are bound to satisfy the three rules of section 2 ( the first two rules by propositions [ spekmspekarekbp ] and [ subsysarekbptoo ] , and the third simply by its definition as the closure under composition of a set of generators ) . it is in this sense that we earlier remarked that * mspek * is a strong candidate for the process category of the toy theory . we achieved our goal stated in the abstract , that is , to provide a rigorous mathematical description of spekkens toy theory , which proves its consistency . this was established both in terms of generators for a dagger symmetric monoidal subcategory of * frel * , consisting of symmetries for the elementary system and a basis structure ( and nothing more ! ) , as well as in terms of an explicit description of these relations as in theorems [ internalnonphasedth ] , [ externalnonphasedth ] and [ mspeknumbers ] . this description meanwhile already has proved to be of great use , for example , in pinpointing what the essential structural difference is between the toy theory and the relevant fragment of quantum theory . in joint work with spekkens in @xcite we showed that the key difference between the toy theory and relevant fragment of quantum theory is the _ phase group _ _ , a group that by pure abstract nonsense can be attributed to each basis structure . in the case of the toy theory this phase group is @xmath368 while in the case of the relevant fragment of quantum theory it is @xmath369 . one can then show that it is this difference that causes the toy theory to be local , while the relevant fragment of quantum theory is non - local . _ _ _ in this context , one may wonder wether there is a general categorical construction which would turn a ` local theory ' like * spek * into a non - local one . we also expect that the construction in this paper can be fairly straightforwardly extended beyond qubit theories , for example to qutrits @xcite . s. abramsky ( 2009 ) _ no - cloning in categorical quantum mechanics__. in : semantic techniques for quantum computation , i. mackie and s. gay ( eds ) , pages 128 , cambridge university press . _ s. abramsky and b. coecke ( 2004 ) _ a categorical semantics of quantum protocols__. in : proceedings of 19th ieee conference on logic in computer science , pages 415425 . ieee press . arxiv : quant - ph/0402130 . revised version ( 2009 ) : _ categorical quantum mechanics__. in : handbook of quantum logic and quantum structures , k. engesser , d. m. gabbay and d. lehmann ( eds ) , pages 261323 . _ _ _ _ _ b. coecke and r. duncan ( 2008 ) _ interacting quantum observables__. in : proceedings of the 35th international colloquium on automata , languages and programming ( icalp ) , pages 298310 , lecture notes in computer science 5126 , springer - verlag . extended version ( 2011 ) : _ interacting quantum observables : categorical algebra and diagrammatics__. new journal of physics * 13 * , 043016 . arxiv:0906.4725 _ _ _ _ _ _ b. coecke and e. o. paquette ( 2011 ) _ categories for the practicing physicist__. in : new structures for physics , b. coecke ( ed ) , pages 167271 . lecture notes in physics , springer - verlag . _ b. coecke and d. pavlovic ( 2007 ) _ quantum measurements without sums__. in : mathematics of quantum computing and technology , g. chen , l. kauffman and s. lamonaco ( eds ) , pages 567604 . taylor and francis . arxiv : quant - ph/0608035 .

the interest in studying record events in a general sense is evident as they by definition tend to be rare events of a special nature . as examples of record - breaking events most readers might think about athletic records like a world record in 100 m dash , or long jump @xcite . however records also play an important role in other naturally occurring and man - made phenomena . in particular , climatic records attract much attention and are frequently mentioned in the media in the context of global warming . the statistics of record temperatures , precipitation amounts and floods have been adressed extensively in recent work @xcite . also in evolutionary biology record events have a certain relevance as was pointed out in @xcite , and they play an important role in the dynamics of disordered systems like spin glasses and dirty superconductors @xcite . last but not least records can be studied in the financial sciences , e.g. when considering record - breaking values or changes in stock prices @xcite . in all these examples not only the rate of record events is of interest but also their correlations : given that a record flood just occurred , can one relax because it will take a while before the next record flood is to be expected ? based on our previous work in @xcite , this question will be addressed in the present paper . most previous studies of record events and record values @xcite have been restricted to the case where the underlying random variables ( rv s ) @xmath0 are independently drawn from a common distribution @xmath1 ( i.i.d . rv s ) . in this case , record events can be shown to be stochastically independent , i.e. the knowledge about previous record events does not indicate whether a future record event is more or less likely . however in many applications , the underlying random variables are not i.i.d . for example , in recent work majumdar and ziff computed the statistics of records in a symmetric random walk @xcite ( see also @xcite ) . here , record events are obviously not independent : in a random walk the probability for a second record to occur immediately after after a previous record is simply @xmath2 , whereas the unconditional probability for a record in the @xmath3th step is @xmath4 @xcite . in this article we consider correlations between records in sequences of rv s that are independent but not identically distributed . an important example is the statistics of hot or cold records in weather data . in @xcite the occurrence of record - breaking temperatures in european and american weather recordings was analyzed by modeling daily temperature measurements as uncorrelated rv s with a linear drift . it was shown that the increase of hot records and the decrease of cold records due to global warming can be described and quantified by this model . we will refer to this particular example of non - i.i.d . rv s , which was originally introduced in the 1980 s in the context of athletic records @xcite , as the linear drift model ( ldm ) . the precise definition of the ldm is as follows : for a positive constant @xmath5 , each random variable @xmath6 has an i.i.d . part @xmath7 with density @xmath1 and an additive linear drift such that @xmath8 in other words , while the shape of the density @xmath9 of the @xmath10 rv is invariant , its overall position is shifted by a constant amount @xmath5 each time a new rv is drawn . as was already noted in @xcite , the record events under this model are not in general independent ( see also @xcite for the related case of rv s with changing variance ) . below we shall see that the sign and magnitude of correlations depend in a highly non - trivial way on the underlying probability density @xmath1 . to quantify the correlations , consider the joint probability @xmath11 of a record occurring in the @xmath12 as well as in the @xmath13 step , @xmath14.\end{aligned}\ ] ] to normalize this joint probability , we divide by the probabilities @xmath15 and @xmath16 of records in the @xmath13 and @xmath17 steps , obtaining the key quantity under consideration throughout this paper , @xmath18 correlations between record events in the @xmath19 and @xmath20 rv for different distributions of the underlying i.i.d . part . for the gumbel case , the well known stochastic independence of record events is found , while simulations for the other distributions show that generally the records are correlated . note in particular the curve for the symmetric lvy - stable distribution with index @xmath21 , which indicates an attraction between record events . ] the simulations shown in fig . [ rec_dep ] illustrate two known results , the fact that record events for i.i.d . rvs are stochastically independent @xcite ( all curves meet at unity for @xmath22 ) , and that for the ldm with gumbel distributed i.i.d . part , record events are stochastically independent for all values of @xmath5 @xcite . however the curves for other distributions show that stochastic independence does not hold in general . for some choices of the underlying probability density @xmath1 record events tend to repel each other ( @xmath23 ) , wheras in other cases the correlations are positive , i.e. record events attract each other ( @xmath24 ) . in particular this last , somewhat counterintuitive numerical finding triggered our interest in this problem . in the next section we will derive a general expression for the correlations between record events as quantified by @xmath25 . for this we employ methods similar to those used in @xcite . in section [ examples ] , we provide some explicit examples illustrating the rich and counterintuitive behavior of the correlations . in particular , we discuss the behavior of correlations with respect to the three classes of extreme value statistics @xcite . we find that while in the weibull class of distributions with finite support the correlations are generally negative , positive correlations mainly arise when the rv s are in the frecht class of distributions with power law tails . in the intermediate gumbel class of distributions with exponential - like tails correlations can be both positive and negative . finally in section [ conclusions ] , we summarize our results and discuss possible applications and open questions . consider a sequence @xmath26 of independent but not necessarily identically distributed rv s , with @xmath27 denoting the probability density of @xmath28 . the probability that the maximum of the first @xmath3 rv s is smaller than a given value @xmath29 then factorizes into the probabilities of the rv s individually being less than @xmath29 , @xmath30=\prod_{l=1}^{n}\int^x{\mathrm{d}}x ' f_{l}(x')=\prod_{l=1}^{n}f_l(x),\ ] ] where @xmath31 denotes the cumulative distribution function for the density @xmath27 , and the omitted lower ( upper ) integral boundary is understood to be the lower ( upper ) bound of the support of @xmath32 respectively . throughout we assume distribution functions to be continuous . using eq . [ factor ] the probability @xmath15 that the @xmath13 rv is a record can be written as @xmath33 and the joint probability that both @xmath34 and @xmath35 are records is @xmath36 for the ldm we have @xmath37 and @xmath38 , where @xmath1 and @xmath39 are the density and distribution function of the i.i.d . part @xmath28 in ( [ ldm ] ) . it then follows that @xcite @xmath40 and similarly @xmath41 the expressions ( [ joint ] ) and ( [ joint_ldm ] ) allow us to verify two known results : both for the i.i.d . case ( @xmath22 ) and for the ldm with gumbel distributed i.i.d . part @xmath28 , record events are independent . we start with the i.i.d . first note that the probability that the @xmath13 rv is a record is the probability that it is the largest among @xmath3 equivalent rv s , so by symmetry @xmath42 . for the i.i.d . case , the subscripts in ( [ joint ] ) can be dropped . using the substitution @xmath43 and @xmath44 , one obtains @xmath45 it is possible to insert an arbitrary spacing between the two record events and still obtain this factorization to yield @xmath46 . similarly , one can consider more than one record event . the factorization property implies that in the i.i.d . case , record events are stochastically independent @xcite . next we consider the ldm with gumbel distributed i.i.d . part , @xmath47 and @xmath48 . using that in this case @xcite @xmath49 the substitution @xmath43 used above yields @xcite @xmath50 with @xmath51 , and @xmath52 again one can introduce an arbitrary number of spaces and intermediate records and still have the corresponding joint probability factorize . this implies that for the ldm with gumbel distributed i.i.d . part , as for arbitrary distributions without drift , record events are stochastically independent . as shown above , for the i.i.d . case @xmath53 , record events are independent . on the other hand , for very large drift , almost every rv is a record , and thus trivially @xmath54 for @xmath55 ( see fig . [ rec_dep ] ) . the behavior of @xmath56 in between these two limits is not as easily characterized . in particular , while for some probability densities the records seem to repel each other for intermediate @xmath5 ( @xmath57 ) , for other probability densities the records seem to _ attract _ ( @xmath24 ) . correlations between record events as the number of rv s @xmath3 increases . for the i.i.d . part drawn from a gaussian distribution with variance @xmath58 , the correlations are negative and for sufficiently large @xmath3 tend to a limiting form . for the symmetric lvy distributed i.i.d . part , the correlations are positive and increase with @xmath3 . ] figure [ no_fse ] shows examples for both types of behavior , illustrating in addition how the correlations evolve with increasing @xmath3 for the cases of lvy and gauss distributed i.i.d . parts . clearly , the correlations are not a finite size effect , but their @xmath3-dependence is markedly different in the two cases . for the gaussian case , the correlations are negative and seem to approach a form independent of @xmath3 . for the lvy case , on the other hand , the correlations seem to increase with @xmath3 . we will argue later that the correlations eventually become @xmath3-independent also for heavy - tailed distributions , but this limit is approached much more slowly than in the gaussian case . a first step towards understanding the behavior of the correlations is to determine the sign with which the curves in figs.[rec_dep ] and [ no_fse ] depart from the i.i.d . case @xmath22 . to address this question , we compute the record rate @xmath59 and the joint probability @xmath60 in the limit of small @xmath5 and @xmath61 , where @xmath62 is the standard deviation or some other measure of the width of the distribution . for the record rate this problem was discussed extensively in @xcite ( see also @xcite ) . knowing that @xmath42 , a taylor expansion for small @xmath5 yields @xmath63 where we have defined the quantity used in @xcite . the integral ( [ i(n ) ] ) also appears in the analysis of the density of near - extreme events @xcite . ] @xmath64 in the same small @xmath5 regime we can also easily determine the joint probability of occurrence of two consecutive records @xmath65 . to leading order in @xmath5 , the expansion of eq.([joint_ldm ] ) yields @xmath66 the first term can be evaluated by partial integration to @xmath67 , and another partial integration yields @xmath68 in addition to the zeroth order term @xmath67 , we find a positive and a negative correction term that describe the effect of the drift . depending on which one of these two terms is larger , positive or negative correlations will result . a similar expansion can be set up for the normalized correlation @xmath25 defined in eq.([lnn ] ) . writing @xmath69 a straightforward computation yields @xmath70\nonumber\\ & & -\frac{n^2(n-1 ) } { 2}i(n-2)-\frac{(n-1)^2(n-2)}{2}i(n-3 ) \end{aligned}\ ] ] again , both positive and negative terms occur in ( [ taylor_mid ] ) , the relative magnitude of which depends explicitly on the underlying probability density @xmath1 . thus from ( [ taylor_mid ] ) , it is not at all obvious how the counterintuitive positive correlations observed numerically arise , how their occurrence depends on the properties of @xmath71 and how the clustering of record events behaves as a function of the number @xmath3 of rv s considered . before turning to the explicit evaluation of @xmath72 in section [ examples ] , we provide some heuristics on the asymptotics of the correlations in the limit of very large @xmath3 , and address the behavior of correlations between rv s that are more than one time step apart . it is clear from eqs . ( [ pn_ldm ] ) and ( [ joint_ldm ] ) that the asymptotic behavior of the record rate @xmath73 and the joint probability @xmath74 hinges on the existence of the function @xmath75 which was rigorously examined by ballerini and resnick . in @xcite , they prove that the limit ( [ g_limit ] ) exists and is nonzero whenever the density @xmath1 has a finite first moment . for completeness we provide a heuristic version of their argument . taking the logarithm of eq . ( [ g_limit ] ) one has to consider the convergence of the series @xmath76 now @xmath77 and for any finite @xmath29 and @xmath78 there is a @xmath79 large enough such that @xmath80 . therefore @xmath81 where @xmath82 is a finite sum and the logarithms in the remaining series were approximated by their taylor expansion around unity . the condition for @xmath1 to have a finite first moment is that @xmath83 decays faster than @xmath84 , which implies convergence of the infinite series in ( [ rest ] ) and hence of @xmath85 ( in the opposite case the series diverges to @xmath86 and @xmath87 ) . since @xmath85 is a probability , it is bounded from above by unity and it follows from ( [ pn_ldm ] ) and ( [ joint_ldm ] ) that @xmath88 and @xmath89 have finite , non - zero limits for @xmath90 . the same statement then applies to the ratio ( [ lnn ] ) . although the focus of this paper is on the case of a positive drift which enhances the occurrence of ( upper ) records , it is instructive to compare the asymptotic behavior for @xmath78 described above to the case @xmath91 . for negative drift both the record rate @xmath73 and the joint probability @xmath74 vanish for @xmath90 , and they do so more rapidly than their i.i.d . counterparts . the asymptotic behavior of the ratio ( [ lnn ] ) is then _ a priori _ undetermined . we consider a simple example where the quantities of interest can be computed explicitly . let the probability density of the i.i.d . variables be given by the negative exponential distribution , @xmath92 and @xmath93 elsewhere . then for @xmath91 the integrand in ( [ pn_ldm ] ) and ( [ joint_ldm ] ) is of the form @xmath94 , \;\;\ ; x \leq 0,\ ] ] and integration yields @xmath95 , \ ; p_{n , n-1}(c ) = \frac{1}{n(n-1)}\exp\left[\frac{c}{2}n(n-1)\right].\ ] ] for @xmath91 the record rate is suppressed superexponentially , such that the expected number of records remains finite for @xmath90 ( see also @xcite ) . however , forming the ratio ( [ lnn ] ) we see that @xmath96\ ] ] which , in contrast to the case of positive drift , grows without bound for @xmath90 . we will see in section [ examples ] that such a behavior is not restricted to this specific example . the non - existence of the @xmath90 limit for @xmath91 suggests that the limiting function @xmath97 may not be smooth for @xmath98 , as will be explicitly demonstrated in section [ examples ] . before embarking on the study of specific distributions , we briefly comment on the behavior of the correlations as a function of the distance between record events . denoting by @xmath99 the joint probability for observing a record in the @xmath100th and the @xmath3th event , it was shown in @xcite that @xmath101 ^ 2 \equiv p(c)^2,\ ] ] which implies that record events become uncorrelated for @xmath102 this leads to the expectation that in a finite time series , the correlations between record events at distance @xmath103 should decay with @xmath103 , in agreement with the simulations shown in fig . [ dist_corr ] . the correlations are seen to be maximal at @xmath104 . in the present paper we therefore focus on the case of nearest neighbor correlations , although extending the computations to @xmath105 is in principle straightforward . correlations between record events at distance @xmath103 from time series of length @xmath106 for gaussian ( @xmath58 ) , lvy - stable ( @xmath107 ) , uniform ( on @xmath108 $ ] ) and pareto ( @xmath107 ) distributions . * left : * full @xmath5 dependence of @xmath109 . * right : * at the drift velocity @xmath110 where the correlations are maximal ( @xmath111 for gaussian and uniform , @xmath112 for the lvy and @xmath113 for the pareto case ) , @xmath114 is shown as function of @xmath103 to illustrate the decay of correlations.,title="fig:",scaledwidth=48.0% ] correlations between record events at distance @xmath103 from time series of length @xmath106 for gaussian ( @xmath58 ) , lvy - stable ( @xmath107 ) , uniform ( on @xmath108 $ ] ) and pareto ( @xmath107 ) distributions . * left : * full @xmath5 dependence of @xmath109 . * right : * at the drift velocity @xmath110 where the correlations are maximal ( @xmath111 for gaussian and uniform , @xmath112 for the lvy and @xmath113 for the pareto case ) , @xmath114 is shown as function of @xmath103 to illustrate the decay of correlations.,title="fig:",scaledwidth=48.0% ] whereas the statistics of record events in sequences of i.i.d . rv s is completely universal @xcite , in the presence of drift one has to distinguish between distributions belonging to the three different universality classes of extreme value theory , the classes of weibull , gumbel and frchet @xcite . for each of these classes we will analyze the correlations in the ldm for a few exemplary distributions , starting with the weibull class , and summarize the observed behavior in a unifying scaling picture in section [ sec : unified ] . the results presented below rely on the study of the record rate @xmath115 in the ldm presented in @xcite , as well as on the expansions for @xmath65 [ eq . ( [ pnm ] ) ] and @xmath116 [ eqs . ( [ taylor_end],[taylor_mid ] ) ] derived above in section [ sec : expansion ] . * left : * numerical simulations of @xmath25 for a uniform distribution of width @xmath117 . for each @xmath5 and @xmath3 we simulated @xmath118 time series . the figure shows data for @xmath119 and @xmath120 . at @xmath121 we find a steep descent of @xmath25 . for @xmath122 and @xmath123 we show the analytic prediction ( [ prediction_uniform ] ) ( steep lines ) . with increasing @xmath5 , @xmath25 appears to become completely independent of @xmath3 . * right : * simulation results for @xmath124 as a function of @xmath3 for @xmath125 and @xmath126 . we simulated @xmath127 series of @xmath128 rv s in each case . for @xmath91 we find a very steep increase of @xmath124 that is in good agreement with the analytical prediction ( [ prediction_uniform ] ) . for @xmath78 , @xmath124 quickly saturates at a constant value and our analytical results are not very useful.,title="fig:",scaledwidth=50.0% ] * left : * numerical simulations of @xmath25 for a uniform distribution of width @xmath117 . for each @xmath5 and @xmath3 we simulated @xmath118 time series . the figure shows data for @xmath119 and @xmath120 . at @xmath121 we find a steep descent of @xmath25 . for @xmath122 and @xmath123 we show the analytic prediction ( [ prediction_uniform ] ) ( steep lines ) . with increasing @xmath5 , @xmath25 appears to become completely independent of @xmath3 . * right : * simulation results for @xmath124 as a function of @xmath3 for @xmath125 and @xmath126 . we simulated @xmath127 series of @xmath128 rv s in each case . for @xmath91 we find a very steep increase of @xmath124 that is in good agreement with the analytical prediction ( [ prediction_uniform ] ) . for @xmath78 , @xmath124 quickly saturates at a constant value and our analytical results are not very useful.,title="fig:",scaledwidth=50.0% ] the weibull class is the class of distributions with bounded support . as a very simple member of the weibull class we consider a uniform distribution on the interval @xmath129 $ ] . we compute @xmath116 in the small @xmath5 regime by making use of ( [ taylor_end],[taylor_mid ] ) and find for @xmath130 @xmath131 the correlations are negative and their magnitude increases rapidly with @xmath3 ( but note that the leading order approximation is valid only for @xmath132 ) . we performed numerical simulations to check this result with @xmath117 and @xmath133 . figure [ fig : uniform ] shows how the descent of @xmath116 for @xmath78 steepens with increasing @xmath3 . in the figure we also plot our prediction ( [ prediction_uniform ] ) for small values of @xmath3 , however the steepness of the descent at zero does not allow us to estimate the quality of our approximation reliably . the figure shows that for large enough @xmath3 and not too small @xmath5 , @xmath116 eventually becomes independent of @xmath3 , as expected from the general arguments given in section [ sec : largen ] . in this case the approach to the @xmath90 limit is particularly rapid , because distributions that are more than @xmath134 steps apart do not overlap . the sharp initial drop of @xmath56 indicates that the limiting distribution @xmath135 is discontinuous at @xmath121 , which is consistent with the non - existence of a limiting function ) is equivalent to the uniform distribution , and therefore the argument of section [ sec : largen ] applies here as well . ] for @xmath91 . as a more general subset of the weibull class we next consider the class of distributions defined by @xmath136 with @xmath137 and @xmath138 . for @xmath139 this produces a uniform distribution similar to the one used above . in @xcite we found that for @xmath140 the integral in ( [ i(n ) ] ) is given by @xmath141 using this and assuming @xmath142 we find the following expression for @xmath116 : @xmath143 and making use of the stirling approximation we finally obtain @xmath144 similar to the uniform case , the correlations between neighboring record events are always negative and their magnitude increases rapidly with increasing @xmath3 , suggesting a discontinuity of the limiting function @xmath135 at @xmath121 . as a first example in the gumbel class we consider the exponential distribution with mean @xmath145 , @xmath146 , @xmath147 . in @xcite it was found that in this case the effect of the linear drift on the record rate is independent of @xmath3 to leading order in @xmath5 . using this result it is straightforward to obtain @xmath148 we compared this result to numerical simulations in fig . [ fig : exp ] and found a very good agreement . apparently @xmath116 assumes a value independent of @xmath3 already for very small @xmath3 . numerical simulations of @xmath25 for an exponential distributions with unit mean . for each @xmath5 and @xmath3 we simulated @xmath118 time series . the figure shows results for @xmath119 and @xmath120 along with our @xmath3-independent analytical prediction ( [ lexp]).,scaledwidth=90.0% ] another important member of the gumbel class is the normal distribution with mean zero and standard deviation @xmath62 . here , the computation is usually a bit more complicated ; however , as in the previous examples , most of the work was already done in @xcite . there we found that for @xmath130 and @xmath149 @xmath150 using ( [ taylor_mid ] ) this result allows us to estimate the effect of the drift on the correlations . we find @xmath151 unlike the case of the exponential distribution , in the gaussian case the correlations are negative and depend strongly on @xmath3 . the behavior is similar to that found in the weibull case , and matches the result ( [ weibull_gen ] ) ( up to logarithimic factors ) for @xmath152 . numerical results for the gaussian distribution are shown in fig . [ fig : gaussian ] . similar to the uniform distribution , @xmath116 shows a steep descent at @xmath121 which becomes steeper for increasing @xmath3 . however , the descent appears to be smoother than in the uniform case . in agreement with the considerations of section [ sec : largen ] , for @xmath78 and large @xmath3 the correlations become independent of @xmath3 . in contrast , for @xmath91 the correlations increase monotonically with @xmath3 and do not appear to saturate . in the case of the gaussian distribution our analytical prediction ( [ prediction_gaussian ] ) is only valid in a small regime for small @xmath78 and relatively small @xmath3 , but for @xmath91 the approximation is significantly better . numerical simulations of @xmath25 for a gaussian distribution of width @xmath153 . * left figure * shows @xmath25 for @xmath119 and @xmath120 . for each @xmath5 and @xmath3 we simulated @xmath127 time series . the figure illustrates the steep descent of the correlator at @xmath121 when @xmath5 is small . * right figure * shows @xmath25 for different , fixed values of @xmath5 together with our analytical results . here , we analyzed @xmath154 series of rv s for each drift rate . again , we find agreement between the simulations and our analytical computations for small @xmath3 and @xmath5 . we manually fitted curves @xmath155 . interestingly for @xmath91 the agreement is a lot better and @xmath25 does not saturate at a constant value.,title="fig:",scaledwidth=50.0% ] numerical simulations of @xmath25 for a gaussian distribution of width @xmath153 . * left figure * shows @xmath25 for @xmath119 and @xmath120 . for each @xmath5 and @xmath3 we simulated @xmath127 time series . the figure illustrates the steep descent of the correlator at @xmath121 when @xmath5 is small . * right figure * shows @xmath25 for different , fixed values of @xmath5 together with our analytical results . here , we analyzed @xmath154 series of rv s for each drift rate . again , we find agreement between the simulations and our analytical computations for small @xmath3 and @xmath5 . we manually fitted curves @xmath155 . interestingly for @xmath91 the agreement is a lot better and @xmath25 does not saturate at a constant value.,title="fig:",scaledwidth=50.0% ] to understand the marked difference between the exponential and gaussian distributions , we analyze the general class of gumbel - type distributions of the form @xmath156 with normalization @xmath157 and @xmath158 . in this case it was found in @xcite that @xmath159 with some further computations this leads us to the following behavior of the leading order correction coefficient @xmath160 : @xmath161 where @xmath162 and @xmath163 are positive constants not depending on @xmath3 . the exact values of @xmath162 and @xmath163 are very difficult to compute and we will not consider them in this article . nevertheless , this expression nicely reproduces our results for the exponential and the gaussian distribution and shows how both positive and negative correlations may emerge in the gumbel class . for @xmath164 the leading first term vanishes and @xmath72 reduces to a positive constant . for all @xmath165 the first term dominates and the correlations grow with @xmath3 , with a sign determined by @xmath166 . for values @xmath167 of distributions that decay faster than the standard exponential we find negative correlations between the records , in agreement with the gaussian example , while for stretched exponential distributions ( @xmath168 ) positive , @xmath3-dependent correlations result ( see fig . [ fig : comp ] for an example ) . in hindsight , the special role of the exponential distribution should not come as a surprise , since the gumbel distribution ( [ gumbeldist ] ) ( for which correlations are completely absent ) has an exponential tail . in the sense of extreme value theory , the exponential distribution is close to the gumbel , and the corresponding records in the ldm are therefore almost uncorrelated ( up to small residual correlations which are independent of @xmath3 ) . as a subset of the distributions in the frchet class of extreme value statistics we consider the well known pareto distribution @xmath169 with @xmath170 and @xmath171 . in @xcite it was found that in this case @xmath172 this allows us to compute @xmath116 for the pareto distribution . we find @xmath173 * left : * numerical simulations of @xmath25 for pareto distributions . for each @xmath5 and @xmath174 we simulated @xmath118 time series of length @xmath123 . the figure shows simulations for @xmath175 and @xmath176 , along with our analytical predictions for small @xmath5 . at @xmath121 we find a steep ascent of @xmath25 which gets steeper with increasing @xmath174 . our analytical result from ( [ prediction_pareto ] ) is in good agreement with the simulations for small enough @xmath5 . * right : * simulation results for a pareto distribution with @xmath177 and different values of @xmath5 . we analyzed @xmath154 series of @xmath178 rv s for each @xmath5 . our analytical work predicts a square - root behavior of @xmath179 , which is in good agreement with the simulations . , title="fig:",scaledwidth=50.0% ] * left : * numerical simulations of @xmath25 for pareto distributions . for each @xmath5 and @xmath174 we simulated @xmath118 time series of length @xmath123 . the figure shows simulations for @xmath175 and @xmath176 , along with our analytical predictions for small @xmath5 . at @xmath121 we find a steep ascent of @xmath25 which gets steeper with increasing @xmath174 . our analytical result from ( [ prediction_pareto ] ) is in good agreement with the simulations for small enough @xmath5 . * right : * simulation results for a pareto distribution with @xmath177 and different values of @xmath5 . we analyzed @xmath154 series of @xmath178 rv s for each @xmath5 . our analytical work predicts a square - root behavior of @xmath179 , which is in good agreement with the simulations . , title="fig:",scaledwidth=50.0% ] which leads to the expression @xmath180 for large @xmath3 . for positive @xmath5 , we thus expect positive correlations between neighbouring record events for all distributions of pareto form . these correlations are increasing with @xmath3 slower than linearly . we expect that , asymptotically , this behavior is universal for all distributions within the frchet class which show negative correlations between record events . we have checked that this behavior is not asymptotic , and that the correlations become positive for larger values of @xmath3 . ] . we compare eq.([prediction_pareto ] ) to numerical simulations in fig . [ fig : pareto ] , finding good agreement for small @xmath5 , both for the @xmath5- and the @xmath3-dependence . the agreement improves when the distribution becomes broader for smaller @xmath174 . the strength of correlations for strongly heavy - tailed distributions of the pareto class is remarkable . for example , for a pareto distribution with coefficient @xmath177 we found correlations @xmath181 for large @xmath3 ( see fig . [ fig : comp ] ) . the correlations displayed in the right panel of fig.[fig : pareto ] increase ( for @xmath182 ) or decrease ( for @xmath183 ) with @xmath3 without showing any sign of saturation , although we know from the general considerations of section [ sec : largen ] that @xmath124 must approach an @xmath3-independent limit for @xmath182 and @xmath184 . as we will see in the next subsection , this reflects the fact that the time scale @xmath185 at which the limit is attained is ( for a given value of @xmath5 ) particularly large for distributions of the frchet class . we also note that because @xmath186 for @xmath183 , the normalized correlations are confined to lie between 0 and 1 in this case . this implies that the divergence of @xmath56 for @xmath90 and @xmath183 , which was demonstrated in section [ sec : largen ] to occur for a specific example in the weibull class , can not happen here . we therefore conjecture that , for distributions in the frchet class , the limit ( [ last ] ) exists also for @xmath183 . the results of the preceding subsections can be summarized in a simple scaling picture . we first recall from @xcite and @xcite that the ldm ( with @xmath78 ) contains a characteristic time scale @xmath185 at which the record rate @xmath88 crosses over from the i.i.d . behavior @xmath187 to the limiting value @xmath188 . for the different distributions discussed above , this time scale diverges for @xmath98 according to @xmath189 ignoring for the moment the logarithmic factor in the gumbel case , these behaviors can be further simplified by expressing the different universality classes in terms of the generalized pareto distribution @xcite @xmath190 for @xmath191 this is of pareto type with @xmath192 , for @xmath193 it reduces to a weibull - type distribution similar to ( [ kuramaswamy ] ) with @xmath194 , and the gumbel class is represented by the exponential distribution which arises from ( [ genpareto ] ) for @xmath195 . using this representation , the different cases in ( 45 ) , ( 46 ) and ( 47 ) reduce to @xmath196 note that @xmath197 in the weibull class but @xmath198 in the frchet class , which explains the slow convergence to the @xmath90 limit in the latter case . moreover , it was shown in @xcite that the integral ( [ i(n ) ] ) behaves for large @xmath3 as ( see also @xcite ) @xmath199 to relate this to the behavior of the correlations , we note that for large @xmath3 eq.([taylor_mid ] ) can be approximately written as @xmath200 inserting ( [ i(n)scal ] ) we thus conclude that , to leading order , @xmath201 which correctly reproduces both the sign and the order of magnitude of the correlations derived in section [ examples ] for the weibull and frchet classes : correlations are positive ( negative ) for @xmath191 ( @xmath193 ) , and they scale sublinearly ( superlinearly ) with @xmath3 in the two cases . a similar calculation using the refined estimate ( [ igumbel ] ) of @xmath202 for the gumbel class shows that the correlations are negative ( positive ) for @xmath203 ( @xmath204 ) , in agreement with ( [ jgumbel ] ) . numerical simulations of @xmath25 with @xmath205 for six different distributions . in each case we averaged over @xmath118 time series of length @xmath206 . the results for @xmath25 were binned logarithmically in order to improve the averaging for large @xmath3 . for the uniform and the gaussian distribution @xmath25 is clearly negative for @xmath78 and it decreases with growing @xmath3 . at some @xmath207 the correlations become independent of @xmath3 . for the special case of the exponential distribution we find a constant value of @xmath25 which is only slightly larger than unity . both the stretched exponential ( with @xmath208 ) and the pareto distributions ( with @xmath209 ) show positive correlations . note the slow convergence for the pareto distribution with @xmath210.,scaledwidth=90.0% ] in this paper we analyzed the effect of a linear drift on the correlations between records drawn from series of independent rv s , as quantified by the normalized joint probability @xmath116 . in section 2 we derived general expressions ( exact and approximate ) for @xmath116 and recalled the fact that records are independent both for i.i.d . rv s ( @xmath121 ) and in the special case where the rv s are drawn from the gumbel distribution . our main analytic results were obtained in section 3 by way of an expansion in the small @xmath5 limit , similar to the approach developed previously in @xcite . using this approach we were able to show that the correlations are generally negative ( ` repulsive ' ) for distributions in the weibull class , and positive ( attractive ) for distributions in the frchet class . in the gumbel class the sign of the correlations depends on the stretching exponent @xmath211 in ( [ fbeta ] ) , with the border between positive and negative correlations being given by the exponential case @xmath212 . in contrast to all other cases , for distributions with an exponential tail the correlations are weak and independent of @xmath3 , which is consistent with the fact that this class of distributions also contains the gumbel distribution , which has no correlations at all . simulation results illustrating the different cases are summarized in fig . [ fig : comp ] . a special role of the exponential distribution in separating two regimes of qualitatively different behaviors has been noted previously in the related context of near - extreme events @xcite . perhaps the most surprising and counterintuitive outcome of our work is the discovery of strong positive record correlations for distributions with a power law tail . in view of the substantial interest in detecting and explaining heavy - tailed distributions in all areas of science @xcite , our finding suggests that drift - induced record correlations could be used as a distribution - free test for detecting power laws or streched exponentials in empirical data @xcite . an interesting open question concerns the structure of the record correlations in the asymptotic limit @xmath90 , where the record rate approaches a nonzero constant and the record process thus becomes stationary . based on the work of ballerini and resnick @xcite , we have argued in section [ sec : largen ] that the limit ( [ last ] ) exists for @xmath78 whenever the underlying distribution has a finite mean , whereas for @xmath183 it is possible that @xmath56 diverges for @xmath90 . the fact that the coefficient of the leading order term in the small @xmath5 expansion generally diverges with @xmath3 indicates that the limiting function @xmath135 may be singular for @xmath98 , and we have presented numerical evidence for the occurrence of a discontinuity at @xmath121 for the weibull class . for heavy - tailed distributions in the frchet class , the large-@xmath3 asymptotics is difficult to ascertain numerically because of the slow convergence , but we have argued that in this case @xmath135 may exist also for @xmath91 . rigorous work addressing these questions along the lines of @xcite would be most welcome . this work has been supported by dfg within the _ bonn cologne graduate school of physics and astronomy _ , by friedrich ebert stiftung through a fellowship to gw , and by studienstiftung des deutschen volkes through a fellowship to jf . spmpsci gembris , d. , taylor , j.g . , suter d. : sports statistics : trends and random fluctuations in athletics . nature * 417 * , 506 ( 2002 ) gembris , d. , taylor , j.g . , suter d. : evolution of athletic records : statistical effects versus real improvements . j. appl . stat . * 34 * , 529 ( 2007 ) bassett , g.w . : breaking recent global temperature records . climatic change * 21 * , 303 ( 1992 ) benestad , r.e . : how often can we expect a record event ? climate research * 25 * , 3 ( 2003 ) redner , s. , petersen , m. r. : role of global warming on the statistics of record - breaking temperatures > phys . e * 74 * , 061114 ( 2006 ) meehl , g.a . , tebaldi , c. , walton , g. , easterling , d. , mcdaniel , l. : relative increase of record high maximum temperatures compared to record low minimum temperatures in the u.s . lett . * 36 * , l23701 ( 2009 ) wergen , g. , krug , j. : record - breaking temperatures reveal a warming climate . epl * 92 * , 30008 ( 2010 ) newman , w.i . , malamud , b.d . , turcotte , d.l . : statistical properties of record - breaking temperatures . e * 82 * , 066111 ( 2010 ) anderson , a. , kostinski , a. : reversible record breaking and variability : temperature distributions across the globe . j. appl . 49 * , 1681 ( 2010 ) de haan , l. , ferreira , a. : extreme value theory - an introduction . springer , new york ( 2006 ) sibani , p. , brandt , m. , alstrm , p. : evolution and extinction dynamics in rugged fitness landscapes . * 12 * , 361 ( 1998 ) krug , j. , jain , k. : breaking records in the evolutionary race . physica a * 358 * , 1 ( 2005 ) park , s .- c . , krug , j. : evolution in random fitness landscapes : the infinite sites model . p04014 ( 2008 ) oliveira , l.p . , jensen , h.j . , nicodemi , m. , sibani , p. : record dynamics and the observed temperature plateau in the magnetic creep - rate of type - ii superconductors . b * 71 * , 104526 ( 2005 ) sibani , p. , rodriguez , g.f . , kenning , g.g . : intermittent quakes and record dynamics in the thermoremanent magnetization of a spin - glass . phys . rev . b * 74 * , 224407 ( 2006 ) wergen , g. , bogner , m. , krug , j. : record statistics for biased random walks , with an application to financial data . phys e * 83 * , 051109 ( 2011 ) franke , j. , wergen , g. , krug , j. : records and sequences of records from random variables with a linear trend . : theor . exp . p10013 ( 2010 ) arnold , b.c . , balakrishnan , n. , nagaraja , h.n . : records . wiley , new york ( 1998 ) nevzorov , v. b. : records : mathematical theory . american mathematical society ( 2001 ) glick , n. : breaking records and breaking boards . monthly * 85 * , 2 ( 1978 ) majumdar , s.n . , ziff , r.m . : universal record statistics of random walks and lvy flights . * 101 * , 050601 ( 2008 ) , p. , wiese , k.j . : driven particle in a random landscape : disorder correlator , avalanche distribution , and extreme value statistics of records . e * 79 * , 051105 ( 2009 ) sabhapandit , s. : record statistics of continuous time random walk . epl * 94 * , 20003 ( 2011 ) ballerini , r. , resnick , s. : records from improving populations . * 22 * , 487 - 502 ( 1985 ) ballerini , r. , resnick , s. : records in the presence of a linear trend . * 19 * , 801 - 828 ( 1987 ) krug , j. : records in a changing world . p07001 ( 2007 ) gumbel , e. j. : statistical theory of extreme values and some practical applications . national bureau of standards applied mathematics series * 33 * ( 1954 ) galambos , j. : the asymptotic theory of extreme order statistics . wiley , new york ( 1987 ) sornette , d. : critical phenomena in natural sciences . springer , berlin ( 2004 ) sabhapandit , s. , majumdar , s.n . : density of near - extreme events . * 98 * , 140201 ( 2007 ) pickands iii , j. : statistical inference using extreme order statistics . . stat . * 3 * , 119 ( 1975 ) christensen , k. , moloney , n.r . : complexity and criticality . imperial college press , london ( 2005 ) franke , j. , wergen , g. , krug , j. ( in preparation )

while the inherent complexity of the solar corona and its dynamic processes has engendered great scientific interest and study over the past half century , it is this complexity that is perhaps the greatest challenge when attempting to model and understand the underlying processes of observed phenomena in a quantitative fashion @xcite . for this reason , the development and use of fully 3d magnetohydrodynamic ( mhd ) models describing the corona and solar wind , which has undergone a veritable renaissance in the computing age , has been critical to furthering our understanding of the solar wind and eruptive events @xcite . a particularly exciting avenue has been the development of 3d observation - based techniques , which provide the opportunity to include realistic conditions of the corona into studies of the global solar wind @xcite and to address how this governs dynamic events @xcite . + with its origin in the long - studied fundamental question of how the solar corona can be heated to and maintain temperatures in the million kelvin ( mk ) regime , a primary focus of this study is to address the issue of energy input into the corona in the context of a fully 3d global model . because non - mhd processes of energy transport and local conditions in the low corona are responsible for coronal heating mechanisms , it is often quite difficult and computationally expensive to include the small - scale micro physics of reconnection and turbulence responsible for coronal heating in 3d models , especially since the exact processes involved are not universally agreed upon . as a result , heating models are often parameterized as a heating term that depends on various local magnetic and thermodynamic properties that is included in the energy equation ( e.g. @xcite ) . however , due to the large number of heating models and their ad - hoc formulations , it is crucial to use real observations to compare and constrain these models and deduce the best combination of models , both for active regions and the quiet sun , that yields the most realistic global simulations . when using such models to study transient events in the solar corona , particularly coronal mass ejections ( cmes ) and coronal waves seen in the extreme ultraviolet regime ( euv waves , @xcite ) , it is a motivating principle that only by constraining physical theories and scenarios through as many observable manifestations available can we provide the best possible avenue to further our knowledge and precision in understanding these dynamic events ( e.g. @xcite ) . the ultimate goal being to strip away layers of empirical approximation with more fundamental physical terms and mechanisms . however , it is equally important to validate and asses the ability of these improvements to adequately represent the global conditions in the solar corona . + to address this issue through a deterministic , data - driven approach , we modify the global corona model of the space weather modeling framework ( swmf ) @xcite to include the transition region between the chromosphere and the corona , where non - mhd thermodynamic terms of energy transport , such as electron heat conduction , radiative losses , and coronal heating all become important . using techniques similar to the pioneering work of @xcite , in this refined version of the model the inner boundary is placed at the chromosphere or upper transition region rather than in the low corona and the now relevant non - mhd terms are added to the governing mhd energy equation . + in section [ section : model ] we discuss the new lower corona ( lc ) model and physical considerations added as part of this work . in section [ section : model_runs ] we overview our runs and results and discuss the application of this model to the study of the corona ( in particular the comparison to euv imaging observations ) . in section [ section : detailed_analysis ] we provide a detailed analysis of a particular run , and benchmark it via comparison to a previous corona model . we conclude in section [ section : conclusion ] . the main model and starting point for this work is the swmf solar corona ( sc ) model @xcite . as far as numerical methods , the code is fully parallelized and designed to be highly customizable in terms the methods , solvers , and equations used . a complete description of the mhd equations and their implementation in the swmf can be found in @xcite as well as the addition of source terms in @xcite . for the current version of the solar corona model , a steady state solar wind solution is obtained using a variable @xmath0 model @xcite . the latter references use the wang - sheeley - arge ( wsa ) model @xcite as an initial condition to derive the variation in the polytropic index . this then achieves a steady state mhd solution after nominal integration ( henceforth refered to as the `` standard sc model '' ) . + [ section : model_swmf ] physically , the most important advantage of this tool and others like it lies in its ability to simulate the complete 3d environment of any event and carrington rotation ( standardized solar rotation number , abbreviated cr ) for which data for the photospheric magnetic field is available throughout the entire solar surface . the initial magnetic configuration is extrapolated using the potential field source surface method ( pfssm ) @xcite , which uses magnetic coefficients derived from observations of the synoptic photospheric magnetic field ( typically high order maps using data from the mdi instrument aboard the soho observatory , or low order maps the wilcox solar observatory ) . using this method , the subsequent evolution of the magnetic field towards a steady state during the simulation is no longer strictly potential . + while the standard version of the sc model has been very successful at reproducing the bi - modal solar wind structure and observations at 1au @xcite , which was the fundamental goal of that empirical model , the lower boundary at the solar surface is quite smooth and nearly uniform . in order to reproduce and study the fine structures of the the low corona we significantly modify the sc model to address the unique physics that take place at this boundary ( henceforth refered to as the the lower corona ( lc ) model ) . this modification of the energy equation takes the form of : @xmath1 where , the subscript @xmath2 refers to the standard mhd terms in the swmf and the additional terms @xmath3 , @xmath4 , @xmath5 are described below . because these terms account for realistic energy transport , the polytropic index is no longer variable , and is set to a uniform value of @xmath6 . the standard form of anisotropic electron heat conduction in the collisional limit is included as an additional energy flux term , @xmath7 , in the energy equation ( equivalent physically to adding a source term of @xmath8 . the parallel component of spitzer conductivity @xcite , this term is especially important in the transition region and low corona and plays a critical role in determining the equilibrium density and temperatures at the base of the corona . we use a value of @xmath9 in cgs units . + we also include a method for broadening the equilibrium scale length of the transition region , which is needed to resolve the entire transition region when using a chromospheric boundary ( described in detail by @xcite ) . this reduces to both fixing @xmath10 to a constant value of @xmath11 and reducing the radiative cooling term by the inverse factor @xmath12 ) for temperatures below a transition region value , @xmath13 . in this work we use @xmath14 . + [ section : heat_conduction ] the next term , @xmath15 , accounts for the radiative losses of the hot coronal plasma in the optically thin limit . the loss function @xmath16 is calculated using the chianti version 5 radiative loss routines @xcite and linearly interpolated as a tabulated function in the model . the assumption of a fully ionized hydrogen gas is already implicit in this model , which gives @xmath17 . it is important to note that the choice of abundances used in calculating the radiative losses function can change the total losses significantly in the coronal temperature regime . for this reason we choose to use the coronal abundances file : ` sun_coronal_ext.abund ` @xcite when calculating the radiative cooling curve with chianti , and do not vary this function in this work . the first empirical coronal heating term , @xmath19 , that we examine is equivalent in form to that used by @xcite and was originally based on work relating the surface unsigned magnetic flux of a convective envelope to the x - ray luminosity of stars @xcite . this model is expressed as : @xmath20 [ heating_model_1 ] where @xmath21 , @xmath22 are fixed parameters , @xmath23 represents the total unsigned magnetic flux at the solar surface , @xmath24 is the local heating weighting function , and @xmath25 is a normalization constant . this formulation calculates the total amount of energy input into the corona and distributes it via @xmath24 , which is chosen to be a function of @xmath18 . the primary benefit of this model is its simplicity , the amount of heating at any location is given by the value of @xmath24 , and the total amount of heat input to the corona ( nearly constant ) is normalized by the global integral in the denominator , @xmath26 , which is carried out over the entire domain . in this implementation , @xmath23 is calculated by integrating the unsigned flux at the model boundary , @xmath27 , which is essentially the photospheric magnetogram used to initiate the model . + because both the relationship between x - ray luminosity and total power input into the corona is not well constrained , and varying magnetogram resolution does not constrain the true unsigned flux , @xmath25 represents a relatively free parameter with which to adjust the total power on a case by case basis . in our model runs we choose @xmath28 and linear weighting with the magnetic field magnitude , @xmath29 . this gives the heating term the operational equivalent of @xmath30 , where @xmath31 is typically around @xmath32 in cgs units , and total integrated power of @xmath33 if applied uniformly . + one drawback of this term ( and of other @xmath18 weighted heating terms in global models ) is that the heating scale height , defined as the point along a given loop where @xmath34 , is not a free parameter and varies with the strength and spatial distribution of @xmath35 . also , note that the value of @xmath31 depends on the specific global magnetic field configuration and the radius to which the volume integral is calculated . in this work , we limit the range of influence of this function to the low corona by multiplying @xmath24 by an exponential envelope function with a scale height of @xmath36 . the second heating model is a simple exponential scale height model . using a standard exponential form we define heating model 2 as : @xmath37}.\ ] ] [ heating_model_2 ] where @xmath38 is the local heating rate at @xmath39 and @xmath40 is the heating scale height . a short heating scale height is consistent with soho and trace euv observations of coronal loops @xcite , and motivated by the `` nanoflare '' class of coronal heating models , in which most of the heat is deposited in the upper chromosphere and transition region @xcite . it is important to note that it is the high thermal conductivity of the coronal plasma that subsequently distributes this energy along the field / flow lines . for heating quiet sun regions , we use the values @xmath41 , and @xmath42 , which give a total power of @xmath43 if applied uniformly ( surface flux @xmath44 ) . the lack of explicit dependence on @xmath18 with an exponential scale height model allows us examine directly the effect that purely the magnetic geometry of the 3d corona has on the resultant thermodynamic equilibrium . to maintain realistic temperatures in the solar wind and coronal holes at large distances , we use the same form of eq . with values of @xmath45 , and @xmath46 giving additional total power @xmath47 . this function is applied uniformly in each model run . while not entirely realistic for the acceleration profiles observed in coronal hole regions and the solar wind , we find that including this term is adequate for reproducing coronal hole emission in the absence of short scale height heating . [ heating_model_ch ] one interesting focus of this study is to isolate the complementary effects that non - uniform heating at the surface and the 3d magnetic topology have in determining the equilibrium structure of the corona . for this reason , as one of the options in the model we include the ability to apply the low scale height heating model ( either 1 or 2 ) to only regions associated with closed field in the corona . to do so , we use the pfssm extrapolation computed to initiate the model to determine at every computational element whether the field threading the location is open ( connects to the source surface and becomes radial ) or closed ( turns back down to the solar surface ) . if the region is associated with open field , then only the long scale height `` coronal hole '' heating model is applied ( section [ heating_model_ch ] ) , while if the region is closed , both are applied . + we would also like to note that any of the empirical heating terms discussed above ( sections [ heating_model_1 ] - [ heating_model_oc ] ) may also be attributed to wave absorption and dissapation , probably due to the ion interaction with alfvn wave turbulence . in this case , the heating may be specified in terms of the wave ( spectral ) energy and their damping coefficient , and the wave turbulence may contribute not only to the energy source but also to the momentum fluxes ( see e.g. @xcite and papers cited therein ) . however , this particular study is focused primarily on the plasma conditions low corona , and we operate under the assumption that alfvn wave pressure will be much lower than the thermal pressure there . [ heating_model_oc ] boundary conditions at the @xmath39 surface are applied to fix constant values temperature and density at the face boundary according to either of the models described below . each of these use a zero flow condition and fix the tangential component of the magnetic field ( similar to those used in @xcite . for the supersonic flow at the outer boundary at @xmath48 we use a floating ( zero - gradient ) condition . the primary boundary condition used in this model is a simple uniform `` chromospheric '' condition . as described in @xcite , by setting chromospheric values of electron temperature , @xmath49 , and density , @xmath50 , and including the transition region broadening method described in section [ section : heat_conduction ] , it becomes possible to resolve the entire transition region everywhere in a global model . this is a critical advantage because the resultant topological equilibrium in the coronal part of the solution ( density enhancements , varying width of the transition region , etc . ) will depend entirely on the included physics and magnetic geometry , and not on its proximity to the boundary of the model . this fact is particularly important for the study of any sort of dynamics in the global corona , as the sharp transition from fast alfvn speed in the quiet corona ( @xmath51 ) to extremely slow speeds in the chromosphere ( @xmath52 ) , provides a realistic dispersion buffer for fast magnetosonic waves . in order to include a realistic lower boundary condition for the corona , while at the same time maintaining decent computational efficiency for the global domain , we use the method of the radiative energy balance model ( reb ) outlined by @xcite ( further formalism , including coronal heating in @xcite ) . in short , this model fixes the base electron temperature to a high transition region value , @xmath53 , and in turn assumes an equilibrium balance throughout the rest of the transition region below to the top of the chromosphere , from which an electron density , @xmath54 , may be obtained . by rearranging the terms in the mhd energy equation and integrating over t at constant pressure , one obtains : @xmath55 where , the integral of the radiative loss function is carried out between @xmath56 and @xmath53 . this allows for the base density @xmath54 to be a function of the physical conditions included in the model ( coronal heating , anisotropic heat conduction , and radiative cooling ) and thus varies spatially along the boundary . most importantly , the numerator of shows how the interplay between magnetic field strength , @xmath57 ( via @xmath19 in the right term ) , and normal orientation at the surface , @xmath58 ( via @xmath59 on the left ) , sets the base density on the solar surface . the ability to capture these features with such a simple , easily computed boundary is a primary motivation in using this particular model . another aspect of this work has been to address the unique geometric concerns that arise when including the transition region and below . a typical solar corona ( sc ) simulation is carried out with an adaptive cartesian grid over a sun - centered @xmath60r@xmath61 cube with @xmath62 cell blocks . the average cell size is smallest at the surface ( @xmath63 r@xmath64 km ) and is incrementally increased with distance to a largest size of @xmath65 r@xmath61 near the outer boundary , giving a few million cells . while this is a relatively large surface cell size , it is adequate considering the precision of the magnetic field boundary ( an observational limitation ) . however , if one is required to resolve the transition region , with dynamic radial spatial scales on the order of a few km , it quickly becomes infeasible to use a cartesian grid ( with every @xmath66 refinement in length the number of surface cells increases by 4 ) . this can be partially alleviated using methods to widen the extent of the model transition region to typical scales of order @xmath67 km without affecting the coronal solution ( section [ section : heat_conduction ] ) , but the problem is still quite substantial . this then represents a balance between height accuracy and computational efficiency ( with a cartesian grid , the number of surface cells increases by a factor of 4 with every successive @xmath66 refinement ) . + to address this scale and resolution issue we adapted the generalized grid capability of the swmf , which has ability to calculate mhd fluxes in arbitrary geometries , to construct a spherical ( r@xmath68 ) grid function with highly non - uniform radial scales near the transition region ( @xmath69 km ) that smoothly transitions to near equal face area at large @xmath70 ( @xmath71 km at @xmath72 ) . because the grid function is continuous , the model maintains the block - adaptive mesh and adaptive mesh refinement ( amr ) capability for flexible local spatial resolution both initially and as the simulation progresses . for this study , focusing on the global structure of the corona , we use a uniform spacing in the angular directions with @xmath73 at the surface , which is coarsened by a factor of two via amr beyond @xmath74 for regions within @xmath75 and @xmath76 outside ( to avoid needlessly small cell sizes at the poles ) . the outer boundary is fixed at @xmath48 . degenerate cells touching the polar axes are treated with azimuthal averaging to avoid discontinuity across the poles . a grid comparison near the surface between this work and an amr cartesian mesh with a similar number of total cells is shown in figure [ fig : geometry ] . to study the euv corona and associated dynamics in the context of real events , we include a proper treatment of integrated coronal emission in the euv range , particularly for the purpose of comparing simulated observations to existing euv observations by a current solar observatory . this process , following the work of @xcite , involves two main steps : ( i ) characterization of detector response based on physical parameters that are calculated in the model and ( ii ) line - of - sight ( los ) integration though the 3d data set to create synthesized image . + [ section : los - synthesis ] the first step is to understand and calculate the instrumental response for coronal material at a given temperature and density . for simplicity and ease to compare to studied observations we chose to model the three coronal band - passes at 171 , 195 , and 284 of the eit instrument on the soho spacecraft @xcite and the almg filter of the sxt instrument @xcite . because of the extremely high temperature and low densities , the emission line euv and soft x - ray radiation of the solar corona above 0.5 mk can be treated as optically thin to a good approximation . as such , the intensity measured by each pixel of an imaging detector can be treated as a los integral through the coronal plasma , namely : @xmath77 where @xmath78 is the pixel response measured in data numbers ( dn ) per second . here @xmath79 follows along the line - of - sight , the square of the electron density , @xmath80 , accounts for the amount of emitting material and @xmath81 is the instrumental response of filter @xmath82 per unit emission measure as a function of both temperature and density . this is equivalent to integrating the resolved differential emission measure ( dem ) distribution along @xmath79 . to calculate the response functions , @xmath81 , we use the chianti version 5 emission line analysis routines @xcite to generate synthetic spectra , and the eit and sxt analysis routines for detector response ( a part of the solarsoft ( ssw ) framework written in idl @xcite ) . in this section we present the application of the lc model to cr 1913 , initialized using a synoptic mdi magnetogram centered on aug 27 , 1996 . occuring during solar minimum , the sun at this time features a conspicuous equatorial coronal hole extending from the north to down past disk - center , as well a large active region . this then conveniently allows us to examine the ability of a given model to describe these three basic regimes : the average quiet - sun , coronal holes , and active regions . additionally , this rotation has been studied in detail using similar methods ( e.g. @xcite ) and thus provides a means of comparison to existing work . + four model runs used to demonstrate the usefulness of the lc model are summarized in table [ table : model_runs ] . as a convenient comparison of their global thermodynamic and topological properties , the los synthesis of euv and soft x - ray emission are shown in figure [ fig : euvcomp_4runs ] . each of the models presented use either @xmath18 weighted heating ( model 1 , section [ heating_model_1 ] ) or exponential heating ( model 2 , section [ heating_model_2 ] ) ) . all four use coronal hole heating ( section [ heating_model_ch ] ) applied uniformly to the domain , while all but run c use the open / closed field weighting ( section [ heating_model_oc ] ) to achieve a more realistic equilibrium in these regions . in order to include strong heating in active regions ( ar ) with high magnetic field strength , run d also includes a modification to the heating function , and transitions smoothly from exponential heating to @xmath18 weighted heating ( @xmath83 ) above @xmath84 gauss . + cccccc a & reb & @xmath85 & yes & @xmath86 + b & chromo & @xmath85 & yes & @xmath86 + c & chromo & exp & no & @xmath87 + d & chromo & exp + ar & yes & @xmath86 + all four model runs show similar global features in the euv and soft x - ray when compared to observations but differ on the precise details . for example , all four runs resolve cool temperatures along the the polarity inversion lines near the surface ( seen to the east and west of the extended equatorial coronal hole near the center of the disk ) . this is naturally due to field line orientation and loop length governing thermodynamic equilibrium . this commonality serves to emphasize the strong role that the 3d magnetic field topology of a given carrington rotation plays in determining dynamic equilibrium in the corona . + in the comparison between run a ( reb model bc ) and run b ( chromospheric bc ) we observe more uniform emission near the surface as well as slightly higher temperature ( as seen in the longer extent of sxt emission ) for the reb case . because the boundary is near the top of the transition region in the reb model , the height of the transition region is less able to vary in response to magnitude of the downward heat conduction flux from above also , since the reb model does not include a region with chromospheric densities ( where the radiative cooling is much higher due to the @xmath88 dependence ) a smaller fraction of the coronal heating power is lost in the transition region , leading to higher equilibrium temperatures . however , it is clear that the reb model , which is more suited for computational efficiency due to the longer equilibrium length scales at @xmath89 , can be used to adequately describe the ambient global corona . + one of the most striking results is perhaps the comparison of run c to the other model runs . in this run , the coronal heating function was applied uniformly in the computational domain ( no dependence on @xmath90 or @xmath23 ) . while the coronal holes and significant ar emission are unsurprisingly not well preserved in the euv , still much of the basic features of the low corona are reproduced . the ability of simple uniform heating model to reproduce obvious features on the surface , such as average quiet sun emission , and lessening near inversion lines , as well as the bi - modal structure between open and closed field regions at higher temperatures ( @xmath91 line and soft x - rays ) speaks to the importance of both the 3d magnetic topology in determining the dynamic equilibrium of the corona and the redistribution of energy via a thermodynamic energy equation including electron heat conduction . + for a more realistic empirical model in run d , we modify run c by including ar heating and the open - field cutoff to better describe heating in active regions in coronal holes . the most interesting finding in this comparison is that by including significantly enhanced heating in the large active region , we observe a pronounced effect on the equilibrium structure of the ar associated closed field / streamer region ( eastern side of the corona in figure [ fig : euvcomp_4runs ] ) . this produces noticeable feedback on the global structure as seen via a larger northward tilt of the heliospheric current sheet near the ar longitudes ( shown in detail figure [ fig : streamer_br_slice ] ) , and suggests a complex relationship both between the thermodynamics of the low corona and the global structure of the solar wind . because this effect can not be fundamentally extracted from extrapolating the magnetic field alone , this provides not only a stong motivation for using thermodynamic models , but also another element with which to constrain and refine ar heating models in future studies ( the one used in in this work being a necessary but crude approximation ) . + another important inference about coronal heating can be gleaned from the comparison of the two heating models directly ( run b and run d ) . although the total heating power is the same in each model , the @xmath18 weighted heating model naturally has a shorter scale height ( due to the falloff of @xmath18 with radius ) and we subsequently observe slightly enhanced pressure in the low corona and lower temperatures ( hence density scale height ) in the closed field regions . however , the fact that the simple exponential heating model applied to the quiet sun in run d can reproduce the mean structure of the low corona with the same fidelity as a more complicated empirical model , demonstrates that perhaps complex empirical models are not as entirely necessary as some might suggest and motivates a need for more physics based models to advance further . [ topological analysis ] in this section we provide a more detailed analysis of run d in particular . we examine the magnetic and thermodynamic structures realized by the lc model and provide a comparison to the standard sc model @xcite . an important advantage of including heat conduction in a global 3d enrionment is the ability to the study complex open / closed field topologies in a self consistent manner , which allows one to study the feedback effect that the magnetic topology has on heat distribution and vice versa . the most striking effect can be seen by examining the differences between heat conduction along open or closed field geometries . because closed field regions represent closed systems for the flow of thermal energy , they acheive higher temperatures than their open - field counterparts and shift into the soft - xray regime ( @xmath92mk ) . this natural correspondence can be clearly seen in figure [ fig : temperature_topology ] where we show a 3d surface at fixed temperature ( @xmath93 mk ) and an los image of soft - x - ray emission . it is also important to examine in detail the effect that the thermodynamic model has on the magnetic structure as a whole . in figure [ fig : magnetic_topology ] we show a comparison between closed field lines at @xmath94 for the pfssm initial condition @xcite , the standard sc model @xcite and run d. immediately obvious are the drastic changes between the closed field / streamer structure of the polytropic sc model and the thermodynamic lc model . unsurprisingly , the high temperatures and thermal energy introduced by coronal heating and conduction along the closed field produces a large thermal stress in the relatively high @xmath95 regions near the streamer cusps along the current sheet . this in turn leads to significant reorganization of the magnetic field as compared to a pfssm or polytropic mhd model . the fact that this sort of stress is highly sensitive to both the local nature of a given coronal heating model ( figure [ fig : streamer_br_slice ] ) and the global magnetic topology ( figure [ fig : magnetic_topology ] ) is a leading motivation for using a fully 3d , thermodynamic model when studying the structure of the global corona . in figure [ fig : euvcomp_run10towsa ] we show los euv and soft - x - ray image synthesis comparisons between run d and the standard sc model . in figure [ fig : eit_slice ] we show two quantitative slice comparisons cutting an average quiet sun region and the large active region . the improved agreement of run d in multiple filter bands is indicative of the improved temperature resolution possible with the lc model . while not an entirely fair comparison because the standard sc model does not include a full energy equation , we demonstrate the ability to quantitatively asses the relative agreement between a given model and observations in the euv regime . this sort of comparison also allows one to study in detail various choices of heating models and , though multi - filter comparison , provide a basis to resolve the inherent degeneracies between modifying the total coronal heating power and the heating scale height within the model . a further examination of some of the limitations imposed by euv synthesis from mhd models with finite resolution as well as future considerations can be found in appendix [ section : appendix ] . it is also instructive to examine the global temperature and density structure as a function of height for both run d and the standard sc model . in figures [ fig : slices_temperature ] and [ fig : slices_density ] we show electron temperature and number density at three separate heights in the low corona . important to emphasize is the fact that with a thermodynamic model , it is possible to resolve both significant changes in temperature with height and with respect to the magnetic field topology in the model . as expected , near the top of the transition region run d resolves cooler temperatures near the inversion lines due the short loop length , while simultaneously higher in the corona , hotter temperatures ( @xmath96 ) are achieved in close field regions , a natural result of energy flowing on a closed path via heat conduction . this greater sensitivity to field geometry ( something difficult for a polytropic model to do without extreme fine tuning of boundary conditions ) is critical when trying to correctly describe the conditions in the low corona . in this paper we demonstrate the clear need for a self - consistent treatment of the thermodynamic energy equation and boundary conditions when studying the global properties of the low corona with an mhd model . we observe that the interplay between coronal heating and electron heat conduction strongly governs the details of important strucures in the low corona , which is something that connot be described by extrapolation of the magnetic field on its own ( e.g. the significant stress of the streamer regions and current sheet in response to active region heating , figures [ fig : streamer_br_slice ] and [ fig : magnetic_topology ] ) . we also showcase the ability of the lc model to examine the effectiveness and applicability of various heating functions , and find that a simple exponential heating profile does quite well in describing the average structure of the quiet sun . + with this foundation , we believe that the lc model can provide a more realistic vehicle with which to study a wide range of aspects of the dynamic corona and solar wind . for example , the time - dependent dynamics of any sort of transient event in the corona , such as a cme or euv wave , depend critically on the magnetic and density structure of the corona ( density scale height being highly sensitive to temperature ) and therefore it is important to use improved models when determining the global ambient medium in which they evolve . additionally because of the general flexibility of the swmf , future coupling of the lc model to complementary solar wind models , such as the sc model based on a full description of alfvn wave turbulence currently in development ( * ? ? ? * in press ) , or one including a multi - species treatement to resolve electron / ion temperature decoupling in the high corona , has become a feasible scenario . ultimately , one thing is clear : with each passing month , year , and decade , our understanding of these fundamental process is constantly refined . in furthering this process it is critical to continue to improve the existing models and validate new efforts as they emerge . this work is primarily supported under the nasa earth and space science fellowship ( nessf ) program nasa nessf08-helio08f-0007 as well as grants : nsf atm-0639335 ( career ) and nasa nnx08aq16 g ( lws ) . computations were performed on the computing cluster at the institute for astronomy ( ifa ) advanced technology research center ( atrc ) pukalani , hi . c. downs would like to thank the entire swmf team for their innovative research and willing collaboration . + soho eit images courtesy of the soho eit consortium ; soho is a project of international cooperation between esa and nasa . yohkoh data courtesy of the nasa - supported yohkoh legacy archive at montana state university . one critical caveat when imaging the solar corona is that remote sensing images are fundamentally ambiguous in the sense that they are 2d realizations of a fully 3d , optically thin data set . for our purposes , it is best to frame the problem with the concepts of modeling an unresolved differential emission measure ( dem ) distribution and hydrostatic weighting bias , well described in @xcite . these issues arise due to two main facts . ( i ) the spatial resolution of current imaging instruments ( e.g. trace , soho , stereo ) all lie below the fundamental width of an isolated flow / field line . and ( ii ) any image is a line of sight integral over a range of heights , which are preferentially biased towards hotter flow lines with longer density scale heights . thus at further distance off of the limb , the observation may be biased towards flow lines that may not be energetically important at the surface . local dem distributions can also be directly observed through coupled filter and tomographic inversions @xcite . + though the los integral for image synthesis includes the range of temperatures and densities in cells along the line of sight , it can not fundamentally account for the dem distribution that is below the physical resolution of the model . in mhd , a given cell has a unique value for its thermodynamic properties , particularly @xmath97 , and @xmath98 , while a physical picture of the low @xmath95 multi - hydrostatic corona below the resolution limit can be treated as a superposition of densities and temperatures , each confined to an isothermal flow / field line . in this sense , the true dem distribution of a parcel of gas that contributes to an euv image ( e.g. the eit instrument ) is instead represented as a delta function in the simulation . one aspect of this limitation can be seen in difference between the data and synthesized limb profiles ( figure [ fig : eit_slice ] ) . because the observational weighting bias shifts to higher temperatures further with distance from the surface , the observed limb profile can not be represented in all four filters simultaneously with only the larger scale temperature equilibrium produced by the mhd model . + to further examine this issue in the context of the lc model we use filter ratios between the euv images , which are nominally sensitive to temperature differences . in figure [ fig : demcurves ] we show filter ratio values for soho eit observations on aug 27 , 1996 and the path in the filter ratio plane spanned by all possible filter ratios of the optically thin response functions @xmath99 and @xmath100 ( section [ section : los - synthesis ] ) . unsurprisingly , the euv data spans a much larger range than that allowed by a single temperature . to simply examine the effect that an unresolved dem distribution may have on the synthesis and subsequent comparison to observational data we construct an ad hoc dem by gaussian convolution of the filter response functions with t : @xmath101 and use @xmath102 for this discussion . while we by no means posit that this dem distribution should resemble the corona everywhere , it includes the basic temperature splitting features of those gaussian dem distributions calculated from observations ( e.g. @xcite . we also show the effect of this modification in the right panel of figure [ fig : demcurves ] . + the effect that this simple induced dem distribution has on the synthesized ratios is shown in figure [ fig : dem_effect_obs ] . it is clear comparing the limb tracks of the synthesized images to the temperature tracks in figure [ fig : demcurves ] that the los synthesis at the current model resolution does not produce a significant dem distribution on its own ( particularly at the limb ) . however , the improvement gained in terms of the relative location with respect to observations for the modified dem synthesis suggests that this effect should not be overlooked when constructing more detailed models in the future . high resolution studies designed to approach the fundamental scale of isothermal separation between coronal loops , particularly those involving local , time - dependent heating for example , should be able to address this fundamental issue . 41 natexlab#1#1 abbett , w. p. 2007 , , 665 , 1469 altschuler , m. d. , levine , r. h. , stix , m. , & harvey , j. 1977 , , 51 , 345 arge , c. , luhmann , j. , odstrcil , d. , schrijver , c. , & li , y. 2004 , journal of atmospheric and solar - terrestrial physics , 66 , 1295 , m. j. , alexander , d. , hurlburt , n. , newmark , j. s. , neupert , w. m. , klimchuk , j. a. , & gary , g. a. 2000 , , 531 , 1129 aschwanden , m. j. & nitta , n. 2000 , , 535 , l59 aschwanden , m. j. & acton , l. w. 2001 , , 550 , 475 aschwanden , m. j. & schrijver , c. j. 2002 , , 142 , 269 aschwanden , m. j. 2008 , in astronomical society of the pacific conference series , vol . 383 , subsurface and atmospheric influences on solar activity , ed . r. howe , r. w. komm , k. s. balasubramaniam , & g. j. d. petrie , 1 bercik , d. j. , fisher , g. h. , johns - krull , c. m. , , & abbett , w. p. 2005 , , 631 , 529 cohen , o. , sokolov , i. v. , roussev , i. i. , arge , c. n. , manchester , w. b. , gombosi , t. i. , frazin , r. a. , park , h. , et al . 2007 , , 654 , l163 cohen , o. , sokolov , i. v. , roussev , i. i. , & gombosi , t. i. 2008 , , 113 , a03104 , o. , attrill , g. d. r. , manchester , w. b. , & wills - davey , m. j. 2009 , , 705 , 587 delaboudinire , artzner , g. e. , brunaud , j. , gabriel , a. h. , hochedez , j. f. , millier , f. , song , x. y. , au , b. , et al . 1995 , , 162 , 291 frazin , r. a. , jacob , m. , manchester , w. b. , morgan , h. , & wakin , m. b. 2009 , , 695 , 636 freeland , s. l. & handy , b. n. 1998 , , 182 , 497 groth , c. p. t. , de zeeuw , d. l. , gombosi , t. i. , & powell , k. g. 2000 , , 105 , 25,053 , c. , roussev , i. i. , lugaz , n. , & poedts , s. 2009 , , 695 , l171 , e. , feldman , u. , & dere , k. p. 2002 , , 139 , 281 landi , e. , zanna , d. g. , young , p. r. , dere , k. p. , mason , h. e. , & landini , m. 2006 , , 162 , 261 lionello , r. , linker , j. a. , , & mikic , z. 2001 , , 546 , 542 lionello , r. , linker , j. a. , & mikic , z. 2009 , , 690 , 902 lugaz , n. , manchester , roussev , i. i. , tth , g. , & gombosi , t. i. 2007 , , 659 , 788 lugaz , n. , vourlidas , a. , roussev , i. i. , jacobs , c. , manchester , & cohen , o. 2008 , , 684 , l111 lugaz , n. , vourlidas , a. , roussev , i. , & morgan , h. 2009 , , 259 , 269 , z. , linker , j. a. , schnack , d. d. , lionello , r. , & tarditi , a. 1999 , physics of plasmas , 6 , 2217 mok , y. , mikic , z. , lionello , r. , , & linker , j. a. 2005 , , 621 , 1098 mok , y. , miki , z. , lionello , r. , & linker , j. a. 2008 , , 679 , l161 moses , d. , clette , f. , delaboudinire , j. p. , artzner , g. e. , bougnet , m. , brunaud , j. , carabetian , c. , gabriel , a. h. , et al . 1997 , , 175 , 571 oran , r. , sokolov , i. v. , roussev , i. i. , van der holst , b. , manchester , w. b. , & gombosi , t. i. 2009 , in astronomical society of the pacific conference series , vol . 406 , numerical modeling of space plasma flows : astronum-2009 proceedings of the 2nd international conference , ed . n. v. pogorelov , e. audit , p. colella , & g. p. zank parker , e. n. 1988 , , 330 , 474 powell , k. 1999 , j. comput . phys . , 154 , 284 roussev , i. i. , gombosi , t. i. , sokolov , i. v. , velli , m. , manchester , w. , dezeeuw , d. l. , liewer , p. , tth , g. , & luhmann , j. 2003 , , 595 , l57 , i. i. , sokolov , i. v. , forbes , t. g. , gombosi , t. i. , lee , m. a. , & sakai , j. i. 2004 , , 605 , l73 roussev , i. i. & sokolov , i. v. 2006 , washington dc american geophysical union geophysical monograph series , 165 , 89 roussev , i. i. , lugaz , n. , & sokolov , i. v. 2007 , . , 668 , l87 schrijver , c. j. , sandman , a. w. , aschwanden , m. j. , & derosa , m. l. 2004 , , 615 , 512 sokolov , i. v. , roussev , i. i. , skender , m. , gombosi , t. i. , & usmanov , a. v. 2009 , , 696 , 261 , l. 1965 , physics of fully ionized gases , 2nd rev . ( new york : interscience publication ) tth , g. , sokolov , i. v. , gombosi , t. i. , chesney , d. r. , clauer , r. c. , de zeeuw , d. l. , hansen , k. c. , kane , k. j. , et al . 2005 , , 110 , a12226 tsuneta , s. , acton , l. , bruner , m. , lemen , j. , brown , w. , caravalho , r. , catura , r. , freeland , et al . 1991 , , 136 , 37 withbroe , g. l. 1988 , , 325 , 442 right : the customized spherical grid with minimum @xmath103 ( part of this work ) . all three directions do not have to be further refined at the surface to enhance the radial resolution when using spherical geometry , leading to significant savings in computational overhead . in this example , both simulation domains have comparable numbers of total cells.,title="fig:",scaledwidth=35.0% ] . right : the customized spherical grid with minimum @xmath103 ( part of this work ) . all three directions do not have to be further refined at the surface to enhance the radial resolution when using spherical geometry , leading to significant savings in computational overhead . in this example , both simulation domains have comparable numbers of total cells.,title="fig:",scaledwidth=35.0% ] ) causes an extreme shift in the northward tilt of the heliospheric current sheet @xmath104 . simultaneously , the region not thermally connected to the active region on the opposite side of the sun remains unchanged . the color contours display the radial component of the magnetic field , @xmath105 . the chosen slice is parallel to the polar axis and intersects the inversion line between the large active region seen on aug 27 , 1996 . the 3d magnetic streamlines intersect the slice at the same location in each figure.,title="fig:",scaledwidth=42.0% ] ) causes an extreme shift in the northward tilt of the heliospheric current sheet @xmath104 . simultaneously , the region not thermally connected to the active region on the opposite side of the sun remains unchanged . the color contours display the radial component of the magnetic field , @xmath105 . the chosen slice is parallel to the polar axis and intersects the inversion line between the large active region seen on aug 27 , 1996 . the 3d magnetic streamlines intersect the slice at the same location in each figure.,title="fig:",scaledwidth=42.0% ] ) causes an extreme shift in the northward tilt of the heliospheric current sheet @xmath104 . simultaneously , the region not thermally connected to the active region on the opposite side of the sun remains unchanged . the color contours display the radial component of the magnetic field , @xmath105 . the chosen slice is parallel to the polar axis and intersects the inversion line between the large active region seen on aug 27 , 1996 . the 3d magnetic streamlines intersect the slice at the same location in each figure.,title="fig:",scaledwidth=8.0% ] and @xmath106 with color contours of t. the black contour lines of normal radial field direction @xmath107 are shown to illustrate how the hot , closed field streamer regions overlay the inversion lines on the surface.,title="fig:",scaledwidth=80.0% ] and @xmath106 with color contours of t. the black contour lines of normal radial field direction @xmath107 are shown to illustrate how the hot , closed field streamer regions overlay the inversion lines on the surface.,title="fig:",scaledwidth=13.0% ] and @xmath108 in each model.,title="fig:",scaledwidth=30.0% ] and @xmath108 in each model.,title="fig:",scaledwidth=30.0% ] and @xmath108 in each model.,title="fig:",scaledwidth=30.0% ] and @xmath108 in each model.,title="fig:",scaledwidth=30.0% ] and @xmath108 in each model.,title="fig:",scaledwidth=30.0% ] and @xmath108 in each model.,title="fig:",scaledwidth=30.0% ] and @xmath108 in each model.,title="fig:",scaledwidth=7.0% ] ( top ) , @xmath109 ( middle ) , and @xmath110 ( bottom)@xmath111 for run d ( left ) and the default sc model ( right ) for cr 1913 ( black lines at @xmath107 ) . the full thermodynamic energy equation is able to capture large changes in temperature over short changes in radius . as expected , near the top of the transition region we resolve cooler temperatures near the inversion lines due the short loop length . simultaneously higher in the corona , hotter temperatures ( @xmath96 ) are achieved in close field regions , a natural result of energy flowing on a closed path via heat conduction.,title="fig:",scaledwidth=44.0% ] ( top ) , @xmath109 ( middle ) , and @xmath110 ( bottom)@xmath111 for run d ( left ) and the default sc model ( right ) for cr 1913 ( black lines at @xmath107 ) . the full thermodynamic energy equation is able to capture large changes in temperature over short changes in radius . as expected , near the top of the transition region we resolve cooler temperatures near the inversion lines due the short loop length . simultaneously higher in the corona , hotter temperatures ( @xmath96 ) are achieved in close field regions , a natural result of energy flowing on a closed path via heat conduction.,title="fig:",scaledwidth=44.0% ] ( top ) , @xmath109 ( middle ) , and @xmath110 ( bottom)@xmath111 for run d ( left ) and the default sc model ( right ) for cr 1913 ( black lines at @xmath107 ) . the full thermodynamic energy equation is able to capture large changes in temperature over short changes in radius . as expected , near the top of the transition region we resolve cooler temperatures near the inversion lines due the short loop length . simultaneously higher in the corona , hotter temperatures ( @xmath96 ) are achieved in close field regions , a natural result of energy flowing on a closed path via heat conduction.,title="fig:",scaledwidth=44.0% ] ( top ) , @xmath109 ( middle ) , and @xmath110 ( bottom)@xmath111 for run d ( left ) and the default sc model ( right ) for cr 1913 ( black lines at @xmath107 ) . the full thermodynamic energy equation is able to capture large changes in temperature over short changes in radius . as expected , near the top of the transition region we resolve cooler temperatures near the inversion lines due the short loop length . simultaneously higher in the corona , hotter temperatures ( @xmath96 ) are achieved in close field regions , a natural result of energy flowing on a closed path via heat conduction.,title="fig:",scaledwidth=44.0% ] ( top ) , @xmath109 ( middle ) , and @xmath110 ( bottom)@xmath111 for run d ( left ) and the default sc model ( right ) for cr 1913 ( black lines at @xmath107 ) . the full thermodynamic energy equation is able to capture large changes in temperature over short changes in radius . as expected , near the top of the transition region we resolve cooler temperatures near the inversion lines due the short loop length . simultaneously higher in the corona , hotter temperatures ( @xmath96 ) are achieved in close field regions , a natural result of energy flowing on a closed path via heat conduction.,title="fig:",scaledwidth=44.0% ] ( top ) , @xmath109 ( middle ) , and @xmath110 ( bottom)@xmath111 for run d ( left ) and the default sc model ( right ) for cr 1913 ( black lines at @xmath107 ) . the full thermodynamic energy equation is able to capture large changes in temperature over short changes in radius . as expected , near the top of the transition region we resolve cooler temperatures near the inversion lines due the short loop length . simultaneously higher in the corona , hotter temperatures ( @xmath96 ) are achieved in close field regions , a natural result of energy flowing on a closed path via heat conduction.,title="fig:",scaledwidth=44.0% ] ( top ) , @xmath109 ( middle ) , and @xmath110 ( bottom)@xmath111 for run d ( left ) and the default sc model ( right ) for cr 1913 ( black lines at @xmath107 ) . the full thermodynamic energy equation is able to capture large changes in temperature over short changes in radius . as expected , near the top of the transition region we resolve cooler temperatures near the inversion lines due the short loop length . simultaneously higher in the corona , hotter temperatures ( @xmath96 ) are achieved in close field regions , a natural result of energy flowing on a closed path via heat conduction.,title="fig:",scaledwidth=7.0% ] ( top ) , @xmath109 ( middle ) , and @xmath110 ( bottom)@xmath111 for run d ( left ) and the standard sc model ( right ) for cr 1913 ( black lines at @xmath107 ) . this demonstrates the high dynamic range in density over small changes in radius resolved with the lc model.,title="fig:",scaledwidth=44.0% ] ( top ) , @xmath109 ( middle ) , and @xmath110 ( bottom)@xmath111 for run d ( left ) and the standard sc model ( right ) for cr 1913 ( black lines at @xmath107 ) . this demonstrates the high dynamic range in density over small changes in radius resolved with the lc model.,title="fig:",scaledwidth=44.0% ] ( top ) , @xmath109 ( middle ) , and @xmath110 ( bottom)@xmath111 for run d ( left ) and the standard sc model ( right ) for cr 1913 ( black lines at @xmath107 ) . this demonstrates the high dynamic range in density over small changes in radius resolved with the lc model.,title="fig:",scaledwidth=44.0% ] ( top ) , @xmath109 ( middle ) , and @xmath110 ( bottom)@xmath111 for run d ( left ) and the standard sc model ( right ) for cr 1913 ( black lines at @xmath107 ) . this demonstrates the high dynamic range in density over small changes in radius resolved with the lc model.,title="fig:",scaledwidth=44.0% ] ( top ) , @xmath109 ( middle ) , and @xmath110 ( bottom)@xmath111 for run d ( left ) and the standard sc model ( right ) for cr 1913 ( black lines at @xmath107 ) . this demonstrates the high dynamic range in density over small changes in radius resolved with the lc model.,title="fig:",scaledwidth=44.0% ] ( top ) , @xmath109 ( middle ) , and @xmath110 ( bottom)@xmath111 for run d ( left ) and the standard sc model ( right ) for cr 1913 ( black lines at @xmath107 ) . this demonstrates the high dynamic range in density over small changes in radius resolved with the lc model.,title="fig:",scaledwidth=44.0% ] ( top ) , @xmath109 ( middle ) , and @xmath110 ( bottom)@xmath111 for run d ( left ) and the standard sc model ( right ) for cr 1913 ( black lines at @xmath107 ) . this demonstrates the high dynamic range in density over small changes in radius resolved with the lc model.,title="fig:",scaledwidth=7.0% ] ( right ) . the blue and green points represent euv data on an off the limb ( binned to 1000 points each ) . the curve path follows the ratio values [ @xmath99 and @xmath112 $ ] from @xmath113k , the width of the curves represents the variation of the ratio values from @xmath114 at a given temperature . this is intended to illustrate the large variation of filter ratio values with respect to an unresolved dem distribution.,title="fig:",scaledwidth=48.5% ] ( right ) . the blue and green points represent euv data on an off the limb ( binned to 1000 points each ) . the curve path follows the ratio values [ @xmath99 and @xmath112 $ ] from @xmath113k , the width of the curves represents the variation of the ratio values from @xmath114 at a given temperature . this is intended to illustrate the large variation of filter ratio values with respect to an unresolved dem distribution.,title="fig:",scaledwidth=49.0% ] ( right ) . while the location of the synthesized filter ratios improves with this dem modification , the fine details of the local dem distribution , realized by the variation of observed eit filter ratio values , are difficult to reproduce with an mhd model involving large , global scales.,title="fig:",scaledwidth=49.0% ] ( right ) . while the location of the synthesized filter ratios improves with this dem modification , the fine details of the local dem distribution , realized by the variation of observed eit filter ratio values , are difficult to reproduce with an mhd model involving large , global scales.,title="fig:",scaledwidth=49.0% ]

starting from the 2007 crisis , one of the most striking features of today s financial environment is represented by the segmentation of interest rate markets . more specifically , while the pre - crisis interest rate market was well described by a _ single _ yield curve and interbank xibor rates associated to different tenors were simply determined by no - arbitrage relations , nowadays the market is segmented in the sense that distinct yield curves are constructed from market instruments that depend on a specific tenor , thus giving rise to _ multiple _ yield curves . the credit and liquidity risks existing in the interbank market , which were perceived as negligible before the crisis , are at the origin of this phenomenon . this is also reflected by the emergence of spreads between ois rates and xibor rates as well as between xibor rates associated to different tenors . in this paper , we propose the most general tractable approach based on affine processes to model multiple yield curves . in our view , it should be understood as _ the affine framework _ for this modeling purpose and not as a specific model class , such as the affine libor models considered in @xcite . we provide a unifying affine methodology within which all interest rate modeling approaches , i.e. , ( i ) short rate , ( ii ) libor market , and ( iii ) heath - jarrow - morton modeling , can be consolidated . on the one hand , this allows us to embed in our framework all affine multiple curve models proposed so far in the literature . on the other hand , the approach allows for novel developments , for instance multiple curve short rate models based on wishart processes as well as interest rate models under alternative numraires . moreover , the generality of the setting does not preclude tractability and appealing practical features , as will be made precise below . we model a general numraire process and , inspired by our previous work @xcite , _ multiplicative spreads _ between ( normalized ) spot xibor rates and ( normalized ) simply compounded ois rates . besides being directly observable from market quotes , multiplicative spreads represent the most convenient modeling quantity in an affine setting and admit a natural interpretation in terms of forward exchange premia . we consider a general interest rate market where ois zero - coupon bonds and fra contracts are traded , for a finite set of tenors and for all maturities , and assume the existence of a numraire - martingale measure couple , thus ensuring absence of arbitrage in a sense precisely specified below . we then model the logarithm of the multiplicative spreads and of the numraire process as affine functions of an underlying affine process . in the special case where the numraire is chosen as the ois bank account , our framework can be regarded as the natural extension of classical affine short rate models to the multi - curve setting . in the special case where the numraire is chosen to be the ois bond with terminal maturity , we can recover an extension of the affine libor market model of @xcite . the proposed framework fully exploits the analytical tractability of affine processes and exhibits several desirable modeling features . indeed , besides its generality , it allows for * an automatic fit to the initially observed term structures of ois bond prices and spreads ; * a transparent characterization of order relations between spreads associated to different tenors ; * efficient valuation formulae for caplets and swaptions in a multi - curve setting . the driving affine process is allowed to take values in a general convex state space . in particular , we analyze a specification based on wishart processes and we derive a novel closed - form formula for the price of a caplet , expressed in terms of the distribution function of weighted sums of independent non - central chi - square - distributed random variables . beyond that , we test the empirical performance of two simple specifications of our setup and show that they achieve a satisfactory fit to market data . referring to the introductory section of @xcite and to the monographies @xcite for a more detailed overview of the relevant literature , we mention here that affine processes have been used to model multiple yield curves in @xcite by adopting a short rate approach and , more recently , in @xcite by extending the affine libor model originally introduced in @xcite . as already mentioned , the flexibility of our approach is highlighted by the fact that , to the best of our knowledge , all existing affine multi - curve models can be recovered as special cases of our framework . for this reason , the present work structurally differs from papers on specific affine modeling approaches , as previously considered in the literature . beyond that , it also provides a broad class of tractable specifications of the abstract hjm approach based on general it semimartingales proposed in @xcite for arbitrage - free modeling of the term structures of ois bonds and multiplicative spreads . the paper is organized as follows . section [ sec : intro2 ] starts by introducing the main market rates and multiplicative spreads and then presents an abstract approach to the arbitrage - free modeling of a general multi - curve interest rate market . in section [ affinespecification ] , we develop a general framework driven by affine processes as well as possible specifications based on a short rate approach . moreover , we relate our setup to several affine multi - curve models recently proposed in the literature . in section [ sec : aff_pricing ] , we derive general semi - closed valuation formulae for caplets and an approximate formula for the price of a swaption . section [ sec : wishart ] contains a detailed analysis of a tractable specification based on wishart processes , deriving an analytical pricing formula for caplets . section [ sec : aff_examples ] presents the calibration results for two specifications of the framework . finally , appendix [ sec : gen_pricing ] presents general pricing formulae in terms of the quantities used in our framework and appendix [ app : proof_prop ] contains the proof of proposition [ prop : bonds_spreads ] . the underlying quantities of most interest rate products are represented by xibor rates . we denote by @xmath0 the xibor rate prevailing at date @xmath1 for the time interval @xmath2 $ ] , where the tenor @xmath3 is typically one day ( 1d ) , one week ( 1w ) , or several months ( 1 m , 2 m , 3 m , 6 m or 12 m ) . referring to ( * chapter 1 ) for a more detailed presentation of post - crisis interest rate markets , we just mention that xibor rates are reference rates determined by a panel of primary financial institutions for unsecured lending and are not based on actual transactions . in the post - crisis fixed income market , xibor rates associated to different tenors started to exhibit distinct behaviors , leading for instance to non - negligible basis spreads ( see appendix a.1 ) . in this paper , we shall consider xibor rates for a generic set of tenors @xmath4 , with @xmath5 , for some @xmath6 . the reference rates for overnight borrowing ( i.e. , for the shortest tenor of 1d ) correspond to the eonia ( euro overnight index average ) rate in the eurozone and to the federal funds rate in the us market . unlike xibor rates , the eonia and the federal funds rates are determined on the basis of actual overnight transactions in the interbank market by a panel of banks ( in the case of the eonia rate , this is the same panel determining the euribor rate ) . overnight rates represent the underlying of overnight indexed swaps ( ois ) , whose market swap rates are referred to as ois rates ( see appendix [ sec : noopt ] ) , typically considered as the best proxy for risk - free rates in market practice . by bootstrapping techniques ( see @xcite ) , ois rates allow to recover the ois term structure @xmath7 , where @xmath8 denotes the price at date @xmath1 of an ois zero - coupon bond with maturity @xmath9 . we define the simply compounded ( risk - free ) ois spot rate by @xmath10 for @xmath3 . in particular , note that the right - hand side of the above formula corresponds to the pre - crisis textbook definition of xibor rate . our main modeling quantities are the following _ spot multiplicative spreads _ : @xmath11 corresponding to multiplicative spreads between ( normalized ) xibor rates and ( normalized ) simply compounded ois spot rates . the multiplicative spreads @xmath12 can be directly inferred from the quoted xibor and ois rates . indeed , the numerator of is determined by the xibor rate quoted on the market , while the quantity appearing in the denominator can be bootstrapped from quoted ois rates , as mentioned above . in the post - crisis environment , the quantities @xmath12 are usually greater than one and increasing with respect to the tenor s length @xmath13 . neglecting liquidity issues , this is related to the fact that xibor rates embed the risk that the average credit quality of the banks included in the xibor panel deteriorates over the term of the loan , while ois rates reflect the average credit quality of a newly refreshed xibor panel ( see , e.g. , @xcite ) . as will be shown below , a key feature of our modeling framework is the facility of generating multiplicative spreads satisfying such requirements . in comparison to additive spreads ( as considered for instance in @xcite ) , multiplicative spreads admit a natural economic interpretation in a multiple curve setting and , as shown in the following sections , represent a convenient modeling quantity in relation with affine processes . referring to ( * ? ? ? * appendix b ) for full details , xibor rates can be associated with artificial risky bond prices @xmath14 , so that @xmath15 , for @xmath16 . if risky bonds are interpreted as bonds of a foreign economy , with @xmath14 representing the price ( in units of the foreign currency ) of a foreign zero - coupon bond , and ois bonds are interpreted as domestic bonds , then the quantity @xmath12 corresponds to the _ forward exchange premium _ between the domestic and the foreign economy over the period @xmath17 $ ] . in this sense , the multiplicative spread @xmath12 can be regarded as a market expectation ( at date @xmath1 ) of the riskiness of the xibor panel over the period @xmath17 $ ] . related foreign exchange analogies have been proposed in @xcite and @xcite . let us now make precise the financial market considered in this paper . among all financial contracts written on xibor rates , forward rate agreements ( fra ) can be rightfully considered as the basic building blocks , due to the simplicity of their payoff and to the fact that all linear interest rate derivatives ( such as interest rate swaps and basis swaps ) can be represented as portfolios of fras . as usually done in the literature , we consider standard `` textbook '' fras , in contrast to the cash - settled fras traded in the market ( compare with ( * ? ? ? * remark 1.3 ) ) . while the payoff of market fras is non - linear , the discrepancy between the two versions of fras is typically negligible for practical purposes , as documented in @xcite . motivated by the above observation , we shall consider a general financial market consisting of the following families of _ basic traded assets _ , with @xmath18 denoting a fixed time horizon : * ois zero - coupon bonds for all maturities @xmath19 $ ] ; * fra contracts for all maturities @xmath19 $ ] and for all tenors @xmath4 . note that , differently from the pre - crisis single - curve setting , fra contracts have to be considered on top of ois bonds since they can not be perfectly replicated by the latter any longer , due to the discrepancy between xibor rates @xmath20 and simply compounded ois rates @xmath21 . moreover , ois bonds play a particularly important role in collateralized transactions . indeed , the collateral rate adopted in collateralized transactions in interest rate markets is typically specified as the ois short rate and , hence , ois bonds represent the natural discount factors for the valuation of interest rate products ( see , e.g. , @xcite and ( * ? ? ? * appendix a ) ) . we denote by @xmath22 the price at date @xmath1 of a fra contract starting at date @xmath9 with maturity @xmath23 , rate @xmath24 and unitary notional , for @xmath25 and @xmath16 . we denote by @xmath26 the corresponding _ fair fra rate _ , i.e. , the rate @xmath27 fixed at date @xmath1 satisfying @xmath28 . we say that a fra contract written at @xmath1 is _ fair _ if its rate @xmath27 is set equal to @xmath26 , i.e. , if the price of the fra contract at the inception date @xmath1 is zero . extending definition of the spot multiplicative spread , we define ( forward ) _ multiplicative spreads _ @xmath29 , for @xmath25 and @xmath16 , by @xmath30 multiplicative spreads of the form have been first introduced in @xcite and further considered as main modeling quantities in @xcite ( see also @xcite ) . we refer to ( * ? ? ? * appendix b ) for the interpretation of the forward multiplicative spread @xmath29 in terms of the foreign exchange analogy introduced at the end of section [ sec : rates ] . a first and fundamental issue is the absence of arbitrage in the financial market composed by the basic traded assets introduced above . to this end , we let @xmath31 be a measurable space endowed with a right - continuous filtration @xmath32 and formulate the following definition . [ def : couple ] let @xmath33 be a probability measure on @xmath31 and @xmath34 a strictly positive adapted process with @xmath35 . we say that @xmath36 is a _ numraire - martingale measure couple _ if the @xmath37-discounted price of every basic traded asset is a martingale on @xmath38 . in other words , if @xmath36 is a numraire - martingale measure couple , the price of every ois zero - coupon bond and fra contract is a @xmath33-martingale when discounted with respect to the numraire process @xmath37 . definition [ def : couple ] is in the spirit of @xcite , where the authors consider term structure models generated by a numraire - martingale measure couple in the above sense . note that we do not assume that @xmath37 is a traded asset and it can therefore be interpreted as a state - price density process when @xmath39 corresponds to the physical probability measure ( compare with @xcite ) . in the present paper , we shall work under the following standing assumption . [ ass : couple ] there exists a numraire - martingale measure couple @xmath36 . in particular , assumption [ ass : couple ] suffices to exclude the existence of arbitrage profits , in the sense of _ no asymptotic free lunch with vanishing risk _ ( naflvr , see @xcite ) , in the large financial market composed by all basic traded assets , i.e. , by all ois zero - coupon bonds and fra contracts . [ rem : ois_num ] in the literature on multiple curves , the numraire is typically chosen to be the ois bank account and @xmath33 is the corresponding ( spot ) martingale measure . while this choice of @xmath36 will be considered in detail in section [ sec : short_rate ] , we want to point out that our framework is not limited to this specification and allows to consider several alternative settings , depending on the choice of the numraire - martingale measure couple ( see section [ sec : aff_model ] below ) . for the moment , we just think of the couple @xmath36 in abstract and general terms . assumption [ ass : couple ] implies that the arbitrage - free price of every basic traded asset can be computed by taking the conditional @xmath33-expectation of its @xmath37-discounted payoff , similarly as in the classical risk - neutral valuation paradigm is the ois bank account and @xmath33 the corresponding martingale measure , as considered in remark [ rem : ois_num ] , computing the price of a derivative by taking the conditional @xmath33-expectation of its @xmath37-discounted payoff corresponds to computing its _ clean price _ , i.e. , neglecting counterparty risk and assuming a funding rate equal to the ois rate ( compare also with the discussion in ( * ? ? ? * appendix a ) ) . ] . throughout the paper , unless explicitly indicated , expectations are taken under the measure @xmath33 , while the measure @xmath40 denotes the @xmath9-forward measure with density @xmath41 , for @xmath19 $ ] . we can now state the following proposition , which is a simple consequence of definition [ def : couple ] ( for completeness , the proof is given in appendix [ app : proof_prop ] ) . [ prop : bonds_spreads ] suppose that assumption [ ass : couple ] holds . then the following hold : 1 . ois zero - coupon bond prices satisfy @xmath42 , \qquad \text { for all } 0\leq t\leq t\leq { \mathbb{t}};\ ] ] 2 . for every @xmath16 , the fair fra rate satisfies @xmath43 , \qquad \text { for all } 0\leq t\leq t\leq { \mathbb{t}};\ ] ] 3 . for every @xmath16 , the multiplicative spread satisfies @xmath44 , \qquad \text { for all } 0\leq t\leq t\leq { \mathbb{t}}.\ ] ] from the modeling perspective , proposition [ prop : bonds_spreads ] together with representation implies that , in order to give an arbitrage - free description of the financial market composed by all ois zero - coupon bonds and fra contracts , it suffices to model the numraire process @xmath37 together with the spot multiplicative spreads @xmath45 , under some reference measure @xmath33 . in the next section , this will be achieved in a flexible and tractable way by letting the numraire process and the spot multiplicative spreads be driven by a common underlying affine process . in this section we develop a general framework based on affine processes for modeling multiple curves . we first recall in section [ sec : aff_gen ] some general results on affine processes , mainly relying on @xcite . in section [ sec : aff_model ] , we provide the general definition of an affine multi - curve model with respect to an abstract numraire - martingale measure couple and derive its fundamental properties . this general setup is then specialized in section [ sec : short_rate ] to the typical setting where the numraire is given by the ois bank account modeled via a short rate . section [ sec : relations ] shows how our setup relates to the existing multi - curve models driven by affine processes . let @xmath46 be a finite - dimensional real vector space with associated scalar product @xmath47 . we denote by @xmath48 a non - empty closed convex subset of @xmath46 , which will serve as state space for the stochastic process to be introduced below , endowed with the borel @xmath49-algebra @xmath50 . typical choices for the state space are represented by @xmath51 or combinations thereof , for some @xmath52 , where @xmath53 denotes the cone of symmetric @xmath54 positive semidefinite matrices . and @xmath55 , the scalar product is given by @xmath56 $ ] , with @xmath57 $ ] being the trace operator . ] on a filtered probability space @xmath58 satisfying the usual conditions , we introduce a stochastic process @xmath59 , starting from an initial value @xmath60 belonging to the interior of @xmath48 . the process @xmath61 is assumed to be an adapted , time - homogeneous and conservative markov process taking values in @xmath48 and we denote by @xmath62;t\in[0,{\mathbb{t}}]\}$ ] the family of its transition kernels . to the process @xmath61 we associate the sets @xmath63<+\infty , \text { for all } t\in[0,t]\bigr\ } \qquad\text { and } \qquad { \mathfrak{u}}:={\mathfrak{u}}_{\mathbb{t}}.\end{aligned}\ ] ] letting @xmath64\times(v+{\ensuremath{\mathsf{i}}}v ) : \zeta\in{\mathfrak{u}}_t\}$ ] , we define as follows the class of affine processes . [ def : affine ] the markov process @xmath65 is said to be _ affine _ if 1 . it is stochastically continuous , i.e. , the transition kernels satisfy @xmath66 weakly on @xmath48 , for every @xmath67\times d$ ] ; 2 . its fourier - laplace transform has exponential affine dependence on the initial state , i.e. , there exist functions @xmath68 and @xmath69 such that , for every initial value @xmath70 and for every @xmath71 , it holds that @xmath72 = \int_de^{\langle u,\xi\rangle}p_t(x , d\xi)=e^{\phi(t , u)+\langle\psi(t , u),x\rangle}.\label{affineproperty}\end{aligned}\ ] ] [ rem : convex_expmom ] 1 . definition [ def : affine ] slightly differs from the usual definition of affine process where the exponential affine form is only assumed to hold on the set of bounded exponentials . indeed , in most theoretical works the definition of affine process only refers to the characteristic function or laplace transform ( for processes with values in a cone ) . then a natural question arises , namely if the exponential affine form can be extended beyond the characteristic function or laplace transform , in our notation to @xmath73 . this important question has been answered positively in @xcite under general assumptions and , therefore , we modified the classical definition . in particular , in view of ( * ? ? ? * lemma 4.2 ) , it holds that @xmath74<+\infty\bigr\ } , \qquad\text { for all } t\in[0,{\mathbb{t}}].\ ] ] 2 . at the expense of greater technicalities , our setup can be extended to affine processes taking values in general ( possibly non - convex ) state spaces . however , the class of affine processes considered in definition [ def : affine ] is sufficiently rich to cover any affine multi - curve model of practical interest . the affine property together with the chapman - kolmogorov equation implies that the functions @xmath75 and @xmath76 have the _ semiflow property _ , i.e. , for every @xmath77 and @xmath78 $ ] with @xmath79 , it holds that @xmath80 the stochastic continuity of the affine process @xmath61 also implies its _ regularity _ , according to ( * ? ? ? * theorem 7 ) ( see also ( * ? ? ? * theorem 4.3 ) in the case @xmath81 , for some @xmath82 , and ( * ? ? ? * proposition 3.4 ) in the case @xmath83 ) , in the sense that the derivatives @xmath84 exist and are continuous at @xmath85 . regularity implies that it is possible to differentiate the semiflow relations with respect to @xmath1 and evaluate them at @xmath86 , thus obtaining the following system of generalized riccati odes @xmath87 we state the following result ( see ( * ? ? ? * theorem 4.10 ) ) , which will be useful for some specifications of our setup . we refer to @xcite for the notion of differential characteristics of an affine process and we denote by @xmath88 and @xmath89 two real vector spaces . [ mixedstate ] let @xmath90 be an affine process taking values in a state space of the form @xmath91 , with initial value @xmath92 and such that the differential characteristics of @xmath93 only depend on @xmath94 . then there exist functions @xmath95 and @xmath96 such that , for every @xmath97 , it holds that @xmath98=e^{\tilde{\phi}(t , u_1,u_2)+\langle \tilde{\psi}(t , u_1,u_2),x^1\rangle+\langle u_2,x^2\rangle}.\ ] ] typical examples where affine processes satisfying the assumptions of lemma [ mixedstate ] are employed are short rate models ( see section [ sec : short_rate ] ) and affine stochastic volatility models ( see , e.g. , ( * ? ? ? * chapter 5 ) ) . as explained in section [ sec : abstract_setting ] , our general setup assumes the existence of a numraire - martingale measure couple @xmath36 , meaning that the @xmath37-discounted price of every traded asset is a @xmath33-martingale . recall also that we consider as basic traded assets the ois zero - coupon bonds and the fra contracts , for all maturities @xmath19 $ ] and for a finite set of tenors @xmath4 , with @xmath5 , for some @xmath6 . letting @xmath65 be an affine process taking values in a state space @xmath99 under a measure @xmath33 , we are now in a position to give the general definition of an _ affine multi - curve model_. [ def : aff_model ] let @xmath100 be a family of functions @xmath101\rightarrow v$ ] , @xmath102 , such that @xmath103 and @xmath104 , for every @xmath16 and @xmath105 $ ] , and @xmath106 a family of functions @xmath107\rightarrow{\mathbb{r}}$ ] , @xmath102 . we say that the triplet @xmath108 is an _ affine multi - curve model _ if 1 . the logarithm of the numraire process @xmath109 satisfies @xmath110;\ ] ] 2 . the logarithm of the spot multiplicative spreads @xmath111 satisfies @xmath112 \text { and } i=1,\ldots , m.\ ] ] depending on the choice of the numraire - martingale measure couple @xmath36 and of the triplet @xmath108 , different modeling approaches can be obtained from the present general setup . for instance : * if the numraire asset @xmath37 is chosen to be the ois bank account , @xmath33 is the corresponding ( spot ) martingale measure and @xmath37 is modeled via an ois short rate @xmath113 , then definition [ def : aff_model ] embeds an affine framework for modeling the ois short rate and the multiplicative spreads . this specification will be analyzed in detail in section [ sec : short_rate ] ( compare also with section [ sec : rel_short_rate ] ) ; * if the ois zero - coupon bond with maturity @xmath114 is chosen as numraire and @xmath33 is the @xmath114-forward measure , then definition [ def : aff_model ] yields an extension of the affine libor model with multiple curves recently introduced in @xcite ( see section [ sec : rel_affinelibor ] ) ; * if @xmath33 is the physical probability measure and @xmath37 is chosen as the _ growth - optimal portfolio _ ( in the spirit of the _ benchmark approach _ to interest rate modeling , see , e.g. , @xcite and , for a general account of the benchmark approach , @xcite ) or , more generally , as a _ state - price density _ ( as recently considered in the rational multi - curve models @xcite ) , then definition [ def : aff_model ] leads to a _ real - world _ approach to the modeling of multiple curves . affine multi - curve models lead to a tractable structure for ois bond prices and multiplicative spreads . more precisely , affine multi - curve models can be characterized via the exponentially affine form of ois bond prices and spreads . to this end , let us give the following definition . [ def : qaffinemodel ] the affine process @xmath61 is said to _ generate exponentially affine ois bond prices and spreads _ if 1 . the @xmath37-discounted prices of ois zero - coupon bonds satisfy @xmath115 for some functions @xmath116\times[0,{\mathbb{t}}]\rightarrow{\mathbb{r}}$ ] and @xmath117\times[0,{\mathbb{t}}]\rightarrow v$ ] , and 2 . the multiplicative spreads satisfy @xmath118 for some functions @xmath119\times[0,{\mathbb{t}}]\rightarrow{\mathbb{r}}$ ] and @xmath120\times[0,{\mathbb{t}}]\rightarrow v$ ] , for all @xmath16 . the following proposition provides the announced characterization of affine multi - curve models , thereby showing that definition [ def : aff_model ] is equivalent to definition [ def : qaffinemodel ] . [ prop : qaffinecharacterization ] let @xmath61 be an affine process . then the following hold : 1 . if @xmath108 is an affine multi - curve model , then @xmath61 generates exponentially affine ois bond prices and spreads and the functions @xmath121 , @xmath122 , @xmath102 , appearing in - are given by @xmath123 for all @xmath25 and @xmath16 , where @xmath75 and @xmath76 denote the characteristic exponents of @xmath61 as in definition [ def : affine ] ; 2 . conversely , if @xmath61 generates exponentially affine ois bond prices and spreads , then @xmath108 is an affine multi - curve model , where the functions @xmath124 , @xmath102 , are given by @xmath125 for all @xmath105 $ ] and @xmath16 . suppose first that @xmath108 is an affine multi - curve model . then , in view of proposition [ prop : bonds_spreads ] , ois bond prices are given by @xmath126 = { \mathbb{e}}\left[e^{v_0(t)+\langle u_0(t),x_t\rangle}\bigr|{\mathcal{f}}_t\right ] = e^{v_0(t)+\phi(t - t , u_0(t))+\langle\psi(t - t , u_0(t)),x_t\rangle},\ ] ] where the last equality follows from together with the assumption that @xmath127 ( see also remark [ rem : convex_expmom ] ) . this proves that representation holds with the functions @xmath128 and @xmath129 being given as in . similarly , proposition [ prop : bonds_spreads ] together with the assumption that @xmath130 , for every @xmath19 $ ] and @xmath16 , implies that the multiplicative spreads satisfy @xmath131 = { \mathbb{e}}^{{\mathbb{q}}^t}\left[e^{v_i(t)+\langle u_i(t),x_t\rangle}\bigr|{\mathcal{f}}_t\right ] \\ & = \frac{b_t}{b(t , t)}{\mathbb{e}}\left[e^{v_i(t)+v_0(t)+\langle u_i(t)+u_0(t),x_t\rangle}\bigr|{\mathcal{f}}_t\right ] \\ & = e^{v_i(t)+\phi(t - t , u_i(t)+u_0(t))-\phi(t - t , u_0(t))+\langle\psi(t - t , u_i(t)+u_0(t))-\psi(t - t , u_0(t)),x_t\rangle},\end{aligned}\ ] ] thus showing that representation holds with the functions @xmath121 and @xmath122 being given as in . conversely , suppose that @xmath61 generates exponentially affine ois bond prices and spreads . then , for every @xmath105 $ ] , letting @xmath132 in - and recalling that @xmath133 , it follows that @xmath134 defining the functions @xmath135 and @xmath136 as in , this proves that representations - hold true . it remains to prove that @xmath137 and @xmath138 , for every @xmath19 $ ] and @xmath16 . this follows from proposition [ prop : bonds_spreads ] , since @xmath139 & = e^{-v_0(t)}{\mathbb{e}}\left[\frac{1}{b_t}\right ] = e^{-v_0(t)}b(0,t ) < + \infty , \\ { \mathbb{e}}\left[e^{\langle u_i(t)+u_0(t),x_t\rangle}\right ] & = e^{-v_i(t)-v_0(t)}{\mathbb{e}}\left[\frac{s^{\delta_i}(t , t)}{b_t}\right ] = e^{-v_i(t)-v_0(t)}b(0,t){\mathbb{e}}^{{\mathbb{q}}^t}[s^{\delta_i}(t , t ) ] \\ & = e^{-v_i(t)-v_0(t)}b(0,t)s^{\delta_i}(0,t ) < + \infty.\end{aligned}\ ] ] together with remark [ rem : convex_expmom ] , this implies that @xmath127 and @xmath130 , for all @xmath19 $ ] and @xmath16 , thus proving that @xmath108 is an affine multi - curve model . in particular , proposition [ prop : qaffinecharacterization ] implies that affine multi - curve models admit explicit valuation formulas for all linear interest rate derivatives ( i.e. , fra contracts , interest rate swaps , overnight swaps and basis swaps ) , since their prices can always be expressed in terms of ois bond prices and multiplicative spreads , as detailed in appendix [ sec : noopt ] . a major issue when modeling multiple curves consists in characterizing when the multiplicative spreads are greater than one and ordered with respect to the tenor s length , as it is the case in typical market scenarios . to this effect , we can establish the following result ( compare also with ( * ? ? ? * corollary 3.17 ) ) . [ prop : x0y ] let @xmath108 be an affine multi - curve model . suppose that the process @xmath61 is of the form @xmath140 and takes values in a state space @xmath141 , with @xmath142 a cone . suppose that @xmath143 , for all @xmath105 $ ] and that the functions @xmath135 are of the form @xmath144 with @xmath145\rightarrow c_z^*$ ] , where @xmath146 denotes the dual cone of @xmath142 , for all @xmath16 . then @xmath147 for all @xmath148 and @xmath149 . moreover , if in addition @xmath150,\ ] ] with @xmath151 denoting the partial order on @xmath146 , then it holds that @xmath152 the assertion is a direct consequence of proposition [ prop : bonds_spreads ] , stating that the process @xmath153 is a @xmath40-martingale , for every @xmath16 and @xmath19 $ ] . it thus satisfies @xmath154={\mathbb{e}}^{{\mathbb{q}}^t}[e^{v_i(t ) + \langle w_i(t ) , z_t \rangle}|{\mathcal{f}}_t ] , \qquad \text { for all } t\in[0,{\mathbb{t}}].\ ] ] besides the modeling flexibility and tractability ensured by affine processes , this facility of generating spreads which are greater than one and ordered with respect to the tenor s length represents one of the main advantages of specifying multiplicative spreads via . the family of functions @xmath106 appearing in definition [ def : aff_model ] can always be chosen in such a way to automatically achieve an exact fit to the initially observed term structures of ois bond prices and multiplicative spreads . we denote by @xmath155 the price of an ois zero - coupon bond with maturity @xmath9 observed on the market at the initial date @xmath86 and , similarly , by @xmath156 the multiplicative spread of tenor @xmath13 observed on the market at @xmath86 , for @xmath16 and @xmath19 $ ] . we say that _ an exact fit to the initially observed term structures is achieved _ if @xmath157 \text { and } i=1,\ldots , m.\ ] ] the following proposition provides a necessary and sufficient condition for an exact fit to hold . if @xmath108 is an affine multi - curve model , we denote by @xmath158 and @xmath159 , for @xmath25 and @xmath16 , the theoretical bond prices and spreads computed according to the model @xmath160 via - , i.e. , with all the functions @xmath136 , @xmath102 , set equal to zero . [ prop : fitting ] an affine multi - curve model @xmath108 achieves an exact fit to the initially observed term structures if and only if the family of functions @xmath106 satisfies @xmath161,\\ v_i(t ) & = \log s^{m,\delta_i}(0,t ) - \log s^{0,\delta_i}(0,t ) , & \qquad & \text { for all } t\in[0,{\mathbb{t } } ] \text { and } i=1,\ldots , m,\end{aligned}\ ] ] with @xmath162 and @xmath163 denoting the theoretical bond prices and spreads computed according to the affine multi - curve model @xmath160 via - , for @xmath105 $ ] and @xmath16 . the claim follows by noting that , for all @xmath105 $ ] and @xmath16 , @xmath164 = e^{v_0(t)}{\mathbb{e}}[e^{\langle u_0(t),x_t\rangle } ] = e^{v_0(t)}b^0(0,t ) , \\ s^{\delta_i}(0,t ) & = { \mathbb{e}}^{{\mathbb{q}}^t}[s^{\delta_i}(t , t ) ] = e^{v_i(t)}{\mathbb{e}}^{{\mathbb{q}}^t}[e^{\langle u_i(t),x_t\rangle } ] = e^{v_i(t)}s^{0,\delta_i}(0,t),\end{aligned}\ ] ] where in the last equality we used the fact that the density of the @xmath1-forward measure @xmath165 is the same for both models @xmath108 and @xmath160 . indeed , for every @xmath105 $ ] , it holds that @xmath166 } = \frac{e^{\langle u_0(t),x_t\rangle}}{{\mathbb{e}}[e^{\langle u_0(t),x_t\rangle } ] } = \frac{1}{b^0_tb^0(0,t)},\ ] ] with @xmath167 denoting the numraire process corresponding to the model @xmath160 . we have so far introduced a general setup based on affine processes for modeling multiple curves . as outlined in the previous subsection , depending on the choice of the numraire - martingale measure couple @xmath36 , different modeling approaches can be obtained from definition [ def : aff_model ] . we now specialize the above general framework to a classical setting where the numraire is given by the ois bank account , defined with respect to the ( risk - free ) ois short rate process @xmath113 , and @xmath33 is the associated ( spot ) martingale measure . more specifically , in this subsection we shall work under the following assumption . [ ass : short_rate ] the numraire @xmath34 satisfies @xmath168 , for all @xmath105 $ ] , where @xmath113 is a real - valued adapted process denoting the ois short rate . the probability measure @xmath33 is such that the @xmath37-discounted price of every basic traded asset is a @xmath33-martingale . throughout this subsection , let @xmath65 be an affine process taking values in @xmath169 . we also introduce the auxiliary affine process @xmath170 defined by @xmath171 , for @xmath105 $ ] , and taking values in @xmath172 . the set @xmath173 is defined as in with respect to @xmath174 . we specialize definition [ def : aff_model ] as follows . [ def : short_rate_model ] let @xmath175 , @xmath176 \to \mathbb{r}$ ] such that @xmath177 , @xmath178 and @xmath179 a family of functions @xmath180 \to \mathbb{r}$ ] , @xmath181 . the tuple @xmath182 is said to be an _ affine short rate multi - curve model _ if @xmath183 is an affine multi - curve model in the sense of definition [ def : aff_model ] , where @xmath184 , \\ v_i(t ) & : = c_i(t ) & \text{and } \quad & u_i(t ) : = ( \gamma_i,0 ) , & \qquad&\text { for all } t\in[0,{\mathbb{t } } ] \text { and } i=1,\ldots , m.\end{aligned}\ ] ] if assumption [ ass : short_rate ] holds , so that @xmath185 , assuming that a tuple @xmath182 is an affine short rate multi - curve model in the sense of definition [ def : short_rate_model ] is equivalent to letting the ois short rate @xmath186 and the spot multiplicative spreads @xmath12 be specified as follows : @xmath187 , \label{eq : short_rate_1}\\ \log s^{\delta_i}(t , t ) & = c_i(t ) + \langle\gamma_i , x_t\rangle , & \text { for all } & t\in[0,{\mathbb{t } } ] \text { and } i=1,\ldots , m . \label{eq : short_rate_2}\end{aligned}\ ] ] note that definition [ def : short_rate_model ] implicitly requires ( via definition [ def : aff_model ] ) that the parameters @xmath188 and @xmath189 satisfy @xmath190 , for all @xmath102 , with @xmath191 . if the function @xmath192 is chosen to be constant and @xmath193 and @xmath194 set to @xmath195 , definition [ def : short_rate_model ] reduces to a classical ( i.e. , single - curve ) time - homogeneous affine short rate model . by additionally allowing @xmath192 and also @xmath193 to be deterministic functions we introduce a time - inhomogeneity which in turn enables a perfect fit to the initial term structures ( see proposition [ prop : deterministicshiftcharacterization ] below ) . definition [ def : short_rate_model ] thus already incorporates _ deterministic shift extended models _ first introduced in @xcite in the single - curve setting and recently taken up in @xcite in the multi - curve setting ( cf . section [ sec : rel_short_rate ] for the relation between @xcite and the present setting ) . in definition [ def : short_rate_model ] , the ois bank account is specified via a short rate modeled as in . by analogy , multiplicative spreads can also be specified via an instantaneous _ short spread rate _ , similarly as in ( * ? ? ? * remark 3.19 ) . this can be achieved by letting the affine process @xmath61 be of the form @xmath196 , where @xmath197 is an affine process on some state space @xmath198 and @xmath199 , with @xmath200 being an affine function , for all @xmath16 . the coefficients @xmath179 and @xmath189 are chosen as @xmath201 and @xmath202 , for all @xmath16 , with @xmath203 being the canonical basis of @xmath204 . the spot multiplicative spreads admit then the representation @xmath205 \text { and } i=1,\ldots , m,\ ] ] where the process @xmath206 can be interpreted as a short spread rate , for @xmath16 . similarly as in the case of general affine multi - curve models , affine short rate multi - curve models admit an equivalent characterization in terms of the structure of ois bond prices and spreads . more specifically , we can establish the following result , which specializes proposition [ prop : qaffinecharacterization ] . let @xmath61 be an affine process and suppose that assumption [ ass : short_rate ] holds . then the following hold : 1 . if @xmath182 is an affine short rate multi - curve model , then @xmath61 generates exponentially affine ois bond prices and spreads , in the sense that @xmath207 where the functions @xmath208 , @xmath209 , @xmath102 , are given by @xmath210 for all @xmath211 and @xmath16 , with @xmath212 and @xmath213 denoting the characteristic exponents of the affine process @xmath214 as in lemma [ mixedstate ] ; 2 . conversely , if ois bond prices and spreads admit the representation for some functions @xmath208 , @xmath209 , @xmath102 , with @xmath215 and @xmath216 differentiable , for all @xmath105 $ ] , then @xmath182 is an affine short rate model , where , for all @xmath16 , @xmath217 noting that the affine process @xmath214 satisfies the assumptions of lemma [ mixedstate ] , part ( i ) easily follows from definition [ def : short_rate_model ] together with proposition [ prop : qaffinecharacterization ] . conversely , if ois bond prices and spreads admit the representation with respect to some differentiable functions @xmath208 and @xmath209 , @xmath102 , then the instantaneous ois forward rate is given by @xmath218 so that the ois short rate @xmath186 satisfies @xmath219.\ ] ] representations - then follow by letting @xmath220 and @xmath221 , for all @xmath16 . finally , the condition @xmath190 , for all @xmath102 , with @xmath191 , follows by the same arguments used in the last part of the proof of proposition [ prop : qaffinecharacterization ] . in view of proposition [ prop : x0y ] , multiplicative spreads greater than one and ordered with respect to the tenor s length can be easily generated by affine short rate multi - curve models . in the current market environment , negative rates have been observed to coexist with spreads which are strictly greater than one and ordered with respect to the tenor s length . this market scenario can be easily captured by letting the affine process @xmath61 be of the form @xmath140 and take value in a state space of the form @xmath222 , for some @xmath223 and where @xmath142 is a cone . if @xmath192 takes values in @xmath224 , @xmath225 and the assumptions of proposition [ prop : x0y ] are satisfied , then spreads will be greater than one and ordered , while the ois short rate will not be restricted to positive values only . similarly to proposition [ prop : fitting ] , we can obtain a necessary and sufficient condition to achieve an exact fit to the initially observed term structures . let us denote by @xmath158 and @xmath159 the bond prices and spreads computed according to the model @xmath226 . moreover , we denote by @xmath227 and @xmath228 the initially observed term structures of ois bond prices and spreads . we also define the corresponding ois forward rates by @xmath229 the following result is a direct consequence of the above definitions together with proposition [ prop : fitting ] . [ prop : deterministicshiftcharacterization ] suppose that assumption [ ass : short_rate ] holds . then an affine short rate multi - curve model @xmath182 achieves an exact fit to the initially observed term structures if and only if @xmath230,\\ c_i(t ) & = \log s^{m,\delta_i}(0,t ) - \log s^{0,\delta_i}(0,t ) , & \qquad & \text { for all } t\in[0,{\mathbb{t } } ] \text { and } i=1,\ldots , m.\end{aligned}\ ] ] in view of proposition [ prop : deterministicshiftcharacterization ] , assuming that the spreads @xmath159 generated by the model @xmath226 are greater than one and ordered , it then easily follows that 1 . for any @xmath231 , if @xmath232 , for all @xmath233 $ ] , then @xmath234 , for all @xmath25 ; 2 . for any @xmath235 with @xmath236 , if @xmath237,\ ] ] then @xmath238 , for all @xmath25 . in this section , we discuss how our setup relates to several multiple curve models based on affine processes that have been recently proposed in the literature . we show that all the existing approaches we are aware of can be recovered as special cases of our framework . as mentioned in the introduction , short rate models based on affine processes have been proposed for modeling multiple curves , see @xcite . a general framework encompassing all the models introduced in these papers has been recently provided in ( * ? ? ? * chapter 2 ) . in that work , the authors assume that the numraire is given as usual by the ois bank account , modeled via a short rate @xmath113 , with a corresponding martingale measure @xmath33 . considering for simplicity of presentation a single tenor @xmath3 , spot libor rates @xmath0 are then specified by @xmath239 where @xmath240 denotes the price at date @xmath1 of an artificial risky bond with maturity @xmath241 . note that risky bonds are only introduced as a modeling tool , their riskiness being referred to the interbank market , and do not represent real traded assets . by analogy to the short rate approach in a classical single - curve setting , ois bond prices and risky bond prices are assumed to be given by @xmath242 \quad\text { and } \quad b^{\delta}(t , t ) = \mathbb{e}\left[e^{-\int_t^t(r_u+s_u)du}\bigr|{\mathcal{f}}_t\right],\ ] ] for all @xmath25 , where @xmath243 denotes an instantaneous spread . the ois short rate @xmath186 and the spread @xmath244 are modeled as affine functions of a common affine process @xmath65 taking values on some state space @xmath245 ( @xcite also consider the case of a wishart process @xmath61 and deterministic shift extensions , meaning that the constant part of the affine function is actually a deterministic function in time ) . this framework can be regarded as a special case of our affine short rate setup introduced in section [ sec : short_rate ] . indeed , it suffices to observe that in the present setting the spot multiplicative spreads can be represented as @xmath246}{\mathbb{e}\left[e^{-\int_t^{t+\delta}(r_u+s_u)du}\bigr|{\mathcal{f}}_t\right ] } = c_{\delta}(t ) + \langle \gamma_{\delta},x_t\rangle,\ ] ] where @xmath247\rightarrow{\mathbb{r}}$ ] and @xmath248 are determined in terms of the characteristic exponents of @xmath61 and the specification of the affine functions for @xmath186 and @xmath244 . this shows that representations - hold with respect to the underlying affine process @xmath61 . similar considerations allow to recover the gaussian exponentially quadratic model recently proposed in @xcite . since our spot multiplicative spread is an observable quantity , unlike the instantaneous spread @xmath244 , it seems to us more natural to define affine short rate multi - curve models directly with respect to spot multiplicative spreads as in definition [ def : short_rate_model ] . the general affine multi - curve setup of section [ sec : aff_model ] allows to recover a generalization of the model recently introduced in @xcite , which extends to a multi - curve setting the affine libor model originally proposed in @xcite . in this setting , the numraire - martingale measure couple @xmath249 is given by the ois zero - coupon bond with maturity @xmath114 and the corresponding @xmath114-forward measure , i.e. , @xmath250 . we explicitly indicate this measure in the expectations below . as in section [ sec : aff_model ] , consider an affine process @xmath251 ( under @xmath252 ) taking values in @xmath253 and denote by @xmath254 the @xmath252-martingale defined by @xmath255 $ ] , for @xmath105 $ ] . for each tenor @xmath3 traded in the market , @xcite consider a finite collection of ordered maturities @xmath256 . for any tenor @xmath3 and @xmath257 , the multiplicative spread @xmath258 is given by @xmath259,\ ] ] where @xmath260 is a sequence in @xmath261 such that @xmath262 , for all @xmath263 , in order to ensure non - negative ois forward rates , and @xmath264 is a sequence in @xmath261 such that @xmath265 , for all @xmath263 , in order to ensure that spreads are greater than one . by its own nature , the model of @xcite is formulated with respect to a finite collection of maturities . however , a continuous extension can be obtained by considering a function @xmath266\rightarrow{\mathfrak{u}}$ ] such that @xmath267 and a function @xmath268\times \mathbb{r}_+ \rightarrow{\mathfrak{u}}$ ] such that @xmath269 , for all @xmath263 and @xmath3 . the idea of a continuous extension first appeared ( in a single - curve setting ) in the unpublished note @xcite and , after the appearance of the present paper , has been further developed in the multi - curve setting in @xcite . in analogy to ( * section 6 ) , let ois bond prices be given as follows : @xmath270 with @xmath271 $ ] . in line with equation , we then model @xmath272 by @xmath273 with @xmath274 $ ] . observe that , in order to ensure that @xmath275 , for all @xmath25 and @xmath3 , it suffices to require that @xmath276 for all @xmath19 $ ] and @xmath3 . similarly , order relations among multiplicative spreads @xmath272 associated to different tenors can be obtained by imposing suitable requirements on the function @xmath277 . recalling that @xmath133 , for all @xmath105 $ ] , representation together with the affine property of @xmath61 implies that the numraire process satisfies @xmath278,\ ] ] where @xmath75 and @xmath76 denote the characteristic exponents of @xmath61 . similarly , representation implies that the spot multiplicative spreads satisfy , for all @xmath105 $ ] and @xmath3 , @xmath279 the model - thus represents a special case of our general affine multi - curve framework . indeed , representations - follow by letting , for all @xmath105 $ ] and @xmath16 , @xmath280 moreover , in view of ( * ? ? ? * lemma 4.3 ) , it holds that @xmath127 and @xmath130 , for all @xmath19 $ ] and @xmath16 , thus showing that this version of the affine libor model with multiple curves can be recovered as a special case of definition [ def : aff_model ] . in our view , our general framework presents significant advantages with respect to the model of @xcite , not only due to the flexibility in the choice of the numraire - martingale measure couple and of affine processes on general convex state spaces . indeed , even under the assumption that the couple @xmath36 is specified in terms of the ois bond with maturity @xmath114 together with the corresponding @xmath114-forward measure as above , our definition [ def : aff_model ] allows for greater generality . this can be seen by noting that , given a family of functions @xmath281 as in definition [ def : aff_model ] , it is not always possible to find a couple of functions @xmath282\times[0,{\mathbb{t}}]\times{\mathbb{r}}_+\rightarrow{\mathfrak{u}}\times{\mathfrak{u}}$ ] such that holds . moreover , as shown in proposition [ prop : fitting ] , our specification - always allows for an automatic fit to the initially observed term structures . in contrast , in @xcite this is only possible under some conditions , notably that the initial ois term structure is decreasing with respect to maturity together with the finiteness of suitable exponential moments of the driving process . furthermore , while in our specification the initial term structures univocally determine the family of functions @xmath283 , in @xcite this is not the case , so that all model parameters have to be jointly determined by fitting the initial term structures and by calibration to market data . multi - curve models based on affine processes have been briefly mentioned in ( * ? ? ? * section 5.3 ) as a simple example of _ hjm - type multiple yield curve models_. the affine specification proposed in @xcite represents a special case of the general setup developed in the present paper . indeed , an affine multi - curve model in the sense of definition [ def : aff_model ] belongs to the family of risk neutral hjm - type multiple yield curve models considered in @xcite only if @xmath284 is absolutely continuous . more specifically , the affine specification considered in @xcite ( see ( * ? ? ? * definition 5.3 ) ) represents a special case of our definition [ def : short_rate_model ] when @xmath192 and @xmath193 are chosen to be constant . for clarity of notation , let us denote all the ingredients appearing in ( * ? ? ? * definition 5.3 ) with fractur letters . then ( * ? ? ? * definition 5.3 ) can be embedded into definition [ def : short_rate_model ] by setting @xmath285 , @xmath286 and @xmath287 , for all @xmath181 . conversely , in the special case where @xmath192 and @xmath193 are constant , definition [ def : short_rate_model ] can be embedded into ( * ? ? ? * definition 5.3 ) by letting @xmath288 , @xmath289 , @xmath290 and @xmath291 , for all @xmath292 and @xmath16 . we now show that affine multi - curve models , in the sense of definition [ def : aff_model ] , lead to tractable general valuation formulae for caplets and swaptions . we compute _ clean prices _ and follow the pricing approach outlined in ( * ? ? ? * appendix a ) , in particular assuming that the collateral account is given by the numraire asset . for simplicity of presentation , we shall consider a fixed maturity @xmath293 and assume that @xmath108 is an affine multi - curve model in the sense of definition [ def : aff_model ] . in the present affine setting , caplets can be easily priced by means of fourier techniques . as a preliminary , let @xmath294 and define the stochastic process @xmath295}$ ] by @xmath296 where the second equality follows from proposition [ prop : qaffinecharacterization ] . we denote by @xmath297 the set @xmath298<+\infty\right\}^{\circ},\ ] ] and introduce the strip of complex numbers @xmath299 . for @xmath300 , we can compute the following expectation , which we call _ modified moment generating function _ of @xmath301 : @xmath302}\\ & = \exp((1- { \ensuremath{\mathsf{i}}}\zeta)(v_0(t+\delta_i)+\phi(\delta_i , u_0(t+\delta_i)))+ { \ensuremath{\mathsf{i}}}\zeta(v_i(t)+v_0(t)))\\ & \quad \times{\mathbb{e}^{}\left[\left.e^{\langle ( 1- { \ensuremath{\mathsf{i}}}\zeta ) \psi(\delta_i , u_0(t+\delta_i))+ { \ensuremath{\mathsf{i}}}\zeta ( u_i(t)+ u_0(t ) ) , x_t\rangle}\right|{\mathcal{f}}_t\right]}\\ & = \exp((1- { \ensuremath{\mathsf{i}}}\zeta)(v_0(t+\delta_i)+\phi(\delta_i , u_0(t+\delta_i)))+ { \ensuremath{\mathsf{i}}}\zeta(v_i(t)+v_0(t)))\\ & \quad \times \exp(\phi(t - t , ( 1- { \ensuremath{\mathsf{i}}}\zeta ) \psi(\delta_i , u_0(t+\delta_i))+ { \ensuremath{\mathsf{i}}}\zeta ( u_i(t)+ u_0(t))))\\ & \quad \times \exp(\langle\psi(t - t , ( 1- { \ensuremath{\mathsf{i}}}\zeta ) \psi(\delta_i , u_0(t+\delta_i))+ { \ensuremath{\mathsf{i}}}\zeta ( u_i(t)+ u_0(t))),x_t\rangle ) . \label{cfchi } \end{split}\end{aligned}\ ] ] in the case of an affine short rate multi - curve model as of definition [ def : short_rate_model ] , expression becomes @xmath303 similarly , expression becomes @xmath304 the sets @xmath297 and @xmath305 depend on the specific choice of the driving process @xmath61 . by relying on ( * theorem 5.1 ) , we now provide a general caplet pricing formula , which is valid for any choice of the underlying affine process @xmath61 and for different choices of the contour of integration . in particular , the next result highlights the tractability of our framework : caplets can be priced by means of univariate fourier integrals . in turn , this implies that a calibration may be obtained with a reasonable amount of computational effort ( which may be further reduced by means of an application of an fft algorithm ) . [ prop : capletbasicaffine ] let @xmath306 , @xmath307 , @xmath308 and assume that @xmath309 . then the price at date @xmath1 of a caplet with notional @xmath310 , reset date @xmath9 and payoff @xmath311 at the settlement date @xmath23 is given by @xmath312 where @xmath313 is given in and @xmath314 is given by @xmath315 as shown in appendix [ sec : option ] , the price of a caplet can be expressed as @xmath316}\\ & = n\,{\mathbb{e}^{}\left[\left.\frac{b_t}{b_t}b(t , t+\delta_i)\left(s^{\delta_i}(t , t)b(t , t+\delta_i)^{-1}-(1+\delta_i k)\right)^+\right|{\mathcal{f}}_t\right]}\\ & = n\,{\mathbb{e}^{}\left[\left.\frac{b_t}{b_t}b(t , t+\delta_i)\left(e^{\mathcal{y}_t}-(1+\delta_i k)\right)^+\right|{\mathcal{f}}_t\right]}\end{aligned}\ ] ] since the modified moment generating function of @xmath301 can be explicitly computed as in , the pricing of a caplet is thus reduced to the pricing of a call option written on an asset whose characteristic function is explicitly known . at this point , a direct application of ( * ? ? ? * theorem 5.1 ) yields the result . note that , in the terminology of ( * ? ? ? * theorem 5.1 ) , the present case corresponds to @xmath317 and @xmath318 , which is an element of @xmath297 by definition [ def : aff_model ] . in the present general setting , swaptions do not admit a closed - form pricing formula . indeed , on the one side , we consider general multi - factor models , so that the jamshidian trick " ( see @xcite ) is not applicable ; on the other side , in a multiple curve setting , a payer ( resp . receiver ) swaption can not be represented as a put ( resp . call ) option on a coupon bond . in this subsection , by relying on fourier methods and along the lines of @xcite , we provide a general analytical approximation which exploits the affine property of our framework . we work in the general setting of section [ sec : aff_model ] , noting that all formulas admit suitable simplifications in the case of affine short - rate multi - curve models . we consider a european payer swaption with maturity @xmath9 , written on a ( payer ) interest rate swap starting at @xmath319 , with payment dates @xmath320 , with @xmath321 for @xmath322 and @xmath294 , with notional @xmath310 . as shown in appendix [ sec : option ] , the value of such a claim at date @xmath1 is given by @xmath323.\ ] ] the general idea underlying our approximation consists in approximating the exercise region by an event defined in terms of an affine function of @xmath324 . more specifically , we have that @xmath325\\ & \geq n\mathbb{e}^{}\left[\frac{b_t}{b_t}\left(\sum_{j=1}^{{\mathcal{n}}}b(t , t_{j-1})s^{\delta_i}(t , t_{j-1})-(1+\delta_i k)b(t , t_j)\right){\mathbf{1}}_{\mathcal{g}}\biggr|{\mathcal{f}}_t\right]\\ & = : \widetilde{\pi}^{swptn}(\alpha , \beta),\end{aligned}\ ] ] where @xmath326 and @xmath327 . let us simplify the notation by introducing @xmath328 so that , in view of proposition [ prop : qaffinecharacterization ] , @xmath329 admits the representation @xmath330.\end{aligned}\ ] ] recall from definition [ def : aff_model ] that @xmath130 , so that @xmath331 and @xmath332 are well - defined . as shown in the following proposition , the quantity @xmath333 can be explicitly computed , analogously to proposition [ prop : capletbasicaffine ] . [ prop : swaptionbasicaffine ] assume @xmath334 @xmath335 . let @xmath307 such that @xmath336 , @xmath335 . then the lower bound in terms of @xmath337 for a payer swaption with notional @xmath310 , maturity @xmath9 , payment dates @xmath338 , with @xmath321 for all @xmath322 is given by @xmath339\right)d\zeta\big ) , \end{aligned } \label{eq : swaptionapproxgeneral}\end{aligned}\ ] ] where @xmath340 is given by @xmath341 , & \mbox{if } \epsilon<0 , \\ \frac{1}{2}\mathbb{e}\left[e^{v_0(t_l)+v_{i , j}(t_l)+\phi(t_l - t , u_{i , j}(t_l)+u_0(t_l))+\langle \psi(t_l - t , u_{i , j}(t_l)+u_0(t_l ) ) , x_t\rangle}\biggr|{\mathcal{f}}_t\right],&\mbox{if } \epsilon = 0,\\ 0,&\mbox{if } \epsilon > 0 . \end{cases}\ ] ] the claim is a direct consequence of ( * ? ? ? * theorem 5.1 ) , noting that each summand appearing in corresponds in the notation of @xcite to the case @xmath342 , @xmath343 , @xmath344 and @xmath345 . proposition [ prop : swaptionbasicaffine ] gives a general lower bound for the price of a swaption , parameterized in terms of @xmath346 . these parameters should be determined in such a way that the lower bound becomes as tight as possible , while at the same time ensuring the finiteness of suitable exponential moments of @xmath324 . as pointed out in @xcite , to which we refer for more details on the numerical implementation , the values of @xmath346 can be chosen in two ways : 1 . by maximizing with respect to @xmath347 , thus providing the lower bound @xmath348 note that this solution can be computationally demanding , especially for highly dimensional models . moreover , for a given choice of @xmath349 , the optimization procedure should be constrained in order to ensure the finiteness of joint exponential moments of @xmath324 . 2 . by considering hyperplane - like approximations and predetermining the best possible values of @xmath347 ( the well - known singleton - umantsev approximation constitutes an example in this sense , see @xcite and compare also with @xcite ) . for notational simplicity we consider here the case of one single tenor @xmath350 and suppose that the driving process @xmath61 is a wishart process on @xmath351 with state space @xmath352 of the form @xmath353 where @xmath354 is @xmath355 matrix of brownian motions , @xmath356 and @xmath357 are @xmath358 matrices . the particular appealing feature of wishart processes is stochastic correlation between the nonnegative diagonal elements of the matrix . on the classical canonical state space @xmath359 this property can not be achieved for positive factors . this possibility is however a crucial ingredient when it comes to modeling spreads which are highly correlated . we consider an affine short rate multi - curve model as of section [ sec : short_rate ] with short rate @xmath360 and spread @xmath361 , where @xmath362 and @xmath363 to guarantee positivity of the spreads . recall that the scalar product @xmath364 is here the trace . by choosing @xmath188 and @xmath365 to be diagonal matrices , the above model represents a natural extension of the classical cir model to the multi - curve setting since the diagonal elements of a wishart process are stochastically correlated cir processes . the goal of this section is to study the pricing of caps for this particular model . as shown in section [ sec : option ] , the price of a caplet with unitary notional can be computed via @xmath366\\ & \quad -(1+\delta k)b(t , t+\delta)\mathbb{q}^{t+\delta}\left[s^{\delta}(t , t)\geq ( 1+\delta k)b(t , t+\delta)\ , |\ , \mathcal{f}_t\right ] , \end{split}\ ] ] where the probability measure @xmath367 is defined via @xmath368 in the case of the above introduced model , it can be easily shown that the process @xmath61 follows under both measures @xmath369 and @xmath370 a non - central wishart distribution with time dependent parameters , which is stated in lemma [ lem : newwish ] below . as the density is ( up to the solution of odes ) explicitly known , this allows obtaining ( semi-)analytical pricing formulas for caplets , similarly as in the cir model . as a preliminary , let us introduce the following definition , in line with ( * ? ? ? * definition a.4 ) . suppose that @xmath356 , @xmath371 and @xmath372 is a @xmath355 matrix such that @xmath373 is symmetric positive semidefinite . a symmetric positive definite random matrix @xmath374 is said to be _ non - centrally wishart distributed _ with @xmath375 degrees of freedom , covariance matrix @xmath376 , and matrix of non - centrality parameter @xmath372 , if its laplace transform satisfies @xmath377=\det(i_d-2 u\sigma)^{-\frac{\delta}{2}}e^{- \langle u(i_d+2u\sigma)^{-1},\sigma\theta\rangle}.\ ] ] in this case , we write @xmath378 . [ lem : newwish ] let @xmath61 be a wishart process of the form under @xmath39 . 1 . under @xmath369 , @xmath324 has a non - central wishart distribution @xmath379 where @xmath380 and @xmath381 are solutions of the following system of ordinary differential equations @xmath382 where @xmath212 and @xmath213 denote the characteristic exponents of the process @xmath383 . the solution of is explicitly given by @xmath384 with @xmath385 . 2 . under @xmath370 , @xmath324 has a non - central wishart distribution @xmath386 , where @xmath387 and @xmath388 are solutions of the following system of ordinary differential equations @xmath389 where @xmath212 and @xmath213 denote the characteristic exponents of the process @xmath383 . the solution of is explicitly given as above . concerning ( i ) , note that the density process @xmath390 of @xmath391 is given by @xmath392= \frac{1}{s^{\delta}(0,t ) b(0,t)}\mathbb{e}\left[\frac{s^{\delta}(t , t ) b(t , t)}{b_t } \big| \mathcal{f}_t\right]\\ & = \frac{1}{s^{\delta}(0,t ) b(0,t)}\mathbb{e}\left[e^{c(t)+\langle \gamma , x_{t}\rangle-\int_0^t \ell(s)ds-\langle\lambda , \int_0^t x_{s}ds\rangle } \big| \mathcal{f}_t\right]\\ & = \frac{1}{s^{\delta}(0,t ) b(0,t)}\exp\left(c(t)-\int_0^t \ell(s)ds+\tilde{\phi}\left(t - t , \gamma,-\lambda \right)+\left\langle\tilde{\psi}\left(t - t , \gamma , -\lambda\right ) , x_t\right\rangle-\langle \lambda , \int_0^t x_{s}ds\rangle\right),\end{aligned}\ ] ] where @xmath212 and @xmath393 denote the characteristic exponents of @xmath383 . note that the diffusion part of @xmath394 is given by @xmath395 so that we can write @xmath396 by girsanov s theorem , under the measure @xmath397 , the linear drift of @xmath61 changes to @xmath398 so that @xmath61 becomes a wishart process with time - varying linear drift under @xmath369 . according to ( * proposition a.6 ) , @xmath324 has a noncentral wishart distribution @xmath399 , where @xmath380 and @xmath381 are solutions of . concerning ( ii ) , we have for the density process of @xmath400 @xmath401&=\frac{b(t , t)}{b_t b(0,t)}\\ & = \frac{1}{b(0,t)}\exp\left(-\int_0^t \ell(u)du+\tilde{\phi}\left(t - t , 0 , -\lambda \right)+\left\langle\tilde{\psi}\left(t - t , 0,-\lambda\right ) , x_t\right\rangle-\langle \lambda , \int_0^t x_{s}ds\rangle \right).\end{aligned}\ ] ] the assertion then follows similarly as for @xmath369 . in view of , we focus here on the computation of @xmath402 \qquad\text{and}\qquad \mathbb{q}^{t+\delta}\left[s^{\delta}(t , t)\geq ( 1+\delta k)b(t , t+\delta)\right].\ ] ] by the specification of our model , the first of the two quantities above becomes @xmath403\ ] ] and similarly for the @xmath370-probability . it thus amounts to compute expressions of the type @xmath404,\ ] ] for some matrix @xmath405 ( note that @xmath406 can be symmetrized to lie in @xmath407 ) and some constant @xmath408 . the following proposition relates linear combination of elements of non - centrally wishart distributed matrices with @xmath409-distributed random variables . [ prop : chisquare ] let @xmath410 with @xmath411 . then @xmath412 where @xmath413 are the eigenvalues of @xmath414 and @xmath415 are independent random variables with @xmath416 , where @xmath417 by ( * ? ? * proposition 6 ) , @xmath61 has the same distribution as @xmath418 where @xmath419 . therefore @xmath420 where @xmath421 . let us now compute the moment generating function of @xmath422 , which is given by ( see , e.g. , ( * ? ? ? * proposition a.5 ) ) @xmath423=\prod_{i=1}^d(1 - 2u\lambda_i)^{-\frac{\kappa}{2}}e^{\frac{u\lambda_iy_{ii}}{1 - 2u\lambda_i}}.\ ] ] however , this corresponds to the moment generating function of @xmath424 , where @xmath425 are independent random variables . indeed , it holds that @xmath426=\prod_{i=1}^d\mathbb{e}\left[e^{u \lambda_i v_{i}}\right ] = \prod_{i=1}^d(1 - 2u\lambda_i)^{-\frac{\kappa}{2}}e^{\frac{u\lambda_iy_{ii}}{1 - 2\lambda_iu}}.\ ] ] this proves that @xmath427 . since @xmath428 , the assertion is proved . [ cor : wish ] let @xmath405 and @xmath61 a wishart process of form . then under @xmath370 it holds that @xmath429 where @xmath430 are the eigenvalues of @xmath431 and @xmath432 are independent random variables with @xmath433 , where @xmath434 and @xmath435 and @xmath436 are given in lemma [ lem : newwish ] . the same assertion holds for @xmath437 with @xmath435 and @xmath436 replaced by @xmath438 and @xmath439 . the assertion is a direct consequence of lemma [ lem : newwish ] and proposition [ prop : chisquare ] . finally , by relying on the above results , we are ready to give a ( semi-)analytical formula for the price of a caplet in the above wishart model . let @xmath61 be a wishart process of the form . consider an affine short rate multi - curve model with * ois short rate @xmath440 , for all @xmath441 $ ] and * logarithmic multiplicative spreads @xmath442 , for all @xmath441 $ ] . then the price of a caplet at date @xmath195 with unitary notional , reset date @xmath9 and payoff @xmath443 at the settlement date @xmath444 , is given by @xmath445 where * the constant @xmath446 is given by @xmath447 with @xmath448 being the constant part in the characteristic exponent of @xmath449 , and * @xmath450 and @xmath451 denote the cumulative distribution functions of a weighted sum of non - centrally @xmath409-distributed random variables corresponding to @xmath452 and @xmath453 as of corollary [ cor : wish ] for @xmath369 and @xmath370 with @xmath454 and @xmath455 the constant part in the characteristic exponent of @xmath449 . the assertion is a consequence of equation together with corollary [ cor : wish ] . in order to practically implement the pricing formula , the following steps are necessary : * compute @xmath455 and integrate it over the interval @xmath456 $ ] to obtain @xmath457 ; * compute the eigenvalues and eigenvectors of @xmath458 and @xmath459 with @xmath454 to get the weights @xmath460 and the non - centrality parameters @xmath461 ; * compute @xmath446 as given in and the distribution function of a positive weighted sum of two independent non - centrally @xmath409-distributed random variables , e.g. , via a laguerre series expansions ( see @xcite ) of the form @xmath462 where @xmath463 is a function depending on @xmath464 and @xmath465 are the coefficients ( depending on @xmath466 ) of the laguerre polynomials @xmath467 ( evaluated at @xmath468 where @xmath469 denotes a constant ) . for an alternative computation based on fourier inversion we refer to @xcite . note that the weights @xmath460 and the non - centrality parameters @xmath470 depend on the maturity ( the degrees of freedom @xmath375 are constant ) while the arguments @xmath446 in the distribution function depend additionally on the strike . in the case where @xmath192 and @xmath471 are constants , @xmath408 only depends on @xmath27 and we write only @xmath472 . in this case the matrix @xmath473 for different maturities and strikes needed for calibration purposes can be obtained by a matrix product @xmath474 where @xmath475 and @xmath476 are defined by @xmath477 provided the initial term structure of spreads and bonds is known , this procedure then gives the prices for all maturities and strike rates . in this section , we discuss the calibration of two simple specifications of our general framework to cap / floor market data . this section aims at illustrating the practical feasibility of the proposed approach and it is not meant to suggest a particular specification of the framework . the first specification is based on a cir - gamma model , while the second specification is driven by a wishart process , as presented in section [ sec : wishart ] . in particular , this represents the first instance of calibration of a wishart short rate model in the multi - curve framework . moreover , we complement our results with a parameter stability analysis in the case of the cir - gamma model , showing a satisfactory degree of stability . we refer to @xcite for calibration results based on swaption data and to @xcite for a calibration approach which relies on atm european swaption and cap quotes . let us start by briefly describing our market data sample . we initially consider a fixed trading date , namely august @xmath478 , 2011 . the data sample consists of a cross section of market quotes ( corresponding to perfectly collateralized transactions ) of linear and non - linear interest rate derivatives . as far as linear products are concerned , we consider market data for overnight indexed swaps and interest rate swaps . on the basis of these market quotes , we construct the ois discount curve @xmath479 and the forward curves @xmath480 , for @xmath481 and @xmath482 . this has been performed by relying on the finmath java library ( see @xcite ) . concerning non - linear interest rate products , the market practice consists in posting a cap / floor composite surface : when the strike price is below the at - the - money ( atm ) level , quotes refer to out - of - the - money ( otm ) floors , whereas if the strike is above atm , quotes refer to otm caps . in this way , the whole surface is constructed from liquidly traded otm options . by relying on the put - call parity between caps and floors , we can however treat all implied volatilities as cap volatilities . to reduce the complexity of the calibration , it is convenient to construct a caplet implied volatility surface , suitably bootstrapped from cap implied volatility data . the surface refers to strike prices ranging between @xmath483 and @xmath484 and maturities between @xmath485 months and @xmath486 years . in the market , caps with maturity larger than two years are indexed to the @xmath485-month forward rate while those with lower expiry are linked to the @xmath487-month curve . a normal implied volatility is obtained by numerically searching for the value of @xmath488 ( where @xmath310 stands for normal ) such that the bachelier pricing formula for a caplet @xmath489}\\ & = b(t , t_i)\delta\sigma_n\sqrt{t_{i-1}-t}\left(\frac{1}{\sqrt{2\pi}}e^{-\frac{z^2}{2}}+zn(z)\right),\end{aligned}\ ] ] where @xmath490 best fits the market price of a given caplet . we now provide a detailed description of the implementation of proposition [ prop : capletbasicaffine ] , by means of the fft algorithm ( see @xcite ) . in the present discussion , we express the integration variable as @xmath491 . let also @xmath492 and @xmath493 the integral term appearing in can then be written as @xmath494 we perform a first approximation by introducing a trapezoidal rule of the form @xmath495 , so that the effective upper limit of integration is given by @xmath496 . as we want to perform a simultaneous evaluation of the integral term for a grid of strike prices , we also introduce a grid for @xmath263 of the form @xmath497 , which gives a mesh covering the interval @xmath498 , with @xmath499 . since we want to apply the fft algorithm , we need to impose the nyquist condition , meaning that we set @xmath500 , thus introducing a tradeoff between the accuracy of the log - strike and integration grids . as suggested by @xcite , we introduce the weights of the simpson rule , in order to obtain a satisfactory accuracy even for large values of @xmath501 . in summary , the integral term ( along a grid of log - strikes ) is approximated via @xmath502,\end{aligned}\ ] ] where @xmath503 denotes a kronecker delta function which is @xmath504 for @xmath505 and zero otherwise . formula can be computed by a direct application of the fft algorithm . in our analysis , we set @xmath506 and @xmath507 . for a given vector of model parameters @xmath508 , belonging to the set of admissible parameters @xmath509 , we compute caplet prices using the above methodology and convert them into model implied normal volatilities , that we denote by @xmath510 . the aim of the calibration procedure is to solve @xmath511 where @xmath512 denotes the market - observed implied volatilities and @xmath513 the euclidean norm . in the following , we illustrate two candidate specifications along with their calibration results . note that both models allow for a perfect fit to the observed term structures via a suitable choice of the functions @xmath192 and @xmath471 in line with proposition [ prop : deterministicshiftcharacterization ] . we first calibrate the following model consisting of a two - dimensional process @xmath90 of the form @xmath514 where @xmath515 , @xmath354 is a brownian motion and @xmath93 is a gamma process with compensating jump measure @xmath516 , with @xmath517 . the short rate is specified as @xmath518 and the spreads are of the form @xmath519 , @xmath520 , with @xmath521 and @xmath522\rightarrow{\mathbb{r}}$ ] . table [ table : cir - vgparams ] reports the calibrated model parameters while figure [ fig : calibrationerrorsdataset08022011 ] illustrates the quality of the fit in terms of prices and implied volatilities . more precisely , the left panel shows the squared error in price while the right panel illustrates the squared error in implied volatilities . despite its simplicity , the model achieves a reasonably good fit to the observed market quotes . looking at squared error in implied volatilities , we observe that the quality of the fit is lower for higher strikes and for the first maturity we considered ( 6 months ) . however , the quality of the fit is acceptable as can be seen by looking at the left panel of figure [ fig : calibrationerrorsdataset08022011 ] , which highlights that the squared error in price in the @xmath485-month maturity and deep otm region is low . observe that the calibrated parameters satisfy @xmath523 , reflecting the fact that @xmath485-month rates embed a higher degree of interbank risk with respect to @xmath487-month rates . this was achieved without imposing a priori constraints in the optimization algorithm . note also that the computational complexity of the procedure is at the same level of a calibration of a standard affine stochastic volatility model for equities . .calibration result for the cir - gamma model . resnorm represents the sum of squared distances between market and model implied volatilites.[table : cir - vgparams ] [ cols="^,^,^,^,^,^ " , ] this appendix presents general pricing formulas for typical interest rate derivatives . as we are going to show , the quantity @xmath272 plays a pivotal role in the valuation of interest rate products . we here derive clean prices in the spirit of ( * ? ? ? * appendix a ) , assuming perfect collateralization with the collateral asset being equal to the numraire . the prices of linear interest rate products ( i.e. , without optionality features ) can be directly expressed in terms of the basic quantities @xmath8 and @xmath272 . + _ forward rate agreement . _ + a forward rate agreement ( fra ) starting at @xmath9 , with maturity @xmath444 , fixed rate @xmath27 and notional @xmath310 is a contract which pays at time @xmath444 the amount @xmath524 the value of such a claim at time @xmath525 is @xmath526}\\ & = n\bigl(b(t , t)s^{\delta}(t , t)-b(t , t+\delta)(1+\delta k)\bigr ) . \end{split}\ ] ] _ overnight indexed swap . _ + an overnight indexed swap ( ois ) is a contract where two counterparties exchange two streams of payments : the first one is computed with respect to a fixed rate @xmath27 , whereas the second one is indexed by an overnight rate ( eonia ) . let us denote by @xmath527 the payment dates , with @xmath528 for all @xmath529 . the swap is initiated at @xmath530 . the value of an ois at @xmath531 with notional @xmath310 can be expressed as ( see e.g. ( * ? ? ? * section 2.5 ) ) @xmath532 therefore , the ois rate @xmath533 , which is by definition the value for @xmath27 such that the ois contract has zero value at inception , is given by @xmath534 _ interest rate swap . _ + in an interest rate swap ( irs ) , two streams of payments are exchanged between two counterparties : the first cash flow is computed with respect to a fixed rate @xmath27 , whereas the second one is indexed by the libor rate . the value of the irs at time @xmath531 , where @xmath535 denotes the inception time , is given by @xmath536 the swap rate @xmath537 , which is by definition the value for @xmath27 such that the contract has zero value at inception , is given by @xmath538 _ basis swap . _ + a basis swap is a special type of interest rate swap where two cash flows related to libor rates associated to different tenors are exchanged between two counterparties . for instance , a typical basis swap may involve the exchange of the 3-month against the 6-month libor rate . following the standard convention for the definition of a basis swap in the euro market ( see @xcite ) , the basis swap is equivalent to a long / short position on two different interest rate swaps which share the same fixed leg . let @xmath539 , @xmath540 and @xmath541 , with @xmath542 , @xmath543 , @xmath544 and corresponding tenor lengths @xmath545 , with no constraints on @xmath546 . the first two tenor structures on the one side and the third on the other are associated to the two floating and to the single fixed leg , respectively . we denote by @xmath310 the notional of the swap , which is initiated at time @xmath547 . the value at time @xmath548 is given by @xmath549 the value @xmath550 ( called _ basis swap spread _ ) such that the value of the contract at initiation is zero is then given by @xmath551 note that , prior to the financial crisis , the value of @xmath550 used to be ( approximately ) zero . in this section , we report general valuation formulas for plain vanilla interest rate products such as ( european ) caplets and swaptions . _ caplet . _ + the price at time @xmath1 of a caplet with strike price @xmath27 , maturity @xmath9 , settled in arrears at @xmath444 , is given by @xmath552}\notag\\ & = n{\mathbb{e}^{}\left[\left.\frac{b_t}{b_t}\bigl(s^{\delta}(t , t)-(1+\delta k)b(t , t+\delta)\bigr)^+\right|{\mathcal{f}}_t\right ] } \label{eq : caplet_gen}.\end{aligned}\ ] ] note that , in the classical single - curve setting ( i.e. , under the assumption that @xmath553 is identically equal to one ) , the valuation formula reduces to the classical relationship between a caplet and a put option on a zero - coupon bond with strike @xmath554 . from , we see that the payoff of a caplet at time @xmath9 and notional @xmath555 corresponds to @xmath556 let us now introduce the following probability measure on @xmath557 : @xmath368 note that @xmath558 and @xmath559=1 $ ] , since @xmath560 is a @xmath39-martingale by proposition [ prop : bonds_spreads ] . by bayes formula , we therefore have @xmath561=s^{\delta}(t , t)b(t , t)\mathbb{e}^{\widetilde{\mathbb{q}}}\left[1_{\{s^{\delta}(t , t)\geq ( 1+\delta k)b(t , t+\delta)\}}\ , \bigr| \ , \mathcal{f}_t\right].\ ] ] similarly , we obtain @xmath562\\=(1+\delta k)b(t , t+\delta)\mathbb{e}^{\mathbb{q}^{t+\delta}}\left[1_{\{s^{\delta}(t , t)\geq ( 1+\delta k)b(t , t+\delta)\}}\ , \bigr|\ , \mathcal{f}_t\right].\end{aligned}\ ] ] hence , the price of a caplet can be computed via @xmath563\\ & \quad -(1+\delta k)b(t , t+\delta)\mathbb{q}^{t+\delta}\left[s^{\delta}(t , t)\geq ( 1+\delta k)b(t , t+\delta)\ , \bigr|\ , \mathcal{f}_t\right ] . \end{split}\ ] ] _ swaption . _ + we consider a standard european payer swaption with maturity @xmath9 , written on a ( payer ) interest rate swap starting at @xmath319 and payment dates @xmath564 , with @xmath528 for all @xmath529 , with notional @xmath310 . the value of such a claim at time @xmath1 is given by @xmath565.\ ] ] under assumption [ ass : couple ] , the process @xmath566 is a martingale , for all @xmath19 $ ] . since @xmath567 , this implies that @xmath568 = { \mathbb{e}}\left[\frac{1}{b_t}\bigr|{\mathcal{f}}_t\right ] , \qquad\text { for all } 0\leq t\leq t\leq { \mathbb{t}},\ ] ] thus proving part ( i ) . in particular , note that this implies that @xmath569 $ ] , thus ensuring that @xmath570 defines a probability measure @xmath571 , for every @xmath19 $ ] . recalling that @xmath572 , for all @xmath25 and @xmath16 , it holds that @xmath573 \\ & = { \mathbb{e}}\left[\frac{l_t(t , t+\delta_i)-l_t(t , t+\delta_i)}{b_{t+\delta_i}}\bigr|{\mathcal{f}}_t\right ] = \frac{b(t , t+\delta_i)}{b_t}\bigl({\mathbb{e}}^{{\mathbb{q}}^{t+\delta_i}}[l_t(t , t+\delta_i)|{\mathcal{f}}_t]-l_t(t , t+\delta_i)\bigr).\end{aligned}\ ] ] since @xmath574 by part ( i ) , the last equality proves part ( ii ) . finally , by bayes formula , the process @xmath153 is a @xmath40-martingale if and only if the process @xmath575 defined by @xmath576 is a @xmath577-martingale , for every @xmath16 . by definition of @xmath29 , it holds that @xmath578 and the desired martingale property then follows from part ( ii ) . f. m. ametrano and m. bianchetti . everything you always wanted to know about multiple interest rate curve bootstrapping but were afraid to ask . preprint ( available at ` http://ssrn.com/abstract=2219548 ` ) , 2013 . g. bormetti , d. brigo , m. francischello , and a. pallavicini . impact of multiple - curve dynamics in credit valuation adjustments . in k. glau , z. grbac , m. scherer , and r. zagst , editors , _ innovations in derivatives markets _ , pages 251266 . springer , 2016 . c. cuchiero and j. teichmann . path properties and regularity of affine processes on general state spaces . in c. donati - martin , a. lejay , and a. rouault , editors , _ sminaire de probabilits xlv _ , volume 2078 of _ lecture notes in mathematics _ , pages 201244 . springer , 2013 . z. grbac , l. meneghello , and w. j. runggaldier . derivative pricing for a multi - curve extension of the gaussian , exponentially quadratic short rate model . in k. glau , z. grbac , m. scherer , and r. zagst , editors , _ innovations in derivatives markets _ , pages 191226 . springer , 2016 . j. kallsen . a didactic note on affine stochastic volatility models . in y. kabanov , r. lipster , and j. stoyanov , editors , _ from stochastic calculus to mathematical finance - the shiryaev festschrift _ , pages 343368 . springer , berlin , 2006 . l. morino and w. j. runggaldier . on multicurve models for the term structure . in r. dieci , x .- z . he , and c. hommes , editors , _ nonlinear economic dynamics and financial modelling : essays in honour of carl chiarella_. springer , 2014 .

the microscopic nature of high-@xmath0 superconductivity in the cuprates is still under heavy debate after more than two decades of intensive studies . it has been widely accepted that the cuprates may be properly considered as doped mott insulators and the superconductivity arises after the antiferromagnetic ( afm ) long range order is destroyed by doping . the challenge is that a typical doped mott insulator is a strongly correlated electron system in which two electrons staying at the same lattice site will cost a huge energy as compared to other energy scales like the hopping energy . with the failure of conventional perturbative many - body methods , to tackle a doped mott insulator one usually resorts to either the so - called gutzwiller projected trial wavefunction or constructing an effective low - energy theory @xcite . in the latter approach , the electron fractionalization @xcite is often utilized in order to implement some key features of a doped mott insulator . indeed , the fractionalization due to the mottness is most transparent at half - filling where only the spin degrees of freedom of the electrons are left , with the charge part being totally frozen out . only by doping can charge carriers emerge which account for a small portion of the total number of the electrons at low doping . the real challenge is that there exist many different ways of introducing fractionalized elementary objects to implement the same local mottness , i.e. , the no double occupancy constraint . these elementary particles are generally further coupled to some emergent gauge field(s ) due to the uncertainty in the fractionalization @xcite . in the lack of precise energetic comparison , a full self - consistent development of a potential fractionalization / gauge theory becomes essential in order for it to be convincingly verified or falsified by experiment as well as by general theoretical considerations . in principle , an inappropriate fractionalization would always lead to strong gauge fluctuations rendering the effective theory intractable . therefore , a stable , well - controlled gauge theory description will be a highly desired candidate for the correct low - energy theory of cuprate superconductors . we will consider a particular electron fractionalization description with a mutual chern - simons gauge structure @xcite . specifically , an electron in this theory is fractionalized into an @xmath1 bosonic spinon and a charge @xmath2 bosonic holon , which are minimally coupled to u(1 ) gauge fields @xmath3 and @xmath4 , which do not have their own dynamics , but are entangled by a mutual chern - simons topological term @xmath5which is ( 2 + 1)d realization of the topological bf theory.@xcite physically it simply means that the spinons and holons will perceive each other as quantized @xmath6 fluxoids @xcite . in the limit of zero hole concentration , the spinon condensation can naturally lead to an afm long - range ordered state . at finite doping , the mutual chern - simons term in eq . ( [ lmcs ] ) captures @xcite a precise statistical sign structure identified @xcite in the @xmath7-@xmath8 model . by the @xmath6 fluxoids carried by the mobile holons , the afm state will be destroyed and eventually a superconducting will set in with the holon condensation , and a zero - temperature phase diagram as a function of doping concentration has been recently discussed in refs . ye1,ye2 . finally , there are other quantum systems dst96,galitski05,cho11,xu09 which have the mutual chern - simons gauge structures in different contexts . in this paper , we focus on the superconducting state to develop an effective description based on the mutual chern - simons gauge theory in sec . ii . instead of a conventional ginzburg - landau theory in description of the fluctuations of the pairing order parameter , we obtain a gauge description of the low - lying collective fluctuations of charge and spin currents , which are essentially the phase fluctuations of the superconducting order parameter . we show that the meissner effect in such a superconductor is basically controlled by a spin gap . in sec . iii , we further identify a non - bcs - type order parameter @xmath9 that characterizes the sc phase coherence . here @xmath9 is related to the correlation function of two wilson loops ( of @xmath4 and @xmath10 ) and is topological as it merely depends on the gaussian linking number under a continuous deformation of the loops . such a relation is similar to the connection between the chern - simons theory and knot theory first studied in ref . . but by coupling to the matter fields , we show that the spin correlations strongly renormalize @xmath9 such that the phase coherence terminates at a critical point where the spin gap vanishes . in sec . iii c , we shall propose a direct experimental probe of such a topological nature of the gauge interaction between the spin and charge degrees of freedom . finally , in sec . iii d , we show that a fermionic excitation can emerge in the superconducting phase , which is gauge neutral and thus coherent , resembling a bogoliubov quasiparticle . the lattice field theory formulation of the phase string theory@xcite for a doped mott insulator has been recently constructed in refs . in terms of partition function : @xmath11d[h^{\dagger } , h , b^{\dagger } , b]e^{-s } \label{partition}\]]in which the action @xmath12 ( @xmath13 denotes spacetime coordinates ) with @xmath14 here the first two terms , @xmath15 and @xmath16 , are the lagrangian densities for the matter fields , holon @xmath17 and spinon @xmath18 , which carry charge and spin degrees of freedom , respectively . they are given by @xmath19equations ( [ lh ] ) and ( [ ls ] ) show that the bosonic holon and spinon fields minimally couple to two lattice u(1 ) gauge fields , @xmath20 and @xmath21 , respectively , in 2 + 1 dimensions ( @xmath22 @xmath23 , with @xmath24 , @xmath25 ) . ( @xmath26 is the external electromagnetic field which solely couples to the holons@xcite . ) the parameters @xmath27 and @xmath28 are effective hopping integral and superexchange coupling , respectively , and @xmath29 is the difference operator defined along the imaginary time axis . the lagrangian multipliers , @xmath30 , control the total numbers of holons and spinons , and the last terms in @xmath31 describe the on - site repulsions , which soften the hard - core boson condition . it is noted that this effective theory is underpinned by a so - called bosonic rvb order parameter : @xmath32 , which describes a short - ranged spin liquid state in the spinion background in the underdoped regime of a doped mott insulator@xcite . the gauge fields , @xmath21 and @xmath20 , capture the essential physics of the phase string effect@xcite in a doped mott insulator via the following lattice mutual chern - simons term in eq . ( [ s ] ) @xmath33with @xmath34 the integer fields due to the compactness of the lattice field theory@xcite , which are important in restoring the spin symmetry . without @xmath34 , eq . ( [ lmcs ] ) is recovered the present theory reduces to the original mutual chern - simons gauge theory formulation of the phase string theory in ref . . superconductivity can naturally arise in a saddle - point state of the above mutual chern - simons gauge theory , in which holons become bose - condensed and spinons are gapped@xcite . in the folllowing , we shall derive a general effective theory governing the low - energy fluctuations of the gauge fields @xmath35 around this saddle - point . this new effective theory will replace the conventional ginzburg - landau theory to describe the generalized rigidity of a non - bcs superconducting state . by minimizing the action @xmath36 in eq . ( [ partition ] ) with regard to @xmath37 and @xmath38 , one can find the saddle - point solution of @xmath39 , i.e. , @xmath40 defined by @xmath41 in the presence of a uniform distribution of the holon condensate with @xmath42 as the holon concentration . one can always choose the temporal components @xmath43 by shifting @xmath30 in eqs . ( [ lh ] ) and ( [ ls ] ) . at the saddle point , one has @xmath44 . equation ( [ sp ] ) can be understood as that a fictitious uniform flux @xmath45 is generated by the average distribution of the holon condensate via @xmath46 , while the vanishing net flux in @xmath47 is due to the fact that the spinons are short - range paired with @xmath48 ( where @xmath49 denotes the expectation value at the saddle - point ) . around the saddle - point , the gauge field @xmath21 may be rewritten as @xmath50then an effective low - energy lagrangian of the weak fluctuations of @xmath20 and @xmath21 around the saddle - point can be derived as follows @xmath51=\frac{g_{\mu } ^{h}}{2\pi ^{2}}% ( a_{\mu } ^{s}+a_{\mu } ^{e})^{2}+\frac{g_{\mu } ^{s}}{2\pi ^{2}}\left ( \epsilon ^{\mu \nu \lambda } \partial _ { \nu } a_{\lambda } ^{h}\right ) ^{2}+% \frac{i}{\pi } \epsilon ^{\mu \nu \lambda } a_{\mu } ^{s}\partial _ { \nu } a_{\lambda } ^{h}\text { \ . } \label{effectivel}\ ] ] here the first term on the right - hand - side ( rhs ) of eq . ( [ effectivel ] ) arises from @xmath15 in eq . ( [ lh ] ) after considering the holon condensation@xcite , with @xmath52 and @xmath53 . note that the gauge field @xmath4 is fixed in the unitary gauge by absorbing goldstone boson field @xmath54 in the holon superfluid @xmath55 , and the superfluid density fluctuation @xmath56 has been integrated out around @xmath57 . the second term on the rhs of eqs . ( [ effectivel ] ) originates from the spinon term @xmath16 in eq . ( [ ls ] ) after integrating out the spinon field @xmath58 , with @xmath59 , @xmath60 ( the details are outlined in appendix [ appendix_s ] ) . here , @xmath61 denotes the spin mass gap which is an important parameter as discussed in appendix a below eq . ( [ ls-1 ] ) . the parameters @xmath62 and @xmath63 as given in eq . ( [ g ] ) . note that in refs . , by assuming that the state is deep inside the superconducting phase , the limit of @xmath64 has been taken such that @xmath65 there . equation ( [ effectivel ] ) corresponds to the general case for a finite spin mass gap @xmath61 . the last term on the rhs of eq . ( [ effectivel ] ) is the mutual chern - simons term with the high - energy fluctuations of @xmath66 omitted . note that the role played by @xmath67 in constructing a topological object like the magnetic flux quantization has been discussed in ref . and the references therein . to see that the effective lagrangian @xmath68 in eq . ( [ effectivel ] ) indeed describes a superconducting state , one may further integrate out @xmath4 and @xmath69 to arrive at a low - energy action @xmath70=\int d^{3}x\frac{m_{p}{}^{2}}{2}({\mathbf{a}}% ^{e\perp } ) ^{2 } , \label{seff}\]]where the summation @xmath71 is replaced by integration in the continuum limit , with the external magnetic field @xmath72 fixed in the coulomb gauge ( @xmath73 ) . here the transverse photon mass @xmath74 is given by @xmath75with the london penetration depth @xmath76 in which @xmath77 . therefore , so long as @xmath78 , the system does exhibit the meissner effect with a finite @xmath79 in response to an external magnetic field . the physical electric current @xmath80 satisfies the london equation : @xmath81}{\delta \mathbf{a}% ^{e\perp } } = \frac{1}{\lambda _ { p}^{2}}\mathbf{a}^{e\perp } \text{. } \label{london}\]]equations ( [ mp ] ) and ( [ lambda - p ] ) will recover the previous result@xcite obtained in the ideal superconductor limit of @xmath64 , where the london penetration depth @xmath79 approaches the shortest @xmath82 . namely , in the large spin - gap limit , the superfluidity of a holon condensate is well protected . however , in the opposite limit of @xmath83 , i.e. , with the spin gap closing up , one finds @xmath84 in eq . ( lambda - p ) such that the meissner effect disappears . it implies that the superconducting phase coherence can be destroyed by the gapless spin excitations at zero temperature , even in the presence of a bare holon condensate with a finite @xmath82 . this is a peculiar manifestation that the neutral spin degrees of freedom can strongly influence the superconducting condensate . the quantum phase transition of superconductivity occurring here is quit different from the conventional transition of bcs - type as in the latter a charge ( superconducting ) gap must be closed up at the fermi surface . physically , with @xmath83 , the proliferation of neutral spinons will disorder the superconducting phase coherence via the @xmath6-vortices bound to them in the mutual chern - simons gauge theory . such a superconducting phase transition can also occur at a finite temperature at @xmath85 @xmath86 which is of non - bcs - type ( see ref . ) . finally it is noted that @xmath87 with a spinon condensation is a necessary , but not a sufficient condition for the emergence of a true afm ordering in the mutual chern - simons gauge theory . a so - called _ bose insulating phase _ ( bi ) with @xmath87 but @xmath88 has been proposed@xcite to appear as an intermediate phase between the superconducting and afm phases . in other words , the present superconducting phase can be terminated at a critical point defined by @xmath83 before a true afm order can set in with further reducing doping concentration near the half - filling . the effective low - energy lagrangian @xmath68 in eq . ( effectivel ) has replaced the conventional ginzburg - landau equation to describe superconductivity in the mutual chern - simons theory . now we examine the following question : what is the essential physical quantity that captures the nature of superconductivity , since the bcs order parameter ( as governed by the conventional ginzburg - landau equation ) no longer explicitly appears in the low - energy effective theory ? to be sure , the off diagonal long order ( odlro ) of superconductivity is still described by the cooper pairing order parameter in the phase string theory of doped mott insulators@xcite . nevertheless , since the cooper pairing amplitude can persist well beyond the superconducting phase , as determined@xcite by the bosonic rvb parameter @xmath89 and the holon condensation @xmath90 , the low - lying excitations and the phase transition of superconducting phase are essentially characterized by the topological phase fluctuations via the mutual chern - simons gauge fields , @xmath4 and @xmath69 , in the low - energy effective lagrangian ( [ effectivel ] ) . physically , @xmath4 and @xmath10 are related to the spin and hole currents in the mutual chern - simon gauge theory by the following equations of motionkou05,ye2 @xmath91where the internal electric fields in the imaginary time representation @xmath92 and the spin and charge currents are defined as @xmath93 . hence the low - lying fluctuations of @xmath4 and @xmath69 in eq . ( [ effectivel ] ) actually describe the current fluctuations of the spin and charge degrees of freedom , while the amplitude fluctuations of the matter fields remain gapped in the superconducting phase . in order to characterize the gauge fluctuations of @xmath94 , a pair of wilson loops have been introduced@xcite : @xmath95\equiv \left \langle \hat{w}^{s , h}[\mathcal{c}]\right \rangle , \label{wilson}\]]where@xmath96e^{-\sum_{x}\mathcal{l}_{eff}} ... \]]and @xmath97\equiv \exp \big [ i\sum_{x}a_{\mu } ^{s , h}(x)j_{% \mathcal{c}}^{\mu } ( x)\big],\ , \ , \label{w - operator}\]]in which the ( 2 + 1)-dimensional unit current , for a test holon or a test spinon , @xmath98 @xmath99 for the link @xmath100 @xmath101 , and is zero otherwise on a close loop @xmath102 . physically , the wilson loop @xmath103 $ ] or @xmath104 $ ] probes the interaction of a pair of test holons or spinons at a distance @xmath105 , defined by @xmath106 or @xmath107 , via@xcite @xmath108 , \label{confinev}\]]if @xmath102 is taken as a spacetime rectangle with length @xmath109 ( @xmath105 ) in the imaginary time ( spatial ) direction . in refs . , in the limit of large spin gap ( @xmath64 ) , one finds @xmath110 and @xmath111 @xmath112 . namely the holons are deconfined while the spinons are logarithmically confined . it is consistent with the present saddle - point state of holon condensation discussed in sec . ii . in the presence of a finite spin gap @xmath61 , we find that the above conclusion remains true , with the the confining potential , between a test spinon and test anti - spinon , modified to @xmath113where @xmath82 is replaced by @xmath79 defined in eq . ( lambda - p ) ( @xmath114 is an ultraviolet cutoff of the distance ) . the result indicates that the spinons will experience a confinement - deconfinement transition at @xmath84 or @xmath83 , coinciding with the superconducting phase transition critical point at @xmath115 . on the other hand , @xmath106 for holons remains deconfined . the wilson loops , @xmath116 $ ] , can provide important information on the confinement / deconfinement of spinons and holons in the superconducting phase as discussed above . however , to understand why @xmath117 $ ] behave as such , the topological nature of the mutual chern - simons effective theory in eq . ( [ effectivel ] ) has to be further revealed . for this purpose we define @xmath118\ , \hat{w}^{h}[\mathcal{c}_{2}]\right \rangle \ , , \label{topo}\]]with the normalization factor @xmath119\,w^{s}[\mathcal{c}_{2 } ] . \label{k}\ ] ] let us first consider the case of a _ pure _ mutual chern - simons term ( [ mcs ] ) without coupling to the matter fields [ i.e. , omitting the first two terms in eq . ( [ effectivel ] ) ] , a straightforward calculation shows @xmath120where @xmath121 , denotes the gaussian linking number , which is the winding number of the close loops , @xmath122 and @xmath123 as shown in fig . [ figure_geometry ] ( a ) and ( b ) . to obtain eq . ( [ link number ] ) , we write down the explicit expression of @xmath124 in a pure mutual chern - simons gauge theory , @xmath125e^{-\int d^{3}x\left ( \frac{i}{\pi } \epsilon ^{\mu \nu \lambda } a_{\mu } ^{s}\partial _ { \nu } a_{\lambda } ^{h}-a_{\mu } ^{s}j_{\mu } ^{\mathcal{c}_{1}}-a_{\mu } ^{h}j_{\mu } ^{\mathcal{c}_{2}}\right ) } \]]where , the unit spacetime currents @xmath126 and @xmath127 keep track of spacetime loops , @xmath122 and @xmath128 , respectively . integrating over @xmath4 leads to a flux constraint on @xmath129 configuration , namely , @xmath130 . it means that the gauge field @xmath69 forms a spacetime closed @xmath131-fluxtube along @xmath122 . @xmath132 denotes the spacetime flux piercing @xmath128 . noting that there are @xmath133 times for the fluxtube @xmath122 piercing @xmath128 , one arrives at @xmath134 , with @xmath135 . ignoring the @xmath136-periodicity , we thus obtain eq . ( [ link number ] ) . in witten s seminal paper@xcite , a similar gaussian linking number structure was found in the chern - simons gauge theory , which is related to the jones polynomial in the knot theory . and @xmath128 in ( 2 + 1)-dimensional euclidean spacetime . @xmath137 is an arbitrary loop but restricted inside spatial plane ( i.e. , the xy plane ) at some imaginary time . @xmath128 is an arbitrary loop with a definite gaussian linking number @xmath138 ( @xmath139 in ( a ) ; @xmath140 in ( b ) ) that characterizes the number of times that @xmath141 winds around @xmath122 . the ball ( in blue ) and the arrow ( in red ) stand for the test holon and test spinon , respectively . the arrows on the two loops indicate the direction of test spinon current and test holon current.,width=377 ] in fig . [ figure_experiment ] , a gedankenexperiment is shown to illustrate the case of winding number @xmath139 in the ( 2 + 1)-dimensional euclidean spacetime , with a pair of test spinon and antispinon creating and annihilating on a close loop @xmath128 . if the background spin degrees of freedom are gapped with @xmath142 , the test spinons can not induce spin excitations from the ground state and thus the net contribution to @xmath143 comes from the @xmath6 flux phase shift ( with @xmath144 mentioned above due to the pure mutual chern - simons action . on the other hand , if the spin gap @xmath61 becomes finite , then the spin excitations can be induced from the spinon background to renomalize @xmath124 . now we consider the full @xmath68 given in eq . ( [ effectivel ] ) . by choosing the close loop @xmath122 of the test holon within the spatial @xmath145-plane , one finally arrives at the following compact result ( for the details , see appendix b ) : @xmath146we see that @xmath9 remains to be real and positive as a topological quantity , which is invariant under a continuous deformation of loops @xmath147 and @xmath128 for a given gaussian linking number @xmath148 , as illustrated in fig . [ figure_geometry ] . but @xmath9 is renormalized by a factor @xmath149 which is solely determined by the spin gap @xmath61 . in particular , in the limit of @xmath83 , one finds @xmath150 . for simplicity ) . ( a - d ) are four snapshots of spatial plane subsequently at imaginary time @xmath151 . ( a ) a pair of test spinons with opposite spin indices are created at a spatial site @xmath152 at @xmath153 . ( b ) two spinons arrive at new spatial positions at @xmath154 via the dashed trajectories . ( c ) at @xmath155 , a holon current circuit @xmath122 is located in plane , indicating that an aharonov - bohm phase around @xmath122 is picked up by the test holon . ( d ) at @xmath156 , the pair of test spinons meet each other again at some position and then are annihilated into vacuum simultaneously . , width=377 ] if one rewrites eq . ( [ i ] ) by @xmath157where @xmath79 becomes a function of topological order parameter @xmath124 for a given gaussian linking number @xmath139 , then the limit @xmath84 , i.e. , the disappearance of superconductivity , will correspond to @xmath158 . in other words , the topological quantity @xmath9 can be regarded as the superconducting order parameter in the effective theory of eq . ( [ effectivel ] ) , and the quantum phase transition at @xmath115 occurs when @xmath9 gets fully screened by the gapless spin excitations at @xmath87 . as illustrated in fig . [ figure_geometry ] , @xmath159 indicates that a test spinon can be always feltby a vortex - like current loop of a test holon if their mutual winding number @xmath160 . physically , since the holons are in the bose condensed state , a spinon excitation will then induce a supercurrent vortex response from the condensed holons throughout the sample to form a so - called spinon - vortex@xcite . this is the origin of spinon confinement discussed above , since each spinon - vortex composite is logarithmically divergent in energy which will result in vortex - antivortex confinement . in a non - superconducting phase where the spinons become deconfined due to the vortex - antivortex disassociation and proliferation , one finds @xmath161 in the gedankenexperiment shown in fig . [ figure_experiment ] . namely , @xmath158 describes a phase disordered superconductor . we propose an experimental setup to probe the topological characterization @xmath124 of the superconducting phase in fig . [ figure_prediction ] . note that @xmath159 reflects the mutual statistical interaction between spin and charge degrees of freedom ( cf . [ figure_geometry ] ) . by using the idea illustrated in the gedankenexperiment of fig . figure_experiment , one may inject free test spinons by heating up the middle semi - island part of the sample in fig . [ figure_prediction ] to a temperature slightly higher than @xmath0 , such that free spinons can be excited and flow towards the lower @xmath109 end connecting to the outside superconducting ring . with @xmath162 inside the ring , one expects that such an injection of free spinons can generate supercurrents in the superconducting ring in a form of significant noise spikes due to the unpolarized spinon - vortex effect . , while the outside part remains below @xmath0 and spinons are confined into vortex - antivortex pairs . the confinement is indicated by the black wavy line between two spinons . the small directional discs ( red arrows ) represent vortices ( spinons ) . in such a mutual chern - simons superconductor at @xmath163 @xmath164 , spinons become deconfined to form free spinon - vortices qi07 . then , the flow of such free spinons from the semi - island towards the cooler end connecting to the outside superconducting ring is expected to generate random supercurrents in the latter as indicated by @xmath80 ( blue dashed line ) . , width=302 ] instead of raising the temperature , one may also apply an in - plance magnetic field @xmath165 to reduce the spinon mass gap @xmath61 inside the semi - island part by @xmath166 , where @xmath167 is the land factor and @xmath168 is the bohr magneton . then a sufficiently strong @xmath165 may drive the semi - island part into a spinon - vortex liquid state with @xmath158 via the zeeman effectqi07 . at the same time , the gradient of @xmath165 can also naturally cause the free spinons flow into the superconducting ring to induce the similar supercurrent response shown in fig . [ figure_prediction ] . the mutual chern - simons theory given in eq . ( [ s ] ) is an all - boson description of doped mott insulators . in the above subsections we have studied the low - lying collective fluctuations of this theory based on the effective lagrangian ( [ effectivel ] ) . an important question that arises naturally is if and how a low - energy fermionic excitation can _ emerge _ in such an all - boson theory , which is independent of the above collective excitations . to give a proof of principle , we first note that the mutual chern - simons term in eq . ( [ effectivel ] ) can be replaced by an auxiliary lagrangian in real - time representation : @xmath169where @xmath170 is an auxiliary 4-component dirac spinor in ( 2 + 1)-dimensional spacetime with flavor indices @xmath171 . @xmath172 . here @xmath173-matrices are defined in 4-dimensional reducible representation of 2-dimensional clifford algebra , i.e. , @xmath174 , where @xmath175 and @xmath176 are pauli matrices . the covariant derivative @xmath177 is defined as @xmath178 the @xmath179 matrix @xmath180 , where @xmath181 is the @xmath182 unit matrix . the second term in eq . ( [ aux ] ) is mass term with @xmath183 . @xmath184 preserves a @xmath185 gauge symmetry where subscripts @xmath186 and @xmath187 denote axial vector type and ordinary vector type @xmath188 gauge invariance , respectively . it originates from the fact that @xmath20 is an ordinary vector field with gauge group generator 1 while @xmath69 is an axial vector field with gauge group generator @xmath189 @xcite . the parity and time - reversal symmetries are conserved even in the presence of finite mass gap which may be expressed as : @xmath190 , where , @xmath191 ( @xmath192 ) is upper ( lower ) 2-component spinor of @xmath170 . here , @xmath193 denotes the chiralityvafa84 . in the case of parity ( @xmath194 is parity operator ) , @xmath195it is easy to check that the mass term is invariant under such parity transformation in ( 2 + 1)-dimensional spacetime . integrating over @xmath170 leads to a pure mutual chern - simons term by ignoring maxwell terms ( @xmath196 momentum squared ) in the one - loop quantum correction ( fig . [ vertex ] ) in the long wavelength limit @xmath197 . the key steps of derivation are given as follows . , @xmath69 , and @xmath198 , correspondingly . the first two diagrams correspond to maxwell terms , while the last one gives the non - vanishing mutual chern - simons vertex.,width=302 ] as shown in fig . [ vertex ] , the vacuum polarization for @xmath4 and @xmath129 at one - loop level@xcite ( @xmath199 is a 3-dimensional euclidean momentum ) : @xmath200.\]]where the total flavor number @xmath201 . the expansion around small @xmath199 is given by @xmath202the propagators of gauge fields can be constructed in the landau gauge and the maxwell terms read out : @xmath203 with @xmath204 and @xmath205 . meanwhile , the nonvanishing mutual chern - simons term for @xmath4 and @xmath69 in fig . [ vertex ] is also radiatively generated without breaking parity and time - reversal symmetries . note that , upper(lower)-component spinor has mass @xmath206 ( @xmath207 ) : @xmath208where , the superscript @xmath209 denoting upper and lower 2-component spinor contributions respectively , and @xmath210 with identity : @xmath211 . in the sufficiently low momentum transfer limit ( @xmath212 ) , it is expanded as power series of @xmath199 : @xmath213 in eq . ( [ eq_sigma ] ) , the sign factor @xmath214 renders vanishing of abelian chern - simons terms ( @xmath215 and @xmath216 ) . but due to the structure of matrix @xmath189 , the crossing term , i.e the mutual chern - simons term survives as expected . mathematically , the above quantum corrections are organized as below : @xmath217where the maxwell terms are ignored in the long wavelength limit . it is important to note that the fermionic auxiliary field @xmath170 is not gauge - neutral as it still couples to @xmath20 and @xmath218 in eq . ( [ aux ] ) . a true fermion excitation of quasiparticle should be gauge - neutral without knowing @xmath4 and @xmath69 . it is only possible if there exists a residual attractive potential such that a holon and a spinon excited from the condensates can form a bound state properly with the auxiliary field @xmath170 . the resulting object @xmath198 will be both _ fermionic _ and _ coherent _ because all statistical gauge charges carried by @xmath170 can be exactly canceled by that of the spinon and holon partners . in fig . [ qploop ] , the spacetime loop of a composite - particle @xmath219 is illustrated as a gauge neutral and thus physically observable elementary excitation at low energies . in particular , it carries the right quantum number as a bogoliubov quasiparticle . as noted before , to produce the right prefactor @xmath220 of the pure mutual chern - simons term , @xmath170 has totally @xmath221 branches of positive energy solutions , which are consistent with the existence of totally four quasiparticle nodal regions in the brillouin zone , although the detailed energy spectrum depends on the residual attractive potential which goes beyond the present long - wavelength effective description . on the other hand , the existence of a _ nodal _ fermionic bogoliubov quasiparticle has been recently demonstrated based on a ground state wavefunction approach to the @xmath7-@xmath8 model in the phase string formulation , in which the quasiparticle as a similar bound state has been explicitly shown@xcite . therefore , the present field - theory formulation provides a _ proof - of - principle _ for a fermionic quasiparticle mode emerging in the mutual chern - simons theory . is composed by a spinon ( red arrow ) , a holon ( blue ball ) and a dirac fermion ( small grey ball @xmath170 ) . its spacetime loop is composed by three loops rendering gauge charge neutrality.,width=302 ] in this work , we have obtained an effective theory of the low - lying fluctuations in the superconducting state of the mutual chern - simons theory . in contrast with a conventional ginzburg - landau action for a bcs superconductor , the present effective description in eq . ( [ effectivel ] ) deals with peculiar phase fluctuations in the doped mott insulator via a pair of mutual chern - simons gauge fields , which are related to the fluctuations of charge and spin currents of the condensate . the superconducting phase coherence is found to be properly characterized by a topological order parameter which measures the mutual entanglement of the charge and spin currents in the low - energy effective theory . in particular , a zero - temperature quantum phase transition is shown to occur , due to the vanishing of this topological order parameter , when the spin gap is reduced to zero at low doping . we have discussed a possible experimental probe into the long - range entanglement between the supercurrents and the neutral spin excitations as a unique signature of this non - bcs superconducting state . we have also given a proof - of - principle how a fermionic quasiparticle can emerge in the mutual chern - simons theory . together with the previously identified flux quantization@xcite at @xmath222 , the non - bcs superconductivity in the mutual chern - simons theory is well established . we acknowledge stimulating discussions with z. c. gu , s .- kou , z .- x . liu , c .- s . tian , y. wang , z. wang , n. xu , and y .- z . you . this work is supported by nsfc grant no . 10834003 , by most national program for basic research grant no . 2009cb929402 , no . 2010cb923003 . 99 for a review , see , p. a. lee , n. nagaosa , and x. g. wen , rev . phys . * 78 * , 17 ( 2006 ) . s. a. kivelson , d. s. rokhsar and j. p. sethna . b * 35 * , 8865 ( 1987 ) . z. zou and p.w . anderson , phys . b * 37 * , 627 ( 1988 ) . t. senthil and matthew p. a. fisher , phys . b * 62 * , 7850 ( 2000 ) . z. y. weng , int . j. mod b * 21 * , 773 ( 2007 ) . p. ye , c. s. tian , x. l. qi , and z. y. weng , phys . lett . * 106 * , 147002 ( 2011 ) ; p. ye , c. s. tian , x. l. qi , and z. y. weng , nucl . b * 854 * , 815 ( 2012 ) . s. p. kou , x. l. qi , and z. y. weng , phys . b * 71 * , 235102 ( 2005 ) . d. birmingham , m. blau , m. rakowski , and g. thompson , phys . rep . * 209 * , 129 ( 1991 ) . z. y. weng , d. n. sheng , y .- c . chen , and c. s. ting , phys . b 55 , 3894 ( 1997 ) . k. wu , z. y. weng , and j. zaanen , phys . b * 77 * , 155102 ( 2008 ) . v. m. galitski , g. refael , m. p. a. fisher , and t. senthil , phys lett . * 95 * , 077002 ( 2005 ) . m. c. diamantini , p. sodano , and c. a. trugenberger , nucl . b * 448 * , 505 ( 1991 ) ; nucl . b * 474 * , 641 ( 1996 ) . g. y. cho and j. e. moore , ann . of phys . * 326 * , 1515 ( 2011 ) . c. xu and s. sachdev , phys . b * 79 * , 064405 ( 2009 ) . e. witten , commun . math . phys . * 121 * , 351 ( 1989 ) . j. w. mei and z. y. weng , phys . b * 81 * , 014507 ( 2010 ) . j. smit , _ introduction to quantum fields on a lattice_. ( cambridge , 2003 ) x. l. qi and z. y. weng , phys . b * 76 * , 104502 ( 2007 ) . c. vafa and e. witten , commun . math . phys . * 95 * , 257 ( 1984 ) . robert d. pisarski , phys . d , * 29 * , 2423 ( 1984 ) . s. deser , r. jackiw , and s. templeton , ann . phys . , * 140 * , 372 ( 1982 ) . z. y. weng , new j. phys . * 13 * , 103039 ( 2011 ) . for a review , see , c. schubert , phys . rep . * 355 * , 73 ( 2001 ) . greiner and reinhardt , _ field quantization _ , ( spinger - verlag , 1996 ) . first of all , we note that the spinon lagrangian @xmath16 in eq . ( [ ls ] ) may be further simplified to a cp@xmath223 form in the long - wavelength , low - energy regime as follows@xcite @xmath224\ , , \label{ls-1}\]]here the temporal component is rescaled by the spin wave velocity @xmath225 with the coupling constant @xmath226 , and @xmath61 is treated as a dynamically generated spinon mass enforcing @xmath227 . the problem of studying the fluctuation of @xmath228 , or @xmath229 , in eq . ( [ ls-1 ] ) is then transformed into calculating the vacuum polarization tensor in ( 2 + 1)-dimensional scalar quantum electrodynamics in the presence of a uniform background magnetic field of @xmath46 defined in eq . ( [ sp ] ) . the technical challenge arises from the fact that because of @xmath46 , the momentum is no longer a good quantum number . nevertheless the issue is resolved by using the so - called string - inspired formalism@xcite . following the procedures and conventions in ref . , we obtain the polarization tensor : @xmath230 \int_{0}^{1}d\ell i^{\mu \nu } \label{bubble}\]]where the prefactor @xmath231 comes from the summation over the spin index , @xmath109 denotes the schwinger proper time , and spacetime dimension @xmath232 . here @xmath233 is a @xmath234 matrix , defined by @xmath235 with @xmath236the determinant in eq . ( [ bubble ] ) can be simply factorized into : @xmath237 = \frac{\pi \ , \delta \,t}{\sinh ( \pi \ , \delta \,t)}\text{,}\]]and the transversality of lorentz tensor @xmath238 is given by @xmath239by further restoring the spin wave velocity @xmath240 , we arrive at the maxwell term on the rhs of eq . ( effectivel ) which describes the fluctuation @xmath241 , with the coefficients given by @xmath242 in which the dimensionless coefficient @xmath243 depends on the dimensionless quantity @xmath244 as follows @xmath245note that @xmath246 in the limit of the absence of magnetic flux@xmath247 in @xmath46 . to obtain @xmath9 , we first note that the lagrangian with holon and spinon currents circulating on loops @xmath248 ( @xmath249 ) @xmath250=\frac{\delta } { 2{m^*}}% \left(a_{\alpha } ^s\right)^2+\frac{1}{2u_1}\left(a_0^s\right)^2 + \frac{i}{% \pi } \epsilon _ { \mu \nu \lambda } a_{\mu } ^h\partial _ { \nu } a_{\lambda } ^s+% \frac{1}{2}\frac{g_{\mu } } { \pi ^2}\left(\epsilon _ { \mu \nu \lambda } \partial _ { \nu } a_{\lambda } ^h\right)^2-i a_{\mu } ^sj_{\mu } ^{\mathcal{c}_1}-i a_{\mu } ^hj_{\mu } ^{\mathcal{c}_2}\ ] ] is quadratic in the gauge fields thus allows rigorously integrating out @xmath251 and @xmath252 in the generating functional @xmath253\hat{w}^h\left[\mathcal{c}_2% \right]\right \rangle = \int d [ a^s a^h]e^{-\int d^3x l_{\text{sc}}\left[% \mathcal{c}_1 , \mathcal{c}_2\right ] } \\ = \text{const}.\times \int d [ a^h]\exp \left \{-\int d^3x\left(\frac{g_{\mu } % } { 2\pi ^2}\left(\epsilon _ { \mu \nu \lambda } \partial _ { \nu } a_{\lambda } ^h\right)^2-i a_{\mu } ^hj_{\mu } \right)\right \}\ ] ] in which we have defined @xmath254 and absorbed terms unrelated to mutual correlation between @xmath255 and @xmath256 into const . which will be canceled by the denominator in the definition of @xmath9 . adopting the `` deformed '' lorentz gauge condition , @xmath257 and integrating out @xmath252 in the feynman gauge@xcite , we get @xmath253\hat{w}^h\left[\mathcal{c}_2% \right]\right \rangle = \\ \text{const}.\times \exp \left \{-\frac{1}{2}\frac{\pi ^2}{g_0g_1g_2}\int d^3x\int d^3y\sum _ { \mu } g_{\mu } j_{\mu } ( x)j_{\mu } ( y)f(x , y)\right \ } \label{eq:6}\ ] ] in which @xmath258 given that @xmath266 , eq . [ eq:10 ] is greatly reduced if @xmath255 lies within a spatial plane , @xmath267 in which @xmath268 is the gaussian linking number of @xmath255 and @xmath269 loops . to obtain the last line of above identity , the definition of @xmath270 and @xmath271 in eq . ( [ lambda - p ] ) is applied .

quantum teleportation is a quantum communication protocol where a sender , alice , aims to transmit to a receiver , bob , an unknown state of a @xmath1-dimensional quantum system ( qudit ) without sending the system itself . in the standard version of the protocol @xcite , alice and bob share two qudits in a pure and maximally entangled state ( the quantum channel ) . she then performs a joint measurement on her half of the entangled pair and the system carrying the state to be teleported . after that , she communicates the outcome of her measurement to bob through a classical channel with @xmath2 bits capacity . once bob learns alice s outcome , he performs a unitary operation on his previously entangled system , which transforms its state into the original one that alice aimed to transmit , and then the teleportation is complete . the consumption of a pure maximally entangled state and @xmath2 bits of classical communication are both necessary and sufficient conditions for deterministic and faithful teleportation @xcite , which means that alice transmits a quantum state to bob with unit probability of success and unity fidelity . when alice and bob share a pure but _ nonmaximally _ two - qudit entangled state , deterministic _ and _ faithful teleportation is no longer possible .- dimensional systems with @xmath3 @xcite . this case will not be addressed here . ] the best they can do is to teleport in a optimal way according to some figure(s ) of merit ( e.g. , average teleportation fidelity , probability of successful teleportation and classical communication cost ) . in this regard , many optimized schemes have been suggested to implement quantum teleportation @xcite . for instance , to obtain the optimal average teleportation fidelity deterministically , alice and bob have just to perform the standard protocol @xcite . on the other hand , if it is crucial for bob to receive from alice a perfect replica of an unknown qudit state , they can resort to the so - called _ perfect conclusive teleportation _ @xcite . in this scheme , faithful teleportation is possible with less than unity probability of success and the optimal protocol is the one that maximizes such probability . so far , several strategies have been proposed to implement the optimal protocol @xcite . in particular , for the purposes of the present work , we highlight the strategy of ref . @xcite where it is shown that to accomplish perfect conclusive teleportation in a optimal way , alice must implement an optimized measurement which discriminates unambiguously between a set of linearly _ independent _ and equally likely symmetric states @xcite . perfect conclusive teleportation applies only for a quantum channel with maximal schmidt rank @xcite - dimensional systems its maximum value is @xmath1 . [ fn : schmidt ] ] . in this article we investigate the problem of teleporting via nonmaximally entangled states with _ schmidt rank . since the success probability of faithful teleportation is zero in this case , we design an optimal _ imperfect conclusive teleportation _ protocol which can be summarized as follows : alice tries to teleport an unknown qudit state to bob and succeeds with the maximum achievable probability . when she succeeds , the teleportation fidelity , although less than unit , is also the maximum achievable for their shared quantum channel . such fidelity depends only on the schmidt rank of this channel and it is larger than the optimal one achieved , deterministically , by the standard protocol @xcite . this is relevant , for instance , in a scenario where it is sufficient for alice s and bob s purposes to teleport with a fidelity above some threshold and their quantum channel allows this only in a probabilistic way . similarly to the approach of ref . @xcite for quantum channels with maximal schmidt rank , we shall see here that if this number is not maximal , in the course of the teleportation process alice is faced with the problem of discriminating among nonorthogonal symmetric states which are linearly _ dependent _ and equally likely . then , we show that by implementing an optimized maximum - confidence ( mc ) measurement @xcite to accomplish this task , she can perform the optimal imperfect conclusive teleportation as described above . the mc discrimination for the aforementioned set of states was studied recently @xcite . specifically , in ref . @xcite we have found the optimal positive operator valued measure ( povm ) that maximizes our confidence in identifying each state in the set and minimizes the probability of obtaining inconclusive results . we also showed that after an inconclusive result , the input states are mapped into a new set of equiprobable symmetric states within a lower - dimensional hilbert space . if such space is not one dimensional , this new set could be submitted to another round of mc measurements , with which we could still gain some information about the input states , although with less confidence than before . this process can be repeated in as many stages as allowed by the input states , and we called it _ sequential maximum - confidence _ ( smc ) measurements . applying smc measurements to our imperfect conclusive teleportation protocol , we shall see that the number of stages that alice is allowed to implement , depends on both the schmidt rank and schmidt coefficients of the quantum channel , which she and bob know in advance . we then show that there are quantum channels for which it is possible to implement a @xmath0-stage smc measurement ( @xmath4 ) such that conclusive events at any stage lead to a teleportation fidelity above the optimal one achieved by the standard protocol @xcite . in this case , there will be an increase of the overall probability of teleportation with a fidelity above that limit and , consequently , alice and bob will be able to save resources for accomplishing the protocol , as we shall discuss later . we shall as well discuss some side effects of our protocol , as for instance , the requirement of ancillary systems , the increase in the classical communication cost , and the reduction of the overall teleportation fidelity ( which involves conclusive and inconclusive events ) when compared with the one achieved by the optimal deterministic protocol . the remainder of this article is organized as follows : in sec . [ review ] we show how the teleportation process via nonmaximally entangled states can be mapped to the problem of discriminating nonorthogonal symmetric states . we then discuss the optimal deterministic and perfect conclusive protocols within this context . in sec . [ results1 ] we briefly review the discrimination of symmetric states via mc measurements and present the optimal imperfect conclusive teleportation protocol . an explicit example is described to illustrate it . in sec . [ sec : results2 ] we describe smc measurements and apply such a strategy to the teleportation process . its benefits as well as its drawbacks are analyzed , and the previous example is used once again to illustrate the protocol . finally , in sec . [ summary ] , we summarize the results , discuss possible extensions of our work , and conclude . in this section we will describe the teleportation protocol within the framework of quantum computing @xcite . doing so , it will be straightforward to establish its connection with the problem of quantum state discrimination . in this regard we shall see that an optimal deterministic protocol and a probabilistic and faithful one can be associated with minimum - error and unambiguous discrimination strategies , respectively ( reminding that the latter case is a known result @xcite ) . this starting point will be important to understand how an optimal probabilistic and unfaithful teleportation protocol is connected with a maximum - confidence discrimination strategy , as we shall see in sec . [ results1 ] . suppose that alice wants to teleport to bob an unknown pure state of a qudit given by @xmath5 and the orthonormal set of states @xmath6 forms the computational basis spanning a @xmath1-dimensional hilbert space @xmath7 . the quantum channel to implement the protocol is given by a pure bipartite entangled state @xmath8 , where system 1 belongs to bob and system 2 to alice . without loss of generality , we assume that both systems live in a @xmath1-dimensional hilbert space , @xmath9 and @xmath10 , respectively . using the schmidt decomposition the state @xmath8 can be written as @xcite @xmath11 where @xmath12 are strictly positive real numbers satisfying @xmath13 . for simplicity we also assume that these coefficients are decreasingly ordered , i.e. , @xmath14 . note that we have included in the sum only the nonvanishing schmidt coefficients , and thereby @xmath15 represents the schmidt rank of the state @xmath8 ( see footnote [ fn : schmidt ] ) , where @xmath16 . as a final assumption , let the schmidt basis , which is composed by @xmath1 orthonormal states @xmath17 , coincide with the computational basis spanning @xmath18 for @xmath19 . if this is not the case in the beginning , alice and bob can apply local unitary rotations to align the schmidt basis with the computational one @xcite . as we mentioned in the introduction , alice must now perform a joint measurement on the quantum systems in her possession , in this case systems 2 and 3 . generalizing the approach of refs . @xcite for qubits to higher - dimensional spaces , this measurement can be described by a two - step process . in the first step , alice implements a unitary operation given by a generalized controlled - not gate , @xmath20 , having system 2 as the control qudit and system 3 as the target qudit ( see fig . [ fig : q_circuit1 ] ) . the action of @xmath20 onto the computational basis is defined by @xcite @xmath21 where @xmath22 denotes subtraction modulo @xmath1 . applying this operator on the initial three - system state @xmath23 and using eqs . ( [ eq : unknown])([eq : gxor ] ) , we obtain @xmath24 where @xmath25 and @xmath26 are the generalized pauli operators , defined by their action on the computational basis as @xmath27 and @xmath28 respectively , where @xmath29 denotes addition modulo @xmath1 . the states @xmath30 in eq . ( [ base ] ) are given by @xmath31 for @xmath32 . they form a set of @xmath1 symmetric states , if they satisfy the following conditions : @xmath33 , @xmath34 @xcite . this is the case for the states in eq . ( [ symmetricstates ] ) with the unitary given by eq . ( [ eq : z ] ) . ] under the action of @xmath35 , and are characterized by the state coefficients @xmath36 defining the quantum channel in eq . ( [ quantumchannel ] ) . additionally , the schmidt rank of this channel , @xmath15 , will determine if the set is linearly independent ( @xmath37 ) or linearly dependent ( @xmath38 ) . the second step of alice s procedure is to perform measurements on the systems 2 and 3 individually . from eq . ( [ base ] ) we see that the state of system 3 can be perfectly determined by a projective measurement in the computational basis . the measurement of system 2 , however , will depend on the features of the quantum channel [ which defines the features of the set @xmath39 in eq . ( [ symmetricstates ] ) ] and a prearrangement between alice and bob regarding which type of teleportation protocol they want ( or they can ) implement . therefore , the teleportation process is mapped to the problem of alice s ability to perform an optimized measurement which discriminates @xmath1 symmetric states @xmath39 occurring with equal prior probabilities @xmath40 . in what follows we discuss the possibilities for teleportation in the context of state discrimination . maximally entangled , @xmath41 and the teleportation fidelity is 1 . otherwise , @xmath42 and the maximum fidelity is the one given by eq . ( [ eq : f_opt ] ) . ] in order to achieve faithful and deterministic teleportation , alice and bob must share a maximally entangled state in the first place . when this is the case , we have in eq . ( [ quantumchannel ] ) @xmath37 and @xmath43 for all @xmath44 . thus , the symmetric states in eq . ( [ symmetricstates ] ) become @xmath45 where @xmath46 is the discrete fourier transform defined by @xmath47 therefore , system 2 is mapped to a set of orthogonal states which can be perfectly and deterministically discriminated . due to this fact , alice can perform faithful and deterministic teleportation . the whole process can be understood from eq . ( [ base ] ) and the quantum circuit shown in fig . [ fig : q_circuit1 ] . alice applies a @xmath20 between systems 2 and 3 , followed by an inverse fourier transform at 2 . then she measures both systems in the computational basis and sends a @xmath2-bit message to bob through a classical channel communicating the result @xmath48 that she has obtained . finally , bob uses this information and apply the unitary transformations @xmath49 on his system whose state becomes a perfect replica of the unknown state @xmath50 . before proceeding , we briefly recall a result that will be useful for the discussions that follow . in ref . @xcite , horodecki _ et al . _ demonstrated that for any given initial entangled state @xmath51 ( pure or mixed ) shared between alice and bob , the optimal teleportation fidelity over all possible protocols performed via local operations an classical communication ( locc ) is given by @xmath52 here , the superscript @xmath53 indicates whether the protocol belongs to a class of deterministic or probabilistic protocols . the parameter @xmath54 , called the _ singlet fraction _ , is defined as @xmath55,\ ] ] and measures the maximum fidelity that can be achieved between @xmath51 and a maximally entangled state @xmath56 when the former is subjected to @xmath53 transformations using locc . in the case of faithful and deterministic teleportation discussed above , we have that the singlet fraction will be unit , and so will be the fidelity in ( [ eq : horodecki ] ) . in the remainder of this work we shall associate @xmath53 not only with the class of the protocol but also with a given state discrimination strategy within that class . in addition , we will see that the singlet fraction is equal to the confidence achieved in the discrimination of the symmetric states in eq . ( [ symmetricstates ] ) when such a strategy is employed . let us now assume that alice and bob share a nonmaximally entangled pure state and they agree beforehand to implement teleportation deterministically . following the discussion in sec . [ subsec : general ] , after applying the @xmath20 , system 2 is mapped to a set of @xmath1 equally likely nonorthogonal symmetric states given by eq . ( [ symmetricstates ] ) . these states can not be discriminated , simultaneously , in a perfect and deterministic way . due to her prior agreement with bob , alice must then abandon the requirement for perfection and implement a measurement with @xmath1 possible outcomes where errors are allowed to occur . in this case , the optimized measurement will be the one that minimizes the probability of making an erroneous identification of the state , which is known as minimum - error ( me ) discrimination @xcite . it turns out that for the aforementioned set of states of system 2 , the physical implementation of the optimal me measurement follows exactly the same procedure of the standard teleportation protocol ( see fig . [ fig : q_circuit1 ] ) , that is , an inverse fourier transform followed by a projective measurement on the computational basis @xcite equally likely symmetric states defined by eq . ( [ symmetricstates ] ) is given by the @xmath1-outcome projective measurement @xmath57 . it applies for both linearly independent and linearly dependent states . in the latter case , this measurement realizes the optimal povm on the lower - dimensional subspace where the symmetric states are restricted in @xcite . [ fn : opt_me ] ] . therefore , when alice applies the @xmath20 , measures system 3 in the computational basis , and implements an optimized me measurement on system 2 , she is just performing the standard teleportation ( which is completed as indicated in the quantum circuit of fig . [ fig : q_circuit1 ] ) . by performing standard teleportation , it is guaranteed by refs . @xcite that the fidelity between bob s final state and alice s unknown state , averaged over all possible input states @xmath50 , will be the maximum achievable @xmath58\equiv f^{\rm me},\ ] ] where @xmath12 are the schmidt coefficients of the entangled state given by eq . ( [ quantumchannel ] ) . the superscript `` me '' now indicates the optimal discrimination strategy adopted by alice to achieve this optimal fidelity in a deterministic protocol . comparing eqs . ( [ eq : horodecki ] ) and ( [ eq : f_opt ] ) we find that @xmath59 the quantity on the right - hand side is precisely the probability of correctly discriminating among equally likely symmetric states via me measurements , @xmath60 , which in turn is equal to the maximum confidence achieved by a me measurement in such discrimination,^{\mathcal{s}}$ ] will represent our _ maximum _ confidence in taking a measurement outcome @xmath61 to indicate the input state as @xmath62 , when the discrimination strategy @xmath53 is employed . [ fn : mc - def ] ] @xmath63^{\rm me}$ ] @xcite , that is , @xmath64^{\rm me}.\ ] ] the connection between optimal deterministic teleportation and optimal me discrimination established here , will be important to understand how the optimal teleportation via mc measurements works in practice , as we will see in sec . [ results1 ] . from the discussion in sec . [ subsec : general ] , it follows that in order to teleport faithfully via a nonmaximally entangled channel , alice must be able to discriminate @xmath1 nonorthogonal states of system 2 [ @xmath30 in eq . ( [ symmetricstates ] ) ] without error . this is possible only if inconclusive outcomes are allowed to occur @xcite and the states are linearly independent @xcite , that is , @xmath37 in eq . ( [ symmetricstates ] ) . the first condition makes the protocol probabilistic , while the second requires the schmidt rank of the channel to be maximal . assuming that the latter condition holds , the optimal teleportation protocol will be the one in which alice s measurement strategy to determine the states @xmath30 minimizes ( maximizes ) the probability of inconclusive ( conclusive ) outcomes . this strategy is known as optimal unambiguous discrimination ( ud ) , and the optimized measurement that implements it for equally likely symmetric states is known @xcite . the teleportation protocol via optimal ud of symmetric states has been studied in ref . it is a perfect conclusive protocol because after a conclusive event in alice s discrimination process , the teleportation proceeds as in the standard scheme and in the end bob receives a perfect replica of the unknown state . thus , the teleportation fidelity will be @xmath65 where the subscript `` s '' labels conclusive ( or successful ) events . in terms of the singlet fraction , we have from eq . ( [ eq : horodecki ] ) @xmath66^{\rm ud}_{\rm s},\ ] ] where we used the fact that the confidence ( see footnote [ fn : mc - def ] ) achieved in the ud strategy is unity for each conclusive event . after an inconclusive event , unambiguous discrimination is forbidden @xcite and , accordingly , perfect teleportation . when it happens , alice and bob just discard the attempt and start the process again with another copy of the shared entangled pair , until she succeeds . in terms of resources this protocol will be more expensive than a deterministic one since it will require the use of ancillary systems ( ancilla ) , one extra bit of classical communication for each teleportation attempt until the successful one and , in general , more copies of the quantum channel to be efficient . the quantum circuits for this protocol are shown in figs . [ fig : q_circuit2](a ) and [ fig : q_circuit2](b ) , and they will be better understood along the discussion of teleportation via mc measurements . when alice and bob share a quantum channel with nonmaximal schmidt rank , i.e. , @xmath38 in eq . ( [ quantumchannel ] ) , the states @xmath30 in eq . ( [ symmetricstates ] ) become linearly dependent . in this case , perfect conclusive teleportation described above is impossible . it is possible , however , to implement optimal _ imperfect _ conclusive teleportation . as we shall see in this section , this is achieved when alice applies mc measurements @xcite to discriminate the states @xmath30 . with such strategy , conclusive events in the discrimination process occur with maximum success probability and lead to a teleportation with the maximum achievable fidelity for the quantum channel . we show that this fidelity depends only on the schmidt rank of the channel and , although less than a unit , it is larger than the one achieved deterministically [ see eq . ( [ eq : f_opt ] ) ] . in the following we shall , first , briefly review mc discrimination for symmetric states @xcite , and then apply this strategy to the teleportation process . the advantages as well as the drawbacks of the protocol are discussed and an example is presented to illustrate it . it is worth mentioning that the problem addressed here could equivalently be thought as the teleportation of @xmath1-dimensional qudit states using two entangled @xmath15-dimensional qudits . in fact , this problem has been discussed in ref . @xcite from a completely different approach and restricted to maximally entangled states only . in our protocol , however , this is just a special case that occurs when all the schmidt coefficients in ( [ quantumchannel ] ) are equal as will be shown later . in the problem of quantum state discrimination , the mc strategy is implemented through an optimized measurement in which each conclusive outcome leads us to identify each state with the maximum possible confidence , while keeping at the minimum the probability of obtaining inconclusive results @xcite . in fact , the mc strategy interpolates between me and ud strategies : when the confidence is the same for all states and there is no inconclusive result , mc and me coincide ; when the confidence is equal to unity for each conclusive result mc corresponds to ud strategy . recently , we have addressed the problem of discriminating with maximum confidence among @xmath1 symmetric states of @xmath15-dimensional qudits ( @xmath38 ) , prepared with equal prior probabilities @xmath40 . for this linearly dependent set , defined in eq . ( [ symmetricstates ] ) for @xmath38 , we found the optimal povm and determined its physical implementation @xcite . the latter is achieved by the following procedure : first , a two - dimensional ancilla , initially prepared in the logical state @xmath67 , is coupled with the qudit ( for convenience labeled as system 2 ) through a unitary operation @xmath68 acting on the tensor product hilbert space @xmath69 . the optimal unitary , derived in ref . @xcite , transforms the input states @xmath30 as follows : @xmath70 where @xmath71 ( the role of the subscript will become clear later ) is the minimum probability of obtaining an inconclusive result , given by @xcite @xmath72 where @xmath73 is the minimum coefficient for the states @xmath30 in eq . ( [ symmetricstates ] ) . the normalized states @xmath74 and @xmath75 can be written as @xmath76 and @xmath77 respectively , with @xmath25 defined in eq . ( [ eq : z ] ) . note that those states are still symmetric and equally likely . each @xmath74 ( @xmath75 ) occurs with probability @xmath78 ( @xmath71 ) . after the unitary coupling , the ancilla is measured on the computational basis , namely @xmath79 . if it is projected on @xmath67 , the initial states are mapped into @xmath74 . this outcome is interpreted as conclusive ( or successful ) in the sense that the transformed states can now be subjected to a measurement that will discriminate them and hence discriminate @xmath30with the maximum achievable confidence . otherwise , if the ancilla is projected on @xmath80 , the initial states are mapped into @xmath75 , and this result is inconclusive ( or a failure ) in the sense that there is no measurement that discriminates them with the maximum achievable confidence . when the result is conclusive , we have shown in ref . @xcite that the mc and me strategies coincide , and so the optimized measurement for both is the same . as we discussed in sec . [ subsec : me_telep ] ( see also footnote [ fn : opt_me ] ) , this measurement is performed by , first , applying an inverse fourier transform [ see eq . ( [ eq : fourier ] ) ] acting on a @xmath1-dimensional hilbert space , and then projecting on the computational basis that spans this space . doing so , our confidence in taking an outcome @xmath61 to identify the input state as @xmath30 will be '' ( for @xmath81 `` s '' or `` f '' and @xmath82 a positive integer ) indicate that the quantity ( or operator ) before them is related to a conclusive ( @xmath83 ) or inconclusive ( @xmath84 ) event at the @xmath82th stage of a possible sequential mc measurement , as will be discussed in sec . [ sec : results2 ] . [ fn : notation ] ] @xmath85_{{\rm s},\mathbbm{1}}^{\rm mc}=\frac{n}{d},\ ] ] which is the maximum achievable for equally likely symmetric pure states @xcite . we show now that by applying the mc strategy described above , alice and bob can perform optimal imperfect conclusive teleportation . following the protocol from the beginning , alice first applies the @xmath20 between systems 2 and 3 in her possession . she then adds a two - dimensional ancilla prepared in the logical state @xmath67 and applies the unitary transformation @xmath68 [ defined in eq . ( [ mc ] ) ] between this ancilla and system 2 . using eqs . ( [ base ] ) and ( [ mc ] ) it can be shown that the four - system state can be written as @xmath86 & & \text{}+\sqrt{p^?_\mathbbm{1}}|\phi^{\rm fail}\rangle_{123}|1\rangle_{\rm a } , \label{eq : mc_transform}\end{aligned}\ ] ] where @xmath71 is given by eq . ( [ eq : min_p ? ] ) and @xmath87 ( @xmath88 ) denotes the normalized three - system state after a conclusive ( an inconclusive ) outcome in the measurement of the ancilla . they are given , respectively , by @xmath89 where @xmath74 is defined by eq . ( [ mcu ] ) , and @xmath90 with @xmath75 defined by eq . ( [ mcxi ] ) . now , alice measures the ancilla on the computational basis and communicates her outcome to bob by sending him one bit of information through a classical channel . the process so far is depicted in the quantum circuit of fig . [ fig : q_circuit2](a ) . each one of the two possible alice s outcomes leads to one of the three - system states above , and each of them allows her to complete the teleportation process with different fidelities as we shall discuss next . in the circuit of fig . [ fig : q_circuit2](a ) this is represented by the box @xmath91 ( see footnote [ fn : notation ] ) . \(a ) + and imperfect when @xmath38 . ( a ) here @xmath92 , where @xmath93 is the identity on the hilbert space of system 3 and the action of @xmath94 is defined by eq . ( [ eq : mc_transform ] ) . the box @xmath91 represents the quantum circuit that completes the protocol after the outcome @xmath95 in the measurement of the ancilla by alice is communicated to bob . @xmath96 is the final state of bob s system which depends on operation @xmath91 . ( b ) in case of conclusive events , i.e. , @xmath97 the teleportation is accomplished as in the standard scheme ( see fig . [ fig : q_circuit1 ] ) . the final state of system 1 will be either a faithful replica of the unknown state @xmath50 ( perfect protocol ) or an unfaithful one with the optimal fidelity given by eq . ( [ strategy - conclusive ] ) ( imperfect protocol ) . [ fig : q_circuit2 ] ] .5 cm ( b ) + and imperfect when @xmath38 . ( a ) here @xmath92 , where @xmath93 is the identity on the hilbert space of system 3 and the action of @xmath94 is defined by eq . ( [ eq : mc_transform ] ) . the box @xmath91 represents the quantum circuit that completes the protocol after the outcome @xmath95 in the measurement of the ancilla by alice is communicated to bob . @xmath96 is the final state of bob s system which depends on operation @xmath91 . ( b ) in case of conclusive events , i.e. , @xmath97 the teleportation is accomplished as in the standard scheme ( see fig . [ fig : q_circuit1 ] ) . the final state of system 1 will be either a faithful replica of the unknown state @xmath50 ( perfect protocol ) or an unfaithful one with the optimal fidelity given by eq . ( [ strategy - conclusive ] ) ( imperfect protocol ) . [ fig : q_circuit2 ] ] .5 cm a conclusive outcome in the measurement of the ancilla makes the transformation @xmath98 [ see eq . ( [ eq : mc_transform ] ) ] or , equivalently , @xmath99 [ see eqs . ( [ symmetricstates ] ) and ( [ mcu ] ) , respectively ] . from eq . ( [ eq : min_p ? ] ) , such event occurs with the maximum success probability @xmath100 . alice now proceeds with the teleportation process by measuring systems 2 and 3 . as usual , system 3 is subjected to a projective measurement on the computational basis . on the other hand , system 2 is subjected to the final measurement that will identify each state @xmath74and accordingly each @xmath30with maximum confidence . as described above and shown in ref . @xcite , this final measurement is exactly the same as the optimized measurement that discriminates among equally likely symmetric states with the minimum probability of error ( see footnote [ fn : opt_me ] ) . therefore , from the discussion in sec . [ subsec : me_telep ] , this implies that alice performs the same measurement of the standard teleportation protocol , and , doing this , it can be ensured by refs . @xcite that she will teleport the unknown state to bob with the _ maximum _ achievable fidelity . this fidelity , denoted by @xmath101 ( see footnote [ fn : notation ] ) , will be given by eq . ( [ eq : f_opt ] ) taking into account that @xmath102 for all @xmath44 . thus , @xmath103 the complete quantum circuit implementing this protocol is obtained by inserting the circuit of fig . [ fig : q_circuit2](b ) into the box @xmath91 in fig . [ fig : q_circuit2](a ) with @xmath104 . the teleportation fidelity in ( [ strategy - conclusive ] ) depends only on the schmidt rank , @xmath15 , of the quantum channel unlike the deterministic protocol whose fidelity depends on both the schmidt rank and schmidt coefficients of the channel [ see eq . ( [ eq : f_opt ] ) ] . consequently , for a given @xmath15 , the fidelity achieved by teleporting via mc measurements will be the same for any entangled state shared between alice and bob . the success probability @xmath105 , on the other hand , depends on the particular quantum channel . in order to compare the fidelities for each protocol , we first find the singlet fraction for the present case . comparing eqs . ( [ eq : horodecki ] ) and ( [ strategy - conclusive ] ) and using eq . ( [ eq : mc_probability ] ) we have @xmath106_{{\rm s},\mathbbm{1}}^{\rm mc}.\ ] ] since @xmath63_{{\rm s},\mathbbm{1}}^{\rm mc}\geqslant [ p(\nu_l|l)]^{\rm me}$ ] @xcite , and so @xmath107 we finally obtain @xmath108 therefore , if alice applies the mc strategy instead of me to discriminate the states @xmath30 given by eq . ( [ symmetricstates ] ) , after a conclusive result she will always teleport with a fidelity larger than the optimal one achieved , deterministically , by standard teleportation . the equality in ( [ eq : fmc - vs - fme ] ) holds only when the entangled state in eq . ( [ quantumchannel])and hence @xmath30 in ( [ symmetricstates])has @xmath109 for all @xmath44 , since in this case mc and me strategies coincide , as discussed above . ( the fidelity for this particular channel has been derived in ref . @xcite using a different approach . ) the price to pay to teleport with better fidelity is that the protocol becomes probabilistic and the required resources increase in the same way as in the perfect conclusive protocol described in sec . [ subsec : ud_telep ] . two extreme cases for the conclusive teleportation fidelity can be seen from eq . ( [ strategy - conclusive ] ) . first , it reaches its minimum value when @xmath110 , that is , when the quantum channel in eq . ( [ quantumchannel ] ) is separable . this value is the maximal fidelity of teleportation via classical channel @xcite @xmath111 consequently , as long as @xmath112 , the conclusive teleportation fidelity @xmath101 will always exceed @xmath113 by an amount of @xmath114 . in the second case , the fidelity reaches its maximum value of 1 when @xmath37 , that is , when the quantum channel in eq . ( [ quantumchannel ] ) has maximal schmidt rank , so the imperfect conclusive protocol becomes perfect . in fact , regarding the success probability and the conclusive teleportation fidelity , the optimal imperfect conclusive protocol presented here interpolates between the optimal deterministic unfaithful ( @xmath115 ) and the perfect conclusive ( @xmath37 ) protocols described in sec . [ subsec : me_telep ] and sec . [ subsec : ud_telep ] , respectively . this is so because , as we mentioned earlier , mc strategy interpolates between me and ud strategies when the above conditions are accomplished . an inconclusive outcome in the measurement of the ancilla makes the transformation @xmath116 [ see eq . ( [ eq : mc_transform ] ) ] or , equivalently , @xmath117 [ see eqs . ( [ symmetricstates ] ) and ( [ mcxi ] ) , respectively ] . such event occurs with the minimum failure probability given by eq . ( [ eq : min_p ? ] ) . since the @xmath1 states @xmath75 form a new set of equally likely symmetric states restricted to a subspace of an @xmath15-dimensional space , they could be subjected to a new round of discrimination . if so , the teleportation process could be completed , although the achieved fidelity would be even lower than the conclusive one , no matter the strategy adopted between alice and bob . let us assume here that in consequence of a prearrangement with bob , alice applies the me strategy to discriminate the states @xmath75 , after an inconclusive result in the mc measurement . from our discussion in sec . [ subsec : me_telep ] and using eq . ( [ mcxi ] ) , the fidelity of the teleported state after the protocol is completed will be given by eq . ( [ eq : f_opt ] ) with @xmath118 for all @xmath44 . denoting this fidelity by @xmath119 ( see footnote [ fn : notation ] ) we obtain @xmath120 ^ 2}},\nonumber\\ & & \label{eq : f_me_?}\end{aligned}\ ] ] with @xmath113 given by eq . ( [ eq : f_clas ] ) . the above fidelity is lower than the optimal @xmath121 given by eq . ( [ eq : f_opt ] ) , and accordingly , lower than @xmath101 in eq . ( [ strategy - conclusive ] ) , as expected . it will be , however , larger than the fidelity of teleportation via classical channel if there exist at least two schmidt coefficients of the entangled state in ( [ quantumchannel ] ) which differ from the minimum . instead of the me strategy , alice could also have applied another round of mc measurements to discriminate the states @xmath75 , after an inconclusive event . this possibility and their consequences will be addressed in sec . [ sec : results2 ] . the overall teleportation fidelity is computed by averaging the fidelities of conclusive and inconclusive events with their respective probabilities . denoting this quantity by @xmath122 , we have for the present protocol @xmath123 where @xmath124 represents the average fidelity when the measurement strategy ( or a sequence of strategies ) @xmath53 is applied by alice after obtaining an inconclusive result in the first stage of the mc measurement . for instance , if alice applies the me strategy as described above we have @xmath125 . thus , using eqs . ( [ eq : min_p ? ] ) , ( [ strategy - conclusive ] ) , and ( [ eq : f_me _ ? ] ) , the overall teleportation fidelity for this particular case will be given by @xmath126 & = & f^{\rm clas}+na_{n-1}^2\left(\frac{n-1}{d+1}\right)\nonumber\\[2 mm ] & & \text{}+\sum_{\stackrel{\scriptstyle m , m'=0}{m\neq m'}}^{n-1 } \sqrt{\frac{(a^2_m - a^2_{n-1})(a^2_{m'}-a^2_{n-1})}{(d+1)^2}}. \label{f2}\end{aligned}\ ] ] this fidelity generalizes the one obtained previously in refs . @xcite for perfect conclusive teleportation , i.e. , for @xmath37 . comparing it with the optimal teleportation fidelity achieved by a deterministic protocol [ see eq . ( [ eq : f_opt ] ) ] , one can show that @xmath127 where , again , the equality holds when the schmidt coefficients of the channel satisfy @xmath109 for all @xmath44 . more generally , using the fact that the largest average fidelity of transformation of a pure bipartite state into another using locc is achieved deterministically rather than probabilistically @xcite , we can write into a final pure state @xmath128 gives the maximal average fidelity with respect to a given target state @xmath56 ( see lemma 2 of ref . when the target state is a maximally entangled one , this means that the _ average _ singlet fraction [ the average of eq . ( [ eq : singlet - fraction ] ) ] achieves its maximum deterministically . therefore , by averaging eq . ( [ eq : horodecki ] ) for deterministic and probabilistic protocols we arrive at eq . ( [ eq : average_fidel_general ] ) . [ fn : overall_fid ] ] @xmath129 for any @xmath53 . therefore , the reduction of the overall teleportation fidelity in comparison with the one achieved in a deterministic protocol is another consequence of teleporting with a fidelity larger than the optimal one given by eq . ( [ eq : f_opt ] ) . to illustrate graphically what has been discussed so far , we consider the teleportation of a four - dimensional qudit state ( ququart ) through an arbitrary quantum channel with schmidt rank @xmath130 . in the following , we plot the fidelities as a function of the schmidt coefficients ( @xmath131 and @xmath132 ) of the entangled state in eq . ( [ quantumchannel ] ) without assuming any ordering between them . from eq . ( [ strategy - conclusive ] ) , the conclusive teleportation fidelity will be @xmath133 for any initial entangled state . this is shown in fig . [ fig : fidel_n3_d4](a ) ( upper and lower panels ) . the optimal fidelity achieved deterministically , @xmath121 , and given by eq . ( [ eq : f_opt ] ) is shown in fig . [ fig : fidel_n3_d4](b ) ( upper panel ) . one clearly observes the relation between those fidelities as established by eq . ( [ eq : fmc - vs - fme ] ) . the floor of this graphic corresponds to the maximal fidelity of teleportation via classical channel , which from eq . ( [ eq : f_clas ] ) will be @xmath134 . . in the upper panel we have : ( a ) the conclusive fidelity @xmath101 given by eq . ( [ strategy - conclusive ] ) , ( b ) the optimal fidelity @xmath121 of a deterministic protocol given by eq . ( [ eq : f_opt ] ) , and ( c ) the conclusive fidelity @xmath135 given by eq . ( [ f_2 ] ) . the floor of the graphic corresponds to the teleportation fidelity via classical channel , @xmath113 , given by eq . ( [ eq : f_clas ] ) . the lower panels show the graphics ( a ) and ( c ) viewed from the top . ] . in the upper panel we have : ( a ) the conclusive fidelity @xmath101 given by eq . ( [ strategy - conclusive ] ) , ( b ) the optimal fidelity @xmath121 of a deterministic protocol given by eq . ( [ eq : f_opt ] ) , and ( c ) the conclusive fidelity @xmath135 given by eq . ( [ f_2 ] ) . the floor of the graphic corresponds to the teleportation fidelity via classical channel , @xmath113 , given by eq . ( [ eq : f_clas ] ) . the lower panels show the graphics ( a ) and ( c ) viewed from the top . ] the overall teleportation fidelities are shown in fig . [ fig : overall_fids ] . in fig . [ fig : overall_fids](a ) we plot the fidelity achieved deterministically , which is given by eq . ( [ eq : f_opt ] ) . figure [ fig : overall_fids](b ) shows the overall fidelity given by eq . ( [ f2 ] ) , which is obtained when alice applies me discrimination to complete the teleportation , after an inconclusive result in the first stage of the mc measurement . the relation between these two quantities established in eq . ( [ eq : overall_me ] ) is clearly observed by comparing figs . [ fig : overall_fids](a ) and [ fig : overall_fids](b ) . following the discussion initiated in sec . [ subsub : telep_inc ] , we will consider that after an inconclusive result in the first stage of the mc measurement , alice in agreement with bob keep trying to accomplish the teleportation by discriminating the `` failure '' states @xmath75 given by eq . ( [ mcxi ] ) . in that section we described the case where she applies the me strategy . here , we will describe the case where she applies another stage(s ) of mc measurements . in this regard , we say that the teleportation is performed via sequential maximum - confidence ( smc ) measurements @xcite . next , we briefly describe this measurement strategy and after that we discuss its application on teleportation . the concept of smc measurement has been introduced in the context of discrimination among linearly dependent and equally likely symmetric pure states @xcite . as we saw in sec . [ subsec : mc - review ] , when this set of states is subjected to a mc measurement , an inconclusive outcome maps the @xmath1 input states @xmath30 in eq . ( [ symmetricstates ] ) into a new set of @xmath1 linearly dependent and equally likely symmetric states @xmath75 in eq . ( [ mcxi ] ) , with a one - to - one correspondence between them . for convenience we rewrite the states @xmath75 as @xmath136 denotes the multiplicity of the smallest coefficient of @xmath30 , namely @xmath73 . therefore , the @xmath1 states @xmath75 are restricted to a @xmath137-dimensional hilbert space . since @xmath138 , we have three possible situations : ( i ) if @xmath139 , mc and me strategies coincide and there is no inconclusive result , so @xmath140 for all @xmath61 . ( ii ) if @xmath141 , the @xmath1 states @xmath75 will be identical , up to a global phase , so no further measurement will allow us to gain information about the input states . ( iii ) if @xmath142 , the mc measurement can be applied again and a conclusive outcome leads us to identify each input state with the confidence @xmath143_{{\rm s},\mathbbm{2}}^{\rm mc}=\frac{n-\mu_1}{d},\ ] ] which is smaller than the one achieved in the first stage [ see eq . ( [ eq : mc_probability ] ) ] , as it should be . on the other hand , following eq . ( [ eq : min_p ? ] ) , an inconclusive outcome occurs with the minimum probability @xmath144 and when it occurs the states @xmath75 are mapped to a new set of equiprobable symmetric states restricted to a ( @xmath145)-dimensional subspace , where @xmath146 accounts for the multiplicity of the second smallest input - state coefficient @xmath36 . thus , the whole analysis above applies again to this new set , and the process can be iterated in as many stages as allowed by the multiplicities @xmath147 of the coefficients @xmath36 . as we show in the appendix , if there exist @xmath148 $ ] sets of equal coefficients , the maximum number of conclusive stages , @xmath149 , will be @xmath150 where @xmath151 is the multiplicity of the largest @xmath36 in eq . ( [ symmetricstates ] ) . this process is referred to as smc measurement @xcite . back to teleportation , let us consider now that after an inconclusive result in the first stage of the mc measurement , alice applies this strategy again to discriminate the states @xmath75 given by eq . ( [ eq : new_xi ] ) . for this , she needs , first , to add another two - dimensional ancilla prepared in the logical state @xmath67 and to implement the unitary coupling @xmath152 between it and system 2 . then she measures the ancilla on the computational basis and communicate the outcome to bob by sending him a one - bit message through a classical channel . this process is depicted in the quantum circuit of fig . [ fig : q_circuit3 ] . a conclusive outcome occurs with maximum probability @xmath153 [ see eq . ( [ eq : p?2 ] ) ] and allows alice and bob to complete the teleportation process as in the standard case [ e.g. , see the quantum circuit of fig . [ fig : q_circuit2](b ) ] . from the discussion of sec . [ subsec : conc_telep ] and using eq . ( [ strategy - conclusive ] ) , the conclusive teleportation fidelity in the second stage of the smc measurement will be @xmath154 which is smaller than the one achieved in the first stage , as expected . however , by the same reasoning used to demonstrate eq . ( [ eq : fmc - vs - fme ] ) , we have that @xmath155 , where @xmath156 is given by eq . ( [ eq : f_me _ ? ] ) and the equality holds when the coefficients of @xmath75 in eq . ( [ eq : new_xi ] ) satisfy @xmath157 for all @xmath0 . thus , it is possible to achieve , with certain probability , a better teleportation fidelity by applying a mc strategy rather than me after an inconclusive outcome in the first stage of the mc measurement . in fig . [ fig : q_circuit2](a ) with @xmath158 . @xmath88 is given by ( [ s2 ] ) and @xmath159 . the box @xmath160 represents the quantum circuit that completes the teleportation after the outcome @xmath95 in the measurement of the ancilla by alice is communicated to bob . if @xmath97 , @xmath161 is given by the circuit of fig . [ fig : q_circuit2](b ) and in the end the teleported state has a fidelity given by eq . ( [ f_2 ] ) . if @xmath162 , @xmath163 will depend on the strategy between alice and bob : a new stage of mc measurements , an inconclusive event , etc . , and @xmath164 is the final state of bob s system which depends on @xmath160 . [ fig : q_circuit3 ] ] let us note that according to the discussion below eq . ( [ eq : new_xi ] ) it can be shown that in order to have at least two stages of mc measurements in the teleportation protocol , the following conditions must hold simultaneously : ( i ) the schmidt rank of the quantum channel in eq . ( [ quantumchannel ] ) must be @xmath165 and ( ii ) the multiplicity of the smallest schmidt coefficient of this channel ( @xmath73 ) must satisfy @xmath142 . as we are considering teleportation of a qudit state with @xmath166 , the condition ( i ) says that the simplest case where smc measurements can be applied to assist the protocol is when @xmath167 and @xmath130 , if condition ( ii ) holds . this case has been addressed in sec . [ subsec : examplei ] and we will return to it later . the @xmath15 schmidt coefficients of the entangled state in eq . ( [ quantumchannel ] ) can be grouped into @xmath148 $ ] sets of equal coefficients . this means that alice can apply , at most , @xmath149 stages of mc measurements , with @xmath149 given by eq . ( [ eq : max - stages ] ) ( see also the appendix ) . at each stage , the procedure is the same described above , and if she gets a conclusive outcome in the @xmath82th one , the teleportation fidelity , after completing the protocol , will be [ eq : fid_sub_eq ] @xmath168 & = & \frac{1}{d+1}\left(1+\sum_{j=\mathbbm{k}}^{d}\mu_j\right ) , \label{eq : fid_con_b}\end{aligned}\ ] ] where @xmath147 denotes the multiplicity of the @xmath95th smallest schmidt coefficient . equation ( [ eq : fid_con_a ] ) shows that the conclusive teleportation fidelity decreases from one stage to the next , which is a consequence of the decrease in the confidence with which alice discriminates the states of system 2 from one stage to the next @xcite . to arrive at ( [ eq : fid_con_b ] ) we use that @xmath169 . comparing this equation with ( [ eq : horodecki ] ) we find the singlet fraction @xmath170_{{\rm s},\mathbbm{k}}^{\rm mc},\ ] ] where the latter equality follows from @xmath63_{{\rm s},\mathbbm{k}}^{\rm mc}=[p(\nu_l|l)]_{{\rm s},\mathbbm{k}-1}^{\rm mc}-\mu_{\mathbbm{k}-1}/d$ ] @xcite . finally , if alice gets a conclusive outcome in the last allowed stage , @xmath149 , the teleportation fidelity will be @xmath171 which is always larger than the maximal one achieved via classical channel and given by eq . ( [ eq : f_clas ] ) . to account for the required resources to accomplish teleportation via smc measurements , let us consider that alice and bob agree beforehand to implement at most @xmath82 among the @xmath149 allowed stages of such strategy ( @xmath172 ) . in the worst case of successful teleportation , alice obtains a conclusive outcome at the @xmath82th stage and the process will consume , in addition to the entangled state and @xmath2 bits of classical communication , another @xmath82 classical bits as well as @xmath82 two - dimensional ancillas . on the other hand , if alice obtains an inconclusive outcome at the @xmath82th stage , all the used resources are lost , i.e. , @xmath82 classical bits and ancillas , the entangled state and the unknown qudit state . in this case , she and bob discard the attempt and start the process again with new copies of the aforementioned resources . therefore , this protocol is always more expensive than a deterministic one . however , as we shall see , in some situations the application of more than one stage of mc measurements enable alice and bob to save resources for teleportation . the probability of teleportation at the @xmath82th stage of a smc measurement , @xmath173 , is given by @xmath174 which takes into account the inconclusive probabilities at all @xmath175 preceding stages . considering again that alice and bob agree beforehand to implement at most @xmath82 among the @xmath149 allowed stages of a smc measurement ( @xmath172 ) , we can compute the overall probability of teleportation , @xmath176 , just by adding the success probability at each stage . thus , using eq . ( [ eq : prob_telep_k ] ) , it is easy to show that @xmath177 the overall teleportation fidelity , assuming that alice implements a given strategy or a sequence of strategies @xmath53 after obtaining an inconclusive outcome at the @xmath82th stage , will be @xmath178 which , for @xmath179 becomes @xmath180 with @xmath113 given by eq . ( [ eq : f_clas ] ) . note that the second term on the right - hand side of eq . ( [ eq : fid_overall_n - stage ] ) contributes only if the multiplicity , @xmath151 , of the largest schmidt coefficient is 1 . according to the discussion in sec . [ subsub : overall_fid ] and using eqs . ( [ eq : overall_me ] ) and ( [ eq : average_fidel_general ] ) ( see also footnote [ fn : overall_fid ] ) we can order the overall teleportation fidelities of the protocols described so far as follows @xmath181 therefore , the effect of adding further stages of mc measurements in the teleportation process is a reduction of the overall fidelity . as discussed so far , in the problem of teleportation via quantum channels with nonmaximal schmidt rank , alice could , in principle , apply an @xmath149-stage smc measurement to assist the protocol . in this case , a conclusive event at any stage @xmath172 leads to a teleportation fidelity @xmath183 given by eq . ( [ eq : fid_sub_eq ] ) . on the other hand , alice in agreement with bob could have chosen to apply a me measurement and teleport deterministically with an optimal fidelity @xmath121 given by eq . ( [ eq : f_opt ] ) . however , let us assume that for alice and bob purposes , it is sufficient to accomplish teleportation with a fidelity of transmission better than a threshold set by @xmath121 . if their shared quantum channel enables alice to implement an @xmath149-stage smc measurement , each stage @xmath82 whose conclusive fidelity fulfills this condition will be useful . this is always the case for the first stage ( unless @xmath109 for all @xmath44 ) , as shown in eq . ( [ eq : fmc - vs - fme ] ) . for the remaining stages the condition @xmath184 , obtained from eqs . ( [ eq : f_opt ] ) and ( [ eq : fid_sub_eq ] ) , reads ) and ( [ eq : sing_frac_smc ] ) to write the inequality ( [ eq : ineq ] ) as @xmath63_{{\rm s},\mathbbm{k}}^{\rm mc}>[p(\nu_l|l)]^{\rm me}.$ ] this expression says that the @xmath82th stage of a smc measurement will be useful for teleporting with a fidelity above @xmath121 , whenever a conclusive outcome in that stage makes alice s confidence in identifying the states @xmath30 [ see eq . ( [ symmetricstates ] ) ] larger than the one achieved via the me strategy . ] @xmath185 given an arbitrary entangled state as in eq . ( [ quantumchannel ] ) , it is not possible to obtain a general expression for the maximum number of useful stages among the @xmath149 allowed ones [ see eq . ( [ eq : max - stages ] ) ] in a smc measurement . however , since alice and bob have complete knowledge of this state , all the parameters in the above inequality are known beforehand . thus , before starting the teleportation , they have to verify whether the @xmath82th stage ( @xmath186 ) will be useful and then establish the number of stages to be implemented in the protocol . assuming that a @xmath82-stage smc measurement enables alice and bob to accomplish teleportation with a fidelity above @xmath121 , the overall probability of success in this case will be given by eq . ( [ eq : probab_overall ] ) . from this equation , we clearly have @xmath187 where the equality holds for @xmath188 . therefore , the implementation of further @xmath175 stages , in addition to the first one , increases the chance of teleportation . this is specially important in a scenario where alice and bob have limited resources , mainly those considered to be more expensive , as for instance , the entangled states . if so , instead of simply discarding the teleportation attempt after an inconclusive outcome in the first stage , they can keep trying to teleport in one of the remaining @xmath175 stages . this increases the overall probability of teleportation with a fidelity above the threshold @xmath121 and , consequently , allows them to save resources . we return now to the example of teleportation of an unknown ququart state through an arbitrary quantum channel with schmidt rank @xmath130 . as mentioned earlier , this is the simplest case where teleportation via smc measurements may be applied . here , we have at most two stages , the first of which has been discussed in sec . [ subsec : examplei ] . if an inconclusive outcome in that stage is followed by a conclusive one in the second , the teleportation fidelity , given by eq . ( [ f_2 ] ) , will be @xmath189 . this is shown in fig . [ fig : fidel_n3_d4](c ) ( upper and lower panels ) . in those graphics , the lines where the fidelity is equal to the classical one represents the entangled states whose smallest schmidt coefficient has multiplicity @xmath190 . the point where the three lines touch each other represents an event that never occurs because @xmath191 , and , hence , there is no inconclusive outcome at the first stage . for the points on the surface [ fig : fidel_n3_d4](b ) that are below the plateau in fig . [ fig : fidel_n3_d4](c ) , the corresponding entangled states enable conclusive fidelities above @xmath121 at the second stage of the smc measurement . figure [ fig : overall_fids](c ) shows the overall teleportation fidelity , @xmath192 , obtained from eq . ( [ eq : fid_overall_n - stage ] ) . comparing this graphic with those from figs . [ fig : overall_fids](a ) and [ fig : overall_fids](b ) , one can verify the relation established in eq . ( [ eq : all_overall ] ) . in fig . [ fig : prob_n3](a ) we plot the probability of teleportation at the first stage of the smc measurement , obtained from eq . ( [ eq : min_p ? ] ) . with those probabilities , we have a teleportation fidelity above @xmath121 for any entangled state [ see eq . ( [ eq : fmc - vs - fme ] ) ] . figure [ fig : prob_n3](b ) shows the _ overall _ probability of teleportation [ see eq . ( [ eq : probab_overall ] ) ] with the condition that the conclusive fidelity of transmission be larger than @xmath121 . the region in this graphic where the probabilities exceeds those ones in fig . [ fig : prob_n3](a ) represents the entangled states where , in addition to the first , the second stage of the smc measurement also satisfy the condition . for these states the implementation of a @xmath193-stage smc measurement is clearly advantageous . . ( a ) probability at the first stage of the smc measurement given by @xmath78 [ see eq . ( [ eq : min_p ? ] ) ] . ( b ) overall probability of teleportation [ see eq . ( [ eq : probab_overall ] ) ] with a fidelity above @xmath121 . ] to summarize , in this work we have described the quantum teleportation protocol via nonmaximally entangled pure states within the framework of quantum state discrimination . more specifically , we showed that a crucial step for teleportation is related to the ability of the sending party , alice , to discriminate nonorthogonal symmetric states which are equally likely . the type of teleportation that she and bob ( the receiving party ) prearrange to implement defines the discrimination strategy that she has to adopt . in this regard , we have shown that to achieve optimal teleportation fidelity in a deterministic way @xcite , alice must implement an optimized me strategy , which is valid for both linearly dependent and linearly independent symmetric states . on the other hand , to achieve unit teleportation fidelity with maximum success probability , alice has to apply an optimized ud strategy . this is a known fact since this protocol , called perfect conclusive teleportation , has been conceived @xcite and intensively studied thereafter @xcite . however , ud applies only for linearly independent states @xcite , which implies that perfect conclusive teleportation is restricted to quantum channels with maximal schmidt rank . here , we investigated the problem where this latter condition does not hold . it was shown that if alice applies an optimized mc measurement @xcite to assist teleportation in this case , the fidelity of transmission at conclusive events , although less than unit , is the maximum achievable for that channel . in fact , as can be seen from eqs . ( [ strategy - conclusive ] ) and ( [ eq : fmc - vs - fme ] ) , we found that this fidelity depends only on the schmidt rank of the channel and it is better than the one achieved deterministically via me measurements . moreover , the probability of successfully teleporting is also maximal . this optimal imperfect conclusive teleportation was shown to interpolate between the optimal deterministic ( when the schmidt coefficients are all equal ) and the perfect conclusive ( when the schmidt rank is maximal ) protocols . subsequently , the teleportation via smc measurements , introduced in ref . @xcite , has also been investigated . in this strategy , one takes into account that it might be possible to extract some information about the input states even after an inconclusive result in the mc measurement . if so , mc can be applied again and the process can be iterated until no further information is available . we have shown that if the quantum channel allows for the maximum of @xmath149 conclusive stages in the smc measurement , the teleportation fidelity decreases from one stage to the next as can be seen in eq . ( [ eq : fid_con_a ] ) ; at the @xmath149th stage , it is still always larger than the maximal fidelity achieved via classical channel . an important feature of this scheme is that there are quantum channels for which it is possible to implement @xmath82 among the @xmath149 allowed stages in the smc measurement such that a conclusive event at any stage leads to a teleportation fidelity above the optimal one achieved deterministically . as a consequence , the overall probability of teleportation increases , allowing alice and bob to save the resources used in the protocol . it is worth noting that smc measurements could also be applied after a failed attempt of perfect conclusive teleportation , since in this case the states to be discriminated becomes linearly dependent and can not be unambiguously identified by any further process . in this work we have shown that the singlet fraction of a _ pure _ bipartite quantum channel , defined by eq . ( [ eq : singlet - fraction ] ) , is equal to the maximum confidence with which alice can discriminate equally likely symmetric pure states by applying a given optimized measurement strategy . this is a direct consequence of the connection between singlet fraction and entanglement concentration procedures @xcite , and the connection between the latter and state discrimination @xcite . it may be fruitful , somehow , to look at the singlet fraction by this new perspective and , for instance , to extend our study here for teleportation via mixed entangled states . finally , another possible extension of the results obtained in this work is to investigate other quantum communication protocols performed via quantum channels with nonmaximal schmidt rank . for instance , perfect conclusive entanglement swapping via ud strategy has been studied in @xcite , and in a forthcoming paper we shall generalize this for an imperfect protocol accomplished via mc measurements . the maximum number of conclusive stages allowed in the smc measurement , i.e. , the number of stages in which it is possible to gain some information about the input states through a measurement , ranges from 1 to @xmath194 . it depends on the multiplicities of the input - state coefficients @xmath195 as follows . the @xmath15 coefficients @xmath36 can be grouped into @xmath196 sets of equal coefficients , where @xmath148 $ ] . the number of elements in each set gives the multiplicity of the coefficient . if @xmath147 ( with @xmath197 ) is the multiplicity of the @xmath95th smallest coefficient , then @xmath151 will be the multiplicity of the largest one . as we saw in sec . [ subsec : smc ] , after each stage of the smc measurement , the dimension of the subspace where the states to be discriminated are restricted , decreases . if there are @xmath196 sets of equal input - state coefficients , the dimension of the hilbert space at each stage will be @xmath198 2^{\rm nd}\ \text{stage } & & \rightarrow & & n-\mu_1 , \\[2 mm ] 3^{\rm rd}\ \text{stage } & & \rightarrow & & n-(\mu_1+\mu_2 ) , \\[2 mm ] \vdots & & & & \hspace{8mm}\vdots \\[2 mm ] % ( d-1)^{\rm th}\ \text{stage } & & \rightarrow & & n-\displaystyle\sum_{j=1}^{d-2}\mu_j=\mu_{d-1}+\mu_d , \\[3 mm ] d^{\rm th}\ \text{stage } & & \rightarrow & & n-\displaystyle\sum_{j=1}^{d-1}\mu_j=\mu_d . \end{array}\ ] ] we note that if @xmath199 the failure states in the last stage will be identical , up to a global phase , so no further measurement allows us to gain information about the input states . otherwise , if @xmath200 the failure states can be discriminated for the last time . in this case , mc and me strategies coincide and there is no further inconclusive result . therefore , the maximum number of conclusive stages in the smc measurement , denoted by @xmath149 , will be @xmath201 \mathbbm{m}=d , & & \text{if } & & \mu_d>1 , \end{array}\ ] ] which , for short , can be written as @xmath202 . in the extreme cases mentioned above , this will be 1 when @xmath203 ( i.e. , all coefficients @xmath36 are equal ) and @xmath194 when @xmath204 ( i.e. , all coefficients @xmath36 differ ) .

the origin of the baryon asymmetry of the universe ( bau ) has been a great puzzle in particle physics and cosmology . from the latest observations of the cosmic microwave background radiations , the supernovae and the large scale structures , the baryon - to - photon ratio is given by @xcite @xmath6 the nonzero bau can be generated if the following three conditions are fulfilled @xcite : ( i ) baryon number ( @xmath7 ) violation , ( ii ) @xmath8 and @xmath0 violation , and ( iii ) departure from thermal equilibrium . electroweak baryogenesis is a scenario in which the bau is explained based only on electroweak physics @xcite . the standard model ( sm ) can in principle accommodate above three conditions . however , it turns out that the sm fails to generate the bau due to the lack of sufficient @xmath0 violation @xcite , and the electroweak phase transition ( ewpt ) is a smooth cross over for @xmath9 gev @xcite , which makes the condition ( iii ) infeasible . this failure motivates us to go beyond the sm . so far , many attempts of electroweak baryogenesis have been made in the various new physics models such as the minimal supersymmetric standard model ( mssm ) @xcite , the two - higgs - doublet model @xcite , the singlet - extended mssms @xcite , and recently the fourth - generation models also have been paid much attention @xcite . among them , the mssm is a theoretically well - motivated model and a good candidate for successful electroweak baryogenesis . as far as the strong first order ewpt which is necessary for condition ( iii ) is concerned , the light higgs boson is generically favored . on the other hand , the negative results of the higgs boson(s ) searches at lep put the constraints on the higgs boson masses in the mssm @xcite . the analysis has been done based on several benchmarks rather than scanning whole parameter space . recently , it is pointed out that the lep constraints can be relaxed in the more general case . in ref . @xcite , it is found that the lightest higgs boson mass can be smaller than the @xmath1 boson mass . such a light higgs boson can be viable since its coupling to @xmath1 boson is modified by the additional higgs bosons and suppressed enough to avoid the experimental observed limits . we call a light higgs boson scenario ( lhs ) when the mass of the lightest higgs boson is less than 114.4 gev . this scenario can accommodate the @xmath10 excess around 98 gev higgs boson mass in the lep experiments @xcite and have parameter space that is consistent with the observed dark matter relic density @xcite . in ref . @xcite , it is claimed that the lhs can provide a solution for a little hierarchy problem as well . as for the electroweak baryogenesis context , since the neutral higgs bosons are relatively light , they may give a new window to a viable baryogenesis scenario . in the electroweak baryogenesis mechanism , a sphaleron process must be decoupled after the ewpt in order to avoid the washout of the generated bau . because of this requirement , the ratio of the higgs vacuum expectation value ( vev ) to the temperature in question should be greater than some value . to search for the region where the strong first order ewpt is realized , and which is also experimentally allowed , not only the theoretical calculations of the zero temperature observables but also those at the finite temperature are mandatory . the accuracy of the former calculations are already beyond at least the tree level . on the other hand , the sphaleron decoupling condition , which is derived using the tree - level higgs potential at zero temperature , has been frequently used to discuss the strength of the strong first order ewpt in the literature . namely , @xmath11 is adopted as a practical criterion , where @xmath12 is the critical temperature at which the higgs potential has two degenerate minima and @xmath13 is the higgs vev at @xmath12 . to obtain more accurate theoretical predictions , the refined calculation of the sphaleron decoupling condition is indispensable . the uncertainties of the condition is expected to be reduced up to about 10% level if we make the following improvements , i.e. , the sphaleron solution at finite temperature is used and the contributions from the zero modes of the fluctuations around sphaleron at finite temperature are taken into account . in addition , the sphaleron decoupling condition must be imposed at the temperature @xmath14 lower than @xmath12 , at which the ewpt terminates . in practice , since it is difficult to determine this temperature , we substitute @xmath15 , a nucleation temperature of the critical bubble , in place of @xmath14 . the critical bubble is defined as the bubble whose surface energy and volume energy becomes balanced . only such bubbles can nucleate and expand in the symmetric phase . in this article , the possibility of the strong first order ewpt in the lhs and the conventional decoupling limit is examined with a particular emphasis on the refinement of the sphaleron decoupling condition . the analysis is conducted taking the latest experimental data into account as well . it is well known that a light stop whose mass is smaller than that of top quark is required for the strong first order ewpt . we search for the parameter space that is consistent with such a light stop . in the lhs , since the masses of all the higgs bosons are relatively light , typically , @xmath16 gev , the @xmath7 physics observables data can restrict the allowed region . here , we consider the constraints from the following processes : @xmath17 , @xmath18 and @xmath19 . the first two modes are relevant when the charged higgs bosons are light . the last one , which has not been observed yet , can give a significant limit when the neutral higgs boson is light . after finding the phenomenologically allowed region , we investigate the strength of the ewpt at not only @xmath12 but also @xmath15 . by using the one - loop effective potential at zero and finite temperatures , we calculate the sphaleron energy and the contributions from zero modes of the fluctuations around sphaleron to evaluate the sphaleron decoupling condition . it is then demonstrated whether or not such an improved sphaleron decoupling condition is satisfied at @xmath12 and @xmath15 . this article is organized as follows : in sec . [ sec : gh ] , we present the lagrangian of the gauge - higgs system for describing the statics and dynamics of the ewpt . the lhs is discussed in sec . [ sec : lhs ] . the qualitative evaluation of the ewpt is outlined in sec . [ sec : ewpt ] . the critical bubbles and the sphaleron decoupling condition are analyzed in secs . [ sec : cri_bub ] and [ sec : sph_dec ] , respectively . the experimental constraints are considered in sec . [ sec : ex_const ] . in sec . [ sec : numerical ] , we show the numerical results . section [ sec : con_dis ] is devoted to the conclusion and discussion . in the appendix [ app : mass_squared ] , we give the mass - squared matrices of the squarks to make our notation clear . to discuss the statics and dynamics of the ewpt , we consider the gauge - higgs system , which is governed by the lagrangian @xmath20 the covariant derivatives are @xmath21 where @xmath22 is the pauli matrices , @xmath23 and @xmath24 are the gauge couplings for @xmath25 and @xmath26 , respectively . @xmath27 is the effective potential , which is composed of the tree and one - loop parts @xmath28 the tree - level part is given by the f- , d- , and the soft supersymmetry ( susy ) breaking terms @xmath29 where @xmath30 the higgs doublets are parameterized as @xmath31 where we assume that @xmath32 is not broken at the vacuum . unlike the general two - higgs - doublet model @xcite , @xmath0 violation can not be accommodated in the higgs potential at the tree level , leading to @xmath33 . the one - loop part comprises the effective potentials at zero and nonzero temperatures @xmath34,\end{aligned}\ ] ] where @xmath35 where @xmath36 is the zero temperature effective potential , which is regularized in the @xmath37 scheme , @xmath38 denotes the degrees of freedom of particle species @xmath39 , @xmath40 is the background - field - dependent mass , and @xmath41 is a renormalization scale , which is determined in such a way that the zero temperature one - loop correction to the effective potential vanishes in the vacuum . in what follows , the contributions of the weak gauge bosons ( @xmath42 ) , the third - generation quarks ( @xmath43 ) , and squarks ( @xmath44 ) are taken into account . the statistical factors of each particle is , respectively , given by @xmath45 where @xmath46 is the number of color . in this article , we use the one - loop mass formulae of the higgs bosons presented in ref . @xcite . to make a discussion on the lhs clear , we rotate the two higgs doublets @xmath47 into @xmath48 by @xmath49 . for simplicity , we assume that @xmath0 is conserved for the moment . the rotated higgs doublets are cast into the form @xmath50 where @xmath51 gev , @xmath39 is the @xmath0-odd higgs boson , @xmath52 is the charged higgs boson and ( @xmath53 ) are the nambu - goldstone bosons . @xmath54 and @xmath55 are related to the physical @xmath0-even higgs bosons ( @xmath56 ) via @xmath57 where @xmath58 is a mixing angle between @xmath59 and @xmath60 . it should be noticed that only @xmath54 possesses the higgs vev in the rotated basis , and we can identify it the sm - like higgs boson , i.e. , @xmath61 . at the tree level , the higgs boson couplings to @xmath1 boson normalized to the corresponding sm values are , respectively , given by @xmath62 from eqs . ( [ rotated_phi ] ) , ( [ rotated_h ] ) , ( [ higgs_zz ] ) and ( [ higgs_za ] ) , one can see that @xmath63 , @xmath64 and @xmath65 for @xmath66 and @xmath67 , @xmath68 and @xmath69 for @xmath70 . the former case ( decoupling limit ) is realized by taking @xmath71 and the latter one ( antidecoupling limit ) is possible for @xmath72 , which corresponds to the lhs . depending on the theory parameters , some of the above four higgs boson couplings can be small enough to avoid the lep exclusion limits . in the lhs , for example , @xmath73 can be as light as about 100 gev for @xmath74 . however , from the sum rules among the higgs boson couplings , i.e. , @xmath75 , not all higgs bosons are allowed to be light . if the higgs boson is the sm - like , the mass of it should be larger than 114.4 gev since its coupling to the @xmath1 boson becomes the sm value . we will discuss the experimental constraints in more detail in sec . [ sec : ex_const ] . finally , we comment on the @xmath0-violating case . although @xmath0 is conserved in the tree - level higgs sector , it can be broken by the radiative corrections such as the top / stop loops . the realization of the maximal @xmath0 violation is called the cpx scenario @xcite . in this scenario , it is possible to accommodate an even lighter higgs boson . for instance , the mass of the lightest higgs boson can be as small as 40 gev , which is experimentally viable . it is interesting to discuss the phenomenological consequences of this scenario . however , the cpx scenario requires the relatively large ratios of @xmath76 assuming @xmath77 , which is not favored in the context of the strong first order ewpt that we will discuss in the next section . to explain how the first order ewpt is strengthened in the mssm , we here give a brief review on the light stop scenario . to simplify the discussion , we use the high temperature expansion in the effective potential @xcite . in the decoupling limit , @xmath73 is responsible for the electroweak symmetry breaking since @xmath78 . on the other hand , @xmath3 is mostly shared by @xmath79 in the lhs . if the ewpt is first order , the critical temperature @xmath80 is given by the temperature at which the effective potential has two degenerate minima . let @xmath13 be the nonzero vev at @xmath12 . from the argument using the high temperature expansion : @xmath81 in @xmath82 , one can see that @xmath83 can be enhanced when the quartic term in @xmath27 gets smaller and/or the coefficient of the cubic term with a negative coefficient ( denoted @xmath84 ) becomes larger . the former implies that the light @xmath85 is favored in the decoupling / lhs limit . the latter can be realized by the additional contributions from the bosons . in fact , the effect of the lighter stop @xmath86 on @xmath84 is sizable due to the large top yukawa coupling constant ( @xmath87 ) and the degrees of freedom , i.e. , @xmath88 . in the high temperature expansion , one can obtain @xmath89 where @xmath90 where @xmath91 and the @xmath92 contributions are neglected in the stop contribution , also the hierarchies between the soft susy breaking masses are implicitly assumed , namely , @xmath93 are taken to be consistent with the lep bounds on the higgs boson mass and the @xmath94 parameter . as a result of @xmath95 , which is necessary for the enhancement of the loop effect , the lighter stop mass becomes less than the top mass @xmath96 . from eq . ( [ dele_hte ] ) , one can easily see that the no - mixing case @xmath97 maximizes the loop effect . if we assume @xmath98 , we obtain @xmath99 , which is about 6 times as large as the sm contributions . to make the analysis on the ewpt precisely , the corrections from the daisy diagrams must be taken into account , which yield the temperature dependent terms in the field - dependent masses of particles . for the light stop , we effectively have the following mass shift to leading order : @xmath100 where @xmath101 is positive and composed of the relevant couplings in the theory . therefore , the ewpt would get weaker even if @xmath102 . nonetheless , as advocated in ref . @xcite the choice of @xmath103 , which can induce the charge - color - breaking ( ccb ) vacuum , may still realize the same enhancement of the stop loop effect as outlined above . for simplicity , we will not pursue this possibility and take @xmath104 putting the daisy diagrams aside in this article . our analysis on the strength of the ewpt is expected to be the same , or might be more optimistic than the more realistic case depending on the model parameters . the high temperature expansion makes it easy to see how the first order phase transition is strengthened analytically . it is , however , untrustworthy to use the approximation when the masses of the particles in loops are somewhat larger than @xmath12 . we thus adopt a different method . although the numerical integrations contained in the definitions of @xmath82 are the standard way , it is an extremely time - consuming task . therefore , we will use the following fitting functions in our numerical analysis : @xmath105 where the coefficients @xmath106 are determined by the least squared method . for @xmath107 , the errors of @xmath108 do not exceed @xmath109 for any @xmath110 , which is sufficient for our purpose . to determine @xmath12 and @xmath13 , we minimize @xmath27 with eq . ( [ i_fit ] ) in the three dimensional space @xmath111 numerically . the first order ewpt begins at somewhat below @xmath12 . for the ewpt to proceed , the radius of the bubble should be larger than some critical size , otherwise it would shrink by the surface tension of the bubble wall . the bubble of this critical size is called the critical bubble . after nucleation of the critical bubbles , they start to percolate and eventually convert the symmetric phase into the broken phase if the supercooling is not too large . from eq . ( [ gauge - higgs ] ) , the energy functional in the temporal gauge takes the form @xmath112.\end{aligned}\ ] ] here , we assume that the least energy has the pure - gauge configurations for @xmath113 and @xmath114 , hence @xmath115 . since a spherically symmetric configuration can give the least energy , the higgs fields depend only on radial coordinate @xmath116 . the classical higgs fields are parameterized as @xmath117 where the gauge - invariant constant phase @xmath94 will be taken to make the boundary value @xmath118 at @xmath119 to vanish . to remove the energy source for the @xmath1 boson from the higgs current , the so - called sourcelessness condition " @xcite @xmath120 must be satisfied . now let us introduce @xmath121 and @xmath122 . one can easily see that only @xmath123 is a gauge - independent phase . equation ( [ sourcelessness ] ) can be rewritten as @xmath124 which is used to eliminate @xmath125 . the energy functional is then reduced to @xmath126.\end{aligned}\ ] ] from this , the equations of motion ( eom ) for @xmath127 and @xmath123 are , respectively , given by @xmath128 the solutions can exist only when the temperature lies in the range @xmath129 , where @xmath130 is the temperature at which the higgs potential at the origin is destabilized in some direction . since we focus on the @xmath0-conserving bubble walls @xcite , @xmath131 is set to be zero in the following discussion . the studies of the @xmath0-violating bubble walls can be found in refs . the boundary conditions for eom are imposed in the symmetric ( @xmath119 ) and broken ( @xmath132 ) phases as @xmath133 it is convenient to parameterize the higgs profiles @xmath134 in terms of the dimensionless quantities @xmath135 then @xmath84 takes the form @xmath136,\end{aligned}\ ] ] where @xmath137 . correspondingly , the eom are rewritten as @xmath138 with the boundary conditions @xmath139 we will comment on the numerical method to solve the eom for @xmath140 and @xmath141 in sec . [ subsec : relaxation ] . the bubble nucleation rate per unit time per unit volume is given by @xcite @xmath142 where @xmath143 is the energy of the critical bubble at temperature @xmath4 . note that this is a rate per unit volume . we define the nucleation temperature @xmath15 as the temperature at which the rate of the nucleation of a critical bubble within a horizon volume is equal to the hubble parameter at that temperature . since the horizon scale is roughly given by @xmath144 , the nucleation temperature is defined by @xmath145 where @xmath146 is the massless degrees of freedom at @xmath15 and @xmath147 is the planck mass ( @xmath148 gev ) . since a single bubble nucleated within the horizon volume can not convert the whole region to the broken phase , the nucleation temperature defined by eq . ( [ def_tn ] ) simply gives an upper bound of the temperature at which the ewpt starts . to determine @xmath15 more precisely , simulations or the methods employed in ref . @xcite are needed . from eq . ( [ def_tn ] ) , it follows that @xmath149 roughly , @xmath150 is necessary for the development of the ewpt . to solve the eom ( [ eom_h1 ] ) and ( [ eom_h2 ] ) under the boundary conditions ( [ bc_h ] ) , we use relaxation methods . " to implement the relaxation methods successfully , the initial configurations of the bubbles are of great importance . here , we take the kink - type ansatz for them , i.e. , @xmath151 where @xmath152 is the radius of the bubbles and @xmath153 is the bubble wall width . since the best values of @xmath152 and @xmath153 can vary as the temperature goes down , it is impossible to know them in advance . nonetheless , it turns out that in most cases @xmath154 can give the successful initial configurations as long as @xmath4 lies somewhere between @xmath15 and @xmath12 . after obtaining the convergent solutions at some temperature , we use them as the initial configurations to search for the critical bubbles at the temperature which is near the previous one . by doing this process iteratively , we find @xmath15 . as shown by manton in ref . @xcite , the static and unstable classical solution with a finite energy can exist in the @xmath155 gauge theory . such a configuration is called sphaleron , which has a maximal energy along the least energy path . similar to the instanton process at zero temperature , the baryon number can be violated by the sphaleron process at finite temperature . although the instanton process , which is quantum tunneling , is highly suppressed and unobservable , the sphaleron process , which is thermal fluctuation , can be active at high temperature , especially before the electroweak symmetry is broken . in order to avoid the washout of the generated bau after the phase transition , the sphaleron process must be decoupled . this condition can be obtained by demanding that the baryon number changing rate be smaller than the hubble parameter @xcite @xmath156 where @xmath157 is number of generation , @xmath158 , @xmath159 is the negative mode of the fluctuations around sphaleron , @xmath160 is the @xmath161 coefficient @xcite , @xmath162 and @xmath163 are contributions from the translational and rotational zero modes , respectively , @xmath164 is the sphaleron energy . if we denote the sphaleron energy as @xmath165 , eq . ( [ sph_dec1 ] ) can be translated into @xmath166 . \label{sph_cond}\end{aligned}\ ] ] it should be noted that model dependent parameters other than the sphaleron energy in the right - hand side of eq . ( [ sph_cond ] ) contribute only logarithmically . in the sm @xcite , the sphaleron solution obtained with the tree - level higgs potential yields @xmath167 @xcite , @xmath168 @xcite , @xmath169 @xcite for @xmath170 . inserting these values into eq . ( [ sph_cond ] ) together with @xmath171 , @xmath172 , and @xmath173 gev , we find @xmath174 where @xmath175 @xcite is used . it is found that the contributions of the zero mode factors to @xmath2 can reach around 10% , while the contributions from the last three terms in the right - hand side of eq . ( [ sph_cond ] ) are only about 1% . now we discuss the sphaleron decoupling condition in the mssm . as an example , we take @xmath176 , @xmath177 gev , @xmath178 gev , @xmath179 gev , @xmath180 gev , @xmath181 gev . this parameter set is experimentally allowed as we will discuss in sec . [ sec : numerical ] . to see the effects of the temperature dependence of the sphaleron solution and the zero mode factors on the sphaleron decoupling condition , we consider the following three cases : * the sphaleron decoupling condition without the zero mode factors based on the zero temperature potential @xmath182 . * the sphaleron decoupling condition with the zero mode factors based on the zero temperature potential @xmath182 . * the sphaleron decoupling condition with the zero mode factors based on the full effective potential @xmath183 . the result of each case is summarized in table [ tab : sph_cond ] . @xmath184 is used in the right - hand side of eq . ( [ sph_cond ] ) . as we see from i and ii , the contribution of the zero mode factors to the sphaleron decoupling condition can be as large as 10% level , which is the same as in the sm case . in case iii , @xmath185 is obtained , which is about 40% stronger than the usual rough estimate used in the literature . we look into this case in more detail in the following . the temperature dependences of @xmath186 , @xmath162 and @xmath163 are plotted in fig . [ fig : sph_profile ] . the endpoints of those curves correspond to those at @xmath12 . the left panel shows that the sphaleron energy decreases as @xmath4 increases . the change of the sphaleron energy is sizable around @xmath12 , hence which can be the dominant error of the sphaleron decoupling condition if we neglect this effect . on the other hand , as shown in the right panel the temperature dependences of @xmath162 and @xmath163 are mild and thus have little effects on @xmath2 . from this analysis , we conclude that in order to evaluate the sphaleron decoupling condition within a 10% accuracy , the sphaleron solution and the zero mode factors at finite temperature must be taken into account . here , we comment on the earlier work @xcite . the authors have evaluated sphaleron at finite temperature using the one - loop finite - temperature effective potential with the daisy resummations . in their analysis , only the energy was calculated and the zero mode factors were missing . to search for the possible region where the ewpt is strong first order , they used the following sphaleron decoupling condition @xcite : @xmath187 in our calculation including the zero mode factors , this condition is translated into @xmath188 in case iii . .the sphaleron decoupling conditions in the three cases i , ii , and iii . [ cols="^,^,^,^",options="header " , ] [ tab : summary_dec ] we have reanalyzed the strong first order ewpt in the mssm considering the experimental results such as the lep and the latest @xmath7 physics data . @xmath2 was evaluated at both @xmath12 and @xmath15 in the lhs and ordinary decoupling limit . in the lhs with a light stop , the no - mixing scenario , which can maximize the stop thermal loop effect , can not be realized due to the severe constraint from @xmath18 . combining with the other experimental data , especially the lep and @xmath189 mode , the allowed region becomes more constrained . the sphaleron decoupling condition was also calculated using the one - loop effective potential at finite temperature . we found that the contributions of the zero mode factors coming from the fluctuations around sphaleron can be as large as about 10% . in the phenomenologically allowed region , the typical values of the sphaleron decoupling conditions at @xmath15 are found to be @xmath190 . it is observed that about 10% enhancement of @xmath2 is possible at @xmath15 in comparison with that at @xmath12 . we find that @xmath191 in the lhs and @xmath192 in the decoupling limit . however , such a strength of the ewpt is not enough to satisfy the sphaleron decoupling condition . although electroweak baryogenesis in the mssm seems to be infeasible , it may be possible to circumvent the above negative conclusion . some comments on the possible loopholes are in order . firstly , the sphaleron decoupling condition is supposed to be imposed at the temperature at which the ewpt ends , which can relax the decoupling condition . secondly , our calculations , which are based on the one - loop effective potential , can be modified by the higher - order loop contributions such as the two - loop effects @xcite . lastly , as we mentioned briefly in sec . [ sec : ewpt ] the scalar potential can be extended in such a way that the lighter stop also has the vev , which can develop the ccb vacuum . in ref . @xcite it is claimed that the ewpt is strong first order when the electroweak vacuum is metastable and the ccb vacuum is the global minimum . since the time scale of the decay of the metastable vacuum is longer than the age of the universe , this scenario is viable . the two - loop contributions are also included in their calculation . it is interesting to investigate whether or not the ewpt is still strong enough to satisfy the sphaleron decoupling conditions that we have considered in this article . a devoted study of the ewpt and the sphaleron decoupling condition based on the two - loop effective potential will be given elsewhere @xcite . here , we give the expressions of the mass - squared matrices of the stop and sbottom @xmath193 where @xmath87 and @xmath194 are the top and bottom yukawa couplings , respectively . c. amsler _ et al . _ [ particle data group ] , phys . b * 667 * , 1 ( 2008 ) . a. d. sakharov , pisma zh . fiz . * 5 * , 32 ( 1967 ) [ jetp lett . * 5 * , 24 ( 1967 sopua,34,392 - 393.1991 ufnaa,161,61 - 64.1991 ) ] . for reviews , see a. g. cohen , d. b. kaplan and a. e. nelson , ann . nucl . part . sci . * 43 * , 27 ( 1993 ) ; m. quiros , helv . acta * 67 * , 451 ( 1994 ) ; v. a. rubakov and m. e. shaposhnikov , usp . fiz . nauk * 166 * , 493 ( 1996 ) [ phys . 39 * , 461 ( 1996 ) ] ; k. funakubo , prog . * 96 * , 475 ( 1996 ) ; m. trodden , rev . phys . * 71 * , 1463 ( 1999 ) ; w. bernreuther , lect . notes phys . * 591 * , 237 ( 2002 ) . m. b. gavela , p. hernandez , j. orloff and o. pene , mod . lett . a * 9 * ( 1994 ) 795 ; m. b. gavela , p. hernandez , j. orloff , o. pene and c. quimbay , nucl . phys . b * 430 * ( 1994 ) 382 ; p. huet and e. sather , phys . d * 51 * ( 1995 ) 379 ; t. konstandin , t. prokopec and m. g. schmidt , nucl . b * 679 * ( 2004 ) 246 . k. kajantie , m. laine , k. rummukainen and m. e. shaposhnikov , phys . lett . * 77 * , 2887 ( 1996 ) ; k. rummukainen , m. tsypin , k. kajantie , m. laine and m. e. shaposhnikov , nucl . b * 532 * , 283 ( 1998 ) ; f. csikor , z. fodor and j. heitger , phys . lett . * 82 * , 21 ( 1999 ) ; y. aoki , f. csikor , z. fodor and a. ukawa , phys . d * 60 * , 013001 ( 1999 ) . m. s. carena , m. quiros and c. e. m. wagner , phys . b * 380 * , 81 ( 1996 ) . d. delepine , j. m. gerard , r. gonzalez felipe and j. weyers , phys . b * 386 * , 183 ( 1996 ) ; p. huet and a. e. nelson , phys . d * 53 * , 4578 ( 1996 ) ; b. de carlos and j. r. espinosa , nucl . b * 503 * , 24 ( 1997 ) ; m. s. carena , m. quiros , a. riotto , i. vilja and c. e. m. wagner , nucl . phys . b * 503 * , 387 ( 1997 ) ; a. riotto , nucl . phys . b * 518 * , 339 ( 1998 ) ; m. s. carena , m. quiros and c. e. m. wagner , nucl . b * 524 * , 3 ( 1998 ) ; k. funakubo , a. kakuto , s. otsuki and f. toyoda , prog . phys . * 99 * , 1045 ( 1998 ) ; a. riotto , phys . d * 58 * , 095009 ( 1998 ) ; k. funakubo , prog . theor . * 101 * , 415 ( 1999 ) ; k. funakubo , s. otsuki and f. toyoda , prog . phys . * 102 * , 389 ( 1999 ) ; j. m. cline , m. joyce and k. kainulainen , jhep * 0007 * , 018 ( 2000 ) ; m. s. carena , j. m. moreno , m. quiros , m. seco and c. e. m. wagner , nucl . phys . b * 599 * , 158 ( 2001 ) ; m. s. carena , m. quiros , m. seco and c. e. m. wagner , nucl . b * 650 * , 24 ( 2003 ) ; t. prokopec , m. g. schmidt and s. weinstock , annals phys . * 314 * , 208 ( 2004 ) ; t. prokopec , m. g. schmidt and s. weinstock , annals phys . * 314 * , 267 ( 2004 ) ; t. konstandin , t. prokopec and m. g. schmidt , nucl . b * 716 * , 373 ( 2005 ) ; c. lee , v. cirigliano and m. j. ramsey - musolf , phys . d * 71 * , 075010 ( 2005 ) ; v. cirigliano , m. j. ramsey - musolf , s. tulin and c. lee , phys . d * 73 * , 115009 ( 2006 ) ; t. konstandin , t. prokopec , m. g. schmidt and m. seco , nucl . b * 738 * , 1 ( 2006 ) ; v. cirigliano , s. profumo and m. j. ramsey - musolf , jhep * 0607 * , 002 ( 2006 ) ; d. j. h. chung , b. garbrecht , m. j. ramsey - musolf and s. tulin , phys . lett . * 102 * , 061301 ( 2009 ) . k. funakubo , s. tao and f. toyoda , prog . theor . * 109 * , 415 ( 2003 ) . m. carena , g. nardini , m. quiros and c. e. m. wagner , nucl . b * 812 * , 243 ( 2009 ) . k. funakubo , a. kakuto and k. takenaga , prog . * 91 * , 341 ( 1994 ) ; m. joyce , t. prokopec and n. turok , phys . d * 53 * , 2930 ( 1996 ) ; m. joyce , t. prokopec and n. turok , phys . d * 53 * , 2958 ( 1996 ) ; j. m. cline , k. kainulainen and a. p. vischer , phys . d * 54 * , 2451 ( 1996 ) ; j. m. cline and p. a. lemieux , phys . d * 55 * , 3873 ( 1997 ) ; s. kanemura , y. okada and e. senaha , phys . b * 606 * , 361 ( 2005 ) ; l. fromme , s. j. huber and m. seniuch , jhep * 0611 * , 038 ( 2006 ) . s. j. huber and m. g. schmidt , nucl . b * 606 * , 183 ( 2001 ) ; a. menon , d. e. morrissey and c. e. m. wagner , phys . d * 70 * , 035005 ( 2004 ) ; j. kang , p. langacker , t. j. li and t. liu , phys . rev . lett . * 94 * , 061801 ( 2005 ) ; k. funakubo , s. tao and f. toyoda , prog . * 114 * , 369 ( 2005 ) ; s. j. huber , t. konstandin , t. prokopec and m. g. schmidt , nucl . b * 757 * , 172 ( 2006 ) ; c. balazs , m. s. carena , a. freitas and c. e. m. wagner , jhep * 0706 * , 066 ( 2007 ) . w. s. hou , arxiv:0803.1234 [ hep - ph ] ; r. fok and g. d. kribs , phys . d * 78 * , 075023 ( 2008 ) ; y. kikukawa , m. kohda and j. yasuda , arxiv:0901.1962 [ hep - ph ] . r. barate _ [ lep working group for higgs boson searches and aleph collaboration and and ] , phys . b * 565 * , 61 ( 2003 ) . s. schael _ et al . _ [ aleph collaboration and delphi collaboration and l3 collaboration and ] , eur . j. c * 47 * , 547 ( 2006 ) . a. belyaev , q. h. cao , d. nomura , k. tobe and c. p. yuan , phys . lett . * 100 * , 061801 ( 2008 ) . m. asano , s. matsumoto , m. senami and h. sugiyama , phys . b * 663 * , 330 ( 2008 ) ; s. g. kim , n. maekawa , k. i. nagao , k. sakurai and t. yoshikawa , phys . rev . d * 78 * , 075010 ( 2008 ) . s. g. kim , n. maekawa , a. matsuzaki , k. sakurai , a. i. sanda and t. yoshikawa , phys . d * 74 * , 115016 ( 2006 ) . j. f. gunion , h. e. haber , g. l. kane and s. dawson , `` the higgs hunter s guide , '' ; `` errata for the higgs hunter s guide , '' arxiv : hep - ph/9302272 . a. pilaftsis and c. e. m. wagner , nucl . b * 553 * , 3 ( 1999 ) ; m. s. carena , j. r. ellis , a. pilaftsis and c. e. m. wagner , nucl . b * 586 * , 92 ( 2000 ) ; m. s. carena , j. r. ellis , a. pilaftsis and c. e. m. wagner , phys . b * 495 * , 155 ( 2000 ) ; m. s. carena , j. r. ellis , a. pilaftsis and c. e. m. wagner , nucl . phys . b * 625 * , 345 ( 2002 ) ; j. s. lee , a. pilaftsis , m. s. carena , s. y. choi , m. drees , j. r. ellis and c. e. m. wagner , comput . commun . * 156 * , 283 ( 2004 ) . j. s. lee , m. carena , j. ellis , a. pilaftsis and c. e. m. wagner , comput . commun . * 180 * , 312 ( 2009 ) . l. dolan and r. jackiw , phys . d * 9 * , 3320 ( 1974 ) . k. funakubo , a. kakuto , s. otsuki , k. takenaga and f. toyoda , prog . theor . phys . * 94 * , 845 ( 1995 ) ; k. funakubo , a. kakuto , s. otsuki and f. toyoda , prog . phys . * 98 * , 427 ( 1997 ) . j. m. moreno , m. quiros and m. seco , nucl . b * 526 * , 489 ( 1998 ) . s. j. huber , p. john , m. laine and m. g. schmidt , phys . b * 475 * , 104 ( 2000 ) ; s. j. huber , p. john and m. g. schmidt , eur . j. c * 20 * , 695 ( 2001 ) . a. d. linde , nucl . b * 216 * ( 1983 ) 421 [ erratum - ibid . b * 223 * ( 1983 ) 544 ] . m. e. carrington and j. i. kapusta , phys . d * 47 * , 5304 ( 1993 ) . n. s. manton , phys . d * 28 * , 2019 ( 1983 ) . p. arnold and l. d. mclerran , phys . d * 36 * , 581 ( 1987 ) . l. carson , x. li , l. d. mclerran and r. t. wang , phys . d * 42 * , 2127 ( 1990 ) . g. d. moore , phys . d * 59 * , 014503 ( 1999 ) . since the effects of the @xmath26 part on the sphaleron energy is less than a few percent @xcite , we neglect it in this article . f. r. klinkhamer and r. laterveer , z. phys . c * 53 * , 247 ( 1992 ) . f. r. klinkhamer and n. s. manton , phys . d * 30 * , 2212 ( 1984 ) . l. carson and l. d. mclerran , phys . d * 41 * , 647 ( 1990 ) . t. akiba , h. kikuchi and t. yanagida , phys . d * 40 * , 588 ( 1989 ) . j. m. moreno , d. h. oaknin and m. quiros , nucl . b * 483 * , 267 ( 1997 ) . m. e. shaposhnikov , nucl . b * 287 * , 757 ( 1987 ) ; m. e. shaposhnikov , nucl . b * 299 * , 797 ( 1988 ) ; a. i. bochkarev and m. e. shaposhnikov , mod . lett . a * 2 * , 417 ( 1987 ) ; m. e. shaposhnikov , phys . b * 277 * , 324 ( 1992 ) [ erratum - ibid . b * 282 * , 483 ( 1992 ) ] . t. n. phenomena , higgs working group , for the cdf collaboration and dzero collaboration , arxiv:0903.4001 [ hep - ex ] . y. okada , m. yamaguchi and t. yanagida , prog . phys . * 85 * , 1 ( 1991 ) ; j. r. ellis , g. ridolfi and f. zwirner , phys . b * 257 * , 83 ( 1991 ) ; r. barbieri and m. frigeni , phys . lett . b * 258 * , 395 ( 1991 ) ; y. okada , m. yamaguchi and t. yanagida , phys . b * 262 * , 54 ( 1991 ) ; h. e. haber and r. hempfling , phys . lett . * 66 * , 1815 ( 1991 ) ; j. r. ellis , g. ridolfi and f. zwirner , phys . b * 262 * , 477 ( 1991 ) . e. barberio _ et al . _ [ heavy flavor averaging group ] , arxiv:0808.1297 [ hep - ex ] . w. s. hou , phys . d * 48 * , 2342 ( 1993 ) . j. r. espinosa , nucl . b * 475 * , 273 ( 1996 ) ; d. bodeker , p. john , m. laine and m. g. schmidt , nucl . b * 497 * , 387 ( 1997 ) . k. funakubo and e. senaha , in preparation .

geometrically local quantum codes provide intriguing models of quantum many - body physics , and also have potential applications to fault - tolerant quantum computation in systems with short - range interactions . there has been impressive recent progress in understanding the properties of such codes . bravyi , poulin , and terhal @xcite showed that for codes defined by geometrically local commuting projectors in @xmath0 dimensions , the code length @xmath4 , distance @xmath1 and number of encoded qubits @xmath5 are related by @xmath6 bravyi and terhal @xcite showed that @xmath7 for subsystem codes with geometrically local gauge generators , and bravyi @xcite showed that @xmath8 for two - dimensional subsystem codes with geometrically local gauge generators . bravyi and terhal @xcite , and kay and colbeck @xcite , also showed that no two - dimensional local stabilizer code can be a _ self - correcting quantum memory _ if we regard the code as a system governed by a local hamiltonian , the energy barrier protecting against logical errors is a constant independent of system size . a self - correcting memory based on a geometrically local stabilizer code is possible in four dimensions @xcite , where the storage time increases sharply as the system size grows . in three dimensions there are codes such that the energy barrier increases logarithmically with system size @xcite , but where the storage time is bounded above by a constant independent of system size @xcite . we address a related but somewhat different question . to illustrate the question , consider the three - dimensional toric code @xcite , on a cubic lattice with linear size @xmath9 . this code provides different degrees of protection against different types of errors . for example , we can arrange for the logical bit flip acting on the code space to have weight @xmath9 ( _ i.e. _ , to be supported on a set of @xmath9 qubits ) , while the logical phase flip has weight @xmath10 . in that case , the energy barrier protecting against logical phase errors grows linearly with @xmath9 , though the energy barrier protecting against bit flips is only a constant . we might say this system is _ partially self correcting _ , meaning it has very robust physical protection against phase errors , but weaker protection against bit flips . we find limitations on partial self correction in two - dimensional local subsystem codes with local stabilizer generators ; in particular the logical phase flip must have weight @xmath11 if the logical bit flip has weight @xmath12 . more generally , we study how the code distance @xmath1 constrains the weight of logical operators , for both local commuting projector codes and subsystem codes , finding that @xmath1 limits not just the weight of the lowest - weight logical operator but also the higher - weight logical operators . let us say that two logical operators are _ equivalent _ if they act in the same way on the protected system . our result , which applies to both local subsystem codes and to local commuting projector codes in @xmath13 dimensions , says that for any logical operator there is an equivalent logical operator with weight @xmath2 such that @xmath14 where @xmath9 is the linear size of the lattice . we call this result the tradeoff theorem for logical operators , since , _ e.g. _ , increasing the weight of the lowest - weight logical operator reduces the upper bound on the weight of other logical operators . one immediate consequence is that , since @xmath15 , @xmath16 a result previously known for local subsystem codes but not for local commuting projector codes with @xmath17 . for @xmath18 the tradeoff becomes @xmath19 , and hence @xmath20 . we also show that for any two - dimensional local commuting projector code there is a nontrivial logical `` string '' operator supported on a narrow strip ( or on a narrow slab in higher dimensions ) , where the operator is only slightly entangling across any cut through the strip . however , we have not settled the question whether two - dimensional local commuting projector codes can be self correcting . we review the theory of stabilizer codes and subsystem codes in sec . ii . in sec . iii we prove a `` cleaning lemma '' for subsystem codes previously stated by bravyi @xcite ; our proof uses tools developed by yoshida and chuang @xcite , and may be of independent interest . we prove the tradeoff theorem for local subsystem codes in sec . iv and for local commuting projector codes in sec . v. in sec . vi we show that any two - dimensional commuting projector code admits a nontrivial logical `` string '' operator supported on a narrow strip . in sec . vii we explain why partial self - correction is impossible for two - dimensional local stabilizer codes with distance @xmath21 . in sec . viii we show that the logical string operator in a two - dimensional local commuting projector code can be chosen to be slightly entangling across any cut through the string . sec . ix contains our conclusions . a _ stabilizer code _ @xcite embeds @xmath5 protected qubits in the hilbert space of @xmath4 physical qubits . the code has a stabilizer group @xmath22 , an abelian subgroup with @xmath23 independent generators of the @xmath4-qubit pauli group @xmath24 , and the code is the simultaneous eigenspace with eigenvalue 1 of all elements of @xmath22 . it is convenient to abelianize @xmath24 by ignoring the phase in the product of two pauli operators , thus obtaining a @xmath25-dimensional vector space over the binary field , which we also denote by @xmath24 . the vector space @xmath24 is equipped with a symplectic form , such that two vectors are orthogonal if and only if the corresponding pauli operators commute . if @xmath26 is a subgroup of @xmath24 , we use the symbol @xmath26 to denote both the subgroup and the corresponding subspace of @xmath24 . viewed as a vector space , @xmath22 is @xmath27-dimensional . we denote by @xmath28 the vector space orthogonal to @xmath22 , which has dimension @xmath29 = @xmath30 . it can be decomposed as a direct sum of @xmath22 and a @xmath31 dimensional vector space corresponding to the logical pauli group , which acts nontrivially on the @xmath5 protected qubits . we define the weight of a pauli operator as the number of qubits on which the operator acts nontrivially , and the distance @xmath1 of the stabilizer code is the minimum weight of a nontrivial logical operator ( one contained in @xmath28 but not in @xmath22 ) . a _ subsystem code _ @xcite can be viewed as a stabilizer code with @xmath32 encoded qubits , but where only @xmath5 of these qubits are used to store protected quantum information . the stabilizer group @xmath22 together with pauli operators acting on the @xmath33 unused qubits generate the code s _ gauge group _ @xmath26 . equivalently , we may say that the subsystem code is defined by its gauge group @xmath34 , and that the code s stabilizer group @xmath35 is the subgroup of @xmath26 that commutes with all elements of @xmath26 . logical operations in the subsystem code preserve the @xmath36-dimensional hilbert space spanned by the @xmath5 protected qubits . we distinguish between _ bare _ logical operators , which act trivially on the gauge qubits , and _ dressed _ logical operators , which may act nontrivially on the gauge qubits as well as the protected qubits . thus , nontrivial bare logical operators are in @xmath37 but not in @xmath26 , while nontrivial dressed logical operators are in @xmath28 but not in @xmath26 . the code distance @xmath1 is the minimum weight of a nontrivial dressed logical operator . a bare logical operator @xmath38 acts trivially on the protected qubits as well as the gauge qubits if and only if @xmath39 ; hence we may regard @xmath40 as the group of bare logical operators . a dressed logical operator @xmath41 acts trivially on the protected qubits ( but perhaps nontrivially on the gauge qubits ) if and only if @xmath42 ; hence we may regard @xmath43 as the group of dressed logical operators , where we regard two dressed logical operators as equivalent if they act the same way on the protected qubits . we denote by @xmath44 $ ] the dimension of the vector space @xmath26 ( the number of independent generators of the corresponding group ) ; by counting the number of independent bare logical operators , we find that the number @xmath5 of protected qubits satisfies @xmath45=[g^\perp ] - [ s ] \\ & = & [ p ] - [ g ] - [ s]= 2n - [ g ] - [ s].\end{aligned}\ ] ] similarly , by counting the number of independent dressed logical operators , we find @xmath46=[s^\perp ] - [ g ] \\ & = & [ p ] - [ s ] - [ g]= 2n - [ s ] - [ g].\end{aligned}\ ] ] a stabilizer code is the special case of a subsystem code in which @xmath47 , and in that case , @xmath48 $ ] . we will also consider stabilizer codes and subsystem codes of the css type @xcite , where each generator of the gauge group , and each logical operator , may be chosen to be either of the @xmath49-type or the @xmath50-type . we use @xmath51 ( @xmath52 ) to denote the group of @xmath49-type ( @xmath50-type ) pauli operators , @xmath53 ( @xmath54 ) to denote the @xmath49-type ( @xmath50-type ) gauge group , and @xmath55 ( @xmath56 ) to denote the @xmath49-type ( @xmath50-type ) stabilizer group . we use ( @xmath57 to denote the subgroup of @xmath52 that commutes with @xmath53 , etc . then the group of bare @xmath50-type logical operators is @xmath58 and the group of bare @xmath49-type logical operators is @xmath59 . therefore the number @xmath5 of protected qubits is @xmath60= n - [ g^x ] - [ s^z],\\ k & = & [ ( g^z)^\perp / s^x]= n - [ g^z ] - [ s^x].\end{aligned}\ ] ] we wish to study stabilizer codes in which the stabilizer generators are geometrically local and subsystem codes in which the gauge generators are geometrically local . to be concrete , we may imagine that the qubits reside at the vertices of a @xmath0-dimensional hypercubic lattice ( with either open or periodic boundary conditions ) , and that each generator acts nontrivially only inside a hypercube ( containing @xmath61 vertices ) with linear size @xmath62 . in fact our results can be easily extended to codes with geometrically local generators defined on any graph embedded in @xmath0-dimensional space . note that for a subsystem code the stabilizer generators might be nonlocal even if the gauge generators are local . some of our results also apply to a larger class of local codes that includes local stabilizer codes . for this class , which we call _ local commuting projector codes _ , the code space is the simultaneous eigenspace with eigenvalue one of a set of mutually commuting geometrically local projection operators , where the projectors do not necessarily project onto eigenspaces of pauli operators . a local stabilizer code , but not a local subsystem code , is a special case of a local commuting projector code . the cleaning lemma for subsystem codes relates the number of independent bare logical operators supported on a set of qubits @xmath63 to the number of independent dressed logical operators supported on the complementary set @xmath64 . the concept of the cleaning lemma was introduced in @xcite , then generalized in @xcite and @xcite . here we use ideas from @xcite to prove a version stated in @xcite . ( see also @xcite . ) as in the sec . [ sec : background ] , we will regard a subgroup of the pauli group as a vector space , allowing us to obtain the cleaning lemma from straightforward dimension counting . we use @xmath65 to denote the subgroup of the pauli group @xmath24 supported on a set @xmath66 of qubits ; likewise for any subgroup @xmath26 of the pauli group @xmath67 , is the subgroup of @xmath26 supported on @xmath66 . we denote by @xmath68 the restriction map that maps a pauli operator to its restriction supported on the set @xmath66 , and we use @xmath69 to denote the number of qubits contained in @xmath66 ; thus @xmath70 = 2|a|$ ] . if we divide @xmath4 qubits into two complementary sets @xmath66 and @xmath71 , then a subgroup @xmath26 of @xmath24 can be decomposed into @xmath72 , @xmath73 , and a `` remainder , '' as follows : _ ( decomposition of pauli subgroups ) _ suppose that @xmath66 and @xmath71 are complementary sets of qubits . then for any subgroup @xmath26 of the pauli group , @xmath74 for some @xmath75 , where @xmath76 & = 2|a| - [ g_a ] - [ g ' ] , \\ [ ( g^\perp)_b ] & = 2|b| - [ g_b ] - [ g']\end{aligned}\ ] ] if @xmath77 is a vector space and @xmath78 is a subspace of @xmath77 , then there is a vector space @xmath79 such that @xmath80 ; we may choose @xmath79 to be the span of the basis vectors that extend a basis for @xmath78 to a basis for @xmath77 . since @xmath72 and @xmath73 are disjoint , i.e. , @xmath81 , @xmath82 is a subspace of @xmath26 , and thus there exists an auxiliary vector space @xmath83 such that @xmath84 the choice of @xmath75 is not canonical , but we need only its existence . since the restriction map @xmath85 obviously annihilates @xmath73 , we may regard it as a map from @xmath86 onto @xmath87 . in fact this map is injective . note that if @xmath88 for some @xmath89 . then since @xmath90 it must be that @xmath91 . but because the sum is direct , i.e. @xmath92 , it follows that @xmath93 , which proves injectivity . hence @xmath94 is an isomorphism . now , we may calculate @xmath95 by solving a system of linear equations . noting that @xmath96 is contained in @xmath37 if and only if @xmath97 commutes with the restriction to @xmath66 of each element of @xmath26 , we see that the number of independent linear constraints is @xmath98 = [ g_a ] + [ g']$ ] ; hence @xmath99=[p_a ] - [ g_a ] - [ g']= 2|a| - [ g_a ] - [ g']$ ] . likewise , @xmath100 is also an isomorphism , and hence @xmath101=[p_b ] - [ g_b ] - [ g']= 2|b| - [ g_b ] - [ g']$ ] . now we are ready to state and prove the cleaning lemma . for a subsystem code , let @xmath102 be the number of independent non - trivial bare logical operators supported on @xmath63 , and let @xmath103 be the number of independent non - trivial dressed logical operators supported on @xmath63 , i.e. , @xmath104 = [ ( g^\perp)_m / s_m ] , \\ g(m ) & = [ s^\perp \cap p_m / g_m ] = [ ( s^\perp)_m / g_m].\end{aligned}\ ] ] likewise , for a css subsystem code , let @xmath105 be the number of independent non - trivial bare @xmath49-type logical operators supported on @xmath63 , and let @xmath106 be the number of independent non - trivial dressed @xmath49-type logical operators supported on @xmath63 , i.e. , @xmath107 , \\ g^x(m ) & = [ ( s^z)^\perp \cap p^x_m / g^x_m],\end{aligned}\ ] ] and similarly for the @xmath50-type logical operators . _ ( cleaning lemma for subsystem codes ) _ for any subsystem code , we have @xmath108 where @xmath63 is any set of qubits and @xmath64 is its complement . moreover , for a css subsystem code @xmath109 [ lem : counting_op ] we use lemma 1 to prove the cleaning lemma by a direct calculation : @xmath110 \\ & = 2|m| - [ g_m ] - [ g ' ] - [ s_m ] , \end{aligned}\ ] ] and @xmath111 \\ & = 2|m^c| - [ s_{m^c } ] - [ s ' ] - [ g_{m^c } ] .\end{aligned}\ ] ] summing , we find @xmath112 + [ g_{m_c } ] + [ g'])\\ & -([s_m ] + [ s_{m_c } ] + [ s'])\end{aligned}\ ] ] and invoking lemma 1 once again , @xmath113 - [ s ] = 2k , \end{aligned}\ ] ] which proves the claim for general subsystem codes . for the css case , we apply the analogue of lemma 1 to the @xmath49-type and @xmath50-type pauli operators , finding @xmath114 \\ & = |m| - [ g^x_{m } ] - [ ( g^x ) ' ] - [ s^z_{m}]\end{aligned}\ ] ] and also @xmath115 \\ & = |m^c| - [ s^z_{m^c } ] - [ ( s^z ) ' ] - [ g^x_{m^c } ] . \end{aligned}\ ] ] summing and using lemma 1 we have @xmath116-[s^z ] = k ; \end{aligned}\ ] ] a similar calculation yields @xmath117-[s^x ] = k , \end{aligned}\ ] ] proving the claim for css subsystem codes . of course , for a stabilizer code there is no distinction between bare and dressed logical operators ; the statement of the cleaning lemma becomes @xmath118 for general stabilizer codes , and @xmath119 for css stabilizer codes . to understand how the cleaning lemma gets its name , note that it implies that if no bare logical operator can be supported on the set @xmath63 then all dressed logical operators can be supported on its complement @xmath64 . that is , any of the code s dressed logical pauli operators can be `` cleaned up '' by applying elements of the gauge group @xmath26 . the cleaned operator acts the same way on the protected qubits as the original operator ( though it might act differently on the gauge qubits ) , and acts trivially on @xmath63 . we say that a region @xmath63 is _ correctable _ if erasure of the qubits in @xmath63 is a correctable error . for a subsystem code , it follows that no nontrivial dressed logical operators are supported on @xmath63 if @xmath63 is correctable ; hence @xmath120 and thus @xmath121 . the cleaning lemma then asserts that all dressed logical operators can be supported on @xmath64 . let us say that two dressed logical operators @xmath97 and @xmath122 are _ equivalent _ if @xmath123 and @xmath124 is an element of the gauge group @xmath26 , so that @xmath97 and @xmath122 act the same way on the protected qubits . we have obtained : _ ( cleaning lemma for dressed logical operators ) _ [ lem : clean - region ] for any subsystem code , if @xmath63 is a correctable region and @xmath97 is a dressed logical operator , then there is a dressed logical operator @xmath122 supported on @xmath64 that is equivalent to @xmath97 . in this section we consider local subsystem codes with qubits residing at the sites of a @xmath0-dimensional hypercubic lattice @xmath125 . the code has _ interaction range _ @xmath62 , meaning that the generators of the gauge group @xmath26 can be chosen so that each generator has support on a hypercube containing @xmath61 sites . [ defn : boundary ] given a set of gauge generators for a subsystem code , and a set of qubits @xmath63 , let @xmath126 denote the support of all the gauge generators that act nontrivially on @xmath63 . the _ external boundary _ of @xmath63 is @xmath127 , where @xmath64 is the complement of @xmath63 , and the _ internal boundary _ of @xmath63 is @xmath128 . the _ boundary _ of @xmath63 is @xmath129 , and the _ interior _ of @xmath63 is @xmath130 . recall that a region ( _ i.e. _ , set of qubits ) @xmath63 is said to be _ correctable _ if no nontrivial dressed logical operation is supported on @xmath63 , in which case erasure of @xmath63 can be corrected . since the code distance @xmath1 is defined as the minimum weight of a dressed logical operator , @xmath63 is certainly correctable if @xmath131 . but in fact much larger regions are also correctable , as follows from this lemma : _ ( expansion lemma for local subsystem codes ) _ for a local subsystem code , if @xmath63 and @xmath66 are both correctable , where @xmath66 contains @xmath132 , then @xmath133 is correctable . [ lem : subsystem - extend ] given a subsystem code @xmath134 with gauge group @xmath26 , we may define a subsystem code @xmath135 on @xmath64 with gauge group @xmath136 , where @xmath137 maps a pauli operator to its restriction supported on @xmath64 . we note that a pauli operator @xmath97 supported on @xmath64 is a bare logical operator for @xmath134 if and only if @xmath97 is a bare logical operator for @xmath135 ; that is , @xmath97 commutes with all elements of @xmath26 if and only if it commutes with all elements of the restriction of @xmath26 to @xmath64 . furthermore , if @xmath97 is a dressed logical operator for @xmath135 supported on @xmath138 , then @xmath97 can be extended to a dressed logical operator @xmath139 for @xmath134 supported on @xmath132 . indeed , suppose @xmath123 , where @xmath122 is a bare logical operator for @xmath135 ( and hence also a bare logical operator for @xmath134 supported on @xmath64 ) , while @xmath124 is an element of the gauge group @xmath140 of @xmath135 . then @xmath124 can be written as a product @xmath141 of generators of @xmath140 , each of which can be expressed as @xmath142 , where @xmath143 is a generator of @xmath26 supported on @xmath144 . thus @xmath145 is a dressed logical operator for @xmath134 supported on @xmath132 . it follows that if @xmath132 is correctable for the code @xmath134 ( _ i.e _ , code @xmath134 has no nontrivial dressed logical operators supported on @xmath132 ) , then @xmath146 is correctable for the code @xmath135 ( @xmath135 has no nontrivial dressed logical operators supported on @xmath146 ) . by similar logic , if @xmath66 is correctable for @xmath134 and contains @xmath132 , then @xmath147 is correctable for @xmath135 . suppose now that the code @xmath134 has @xmath5 encoded qubits and that @xmath63 is correctable , _ i.e. _ @xmath148 . therefore , applying lemma [ lem : counting_op ] to the code @xmath134 , @xmath149 . suppose further that the set @xmath66 containing @xmath132 is correctable for @xmath134 , implying that @xmath147 is correctable for @xmath135 , _ i.e. _ @xmath150 . then applying lemma [ lem : counting_op ] to the code @xmath135 , we conclude that @xmath151 . since each bare logical operator for @xmath135 , supported on @xmath152 , is also a bare logical operator for @xmath134 , supported on @xmath152 , we can now apply lemma [ lem : counting_op ] once again to the code @xmath134 , using the partition into @xmath152 and @xmath133 , finding @xmath153 . thus @xmath154 is correctable . if the interaction range is @xmath62 , and @xmath63 is a correctable hypercube with linear size @xmath155 , then we may choose @xmath156 so that @xmath133 is a hypercube with linear size @xmath157 and @xmath158 is a hypercube with linear size @xmath159 . then @xmath66 contains @xmath160^d \le 4(w-1)dl^{d-1}\ ] ] qubits , and @xmath66 is surely correctable provided @xmath161 , where @xmath1 is the code distance . suppose that @xmath162 , so a single site is correctable . applying lemma [ lem : subsystem - extend ] repeatedly , we can build up larger and larger correctable hypercubes , with linear size @xmath163 . this process continues as long as @xmath164 . we conclude : _ ( holographic principle for local subsystem codes ) _ [ lem : subsystem - hypercube ] for a @xmath0-dimensional local subsystem code with interaction range @xmath165 and distance @xmath166 , a hypercube with linear size @xmath157 is correctable if @xmath167 thus ( roughly speaking ) for the hypercube to be correctable it suffices for its @xmath168$]-thickened _ boundary _ , rather than its volume , to be smaller than the code distance . bravyi @xcite calls this property `` the holographic principle for error correction , '' because the _ absence _ of information encoded at the boundary of a region ensures that no information is encoded in the `` bulk . '' for local stabilizer codes , the criterion for correctability is slightly weaker than for local subsystem codes , as we discuss in appendix [ app : holographic_lemma_stabilizer_codes ] . now we are ready to prove our first tradeoff theorem . and the white gap between squares has width @xmath169 . the solid blue curve represents the support of a nontrivial logical operator ; because the square @xmath170 is correctable , this square can be `` cleaned '' we can find an equivalent logical operator supported on @xmath171 , the complement of @xmath170 . when all squares are cleaned , the logical operator is supported on the narrow strips between the squares . , scaledwidth=42.0% ] _ ( tradeoff theorem for local subsystem codes ) _ for a local subsystem code in @xmath13 dimensions with interaction range @xmath165 and distance @xmath172 , defined on a hypercubic lattice with linear size @xmath9 , every dressed logical operator is equivalent to an operator with weight @xmath2 satisfying @xmath173 where @xmath174 is a constant depending on @xmath62 and @xmath0 . [ thm : subsystem - tradeoff ] as shown in fig . [ fig : cleaning ] , we fill the lattice with hypercubes , separated by distance @xmath169 , such that each hypercube has linear size @xmath157 satisfying eq .. ( by `` distance '' we mean the number of sites in between _ e.g. _ we say that adjacent sites are `` distance zero '' apart . ) thus no gauge generator acts nontrivially on more than one hypercube , and each hypercube is correctable by lemma [ lem : subsystem - hypercube ] . consider any nontrivial dressed logical operator @xmath97 , and label the hypercubes @xmath175 . by lemma 3 there exists a gauge operator @xmath176 that `` cleans '' the logical operator in the hypercube @xmath170 , _ i.e. _ , such that @xmath177 acts trivially in @xmath170 . furthermore , since no gauge generator acts nontrivially on more than one hypercube , we can choose @xmath176 so that it acts trivially in all other hypercubes . taking the product of all the @xmath176 s we construct a gauge operator that cleans all hypercubes simultaneously ; thus @xmath178 is equivalent to @xmath97 and supported on the complement of the union of hypercubes @xmath179 . therefore , the weight @xmath2 of @xmath180 is upper bounded by @xmath181 . the lattice is covered by hypercubes of linear size @xmath182 , each centered about one of the @xmath170 s . there are @xmath183^d$ ] such hypercubes in this union , each containing no more than @xmath184^d - l^d \le ( w-1)d\left[l+(w-1)\right]^{d-1}$ ] elements of @xmath64 . thus @xmath185^{d-1}\frac{l^d}{\left[l+(w-1)\right]^d } \nonumber\\ & = \frac{(w-1)d}{l+(w-1)}l^d.\end{aligned}\ ] ] we optimize this upper bound on @xmath2 by choosing @xmath157 to be the largest integer such that a hypercube with linear size @xmath157 is known to be correctable , _ i.e. _ , satisfying @xmath186 thus obtaining eq .. note that eq . is trivial if @xmath1 is a constant independent of @xmath9 , since the weight @xmath2 can not be larger than @xmath187 . in this section we consider a local commuting projector code , defined as the simultaneous eigenspace with eigenvalue one of a set of commuting projectors . as in sec . [ sec : subsystem - tradeoff ] we assume that the qubits reside on a hypercubic lattice @xmath125 and that each projector acts trivially outside a hypercube of linear size @xmath62 , where @xmath62 is the interaction range . by a _ logical operator _ we mean any transformation that preserves the code space , and we say that two logical operators are _ equivalent _ if they have the same action on the code space . the weight of a logical operator is the number of qubits on which it acts nontrivially . we say that a set of qubits @xmath63 is correctable if erasure of @xmath63 can be reversed by a trace - preserving completely positive recovery map . the distance @xmath1 of the code is the minimum size of a noncorrectable set of qubits . bravyi , poulin , and terhal @xcite proved some useful properties of these codes . to state their results , we use the definition [ defn : boundary - projector ] given a set of commuting projectors defining a code , and a set of qubits @xmath63 , let @xmath126 denote the support of all the projectors that act nontrivially on @xmath63 . the _ external boundary _ of @xmath63 is @xmath127 , where @xmath64 is the complement of @xmath63 , and the _ internal boundary _ of @xmath63 is @xmath128 . the _ boundary _ of @xmath63 is @xmath129 , and the _ interior _ of @xmath63 is @xmath130 . [ lem : disentangling ] _ ( disentangling lemma @xcite ) _ consider a local commuting projector code and suppose that @xmath63 and @xmath138 are both correctable regions . then there exists a unitary operator @xmath188 acting only on the boundary @xmath132 such that , for any pure code vector @xmath189 , @xmath190 here @xmath191 , supported on @xmath63 , does not depend on the code vector @xmath189 , while @xmath192 , supported on @xmath64 , does depend on @xmath189 . the disentangling lemma says that , if @xmath63 and @xmath138 are both correctable , then the entanglement of code vectors across the cut between @xmath63 and @xmath64 is localized in @xmath132 and can be removed by a unitary transformation acting on only @xmath132 . furthermore , in the resulting product state , no information distinguishing one code vector from another is available in @xmath63 . this lemma has a simple but important corollary : _ ( expansion lemma for local commuting projector codes @xcite ) _ for a local commuting projector code , if @xmath63 and @xmath66 are both correctable , where @xmath66 contains @xmath132 , then @xmath133 is correctable . [ lem : commuting - projector - extend ] by eq.([eq : disentangling ] ) , if @xmath66 is erased the resulting state on @xmath158 is independent of the code vector @xmath189 ; all the information needed to reconstruct @xmath189 resides in @xmath152 . therefore , we can erase @xmath158 as well without compromising our ability to reconstruct @xmath189 ; that is , @xmath133 is correctable . definition [ defn : boundary - projector ] and lemma [ lem : commuting - projector - extend ] for commuting projector codes are parallel to definition [ defn : boundary ] and lemma [ lem : subsystem - extend ] for subsystem codes . arguing as in the proof of lemma [ lem : subsystem - hypercube ] , we see that one consequence is a holographic principle for these codes : _ ( holographic principle for local commuting projector codes ) _ [ lem : projector - hypercube ] for a @xmath0-dimensional local commuting projector code with interaction range @xmath165 and distance @xmath166 , a hypercube with linear size @xmath157 is correctable if @xmath193 we will need an analog of the cleaning lemma to analyze the logical operator tradeoff for local commuting projector codes ; it can be derived from the disentangling lemma . _ ( cleaning lemma for local commuting projector codes ) _ consider a local commuting projector code , and suppose that @xmath63 and @xmath146 are both correctable . for any logical operator @xmath78 there exists an equivalent logical operator @xmath77 supported on the complement of the interior @xmath194 of @xmath63 . if @xmath78 is an isometry , then @xmath77 can be chosen to be unitary . [ lem : cleaning - projector ] let us name the regions : @xmath195 let @xmath196 be an orthonormal basis for the code space . by lemma [ lem : disentangling ] , there exists a unitary transformation @xmath197 , and vectors @xmath198 such that @xmath199 where the normalized vector @xmath200 does not depend on @xmath201 and the vectors @xmath202 are normalized and mutually orthogonal . because @xmath78 is a logical operator , @xmath203 is also a code vector , and therefore @xmath204 where @xmath205 is another set of vectors ; if @xmath78 is an isometry then these vectors , too , are normalized and mutually orthogonal . define a transformation @xmath79 by @xmath206 , and choose an arbitrary extension so that @xmath79 becomes an operator on @xmath207 . if @xmath78 is an isometry , then this extension @xmath208 can be chosen to be unitary . we now have @xmath209 for all @xmath201 . defining @xmath210 we observe that @xmath211 acts trivially on @xmath212 and has the same action on code vectors as @xmath78 , completing the proof . to prove the tradeoff theorem we will need a further lemma establishing that a union of correctable sets is correctable under suitable conditions . recall that we say a set of qubits @xmath63 is correctable if and only if erasure of @xmath63 can be corrected . equivalently , @xmath63 is correctable if and only if , for any operator @xmath213 supported on @xmath63 , @xmath214 where @xmath215 denotes the projector onto the code space and @xmath216 is a constant ( possibly zero ) depending on @xmath213 @xcite . _ ( the union of separated correctable regions is correctable ) _ for a local commuting projector code , suppose that @xmath63 and @xmath217 are correctable regions such that no projector acts nontrivially on both @xmath63 and @xmath217 . then @xmath218 is also correctable . [ lem : extension_of_correctable_region ] a weaker version of this lemma was proved in @xcite . let @xmath219 be the set of local commuting projectors that define the code . we denote by @xmath220 the set of projectors in @xmath219 that act nontrivially on @xmath217 . define @xmath221 and note that the projector onto the code space is @xmath222 also note that the support of @xmath223 does not intersect @xmath63 and the support of @xmath224 does not intersect @xmath217 . let @xmath213 be an arbitrary operator supported on @xmath218 ; we will show that @xmath213 satisfies eq .. since @xmath63 and @xmath217 are disjoint , @xmath213 has a schmidt decomposition @xmath225 where each @xmath226 is supported on @xmath63 and each @xmath227 is supported on @xmath217 . since @xmath223 commutes with @xmath226 and @xmath224 commutes with @xmath227 , @xmath228 where in the fourth equality we used the correctability of @xmath63 and @xmath217 . thus @xmath213 obeys eq . , and @xmath218 is correctable . now we are ready to state and prove our second tradeoff theorem . _ ( tradeoff theorem for local commuting projector codes ) _ for a local commuting projector code in @xmath13 dimensions with interaction range @xmath165 and distance @xmath172 , defined on a hypercubic lattice with linear size @xmath9 , every logical operator is equivalent to an operator with weight @xmath2 satisfying @xmath229 where @xmath174 is a constant depending on @xmath62 and @xmath0 . [ thm : commuting_tradeoff ] the proof is similar to the proof of theorem [ thm : subsystem - tradeoff ] . we fill the lattice with hypercubes , separated by distance @xmath169 , where each hypercube @xmath170 has linear size @xmath157 sufficiently small so that @xmath170 and @xmath230 are both correctable . applying lemma [ lem : extension_of_correctable_region ] repeatedly , we conclude that the union @xmath63 of all @xmath170 is correctable , and the union @xmath138 of all @xmath231 is correctable . for any logical operator , lemma [ lem : cleaning - projector ] now ensures the existence of an equivalent logical operator supported outside the interior @xmath232 of @xmath63 , and hence the weight @xmath2 of this equivalent logical operator is bounded above by @xmath233 . the lattice is covered by hypercubes with linear size @xmath234 , each centered about one of the @xmath170 , and there are @xmath183^d$ ] such hypercubes , each containing no more than @xmath235^d - \left[l-2(w-1)\right]^d \nonumber\\ \le 3(w-1)d\left[l+(w-1)\right]^{d-1}\end{aligned}\ ] ] elements of @xmath236 ; therefore , @xmath237^{d-1}\frac{l^d}{\left[l+(w-1)\right]^d}\\ & = \frac{3(w-1)d}{l+(w-1)}l^d.\end{aligned}\ ] ] to ensure that @xmath170 and @xmath231 are correctable , it suffices that @xmath238 , where @xmath1 is the code distance , or @xmath239^d - \left[l-2(w-1)\right]^d \\ & \le 4(w-1)d\left[l+2(w-1)\right]^{d-1 } < d.\end{aligned}\ ] ] we choose the largest such integer value of @xmath157 , obtaining eq .. because the code distance @xmath1 is defined as the size of the smallest noncorrectable set , and because a set supporting a nontrivial logical operator is noncorrectable , we have @xmath240 and hence theorem [ thm : commuting_tradeoff ] implies @xmath241 in fact we can make a stronger statement , specifying the geometry of a region that supports a nontrivial logical operator with weight @xmath242 . on the hypercube @xmath243 , we refer to the set @xmath244 as a _ slab _ of width @xmath245 . let us say that a code is nontrivial if the code space dimension is greater than one . then : _ ( existence of a noncorrectable thin slab ) _ for a nontrivial local commuting projector code in @xmath246 dimensions with interaction range @xmath165 , there is a noncorrectable slab of width @xmath247 . [ lem : slab ] suppose , contrary to the claim , that any slab of width @xmath247 is correctable . choose a correctable slab @xmath63 of width @xmath247 . the boundary @xmath132 of @xmath63 is contained in two slabs @xmath248 and @xmath249 , each of width @xmath250 . hence @xmath248 and @xmath249 are both correctable , and since @xmath63 has width @xmath247 , @xmath248 and @xmath249 are separated by @xmath169 . therefore , no local projector acts on both @xmath248 and @xmath249 , and by lemma [ lem : extension_of_correctable_region ] , @xmath251 is correctable . then lemma [ lem : commuting - projector - extend ] implies that the slab @xmath252 of width @xmath253 is correctable . repeating the argument , we see that if a slab @xmath63 of width @xmath245 is correctable , so is the slab of width @xmath254 containing @xmath63 . if the system obeys open boundary conditions , then by induction the entire lattice is correctable . if the lattice is periodic , we may consider two thick correctable slabs @xmath255 such that @xmath256 is the entire lattice and @xmath257 ; in that case lemma [ lem : commuting - projector - extend ] implies that the entire lattice @xmath256 is correctable . for either type of boundary condition , then , there are no nontrivial logical operators at all . but we assumed that the code is nontrivial , and therefore reach a contradiction . it follows from lemma [ lem : slab ] that the distance @xmath1 of a local commuting projector code satisfies @xmath258 it was previously known that @xmath259 for a local stabilizer code @xcite and @xmath260 for a local subsystem code @xcite . now we may wonder about the geometry of a set that supports a nontrivial logical operator . for a subsystem code , there is a nontrivial logical operator supported by any noncorrectable set , but this statement is not true for general codes ( see appendix [ app : counter_example_nolop_noncorrectable ] ) . we say that an operator @xmath213 is a _ logical _ operator if it preserves the code space , and that it is a _ nontrivial _ logical operator if it preserves the code space and its restriction to the code space is not proportional to the identity . from the definition of correctability , then , @xmath63 is not correctable if it supports a nontrivial logical operator . but for some codes the converse is false . if @xmath63 is not correctable , then an operator @xmath213 exists that fails to satisfy eq . ; however @xmath213 might not preserve the code space . but for a local commuting projector code , a correctable set can be extended to a slightly larger set that does support a nontrivial logical operator . suppose the code is the simultaneous eigenspace with eigenvalue one of a set of commuting projectors @xmath261 . for any set of qubits @xmath63 , we define @xmath126 as the support of all the projectors that act nontrivially on @xmath63 . then if @xmath63 is noncorrectable a nontrivial unitary logical operator is supported on @xmath126 . [ lem : projector - support ] _ ( support for nontrivial logical operator ) _ for a commuting projector code , if the set @xmath63 is not correctable , then there is a nontrivial unitary logical operator supported on @xmath126 that commutes with every projector in @xmath219 . let @xmath262 be the projector onto the code space . we claim that there exists a pauli operator @xmath263 supported on @xmath63 such that @xmath264 is not proportional to @xmath215 . indeed , if @xmath63 is not correctable , then there exists an operator @xmath265 supported on @xmath63 such that @xmath266 . expanding @xmath267 as a linear combination of pauli operators , we see that at least one pauli operator @xmath268 must satisfy @xmath269 . we denote by @xmath270 the set of projectors in @xmath219 that act nontrivially on @xmath63 , and define @xmath271 we claim that @xmath272 is a nontrivial hermitian logical operator supported on @xmath126 . to see that @xmath273 is a logical operator , note that if @xmath274 , then @xmath275 , because @xmath276 ; hence @xmath277 _ i.e. _ , @xmath278 commutes with @xmath273 . if @xmath279 , then @xmath278 is supported in the complement @xmath64 of @xmath63 ; hence it commutes trivially with @xmath263 , and therefore also with @xmath280 . since @xmath280 commutes with each projector in @xmath219 , it certainly commutes with @xmath215 and hence preserves the code space . furthermore , because @xmath281 @xmath273 acts on the code space in the same way as @xmath282 , and therefore must be nontrivial . thus @xmath283 preserves the code space and is unitary for any real @xmath284 . since @xmath273 , restricted to the code space , has at least two distinct eigenvalues , the same is true of @xmath285 for a generic choice of @xmath284 ; _ i.e. _ , @xmath285 is a nontrivial unitary logical operator . lemmas [ lem : slab ] and [ lem : projector - support ] now imply : _ ( a logical operator is supported on one thin slab ) _ for a nontrivial local commuting projector code in @xmath246 dimensions , with interaction range @xmath286 , there is a nontrivial unitary logical operator ( commuting with all projectors ) supported on a slab of width @xmath253 . [ thm : slab ] note that , though the proof of theorem [ thm : slab ] establishes the existence of a logical operator supported on a slab of constant width , it provides no algorithm for constructing the operator . in @xmath18 dimensions , the slab becomes a strip of constant width stretching across the @xmath287 code block , and the logical operator supported on the strip may be called a `` string '' operator . it was previously known that for @xmath18 a string operator can be supported on a strip of width @xmath62 in a local stabilizer code @xcite , and of width @xmath288 in a local subsystem code @xcite . theorems 1 and 2 constrain the weight of logical operators , but the proofs tell us more they specify the _ geometry _ of a region that supports a logical operator . this geometry has further implications for the physical robustness of quantum memories . consider a subsystem code whose stabilizer group @xmath22 has a set of geometrically local generators @xmath289 , where the qubits reside at the sites of a @xmath0-dimensional hypercubic lattice with linear size @xmath9 . the generating set @xmath289 might be overcomplete , but we assume that the number of generators acting nontrivially on each qubit is a constant independent of @xmath9 . the local hamiltonian @xmath290 has a @xmath291-fold degenerate ground state with energy @xmath292 , where @xmath5 is the number of protected qubits and @xmath33 is the number of gauge qubits of the subsystem code each ground state is a simultaneous eigenstate with eigenvalue one of all elements of @xmath289 . if a quantum memory governed by this hamiltonian is subjected to thermal noise , how well protected is the @xmath36-dimensional code space ? if @xmath189 is a zero - energy eigenstate of @xmath293 and @xmath294 , then @xmath295 is an eigenstate of @xmath293 with eigenvalue @xmath296 , where @xmath296 is the number of elements of @xmath289 that anticommute with @xmath97 . thermal fluctuations may excite the memory , but excitations with energy cost @xmath297 are suppressed by the boltzmann factor @xmath298 where @xmath299 is the temperature ( and boltzmann s constant @xmath300 has been set to one ) . following @xcite , we suppose that the environment applies a sequence of weight - one pauli operators to the system , so that the error history after @xmath301 steps can be described as a walk on the pauli group , starting at the identity : @xmath302 where @xmath303 , and @xmath304 has weight one . let @xmath305 denote the set of all such walks , with any number of steps , that start at @xmath306 and terminate at @xmath307 . we define @xmath308 the minimum energy barrier that must be surmounted by any walk that reaches pauli operator @xmath124 . thus such walks occur with a probability per unit time suppressed by the boltzmann factor @xmath309 . we also define @xmath310 here @xmath311 is the lowest energy barrier protecting any nontrivial dressed logical operator ( representing a nontrivial coset of @xmath312 ) , and @xmath313 is the highest such energy barrier . we say that a quantum memory is _ self correcting _ if @xmath311 grows faster than logarithmically with @xmath9 . in that case _ all _ nontrivial logical operators are suppressed by a boltzmann factor whose reciprocal grows super - polynomially with @xmath9 . we say that the quantum memory is _ partially self correcting _ if @xmath313 grows faster than logarithmically with @xmath9 . in that case _ at least one _ logical operator is protected by an energy barrier that increases with system size . though the pauli walk may not be a particularly accurate description of noise in realistic systems , it allows us to define the notion of barrier height precisely , and to state the criteria for self correction and partial self correction simply . furthermore , we expect the boltzmann factor @xmath314 suppressing the pauli walk to provide a reasonable ( though crude ) estimate of the logical error rate for more realistic noise models , assuming that the system attains thermal equilibrium . bravyi and terhal @xcite , and kay and colbeck @xcite , showed that no two - dimensional local subsystem code with local stabilizer generators can be self correcting . on the other hand , partially self - correcting quantum memories are certainly possible in two dimensions the ising model , regarded as a quantum repetition code , is an example . in the ising model , the logical bit flip operator flips every qubit , hence @xmath315 . in the pauli walk that reaches the logical bit flip and traverses the lowest barrier , a domain wall of length @xmath12 sweeps across the system ; hence @xmath316 . theorem [ thm : subsystem - tradeoff ] shows that this high value of @xmath2 for the logical bit flip is possible only because the code distance @xmath1 is @xmath317 , and hence a logical phase flip can be realized by an operator of constant weight . but suppose that , as in the toric code @xcite , a logical phase flip can occur only if a thermally activated localized quasiparticle propagates across the system . thus @xmath318 for the logical phase flip . can the logical bit flip still be protected by a high barrier ? arguing as in @xcite , and invoking theorem [ thm : subsystem - tradeoff ] , we see that under this condition robust protection against bit flips can not be achieved using a local subsystem code with local stabilizer generators . _ ( limitation on partial self correction in local subsystem codes ) _ for a two - dimensional local subsystem code , with qubits residing at sites of an @xmath287 square lattice , suppose that @xmath289 is a ( possibly overcomplete ) set of _ geometrically local _ stabilizer generators , where the number of generators acting on each qubit is an @xmath9-independent constant . consider a quantum memory governed by the hamiltonian eq .. if the code distance is @xmath21 , then the memory is not partially self correcting _ i.e. _ , @xmath319 . more generally , if the code distance is @xmath320 in @xmath0 spatial dimensions , then @xmath321 , where @xmath322 . [ thm : no - partial ] let @xmath62 be the interaction range of the gauge generators of the subsystem code and let @xmath323 be the interaction range of the stabilizer generators . for any dressed logical operator @xmath97 supported on this set , we may build a pauli walk that starts at @xmath306 and ends at @xmath97 by first building the horizontal strings column by column and then building the vertical strings row by row . at each stage of this walk , any `` excited '' local stabilizer @xmath324 such that @xmath325 acts only on qubits in a @xmath326 square that contains qubits either at the boundary of the walk or in the intersection of a horizontal and vertical string . the number of such qubits is @xmath317 and the total number of stabilizer generators acting on these qubits is @xmath317 . therefore , the energy cost of the partially completed walk , and hence @xmath313 , are @xmath317 . in @xmath0 spatial dimensions , the proof of theorem [ thm : subsystem - tradeoff ] shows that the support of any dressed logical operator can be reduced to a network of overlapping @xmath327-dimensional slabs , where each slab has constant width and slabs with the same orientation are separated by distance @xmath157 such that @xmath328 ; hence @xmath329 if @xmath330 . for any dressed logical operator supported on this set of slabs , we may build a pauli walk that sweeps across the system , such that at each stage of the walk the excited stabilizer generators are confined to a @xmath327-dimensional `` surface . '' this surface may be oriented such that it cuts across each slab on a @xmath331-dimensional surface with weight @xmath332 . there are @xmath333 such intersections ; therefore during the walk the total number of excited stabilizer generators ( and hence the energy cost ) is @xmath334 , where @xmath335 . we needed to assume that each @xmath324 is geometrically local to ensure that eq . is a geometrically local hamiltonian . for any local subsystem code with geometrically local gauge generators , whether or not the stabilizer generators are also geometrically local , the hamiltonian @xcite @xmath336 is geometrically local , where now @xmath337 is the set of gauge generators . however , because the gauge generators are not mutually commuting , the energetics of a pauli walk is not easy to study in this model , which is beyond the scope of theorem [ thm : no - partial ] . for a two - dimensional local commuting projector code , the simultaneous eigenspace with eigenvalue one of the projectors @xmath338 , the code space is the degenerate ground state with energy @xmath292 of the hamiltonian @xmath339 if only a constant number of projectors act on each qubit , then an operator supported on a set @xmath63 can increase the energy by at most @xmath340 , where @xmath174 is a constant . since theorem [ thm : slab ] establishes the existence of a nontrivial logical operator supported on a narrow strip , one might anticipate that , by arguing as in the proof of theorem [ thm : no - partial ] , we can show that this system is not self correcting or partially self correcting . we may envision a sequence of operations interpolating between the identity and a nontrivial logical operator , where each operation in the sequence could plausibly evolve from the previous operation due to the action of a thermal bath . in the strip @xmath63 of constant width that supports a nontrivial logical operator @xmath213 , we can divide the qubits into two subsets @xmath66 and @xmath341 , imagining that the interface between @xmath66 and @xmath71 gradually creeps along the strip . now , however , we encounter an important distinction between stabilizer codes and more general commuting projector codes . for a stabilizer code , the nontrivial logical operator supported in @xmath63 can be chosen to be a pauli operator , and hence the product of an operator supported in @xmath66 and an operator supported in @xmath71 . for a commuting projector code , a logical operator supported in @xmath63 may actually be _ entangling _ across the @xmath66-@xmath71 cut . are we assured that this entangling operation can be built up gradually due to the effects of local noise ? we have not been able to settle this question . we _ can _ say that in any two - dimensional local commuting projector code there exists a nontrivial logical operator that is only _ slightly entangling _ across any cut through the strip . this property , however , might not suffice to guarantee that the logical operator can be constructed as a product of physical operations , where each operation acts on a constant number of system qubits near the @xmath66-@xmath71 cut and also on a constant number of ancillary qubits in the `` environment . '' to define the notion of `` slightly entangling '' for an operator @xmath213 supported on @xmath342 , we perform a schmidt decomposition @xmath343 here @xmath344 is a set of nonnegative real numbers , while @xmath345 is a set of operators supported on @xmath66 and @xmath346 is a set of operators supported on @xmath71 , with the normalization conditions @xmath347 the number of nonzero terms in the schmidt decomposition is the schmidt rank of @xmath213 , and we say that @xmath213 is slightly entangling if its schmidt rank is a constant independent of system size . as we know from theorem [ thm : slab ] , for a two - dimensional local commuting projector code on an @xmath287 lattice , there is a nontrivial logical operator supported on a vertical strip @xmath63 with dimensions @xmath348 , where @xmath245 is a constant . @xmath63 can be regarded as the disjoint union of an @xmath349 rectangle @xmath66 covering the bottom of @xmath63 and an @xmath350 rectangle @xmath71 covering the top of @xmath63 . we can prove [ lem : e_slightly_entangling_lop ] _ ( existence of slightly entangling logical operators ) _ for a nontrivial two - dimensional local commuting projector code , there is a nontrivial hermitian logical operator @xmath273 supported on a strip of constant width @xmath126 such that , for any division of @xmath126 into constant - width rectangles @xmath66 and @xmath71 , @xmath273 is slightly entangling across the @xmath66-@xmath71 cut . we know from lemmas [ lem : slab ] and [ lem : projector - support ] that there is a noncorrectable constant - width strip @xmath63 and a pauli operator @xmath263 supported on @xmath63 such that @xmath272 is a nontrivial hermitian logical operator supported on @xmath126 ; here @xmath351 and @xmath270 is the set of projectors that act nontrivially on @xmath63 . the pauli operator @xmath263 is a product operator , with schmidt number one across the @xmath66-@xmath71 cut . among the local projectors occurring in the product @xmath352 , those fully supported on either @xmath66 or @xmath71 have no effect on the schmidt number of @xmath353 , and only a constant number of the projectors act nontrivially on both @xmath66 and @xmath71 . since each such @xmath278 is supported on a constant number of qubits , the action of @xmath278 increases the schmidt number by a constant . thus @xmath273 has constant schmidt number , _ i.e. _ is slightly entangling . we may relax the notion of slightly entangling , regarding an operator @xmath213 as slightly entangling if it may be _ well approximated _ by an operator with constant schmidt rank . in this sense the unitary logical operator @xmath354 is also slightly entangling . we may expand the exponential as a power series where each term has a schmidt rank independent of system size ; furthermore , the power series expansion truncated at constant order approximates the exponential function very well with respect to the operator norm . now we might hope to construct a slightly entangling logical operator @xmath213 , supported on a constant - width vertical strip , by gradually building its support one horizontal row of qubits at a time . however , lamata _ @xcite showed that , if @xmath213 is entangling , then it can not be obtained as a product of _ unitary _ operators where each of these operators acts on just a few rows of system qubits and on a shared ancillary system . an alternative procedure for gradually building a nontrivial logical error has been proposed by landon - cardinal and poulin @xcite . they envision a walk along the strip such that , in each step of the walk , first a constant size set of qubits depolarizes , and then the code projectors acting on that set are applied . if the projection fails to accept the state , the step can be repeated until the projection succeeds . this procedure could fail if at some stage the projection succeeds with zero probability . but landon - cardinal and poulin @xcite have shown that their procedure eventually generates a nontrivial logical error ( and hence that the code is not self correcting ) for any local commuting projector code obeying a `` local topological order '' criterion @xcite . whether self - correcting two - dimensional local commuting projector codes are possible remains open , though , because topologically ordered codes that violate _ local _ topological order have not been ruled out . the quantum accuracy threshold theorem @xcite shows that quantum information can be reliably stored and processed by a noisy physical system if the noise is not too strong . but can quantum information be protected `` passively '' in a macroscopic physical system governed by a static local hamiltonian , at a sufficiently low nonzero temperature ? this question @xcite , aside from its far - reaching potential implications for future quantum technologies , is also a fundamental issue in quantum many - body physics . hamiltonians derived from local quantum codes , whose properties are relatively easy to discern , can provide us with valuable insights . a two - dimensional ferromagnet can be a self - correcting classical memory , but a hamiltonian based on a two - dimensional local subsystem code with local stabilizer generators can not be a self - correcting quantum memory @xcite . we have shown that for two - dimensional local subsystem code with local stabilizer generators on an @xmath355 square lattice , robust _ classical _ protection is impossible if the code distance is @xmath21 , as expected for a topologically ordered two - dimensional system . more generally , we have studied how the code distance @xmath1 limits the size of the support of arbitrary nontrivial logical operators , in both local subsystem codes and local commuting projector codes . in view of the upper bound @xmath356 on the code distance , we may write @xmath357 where @xmath358 , and thus our upper bound eq . on the weight of logical operators becomes @xmath359 in particular , in three dimensions , @xmath21 implies @xmath360 . we have also shown that any two - dimensional local commuting projector code admits a nontrivial logical string operator which is only slightly entangling across any cut through the string . our arguments modestly extend the findings of @xcite , and use similar ideas . in passing , we also proved a cleaning lemma for subsystem codes based on ideas from @xcite , proved a cleaning lemma for local commuting projector codes . our methods might find further applications in future studies of quantum memories based on local codes . we are grateful to salman beigi , alexei kitaev , robert knig , olivier landon - cardinal , and norbert schuch for helpful discussions , and we especially thank david poulin for useful comments on the manuscript . this research was supported in part by nsf under grant no . phy-0803371 , by doe under grant no . de - fg03 - 92-er40701 , by nsa / aro under grant no . w911nf-09 - 1 - 0442 , and by the korea foundation for advanced studies . the institute for quantum information and matter ( iqim ) is an nsf physics frontiers center with support from the gordon and betty moore foundation . we say that a local stabilizer code has interaction range @xmath62 if each stabilizer generator has support on a hypercube containing @xmath61 sites . for this case , we can improve the criterion for correctability of a hypercube , found for local subsystem codes in lemma [ lem : subsystem - hypercube ] . _ ( expansion lemma for local stabilizer codes ) _ for a local stabilizer code , suppose that @xmath138 , @xmath66 , and @xmath158 are all correctable , where @xmath361 . then @xmath63 is also correctable . suppose , contrary to the claim , that there is a nontrivial logical operator @xmath97 supported on @xmath63 . then , because @xmath66 is correctable , lemma [ lem : clean - region ] implies that there is a stabilizer generator @xmath122 such that @xmath362 acts trivially on @xmath66 . furthermore , @xmath122 can be expressed as a product of local stabilizer generators , each supported on @xmath363 . thus @xmath362 is a product of two factors , one supported on @xmath158 and the other supported on @xmath138 . because @xmath364 , no local stabilizer generator acts nontrivially on both @xmath158 and @xmath138 ; therefore , each factor commutes with all stabilizer generators and hence is a logical operator . because @xmath158 and @xmath138 are both correctable , each factor is a trivial logical operator and therefore @xmath362 is also trivial . it follows that @xmath97 is trivial , a contradiction . now , if the interaction range is @xmath62 and @xmath63 is a hypercube with linear size @xmath157 , we choose @xmath66 so that @xmath158 is a hypercube with linear size @xmath155 , and we notice that @xmath138 is contained in a hypercube with linear size @xmath365 . thus both @xmath158 and @xmath146 are correctable provided that @xmath366^{d-1 } < d.\end{aligned}\ ] ] reasoning as in the proof of lemma [ lem : subsystem - hypercube ] , we conclude that : _ ( holographic principle for local stabilizer codes ) _ [ lem : stabilizer - hypercube ] for a @xmath0-dimensional local stabilizer code with interaction range @xmath165 and distance @xmath166 , a hypercube with linear size @xmath157 is correctable if @xmath367^{d-1 } < d.\ ] ] to ensure that the hypercube @xmath63 is correctable , it suffices for its @xmath368-thickened boundary , rather than its @xmath168$]-thickened boundary , to be smaller than the code distance . here we give a simple example illustrating that for some quantum codes a noncorrectable set need not support a nontrivial logical operator . for @xmath369 qubits , consider the three - dimensional code space spanned by the orthogonal vectors @xmath370 this is the eigenspace with eigenvalue 1 of the projector @xmath371 if the first qubit is mapped to @xmath372 , then @xmath373 is no longer perfectly distinguishable from @xmath189 or @xmath374 ; hence erasure of this qubit is not correctable . ( similarly , the second qubit is also a noncorrectable set . ) is there a logical operator supported on the first qubit ? suppose that @xmath375 is an operator acting on the first qubit . then @xmath376 is a code vector only if @xmath377 , and @xmath378 is a code vector only if @xmath379 . furthermore , if @xmath380 , then @xmath381 is a code vector only if @xmath382 . thus @xmath9 is a multiple of the identity , a trivial operator . s. bravyi , d. poulin , and b. terhal , tradeoffs for reliable quantum information storage in 2d systems , phys . rev . lett . 104 , 050503 ( 2010 ) , arxiv:0909.5200 . s. bravyi and b. terhal , no - go theorem for two - dimensional self - correcting quantum memory based on stabilizer codes , new j. phys . 11 , 043029 ( 2009 ) , arxiv:0810.1983 . s. bravyi , subsystem codes with spatially local generators , arxiv:1008.1029 ( 2010 ) . a. kay and r. colbeck , quantum self - correcting stabilizer codes , arxiv:0810.3557 ( 2008 ) . e. dennis , a. kitaev , a. landahl , and j. preskill , topological quantum memory , j. math . 43 , 4452 - 4505 ( 2002 ) , arxiv : quant - ph/0110143 . r. alicki , m. horodecki , p. horodecki , and r. horodecki , on thermal stability of topological qubit in kitaev s 4d model , open syst . 17 , 1 ( 2010 ) , arxiv:0811.0033 . j. haah , local stabilizer codes in three dimensions without string logical operators , phys . a 83 , 042330 ( 2011 ) , arxiv:1101.1962 . s. bravyi and j. haah , on the energy landscape of 3d spin hamiltonians with topological order , phys . 107 , 150504 ( 2011 ) , arxiv:1105.4159 . s. bravyi and j. haah , analytic and numerical demonstration of quantum self - correction in the 3d cubic code , arxiv:1112.3252 ( 2011 ) . c. castelnovo and c. chamon , topological order in a 3d toric code at finite temperature , phys b 78 , 155120 ( 2008 ) , arxiv:0804.3591 . b. yoshida and i. l. chuang , framework for classifying logical operators in stabilizer codes , phys . a 81 , 052302 ( 2010 ) , arxiv:1002.0085 . a. r. calderbank , e. m. rains , p. w. shor , and n. j. a. sloane , quantum error correction and orthogonal geometry , phys . lett 78 , 405 - 408 ( 1997 ) , arxiv : quant - ph/9605005 . d. gottesman , class of quantum error - correcting codes saturating the quantum hamming bound , phys . a 54 , 18621868 ( 1996 ) , arxiv : quant - ph/9604038 . d. bacon , operator quantum error - correcting subsystems for self - correcting quantum memories , phys . a 73 , 012340 ( 2006 ) , arxiv:/quant - ph/0506023 . d. poulin , stabilizer formalism for operator quantum error correction , phys . lett 95 , 230504 ( 2005 ) , arxiv : quant - ph/0508131 . a. r. calderbank and p. w. shor , good quantum error - correcting codes exist , phys . rev . a 54 , 10981105 ( 1996 ) , arxiv : quant - ph/9512032 . a. steane , multiple particle interference and quantum error correction , proc . a a452 , 25512577 ( 1996 ) , arxiv : quant - ph/9601029 . m. wilde and d. fattel , nonlocal quantum information in bipartite quantum error correction , quant . inf . proc . 9 , 591 - 610 ( 2010 ) , arxiv:0912.2150 . d. gottesman , an introduction to quantum error correction and fault - tolerant quantum computation , arxiv:0904.2557 ( 2009 ) , and references therein . a. yu . kitaev , fault - tolerant quantum computation by anyons , ann . 303 , 2 - 30 ( 2003 ) , arxiv : quant - ph/9707021 . l. lamata , j. len , d. prez - garcia , d. salgado , and e. solano , sequential implementation of global quantum operations , phys . rev . lett . 101 , 180506 ( 2008 ) , arxiv:0711.3652 . o. landon - cardinal and d. poulin , unpublished ( 2012 ) . s. bravyi and m. b. hastings , a short proof of stability of topological order under local perturbations , arxiv:1001.4363 ( 2010 ) .

euv spectroscopic observations often reveal patches of blueshifted emission at boundaries of some active regions ( ars , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? these authors used a single gaussian fit to approximate coronal emission line profiles and derived a blue shift of the order of 20 km s@xmath0 . large line widths have also been found in these blueshifted regions ( e.g. , * ? ? ? there are suggestions that these blue shifts are indicators of the nascent slow solar wind @xcite . recent investigations have revealed the ubiquitous presence of blueward asymmetries of emission lines formed in the solar corona and transition region ( e.g. , * ? ? ? * ; * ? ? ? the enhancement in the blue wings of line profiles are most prominent in loop footpoint regions , such as ar boundaries @xcite . the discovery of these asymmetric line profiles suggests the presence of a highly blue - shifted ( much larger than 20 km s@xmath0 ) emission component besides the primary emission component , and thus provides a significant challenge to previous results and interpretations which are based on a single gaussian fit . @xcite and @xcite suggested that this secondary emission component is associated with type - ii spicules @xcite or rapid blue - shifted events ( rbes , * ? ? ? * ; * ? ? ? * ) in the chromosphere , and that they play an important role in replenishing the corona with hot plasma ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? the so - called rb ( red - blue ) asymmetry analysis @xcite is based on a comparison of the two wings at same velocity ranges , and detailed analysis indicates that both the speed ( 50 - 150 km s@xmath0 ) and the relative intensity ( a few percent of the total emission ) of the secondary component are roughly consistent across a temperature range from 100,000 to several million degrees @xcite . through joint imaging and spectroscopic observations of the corona , @xcite and @xcite have suggested that the secondary emission component is caused by high - speed repetitive upflows in the form of upward propagating disturbances in euv and x - ray imaging observations . such disturbances were previously interpreted as slow mode magnetoacoustic waves ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) or slow - speed solar wind outflows ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? there is also a suggestion that they are warps in two - dimensional sheet - like structures @xcite . besides the rb asymmetry analysis , the double gaussian fit technique has also been used to resolve the second emission component from line profiles @xcite . @xcite and @xcite applied a rb guided double gaussian fit to spectra with a high signal - to - noise ratio and obvious blueward asymmetry . they used the velocity derived from the rb asymmetry analysis as an initial guess of the velocity of the secondary component . the algorithm undertakes a global minimization of the difference between the observed spectrum and the fit by allowing both primary and secondary component centroids to move by one spectral pixel ( @xmath130 km s@xmath0 ) to the blue or red of the initial positions . the speed of the secondary component was found to be @xmath160 km s@xmath0 in @xcite and @xmath1100 km s@xmath0 in @xcite . they both found that the widths of the two components are similar . the widths of the two gaussian components were forced to be the same in the double gaussian fit algorithm of @xcite , who mentioned that the velocity of the secondary component is often as high as 200 km s@xmath0 and that the primary component is also blue - shifted by @xmath110 km s@xmath0 . however , they claimed that the double gaussian fit is a good approximation for only the fe 195.12 and fe 202.02 lines and that other lines are fine with a single gaussian fit . by applying a completely free double gaussian fit to spectral profiles of the fe 284 line , @xcite found that the secondary component usually contributes 10% to 20% to the total emission , is usually blue - shifted by @xmath140 km s@xmath0 and twice as broad as the primary component , especially in loop footpoint regions . however , different settings of initial values and allowable ranges of the seven free parameters can lead to very different fitting results , and thus reasonable initial values and constraints to some free parameters are highly desirable when performing the double gaussian fit . here we slightly modify the initially developed technique of rb asymmetry analysis and apply it to both artificial and observed spectra by the euv imaging spectrometer ( eis , * ? ? ? * ) onboard _ hinode _ , and demonstrate that the speed of the secondary component usually does not reach 200 km s@xmath0 and that the widths of the two components are comparable . we then use parameters determined from the rb asymmetry analysis to guide the double gaussian fit to each spectrum that is observed to have an obvious blueward asymmetry . both the rb asymmetry analysis and the double gaussian fit yield consistent results . we also use imaging observations simultaneously performed by the atmospheric imaging assembly ( aia , * ? ? ? * ) onboard the solar dynamics observatory ( _ sdo _ ) to demonstrate that the propagating disturbances at ar edges are plasma upflows which are indeed responsible for the blueward asymmetries of line profiles . the technique of rb asymmetry analysis was first introduced by @xcite . in our previous work @xcite , we first interpolated the line profile to a spectral resolution ten times greater than the original one , then subtracted the blue wing emission integrated over a narrow spectral range from that at the same position and over the same range in the red wing . the range of integration was then sequentially stepped outward from the line centroid to build an rb asymmetry profile ( simply rb profile ) . the rb asymmetry for an offset velocity @xmath2 can be expressed as the following @xcite : @xmath3 where @xmath2 , @xmath4 , and @xmath5 represent the velocity from the line centroid , the velocity range over which the rb asymmetry is determined , and the spectral intensity , respectively . the value of @xmath4 is usually set as 20 km s@xmath0 . in our previous work , we used the single gaussian fit to determine the line centroid . the resulting rb profile can be normalized to the peak intensity derived from the single gaussian fit . here we slightly modify this technique by using the spectral position corresponding to the peak intensity as the line centroid . the resulting rb profile is then normalized to the peak intensity of the observed line profile . in the following the rb profiles all refer to normalized rb profiles . the originally defined and modified techniques are designated as rb@xmath6 and rb@xmath7 , respectively . in the following we apply these two techniques to artificial spectra composed of two gaussian components to test the ability of the rb asymmetry analysis to resolve the secondary component . similar to @xcite , each artificial spectrum consists of a fixed primary component and a much weaker secondary component . however , to mimic the observed eis spectra , the gaussian parameters we used here are very different from those in @xcite . the spectral pixel size is set as 24 km s@xmath0 , similar to that of the eis spectra at around 274 . the relative peak intensity of the secondary component with respect to the primary one can be set as any value below 30% , while in figure fig.1 we only show the case of 20% . the gaussian width ( 1/e width ) of the primary component is set to be 56 km s@xmath0 , a value comparable to the coronal line width ( including instrumental , thermal , and non - thermal width ) observed by eis . for the secondary component , three values of the width are chosen ( 28 , 56 , 112 km s@xmath0 ) . the spectral position of the secondary component is shifted by -20 , -40 , -60 , ... , -180 km s@xmath0 with respect to the major component and they are shown as different colors in figure fig.1 . after applying the rb@xmath6 and rb@xmath7 techniques to each artificial spectrum , we obtain rb asymmetry profiles of each spectrum . the same colors are used in the rb asymmetry plots of figure fig.1 . in the third and fourth rows of figure fig.1 we plot the peak intensity , velocity and 1/e width derived from the rb analysis as a function of the velocity offset of the secondary component . as noticed by @xcite , when the width of the secondary component is considerably smaller ( 28 km s@xmath0 ) than that of the primary one and the secondary component is centered at velocities lower than the gaussian width of the primary component , the rb@xmath6 asymmetry profile is blueward at low velocities and redward at higher velocities . the asymmetry profile reverses when the offset velocity of the secondary component is larger than the width of the primary component . here we find that the rb@xmath7 asymmetry profile is redward at low velocities and blueward at higher velocities when the secondary component is too close to the primary one . as the offset velocity increases to a value larger than the width of the primary component , the rb@xmath7 asymmetry profile is blueward at all velocities . we find that when the velocity offset is larger than the width of the primary component , the velocity , width , and relative intensity derived from the rb@xmath7 technique are very close to the true velocity and width of the secondary component . the rb@xmath6 technique does slightly worse but still can give relatively accurate values of the parameters . when the widths of two components are the same ( 56 km s@xmath0 ) , no matter where the secondary component is centered , the rb@xmath6 asymmetry profile is redward at low velocities and blueward at higher velocities , and the rb@xmath7 asymmetry profile is blueward at all velocities . by comparing the velocity , width , and relative intensity derived from the rb techniques with the true parameters of the secondary component , we can still conclude that the rb techniques can provide relatively accurate estimate of the properties of the secondary component provided the offset velocity is larger than the width of the primary component . we notice that compared to the rb@xmath6 technique , the rb@xmath7 technique has a better ability to reproduce the parameters of the secondary component , especially when the offset velocity is smaller than twice the width of the primary component . the improvement is around 10 km s@xmath0 in the velocity and width , and 4% in the relative intensity . when the secondary component is considerably broader ( 112 km s@xmath0 ) than that of the primary one , the rb profiles show morphologies similar to the case of 56 km s@xmath0 . however , both the rb@xmath6 and rb@xmath7 techniques can not reproduce an offset velocity smaller than 100 km s@xmath0 . the velocity derived from rb analysis is always larger than 100 km s@xmath0 if the secondary component is twice as broad as the primary one . there is also a large discrepancy between the width derived from rb analysis and the true width of the secondary component if the offset velocity is smaller than twice the width of the primary component ( 112 km s@xmath0 ) . we notice that in all the three cases , the rb techniques can give an extremely accurate estimate of the velocity when the offset velocity of the secondary component is larger than 112 km s@xmath0 . we have varied the intensity ratio of the two components in the range of 5 - 30% and obtain basically the same results , except for the different values of the calculated relative intensities . one thing we have noticed from figure fig.1 is that the rb technique tends to underestimate the value of the relative intensity of the secondary component at smaller offset velocities ( e.g. , 56 - 112 km s@xmath0 ) . this is especially the case for the rb@xmath6 technique which we used in our previous work . from the third row of figure fig.1 we can see that in the velocity range of @xmath156 - 112 km s@xmath0 the relative intensity of the secondary component recovered from the rb@xmath7 technique is much closer to the real value ( -0.20 in figure fig.1 ) , as compared to the rb@xmath6 technique . the underestimation of relative intensity might be improved by using the chi - square method @xcite . the fact that the calculated rb velocities and relative intensities obviously deviate from the real values when the offset velocities of the secondary component are smaller than the width of the primary component has an important implication for the development of future euv spectrographs . the eis instrument has a large instrumental width of @xmath135 km s@xmath0 ( e.g. , * ? ? ? * ) , which contributes almost half of the total line width . the interface region imaging spectrograph ( iris ) , which is scheduled to be launched at the end of 2012 , will have a much smaller instrumental width ( a gaussian width of @xmath15 km s@xmath0 at 1400 ) and our rb technique should be able to accurately resolve the secondary component at smaller offset velocities . the very high spectral resolution ( @xmath13 km s@xmath0 ) of the iris spectrograph will greatly reduce the error caused by the interpolation of line profiles . in the following we pick out two ars and investigate properties of the secondary component from the spectra obtained by eis . since the modified rb technique , rb@xmath7 , has a better ability to accurately resolve the blueshifted secondary component as compared to the originally defined rb@xmath6 technique , here we mainly present results by appling the rb@xmath7 technique instead of the rb@xmath6 technique to the real data . a comparison between the rb@xmath7 and rb@xmath6 results for the 2010 september 16 observation ( see the details of this observation in section 4 ) is presented in appendix a. the @xmath8 slit of eis was used to scan ar noaa 10938 from 18:12 to 20:27 on 2007 january 18 , with an exposure time of 30 s and a step size of @xmath9 . this data was previously analyzed by @xcite and @xcite for the purpose of investigating properties of asymmetric line profiles . @xcite mainly used the fe 274.20 and fe 284.16 lines which are formed around a temperature of 2 mk , while @xcite only focused on the fe 284.16 line . here we choose both lines for our study , but focus on the fe 274.20 line . @xcite mentioned that the fe 284.16 line is blended with al 284.03 and that the latter usually contributes no more than 5% to the deviation from a single gaussian profile . however , as we will discuss later , in some locations the blend can greatly complicate the asymmetries of line profiles . the fe 274.20 line is also a strong line and it is blended with si 274.18 . the blend is much weaker than fe 274.20 and can safely be ignored in active region conditions @xcite . moreover , this blend is very close ( less than 1 spectral pixel ) to the line center of fe 274.20 and thus it should not have an important influence on the results of our rb asymmetry analysis and gaussian fit . further confidence is given by the highly similar behavior of the line moments and profile asymmetries between fe 274.20 and the weaker fe 264.78 line in most ars we analyzed ( see appendix b ) , although the fe 264.78 line is usually noisier than fe 274.20 . the ssw routine _ eis_prep.pro _ was applied to correct and calibrate the eis data . this includes ccd pedestal and dark current subtraction , cosmic ray removing , warm and hot pixels identification , absolute calibration , error estimation , and so on . the effects of slit tilt and orbital variation ( thermal drift ) were estimated by using the ssw routine _ eis_wave_corr.pro _ and removed from the data . after that , a running average over 5 pixels along the slit and 3 pixels across the slit was applied to the spectra to improve the signal to noise ratio . as a common practice , a single gaussian fit was applied to each spectrum and figure fig.2(a)-(c ) show the spatial distributions of the gaussian parameters for fe 274.20 . figure fig.2(d ) shows the map of reduced @xmath10 for the fit . the reduced @xmath10 is defined as the following ( e.g. , * ? ? ? * ; * ? ? ? * ) : @xmath11 where @xmath12 , @xmath13 , and @xmath14 denote the observed spectral radiance , fitted spectral radiance , and measurement error calculated by using the ssw routine _ eis_prep.pro _ ( mainly poisson error ) , respectively . here @xmath15 represents the spectral position and the summation is performed over all the @xmath16 spectral positions . the degree of freedom is given by @xmath17 and equals to 4 or 7 for a single or double gaussian fit . we assume zero shift of the profile averaged over the entire observation region . we can see that the loop footpoint regions , or boundaries of the ar , are characterized by a blueshift of @xmath120 km s@xmath0 and an enhancement of the line width , a well - known phenomenon in the hinode era ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? however , the goodness of the fit , the @xmath10 is also clearly enhanced mainly in loop footpoint regions , indicating an obvious deviation from a single gaussian profile there ( e.g. , * ? ? ? * ; * ? ? ? note that the definition of @xmath10 in @xcite is not correct and that it is probably a typographic error . we applied the modified rb technique rb@xmath7 to each spectrum and calculated the average relative intensity in the velocity interval of 60 - 110 km s@xmath0 from the rb asymmetry profile . figure fig.2(h ) shows the spatial distribution of this average rb asymmetry . here a negative / positive value indicates an enhancement of the blue / red wing in this velocity interval . we can see that the blueward asymmetry is most prominent at loop footpoint regions , generally coincident with the blueshift , width enhancement , and @xmath10 enhancement . for further analysis , we only selected those locations where this average rb asymmetry is smaller than -0.05 ( obvious blueward asymmetry ) and the signal to noise ratio of the profile ( defined as the ratio of the peak and background intensities ) is larger than 5 . we have to mention that such a cutoff excludes the analysis of spectra in the ar core and edges where the magnetic field lines largely incline with respect to the line of sight , since an inclination of the magnetic field line usually leads to a smaller offset velocity of the secondary component in the line of sight direction , which would reduce the value of the relative intensity derived from the rb technique ( see figure fig.1 ) . figure fig.2(e)-(g ) show the peak relative intensity , velocity , and 1/e width derived from rb@xmath7 asymmetry profiles at these locations . the distributions of these three parameters are presented as the red histograms in figure fig.6(a)-(c ) . in figure fig.3 we show the scatter plots of the relationship between doppler shift / gaussian width derived from the single gaussian fit and the peak relative intensity derived from rb@xmath7 asymmetry profiles for fe 274.20 . here only data points with blue shift at those selected locations are shown . a clear correlation is found in each of the two panels in figure fig.3 . such a correlation strongly suggests that the clear blue shift and enhanced line width in the loop footpoint regions are related to , or caused by the blueward asymmetries . we can simply imagine a faint high - speed upflow superimposed on a strong and almost stationary ( or slightly shifted ) background in the line of sight . such a scenario would naturally lead to a blueward asymmetric line profile . and a single gaussian fit to the total emission line profile would result in a blueshift and enhanced line width , as compared to the line profile of the background emission - a similar conclusion was reached by @xcite . so it is clear that the blue shift at ar boundaries , which was reported in many previous investigations @xcite , is actually a composite effect of at least two emission components and can not reflect the real physical process @xcite . some authors claim that the blue shifts derived from the single gaussian fit at ar boundaries are indicators of the nascent slow solar wind @xcite . if this is the case , the wind speed must have been largely underestimated if the source of the mass supply to the solar wind is the secondary component , or overestimated if the primary component dominates the mass supply to the solar wind . our results also suggest that the enhanced line width can be the result of the superposition of different emission components with different velocities and is not necessarily related to the increase in the alfvn wave amplitude @xcite . we think that both the inhomogeneities of flow velocities and alfvn waves may contribute to the observed non - thermal width of an emission line , and that caution must be taken when interpreting the non - thermal width as purely cased by alfvn waves ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) . from figure fig.6(a ) we can see that the distribution of the intensity ratio ( relative intensity of the secondary component ) peaks around 0.07 but can reach as high as 0.3 . this is generally consistent with previous result that the blueward excess emission is often a few percent of the total emission , which is based on the rb@xmath6 technique ( e.g. , * ? ? ? * ; * ? ? ? * ) . while @xcite found a relative intensity of some 10% to 20% based on his free double gaussian fit . as mentioned above , when the width of the secondary component is not considerably larger than that of the primary one , the rb@xmath7 technique can provide an accurate estimate of the gaussian parameters of the secondary component provided the offset velocity is larger than the width ( @xmath156 km s@xmath0 ) of the primary component . as seen from figure fig.6(b ) , the rb@xmath7 velocity seldom reaches 100 km s@xmath0 and its distribution peaks at @xmath175 km s@xmath0 . such a result would exclude the possibility of a very broad secondary component , as claimed by @xcite based on a free double gaussian fit this is because the velocity derived from rb analysis is always larger than 100 km s@xmath0 when the secondary component is twice as broad as the primary one ( see figure fig.1 ) . in figure fig.6(c ) we can see that the rb@xmath7 width is usually smaller than 60 km s@xmath0 and larger than 35 km s@xmath0 , and that its distribution peaks around 50 km s@xmath0 , which is also not consistent with what we expect if the secondary component is twice as broad as the primary one . from the last panel of figure fig.1 we can see that the rb@xmath7 width is always larger than 60 km s@xmath0 in such a case . from figure fig.1 we can also see that when the width of the secondary component is considerably smaller ( 28 km s@xmath0 ) than that of the primary one , the rb@xmath7 width is always smaller than 35 km s@xmath0 , which is not consistent with the observed histogram in figure fig.6(c ) . these inconsistencies strongly suggest that the secondary component can not be considerably broader or narrower than the primary one . the widths of the two components should be comparable . this conclusion implies that the assumption of the same width of the two components in the double gaussian fit algorithm of @xcite is reasonable . the velocity distribution presented in figure fig.6(b ) is not consistent with the scenario of a very high - speed ( @xmath1200 km s@xmath0 ) secondary component , as suggested by @xcite after analyzing another data set . as seen from figure fig.1 , the velocity derived from the rb@xmath7 technique is extremely close to ( usually less than 5 km s@xmath0 ) the true velocity of the secondary component when the latter is larger than 112 km s@xmath0 . if very large velocities exist , the rb@xmath7 technique should yield such large numbers of velocities . however , the observed velocity derived by using the rb@xmath7 technique seldom reaches higher than 150 km s@xmath0 , indicating that the velocity of the secondary component in our observation can not be as high as 200 km s@xmath0 . we have to mention that the magnitude of velocity also depends on the viewing angle . however , in loop footpoint regions usually a large portion of the magnetic field lines are almost vertical so that the line of sight effect should not be very significant at disk center . we also applied the rb asymmetry analysis to the data analyzed by @xcite and found larger velocities ( @xmath195 km s@xmath0 ) of the secondary component in that ar ( see appendix c ) . but still the velocity seldom reaches 200 km s@xmath0 . following the initial idea of using the rb@xmath6 analysis result to guide double gaussian fit @xcite , we designed a similar but modified algorithm . the peak intensity , velocity , and width derived from a single gaussian fit were used as the initial guess of those parameters of the primary gaussian component . the single gaussian background was also used as the initial value of the double gaussian background . meanwhile , we used the peak intensity and velocity obtained through the rb@xmath7 analysis as the initial values of the same parameters of the secondary gaussian component . the width of the secondary component was initially set as the same as the primary component . during the iterations , we allowed all the three intensities ( peak intensities of the two components and the background intensity ) to vary within the range of 75%-125% of the corresponding initial values . the velocity was allowed to move to the blue or red of the initial position by one spectral pixel ( @xmath124 km s@xmath0 at 274 ) for the primary component and two pixels for the secondary component . we also allowed the width to increase or decrease by one spectral pixel for the primary component and two pixels for the secondary component . the algorithm undertook a global minimization of the difference between the observed and fitted spectrum . we present several examples of the observed and fitted profiles of both the fe 274.20 and fe 284.16 lines in figure fig.4 . the profiles in figure fig.4(a ) & and ( b ) are the profiles averaged respectively over region 1 and 2 marked in figure fig.2 . by comparing the observed profiles with the different fitting profiles , we can clearly see the better performance of the double gaussian fit and the deviations of the observed profiles from the single gaussian fits . after the double gaussian fit , we took the velocity of the primary component as the line centroid and calculated the rb asymmetry profile ( here rb@xmath18 ) . the lower panels of figure fig.4 show the three rb asymmetry profiles ( rb@xmath6 , rb@xmath7 , rb@xmath18 ) . we can see that rb@xmath7 and rb@xmath18 show similar behaviors and values of the three parameters are also very close . it is known that the fe 284.16 line is blended with al 284.03 ( e.g. , * ? ? ? * ; * ? ? ? the latter is sitting at @xmath1150 km s@xmath0 away from the spectral position of fe 284.16 so that this blend is potentially contaminating the blue wing of the line and may complicate the rb analysis and gaussian fit . for example , the line profiles in both figure fig.4(a ) and ( b ) were obtained around loop footpoint regions ( marked in figure fig.2 ) . the magnitude of asymmetry is comparable in both the fe 274.20 and fe 284.16 line profiles shown in figure fig.4(a ) , indicating that the blend al 284.03 can be ignored . however , in figure fig.4(b ) we see a @xmath12% blueward asymmetry in fe 274.20 but a @xmath116% blueward asymmetry in fe 284.16 . since the two lines have similar formation temperatures , the completely different magnitude of asymmetries in the two lines strongly suggests that the bump at the blue wing of fe 284.16 is dominated by the blend al 284.03 . therefore , we focus on the fe 274.20 line instead of fe 284.16 which was used by @xcite . we then applied the rb@xmath7 guided gaussian fit and rb@xmath18 analysis to all spectra of fe 274.20 with obvious blueward asymmetries and large signal to noise ratio . we present the spatial distributions of the three gaussian parameters of both components in figure fig.5 . in addition , the intensity ratio and velocity difference of the two components are also shown in figure fig.5 . we found that in most locations the primary component is slightly blue shifted ( @xmath110 km s@xmath0 ) , a result also found by @xcite , who interpreted it as the nascent slow solar wind from ar boundaries @xcite . while for the secondary component , the velocity is largely blue shifted ( @xmath180 km s@xmath0 ) at almost all locations . the maps of parameters derived from rb@xmath18 asymmetry profiles are very similar to those derived from rb@xmath7 and thus are not shown here . histograms of the intensity ratio ( relative intensity ) , velocity difference ( velocity ) , and gaussian width of the secondary component as derived from double gaussian fit and rb asymmetry analysis ( rb@xmath7,rb@xmath18 ) are presented in figure fig.6 . we can see that they are very similar , especially rb@xmath7 and rb@xmath18 . the intensity ratio peaks around 0.07 and can also reach as high as 0.3 . for rb@xmath7 and rb@xmath18 , the velocity and width both have a narrow distribution which peaks at @xmath175 km s@xmath0 and @xmath150 km s@xmath0 , respectively . while for the double gaussian fit , most of the width values are still much smaller than @xmath1100 km s@xmath0 and the width distribution peaks slightly higher . the velocity difference is usually in the range of 50 - 100 km s@xmath0 , but the double gaussian fit algorithm also produces some relatively small values of velocity difference ( 10 - 50 km s@xmath0 ) . however , these relatively small blue shifts are found at only a few locations and they are not prevalent . the @xmath10 value of the double gaussian fit is usually smaller than unity - similar result was also found by @xcite , who attributed it to the overestimation of the measurement error . from figure fig.6(d ) we can see that the @xmath10 ratio between the double and single gaussian fit is usually less than 0.5 in the regions characteristic of obvious blueward asymmetries . such a result indicates that the double gaussian fit does at least two times better than the single gaussian fit for these asymmetric line profiles . it is likely that even more components , e.g. , a component from cooling downflows , might be present in the observed emission . however , a reliable decomposition of these additional components can not be made due to the relatively large instrumental width and modest spectral resolution of eis . with the upcoming iris data , we may be able to resolve more coronal emission components . the other observation was performed by eis on 2010 september 16 , with simultaneous observations by aia . the eis instrument scanned noaa ar 11106 from 10:38 to 11:57 ut . the @xmath19 slit was used for the observation , with a 60 s exposure . after standard correction and calibration of the eis data , a running average over 5 pixels along the slit and 3 pixels across the slit was applied to the spectra to improve the signal to noise ratio . four relatively strong and clean emission lines in the spectral window were selected for our study : si 275.35 , si 258.37 , fe 202.04 , and fe 274.20 . their formation temperatures are log(_t_/k)=5.8 , 6.1 , 6.2 , and 6.3 , respectively @xcite . but we will mainly concentrate on the hottest strong line , fe 274.20 . a single gaussian fit was first applied to each spectrum to derive the line intensity , centroid and gaussian width . by assuming the average doppler shift of each line is zero over the entire region , we calculated the doppler shift from the line centroid for each spectral profile . maps of the gaussian parameters and @xmath10 for fe 274.20 are presented in figure fig.7(a)-(d ) . again , we clearly see that the loop footpoint regions , or boundaries of the ar , are characterized by a blueshift of @xmath120 km s@xmath0 and an enhancement of the line width . the deviation from a single gaussian profile , which is quantified by the enhancement of @xmath10 , is also clearly seen in the loop footpoint regions . the si 258.37 and fe 202.04 lines reveal basically the same structures in the maps of gaussian parameters . the cool line si 275.35 shows prominent redshifts in the fan regions ( not shown here ) , a phenomenon found also by @xcite . such redshifts could be the downflows after cooling of the high - speed upflows and evidence for the mass circulation in coronal loops @xcite . similarly to before , we applied the rb@xmath7 technique to each spectrum and calculated the average relative intensity in the velocity interval of 60 - 110 km s@xmath0 from the rb asymmetry profile . figure fig.7(h ) shows the spatial distribution of this average rb asymmetry . it is clear that the blueward asymmetry is most prominent at the two edges of the ar , generally coincident with the enhancement of blueshift , line width , and @xmath10 . however , the blueward asymmetry is clearly not only present at the ar edges , but also existing in other plage / loop footpoint regions . for example , the tongue - like patch ( x=@xmath20,y=@xmath21 ) of blueward asymmetries in figure fig.7(h ) seems to be coincident with moss and it is not a typical outflow region . the fact is that simple magnetic field structures which are aligned with the line of sight are often not present in ar cores so that the secondary component there is often not revealed as clear blueward asymmetries of line profiles . an obvious correlation between the doppler shift / gaussian width derived from single gaussian fit and peak relative intensity derived from rb@xmath7 asymmetry profiles is found in figure fig.8 , suggesting that the blueshift and enhanced line width are caused by the high - speed secondary emission component and that the single gaussian fit can not accurately reflect the real physical process here . in the following , only those pixels with obvious blueward asymmetry ( the average rb asymmetry in 60 - 110 km s@xmath0 smaller than -0.05 ) and significant signal to noise ratio ( larger than 5 ) were selected . clearly , such a cutoff excludes the analysis of spectra in the ar core and edges where the magnetic field lines make large angles relative to the line of sight . the peak relative intensity , velocity , and 1/e width derived from rb@xmath7 asymmetry profiles of fe 274.20 at these pixels are shown in figure fig.7(e)-(g ) . the distributions of these three parameters are shown as the red histograms in figure fig.11(a)-(c ) and they are very similar to those in figure fig.6 . the only obvious difference is perhaps the slightly larger peak velocity and gaussian width . in the 2007 january 18 observation , the velocity and width peak at @xmath175 km s@xmath0 and @xmath150 km s@xmath0 , respectively . but in the 2010 september 16 observation , the velocity peaks at @xmath185 km s@xmath0 and the width peaks at @xmath160 km s@xmath0 . the major part of the velocity distribution is still on the left side of 100 km s@xmath0 in the 2010 september 16 observation , which is not consistent with the scenario of a considerably broader secondary component @xcite . the scenario of a considerably narrow secondary component can also be excluded since the rb@xmath7 width is almost always larger than 35 km s@xmath0 . similar to the 2007 january 18 observation , the rb@xmath7 velocity in this recent observation also seldom reaches higher than 150 km s@xmath0 , indicating that the line of sight velocity of the secondary component is unlikely to be as high as 200 km s@xmath0 in this observation . so with the rb asymmetry analysis ( particularly rb@xmath7 ) , we can conclude that the secondary emission component , which is usually less than 20% and often a few percent of the primary one in intensity , has a speed around @xmath180 km s@xmath0 and a width comparable to the primary component . the same algorithm of rb@xmath7 guided double gaussian fit was adopted to the asymmetric profiles obtained in the 2010 september 16 observation . several examples of the observed and fitted line profiles are presented in figure fig.9 . in figure fig.9(a ) we show profiles of four emission lines averaged over region 1 which is marked in figure fig.7 . blueward asymmetries and deviations from single gaussian profiles are clearly present in all of the four lines . the double gaussian algorithm yields a reasonably good fit to the observed spectra . this spatial average is necessary because individual profiles of the si 275.35 and si 258.37 lines are too noisy to allow reliable gaussian fits and rb asymmetry analysis in the ar boundaries . we find that the relative intensity and velocity of the secondary component stay relatively stable across the temperature range of log(_t_/k)=5.8 - 6.3 , consistent with the scenario of multi - thermal high - speed upflows , as suggested by our previous work @xcite . the clear presence of blueward asymmetry of the cool si 275.35 line is consistent with the finding of @xcite . compared to other lines , the si 275.35 line reveals a smaller velocity ( @xmath165 km s@xmath0 ) of the secondary component . one possible reason could be that the si 275.35 line reflects emission from much cooler plasmas so that its line profiles are complicated by cooling downflows ( see also * ? ? ? fe 274.20 line profiles at four single pixels ( without spatial average ) are presented in figure fig.9(b)-(e ) . the secondary component resolved by the double gaussian fit has different relative intensity and velocity . the magnitude of velocity is likely to be related to the viewing angle of the upflows . if the magnetic field lines associated with the upflows make a smaller angle with respect to the line of sight , the observed velocity should be closer to the real velocity . the location of the profile presented in figure fig.9(b ) is in a weak emission region surrounded by hot loops . this weak emission region might be related to less compact ( larger ) loops or open field lines which are usually more vertical at lower heights compared to compact and smaller loops . this may explain the large value of velocity ( @xmath1120 km s@xmath0 ) . the gaussian widths of the two components are generally comparable in all these four cases . the rb@xmath18 profiles were then calculated by taking the velocity of the primary component as the line centroid . in figure fig.9 we also show the three rb asymmetry profiles ( rb@xmath6 , rb@xmath7 , rb@xmath18 ) for the corresponding spectra . the parameters of the secondary component obtained through rb@xmath7 and rb@xmath18 techniques are usually close to those obtained by using the double gaussian fit . the rb@xmath7 guided double gaussian fit and rb@xmath18 analysis were then applied to all spectra of fe 274.20 with obvious blueward asymmetries . figure fig.10 shows the spatial distributions of the three gaussian parameters of both components as well as the intensity ratio and velocity difference of the two components . again , we found that the primary component is usually slightly blue shifted by @xmath110 km s@xmath0 in this observation . the velocity of the secondary component is blue shifted by much larger values . the maps of parameters derived from rb@xmath18 asymmetry profiles are very similar to those derived from rb@xmath7 , which have been presented in figure fig.7 and thus are not shown here . structures on the maps of rb@xmath7 peak relative intensity , velocity , and width shown in figure fig.7 generally coincide with those on the maps of intensity ratio , velocity difference , and secondary component width as presented in figure fig.10 , indicating that both the rb@xmath7 technique and rb@xmath7 guided double gaussian fit are able to extract the parameters of the secondary component . figure fig.11(a)-(c ) shows the distributions of the intensity ratio ( relative intensity ) , velocity difference ( velocity ) , and gaussian width of the secondary component as derived from double gaussian fit and rb asymmetry analysis ( rb@xmath7,rb@xmath18 ) . for each of the parameters , the three histograms are very similar . all these histograms show no big difference from those presented in figure fig.6 , except the slightly larger peak velocity and width . the velocity of the secondary component is usually in the range of 50 - 120 km s@xmath0 in this observation , which is about twice as high as that suggested by @xcite . the distribution of the @xmath10 ratio between the double and single gaussian fit in figure fig.11(d ) is also very similar to that in figure fig.6(d ) . the ratio is usually less than 0.5 , indicating a significant improvement of the fitting by using the double gaussian fit instead of the single gaussian fit . during the eis scan , the aia instrument took full - disk images of the sun in several euv channels at a cadence of 12 s. we selected images in the passbands of 171 , 193 and 211 for our analysis . the dominant emission of these passbands has a temperature around 0.8 mk , 1.3 mk , and 2 mk , respectively . there were a few missing frames in each passband so that an linear interpolation was applied to the data . the pixel size is about 0.6@xmath22 . we extracted a sub - region enclosing the ar observed by eis and co - aligned different image frames . a movie of aia 193 images and the running difference images is available online ( aia193.mpeg ) . fig.12(a ) shows an image of aia 193 at 12:06 . the rectangle indicates the region scanned by eis . the cross correlation technique was used to coalign aia 193 and eis fe 274.20 . upward propagating disturbances ( pds ) are clearly present in the aia movie . these disturbances are mainly found at the two boundaries of the ar and coincide with obvious blueward asymmetries as seen from figure fig.7(h ) , suggesting a connection of the two . the speed of these disturbances can be estimated by placing a virtual slit along the propagation direction of the disturbances and calculating the slope of a bright strip in the space - time ( s - t ) plot ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? figure [ fig.12](b)-(g ) show the s - t plots of the original and detrended intensities for the dotted line shown in figure fig.12(a ) . the detrended intensities were used to better reveal the faint outflow signatures and they were obtained by first subtracting a 8-minute running average from the original intensity time series and then normalized to the running average at each location of the slit . one example of pd is indicated by the inclined dashed line . the speed of this pd , the slope of the dashed line , is estimated to be 91 km s@xmath0 . as the pds are usually present at different passbands of aia , in principle we should be able to compare the speed of a certain pd at different temperatures using aia data . for a certain virtual slit , we often find similar patterns of the s - t plots ( both the original and detrended intensities ) in the 193 passband and the other two passbands ( particularly the 211 passband ) . in most cases we did not find an obvious increasing trend in the speed with increasing temperature , which seems to support the interpretation of pds as multi - thermal high - speed upflows ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) rather than slow mode waves ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? however , a recent investigation suggests that non - dominant cool ions such as o may contribute significantly to the total emission of the 193 and 211 passbands @xcite . but cool ions such as o v would not be expected to show obvious emission over a large height range . for example , pds are also seen off - limb in a coronal hole up to heights of 50 - 100 mm with aia @xcite , i.e. , heights where emission from cool lines does not typically occur at sufficient strength . this renders the impact of non - dominant cool ions less likely . however , we can not rule out the possibility of slow waves . perhaps both slow waves and upflows are existing in our observations . in fact waves can be exited by high - speed flows . but our eis result that lines with different formation temperatures basically show similar velocities seems to suggest that upflows dominate the emission . we note that for some virtual slits the s - t plots show some differences in different aia passbands , especially between the 171 passband and the other two passbands . the emission in the 171 passband is dominated by fe 171.107 @xcite , which is formed in the upper transition region and thus may have a smaller scale height compared to the hotter emission in the other two passbands . moreover , due to the different sensitivity , signal to noise ratio , and emission contrast in different passbands we may not always be able to clearly identify every outflow event in all the three passbands from the s - t plots . the 211 passband samples plasma with a temperature much closer to the formation temperature of fe 274.20 . however , the 211 images have a lower signal to noise ratio and are thus noisier compared to the 193 images . since the s - t plots in the 211 passband are highly similar to those in the 193 passband , in the following we concentrate on the 193 images . we identified 39 well - isolated disturbances propagating along the fan - like structure at the left boundary of the ar and calculated their propagating speeds in the 193 passband . a speed distribution was then obtained . a comparison of this speed distribution with the distribution of the velocities derived from eis fe 202.04 line profiles ( through double gaussian fit , rb@xmath7 , and rb@xmath18 analysis ) is presented in figure [ fig.13](a ) & ( b ) . here only line profiles with obvious blueward asymmetries in the fan - like structure at the left boundary of the ar were used . by using the potential field source surface ( pfss ) model @xcite , the inclination angles of the magnetic field lines in the fan - like structure at the left boundary of the ar were estimated to be around 55@xmath23 with respect to the line of sight . disturbances propagating at a speed of @xmath24 along these magnetic field lines should exhibit a plane - of - sky speed of @xmath24@xmath25 and a line - of - sight speed of @xmath24@xmath26 . the ratio of the two speed components is 1.43 . since the speeds derived from s - t plots of aia images are the real speeds projected onto the plane of sky and the speeds derived from eis line profiles are the line - of - sight component of the real speeds , the ratio of the two should be around 1.43 . from figure [ fig.13 ] we find that this is indeed the case . the speeds derived by using aia data are systematically larger than those derived by using eis data . the ratio of the two average speeds is around 135/85=1.58 , which is close to 1.43 . for comparison , the rb@xmath7 velocity histogram is multiplied by 1.58 and overplotted as the dotted line in figure [ fig.13](b ) . the small difference between 1.58 and 1.43 could easily be caused by the fact that the pfss model is not expected to give a perfect representation of what viewing angles the real sun presents . we think that in the future 3-d reconstruction ( e.g. , * ? ? ? * ; * ? ? ? * ) could be considered as an alternate method to help relate the velocities derived from imaging and spectroscopic observations . in figure [ fig.13 ] we can also compare histograms of the relative intensities derived from eis fe 202.04 line profiles ( in panel ( c ) , through double gaussian fit , rb@xmath7 , and rb@xmath18 analysis ) and those derived from s - t plots ( d ) . the latter was obtained through the following process : we first smoothed the original time series with a 1-hour kernel . after that we subtracted the smoothed time series from the original one and obtained a time series of intensity change . the relative intensity was then obtained by normalising the maximum intensity change to the average intensity at each slit location . from figure [ fig.13](d ) we can see that the relative intensity is usually a few percent but can also be larger than 10% , which is consistent with magnitude of the relative intensity of the secondary component revealed by rb asymmetry analysis and double gaussian fit . for a better comparison , the rb@xmath7 relative intensity histogram is overplotted as the dotted line in figure [ fig.13](d ) . note that the lack of small values of the relative intensity derived from eis line profiles are due to the fact that we only selected profiles with significant blueward asymmetries . these results provide further support to the argument that the blueward asymmetries of coronal line profiles are caused by the upward propagating disturbances at ar boundaries @xcite . as mentioned by @xcite , although there are striking similarities , it is not clear how the outflows inferred from eis line profiles and the pds seen in euv imaging observations are related . @xcite pointed out that the blue shifts ( from single gaussian fit ) of eis lines are only of the order of 20 - 50 km s@xmath0 and no periodicity has been reported . while the pds in imaging observations have a speed of @xmath1100 km s@xmath0 and they are often quasi - periodic . now we understand that the eis line profile consists of two components and that the secondary component has a velocity of @xmath1100 km s@xmath0 . and we have already demonstrated that the secondary component is also quasi - periodically enhanced or weakened @xcite . the consistency between properties of the pds observed by aia and those of the secondary component revealed by rb asymmetry analysis and double gaussian fit of eis line profiles suggests a close connection of the two . it is thus natural to suggest that the pds are responsible for the blueward asymmetries of line profiles . the non - thermal width of the secondary component is probably caused by both the superposition of multiple unresolved upflows and the alfvenic motions associated with the upflows @xcite . we have to mention that in figure [ fig.13 ] we only present the results for the fan - like structure at the northeast boundary of the ar because of the simple magnetic field structure there . from the online movie ( aia193.mpeg ) we can see that outward propagating disturbances are also clearly present in the fan - like structure to the southwest of the ar . we produced several s - t plots and found generally similar s - t patterns in different passbands . however , the footpoints of these fan loops ( around x=@xmath27 , y=@xmath28 ) are mixed together and the higher parts of different loops are not well - separated , making it difficult to identify a well - isolated propagating disturbance in the s - t plots . in other parts of the ar , magnetic field structures are either too complex or only slightly inclined with respect to the line of sight so that reliable speeds may not be obtained through s - t plots . in conclusion , the presence of asymmetric coronal line profiles in loop footpoint regions suggests that at least two emission components are present : an almost stationary primary component and a high - speed secondary component . the secondary emission component might play an important role in supplying the corona with hot plasma . we have generated artificial spectra composed of two gaussian components to test the ability of extracting the gaussian parameters ( intensity , velocity , and width ) of the secondary component through the originally defined ( use the single gaussian fit to determine the line centroid ) and a slightly modified ( use the spectral position corresponding to the peak intensity as the line centroid ) technique of the rb asymmetry analysis , and find that this technique , especially the modified one , can provide a relatively accurate estimate of the parameters of the secondary component if the velocity is larger and the width is not considerably larger than the primary component width . we have applied the modified technique to the spectra obtained by the euv imaging spectrometer ( eis ) onboard hinode in two active regions and found that the distributions of the velocity / width determined from the rb asymmetry analysis peak around 80 km s@xmath0/60 km s@xmath0 . a comparison of such distributions with the values of velocity / width calculated from the artificial spectra suggests that the secondary component can not be very broad or narrow and that the widths of the two components are comparable . the velocity of the secondary component is usually within the range of 50 - 150 km s@xmath0 and seldom reaches as high as 200 km s@xmath0 or as low as 40 km s@xmath0 in our eis observations . the relative intensity of the secondary component with respect to the primary one is often a few percent but can sometimes reach 30% . our conclusions are very different from those of @xcite , who claimed that the secondary component contributes some 10% to 20% to the total emission , is about twice as broad as the primary component and blueshifted by up to 50 km s@xmath0 . the velocity range of the secondary component we derived here is roughly consistent with , yet slightly smaller than that derived by @xcite . however , we find that the very high speed of 200 km s@xmath0 mentioned by @xcite seems to be not very likely . using the parameters determined from the rb asymmetry analysis and single gaussian fit as initial values for several free parameters , we have performed a double gaussian fit to the spectra that were observed to have obvious blueward asymmetries . the rb asymmetry analysis has been performed for the third time , but then using the centroid of the secondary component determined from the double gaussian fit as the line centroid . the double gaussian fit and the rb asymmetry analyses have given basically consistent results . the properties of the secondary component stay relatively stable across the temperature range of log(_t_/k)=5.8 - 6.3 . the double gaussian fit also shows that the velocity of the primary component is often blue shifted by @xmath110 km s@xmath0 . we have also used imaging observation simultaneously performed by the aia instrument onboard _ sdo _ and demonstrated that the propagating disturbances coincide with obvious blueward asymmetries of line profiles , show an intensity change similar to the relative intensity of the secondary component of line profiles and an average velocity consistent with that derived from emission line profiles . such results suggest that the upward propagating disturbances are plasma upflows responsible for the blueward asymmetries of line profiles . _ sdo _ is the first mission of nasa@xmath29s living with a star ( lws ) program . eis is an instrument onboard _ , a japanese mission developed and launched by isas / jaxa , with naoj as domestic partner and nasa and stfc ( uk ) as international partners . it is operated by these agencies in cooperation with esa and nsc ( norway ) . scott w. mcintosh is supported by nasa ( nnx08al22 g , nnx08ba99 g ) and nsf ( atm-0541567 , atm-0925177 ) . bart de pontieu is supported by nasa grants nnx08al22 g and nnx08ba99 g . hui tian is supported by the asp postdoctoral fellowship program of ncar . the national center for atmospheric research is sponsored by the national science foundation . hui tian thanks marc derosa for his assistance in using the pfss package . we appreciate the efforts of the referee to improve the manuscript . as we mentioned in section 2 , both the rb@xmath7 and rb@xmath6 techniques can provide a relatively accurate estimate of the gaussian parameters of the secondary component when the offset velocity is larger than the gaussian width of the primary component . we have also found that the modified rb technique , rb@xmath7 , has a better ability to accurately quantify the properties of the secondary component as compared to the originally defined rb@xmath6 technique . figure [ fig.s1 ] shows a comparison between the rb@xmath7 and rb@xmath6 results for the 2010 september 16 observation . we can see that the spatial structures in all of the four parameters are very similar . the most prominent difference is the smaller values of relative intensity and width , and larger upflow velocity for rb@xmath6 . such results are fully consistent with our forward modeling results in figure [ fig.1 ] and imply that in our previous work the relative intensity and width might be underestimated while the upflow velocity was likely to be slightly overestimated . we have mentioned that the strong fe 274.20 line is blended with the weak si 274.18 line and that the latter can safely be ignored in active region conditions @xcite . the very close wavelengths ( less than 1 spectral pixel ) of the two also suggest that the blend is very unlikely to have an important impact on the high - speed ( of the order of 100 km s@xmath0 ) upflow . the fe 264.78 line spectra were not obtained in the 2007 january 18 observation so that we can not provide a direct comparison of it with the fe 274.20 line . in 2010 september 16 both fe lines were used in the observation and figure [ fig.s2 ] shows the spatial distributions of the line moments and profile asymmetries . note that the spectra ( especially for the fe 264.78 line ) in the upper left corner , the higher part of the fan , have low signal to noise ratio and thus the line moments and profile asymmetries are noisy there . the very similar behavior of line moments and profile asymmetries between fe 274.20 and the weaker but clean fe 264.78 line provides further confidence that the blend of si 274.18 line can be safely ignored in our analysis . we also applied the techniques of rb asymmetry analysis and double gaussian fit to an eis observation of ar noaa 10978 from 11:43 to 17:03 on 2007 december 12 . @xcite found that the secondary emission component in this ar has velocities around 110 km s@xmath0 and , as they stated in their paper , often as high as 200 km s@xmath0 . @xcite used the fe 195.12 line and they mentioned that the blend of this line , fe 195.18 , should not significantly affect the fe 195.12 line in the low - density @xcite outflow region ( ar edges ) . however , to avoid the complexity introduced by this blend we decided to turn to other clean and strong fe lines . we found that in the ar edges some other lines at the far blue wing of the fe 192.39 line can be relatively strong and thus affect the results of rb asymmetry analysis and double gaussian fit . so here we mainly present the results of the fe 193.51 line , which is clean and about 2/3 as strong as fe 195.12 . since the line is very strong , there is no need to perform a spatial average of the line profiles . figure [ fig.s3 ] shows the spatial distributions of the single gaussian parameters and profile asymmetry . similar to ar 10938 and ar 11106 , the two loop footpoint regions are characterized by an enhancement in the blue shift , line width , and blueward profile asymmetry , and there is an obvious correlation between the doppler shift / gaussian width and profile asymmetry . note that the white strip in each panel mark locations where the observed data is lost . two examples of line profiles and the corresponding rb asymmetry profiles of the fe 193.51 and si 275.35 lines are shown in figure fig.s4 . the profiles were averaged over 3 pixels centered at the locations marked in figure fig.s3(a ) in both spatial dimensions . @xcite mentioned that cool lines like si 275.35 do not have any significant asymmetries and that their profiles can be accurately represented by a single gaussian . we found that this is not always the case . similar to what we found in ar 11106 , at some locations the si 275.35 line profiles clearly exhibit obvious asymmetries . as shown in figure fig.s4 , usually the primary component is redshifted by @xmath115 km s@xmath0 and the velocity difference between the two components is smaller than that of fe 193.51 , which might be due to the complexity introduced by cooling downflows . similar to what we did for ar 10938 and ar 11106 , the techniques of rb asymmetry analysis and rb@xmath7 guided double gaussian fit were applied to each spectral profile with an obvious blueward asymmetry ( the average rb asymmetry in 60 - 110 km s@xmath0 smaller than -0.05 ) . spatial distributions of the double gaussian parameters are presented in figure fig.s5 . spatial distributions of the gaussian parameters derived from the rb@xmath7 and rb@xmath18 analyses are similar to those of the secondary component shown in figure fig.s5 , and thus are not shown here . the distributions of the intensity ratio ( relative intensity ) , velocity difference ( velocity ) , and gaussian width of the secondary component as derived from double gaussian fit and rb asymmetry analysis ( rb@xmath7,rb@xmath18 ) , as shown in figure fig.s6(a)-(c ) , reveal no big difference from those in figure fig.11 . however , the velocity distributions seem to peak at @xmath195 km s@xmath0 , a value larger than those in ar 10938 ( @xmath175 km s@xmath0 ) and ar 11106 ( @xmath185 km s@xmath0 ) and close to that derived by @xcite ( @xmath1110 km s@xmath0 ) . the velocity seldom reaches as high as 200 km s@xmath0 . in fact from figure 7 of @xcite one can conclude that the velocity of the secondary component is usually in the range of 60 - 170 km s@xmath0 and that only a very minor portion of the velocities exceeds 200 km s@xmath0 . as seen from figure fig.s6(d ) , the @xmath10 ratio between the double and single gaussian fit is generally smaller than 1 , indicating an improvement of the fitting by using the double gaussian fit instead of the single gaussian fit . aschwanden , m. j. , wlser , j .- p . , nitta , n. v. , & lemen , j. r. 2008 , apj , 679 , 827 banerjee , d. , teriaca , l. , doyle , j. g. , & wilhelm , k. 1998 , a&a , 339 , 208 banerjee , d. , prez - surez , d. , doyle , j. g. , a&a , 501 , l15 bevington , p. r. , & robinson , d. k. 1992 , data reduction and error analysis for the physics sciences ( mcgraw - hill , new york ) boerner , p. , et al . 2010 , sol . phys . , submitted brooks , d. h. & warren , h. p. 2011 , apj , 727 , l13 brown , c.m . , et al . 2008 , apjs , 176 , 511 bryans , p. , young , p. r. , & doschek , g. a. 2010 , apj , 715 , 1012 culhane , j. l. , harra , l. k. , james , a. m. , et al . 2007 , solar physics , 243 , 19 del zanna , g. 2008 , a&a , 481 , l49 del zanna , g. , aulanier , g. , klein , k .- trk , t. 2011 , a&a , 526 , a137 de moortel , i. , ireland , j. , & walsh , r. w. 2000 , a&a , 355 , l23 de moortel , i. , ireland , j. , hood , a. w. , & walsh , r. w. 2002 , a&a , 387 , l13 de moortel , i. 2009 , space sci . rev . 2009 , 149 , 65 de pontieu , b. , mcintosh , s. w. , hansteen , v. h. , carlsson , m. , schrijver , c. j. , et al . 2007 , pasj , 59 , s655 de pontieu , b. , mcintosh , s. w. , hansteen , v. h. , & schrijver , c. j. , 2009 , apj , 701 , l1 de pontieu , b. , & mcintosh , s. w. 2010 , apj , 722 , 1013 de pontieu , b. , et al . 2011 , science , 331 , 55 dolla , l. r. , & zhukov , a. n. 2011 , apj , 730 , 113 doschek , g. a. , mariska , j. t. , warren , h. p. , et al . 2007 , apj , 667 , l109 doschek , g. a. , warren , h. p. , mariska , j. t. , et al . 2008 , apj , 686 , 1362 feng , l. , inhester , b. , solanki , s. k , et al . 2007 , apj , 671 , l205 guo , l. j. , tian , h. , & he , j .- s . , 2010 , raa , 10 , 1307 hansteen , v. h. , hara , h. , de pontieu , b. , & carlsson , m. 2010 , apj , 718 , 1070 hara , h. , watanabe , t. , harra , l.k . , et al . 2008 , apj , 678 , 67 harra , l. k. , et al . 2008 , apj , 676 , l147 he , j .- s . , marsch , e. , tu , c .- y . , guo , l .- j . , tian , h. 2010 , a&a , 516 , a14 judge , p. g. , tritschler , a. , & low , b. c. 2011 , apj , 730 , l4 king , d. b. , nakariakov , v. m. , deluca , e. e. , et al . 2003 , a&a , 404 , l1 mariska , j. t. , & muglach , k. 2010 , apj , 713 , 573 marsch , e. , wiegelmann , t. , & xia , l. d. 2004 , a&a , 428 , 629 marsch , e. , tian , h. , sun , j. , curdt , w. , & wiegelmann , t. 2008 , apj , 684 , 1262 marsh , m. s. , walsh , r. w. , & plunkett , s. 2009 , apj , 697 , 1674 marsh , m. s. , de moortel , i. , & walsh , r. w. 2011 , apj , 734 , 81 martnez - sykora , j. , de pontieu , b. , hansteen , v. , & mcintosh , s. w. 2011a , apj , 732 , 84 martnez - sykora , j. , et al . 2011b , in preparation mcewan , m. p. , & de moortel , i. 2006 , a&a , 448 , 763 mcintosh , s. w. , & de pontieu , b. 2009a , apj , 706 , l80 mcintosh , s. w. , & de pontieu , b. 2009b , apj , 707 , 524 mcintosh , s. w. , innes , d. e. , de pontieu , b. , leamon , r. j. 2010a , a&a , 510 , l2 mcintosh , s. w. , leamon , r. j. , & de pontieu , b. 2010b , apj , in press murray , m. j. , baker , d. , van driel - gesztelyi , l. , & sun , j. 2010 , sol . 261 , 253 o@xmath29dwyer , b. , del zanna , g. , mason , h. e. , weber , m. a. , & tripathi , d. 2010 , a&a , 521 , a21 peter , h. 2001 , a&a , 374 , 1108 peter , h. 2010 , a&a , 521 , a51 robbrecht , e. , verwichte , e. , hochedez , j. f. , et al . 2001 , a&a , 370 , 591 rouppe van der voort , l. , leenaarts , j. , de pontieu , b. , carlsson , m. , vissers , g. 2009 , apj , 705 , 272 sakao , t. , kano , r. , narukage , n. , et al . 2007 , science , 318 , 1585 schatten , k. h. , wilcox , j. m. , & ness , n. f. 1969 , sol . 6 , 442 schrijver , c. j. , & derosa , m. l. 2003 , sol . phys . , 212 , 165 stenborg , g. , marsch , e. , vourlidas , a. , howard , r. , & baldwin , k. 2011 , a&a , 526 , a58 tian , h. , mcintosh , s. w. , & de pontieu , b. 2011a , apj , 727 , l37 tian , h. , mcintosh , s. w. , habbal , s. r. , & he , j .- s . 2011b , apj , in press tripathi , d. , mason , h. e. , dwivedi , b. n. , del zanna , g. , & young , p. r. 2009 , apj , 694 , 1256 ugarte - urra , i. , & warren , h. 2011 , apj , 730 , 37 verwichte , e. , marsh , m. foullon , c. , et al . 2010 , apj , 724 , 194 wang , t. j. , ofman , l. , & davila , j. m. 2009a , apj , 696 , 1448 wang , t. j. , ofman , l. , davila , j. m. , & mariska , j. t. 2009b , a&a , 503 , l25 wang , y .- m . , ko , y .- k . , & grappin , r. 2009 , apj , 691 , 760 warren h. p. , ugarte - urra , i. , young , p. r. , & stenborg , g. 2011 , apj , 727 , 58 young , p. r. , et al . 2007 , pasj , 59 , s857

the process of star formation in molecular gas is often described as multi - physics . a myriad of physical effects are thought to be important at regulating the process , through establishing the structure of molecular gas and the subsequent distribution of stellar masses that result . fluid dynamics shaped by gravitation , magnetic fields , chemistry , and radiation lies at the heart of the process and each of these effects is thought to contribute significantly to the outcome . while analytic approaches guide our thinking , numerical simulations have been essential for modeling the nonlinear interaction of these effects . as numerical simulations become more sophisticated , they provide a better emulation of the processes at work in molecular clouds . by finding a good match between the observations and simulations , we recognize the underlying physical parameters governing the simulation as a possible ( though not unique ) set of conditions within molecular clouds . given this improvement in numerical methods , many groups have begun to compare simulations of star forming molecular clouds to observations of the same ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? these numerical studies all produced _ synthetic observations _ , which can be compared to observations of actual molecular clouds . the synthetic observation approach avoids the complexity of inferring the real physical properties of molecular gas from the incomplete information present in observations . these comparisons usually leverage an underlying statistic to gauge the degree of similarity . the most common route is to identify a population of objects within the simulation ( e.g. , dense molecular cores ) and compare their properties to those of the known population . the studies claim a successful emulation when the distributions of certain properties ( mass , angular momentum , size , etc . ) mimic those of the observed population . population comparisons work well when there is a well - defined population of observed objects to which the simulations can be compared . when comparing continuous structure such as that of the gas in a molecular cloud , different sets of statistics that rely on the continuous nature of the data can be employed . in general , these statistics can be broadly divided into those that rely on the theoretical understanding of turbulence @xcite and those inspired by observations @xcite . in addition to those statistics that are constructed around the continuous nature of the star forming clouds , catalog - based approaches are commonly adopted . the canonical approach is that of clump finding , which entails dividing or extracting a catalog of substructures from the molecular cloud ( called clumps ) , and comparing the resulting population of objects @xcite . with the improving quality of both simulations and observations , there is a growing need for a general framework by which the comparisons between observations and simulations can be rigorously assessed . many statistical tools are developed _ ad hoc _ , and it is unclear what each individual tool is telling us . a given tool may not be sensitive to different physical parameters . for example , fractal statistics may not be significantly altered by changes in the magnetic field . similarly , simulation efforts may claim to reproduce the structure of molecular clouds , but the assessment of that similarity is frequently qualitative . in this paper , we present a framework by which the sensitivity of different statistics to physical parameters can be measured . we further use this framework to evaluate the quality of these statistics . we focus this work on the structure of star - forming molecular clouds . observations of molecular cloud structure use both spectral lines ( typically co ) and the dust continuum . of these , we focus on the co line mapping of clouds , which produces position - position - velocity ( ppv ) data cubes containing information about both the distribution and dynamics of molecular clouds . such ppv data sets display a wealth of structure , representing a challenging modeling problem for numerical simulations . in this paper , we will outline a basic analysis framework and provide several candidate statistical measures for comparing data sets ( [ section - comparison ] ) . we will then present a suite of numerical simulations that exhaustively probe a simple parameter space ( [ section - simulations ] ) . using this set of simulations , we describe how _ linear models _ provide a means to analyze the significance of the statistical measures ( [ section - linearmodel ] ) . using the results of the analysis , we critically evaluate the different proposed statistics in [ section : results ] . this work emphasizes the comparison of a family of simulations to each other . future work will use this approach to compare simulations to observations and the optimization of simulation parameters to match observations . the comparison between simulations and observations is best carried out in the observational domain . our observational window onto reality is restricted to information that can be inferred from the observed emission of these molecular clouds . inferring the full suite of physical cloud properties ( density , temperature , velocities ) from the emergent radiation is complicated by radiative transfer effects ( particularly in optically thick emission ) , chemistry , spatially and temporally varying heating and cooling as well as imperfections in observations . compensating for these effects is difficult and always leaves caveats attached to the conclusions . by creating synthetic observations from the known physics in simulations , we avoid having to invert these non - linear effects . instead , we rely on the same caveats affecting both our emulated and actual views of star formation . thus , a rigorous comparison of simulations and observations necessarily requires the comparison of synthetic observations with actual observations ( e.g. , * ? ? ? even though the present work does not directly incorporate observational data , we operate on synthetic observations and deploy statistical tools that can also operate on observational data . given two ppv data sets , @xmath0 and @xmath1 , we propose a formalism based around a pseudo - distance metric that measures the differences between them . like other distance metrics , we require @xmath2 and , ideally , this pseudo - distance measures the difference in the underlying physics producing the data sets , such as temperature , magnetic field , and mach number of the fluid flow . in contrast , it should not be affected by other features such as coordinate shifts . indeed , a good similarity statistic will be unaffected by any representation effects such as the pixelization ( though it should depend on the resolution ) . for example , @xmath3 if @xmath1 is just @xmath0 shifted along any of its axes , e.g. , @xmath4 . any statistic that ignores the spatial relationships between pixels will trivially fulfill this criterion . the distance metric should be sensitive to difference in physical scale for a scale factor @xmath5 : @xmath6 . for observational data , the statistic should be independent of the noise levels of the data set . we denote the normal distribution with mean @xmath7 and variance @xmath8 as @xmath9 . for example , @xmath10 if @xmath11 and @xmath12 for the same @xmath13 but different @xmath14 and @xmath15 . practically , this means any statistic involving simulation data should not rely on any data that are at low intensity levels with respect to the maximum . such data would be impossible to measure in real observational data sets . the distance metric should be near zero for different realizations of the same physical processes . in our notation , @xmath10 if @xmath0 and @xmath1 are synthetic observations generated with the same physical conditions but different random seeds . likewise , observations of regions in a similar evolutionary state should also show minimal difference under these statistics . finally , there is significant physical information in the spatial ordering of the pixels . self - gravitation , for example , leads to neighboring pixels begin correlated in their density and consequently their emission . thus , @xmath16 if @xmath17 is the same data as @xmath0 with the positions randomized ( either in position or in position plus velocity ) . given these considerations , we propose several possible similarity statistics as follows . not all statistics fulfill all of the above criteria , and the different features they incorporate lead to different responses under the changing physical conditions ( [ section : results ] ) . @xcite first proposed an application of principal component analysis ( pca ) to spectral line data cubes of molecular clouds . subsequently , pca has been shown to be sensitive to the structure function of the underlying turbulent flow @xcite . we adapt part of the pca formalism to parameterize the physical differences between datasets . given two data sets , @xmath18 and @xmath19 , we construct a covariance matrix , comparing the variance between different velocity channels in the data cube : @xmath20 and a parallel construction for @xmath1 yielding @xmath21 . following the pca approach , we compute eigenvalues @xmath22 and eigenvectors @xmath23 following @xmath24 . to construct a distance measure , we sort the vectors eigenvalues in order of absolute value ( @xmath25 , @xmath26 ) and propose a distance metric : @xmath27^{1/2}. \label{equation - dpca}\ ] ] here , the sums are carried out over the elements of the vectors @xmath28 . the length of these vectors is equal to the number of velocity channels in the data . in practice , nearly all of the power is found in the first @xmath29 eigenvalues , and thus we truncate the sums in equation [ equation - dpca ] after 50 terms . this facilitates comparing the results to cubes with different extent in velocity . the spectral correlation function ( scf ) was first proposed by @xcite , and subsequently demonstrated to be a useful discriminant between the properties of turbulence in comparing simulations and observations @xcite . the scf has a variety of different forms ; here we express the scf in terms of a normalized root - mean - square difference between spectra separated by an offset vector @xmath30 : @xmath31 here , @xmath32 is a vector representing the two spatial coordinates within a datacube . note that @xmath33 and generally decreases to zero as @xmath34 grows large . computing the scf for a pair of data cubes will yield two different surfaces . we calculate a distance metric as @xmath35 ^ 2}.\ ] ] again , we truncate the summation over an area where the scf is significant . we chose a 23-pixel squared patch for calculating @xmath36 and subsequently @xmath37 . motivated by the need for multivariate two - sample testing , the cramer test statistic was first proposed by @xcite . given two dimensional data sets @xmath38 and @xmath39 with sizes @xmath40 and @xmath41 respectively , the test statistic is calculated by viewing each row as a point in a @xmath42-dimensional space . the test statistic is computed by comparing the typical euclidean inter - point distances between the data @xmath38 and the data in @xmath39 to the distances between the data within each of @xmath38 and @xmath39 respectively : @xmath43 the above norm is the euclidean norm between two points in @xmath42 dimensions : @xmath44 . one advantage of the cramer statistic is that its significance can be determined by bootstrapping . the bootstrap draws two mock data sets of the same sizes as @xmath38 and @xmath39 from the combined set @xmath45 and evaluates the statistic over a large number of repetitions . additionally , the statistic has been shown to be more sensitive than the multivariate kolmogorov - smirnov test . in the context of the current work , we must recast ppv data cubes into sets of @xmath42-dimensional points suitable for the cramer test . we experimented with several methods for representing a ppv data cube as one or more @xmath46-dimensional points . given a data cube @xmath47 , one approach is to consider each spectrum with @xmath42 velocity channels as a datum and consider the sets @xmath38 and @xmath39 as spectra drawn from the two different data cubes . this comparison requires that the data sets being compared have matched spectral resolution and eliminates spatial information in the data . alternatively , we construct a vector from each velocity channel in the data cube by sorting the data into a vector of intensity values for each channel . we found that only retaining the top quartile of the brightness data maintained the sensitivity of the statistic while improving performance . we use the latter approach in this initial study and evaluate the application of the cramer test to ppv data cubes in other work ( flynn et al . , in prep . ) . the cramer statistic is sensitive to the absolute values of the data set . in order to eliminate this sensitivity , we normalize the transformed arrays by their respective spectral norms . the spectral norm of an array @xmath38 is constructed by first taking the eigenvalues , @xmath28 of the square array @xmath48 . the spectral norm is then @xmath49 . we note that clearly the cramer statistic is not sensitive to pixel ordering in our implementation . the statistic performs well in the task of classifying similar and dissimilar data cubes , and experimentally satisfies the requirements of a well behaved distance metric . we ran 16 simulations of self - gravitating , magnetized turbulence using enzo version 1.9 @xcite . the simulations were designed to emulate a star forming cloud undergoing gravitational collapse . we used a base grid with dimensions 128@xmath50 over a box size of @xmath51 pc with periodic boundary conditions . we set the initial density field of the box to be uniform @xmath52 and the initial direction of the magnetic field to be parallel to one face of the cube . for each of the simulations , we initially ran the simulations without self - gravity to allow the driving mechanism to develop turbulence . the random driving pattern for the turbulence is solenoidal with power on wavenumbers from @xmath53 to @xmath54 . depending on the run , we set the shape of the initial velocity spectrum to have power on a wide range of scale ( from @xmath55 to @xmath56 ) or only large scale velocity structure ( @xmath57 to @xmath55 ) . we scale the time of the simulation to the crossing time @xmath58 where @xmath59 is the mach number and @xmath60 is the sound speed . at @xmath61 , we turned on self - gravity , adaptive mesh refinement and enabled the formation of sink particles . we used a jeans length refinement criterion , refining the grid by a factor of two whenever the local jeans length became shorter than four cells @xcite . we limited the number of refinement levels to four . we systematically varied the input mach number ( @xmath59 , which sets the amplitude of the forcing field ) , initial magnetic field strength ( @xmath62 ) , initial velocity spectrum shape ( @xmath63 ) , and temperature ( @xmath64 ) as described in table [ table : codes ] . the table also reports the `` text codes '' we give to describe the varying parameters in the results . we analyze the simulations at intervals of 0.1 crossing times from @xmath65 to @xmath66 . ccccc low & 6 & 1 & @xmath67 $ ] & 10 + high & 18 & 100 & @xmath68 $ ] & 30 + + & m & b & k & t [ table : codes ] we ran one simulation for each of the @xmath69 possible combinations of the parameter settings . we post - processed the data using the radmc-3d code @xcite to produce ppv data cubes of @xmath70 emission . in radmc-3d , the level populations are calculated using the large velocity gradient ( lvg ) approximation @xcite . the @xmath71co collisional and emission coefficients are taken from the leiden atomic and molecular database ( lamda ; * ? ? ? the simulated observations are generated as if the simulation domain were located a distance of 260 pc ( the distance of the perseus molecular cloud , chosen for future comparison with observational data from the complete survey ; * ? ? ? we also selected @xmath71co as a tracer since it highlights the main structure of a molecular cloud while not being as limited by opacity effects in the @xmath72co data . in the radiative transfer , we include `` microturbulence '' with @xmath73 . microturbulence represents turbulent motions on scales too small to be resolved by the hydrodynamic grid and contributes to line broadening . the velocity resolution of the data cubes is @xmath74 . we produced ppv data sets for each of the three orthogonal orientations of the cube to capture the effects of observations along and parallel to the initial magnetic field . regions where the adaptive mesh undergoes refinement are averaged over so that the ppv data have the same spatial size as the base grid , namely 10 pc resolved into 128 resolution elements . our final data suite consists of ( 16 simulations)@xmath75(8 time steps)@xmath75(3 orientations)@xmath76 synthetic ppv data cubes . figure [ fig : compcube ] shows the @xmath77 slice through the data cubes for four different parameter settings at @xmath66 . additional data from the simulations are presented in appendix [ appendix - simprops ] . the work leverages the framework from the statistical subfield of experimental design . the goal of this analysis is to determine how changes in physical conditions manifest in the synthetic data . we refer to these changes as _ effects _ and aim to estimate their significance . to make these estimates , we use similarity metrics to compare the values at the fiducial ( or low ) parameter values to those at the high values ( table [ table : codes ] ) . we define a scalar response when comparing the synthetic data from one simulation to another . since there are multiple timesteps available for each set of physical parameters , these need to be aggregated into a single statistic typifying how distinct simulations with different parameters are . we define a scalar response value for the @xmath78th simulation compared to a fiducial simulation as @xmath79 . \label{eq : response}\ ] ] this is the mean of a given comparison statistic over the last @xmath80 outputs of the simulation ( i.e. , from @xmath65 to 3.0 ) . each simulation is compared to the fiducial simulation @xmath0 , which has @xmath81 , @xmath82 , @xmath83 k and @xmath84 $ ] . the averaging occurs when self - gravity is active , and some simulations are evolving significantly . the use of an average over time aggregates these effects , but the time evolution can be significant ( see [ subsection - evolution ] ) . typically , parameter studies of star formation use `` one factor at a time '' ( ofat ) variation of parameters ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? one parameter ( e.g. , the magnetic field strength ) is varied while all the other factors in the simulation are held constant . the response of the simulation to the variation of this single parameter is then interpreted as the effect of changing this parameter . this approach is reasonable when simultaneous parameter changes will not result in a different degree response relative to individual changes , i.e. , when there are no interactions between the parameters . analyzing a suite of ofat parameter studies may be misleading when parameter interactions exist . we establish quantitative measures of parameter changes by studying the resulting runs as a standard linear model . we consider the case where we have @xmath85 different numerical simulations and @xmath46 different combinations of parameter settings . note that repeated experiments with the same parameter settings but different random seeds will increase @xmath85 but not @xmath46 . in matrix form we have @xmath86 where @xmath87 is an @xmath88 model matrix , @xmath89 is the @xmath90 vector of responses ( i.e. , equation [ eq : response ] ) , @xmath91 is the @xmath92 vector of independent normal errors and @xmath93 is the @xmath94 vector of unknown parameters ( including the intercept ) . these fit parameters characterize the response of the simulations to changing the physical conditions in the simulation . in this framework , the ofat design has a model matrix @xmath95 where the first column corresponds to a constant intercept , while the latter four columns code the factor setting for the four factors used in this study ( see table [ table : codes ] ) . in ofat , we represent the low / fiducial settings for parameters with a _ code _ of 0 and the high setting of parameters with a code of 1 . we utilize codes instead of the actual parameter values for simplicity in developing the design matrix ( equation [ eq : base_ofat ] ) and interpreting the statistical properties thereof . there is substantial latitude in choice of coding since the values of @xmath96 . under certain coding schemes , the values of these effect estimates can be interpreted in terms of quantitative effects @xcite , but here we are only interested in the significance of the values . the standard least squares estimator , @xmath97 , of the fit parameters @xmath96 is given by : @xmath98 the covariance matrix for @xmath99 is given by @xmath100 where @xmath101 is the unknown error variance characterizing @xmath91 . determining @xmath8 usually requires running multiple experiments with the same physical parameter settings but different random seeds . the repeated experiments indicate what fraction of the measured variation is the result of statistical fluctuations , enabling these effects to be distinguished from changes that can be attributed to changing simulation parameters . the covariance matrix for the parameters @xmath91 is given by @xmath102 indicating that the parameter estimates in the ofat approach are correlated with each other meaning that changes in one will affect changes in another parameter . in the case of an experimental setup and measuring of parameter significance it is desirable to have independent ( i.e. , uncorrelated ) parameter estimates . additionally , the above model does not allow for interactions between the parameters to be studied through that set of simulations . we run a _ full factorial _ design where all possible changes of the parameters are considered . using only two values for each of the four factors , this setup results in @xmath103 runs . in order to aid the resulting analysis , we adopt a different coding for the levels , which utilizes the linear independence between the columns of orthogonal arrays to produce independent estimates of the importance of different physical effects . we note again that the choice of coding is arbitrary as we only seek to measure the significance of the effect estimates . the fiducial / low will be coded as @xmath104 ( previously 0 ) and a @xmath105 is used for the high level . the first five columns of the model matrix are given as @xmath106 where the first column represents the intercept and the remaining 4 columns show the level setting for the parameters . for example the third row shows that enzo would be run with @xmath81 , @xmath82 , @xmath107 $ ] and @xmath108 k. also notice that rows 15 , 16 , 14 , 10 and 7 respectively are the 5 runs in the ofat design described in equation [ eq : base_ofat ] . the above design allows for the estimation of all of the two , three and four factor interactions between the resulting parameter settings . these terms are included in the model matrix by adding an additional 10 columns to @xmath87 that are constructed by taking the product of all possible subsets of the columns 2 through 5 in equation [ eq : ortho_full ] . when using this parameterization , @xmath109 where @xmath110 is the identity matrix , so we describe this coding as an _ orthogonal design_. in this case , the covariance matrix ( corresponding to equation [ eq : cov_ofat ] ) is diagonal , and the estimated effects @xmath97 are statistically independent . under this framework each parameter estimate is given by @xmath111 where @xmath112 and @xmath113 are the means of the responses at the high and low values of the parameter respectively ( as defined in table [ table : codes ] ) . the goal is to evaluate the statistical significance of these effects . without a measure of the run - to - run variation in the response statistics , any derived effect could just be the result of random fluctuations . ideally , there would be several replicates of each experimental run which would allow for the estimation of @xmath8 in the model . replicated simulations are runs with the same physical parameters , but different random seeds for the turbulent driving . with an estimate of @xmath8 , the statistical significance of each of the regression coefficients ( @xmath114 ) can be assessed to determine what parameters are causing the real changes in the underlying ppv data cubes . usually , replicated numerical experiments entail a considerable investment of computational resources and using those resources to produce a different experimental run can provide substantially more information . in the case of the two - level full factorial design , parameter significance can still be determined even without replicated runs . @xcite introduced a formal approach to judge parameter significance in unreplicated experiments . lenth s method assumes that most effects and interactions are insignificant so they can be used to estimate the `` noise '' in the effect estimates . this assumption is checked after analysis to verify that only a small number of significant effects are indeed significant . the method can not be relied upon if it finds nearly all or none the effects are significant . a pseudo - standard error is calculated using a robust estimator @xmath115 given by @xmath116 where @xmath117 the approach proceeds by constructing statistic similar to a student s t : @xmath118 and comparing @xmath119 to a critical value set by the experimental design . in the case of a 16 run experiment with 15 factors , any @xmath120 is judged as significant , where the critical value is determined using the so - called individual error rate and changes depending in the number of runs @xcite . using the simulations described above , we calculate response vectors @xmath89 for each of the three similarity metrics described in [ section - comparison ] . in this section , we examine how the different similarity metrics respond to changes in the physical parameters defined in the initial conditions of the simulation . we begin by examining what can be determined from the full suite of 16 simulations , representing all parameters and then we examine how the results change for an ofat study . finally , we propose using fractional factorial designs , where a carefully chosen subset of the 16 simulations is used in the analysis . using the full factorial design , we can estimate the effects @xmath121 and their statistical significance . the effects measure how responsive a statistic is to changing physical parameters in the simulations . the influence of the four principal parameters ( known as _ main effects _ in the design literature ) and their interactions is probed with a full factorial design . for example , the analysis measured how the scf statistic responds to changes in input mach number and temperature as well as the interaction term between these effects ( coded as m : t ) that measures the change in the response statistic in excess of the change seen from the main effects alone . these additional interaction terms probe cases where the physical parameters amplify the effects of each other . figure [ fig : full_lenth ] shows the analysis of the full factorial design across all three similarity metrics . the coefficients @xmath122 are shown as the lengths of the bars . the differences in the range of values shown for each @xmath114 arises because the individual statistics have different numerical ranges in their response . only their value relative to the significance threshold of @xmath123 for each statistic is important . the degree by which @xmath122 exceeds the threshold can be interpreted as a degree of significance . for example , the scaling of the three statistics with changing the input mach number is more significant than any other effect for the three response statistics . the analysis of the three comparison statistics consistently highlights two significant main effects : input mach number and temperature . notably , there is no significant effect due to changing magnetic field alone . since the input mach number of the turbulent driving in the simulation determines the density contrasts in shocks , this result is not surprising . note that the input mach number is most closely tied to the amplitude of the turbulent forcing field and does not mean that actual mach number of the flows in the simulation is this value ( see [ subsection - driving ] ) . the temperature changes manifest in slightly larger line widths due to thermal pressure , but these effects should be insignificant because the sonic mach numbers are large . the temperature increase also controls the amplitude of the spectral lines after radiative transfer ( see appendix a ) . while the amplitude of the spectral lines is compensated for in the three statistics under comparison , the compensation has varying degrees of success . the full - factorial analysis indicates an interaction effect between the input mach number and the temperature which is significant for pca and cramer statistics but not for the scf . the interaction effect is completely eliminated because of the nature of the scf normalization which compensates for absolute values of the brightness in the cube comparison on a comparison - by - comparison basis . the cramer statistic also normalizes the data but appears to be more sensitive to variations due to the brightness of the data sets . given the strong effects of the magnetic field visually apparent in figure [ fig : compcube ] , it is surprising that the magnetic field does not manifest as an active parameter in the response statistics . this indicates a shortcoming of the statistics as formulated in the analysis : they are not sensitive to magnetic effects . even though the principal analysis is formulated for a line of sight parallel to the initial magnetic field , we performed a parallel analysis for the two faces perpendicular to the magnetic field and found no significant magnetic effects . this lack of sensitivity to the strong effect on the emission structure in the highly magnetized case can be explained because the spectral line profiles for the @xmath124 are not significantly different from the fiducial case ( see figure [ fig - samplespec ] ) . again , the analysis highlights the relative merits of the comparison statistics and indicates different statistics will need to be developed in order to measure the sensitivity to magnetic field . despite this lack of sensitivity to the main effect , both the pca and scf indicate a negative interaction effect between the magnetic field and the input mach number . this interaction indicates that joint increases in both the input mach number and magnetic field produce simulations that these similarity measures find are more similar to the fiducial value than if these parameters are increased independently . we attribute these significant interaction effects to a difference between the input mach number , which is the parameter in the simulation , and the actual mach number which establishes the density structure in the simulation ( see [ subsection - driving ] ) . despite the comparatively low resolution of the simulation suite , the analysis of the full factorial design indicates the initial success at adopting the formalism . the results show that the framework can identify some of the significant physical effects ( input mach number and temperature ) that govern the structure of ppv data cubes . however , the nondetection of effects that are clearly significant ( magnetic field ) shows that the construction of the similarity statistics needs to be adapted to detect magnetism . alternatively , other similarity measures which emphasize magnetic effects could be recast into this framework ( e.g. , * ? ? ? * ; * ? ? ? we emphasize that the results identify the significance of both the effects and their interactions , representing the primary utility in employing the fully developed statistical framework to this analysis . .effect estimates of the various statistics based on the one factor at a time experiment ( ofat ) . [ cols=">,>,>,>",options="header " , ] where the full factorial analysis identifies significant main effects as well as their interaction terms , ofat analysis can only detect main effects . furthermore , ofat analysis without multiple realizations of the physical effects yields no estimate of the variance that arises because of the random fluctuations between simulations with the same physical conditions . the ofat analysis can be used to compare the magnitude of the main effects as given in table [ tb : ofat ] for the three different similarity statistics . as with the full factorial design , input mach number remains the dominant effect . temperature effects and magnetic effects also seem to be detectable and the initial velocity spectrum is least important affect . the intercept value is estimated as 0.0 for all statistics because of the definition of the response , namely @xmath125 when the fiducial simulation is compared to itself . we note that , in practice , the fiducial data will likely be observational data sets to which the simulations are being compared . for analyzing a suite of simulations without observational data , a simulation at the center of the parameter space would make the best choice for a fiducial simulation . the effect estimate for the magnetic field seems like it could be detected by the similarity statistics ; however , the full - factorial design indicates that this apparent detection arises because of the interaction effects seen between magnetic field strength and input mach number . a purely ofat analysis will incorrectly conclude that the influence of the magnetic field strength has been detected . furthermore , ofat provides no assessment of the variance in the similarity statistics and the importance of each of these effect estimates can not be determined . this analysis highlights the shortcomings of the ofat approach . specifically , the analysis can be misled in the presence of parameter interactions and is inefficient at testing the significance of different simulation parameters @xcite . the full factorial design provides a more nuanced interpretation of the responses as a function of changing initial conditions . however , running a full factorial design can require substantial computational resources . we ran the two - level , full factorial design , because we were using 128@xmath50 resolution simulations . in addition , we chose a model where there are only four factors and each factor only has two values . there are situations where one would like to run simulations with higher resolution , more factors , and a continuous range of values . to reduce the computational cost , it is possible to run a subset of the full factorial design with the ability to estimate both the main effects and some of the interactions . in many situations , prior knowledge can be used to identify a limited subset of the interactions that are likely to be important . starting from the full factorial design of 16 runs , we select the 8 runs that correspond to having all factors ( m , b , k , t ) at the high level . this results in the following 8 runs , which are commonly referred to as a @xmath126 fractional factorial design . @xmath127 again the columns 2 through 5 are used to assign the level setting for each of the four factors . since we have reduced the run size by a factor of two we lose the ability to estimate 8 of the 15 effects that were estimated in [ section - fullfac ] . of the 7 remaining degrees of freedom , one can attribute the significance of the estimated effect to one of two terms . for example , if the effect of column 2 is significant then either the main effect of input mach number ( m ) or the b : k : t interaction is important . we assume third and higher order terms are negligible , so we can estimate each of the four main effects . the remaining three columns can be used to estimate the following pairs of effects , m : b = k : t , m : t = k : b and b : t = k : m . the notation m : b = k : t is an indication that , in fractional factorial design , estimates of the two - factor interaction m : b are aliased with the k : t interaction . if significant , we would not be able to discriminate between these two effects . this aliasing is a cost of running a factional factorial design . however , we can use our prior knowledge to decide whether a given significant effect should be interpreted as , for example , a m : t interaction or a k : b interaction . we believe that the effects from the initial velocity spectrum are negligible because those effects are washed out by the turbulent driving . our assumption is well justified by the insignificance of velocity spectrum effects in the full factorial response analysis . in larger simulations , it is possible the velocity spectrum and magnetic field could show dependence . thus , for this set of simulations , we believe that the significant interactions would be attributed to m : b , m : t and b : t as opposed to the effects involving the initial velocity spectrum ( k ) . ignoring the aliasing of the parameters in the above design , the 8 runs simply correspond to a full factorial in three factors . hence , one can use the same unreplicated analysis proposed by @xcite to determine which parameters are significant . in this case the appropriate critical value is @xmath128 . figure [ fig : frac_lenth ] shows the results from the fractional factorial analysis . [ fig : frac_lenth ] the significant parameters from both the full factorial and fractional factorial are consistent in relative magnitude and sign . the effect significance , particularly for the pca is reduced under the fractional factorial design . some information is lost when the fractional factorial is used over the full factorial , but the fractional factorial does provide some information about the interactions . in addition , it only requires an additional 3 runs in comparison to the ofat design , while establishing an understanding of the parameter interactions that compares well to running the full suite simulations at twice the computational cost . the results from this design illustrate how a carefully chosen design can maximize information from a parameter studies that are subject to limited computational resources . the savings of a fractional factorial design over a full factorial grows as the number of parameters explored or the number of level settings increases . the dominant physical effect we see in our simulations is the response of our statistics to the input mach number parameter set . this input mach number does not actually set the true mach number of the fluid flow in the simulation . rather , it determines the amplitude of the driving field used to establish the turbulent velocity field ( * ? ? ? * outlines an approach similar to that used in enzo ) . the default enzo distribution uses a heuristic formulation to specify the amplitude of the turbulent driving force per unit mass : @xmath129 , where @xmath130 is a _ driving efficiency _ , @xmath60 is the sound speed , @xmath131 is the initial rms velocity measured in the simulation and @xmath132 is the mean free - fall time in the domain . the parameter @xmath59 is the input mach number for the simulation but does not necessarily mean that the mach number of the simulation will have that value . the driving efficiency was taken as @xmath133 for all simulations , but this value only produces real mach numbers close to the initial mach number parameter for weak @xmath134-field cases . the initial mach number should thus be thought as a characteristic of the strength of the driving field rather than the true mach number characterizing the simulation . when compared to the input mach number , we found that the actual mach numbers were between 37% ( @xmath124 , @xmath135 $ ] ) and 95% ( fiducial ) of the input mach number . of note , the input mach number parameter also establishes the crossing time for the simulation . the actual crossing time will be different from the crossing time reported in the simulation . we also constructed a linear model of the actual mach number in the simulation as a function of the changing input parameters . as expected , the input mach number was the dominant effect in establishing the real mach number , but the initial magnetic field strength has a strong negative effect ( increasing @xmath62 decreased actual mach number relative to the input mach number ) . the damping effect of the magnetic field is expected on physical grounds and serves to explain why the magnetic field appears significant in ofat analysis , but the effect is not present in the full - factorial design . instead , interaction effects between the magnetic field and temperature or input mach number appear significant . the interaction effects in the actual simulation response can be attributed partly to the magnetic field driving a discrepancy between input and actual mach number . additionally , a modest mach - temperature interaction emerged from the model which may explain the same effect seen in the response of the comparison metrics . our primary analysis averages over the last eight time steps of the simulation ( from @xmath65 to 3.0 ) . since self - gravity is turned on at time @xmath136 , simulations that are undergoing gravitational collapse will see these effects manifest over the time period contributing to the average . real effects may be averaged over or diminished by using the aggregate of all the response values over the self - gravitating phase of the simulation . to address this concern , we plot the main effect estimates as a function of time in figure [ fig : timevar ] . this figure compares each simulation at a fixed time step ( though note that @xmath137 will vary by simulation based on the input mach number as above ) . figure [ fig : timevar ] shows that there is not a significant change in the main effect estimates over the period when the simulation is self - gravitating . there is some variation in the sensitivity of the scf to the magnetic field at the last timesteps of the simulation , but this effect is modest . the significance of the effects for mach and temperature changes appears significant over the whole period . in this work , we have proposed a formalism by which ppv data cubes of spectral line data can be compared in order to assess physical effects . the formalism relies on the methods utilized in the statistical subfield of experimental design . we have proposed three similarity metrics based on existing techniques in the literature and assess their reliability using the formalism applied to a suite of simulations . the three similarity metrics are principal component analysis , the spectral correlation function and a new method using the cramer multivariate two - sample test . all three similarity statistics showed consistent results in our analysis , though the relative sensitivities of these measures varied . in particular , the statistics are most sensitive to mach number and temperature effects . we have also examined the benefits and liabilities for using the `` one factor at a time '' ( ofat ) analysis that is common in the astrophysical literature . while the results of ofat analysis are relatively easy to interpret , the analysis is unable to assess whether the physical effects explored in the simulations interact with each other . this can lead to overinterpreting the importance of certain parameters in the simulation . for example , we found that our measurement statistics appeared sensitive to magnetic field strength based on the ofat analysis . however , in exploring the parameter space fully , the statistics were found to be unable to detect changes in the magnetic field strength , instead finding an interaction effect between magnetic field strength and the input mach number . we attribute this interaction to the reduction of the real mach number relative to the input for increasing magnetic field strength . a full exploration of parameter space is invariably computationally prohibitive for high quality simulations . thus , we advocate selecting a fractional factorial design that is able to recover those effects most likely to be significant through the study . a major shortcoming of this study is the exaggerated ranges used for the parameters ( e.g. , 100 @xmath7 g fields on 10 pc scales are not observed ) . even so , this work demonstrates proof - of - concept for three avenues of study . first , the comparison of simulations with observations can be explored under this framework . frequently , simulations are compared to observations and the conditions for a best match are assessed under qualitative terms . by establishing observations as the fiducial to which the simulation suite is compared , we can establish those effects which most closely govern similarity to simulations . while this would formalize the comparison between simulations and observations , it will require amending the similarity statistics to account for the effects of telescope observations ( variable data set sizes , limited resolution , receiver noise ) . the second set of work that this study enables is the ability to assess the reliability of comparison metrics . there are a wealth of such metrics present in the literature , but they are only validated under a limited set of conditions . using this test suite of simulations or a higher quality suite that studies a full range of parameter combinations , we can assess how reliable the simulations are at detecting certain physical effects . the framework also reveals when a statistic can be conflating the effects of two parameters . for example , the cramer statistic is primarily sensitive to mach and temperature effects and is surprisingly immune to interaction effects between these parameters and the magnetic field . the pca and scf metrics detect significant statistical interactions between these parameters and the magnetic field strength . however , none of these statistics detect the magnetic field as a main effect indicating limitations in their current formulation . finally , this study offers guidance for explorations of parameter space in suites of simulations . given the prevalence of ofat analysis in astronomy and the complexity of interpreting the simultaneous variation of multiple physical effects , using a statistical response model to assess parameter effects offers some clear guidance . extensions to this framework allow for multiple values or continuous exploration of parameter spaces , non - linear response models and efficient exploration of high - dimensional parameter spaces for simulation . this project was supported by discovery grants from nserc of canada and a collaborative research grant from the i.k . barber school of arts and sciences at ubc . ssro is supported by a nasa through a hubble fellowship grant # 51311.01 awarded by the space telescope science institute , which is operated by the association of universities for research in astronomy , inc . , for nasa , under contract nas 5 - 26555 . we are grateful for simulation assistance from tyler berg . we acknowledge productive discussions with doug johnstone and alyssa goodman in formulating this research . westgrid and compute canada resources were used in executing the simulations . this research made heavy use freely - available , community - developed software , including enzo ( http://enzo-project.org ) , yt ( http://yt-project.org/ ; * ? ? ? * ) and astropy ( http://astropy.org ; * ? ? ? * ) and we are grateful for the developer efforts on those projects . this research has made use of nasa s astrophysics data system . we acknowledge the constructive comments from the anonymous referee , which improved the quality and presentation of the paper . 42 natexlab#1#1 attwood , r. e. , goodwin , s. p. , stamatellos , d. , & whitworth , a. p. 2009 , a&a , 495 , 201 baringhaus , l. & franz , c. 2004 , journal of multivariate analysis box , g. , hunter , j. s. , & hunter , w. g. 2005 , statistics for experimenters : design , innovation , and discovery , 2nd edition ( wiley ) brunt , c. m. & heyer , m. h. 2002 , apj , 566 , 276 . 2002 , apj , 566 , 289 burkhart , b. , falceta - gonalves , d. , kowal , g. , & lazarian , a. 2009 , apj , 693 , 250 burkhart , b. , lazarian , a. , goodman , a. , & rosolowsky , e. 2013 , apj , 770 , 141 dib , s. , hennebelle , p. , pineda , j. e. , csengeri , t. , bontemps , s. , audit , e. , & goodman , a. a. 2010 , apj , 723 , 425 dickman , r. l. & kleiner , s. c. 1985 , apj , 295 , 479 dullemond , c. p. 2012 , astrophysics source code library , 02015 falgarone , e. , lis , d. c. , phillips , t. g. , pouquet , a. , porter , d. h. , & woodward , p. r. 1994 , apj , 436 , 728 falgarone , e. , phillips , t. g. , & walker , c. k. 1991 , apj , 378 , 186 goodman , a. a. 2011 , computational star formation , 270 , 511 heyer , m. , gong , h. , ostriker , e. , & brunt , c. 2008 , apj , 680 , 420 heyer , m. h. & schloerb , f. p. 1997 , apj , 475 , 173 houlahan , p. & scalo , j. 1990 , apjs , 72 , 133 kleiner , s. c. & dickman , r. l. 1985 , apj , 295 , 466 kramer , c. , stutzki , j. , rohrig , r. , & corneliussen , u. 1998 , a&a , 329 , 249 lazarian , a. & pogosyan , d. 2000 , apj , 537 , 720 . 2004 , apj , 616 , 943 lenth , r. v. 1989 , technometrics , 31 , 469 mac low , m. & klessen , r. s. 2004 , reviews of modern physics , 76 , 125 mac low , m .- m . 1999 , apj , 524 , 169 miesch , m. s. & bally , j. 1994 , apj , 429 , 645 offner , s. s. r. , bisbas , t. g. , viti , s. , & bell , t. a. 2013 , apj , 770 , 49 offner , s. s. r. , krumholz , m. r. , klein , r. i. , & mckee , c. f. 2008 , aj , 136 , 404 oshea , b. w. , bryan , g. , bordner , j. , norman , m. l. , abel , t. , harkness , r. , & kritsuk , a. 2004 , arxiv , 3044 padoan , p. , goodman , a. a. , & juvela , m. 2003 , apj , 588 , 881 padoan , p. , juvela , m. , kritsuk , a. , & norman , m. l. 2006 , apjl , 653 , l125 ridge , n. a. , di francesco , j. , kirk , h. , li , d. , goodman , a. a. , alves , j. f. , arce , h. g. , borkin , m. a. , caselli , p. , foster , j. b. , heyer , m. h. , johnstone , d. , kosslyn , d. a. , lombardi , m. , pineda , j. e. , schnee , s. l. , & tafalla , m. 2006 , aj , 131 , 2921 rosolowsky , e. w. , goodman , a. a. , wilner , d. j. , & williams , j. p. 1999 , apj , 524 , 887 scalo , j. m. 1984 , apj , 277 , 556 schier , f. l. , van der tak , f. f. s. , van dishoeck , e. f. , & black , j. h. 2005 , a&a , 432 , 369 shetty , r. , glover , s. c. , dullemond , c. p. , & klessen , r. s. 2011 , mnras , 412 , 1686 stutzki , j. , bensch , f. , heithausen , a. , ossenkopf , v. , & zielinsky , m. 1998 , a&a , 336 , 697 stutzki , j. & gsten , r. 1990 , apj , 356 , 513 . 2013 , arxiv , 6212 truelove , j. k. , klein , r. i. , mckee , c. f. , holliman , j. h. i. , howell , l. h. , & greenough , j. a. 1997 , astrophysical journal letters v.489 , 489 , l179 turk , m. j. , smith , b. d. , oishi , j. s. , skory , s. , skillman , s. w. , abel , t. , & norman , m. l. 2011 , apjs , 192 , 9 vzquez - semadeni , e. , gonzlez , r. f. , ballesteros - paredes , j. , gazol , a. , & kim , j. 2008 , mnras , 390 , 769 williams , j. p. , de geus , e. j. , & blitz , l. 1994 , apj , 428 , 693 wu , c. f. j. & hamada , m. s. 2011 , experiments , planning , analysis , and optimization ( wiley ) in figure [ fig - samplespec ] we present two summaries of the simulations in the simulation suite . rather than plotting curves for all 16 of the simulations , we only illustrate the fiducial case with all parameters at the low setting , the curves for only one factor changing from its low to high setting and finally the curves for all parameters set to their high value . the two summaries present density pdfs and spectra for these simulations . of note , the density pdfs show that only the fiducial case and the large scale initial spectrum ( @xmath138 $ ] ) show significant gravitational collapse and a tail of values off to high densities @xcite . in the other simulations , including those not shown in the figure , the local jeans refinement condition is never met and there is no local gravitational collapse . the initial driving scale likely plays little role at governing gravitational collapse since the fixed driving pattern dominates the structure of the turbulence at the time of gravitational collapse . the average spectra of the simulated observations also show roughly expected behavior . the line width is larger for higher mach numbers and temperature . the amplitude of the line is also larger for the higher temperatures as expected from radiative transfer effects . spectra along individual lines of sight show significant non - gaussian structure as is frequently seen in simulated data .

the yee scheme ( also known as finite - difference time - domain , or fdtd ) was introduced in @xcite and remains one of the most successful numerical methods used in the field of computational electromagnetics , particularly in the area of microwave problems . although it is not a `` high - order '' method , it is still preferred for many applications because it preserves important structural features of maxwell s equations that other methods fail to capture . among these distinguishing attributes are that the gauss constraint @xmath1 is exactly conserved in a discrete sense , and electrostatic solutions of the form @xmath2 indeed remain stationary in time ( see * ? ? ? * ) . in this paper , we show that these desirable properties are direct consequences of the variational and discrete differential structure of the yee scheme , which mirrors the geometry of maxwell s equations . moreover , we will show how to construct other variational methods that , as a result , share these same numerical properties , while at the same time applying to more general domains . geometric numerical integrators have been used primarily for the simulation of classical mechanical systems , where features such as symplecticity , conservation of momentum , and conservation of energy are essential . ( for a survey of various methods and applications , see @xcite . ) among these , _ variational integrators _ are developed by discretizing the lagrangian variational principle of a system , and then requiring that numerical trajectories satisfy a discrete version of hamilton s stationary - action principle . these methods are automatically symplectic , and they exactly preserve discrete momenta associated to symmetries of the lagrangian : for instance , systems with translational invariance will conserve a discrete linear momentum , those with rotational invariance will conserve a discrete angular momentum , etc . in addition , variational integrators can be seen to display good long - time energy behavior , without artificial numerical damping ( see * ? ? ? * for a comprehensive overview of key results ) . this variational approach was extended to discretizing general multisymplectic field theories , with an application to nonlinear wave equations , in @xcite and @xcite , which developed the multisymplectic approach for continuum mechanics . building on this work , @xcite introduced _ asynchronous variational integrators _ ( avis ) , with which it becomes possible to choose a different time step size for each element of the spatial mesh , while still preserving the same variational and geometric structure as uniform - time - stepping schemes . these methods were implemented and shown to be not only practical , but in many cases superior to existing methods for problems such as nonlinear elastodynamics . some further developments are given in @xcite . while there have been attempts to apply the existing avi theory to computational electromagnetics , these efforts encountered a fundamental obstacle . the key symmetry of maxwell s equations is not rotational or translational symmetry , as in mechanics , but a _ differential gauge symmetry_. without taking additional care to preserve this gauge structure , even variational integrators can not be expected to capture the geometry of maxwell s equations . as will be explained , we overcome this obstacle by combining variational methods with discrete differential forms and operators . this differential / gauge structure also turns out to be important for the numerical performance of the method , and is one of the hallmarks of the yee scheme . as motivation , consider the basic relation @xmath3 , where @xmath4 is the magnetic flux and @xmath5 is the magnetic vector potential . because of the vector calculus identities @xmath6 and @xmath7 , this equation has two immediate and important consequences . first , @xmath4 is automatically divergence - free . second , any transformation @xmath8 has no effect on @xmath4 ; this describes a gauge symmetry , for which the associated conserved momentum is @xmath9 ( which must be zero by gauss law ) . a similar argument also explains the invariance of electrostatic solutions , since @xmath10 is curl - free and invariant under constant shifts in the scalar potential @xmath11 . therefore , a proper variational integrator for electromagnetism should also preserve a discrete analog of these differential identities . this can be done by viewing the objects of electromagnetism not as vector fields , but as _ differential forms _ in @xmath0-dimensional spacetime , as is typically done in the literature on classical field theory . using a discrete exterior calculus ( called dec ) as the framework to discretize these differential forms , we find that the resulting variational integrators automatically respect discrete differential identities such as @xmath12 ( which encapsulates the previous div - curl - grad relations ) and stokes theorem . consequently , they also respect the gauge symmetry of maxwell s equations , and therefore preserve the associated discrete momentum . the yee scheme , as we will show , is a method of precisely this type , which gives a new explanation for many of its previously observed _ a posteriori _ numerical qualities . for instance , one of its notable features is that the electric field @xmath13 and magnetic field @xmath14 do not live at the same discrete space or time locations , but at separate nodes on a staggered lattice . the reason why this particular setup leads to improved numerics is not obvious : if we view @xmath13 and @xmath14 simply as vector fields in @xmath15-space the exact same type of mathematical object why should nt they live at the same points ? indeed , many finite element method ( fem ) approaches do exactly this , resulting in a `` nodal '' discretization . however , from the perspective of differential forms in spacetime , it becomes clear that the staggered - grid approach is more faithful to the structure of maxwell s equations : as we will see , @xmath13 and @xmath14 come from objects that are dual to one another ( the spacetime forms @xmath16 and @xmath17 ) , and hence they naturally live on two staggered , dual meshes . the argument for this approach is not merely a matter of theoretical interest : the geometry of maxwell s equations has important practical implications for numerical performance . for instance , the vector - field - based discretization , used in nodal fem , results in spurious @xmath15-d artifacts due to its failure to respect the underlying geometric structure . the yee scheme , on the other hand , produces resonance spectra in agreement with theory , without spurious modes ( see * ? ? ? * ) . furthermore , it has been shown in @xcite that staggered - grid methods can be used to develop fast numerical methods for electromagnetism , even for problems in heterogeneous media with highly discontinuous material parameters such as conductivity and permeability . by developing a structure - preserving , geometric discretization of maxwell s equations , not only can we better understand the yee scheme and its characteristic advantages , but we can also construct more general methods that share its desirable properties . this family of methods includes the `` yee - like '' scheme of @xcite , which presented the first extension of yee s scheme to unstructured grids ( e.g. , simplicial meshes rather than rectangular lattices ) . general methods like these are highly desirable : rectangular meshes are not always practical or appropriate to use in applications where domains with curved and oblique boundaries are needed ( see , for instance * ? ? ? * ) . by allowing general discretizations while still preserving geometry , one can combine the best attributes of the fem and yee schemes . using dec as a structure - preserving , geometric framework for general discrete meshes , we have obtained the following results : 1 . the yee scheme is actually a variational integrator : that is , it can be obtained by applying hamilton s principle of stationary action to a discrete lagrangian . 2 . consequently , the yee scheme is _ multisymplectic _ and _ preserves discrete momentum maps _ ( i.e. , conserved quantities analogous to the continuous case of electromagnetism ) . in particular , the gauss constraint is understood as a discrete momentum map of this integrator , while the preservation of electrostatic potential solutions corresponds to the identity @xmath18 , where @xmath19 is the discrete exterior derivative operator . we also create a foundation for more general schemes , allowing _ arbitrary discretizations of spacetime _ , not just uniform time steps on a spatial mesh . one such scheme , introduced here , is a new asynchronous variational integrator ( avi ) for maxwell s equations , where each spatial element is assigned its own time step size and evolves `` asynchronously '' with its neighbors . this means that one can choose to take small steps where greater refinement is needed , while still using larger steps for other elements . since refining one part of the mesh does not restrict the time steps taken elsewhere , an avi can be computationally efficient and numerically stable with fewer total iterations . in addition to the avi scheme , we briefly sketch how completely covariant spacetime integrators for electromagnetism can be implemented , without even requiring a 3 + 1 split into space and time components . we will begin by reviewing maxwell s equations : first developing the differential forms expression from a lagrangian variational principle , and next showing how this is equivalent to the familiar vector calculus formulation . we will then motivate the use of dec for computational electromagnetics , explaining how electromagnetic quantities can be modeled using discrete differential forms and operators on a spacetime mesh . these dec tools will then be used to set up the discrete maxwell s equations , and to show that the resulting numerical algorithm yields the yee and bossavit kettunen schemes as special cases , as well as a new avi method . finally , we will demonstrate that the discrete maxwell s equations can also be derived from a discrete variational principle , and will explore its other discrete geometric properties , including multisymplecticity and momentum map preservation . this section quickly reviews the differential forms approach to electromagnetism , in preparation for the associated discrete formulation given in the next section . for more details , the reader can refer to @xcite and @xcite . maxwell s equations , without free sources of charge or current , are traditionally expressed in terms of four vector fields in @xmath15-space : the electric field @xmath13 , magnetic field @xmath14 , electric flux density @xmath20 , and magnetic flux density @xmath4 . to translate these into the language of differential forms , we begin by replacing the electric field with a @xmath21-form @xmath22 and the magnetic flux density by a @xmath23-form @xmath24 . these have the coordinate expressions @xmath25 where @xmath26 and @xmath27 . the motivation for choosing @xmath22 as a @xmath21-form and @xmath24 as a @xmath23-form comes from the integral formulation of faraday s law , @xmath28 where @xmath13 is integrated over curves and @xmath4 is integrated over surfaces . similarly , ampre s law , @xmath29 integrates @xmath14 over curves and @xmath20 over surfaces , so we can likewise introduce a @xmath21-form @xmath30 and a @xmath23-form @xmath31 . now , @xmath13 and @xmath4 are related to @xmath20 and @xmath14 through the usual constitutive relations @xmath32 as shown in @xcite , we can view @xmath33 and @xmath34 as corresponding to hodge operators @xmath35 and @xmath36 , which map the @xmath21-form `` fields '' to @xmath23-form `` fluxes '' in space . therefore , this is compatible with viewing @xmath22 and @xmath30 as @xmath21-forms , and @xmath31 and @xmath24 as @xmath23-forms . note that in a vacuum , with @xmath37 and @xmath38 constant , one can simply express the equations in terms of @xmath13 and @xmath4 , choosing appropriate geometrized units such that @xmath39 , and hence ignoring the distinction between @xmath13 and @xmath20 and between @xmath4 and @xmath14 . this is typically the most familiar form of maxwell s equations , and the one that most students of electromagnetism first encounter . in this presentation , we will restrict ourselves to the vacuum case with geometrized units ; for geometric clarity , however , we will always distinguish between the @xmath21-forms @xmath22 and @xmath30 and the @xmath23-forms @xmath31 and @xmath24 . finally , we can incorporate free sources of charge and current by introducing the _ charge density @xmath15-form _ @xmath40 , as well as the _ current density @xmath23-form _ @xmath41 . these are required to satisfy the continuity of charge condition @xmath42 , which can be understood as a conservation law ( in the finite volume sense ) . in lorentzian spacetime , we can now combine @xmath22 and @xmath24 into a single object , the faraday @xmath23-form @xmath43 there is a theoretical advantage to combining the electric field and magnetic flux into a single spacetime object : this way , electromagnetic phenomena can be described in a relativistically covariant way , without favoring a particular split of spacetime into space and time components . in fact , we can turn the previous construction around : take @xmath16 to be the fundamental object , with @xmath22 and @xmath24 only emerging when we choose a particular coordinate frame . taking the hodge star of @xmath16 , we also get a dual @xmath23-form @xmath44 called the maxwell @xmath23-form . the equation @xmath17 describes the dual relationship between @xmath22 and @xmath24 on one hand , and @xmath31 and @xmath30 on the other , that is expressed in the constitutive relations . likewise , the charge density @xmath45 and current density @xmath46 can be combined into a single spacetime object , the source @xmath15-form @xmath47 having defined @xmath48 in this way , the continuity of charge condition simply requires that @xmath48 be closed , i.e. , @xmath49 . let @xmath50 be the electromagnetic potential @xmath21-form , satisfying @xmath51 , over the spacetime manifold @xmath52 . then define the @xmath0-form lagrangian density @xmath53 and its associated action functional @xmath54 = \int _ x \mathcal{l}.\ ] ] now , take a variation @xmath55 of @xmath50 , where @xmath55 vanishes on the boundary @xmath56 . then the variation of the action functional along @xmath55 is @xmath57 \cdot \alpha & = \left . \frac{\mathrm{d } } { \mathrm{d } \epsilon } \right\rvert _ { \epsilon = 0 } s[a + \epsilon \alpha ] \\ & = \int _ x \left ( - \mathrm{d } \alpha \wedge * \mathrm{d } a + \alpha \wedge \mathcal{j } \right ) \\ & = \int _ x \alpha \wedge \left ( - \mathrm{d } { * \mathrm{d}a } + \mathcal{j } \right ) , \end{aligned}\ ] ] where in this last equality we have integrated by parts , using the fact that @xmath55 vanishes on the boundary . hamilton s principle of stationary action requires this variation to be equal to zero for arbitrary @xmath55 , thus implying the electromagnetic euler lagrange equation , @xmath58 since @xmath59 , then clearly is equivalent to @xmath60 . furthermore , since @xmath61 , it follows that @xmath62 . hence , maxwell s equations with respect to the maxwell and faraday @xmath23-forms can be written as @xmath63 suppose now we choose the standard coordinate system @xmath64 on minkowski space @xmath65 , and define @xmath22 and @xmath24 through the relation @xmath66 . then a straightforward calculation shows that is equivalent to @xmath67 likewise , if @xmath68 , then is equivalent to @xmath69 hence this lagrangian , differential forms approach to maxwell s equations is _ strictly equivalent _ to the more classical vector calculus formulation in smooth spacetime . however , in discrete spacetime , we will see that the differential forms version is _ not _ equivalent to an arbitrary vector field discretization , but rather implies a particular choice of discrete objects . we can also derive maxwell s equations by using a mixed variational principle , similar to the hamilton pontryagin principle introduced by @xcite for classical lagrangian mechanics . to do this , we treat @xmath50 and @xmath16 as separate fields , while @xmath70 acts as a lagrange multiplier , weakly enforcing the constraint @xmath71 . define the extended action to be @xmath72 = \int _ x \left [ - \frac{1}{2 } f \wedge * f + a \wedge \mathcal{j } + \left ( f - \mathrm{d } a \right ) \wedge g \right ] .\ ] ] then , taking the variation of the action along some @xmath73 ( vanishing on @xmath56 ) , we have @xmath74 \cdot \left ( \alpha , \phi , \gamma \right ) & = \int _ x \left [ - \phi \wedge * f + \alpha \wedge \mathcal{j } + \left ( \phi - \mathrm{d } \alpha \right ) \wedge g + \left ( f - \mathrm{d } a \right ) \wedge \gamma \right ] \\ & = \int _ x \left [ \alpha \wedge \left ( \mathcal{j } - \mathrm{d } g \right ) + \phi \wedge \left ( g - * f \right ) + \left ( f - \mathrm{d } a \right ) \wedge \gamma \right ] .\end{aligned}\ ] ] therefore , setting this equal to zero , we get the equations @xmath75 this is precisely equivalent to maxwell s equations , as derived above . however , this approach provides some additional insight into the geometric structure of electromagnetics : the gauge condition @xmath51 and constitutive relations @xmath17 are _ explicitly included in the equations of motion _ , as a direct result of the variational principle . when solving an initial value problem , it is not necessary to use all of maxwell s equations to evolve the system forward in time . in fact , the curl equations and automatically conserve the quantities @xmath76 and @xmath77 . therefore , the divergence equations and can be viewed simply as constraints on initial conditions , while the curl equations completely describe the time evolution of the system . there are a number of ways to see why we can justify eliminating the divergence equations . a straightforward way is to take the divergence of equations and . since @xmath78 , we are left with @xmath79 therefore , if the divergence constraints are satisfied at the initial time , then they are satisfied for all time , since the divergence terms are constant . another approach is to notice that maxwell s equations depend only on the exterior derivative @xmath80 of the electromagnetic potential , and not on the value of @xmath50 itself . therefore , the system has a _ gauge symmetry _ : any gauge transformation @xmath81 leaves @xmath80 , and hence maxwell s equations , unchanged . choosing a time coordinate , we can then partially fix the gauge so that the electric scalar potential @xmath82 ( the so - called weyl gauge or temporal gauge ) , and so @xmath50 has only spatial components . in fact , these three remaining components correspond to those of the usual vector potential @xmath5 . the reduced euler lagrange equations in this gauge consist only of , while the remaining gauge symmetry @xmath83 yields a momentum map that automatically preserves @xmath84 in time . equations and are automatically preserved by the identity @xmath85 ; they are not actually part of the euler lagrange equations . a more detailed exposition of these calculations will be given in . in this section , we give a quick review of the fundamental objects and operations of discrete exterior calculus ( dec ) , a structure - preserving calculus of discrete differential forms . by construction , dec automatically preserves a number of important geometric structures , and hence it provides a fully discrete analog of the tools used in the previous section to express the differential forms version of maxwell s equations . in subsequent sections , we will use this framework to formulate maxwell s equations discretely , emulating the continuous version . modern computational electromagnetism started in the 1960s , when the finite element method ( fem ) , based on _ nodal basis functions _ , was used successfully to discretize the differential equations governing @xmath23-d static problems formulated in terms of a scalar potential . unfortunately , the initial success of the fem approach appeared unable to carry over to @xmath15-d problems without spurious numerical artifacts . with the introduction of _ edge elements _ in @xcite came the realization that a better discretization of the geometric structure of maxwell s electromagnetic theory was key to overcoming this obstacle ( see @xcite for more historical details ) . mathematical tools developed by weyl and whitney in the 1950s , in the context of algebraic topology , turned out to provide the necessary foundations on which robust numerical techniques for electromagnetism can be built , as detailed in @xcite . in this section , we show how to define differential forms and operators on a discrete mesh , in preparation to use this framework for computational modeling of classical fields . by construction , the calculus of discrete differential forms automatically preserves a number of important geometric structures , including stokes theorem , integration by parts ( with a proper treatment of boundaries ) , the de rham complex , poincar duality , poincar s lemma , and hodge theory . therefore , this provides a suitable foundation for the coordinate - free discretization of geometric field theories . in subsequent chapters , we will also use these discrete differential forms as the space of fields on which we will define discrete lagrangian variational principles . the particular `` flavor '' of discrete differential forms and operators we will be using is known as _ discrete exterior calculus _ , or _ dec _ for short ; see @xcite . ( for related efforts in this direction , see also @xcite and @xcite . ) guided by cartan s exterior calculus of differential forms on smooth manifolds , dec is a discrete calculus developed , _ ab initio _ , on discrete manifolds , so as to maintain the covariant nature of the quantities involved . this computational tool is based on the notion of discrete chains and cochains , used as basic building blocks for compatible discretizations of important geometric structures such as the de rham complex @xcite . the chain and cochain representations are not only attractive from a computational perspective due to their conceptual simplicity and elegance ; as we will see , they also originate from a theoretical framework defined by @xcite , who introduced the whitney and de rham maps that establish an isomorphism between simplicial cochains and lipschitz differential forms . dec is concerned with problems in which the smooth @xmath86-dimensional manifold @xmath52 is replaced by a discrete mesh precisely , by a cell complex that is manifold , admits a metric , and is orientable . the simplest example of such a mesh is a finite simplicial complex , such as a triangulation of a @xmath23-dimensional surface . we will generally denote the complex by @xmath87 , and a cell in the complex by @xmath88 . given a mesh @xmath87 , one can construct a _ dual mesh _ @xmath89 , where each @xmath90-cell @xmath91 corresponds to a dual @xmath92-cell @xmath93 . ( @xmath89 is `` dual '' to @xmath94 in the sense of a graph dual . ) one way to do this is as follows : place a dual vertex at the circumcenter of each @xmath86-simplex , then connect two dual vertices by an edge wherever the corresponding @xmath86-simplices share an @xmath95-simplex , and so on . this is called the _ circumcentric dual _ , and it has the important property that primal and dual cells are automatically orthogonal to one another , which is advantageous when defining an inner product ( as we will see later in this section ) . for example , the circumcentric dual of a delaunay triangulation , with the euclidean metric , is its corresponding voronoi diagram ( see ) . for more on the dual relationship between delaunay triangulations and voronoi diagrams , a standard reference is @xcite . a similar construction of the circumcenter can be carried out for higher - dimensional euclidean simplicial complexes , as well as for simplicial meshes in minkowski space . note that , in both the euclidean and lorentzian cases , the circumcenter may actually lie _ outside _ the simplex if it has a very bad aspect ratio , underscoring the importance of mesh quality for good numerical results . there are alternative ways to define the dual mesh for example , placing dual vertices at the barycenter rather than the circumcenter but we will use the circumcentric dual unless otherwise noted . note that a refined definition of the dual mesh , where dual cells at the boundary are restricted to @xmath87 , will be discussed in to allow proper enforcement of boundary conditions in computational electromagnetics . -d simplicial mesh ( left ) , we can construct its circumcentric dual mesh , called the voronoi diagram of the primal mesh ( right ) . in bold , we show one particular primal edge @xmath96 ( left ) and its corresponding dual edge @xmath97 ( right ) ; the convex hull of these cells @xmath98 is shaded dark grey . ] the fundamental objects of dec are discrete differential forms . a discrete @xmath90-form @xmath99 assigns a real number to each oriented @xmath90-dimensional cell @xmath100 in the mesh @xmath87 . ( the superscripts @xmath90 are not actually required by the notation , but they are often useful as reminders of what order of form or cell we are dealing with . ) this value is denoted by @xmath101 , and can be thought of as the value of @xmath102 `` integrated over '' the element @xmath100 , i.e. , @xmath103 for example , @xmath104-forms assign values to vertices , @xmath21-forms assign values to edges , etc . we can extend this to integrate over discrete paths by linearity : simply add the form s values on each cell in the path , taking care to flip the sign if the path is oriented opposite the cell . formally , these `` paths '' of @xmath90-dimensional elements are called _ chains _ , and discrete differential forms are _ cochains _ , where @xmath105 is the pairing between cochains and chains . differential forms can be defined either on the mesh @xmath87 or on its dual @xmath89 . we will refer to these as _ primal forms _ and _ dual forms _ respectively . note that there is a natural correspondence between primal @xmath90-forms and dual @xmath92-forms , since each primal @xmath90-cell has a dual @xmath92-cell . this is an important property that will be used below to define the discrete hodge star operator . the discrete exterior derivative @xmath19 is constructed to satisfy stokes theorem , which in the continuous sense is written @xmath106 therefore , if @xmath55 is a discrete differential @xmath90-form , then the @xmath107-form @xmath108 is defined on any @xmath109-chain @xmath91 by @xmath110 where @xmath111 is the @xmath112-chain boundary of @xmath113 . for this reason , @xmath19 is often called the _ coboundary _ operator in cohomology theory . the discrete hodge star transforms @xmath90-forms on the primal mesh into @xmath92-forms on the dual mesh , and vice - versa . in our setup , we will use the so - called diagonal ( or mass - lumped ) approximation of the hodge star @xcite because of its simplicity , but note that higher - order accurate versions can be substituted . given a discrete form @xmath55 , its _ hodge star _ @xmath114 is defined by the relation @xmath115 where @xmath116 and @xmath117 are the volumes of these elements , and @xmath118 is the causality operator , which equals @xmath119 when @xmath113 is spacelike and @xmath120 otherwise . ( for more information on alternative discrete hodge operators , the reader may refer to , e.g. , @xcite . ) define the _ @xmath121 inner product _ @xmath122 between two primal @xmath90-forms to be @xmath123 where the sum is taken over all @xmath90-dimensional elements @xmath91 , and @xmath124 is the @xmath86-dimensional convex hull of @xmath125 ( see ) . the final equality holds as a result of using the circumcentric dual , since @xmath91 and @xmath126 are orthogonal to one another , and hence @xmath127 . ( indeed , this is one of the advantages of using the circumcentric dual , since one only needs to store volume information about the primal and dual cells themselves , and not about these primal - dual convex hulls . ) this inner product can be expressed in terms of @xmath128 , as in the continuous case , for a particular choice of the discrete primal - dual wedge product ; see @xcite . note that since we have already defined a discrete version of the operators @xmath129 and @xmath130 , we immediately have a discrete codifferential @xmath131 , with the same formal expression as given previously . see for a visual diagram of primal and dual discrete forms , along with the corresponding operators @xmath132 , for the case where @xmath87 is a @xmath15-d tetrahedral mesh . -d simplicial mesh . in the top row , we see how a discrete @xmath90-form lives on @xmath90-cells of the primal mesh , for @xmath133 ; the bottom row shows the location of the corresponding dual @xmath92-forms on the dual mesh . the differential operators @xmath129 and @xmath131 map `` horizontally '' between @xmath90 and @xmath134 forms , while the hodge star @xmath130 and its inverse @xmath135 map `` vertically '' between primal and dual forms . ] dec can be implemented simply and efficiently using linear algebra . a @xmath90-form @xmath55 can be stored as a vector , where its entries are the values of @xmath55 on each @xmath90-cell of the mesh . that is , given a list of @xmath90-cells @xmath136 , the entries of the vector are @xmath137 . the exterior derivative @xmath129 , taking @xmath90-forms to @xmath134-forms , is then represented as a matrix : in fact , it is precisely the _ incidence matrix _ between @xmath90-cells and @xmath134-cells in the mesh , with sparse entries @xmath138 . the hodge star taking primal @xmath90-forms to dual @xmath92-forms becomes a square matrix , and in the case of the diagonal hodge star , it is the diagonal matrix with entries @xmath139 . the discrete inner product is then simply the hodge star matrix taken as a quadratic form . because of this straightforward isomorphism between dec and linear algebra , problems posed in the language of dec can take advantage of existing numerical linear algebra codes . for more details on programming and implementation , refer to @xcite . particular care is required to properly enforce initial and boundary conditions on the discrete spacetime boundary @xmath140 . for example , in electromagnetism , we may wish to set initial conditions for @xmath22 and @xmath24 at time @xmath141 but while @xmath24 is defined on @xmath142 at @xmath141 , @xmath22 is not . in fact , as we will see , @xmath22 lives on edges that are extruded between the time slices @xmath143 and @xmath144 , so unless we modify our definitions , we can only initialize @xmath22 at the half - step @xmath145 . ( this half - step issue also arises with the standard yee scheme . ) there are some applications where it may be acceptable to initialize @xmath22 and @xmath24 at separate times ( for example , when the fields are initialized randomly and integrated for a long time to compute a resonance spectrum ) , but we wish to be able to handle the more general case . although our previous exposition of dec thus far applies anywhere away from a boundary , notions as simple as `` dual cell '' need to be defined carefully on or near @xmath140 . for a primal mesh @xmath87 , the dual mesh @xmath146 is defined as _ the voronoi dual of @xmath87 restricted to @xmath87_. this truncates the portion of the dual cells extending outside of @xmath147 ; compare with the earlier . this new definition results in the addition of a dual vertex at the circumcenter of each boundary @xmath95-simplex , in addition to the interior @xmath86-simplices as previously defined . to complete the dual mesh @xmath146 , we add a dual edge between adjacent dual vertices on the boundary , as well as between dual boundary vertices and their neighboring interior dual vertices , and proceed similarly with higher - dimensional dual cells . for intuition , one can imagine the @xmath95-dimensional boundary to be a vanishingly thin @xmath86-dimensional shell . that is , each boundary @xmath148-simplex can be thought of as a prismal @xmath90-cell that has been `` squashed flat '' along the boundary normal direction . this process is quite similar to the use of `` ghost cells '' at the boundary , as is commonly done for finite volume methods ( see * ? ? ? note that these additional dual cells provide the boundary @xmath149 with its own dual mesh @xmath150 . in fact , the boundary of the dual is now equal to the dual of the boundary , i.e. , @xmath151 . -d example , the dual mesh is properly defined near the boundary by adding dual vertices on the boundary edges . the restricted voronoi cells of the primal boundary vertices ( shaded at right ) thus have boundaries containing both dual edges ( dashed lines ) _ and _ primal boundary half - edges . ] returning to the example of initial conditions on @xmath22 and @xmath24 , we recall that @xmath22 is defined on extruded faces normal to the time slice @xmath143 . therefore , thanks to the proper restriction of the voronoi diagram to the domain , we can now define @xmath22 on _ edges _ in @xmath152 at time @xmath143 , where these edges can be understood as vanishingly thin faces ( i.e. , extruded between some @xmath153 and @xmath143 for @xmath154 ) . notice finally that with this construction of @xmath146 , there is a dual relationship between dirichlet conditions on the dual mesh and neumann conditions on the primal mesh , e.g. , between primal fields and dual fluxes , as expected . with the dual mesh properly defined , dual forms can now be defined on the boundary . therefore , the discrete duality between @xmath155 and @xmath156 can be generalized to include nonvanishing boundary terms . if @xmath55 is a primal @xmath148-form and @xmath157 is a primal @xmath90-form , then @xmath158 in the boundary integral , @xmath55 is still a primal @xmath148-form on @xmath159 , while @xmath160 is an @xmath92-form taken on the boundary dual @xmath161 . formula is readily proved using the familiar method of discrete `` summation by parts , '' and thus agrees with the integration by parts formula for smooth differential forms . in this section , we explain how to obtain numerical algorithms for solving maxwell s equations with dec . to do so , we will proceed in the following order . first , we will find a sensible way to define the discrete forms @xmath16 , @xmath70 , and @xmath48 on a spacetime mesh . next , we will use the dec version of the operators @xmath19 and @xmath162 to obtain the discrete maxwell s equations . while we havent yet shown that these equations are variational in the discrete sense , we will show later in that the lagrangian derivation of the smooth maxwell s equations also holds with the dec operators , in precisely the same way . finally , we will discuss how these equations can be used to define a numerical method for computational electromagnetics . in particular , for a rectangular grid , we will show that our setup results in the traditional yee scheme . for a general triangulation of space with equal time steps , the resulting scheme will be bossavit and kettunen s scheme . we will then develop an avi method , where each spatial element can be assigned a different time step , and the time integration of maxwell s equations can be performed on the elements asynchronously . finally , we will comment on the equations for fully generalized spacetime meshes , e.g. , an arbitrary meshing of @xmath163 by @xmath0-simplices . note that the idea of discretizing maxwell s equations using spacetime cochains was mentioned in , e.g. , @xcite , as well as in a paper by @xcite taking the more abstract perspective of higher - level `` @xmath164-form '' versions of electromagnetism and category theory . suppose that we have a rectangular grid in @xmath165 , oriented along the axes @xmath64 . to simplify this exposition ( although it is not necessary ) , let us also suppose that the grid has uniform space and time steps @xmath166 . note that the dec setup applies directly to a non - simplicial rectangular mesh , since an @xmath86-rectangle does in fact have a circumcenter . since @xmath16 is a @xmath23-form , its values should live on @xmath23-faces in this grid . following the continuous expression of @xmath16 @xmath167 and due to the tensor product nature of the regular grid , the exact assignment of each @xmath23-face becomes simple : _ the six components of @xmath16 correspond precisely to the six types of @xmath23-faces in a @xmath0-d rectangular grid_. simply assign the values @xmath168 to faces parallel to the @xmath169-plane , @xmath170 to faces parallel to the @xmath171-plane , and @xmath172 to faces parallel to the @xmath173-plane . likewise , assign @xmath174 to faces parallel to the @xmath175-plane , @xmath176 to faces parallel to the @xmath177-plane , and @xmath178 to faces parallel to the @xmath179-plane . this is pictured in . are stored on the primal @xmath23-faces of a @xmath0-d rectangular grid . shown here are the three mixed space / time @xmath15-cells , and the one purely spatial @xmath15-cell ( lower right ) . ] let us look at these values on the faces of a typical @xmath0-rectangle @xmath180 \times \left [ y _ l , y _ { l + 1 } \right ] \times \left [ z _ m , z _ { m + 1 } \right ] \times \left [ t _ n , t _ { n + 1 } \right ] $ ] . to simplify the notation , we can index each value of @xmath16 by the midpoint of the @xmath23-face on which it lives : for example , @xmath181 is stored on the face @xmath182 \times \left\{y _ l\right\ } \times \left\{z _ m\right\ } \times \left [ t _ n , t _ { n + 1 } \right ] $ ] , parallel to the @xmath183-plane . hence , the following values are assigned to the corresponding faces : @xmath184 @xmath185 we see that a `` staggered grid '' arises from the fact that @xmath22 and @xmath24 naturally live on @xmath23-faces , not at vertices or @xmath0-cells . the discrete equations of motion are , as in the continuous case , @xmath186 where now these equations are interpreted in the sense of dec . let us first look at the dec interpretation of @xmath187 . since @xmath188 is a discrete @xmath15-form , it takes values on the @xmath15-faces of each @xmath0-rectangle . its values are as follows : @xmath189 @xmath190 @xmath191 @xmath192 setting each of these equal to zero , we arrive at the following four equations : @xmath193 @xmath194 @xmath195 and @xmath196 these equations are the discrete version of the equations @xmath197 moreover , since @xmath22 and @xmath24 are differential forms , this can also be seen as a discretization of the _ integral version _ of maxwell s equations as well ! because dec satisfies a discrete stokes theorem , this automatically preserves the equivalence between the differential and integral formulations of electromagnetism . doing the same with the equation @xmath60 , evaluating on dual @xmath15-faces this time , we arrive at four more equations : @xmath198 @xmath199 @xmath200 and @xmath201 this results from storing @xmath202 on the dual grid , as shown in . this set of equations is the discrete version of @xmath203 are stored on dual @xmath23-faces in a rectangular grid . shown here are a mixed space / time dual @xmath15-cell ( left ) , corresponding to a spacelike primal edge ; and a purely spatial dual @xmath15-cell ( right ) , corresponding to a timelike primal edge . there are also two other mixed space / time cells , as in , that are not shown here . ] after eliminating the redundant divergence equations and ( see for details ) and making the substitutions @xmath204 , @xmath205 , the remaining equations are precisely the yee scheme , as formulated in @xcite . we now consider the case of an unstructured grid in space , but with uniform steps in time as advocated in , e.g. , @xcite . suppose that , instead of a rectangular grid for both space and time , we have an arbitrary space discretization on which we would like to take uniform time steps . ( for example , we may be given a tetrahedral mesh of the spatial domain . ) this mesh contains two distinct types of @xmath23-faces . first , there are triangular faces that live entirely in the space mesh at a single position in time . every edge of such a face is spacelike that is , it has positive length so the causality operator defined in takes the value @xmath206 . second , there are rectangular faces that live between time steps . these faces consist of a single spacelike edge extruded by one time step . because they have one timelike edge , these faces satisfy @xmath207 . again , the circumcentric - dual dec framework applies directly to this type of mesh , since the prismal extrusion of a @xmath15-simplex still has a circumcenter . again , we can characterize the discrete values of @xmath16 by looking at the continuous expression @xmath208 therefore , let us assign @xmath24 to the purely spacelike faces and @xmath209 to the mixed space / time faces . looking at @xmath17 shows that mixed dual faces should store @xmath210 and spacelike dual faces should store @xmath31 ; see . is stored on primal @xmath23-faces ( left ) , while @xmath17 is stored on dual @xmath23-faces ( right ) . shown here are the values on mixed space / time @xmath15-cells . ( the purely spatial @xmath15-cells , which correspond to the divergence equations and do not contribute to the equations of motion , are not shown . ) ] as in @xcite , we can store the values of each differential form over every spatial element in an array , using the method described in . this leads to the arrays @xmath211 and @xmath212 at whole time steps @xmath86 , and @xmath213 and @xmath214 at half time steps . let @xmath215 denote the edges - to - faces incidence matrix for the spatial domain . that is , @xmath216 is the matrix corresponding to the discrete exterior derivative , taken only in space , from primal @xmath21-forms to primal @xmath23-forms . similarly , the transpose @xmath217 corresponds to the exterior derivative from spatial dual @xmath21-forms to dual @xmath23-forms . then the equation @xmath218 , evaluated on all prismal @xmath15-faces becomes @xmath219 likewise , the equation @xmath60 , evaluated on all space / time @xmath15-faces in the dual mesh , becomes @xmath220 we can also evaluate @xmath218 and @xmath221 on spacelike @xmath15-faces , e.g , tetrahedra ; these simply yield the discrete versions of the divergence conditions for @xmath24 and @xmath31 , which can be eliminated . therefore , the dec scheme for such a mesh is equivalent to bossavit and kettunen s yee - like scheme ; additionally , when the spatial mesh is taken to be rectangular , this integrator reduces to the standard yee scheme . however , we now have solid foundations to extend this integrator to handle asynchronous updates for improved efficiency . instead of picking the same time step size for every element of the spatial mesh , as in the previous two sections , it is often more efficient to assign each element its own , optimized time step , as done in @xcite for problems in elastodynamics . in this case , rather than the entire mesh evolving forward in time simultaneously , individual elements advance one - by - one , asynchronously hence the name _ asynchronous variational integrator _ ( avi ) . as we will prove in , this asynchronous update process will maintain the variational nature of the integration scheme . here , we again allow the spatial mesh to be unstructured . after choosing a primal space mesh , assign each spatial @xmath23-face ( e.g. , triangle ) @xmath91 its own discrete time set @xmath222 for example , one might assign each face a fixed time step size @xmath223 , taking equal time steps _ within _ each element , but with @xmath224 varying _ across _ elements . we further require for simplicity of explanation that , except for the initial time , no two faces take the same time step : that is , @xmath225 for @xmath226 . in order to keep proper time at the edges @xmath227 where multiple faces with different time sets meet , we let @xmath228 therefore the mixed space - time @xmath23-faces , which correspond to the edge @xmath227 extruded over a time step , are assigned the set of intermediate times @xmath229 where @xmath230 . the values stored on a primal avi mesh are shown in . since @xmath231 when @xmath232 , each spatial edge @xmath227 takes more time steps than any one of its incident faces @xmath91 ; as a result , it is not possible in general to construct a circumcentric dual on the entire spacetime avi mesh , since the mesh is not prismal and hence the circumcenter may not exist . instead , we find the circumcentric dual to the _ spatial _ mesh , and assign same time steps to the primal and dual elements @xmath233 this results in well - defined primal and dual cells for each @xmath23-element in spacetime , and hence a hodge star for this order . ( a hodge star on forms of different order is not needed to formulate maxwell s equations . ) the equation @xmath218 , evaluated on a mixed space / time @xmath15-cell , becomes @xmath234 similarly , the equation @xmath60 becomes @xmath235 where @xmath236 equals @xmath21 when face @xmath91 has @xmath237 for some @xmath86 , and @xmath104 otherwise . ( that is , the indicator function `` picks out '' the incident face that lives at the same time step as this edge . ) solving an initial value problem can then be summarized by the following update loop : 1 . pick the minimum time @xmath238 where @xmath239 has not yet been computed . 2 . advance @xmath240 according to . update @xmath241 . 4 . advance @xmath242 on neighboring edges @xmath243 according to . update @xmath244 . as detailed in @xcite for elastodynamics , the explicit avi update scheme can be implemented by selecting mesh elements from a priority queue , sorted by time , and iterating forward . however , as written above , the scheme is not strictly iterative , since depends on past values of @xmath22 . this can be easily fixed by rewriting the avi scheme to advance in the variables @xmath50 and @xmath22 instead , where the potential @xmath50 effectively stores the cumulative contribution of @xmath245 to the value of @xmath246 on neighboring faces . compared to the avi for elasticity , @xmath50 plays the role of the positions @xmath247 , while @xmath22 plays the role of the ( negative ) velocities @xmath248 . the algorithm is given as pseudocode in . note that if all elements take uniform time steps , the avi reduces to the bossavit kettunen scheme . ' '' '' @xmath249 _ // store initial field values and times _ @xmath250 compute the next update time @xmath251 @xmath252.push@xmath253 _ // push element onto queue with its next update time _ @xmath254 @xmath252.pop ( ) _ // pop next element @xmath91 and time @xmath255 from queue _ @xmath256 _ // update neighboring values of @xmath50 at time @xmath255 _ @xmath257 @xmath258 @xmath259 @xmath260 @xmath261 @xmath262 _ // update element s time _ compute the next update time @xmath263 @xmath252.push@xmath264 _ // schedule @xmath91 for next update _ ' '' '' we first present a simple numerical example demonstrating the good energy behavior of our asynchronous integrator . the avi was used to integrate in time over a @xmath23-d rectangular cavity with perfectly electrically conducting ( pec ) boundaries , so that @xmath22 vanishes at the boundary of the domain . @xmath22 was given random values at the initial time , so as to excite all frequency modes , and integrated for 8 seconds . each spatial element was given a time step equal to 1/10 of the stability - limiting time step determined by the cfl condition . this simulation was done for two different spatial discretizations . the first is a uniform discretization so that each element has identical time step size , which coincides exactly with the yee scheme . the second discretization randomly partitioned the @xmath265- and @xmath266-axes , so that each element has completely unique spatial dimensions and time step size , and so the update rule is truly asynchronous . the energy plot for the uniform yee discretization is shown in , while the energy for the random discretization is shown in . even for a completely random , irregular mesh , our asynchronous integrator displays near - energy preservation qualities . such numerical behavior stems from the variational nature of our integrator , which will be detailed in . in addition , we tested the performance of the avi method with regard to computing the resonant frequencies of a @xmath15-d rectangular cavity , but using an _ unstructured _ tetrahedral spatial mesh . while the resonant frequencies are relatively simple to compute analytically , nodal finite element methods are well known to produce spurious modes for this type of simulation . by contrast , as shown in , the avi simulation produces a resonance spectrum in close agreement with theory . furthermore , by refining the mesh close to the spatial boundary , while using a coarser discretization in the interior , we were able to achieve these results with less computational effort than a uniformly fine mesh would require , since the time steps were selected to be proportional to the respective element sizes . finally , we look at the most general possible case : an arbitrary discretization of spacetime , such as a simplicial @xmath0-complex . such a mesh is completely relativistically covariant , so that @xmath16 can not be objectively separated into the components @xmath22 and @xmath24 without a coordinate frame . in most engineering applications , relativistic effects are insignificant , so a 3 + 1 mesh ( as in the previous subsections ) is almost always adequate , and avoids the additional complications of spacetime mesh construction . still , we expect that there are scientific applications where a covariant discretization of electromagnetism may be very useful . for example , many implementations of numerical general relativity ( using regge calculus for instance ) are formulated on simplicial @xmath0-complexes ; one might wish to simulate the interaction of gravity with the electromagnetic field , or charged matter , on such a mesh . first , a quick caution on mesh construction : since the lorentz metric is not positive definite , it is possible to create edges that have length @xmath104 , despite connecting two distinct points in @xmath267 ( so - called `` null '' or `` lightlike '' edges ) . meshes containing such edges are degenerate akin to a euclidean mesh containing a triangle with two identical points . in particular , the dec hodge star is undefined for @xmath104-volume elements ( due to division by zero ) . even without @xmath104-volume elements , it is still possible for a spacetime mesh to violate causality , so extra care should be taken . methods to construct causality - respecting spacetime meshes over a given spatial domain can be found in , e.g. , @xcite and @xcite . when the mesh contains no inherent choice of a time direction , there is no canonical way to split @xmath16 into @xmath22 and @xmath24 . therefore , one must set up the problem by assigning values of @xmath16 directly to @xmath23-cells ( or equivalently , assigning values of @xmath50 to @xmath21-cells ) . for initial boundary value problems , one might choose to have the initial and final time steps be prismal , so that @xmath22 and @xmath24 can be used for initial and final values , while the internal discretization is general . the equations @xmath218 and @xmath60 can be implemented directly in dec . since this mesh is generally unstructured , there is no simple algorithm as the ones we presented above . instead , the equations on @xmath16 results in a sparse linear system which , given proper boundary conditions , can be solved globally with direct or iterative solvers . however , it is clear that the previous three examples that the methods of yee , bossavit kettunen , and our avi integrator are special cases where the global solution is particularly simple to compute via synchronous or asynchronous time updates . it is known that , while variational integrators in mechanics do not preserve energy exactly , they have excellent energy behavior , in that it tends to oscillate close to the exact value . this is only true , however , when the integrator takes time steps of uniform size ; adaptive and other non - uniform stepping approaches can give poor results unless additional measures are taken to enforce good energy behavior . ( see ( * ? ? ? * chapter viii ) for a good discussion of this problem for mechanics applications . ) therefore , there is no reason to expect that _ arbitrary _ meshes of spacetime will yield energy results as good as the yee , bossavit kettunen , and avi schemes . however , if one is taking a truly covariant approach to spacetime , `` energy '' is not even defined without specifying a time coordinate . likewise , one would not necessarily expect good energy behavior from the other methods with respect to an arbitrary transformation of spatial coordinates . which sort of mesh to choose is thus highly application - dependent . in this section , we complete our exposition with a number of theoretical results about the discrete and continuous maxwell s equations . in particular , we show that the dec formulation of electromagnetism derives from a discrete lagrangian variational principle , and that this formulation is consequently multisymplectic . furthermore , we explore the gauge symmetry of maxwell s equations , and detail how a particular choice of gauge eliminates the equation for @xmath84 from the euler lagrange equations , while preserving it automatically as a momentum map . the discrete maxwell s equations are variational . the idea of this proof is to emulate the derivation of the continuous maxwell s equations from . interpreting this in the sense of dec , we will obtain the discrete maxwell s equations . given a discrete @xmath21-form @xmath50 and dual source @xmath15-form @xmath48 , define the discrete lagrangian @xmath0-form @xmath268 with the corresponding discrete action principle @xmath269 = \left\langle \mathcal{l } _ d , k \right\rangle . \nonumber\ ] ] then , taking a discrete @xmath21-form variation @xmath55 vanishing on the boundary , the corresponding variation of the action is @xmath270 \cdot \alpha = \left\langle - \mathrm{d } \alpha \wedge * \mathrm{d } a + \alpha \wedge \mathcal{j } , k \right\rangle = \left\langle \alpha \wedge \left ( - \mathrm{d } { * \mathrm{d } a } + \mathcal{j } \right ) , k \right\rangle . \nonumber\ ] ] ( here we use the bold @xmath271 to indicate that we are differentiating over the _ smooth space of discrete forms _ @xmath50 , as opposed to differentiating over discrete spacetime , for which we use @xmath19 . ) setting this equal to @xmath104 for all variations @xmath55 , the resulting discrete euler lagrange equations are therefore @xmath272 . defining the discrete @xmath23-forms @xmath71 and @xmath17 , this implies @xmath273 and @xmath274 , the discrete maxwell s equations . the concept of _ multisymplecticity _ for lagrangian field theories was developed in @xcite , where it was shown to arise from the boundary terms for general variations of the action , i.e. , those not restricted to vanish at the boundary . as originally presented , the cartan form @xmath275 is an @xmath276-form , where the @xmath86-dimensional boundary integral is then obtained by contracting @xmath275 with a variation . the multisymplectic @xmath277-form @xmath278 is then given by @xmath279 . contracting @xmath278 with two arbitrary variations gives an @xmath86-form that vanishes when integrated over the boundary , a result called the _ multisymplectic form formula _ , which results from the identity @xmath280 . in the special case of mechanics , where @xmath281 , the boundary consists of the initial and final time points ; hence , this implies the usual result that the symplectic @xmath23-form @xmath282 is preserved by the time flow . alternatively , as communicated to us by @xcite , one can view the cartan form @xmath283 as an @xmath86-form - valued @xmath21-form , and the multisymplectic form @xmath278 as an @xmath86-form - valued @xmath23-form . therefore , one simply evaluates these forms on tangent variations to obtain a boundary integral , rather than taking contractions . these two formulations are equivalent on smooth spaces . however , we will adopt patrick s latter definition , since it is more easily adapted to problems on discrete meshes : @xmath284 and @xmath278 remain smooth @xmath21- and @xmath23-forms , respectively , but their @xmath86-form values are now taken to be discrete . see for an illustration of the discrete multisymplectic form formula . the discrete maxwell s equations are multisymplectic . let @xmath285 be an arbitrary subcomplex , and consider the discrete action functional @xmath286 restricted to @xmath87 . suppose now that we take a discrete variation @xmath55 , _ without _ requiring it to vanish on the boundary @xmath142 . then variations of the action contain an additional boundary term @xmath270 \cdot \alpha = \left\langle \alpha \wedge \left ( - \mathrm{d } { * \mathrm{d } a } + \mathcal{j } \right ) , k \right\rangle + \left\langle \alpha \wedge * d a , \partial k \right\rangle . \nonumber\ ] ] restricting to the space of potentials @xmath50 that satisfy the discrete euler lagrange equations , the first term vanishes , leaving only @xmath287 then we can define the _ cartan form _ @xmath288 by @xmath289 since @xmath290 takes a tangent vector @xmath55 and produces a discrete @xmath15-form on the boundary of the subcomplex , it is a _ smooth 1-form taking discrete @xmath15-form values_. now , since the _ space of discrete forms _ is itself actually continuous , we can take the exterior derivative in the smooth sense on both sides of . evaluating along another first variation @xmath157 ( again restricted to the space of euler lagrange solutions ) , we then get @xmath291 \cdot \alpha \cdot \beta = \left\langle \mathbf{d } \theta \cdot \alpha \cdot \beta , \partial k \right\rangle . \nonumber\ ] ] finally , defining the multisymplectic form @xmath292 , and using the fact that @xmath293 , we get the relation @xmath294 for all variations @xmath295 ; is a discrete version of the multisymplectic form formula . since this holds for any subcomplex @xmath87 , it follows that these schemes are multisymplectic . -d asynchronous - time mesh @xmath296 , where the shaded region is an arbitrary subcomplex @xmath285 . given any two variations @xmath297 of the field , and the multisymplectic form @xmath298 , the formula states that @xmath299 vanishes when integrated over the boundary @xmath140 ( shown in bold ) . ] we now explore the symmetry of maxwell s equations under gauge transformations . this symmetry allows us to reduce the equations by eliminating the time component of @xmath50 ( for some chosen time coordinate ) , effectively fixing the electric scalar potential to zero . because this is an incomplete gauge , there is a remaining gauge symmetry , and hence a conserved momentum map . this conserved quantity turns out to be the charge density @xmath300 , which justifies its elimination from the euler lagrange equations . these calculations are done with differential forms and exterior calculus , hence they apply equally to the smooth and discrete cases of electromagnetism . because maxwell s equations only depend on @xmath80 , they are invariant under gauge transformations of the form @xmath81 , where @xmath301 is any scalar function on spacetime . if we fix a time coordinate , we can now choose the weyl gauge , so that the time component @xmath302 . therefore , we can assume that @xmath303 in fact , @xmath304 are precisely the components of the familiar vector potential @xmath5 , i.e. , @xmath305 . having fixed the gauge and chosen a time coordinate , we can now define two new `` partial exterior derivative '' operators , @xmath306 ( time ) and @xmath307 ( space ) , where @xmath308 . since @xmath50 contains no @xmath309 terms , @xmath310 is a @xmath23-form containing only the space terms of @xmath311 , while @xmath312 contains the terms involving both space and time . that is , @xmath313 restricted to this subspace of potentials , the lagrangian density then becomes @xmath314 next , varying the action along a restricted variation @xmath55 that vanishes on @xmath56 , @xmath315 \cdot \alpha & = \int _ x \left ( \mathrm{d } _ t \alpha \wedge { d } - \mathrm{d } _ s \alpha \wedge h \wedge \mathrm{d } t + \alpha \wedge j \wedge \mathrm{d } t \right ) \label{eqn : weylaction}\\ & = \int _ x \alpha \wedge \left ( \mathrm{d } _ t d - \mathrm{d } _ s h \wedge \mathrm{d } t + j \wedge \mathrm{d } t \right ) .\nonumber\end{aligned}\ ] ] setting this equal to zero by hamilton s principle , one immediately gets ampre s law as the sole euler lagrange equation . the divergence constraint @xmath316 , corresponding to gauss law , has been eliminated via the restriction to the weyl gauge . let us restrict @xmath50 to be an euler lagrange solution in the weyl gauge , but remove the previous requirement that variations @xmath55 be fixed at the initial time @xmath143 and final time @xmath317 . then , varying the action along this new @xmath55 , the euler lagrange term disappears , but we now pick up an additional boundary term due to integration by parts @xmath318 \cdot \alpha = \left . \int _ { \sigma } \alpha \wedge { d } \,\right\rvert _ { t _ 0 } ^ { t _ f } , \ ] ] where @xmath319 denotes a cauchy surface of @xmath52 , corresponding to the spatial domain . if we vary along a gauge transformation @xmath320 , then this becomes @xmath318 \cdot \mathrm{d } _ s f = \left . \sigma \mathrm{d } _ s f \wedge { d } \,\right\rvert _ { t _ 0 } ^ { t _ f } = - \left . \int _ \sigma f \wedge \mathrm{d } _ s { d } \,\right\rvert _ { t _ 0 } ^ { t _ f } \ ] ] alternatively , plugging @xmath321 into , we get @xmath318 \cdot \mathrm{d } _ s f = \int _ x \mathrm{d } _ s f \wedge j \wedge \mathrm{d } t = - \int _ x f \wedge \mathrm{d } _ s j \wedge \mathrm{d } t = - \int _ x f \wedge \mathrm{d } _ t \rho = - \left . \int _ \sigma f \wedge \rho \,\right\rvert _ { t _ 0 } ^ { t _ f } .\ ] ] since these two expressions are equal , and @xmath301 is an arbitrary function , it follows that @xmath322 this indicates that @xmath323 is a conserved quantity , a momentum map , so if gauss law holds at the initial time , then it holds for all subsequent times as well . it should be noted that the variational structure and symmetry of maxwell s equations may be affected by the boundary conditions that one chooses to impose . there are many boundary conditions that one can specify independent of the initial values , such as the pec condition used in the numerical example in . however , one can imagine more complicated boundary conditions where which the boundary interacts nontrivially with the interior of the domain such as dissipative or forced boundary conditions , where energy / momentum is removed from or added to the system . in these cases , one will obviously _ not _ conclude that the charge density @xmath324 is conserved , but more generally that the _ change _ in charge is related to the flux through the spatial boundary . this is because , in the momentum map derivation above , the values of @xmath301 on the initial time slice causally affects its values on the spatial boundary at intermediate times , not just on the final time slice . thus , the spatial part of @xmath56 can not be neglected for arbitrary boundary conditions . the continued success of the yee scheme for many applications of computational electromagnetism , for over four decades , illustrates the value of structure - preserving numerical integrators for maxwell s equations . recent advances by , among others , bossavit and kettunen , and gross and kotiuga , have demonstrated the important role of compatible spatial discretization using differential forms , allowing for yee - like schemes that apply on generalized spatial meshes . in this paper , we have extended this approach by considering discrete forms on _ spacetime _ , encapsulating both space and time discretization , and have derived a general family of geometric numerical integrators for maxwell s equations . furthermore , since we have derived these integrators from a discrete variational principle , the resulting methods are provably multisymplectic and momentum - map - preserving , and they experimentally show correct global energy behavior . besides proving the variational nature of well - known techniques such as the yee and bossavit kettunen schemes , we have also introduced a new asynchronous integrator , so that time step sizes can be taken non - uniformly over the spatial domain for increased efficiency , while still maintaining the desirable variational and energy behavior of the other methods . one promising avenue for future work involves increasing the order of accuracy of these methods by deriving higher - order discrete hodge star operators . while this would involve redefining the hodge star matrix to be non - diagonal , the discrete maxwell s equations would remain formally the same , and hence there would be no change in the variational or multisymplectic properties proven here . we are currently exploring the development of a spectrally accurate spatial hodge star , which might make these geometric schemes competitive for applications where non - variational spectral codes are currently favored . additionally , the recent work of @xcite has shown that avis can be implemented as parallel algorithms for solid mechanics simulations . this uses the fact that , due to the asynchronous update procedure , an element does not need information from every one of its neighbors at every time step , which lessens the need for communication among parallel nodes . the resulting parallel avis , or pavis , can therefore take advantage of parallel computing architecture for improved efficiency . it is reasonable to expect that the same might be done in the case of our electromagnetic avi . while we have experimentally observed the fact that variational integrators exhibit near - energy conservation , little is known about this behavior from a theoretical standpoint . in the case of odes in mechanics , backwards error analysis has shown that these methods exactly integrate a nearby smooth hamiltonian system , although not much known about how this relates to the discrete variational principle on the lagrangian side . some initial work has been done in @xcite to understand , also by a backward error analysis approach , why discrete multisymplectic methods also display good energy behavior . finally , variational methods using discrete spacetime forms may be developed for field theories other than electromagnetism . promising candidates include numerical general relativity and fluid dynamics , although the latter is complicated by the difficulty in finding a proper discretization of the infinite - dimensional diffeomorphism group . if discrete lagrangian densities are developed for these theories , it should be straightforward to combine them with the electromagnetic lagrangian , resulting in numerical methods to simulate , e.g. , gravity coupled with an electromagnetic field , or the dynamics of a charged or magnetic fluid . we would like to thank several people for their inspiration and suggestions . first of all , alain bossavit for suggesting many years ago that we take the present dec approach to computational electromagnetism , and for his excellent lectures at caltech on the subject . second , michael ortiz and eva kanso for their ongoing interactions on related topics and suggestions . we also thank doug arnold , uri ascher , robert kotiuga , melvin leok , adrian lew , and matt west for their feedback and encouragement . in addition , the @xmath15-d avi simulations shown in were programmed and implemented by patrick xia , as part of a summer undergraduate research fellowship at caltech . desbrun , m. , a. n. hirani , and j. e. marsden ( 2003 ) , discrete exterior calculus for variational problems in computer vision and graphics . in _ proceedings of the 42nd ieee conference on decision and control ( cdc ) _ , volume 5 , pages 49024907 . ieee press , washington , dc . desbrun , m. , e. kanso , and y. tong ( 2008 ) , discrete differential forms for computational modeling . in _ discrete differential geometry _ , volume 38 of _ oberwolfach semin . _ , pages 287324 . birkhuser , basel . elcott , s. , and p. schrder ( 2005 ) , building your own dec at home . in _ siggraph 05 : acm siggraph 2005 courses_. acm press , new york . http://dx.doi.org/10.1145/1198555.1198667[`doi:10.1145/1198555.1198667 ` ] . erickson , j. , d. guoy , j. sullivan , and a. ngr ( 2005 ) , building spacetime meshes over arbitrary spatial domains . _ engineering with computers _ , * 20 * ( 4 ) , 342353 . http://dx.doi.org/10.1007/s00366-005-0303-0[`doi:10.1007/s00366-005-0303-0 ` ] . gross , p. w. , and p. r. kotiuga ( 2004 ) , _ electromagnetic theory and computation : a topological approach _ , volume 48 of _ mathematical sciences research institute publications_. cambridge university press , cambridge . hairer , e. , c. lubich , and g. wanner ( 2006 ) , _ geometric numerical integration _ , volume 31 of _ springer series in computational mathematics_. springer - verlag , berlin , second edition . structure - preserving algorithms for ordinary differential equations . wang , k. , weiwei , y. tong , m. desbrun , and p. schrder ( 2006 ) , edge subdivision schemes and the construction of smooth vector fields . in _ siggraph 06 : acm siggraph 2006 papers _ , pages 10411048 . acm press , new york . http://dx.doi.org/10.1145/1179352.1141991[`doi:10.1145/1179352.1141991 ` ] .

it is well known that the cooling of a neutron star ( ns ) is strongly affected by the relationship between the internal temperature of the star , @xmath5 , and its effective surface temperature @xmath0 . throughout this paper , @xmath5 denotes the temperature at the outer boundary of the isothermal internal region . the @xmath5@xmath0 relationship is determined by equation of state ( eos ) and thermal conductivity of matter in the outer ns envelope . for non - magnetized nss with envelopes composed of iron , this relationship has been thoroughly studied in a classical article of gudmundsson et al . ( 1983 , hereafter gpe ) . several papers ( hernquist 1984 , 1985 , van riper 1988 , 1991 , schaaf 1988 , 1990 ) have considered strongly magnetized ns envelopes . in this article , we will reconsider thermal transport in non - magnetized envelopes of nss and extend the results of the preceding authors in two respects . first , we analyze matter composed not only of iron , but of light elements as well . the light elements can be provided by an accretion from a supernova remnant ( e.g. , chevalier 1989 , 1996 , brown & weingartner 1994 ) , from interstellar medium ( e.g. , miralda - escud et al . 1990 , blaes et al . 1992 , nelson et al . 1993 , morley 1996 ) , from a distant binary component , or by comets . freshly accreted matter burns then into heavier elements ( he , c , o , fe ) while sinking within the ns . recent multiwavelength observations of the geminga pulsar ( 1e 0630 + 17.8 ) suggest a possible h or he cyclotron feature in its spectrum ( bignami et al . if confirmed , it may be a direct observational evidence of the presence of the light elements in the pulsar atmosphere . chemical composition affects the eos and thermal conduction , and , therefore , the thermal structure and cooling of a ns . as a reference case , we reconsider the outer ns envelopes composed of iron . second , we implement new , advanced theoretical data on eos and thermal conductivity of dense matter . specifically , we employ the opacity library ( opal ) radiative opacities for h , he , and c , improved considerably with respect to the los - alamos opacities used in the previous studies . in the outermost ns layers , we also implement the modern opal eos for fe and the saumon chabrier eos for h and he . we use improved thermal conductivities of degenerate electrons . for solidified matter , we employ the thermal conductivity due to the electron - phonon scattering obtained with the inclusion of the debye waller factor . for liquid matter , we recalculate and implement the thermal conductivity due to coulomb electron - ion and electron - electron collisions . the electron - ion scattering is described with the exact ( non - born ) coulomb cross sections and with the ion structure factors calculated when taking into account the response of the electron background . the new physics input allows us to extend the results of previous studies to colder nss , with @xmath0 down to 50000 k. the physics input is described in sec . 2 . in sec . 3 , we calculate the @xmath5@xmath0 relationships for non - accreted and partly accreted nss and analyze the sensitivity of the results to the uncertainty in our knowledge of the electron thermal conductivity . in addition , we simulate the cooling of nss with standard and enhanced neutrino energy losses . we show that the cooling of a ns with the accreted envelope can be quite different from the cooling of a non - accreted ns . this can change the conclusions on the internal structure of nss deduced from comparison of theoretical cooling curves with observations of ns thermal radiation . useful analytical formulae for the physics input are given in the appendix . consider thermal transport throughout the outer envelope of a ns that extends from the surface to the layers with the density @xmath6 g @xmath7 . this envelope , of which study is the aim of the present paper , provides the main thermal insulation of the ns interior . the envelope is thin ( @xmath8 m deep ) and contains very small fraction ( @xmath9 ) of the ns mass ( gpe ) . in these layers , one can neglect the nonuniformity of the energy flux due to the neutrino emission and the variation of the gravitational acceleration . then the temperature profile obeys the _ thermal structure equation _ ( gpe ) , which can be written as ( e.g. , van riper 1988 ) @xmath10 here , @xmath11 is the pressure , @xmath0 is the effective temperature , @xmath12 is the surface gravity , @xmath13 is the stellar mass , @xmath14 the stellar radius , @xmath15 the gravitational constant , and @xmath16 is the gravitational radius of the star . furthermore , @xmath17 is the opacity , @xmath18 composed of the radiative opacity @xmath19 and the equivalent electron conduction opacity @xmath20 . the latter is related to the electron thermal conductivity @xmath21 as @xmath22 where @xmath23 is the stefan boltzmann constant . the effective temperature @xmath0 is defined through the stefan s law , @xmath24 where @xmath25 is the local radiative luminosity . the apparent luminosity measured by a distant observer is @xmath26 , and the apparent surface temperature inferred by the observer from the spectrum is @xmath27 ( e.g. , thorne 1977 ) . we adopt the standard ( gpe ) outer boundary condition to eq . ( [ th - str ] ) by equating @xmath0 to the temperature @xmath28 at the stellar surface in the eddington approximation . the radiative surface , determined in this way , lies at the optical depth @xmath29 , that is approximately at @xmath30 here and hereafter the subscript `` s '' denotes quantities taken at the surface , while the subscript `` b '' is assigned to those at the inner boundary @xmath31 . previous calculations of the radiative transfer in the outer part of the atmosphere ( zavlin et al . 1996 ) have shown that , for the parameters of interest , there is no convective instability at @xmath32 which could invalidate the eddington approximation . starting from the surface specified by eq . ( [ ps ] ) , we integrate eq.([th - str ] ) inward , using the classical 4-step 4th - order runge kutta algorithm ( e.g. , fletcher 1988 ) . the integration step varies with depth and temperature to ensure the variations of @xmath33 and @xmath17 to be smaller than 5% at one step . following previous studies ( e.g. , gpe , hernquist 1985 , van riper 1988 ) , we terminate the integration at @xmath34 g @xmath7 . a comparison with analytically solvable models ( e.g. , hernquist & applegate 1984 ) shows that our numerical procedure determines @xmath5 at a given @xmath28 with an error @xmath35 1% . although the density @xmath36 does not enter eq.([th - str ] ) explicitly , its value is required to determine the opacity @xmath17 at given @xmath37 . thus , we need an eos of matter in the outer ns envelope . in this paper , we are specifically interested in h , he , c , o , and fe plasmas , which are likely to compose partly accreted ns envelopes . the parameters of interest cover the region which extends from the partially ionized non - degenerate non - relativistic plasma near the surface , through the domain of pressure ionization , to high - density layers of fully ionized relativistic plasma . we assume a strongly stratified envelope , i.e. , that at each given density , the plasma is composed of electrons and of ions of one chemical element which can be in different ionization stages . the plasma density can be characterized by the parameter @xmath38 where @xmath39 is the mean inter - electron distance , @xmath40 is the number density of electrons , and @xmath41 is the bohr radius . the parameter @xmath42 is directly related to the relativistic parameter of degenerate electrons @xmath43 ( e.g. , yakovlev & shalybkov 1989 ) : @xmath44 . here , @xmath45 is the electron fermi momentum , @xmath46 is the electron mass , @xmath47 g @xmath7 , and @xmath48 and @xmath49 are the mean charge and mass number of ions , respectively . in our case , the averaging is over different ionization stages of ions of the same chemical element . the electron degeneracy temperature is given by @xmath50 m_{\rm e}c^2 / k_{\rm b},\ ] ] where @xmath51 is the boltzmann constant . let us also introduce the ion coupling parameter , @xmath52 , which measures typical energy of ion coulomb interaction relative to the thermal energy ( @xmath53 being the electron charge ) . let us specify the high density regime as @xmath54 . this corresponds to @xmath55 g @xmath7 , @xmath56 and @xmath57 being the charge and mass numbers of atomic nuclei . in this regime , @xmath58 is considerably smaller than the @xmath17-shell radius of an atom , and we have a fully ionized one - component plasma ( ocp ) of ions immersed in the `` rigid '' electron background . then the pressure is @xmath59 where @xmath60 is the electron pressure , @xmath61 is the pressure of ideal gas of ions , and @xmath62 is the coulomb term . the pressure of the ideal gas of electrons of any degeneracy is given by ( e.g. , landau & lifshitz 1986 ) , @xmath63 p^2{\rm d}p , \label{p_e}\ ] ] where @xmath64 is the energy of an electron with a momentum @xmath65 . the electron chemical potential @xmath66 has to be determined from the equation @xmath67}. \label{ne}\ ] ] the pressure contribution from the ideal gas of ions in eq.([phigh ] ) is @xmath68 where @xmath69 is the ion number density . the coulomb correction accounts for electrostatic interactions of plasma particles . according to hansen & vieillefosse ( 1975 ) , the results of monte carlo simulations of the ocp of ions immersed in a rigid electron background can be fitted by @xmath70 , \label{p_c}\ ] ] where @xmath71 , @xmath72 , @xmath73 , and @xmath74 . the fit error is smaller than 1% at @xmath75 , as confirmed by comparison with the recent monte carlo results of stringfellow et al . ( 1990 ) and dewitt et al . equation ( [ p_c ] ) reproduces also the correct debye hckel limit at @xmath76 . equations ( [ phigh])([p_c ] ) can not be used at @xmath77 and @xmath78 because of the failure of the rigid electron background approximation . they lead even to negative pressure at low temperatures and intermediate densities . van riper ( 1988 ) modified the coulomb correction ( [ p_c ] ) in an _ ad hoc _ manner to avoid negative pressures . we use another approach described below in sect . [ sect - interdens ] . in the _ low density regime _ , @xmath79 and @xmath76 , the plasma is nearly ideal and may be partially ionized , depending on @xmath33 and @xmath36 . theoretical description of partially ionized plasmas can be based either on the physical picture or on the chemical picture of the plasma ( e.g. , ebeling et al . 1977 ) . in the chemical picture , the bound species ( atoms , molecules , ions ) are treated as elementary members of the thermodynamic ensemble , along with free electrons and nuclei . in the physical picture , nuclei and electrons ( free and bound ) are the only constituents of the ensemble . both pictures can be thermodynamically self - consistent , though the chemical picture has limited microscopic consistency ( see , e.g. , chabrier & schatzman 1994 , for review ) . the most advanced results based on the physical picture are those obtained in the opal project ( rogers et al . however , as an intrinsic limitation of the formalism , they are restricted to low densities and/or high temperatures . we employ the opal eos for partially ionized iron , which composes non - accreted envelopes . for the outermost layers of accreted envelopes , composed of h and he , we use the eos derived by saumon et al . ( 1995 ) ( scvh ) within the framework of the chemical picture . the second - order thermodynamic quantities provided by scvh suffer from some thermodynamic inconsistency in some regions of the phase diagram ( see scvh and potekhin 1996 , for a discussion ) , but the inconsistency is quite small at low densities . a comparison of the scvh and opal eoss for h and he reveals no discrepancies which could noticeably affect the temperature profiles in a ns envelope under the conditions of interest . furthermore , the scvh tables cover a larger @xmath80 domain and provide also , along with the first- and second - order thermodynamic quantities , the fractions of h and he atoms , h@xmath81 molecules , and h@xmath82 , he@xmath83 , he@xmath84 ions , which enables us to determine an average ion charge , required for our calculations of the conductivities ( sect . [ sect - conduct ] ) . the intermediate density domain ( @xmath85 ) is most complicated because of the onset of recombination . previous studies of the thermal structure of the ns envelopes ( gpe , van riper 1988 ) employed the model described by eqs.([phigh])([p_c ] ) , with the free electron density @xmath40 taken from the los alamos astrophysical opacity library ( huebner et al . 1977 , hereafter lao ) . however , this approach fails at intermediate densities and relatively low temperatures ( see , e.g. , van riper 1988 ) , where it leads to unrealistic eos owing to too large values of the coulomb correction . in the afore - mentioned density region , the less dense and more strongly correlated electron gas is polarized by the external ionic field and eventually the electrons become bound by pressure recombination . equation ( [ p_c ] ) becomes inadequate in this case , since it applies only to a _ fully ionized _ plasma , immersed in a _ rigid _ electron background . fortunately , at temperatures of present interest ( @xmath86 } \ga 4.7 $ ] ) , the scvh tables extend up to @xmath87 ( @xmath88 g @xmath7 for he ) , i.e. , to the completely ionized region . thus , they fully cover the intermediate density range for h and he . we use these tables up to the highest tabulated densities , and we use the high - density eos given by eqs.([phigh])([p_c ] ) for still higher @xmath36 . the high - density domain for c and o starts from @xmath89 g @xmath7 . we do not need to consider c and o at lower densities in the present work . the case of iron is more difficult . the opal tables are bound from low-@xmath33 high-@xmath36 side approximately by the line @xmath90 g @xmath7 . in order to extend the eos beyond this boundary , we adopt the approach of fontaine et al . ( 1977 ) , who circumvented a similar difficulty by interpolating pressure logarithm along isotherms over the intermediate density region . however , the conventional cubic spline interpolation of @xmath91 vs @xmath92 violates the condition @xmath93 although this `` normality condition '' is not required by the thermodynamic consistency , it is supposed to hold at intermediate densities ( see , e.g. , fontaine et al . 1977 ) . in order to keep this condition , we introduce the following modification of the interpolation procedure . ( i ) pressure isotherms from @xmath94 to 8 are taken from the opal data at @xmath95 and calculated from eqs.([phigh])([p_c ] ) at @xmath96 . ( ii ) the differences @xmath97 between @xmath91 for neighbouring isotherms are interpolated across the gaps in @xmath92 between @xmath98 and @xmath99 by cubic splines . ( iii ) starting with the isotherm @xmath100 , for which eqs . ( [ phigh])([p_c ] ) are sufficiently accurate , the pressure is consecutively calculated between @xmath98 and @xmath99 along the isotherms @xmath101 , using the differences @xmath97 obtained at the preceding step : @xmath102 , where @xmath103 is the pressure at the preceding ( hotter ) isotherm . in order to verify this procedure , we have applied it to he and found that it gives much better agreement with the scvh eos than the straightforward @xmath91@xmath92 interpolation across the intermediate densities . we will treat the electron heat conduction in the mean ion approximation , that is , we consider the plasma as composed of electrons and one ionic species with an effective charge @xmath104 . in the case of h and he , we adopt the mean ion charge @xmath105 provided by the scvh tables , as discussed above . for carbon and oxygen , only the high - density , fully ionized regime is involved , for which @xmath106 and 8 , respectively . finally , for iron we adjust the effective number density of free electrons @xmath40 for the pressure @xmath107 , calculated formally from eqs . ( [ p_e])([p_i ] ) , to reproduce the pressure @xmath11 given by the accurate eos at the same @xmath36 and @xmath33 . the mean ion charge , that corresponds to @xmath40 , is then assumed to be equal to @xmath104 . figure [ fig - zeff ] shows the isotherms of @xmath104 corresponding to the interpolated fe eos . the decreasing parts of the curves reflect the electron recombination due to the reduction of phase space per particle with increasing density . in this low - density region , our @xmath104 coincides with @xmath48 given by the opal data ( when available ) within @xmath108% . the increase of @xmath104 at high densities corresponds to pressure ionization . the non - monotonic behaviour at intermediate densities reflects consecutive pressure destruction of different electron shells of fe atoms . in recent years , a significant progress has been made in calculations of the radiative opacities in dense atmospheres . we use the most recent opal opacity library ( rogers et al . 1996 ) . for iron and hydrogen , it displays the tables of the spectral opacities , which we convert into the rosseland opacities @xmath19 using the standard frequency - averaging procedure ( e.g. , schwarzschild 1958 ) . for helium , we implement the rosseland opacities readily available in the opal library . for @xmath109 values within the table range ( @xmath95 ) , we have used bilinear interpolation of @xmath110 between the table entries in @xmath92 and @xmath111 . for @xmath112 , we employ similar bilinear extrapolation based on the nearest pairs of table entries in @xmath92 and @xmath111 . the extrapolated values of @xmath19 have practically no effect on temperature profiles , because at high densities the heat transport is provided by electrons ( @xmath113 ) . previous studies of the temperature profiles in ns envelopes have mainly used the electron thermal conductivity of completely ionized , strongly degenerate plasma . in particular , one has often taken the conductive opacities from urpin & yakovlev ( 1980 ) and yakovlev & urpin ( 1980 ) ( hereafter yu ) . however , in a cold enough ns , the thermal conductivity of non - degenerate or partly degenerate electrons may become important . a comparison with the tabular data of hubbard & lampe ( 1969 ) reveals that a straightforward extrapolation of the yu formulae from their validity domain ( fully ionized , degenerate plasma ) to the case of non - degenerate matter may lead to an underestimation of the electron thermal conductivity by orders of magnitude . as will be shown in sect . [ sect - res ] , this may strongly affect the temperature profiles in the ns envelopes at @xmath114 k. in order to determine the conductive opacities for any degeneracy , we employ the numerical code developed recently by potekhin & yakovlev ( 1996 ) ( hereafter py ) for calculating the electron transport coefficients along magnetic fields in the magnetized ns envelopes . the code performs numerical thermal averaging of the effective energy - dependent electron relaxation time with the fermi dirac distribution at any degeneracy . this enables us to consider not only strongly degenerate , but also mildly degenerate matter . in addition , py have improved the approach of yu by a more accurate calculation of the coulomb logarithm for the electron - ion scattering in the liquid phase of ns matter and by incorporating the debye waller reduction factor ( e.g. , itoh et al . 1984 ) for the electron - phonon scattering in the solid phase . in the present article , we apply the above code for the particular case of zero magnetic field . for this purpose , further modifications have been made . first , in addition to the thermal conduction produced by the electron - ion scattering considered by py , we have incorporated also the contribution from the electron - electron scattering in the liquid regime ( see sect . [ app - ee ] ) . this process can be important for light elements ( h , he ) in a weakly degenerate matter . second , we have improved the coulomb logarithm which determines the collision rate of strongly degenerate electrons with ions ( see sect . [ sect - coulog ] ) . third , in the solid regime , we have used the new expression for the electron thermal conductivity , derived recently by baiko & yakovlev ( 1995 ) . note that at temperatures much lower than the melting temperatures , the coulomb scattering of electrons by charged impurities may become important ( e.g. , yu ) . it depends on the impurity charge and , most important , on its number density @xmath115 which is generally unknown in the ns envelopes . we will present the results obtained for pure crystals ( @xmath116 ) . our calculations including @xmath117ni impurities in iron envelopes show that the electron - impurity scattering may affect the temperature profiles in the crusts of cooling nss , whenever the impurity concentration is sufficiently high , @xmath118 . finally , in the non - degenerate layers , we have performed the thermal averaging introducing modifications into the effective width of the landau levels which was used by py to account for collisional and inelastic broadenings of the landau levels in strongly degenerate matter . the effective width proposed by py becomes inadequate in the non - degenerate regime , and we have switched it off by multiplying by the factor @xmath119 . a comparison of our thermal conductivities with the tables of hubbard & lampe ( 1969 ) ( at @xmath120 g @xmath7 , where the tables are adequate ) shows maximum discrepancy up to 30% for carbon and up to 70% for lighter elements in the non - degenerate regime , and still smaller discrepancies in mildly and strongly degenerate matter . more accurate values of the coulomb logarithm obtained in sect . [ sect - coulog ] modify @xmath20 by about 530% , which has only a little effect on the thermal structure of the envelope ( see sect . [ sect - res ] ) . the thermal conductivity @xmath21 of degenerate electrons ( @xmath121 ) in a liquid or a gas of ions with mass and charge numbers @xmath57 and @xmath56 can be written as @xmath122 where @xmath123 , @xmath124 is the effective electron collision frequency , @xmath125 is determined by the coulomb electron - ion collisions , and @xmath126 is determined by the electron - electron collisions . new calculations and fits of @xmath126 are presented in sect . [ app - ee ] . in this section , we examine @xmath125 . it can be expressed as ( e.g. , yu ) @xmath127 where @xmath128 is the coulomb logarithm to be evaluated . it is a slowly varying function of @xmath36 and @xmath33 . first consider a strongly coupled coulomb liquid of ions , at @xmath129 , where @xmath130 is the critical value of the ion coupling parameter @xmath131 at which a coulomb crystal melts ( @xmath130 = 172 , for a classical ion system , nagara et al . 1987 ; the deviations from the classical melting curve due to the quantum corrections turn out to be small at @xmath132 g @xmath7 for carbon and heavier ions see chabrier 1993 ) . in this case , one usually assumes that the screening of the electron - ion coulomb interaction is static and produced by the ion - ion correlations . the screening is described by the ion - ion structure factor @xmath133 . extensive calculations of @xmath128 in the born approximation for different ion species , @xmath56 and @xmath57 , were performed by itoh et al . the authors used the structure factors obtained in the approximation of rigid electron background . they fitted their results by complicated expressions . later yakovlev ( 1987 ) evaluated non - born corrections to @xmath128 , using the weak screening approximation . now we recalculate the coulomb logarithm for strongly degenerate electrons at @xmath129 from the expression @xmath134 where @xmath135 , @xmath136 , @xmath137 is a momentum transfer in an @xmath138 collision event , @xmath139 is the static longitudinal dielectric function of electrons , @xmath140 is the nuclear form factor to allow for finite nuclear size , and @xmath141 is the non - born correction factor specified below . the dielectric function has been calculated for the electron gas at any degeneracy . however , we have verified that , for all the cases in the present study , this dielectric function yields the same results as the function obtained in the limit of strong electron degeneracy for the relativistic electron gas ( jancovici 1962 ) . thus the bulk of the calculations has been done with this latter function . we use the nuclear form factor @xmath140 appropriate to a uniform proton core of an atomic nucleus , with the same core radii as in itoh et al . the finite sizes of the nuclei appear to have almost no effect on @xmath128 , for the conditions of study . nevertheless we have kept @xmath140 in our calculations . following yakovlev ( 1987 ) , the non - born factor has been taken as @xmath142 , where @xmath143 is the exact differential coulomb scattering cross section for a momentum transfer @xmath144 , and @xmath145 is the cross section in the born approximation . the exact cross sections are taken from doggett & spencer ( 1956 ) . in our calculations from eq . ( [ l ] ) , we have used the ion structure factors evaluated for a _ responsive _ electron background including the local field corrections between electrons ( chabrier 1990 ) . our calculations extend those of itoh et al . ( 1983 ) by including the non - born corrections and the effects of electron response on the ion structure factors . on the other hand , we improve the results of yakovlev ( 1987 ) since we evaluate the non - born corrections beyond the weak screening approximation . we have evaluated @xmath128 from eq . ( [ l ] ) for chemical elements with @xmath146 and for a dense grid of @xmath147 and @xmath148 g @xmath149 g @xmath7 provided @xmath150 . in the case of hydrogen , we have restricted ourselves to @xmath151 g @xmath7 since hydrogen should be completely burnt at higher densities . for light elements , the highest temperatures included in these data correspond to @xmath152 and appear to be much below @xmath153 . accordingly , there exists a large temperature interval in the electron degeneracy domain where @xmath154 , the ion coupling ceases to be strong , the ion screening can not be treated in terms of ion - ion correlations , and eq . ( [ l ] ) is invalid . in principle , the coulomb logarithm for mildly - coupled ion plasma ( @xmath155 ) can be evaluated using the formalism developed by boerker et al . ( 1982 ) . in order to simplify our analysis , we will restrict ourselves to the weak coupling case ( @xmath76 ) , in which the ions constitute a boltzmann gas , the weak ion screening is dynamical and can be described by the dielectric function formalism , while the electron screening remains static ( e.g. , lampe 1968 ) . at @xmath76 and @xmath121 , we propose the expression @xmath156 where @xmath157 is the ion screening parameter in the weak coupling regime , @xmath158 is the debye radius of ions , @xmath159 , @xmath160 is the electron screening parameter , @xmath161 being the inverse screening length of a charge by degenerate plasma electrons and @xmath162 the fine - structure constant . the first three terms represent the coulomb logarithm obtained for degenerate electrons and weakly coupled ions with the non - relativistic @xmath138 scattering cross section . the expression presented is valid for @xmath163 ; it can be derived using the formalism of williams & dewitt ( 1969 ) to account for the dynamic ion screening and to integrate the @xmath138 collision rate over velocities of ions . the fourth term is the relativistic correction in the weak coupling limit ( e.g. , yu ) . the fifth term is the second born correction in the weak coupling limit ( yakovlev 1987 ) . thus , we have supplemented our table of the coulomb logarithms calculated for @xmath164 ( see above ) with new values evaluated from eq . ( [ dewitt ] ) for @xmath165 and @xmath121 . analytic fits to the full set of about 31,000 values of @xmath128 are presented in sect . [ app - l ] . the tables of @xmath128 are freely available in the electronic form . for illustration , in fig . [ fig - l ] ( left panel ) we compare the coulomb logarithms in fully ionized fe and c plasmas with those calculated in the born approximation ( itoh et al . 1983 ) . the non - born corrections enhance the small angle coulomb scattering cross section , and increase @xmath128 . the increase is especially pronounced for the heavier element , fe ( exceeds 20% ) , at densities @xmath166 g @xmath7 , where the electrons are relativistic . the non - born corrections become less important with decreasing @xmath56 , and they are negligible for light elements ( h , he ) . the inclusion of the electron response in the calculations of the ionic structure factors affects the coulomb logarithm @xmath128 for h and he at @xmath167 , as demonstrated on the right panel of fig . [ fig - l ] . one can observe an increase of @xmath128 up to 15% for h at @xmath168 g @xmath169 and @xmath170 with respect to the case of rigid electron background . actually , the increase grows with @xmath131 and reaches about 40% at @xmath171 . the effect is weaker for he , and small for heavier elements ( in the adopted parameter range ) . thus , for the elements heavier than he , we have used the structure factors calculated in the rigid background approximation . note that calculated values of @xmath128 determine also the electric conductivity of degenerate matter in a liquid or gaseous ionic plasma , @xmath172 . in this section , we consider a non - accreted ns envelope composed solely of iron , which may be at various ionization stages . we use the opal iron eos where available , the ideal - gas eos with the coulomb correction ( [ p_c ] ) at high densities , and the interpolated eos at intermediate densities as described in sect . [ sect - interdens ] . the conduction properties are described in the mean ion approximation with the effective ion charge @xmath104 determined consistently with the eos ( sect . [ sect - zeff ] ) . figure [ fig - prfe ] presents density dependence of the temperature in the ns envelope at various @xmath0 . the integration of eq . ( [ th - str ] ) has been started with the surface density @xmath173 , shown in fig . [ fig - rho - s ] , and terminated at @xmath34 g @xmath7 . the value @xmath174 chosen in fig . [ fig - prfe ] corresponds to a ns with a radius @xmath175 km and a mass @xmath176 . circles on the curves are the points of equal radiative and conductive opacities . the thermal flux is mainly carried by radiation at lower densities and by electrons at higher densities . we also show the electron degeneracy curve and the melting curve . in a wide range of @xmath0 , some layers ( fig . [ fig - prfe ] ) turn out to be convectively unstable ( zavlin et al . 1996 , rajagopal & romani 1996 ) . in these layers , the energy transport is dominated by convection rather than by electron or radiative conduction . we describe the convective energy flux in the adiabatic approximation ( schwarzschild 1958 ) , which assumes that the convective heat transport is much more efficient than other mechanisms . accordingly , in the convective zones , we replace the temperature gradient given by eq . ( [ th - str ] ) by the adiabatic gradient provided by the eos tables . in order to check the effect of convection , we have performed also calculations with `` frozen '' convection ( i.e. , neglecting the convection effect ) . this is the extreme case opposite to the adiabatic one . in this approximation , we obtain ( fig . [ fig - prfe ] ) slightly higher temperatures inside the convective part of the atmosphere . the atmospheric temperature profiles derived by zavlin et al . ( 1996 ) , who solved numerically the radiative transfer equation at moderate optical depths and described the convection using the mixing - length theory , lie between our two extremes . in deeper layers , the two extreme profiles tend to merge , as clearly seen in fig . [ fig - prfe ] . this merging occurs because the factor @xmath177 on the r.h.s . of eq . ( [ th - str ] ) rapidly decreases , with increasing @xmath33 , and thus ensures the isothermal profile at higher @xmath33 behind the convective zone ( at fixed @xmath0 ) . thus , the thermal structure of the ns envelope at @xmath178 g @xmath7 is practically unaffected by the convection . the dot - dashed curves in fig . [ fig - prfe ] are obtained with @xmath20 calculated according to py ( sect . [ sect - conduct ] ) . these curves almost coincide with the solid lines , calculated more accurately . this confirms the validity of the py approximations for the non - accreted envelopes . if @xmath179 k , the electron conduction becomes important not only in degenerate matter , but also in a non - degenerate region ( fig . [ fig - prfe ] ) . then the straight - line logarithmic extrapolation of the yu conductive opacities into the non - degenerate regions leads to a significant overestimation of the internal temperature ( dotted curves ) . this effect is seen also in fig . [ fig - link - fe ] , that shows @xmath180 as a function of @xmath181 for four values of the surface gravity . if @xmath182>5.5 $ ] , the well - known simple fit of gpe , @xmath183 is rather accurate , but at lower @xmath28 the deviations from this fit are quite pronounced . nevertheless , the scaling law @xmath184 holds with a good accuracy in the whole temperature gravity range presented in this figure , except for the lowest @xmath5 values . the fit which reproduces these new features is presented in sect . [ app - link ] . the chemical composition of an accreted envelope depends on the composition of the infalling matter and on the nuclear transformations ( thermo- or pycnonuclear burning , beta - captures ) which occur while the matter sinks within the ns . the structure and stability of the accreted envelopes have been considered in number of works devoted to bursting nss ( see , e.g. , ayasli & joss 1982 , paczyski 1983 , ergma 1986 , miralda - escud et al . 1990 , blaes et al . 1992 , and references therein ) . however , most of the work focused either on high accretion rates or on slowly accreting old nss whose internal temperature is solely determined by the accretion . on the contrary , we are mainly interested in not too old , cooling nss with thin accreted envelopes , in which the heat release due to the nuclear transformations has no effect on the thermal structure . let us assume that an accreted envelope consists of shells with different chemical compositions . for an accretion of h / he , it is reasonable to expect that the outermost accreted layers are built of pure h , owing to the strong gravitational stratification ( alcock & illarionov 1980 ) , whereas he may exist in the deeper layers . the accretion is accompanied by the nuclear transformations of h / he into heavier elements . the details of these transformations are still uncertain , and we consider several models of accreted envelopes . as a first example , we take the typical temperatures and densities of the nuclear burning from ergma ( 1986 ) and assume the burning to be non - explosive . if @xmath0 is high and the accreted hydrogen layer reaches the region with @xmath185 k , then h burns efficiently into he in the thermonuclear regime . if temperature is lower and the hydrogen layer reaches the density @xmath186 g @xmath7 , a pycnonuclear burning of h into he proceeds . at higher @xmath187 g @xmath7 and/or @xmath188 k , helium burns into carbon ( schramm et al . the pycnonuclear carbon burning occurs at @xmath189 g @xmath7 ( e.g. , yakovlev 1994 ) , which is already inside the isothermal region of the envelope . figure [ fig - prc ] displays the thermal structure of a fully accreted envelope ( the accreted mass @xmath190 fills the layers up to @xmath191 ) of a ns with the surface gravity @xmath192 cm s@xmath193 . the outer , intermediate , and inner shells of this envelope , separated by asterisks , are composed of h , he , and c , respectively . one can observe significant differences from the thermal structure of a non - accreted fe envelope , as explained below . for a not too cold ns ( @xmath194 k ) , the main temperature gradient occurs in a layer of degenerate electron gas with ions in the liquid state , where the thermal conduction is mostly provided by the electrons due to the electron - ion collisions . the heavier the element , the smaller the thermal conductivity , and the steeper is the temperature growth inside the star . however , with decreasing @xmath0 , the width of the heat insulating degenerate layer becomes smaller , and the effect is less pronounced . in a cooler ns ( @xmath195 k ) , the main temperature gradient shifts into the ns atmosphere ( to the optical depths @xmath196 ) . for heavier elements , the atmospheric layers appear generally denser ( at the same @xmath0 , see fig . [ fig - rho - s ] ) . then the internal temperature gradient is weaker and the temperature grows slower inside the star . thus , a not too cold accreted envelope is more heat transparent than the iron envelope , whereas a cold accreted envelope is less heat - transparent . as seen from fig . [ fig - rho - s ] , the effective surface temperature which separates these two regimes is almost independent of the surface gravity . figure [ fig - py ] demonstrates the effect of the improvements in the conductive opacities . the present results are compared with those in which the electron thermal conductivity is calculated with the py code . although the difference is not too large , it is higher than for the non - accreted fe envelopes ( fig . [ fig - prfe ] ) . the difference comes mostly from the electron - electron collisions ( neglected in py ) . dots in fig . [ fig - py ] show the results obtained with the py code supplemented by the contribution from the electron - electron collisions to the thermal conduction ( sect . [ app - ee ] ) . the difference from the accurate results practically disappears . now let us analyze the sensitivity of the temperature profiles to the positions of the h / he and he / c interfaces . these interfaces , indicated by asterisks in fig . [ fig - prc ] , are not well theoretically established . for example , the densities of nuclear h and he burning presented by iben ( 1974 ) ( see also kippenhahn & weigert 1990 ) are much lower than those given by ergma ( 1986 ) . we have significantly shifted the interfaces according to the results of iben ( 1974 ) , but the @xmath0@xmath5 relationship remains practically unchanged . we have also calculated the temperature profiles for a purely helium ns envelope . these profiles are slightly different from those in a burning accreted matter described above . however , the @xmath0@xmath5 relationships remain practically the same . recently the thermal structure of old , slowly accreting nss has been examined in several articles ( see miralda - escud et al . 1990 , blaes et al . 1992 , and references therein ) . in these studies , the internal stellar temperature is assumed to be determined by the accretion rate . the composition of the accreted envelopes appears to be similar to the ones considered above except for the innermost layers . for instance , miralda - escud et al . ( 1990 ) assume that he transforms directly into fe while blaes et al . ( 1992 ) state that the burning of he into c at @xmath197 g @xmath7 is followed by the carbon transformation through the rapid @xmath198c@xmath199o process . accordingly , we have recalculated the temperature profiles replacing @xmath198c by @xmath117fe or @xmath200o and found only very small differences . the insensitivity of the temperature profiles to the composition of the innermost layers of the envelope is quite natural , because these layers are nearly isothermal . summarizing the above discussion we can conclude that the @xmath0@xmath5 relationship is remarkably insensitive to the parameters of nuclear burning and to the chemical composition of accreted matter ( h and/or he ) but differs from the one in a non - accreted ns . this conclusion will not change if h burning is explosive ( see , e.g. , paczyski 1983 , miralda - escud et al . 1990 ) . during an outburst , the h - shell burns into he , without any noticeable effect on the @xmath0@xmath5 relation . according to miralda - escud et al . ( 1990 ) helium burning in slowly accreting nss is unstable ; it produces nuclear outbursts with burning of all accreted material . nevertheless , the helium layer should reach a very high density , @xmath187 g @xmath7 to trigger the instability in a not too hot ns . if the same is true for the conditions of our interest , our results are valid during a long period of time until the instability is reached . however , according to paczyski ( 1983 ) , the presence of the heat outflow from a not too cold , cooling ns essentially stabilizes nuclear burning in accreted envelope , and the he burning is mostly non - explosive for the parameters we are interested in . figure [ fig - link - a ] shows @xmath180 as a function of @xmath181 for various amounts of accreted matter . the dot - dashed line represents @xmath180 for a non - accreted envelope from fig . [ fig - link - fe ] . this function is strongly affected by the accretion . even a thin hydrogen ( or he ) mantle of mass @xmath201 , which extends only to @xmath202 g @xmath7 , modifies strongly the @xmath5@xmath0 relation . analytical fits of @xmath0 as a function of the internal temperature @xmath5 , surface gravity @xmath203 , and accreted mass @xmath204 are given in sect . [ app - link ] . consider briefly the effects of the possible presence of accreted matter in the ns envelopes on the thermal history of nss . for this purpose , we simulate the cooling of nss using two ns models . we use the cooling code of gnedin & yakovlev ( 1993 ) based on the approximation of isothermal ns interior . the code follows accurately the cooling of a not too young ns whose thermal relaxation is already over ( @xmath205 yrs , see nomoto & tsuruta 1987 , umeda et al . we adopt a moderately stiff eos of prakash et al . ( 1988 ) to describe matter in the ns cores . this matter is assumed to consist of neutrons , protons , and electrons ( no exotic cooling agents such as quarks or kaon condensates ) . the maximum ns mass , for this eos , is about 1.7 @xmath206 . we examine two ns models : ( 1 ) @xmath207 , @xmath208 km , the central density @xmath209 g @xmath7 ; ( 2 ) @xmath210 , @xmath211 km , @xmath212 g @xmath7 . the first model is a typical example of the _ standard _ cooling . the central density is insufficient to switch on the most powerful neutrino cooling due to the direct urca process ( lattimer et al . 1991 ) . in this case the main neutrino energy loss mechanisms in the ns cores are the modified urca reactions ( neutron and proton branches ) and the nucleon - nucleon bremsstrahlung ( friman & maxwell 1979 , yakovlev & levenfish 1995 ) . our second ns model is an example of the _ rapid _ neutrino cooling . in this model , the central density slightly exceeds the threshold density @xmath213 g @xmath7 of the direct urca process ( for the given eos ) . then the ns has a small central kernel ( of radius 2.32 km and mass 0.035 @xmath206 ) where the direct urca process operates , and the neutrino luminosity exceeds the standard one by several orders of magnitude . all the neutrino energy loss rates included in the calculations are described by levenfish & yakovlev ( 1996 ) . to simplify our analysis , we assume that the ns cores are non - superfluid , and there is no internal reheating ( e.g. , energy dissipation due to differential rotation ) . the results are presented in fig . [ fig - cooling ] in the form of the cooling curves effective temperatures @xmath214 as measured by a distant observer ( see sect . [ sect - th - str - eq ] ) versus ns age @xmath215 . figure [ fig - cooling ] presents the cooling curves under the assumption that all the accreted material is accumulated at the ns surface during the ns birth . the accretion can occur , for instance , at the post - supernova stage ( e.g. , chevalier 1989 , 1996 , brown & weingartner 1994 ) especially if the pulsar activity is suppressed initially by burying the pulsar magnetic field under the fallen - back matter ( muslimov & page 1995 ) . we show the standard and rapid neutrino cooling for different amount of accreted mass @xmath204 . the fraction of accreted mass @xmath216 varies from 0 ( non - accreted nss ) to @xmath9 ( fully accreted ns envelope ) . it is evident that further increase of @xmath204 does not affect the cooling . the change of slopes of the cooling curves at @xmath217@xmath218 yrs reflects the change of the cooling regime . initially , a ns is at the neutrino cooling stage . it cools mainly via the neutrino emission ; the internal stellar temperature is ruled by this emission and is thus independent of the thermal insulation of the ns envelope . the surface photon emission is determined by the @xmath5@xmath0 relation . since the accreted envelopes of not too cold nss are more thermally transparent ( sect . [ sect - accr ] ) , the surface temperature of an accreted ns is noticeably higher than that of a non - accreted ns . one can see that even a very small fraction of accreted matter , such as @xmath219 , may change appreciably the thermal history of the star . the colder the star , the smaller is the fraction of accreted material which yields the same cooling curve as the fully accreted ns envelope . the effect is naturally more pronounced for the rapid cooling ( for colder nss ) . if @xmath220@xmath218 yrs , the neutrino luminosity decreases , and the ns cools mainly via the surface emission of photons ( the photon cooling stage ) . since the envelopes of the accreting ( and not too cold ) nss have lower thermal insulation , these stars cool now faster than the non - accreted ones ( fig . [ fig - cooling ] ) . the change of the thermal history of a ns by the accreted material should be taken into account in theoretical interpretation of observational data on thermal radiation from nss . for illustration , in fig . [ fig - cooling ] we present the effective temperatures with @xmath221 confidence intervals of the pulsars 083345 vela ( gelman et al . 1993 , gelman 1995 ) , 0630 + 18 geminga ( halpern & ruderman 1993 ) , 0656 + 14 and 105552 ( greiveldinger et al . 1996 ) , rxj 0002 + 6246 ( hailey & craig 1995 ) , and the @xmath222 upper limit for the crab pulsar ( psr 0531 + 21 , becker & aschenbach 1995 ) . these data have been deduced from the fits of spectral observations of the _ rosat _ x - ray observatory . for geminga , psr 105552 , and psr 0656 + 14 , we plot the values of @xmath0 obtained for `` soft '' components of the two- and three - component x - ray spectra models , since it is the soft component that is associated with the thermal flux from the stellar interior . for vela , we extend downward the confidence interval obtained by gelman et al . ( 1993 ) according to the more recent spectral fitting of gelman ( 1995 ) , that gives a more plausible ns radius than the previous fit . the age of the crab pulsar in fig . [ fig - cooling ] is its true age known from the chronicles , the age of vela is its possible true age determined recently by lyne et al . ( 1996 ) , the age of rxj 0002 + 6246 is an estimate for the related supernova remnant g117.7 + 06 , while for the three other pulsars we adopt their characteristic ages . we plot the values of @xmath0 derived by the authors by fitting the assumed blackbody thermal spectrum to the measured x - ray spectrum . the alternative often used is to derive @xmath0 from the apparent luminosity in the whole spectral range observed . however , this latter way is less certain , because the inferred temperature strongly depends on the assumed ns distance ( usually poorly known ) and model - dependent radius . note that the blackbody fits do not take into account the influence of the magnetized atmosphere . the fits for psr 0656 + 14 ( anderson et al . 1993 ) , geminga ( meyer et al . 1994 ) , and vela ( page et al . 1996 ) , which employ the models of magnetized atmospheres , have resulted in a few times lower @xmath0 than the blackbody models . our justification of using here the blackbody data is that the so far developed models of magnetized atmospheres do not take proper account of absorption by neutral atoms . the neutral fraction can be significant at @xmath223 k , because the magnetic fields @xmath224 g increase the ground - state binding energy by about one order of magnitude . the opacities of hydrogen in such strong fields are still under investigation ( see pavlov & potekhin ( 1995 ) and references therein ) . the magnetic fields affect also the thermal insulation of the ns envelopes and , therefore , the cooling ( e.g. , yakovlev & kaminker 1994 ) . we do not take this effect into account in the present article . as suggested by hernquist ( 1985 ) and shown recently by shibanov & yakovlev ( 1996 ) , the dipole surface magnetic fields @xmath225 g affect the cooling much more weakly than predicted by the studies of simplified magnetic field geometries ( e.g. , van riper 1988 , 1991 ) . as seen from fig . [ fig - cooling ] , for the pulsars vela , geminga , and rxj 0002 + 6246 , the presence of accreted matter would simplify the explanation of the cooling by the standard neutrino emission without invoking superfluidity of ns matter , as favoured in previous work ( page 1994 , levenfish & yakovlev 1996 ) . interestingly , the accretion rate of vela estimated by morley ( 1996 ) , multiplied by its age , yields @xmath226 , that just fits the middle of the error bar in fig . [ fig - cooling ] . for psr 0656 + 14 , the scenarios with and without the accretion are equally consistent with the result of the fitting shown if fig . [ fig - cooling ] , especially if we assume that its braking index is somewhat lower than 3 ( as happens for all 4 pulsars with known braking indices , see lyne et al . 1996 ) and therefore its true age is somewhat higher than its characteristic age . the relatively high temperature of psr 105552 can not be explained by the considered models of the cooling , that probably suggests some reheating process for this relatively old pulsar . we have used , for illustration , the standard ns models and simplified blackbody fits to the observed spectra . more sophisticated models and fits can also gain from the assumption of accreted envelopes . for instance , page ( 1996 ) considers the cooling of nss with kaon condensed cores and reheating in the outer crusts , using our models of the envelopes . we have considered thermal structure and evolution of nss whose envelopes are composed of non - accreted or accreted matter . we have used new , state - of - the - art calculations of eos and opacities of ns envelopes , described in sect . 2 . in particular , we have recalculated ( sects . [ sect - coulog ] , [ app - ee ] , and [ app - l ] ) the electron - electron and electron - ion collision frequencies , that determine the electron thermal conductivity , for a wide range of densities and temperatures of a degenerate electron gas and ionic liquid plasmas . using this new physics input , we have calculated ( sects . [ sect - non - accr ] , [ sect - accr ] ) the temperature profiles in the envelopes of non - accreted and partly accreted nss and obtained the relationships between the internal and effective temperatures of nss , @xmath5 and @xmath0 . these relationships are important for simulating the ns cooling ; they are fitted by simple analytic expressions in sect . [ app - link ] . they appear to be very sensitive to the presence of accreted matter in the ns envelope ; even a small amount of the accreted matter , @xmath227 , can reduce substantially the thermal insulation of the envelopes . for a non - accreting ns , our relationship between @xmath5 and @xmath0 extends the well known result of gpe to lower temperatures ( down to @xmath228 k ) . in sect . [ sect - cooling ] , we have examined briefly the effect of the possible presence of accreted matter on the ns cooling . we show that the accreted matter may increase the surface temperature ( photon thermal luminosity ) at the neutrino cooling stage , and decrease them at the subsequent photon cooling stage , as compared to the nss without accreted envelopes . we have shown that these results can be important for a proper interpretation of observed thermal radiation from nss . in particular , the presence of accreted matter facilitates the explanation of recent observational results concerning the pulsars vela and geminga , and rxj 0002 + 6246 , in the framework of the standard neutrino emission model ( without exotic matter , superfluidity , or direct urca processes ) . in this paper , we have neglected effects of magnetic fields on the eos and the thermal conduction of matter , which can be significant ( e.g. , yakovlev & kaminker 1994 ) . they do deserve further studies using improved eos and thermal conductivities of magnetized ns envelope ( e.g. , py ) and improved radiative opacities of magnetized ns atmosphere ( e.g. , pavlov & potekhin 1995 ) , with allowance for the possible presence of light elements in the surface layers . finally , it is worth noting that the physics input used in the present calculations can be applied to a variety of other astrophysical problems concerning dense stellar matter , e.g. the thermal structure and bursting activity of x - ray bursters ( see , e.g. , miralda - escud et al . 1990 and references therein ) and the cooling of white dwarfs ( segretain et al . 1994 ) . we are grateful to h. dewitt for pointing out the article of boerker et al . ( 1982 ) , to dany page and yu.a . shibanov for fruitful discussions , and to d. page for reading the manuscript and making the most useful critical remarks . this work was supported in part by the rbrf grant no . 96 - 02 - 16870a , intas grant no . 94 - 3834 , and dfg rbrf grant no . 96 - 02 - 00177 g . ayp gratefully acknowledges the hospitality of the theoretical astrophysics group of the ecole normale suprieure de lyon . the effective collision frequency @xmath126 of non - relativistic degenerate electrons ( @xmath229 , @xmath121 ) was analyzed by lampe ( 1968 ) using the formalism of the dynamic screening of the electron electron interaction . the expression of @xmath126 for the relativistic degenerate electrons at @xmath230 was obtained by flowers & itoh ( 1976 ) . here , @xmath231 is the electron plasma temperature determined by the electron plasma frequency @xmath232 , @xmath233 @xmath234 . the degeneracy temperature @xmath153 in the ns envelopes is typically higher than @xmath231 . urpin & yakovlev ( 1980 ) extended the results of flowers & itoh ( 1976 ) to higher temperatures , @xmath121 . in the approximation of static electron screening of the coulomb interaction , urpin & yakovlev ( 1980 ) obtained @xmath235 where @xmath236 , @xmath237 , @xmath238 and @xmath239 ( see eq . ( [ dewitt ] ) ) . now it is sufficient to calculate the function @xmath240 , presented by urpin and yakovlev ( 1980 ) as a 2d integral which depends parametrically on the relativistic parameter @xmath241 . lampe ( 1968 ) analyzed this function in the static screening approximation at @xmath229 . the asymptotes of @xmath242 were obtained by lampe ( 1968 ) for @xmath229 at @xmath243 and @xmath244 , by flowers & itoh ( 1976 ) for @xmath244 at any @xmath245 , and by urpin & yakovlev ( 1980 ) for @xmath243 and @xmath246 . timmes ( 1992 ) calculated @xmath240 numerically in the limit of @xmath246 . however , the unified expression of @xmath240 at @xmath247 valid equally for relativistic and non - relativistic electrons has been absent . note that the fit expression of @xmath240 obtained by timmes ( 1992 ) ( his eq . ( 10 ) ) is valid only at @xmath248 . we have calculated @xmath242 numerically for a dense grid of @xmath241 and @xmath249 in the intervals @xmath250 and @xmath251 . the results are fitted by the expression : @xmath252 \label{jfit}\end{aligned}\ ] ] which reproduces also all the asymptotic limits mentioned above . the mean error of the fits is 3.7% , and the maximum error of 11% takes place at @xmath253 and @xmath254 . the tables of the coulomb logarithms calculated from eqs . ( [ l ] ) and ( [ dewitt ] ) as described in sect . [ sect - coulog ] are fitted by @xmath255 where @xmath256 and @xmath257 for @xmath258 , the fit parameters are given by : + @xmath259 , + @xmath260 , + @xmath261 , + @xmath262 , + @xmath263 , + @xmath264 , + @xmath265 . if @xmath274 , the rms fit error over all our data set is about @xmath275 3.2% , and the maximum error of @xmath276 7.2% takes place at @xmath277 g @xmath7 and @xmath278 . if @xmath279 we have @xmath275 3.7% , @xmath276 7.8% at @xmath280 g @xmath7 and @xmath281 . for @xmath282 , we obtain @xmath283 2.4% , @xmath276 7.0% at @xmath284 ) \approx 2.5 $ ] and @xmath281 . for higher @xmath56 , the rms error remains about 1.51.7% , and the maximum error mainly decreases . for instance , we have @xmath276 5.1% at @xmath285 and @xmath286 for @xmath287 ; @xmath276 3.7% at @xmath288 and @xmath289 for @xmath290 ; and @xmath276 3.4% at @xmath291 and @xmath292 for @xmath293 . this fit accuracy is quite sufficient for studying the thermal structure of nss . let us define @xmath298 k , @xmath299 k , and @xmath300 where @xmath13 is the ns mass and @xmath204 is the accreted mass . according to gpe , @xmath301 where @xmath302 is the pressure at the bottom of the accreted envelope . for a purely iron ( non - accreted ) envelope , a very crude estimate ( with an error @xmath303 30% ) yields @xmath304 then @xmath305 is approximately an increase of the temperature through the iron envelope ( in @xmath306 k ) . a more accurate fitting formula reads @xmath307 . \label{t0}\ ] ] the typical fit error of @xmath308 is about 2% , with maximum 4.2% , over the @xmath309 domain indicated above . finally , for the partially accreted envelopes at any temperatures within the indicated range , we have @xmath312 where @xmath313\,t_{\rm b9}^{5/3}.\ ] ] the typical fit error of eq.([ttfit ] ) for @xmath314 is about 3% , with maximum 5.2% , for all possible values of @xmath315 and any values of @xmath203 and @xmath0 within the indicated range . the dependence ( [ t0 ] ) is recovered not only at sufficiently low accreted mass ( @xmath316 ) , but also at sufficiently high @xmath5 . the latter result reflects the fact ( first demonstrated by gpe ) that at high @xmath0 the thermal insulation is mostly produced by the conductive opacities in the deep and hot layers of the envelope , in which light elements ( h , he ) burn into heavier ones . on the other hand , even at very low accreted mass ( @xmath317 ) , the approximation of fully accreted crust is good enough at sufficiently low temperature because in this case the thermal insulation is provided mainly by the low - density accreted surface layers . page d. , shibanov yu.a . , zavlin v.e . 1996 , temperature , distance and cooling of the vela pulsar . in : zimmermann h .- u . , trmper j. , yorke h. ( eds ) proc . conf . on x - ray astronomy and astrophysics , rntgenstrahlung from the universe . mpe report 263 , garching , p. 173 yakovlev d.g . , kaminker a.d . , 1994 , neutron star crusts with magnetic fields . in : chabrier g. , schatzman e. ( eds ) proc . 147 , the equation of state in astrophysics . cambridge univ . press , cambridge , p. 214 zavlin v.e . , pavlov g.g . , shibanov yu.a . , rogers f.j . , iglesias c.a . , 1996 , x - ray spectra from convective photospheres of neutron stars . in : zimmermann h .- u . , trmper j. , yorke h. ( eds ) proc . conf . on x - ray astronomy and astrophysics , rntgenstrahlung from the universe . mpe report 263 , garching , p. 209

the astrophysicist subrahmanyan chandrasekhar once referred to black holes as `` the most perfect macroscopic objects there are in the universe : the only elements in their construction are our concepts of space and time.''@xcite this special issue of journal of physics a is devoted to a completely different topic , the quantum - mechanical amplitudes for the scattering of relativistic particles , in which a revolution in our understanding has been taking place recently . however , scattering amplitudes represent a kind of short - distance complement to the perfection chandrasekhar saw in black holes . we might consider them to be the most perfect microscopic structures in the universe . black holes are gigantic and ( almost ) eternal . scattering amplitudes , in contrast , portray events that wink in and out of existence much faster than the blink of an eye . they describe processes at the shortest distance scales that can be probed in the laboratory . when gravitons scatter , interacting via einstein s field equations , again in some sense the only elements in the construction are our concepts of space and time although the boundary conditions at infinity can be more detailed than for black holes , involving multiple plane - wave ripples ( the gravitons ) propagating in various directions through minkowski space - time . when other relativistic particles scatter , as in non - abelian gauge theory , more information has to be provided , such as the gauge group , the coupling constant , the matter content of the theory , the particle masses , and the precise set of particle types and spins being scattered . however , in the ultra - relativistic limit , if there are only gauge interactions , much of the group - theoretical information can be stripped from the amplitudes , and certain color - ordered `` primitive '' amplitudes can be defined which are quite universal . these amplitudes have intricate analytic properties , which have been teased out over the decades , but especially in the past few years . a variety of new , efficient techniques have been found to construct amplitudes , and hidden symmetries have been identified which illustrate the amplitudes `` perfection . '' in the real universe , black holes are not found in an absolute vacuum . typically they are surrounded by accretion disks of gas and dust , which muddy their perfection somewhat , but also enrich the range of phenomena involving them , and make them much easier to observe , albeit indirectly . similarly , scattering amplitudes for quantum chromodynamics , or qcd , the most relativistic non - abelian gauge theory we can study in the laboratory , are shrouded from direct view by confinement . the quarks and gluons that theorists scatter mentally are never seen experimentally . in the initial state of a hadronic collision they are bound into a proton or other hadron . in the final state they emerge as collimated jets of particles . despite this fact , quantities that are sensitive only to very short distances , so - called infrared - safe observables such as jet production cross sections , can be computed reliably in terms of quarks and gluons , in a systematic expansion in the strong coupling @xmath1 . if we could go to asymptotically high energies , so that @xmath1 were infinitesimal , it would be enough to consider just the leading order in this expansion , which is generated by the born approximation , or tree - level amplitudes . the tree - level amplitudes of qcd have even more `` perfection '' than the generic , loop - level amplitudes , because they coincide with the tree amplitudes of a much more symmetric theory , @xmath0 supersymmetric yang - mills theory ( @xmath0 sym ) . this theory , which is the subject of many of the articles in this special issue , has the maximum amount of supersymmetry possible in a gauge theory . its four supercharges can be used to transform the helicity @xmath2 gluon state , by @xmath3 unit of helicity at a time , all the way to the helicity @xmath4 cpt conjugate gluon state . all states in the theory belong to a single supermultiplet , transforming in the adjoint representation of the gauge group . therefore all interactions are related by supersymmetry to the triple - gluon vertex , and the theory has a single dimensionless coupling @xmath5 . the @xmath6 function for @xmath0 sym vanishes exactly for any gauge group , @xmath7 , so that the theory is exactly conformally invariant at the quantum level , as well as classically . ( the classical conformal invariance of qcd with massless quarks is of course spoiled by its nonvanishing one - loop @xmath6 function , which leads to asymptotic freedom . ) in investigations of scattering amplitudes in @xmath0 sym , the gauge group under consideration is usually su@xmath8 . quite often , the limit @xmath9 is also taken , at a fixed value of the t hooft coupling , @xmath10 . for @xmath11 , perturbation theory can be applied , and for large @xmath12 only planar feynman diagrams contribute . for @xmath13 , perturbation theory breaks down , but the ads / cft correspondence @xcite can be used . in the large-@xmath12 or planar limit of @xmath0 sym , even more remarkable symmetries and relations emerge for scattering amplitudes . as we will discuss further below , these properties include dual conformal ( super)symmetry , which is part of a larger yangian invariance ; a description of amplitudes in terms of spaces of complex planes ( grassmannians ) ; and in terms of various types of twistors ; and a relation between scattering amplitudes and the expectation values of wilson loops for closed polygons bounded by light - like edges . many of these properties are intimately connected with a strong - coupling picture of gluon scattering in terms of strings moving in anti - de - sitter space @xcite . the exceptional simplicity and numerous hidden symmetries of @xmath0 sym have made it a playground for theorists interested in scattering amplitudes for a couple of decades . this interest has accelerated rapidly in the past few years . because the tree - level amplitudes of qcd and @xmath0 sym coincide , the early discoveries about qcd helicity amplitudes , motivated by the physics of jet production , were also discoveries about @xmath0 sym . for example , the parke - taylor formula @xcite for the maximal - helicity - violating ( mhv ) sequence of @xmath14-gluon amplitudes , which is just a single - term expression , even for an arbitrarily large value of @xmath14 , is also consistent with all the symmetries of @xmath0 sym @xcite , including some that were only unveiled decades later . many general properties of tree amplitudes were understood early on , such as their factorization properties @xcite and the ( @xmath15 or @xmath16 ) supersymmetry ward identities that they obey @xcite . however , a fuller understanding and exploitation of these properties have come only in recent years . for instance , the complete solutions of the long - known @xmath0 supersymmetry ward identities , and the corresponding @xmath17 identities in supergravity , have been worked out recently by elvang , freedman and kiermaier @xcite , and are reviewed here @xcite . several years ago , witten @xcite discovered that the twistor space developed decades earlier by penrose @xcite gave a natural description of tree - level amplitudes in massless gauge theory . as discussed in , such amplitudes have a natural description in terms of two - component weyl spinors , a left - handed and a right - handed spinor for each external state . to go to twistor space , one performs a fourier transform on half the variables , namely the left - handed spinors , leaving the right - handed ones alone . witten found that gauge theory could be reformulated in terms of a topological string moving in twistor space . gauge amplitudes turn out to be localized on particular curves in twistor space . in one approach , the curves are intersecting lines : a single line for the simplest mhv amplitudes , a pair of lines for the next - to - maximally - helicity - violating ( nmhv ) amplitudes , and so on . this version gave rise to the csw or mhv rules for gauge theory @xcite , whose mhv vertices are a particular off - shell continuation of the parke - taylor formulae . these developments are reviewed here by brandhuber , spence and travaglini @xcite , and by adamo , bullimore , mason and skinner @xcite . the latter reference also discusses another class of twistors , namely momentum twistors , which also provides a natural set of kinematic variables for amplitudes , in that they automatically satisfy the kinematic constraints of momentum conservation and the mass - shell conditions . finally , the correspondence between wilson loops and scattering amplitudes is described in this article @xcite from the perspective of twistor space . britto , cachazo , feng and witten ( bcfw ) @xcite recognized that factorization properties are powerful enough to allow the derivation of recursion relations @xcite for tree amplitudes , in terms of on - shell lower - point amplitudes , which are evaluated at particular complex momenta . this observation , reviewed here by brandhuber , spence and travaglini @xcite , makes essential use of the analyticity , or _ plasticity _ , of amplitudes ; that is , how they vary under smooth deformations of the kinematics . bcfw embedded an amplitude @xmath18 into a one - complex - parameter family of on - shell amplitudes @xmath19 with shifted complex momenta , and associated the poles of @xmath19 in the @xmath20 plane with factorization of the amplitude onto simpler lower - point amplitudes . one can also construct a supersymmetrized version of the bcfw recursion relation , in which grassmann variables @xmath21 associated with @xmath0 supersymmetry are shifted along with the momenta @xcite . this recursion relation can be solved explicitly to yield all tree amplitudes in @xmath0 sym @xcite ; the solution and its many symmetries are reviewed here by drummond @xcite . scattering amplitudes in massless gauge theories all have quite similar structure at tree level , once the color factors have been stripped off , as discussed in . on the other hand , the structure of loop amplitudes depends critically on the theory . the simplicity of one - loop multi - gluon amplitudes in @xmath0 sym played a key role in the development of the unitarity method @xcite . this method reconstructs loop amplitudes from their unitarity cuts , again exploiting analyticity . in general at one loop , as explained in this issue by britto @xcite and ita @xcite , unitarity cuts are matched against a decomposition of the amplitude in terms of a set of scalar integrals boxes , triangles , bubbles and ( sometimes ) tadpoles in order to determine the coefficients of the integrals . in @xmath0 sym , the high degree of supersymmetry implies that only boxes have non - vanishing coefficients . using unitarity , an infinite sequence of one - loop amplitudes ( the mhv amplitudes ) could be determined in @xmath0 sym from just the product of two tree - level mhv ( parke - taylor ) amplitudes @xcite . later , generalized unitarity was applied to the problem of computing one - loop amplitudes , by extracting additional information from products of three and four tree amplitudes , the so - called triple @xcite and quadruple @xcite cuts . quadruple cuts have four on - shell conditions . these four equations determine the four components of the ( four - dimensional ) loop momenta completely , up to a ( possible ) two - fold degeneracy . this realization allows the box coefficients to be computed very simply in terms of the product of four tree amplitudes , glued together as the four corners of a box , and evaluated at each of the two solutions for the loop momenta @xcite . for @xmath0 sym , there are no other coefficients to determine , so the one - loop problem has been `` reduced to quadrature . '' ( it turns out that maximal @xmath17 supergravity can be proven to have the same `` no - triangle '' property @xcite , so it too is solved at one loop by the quadruple cuts . ) generalized unitarity can also be applied very effectively at the multi - loop level . in principle it can be used for any gauge ( or gravitational ) theory . as an example , the two - loop four - gluon scattering amplitudes in qcd have been computed in this way @xcite . in practice the method has been pushed the furthest in @xmath0 sym . the basic techniques of multi - loop generalized unitarity are reviewed here by bern and huang @xcite , and by carrasco and johansson @xcite . returning now to one - loop amplitudes for generic gauge theories , including qcd , the triangle and bubble coefficients have to be determined , as well as certain _ rational parts _ , which have no unitarity cuts in four dimensions . the triple cuts contain information about the triangle coefficients , but they also receive contributions from boxes . similarly , the ordinary double cuts determine the bubble coefficients , once the contributions of triangles and boxes are removed . this separation can be done analytically , by making use of the different analytic behavior of the different types of cut integrals , as reviewed by britto @xcite . it can also be done numerically , by a suitable sampling of the continuum of loop momenta solving the triple or double cut on - shell conditions @xcite , as described by ita @xcite . these two articles also review methods for computing the rational parts of one - loop amplitudes . although the rational parts have no unitarity cuts in four dimensions ( by definition ) , they can be constructed via their cuts in dimensional regularization in @xmath22 dimensions . alternatively , it is possible to construct them recursively in the number of legs @xcite . on - shell loop amplitudes in massless non - abelian gauge theories always contain infrared divergences , due to the exchange of soft gluons or virtual collinear splittings . in qcd it is conventional to regulate these divergences , as well as the ultraviolet ones , using dimensional regularization . in supersymmetric theories , dimensional reduction @xcite , or the related four - dimensional helicity scheme @xcite , can be used to keep the number of bosonic and fermionic states the same , and preserve ward identities for ordinary supersymmetry . the general structure of the infrared divergences is well understood from decades of work in qed as well as qcd @xcite . it has been worked out in the most detail in the context of dimensional regularization @xcite . the basic picture is shown in . soft divergences and collinear divergences associated with the amplitude @xmath23 each have a universal form . they can be factorized from each other and from a hard , short - distance part of the amplitude . soft divergences , denoted by the blob marked @xmath24 , come from exchange of long - wavelength gluons . these gluons do not have the resolving power to probe the internal structure of the `` jets '' @xmath25 of virtual collinear particles which capture the collinear divergences . soft gluons can only see the overall color charges of the jets . an individual jet function @xmath26 depends on the type of particle @xmath27 , but not on the full amplitude kinematics . the soft function @xmath24 does not depend on the particle types , but only on their momenta and color quantum numbers . in general these quantum numbers can be mixed by gluon exchange , so @xmath24 is a matrix in color space . the hard function @xmath28 has no infrared singularities , but generically depends on the particle types , colors and kinematics . 4.5 truein in the planar or large-@xmath12 limit , the picture simplifies considerably , to that shown in . now @xmath23 represents the coefficient of a particular color structure , such as @xmath29 ( assuming that all external states are in the adjoint representation ; see ) . in the planar limit , individual soft gluons can only connect color - adjacent external partons . there is no mixing of different color structures at large @xmath12 . one can absorb the entire soft function @xmath24 into jet functions , which corresponds to breaking up the right - hand side of into @xmath14 wedges . each wedge is bounded by two hard lines , and is composed of `` half '' of each of the two jet functions , as well as the soft gluons exchanged between them . up to nonsingular terms , the wedge controlling the infrared divergences represents the square root of the sudakov form factor , which is defined as the amplitude for a color - singlet state to decay into a pair of ( adjoint ) gluons . 4.5 truein in dimensional regularization , the sudakov form factor obeys a particular differential equation @xcite , whose source term is called the cusp anomalous dimension @xmath30 @xcite . there is an additional constant of integration , called @xmath31 . these two functions ( along with the @xmath6 function in a non - conformal theory ) control the infrared divergences to all orders for any amplitude in a planar massless gauge theory . in planar @xmath0 sym , the cusp anomalous dimension has been determined to all orders in the coupling using integrability @xcite . the bds ansatz @xcite was built up from this description of the infrared singularities of planar amplitudes , plus the observation for the two- and three - loop four - point amplitudes that the hard function @xmath28 essentially reduces to a constant , independent of the kinematics . besides the cusp anomalous dimension , another place that integrability certainly enters multi - loop scattering amplitudes is in multi - regge - kinematics , a particular class of high - energy or small - angle limits of @xmath32 particle scatterings . these kinematics gave the first indication that the bds ansatz for mhv amplitudes has to be corrected beginning at the six - point level @xcite . ( it was previously argued @xcite using the properties of wilson loops , that the ansatz should be corrected at two loops for a large value of @xmath14 , unless the correspondence between mhv amplitudes and wilson loops were to break down . the hexagon wilson loop was then computed and found to differ from the ansatz prediction @xcite . ) the dynamics of gluons in the transverse plane is given , in the leading - logarithmic approximation , by the hamiltonian for a integrable open spin chain , as reviewed here by bartels , lipatov and prygarin @xcite . in planar @xmath0 sym , another infrared regulator is more convenient for many purposes , in particular for exploring the loop - level consequences of dual conformal symmetry , which is not preserved by dimensional regularization . the features of a recently - developed `` higgs regulator '' @xcite are reviewed in this issue by henn @xcite . for this regulator , vacuum expectation values are given to some of the adjoint scalar fields in the theory , breaking the gauge symmetry in such a way that ( in the planar limit ) the amplitudes are regulated by massive particles circulating around the outside of the loop diagrams . dual conformal symmetry remains exact in a certain sense . when the particle masses @xmath33 are taken to be much less than the momentum - invariants for the scattering process , logarithmic divergences develop , the analogs of the @xmath34 infrared poles in dimensional regularization . there is a hierarchy of simplicity in the scattering amplitudes for various types of gauge theory , as sketched in . this hierarchy begins to be revealed at one loop . the outer region of the diagram stands for a generic gauge theory with massive matter fields , and perhaps massive gauge bosons , if the gauge symmetry is spontaneously broken , as in electroweak theory . one - loop amplitudes in such a theory generically contain tadpole integrals . one - particle cuts are nontrivial , and are particularly delicate because of external - leg contributions @xcite . the cut structure of loop integrals containing massive propagators in the loop is generically somewhat more complicated than the purely massless case . massive particles in the loop can be unstable , which usually necessitates complex masses . when one enters the `` massless '' ring in , corresponding to massless gauge bosons and matter fields , most of these complications vanish , although there are still generically rational parts to compute . the ring `` sym '' stands for supersymmetric gauge theories . their one - loop amplitudes can be constructed from four - dimensional unitarity cuts alone , _ i.e. _ there are no non - trivial rational parts @xcite . 5.2 truein moving further inward in , we arrive at @xmath0 sym . as mentioned earlier , at one loop the coefficients of bubble and triangle integrals now vanish , as well as the independent rational parts . ( there are other gauge theories with vanishing bubble and triangle coefficients , at least for their @xmath14-gluon amplitudes @xcite . ) the theory becomes conformally invariant . it has been conjectured that the _ leading singularities _ the multi - loop analogs of the quadruple cuts are sufficient to determine the amplitudes at any loop order @xcite . in addition , scattering amplitudes have empirically a predictable , uniform transcendental weight @xcite . this weight refers to their construction out of polylogarithms , logarithms , and riemann @xmath35 values . for example , the finite ( @xmath36 ) terms in one - loop @xmath0 sym amplitudes are of weight two : they contain some terms proportional to the polylogarithm @xmath37 , and others which are products of two logarithms , or proportional to @xmath38 , but they do not contain any terms of lower transcendentality . ( at one loop this result just follows from the absence of bubble and tadpole integrals . ) wherever analytic results are available , this uniform transcendentality property holds . for example , at two loops ( weight four for the finite terms ) it holds for the non - planar , subleading - color terms as well as the planar ones @xcite . finally , for planar @xmath0 sym , as mentioned earlier , additional symmetries appear , dual superconformal invariance and an associated yangian symmetry . the extra symmetries are related to integrability . they give rise to the prospect of adapting methods developed for other integrable systems to determine the @xmath24-matrix exactly , at least in the large @xmath12 limit . indeed , as reviewed by drummond @xcite , an integrand exhibiting manifest yangian invariance has been proposed recently @xcite for the general @xmath39-loop @xmath14-point amplitude , based on the bcfw recursion relations ; see also ref . @xcite . at the level of integrated amplitudes , dual conformal invariance has an anomaly due to the need for an infrared regulator . the anomaly was first understood in terms of wilson loops rather than amplitudes @xcite ( for which divergences are ultraviolet in nature , rather than infrared ) . nevertheless , dual conformal symmetry completely fixes the form of the four- and five - point amplitudes . the form to which they are fixed turns out to be precisely the abdk / bds ansatz @xcite , which was developed earlier based on the structure of infrared divergences of amplitudes , plus patterns observed at two and three loops . as discussed in , for six external legs one can build three different combinations of kinematic variables that are invariant under all dual conformal transformations , the cross - ratios @xmath40 , @xmath41 and @xmath42 . at the six - point level , the abdk / bds ansatz fails at the first nontrivial order , two loops @xcite , and dual conformal symmetry allows for an additional `` remainder function '' , @xmath43 , which depends on the three cross - ratios . at strong coupling , large @xmath44 and large-@xmath12 , alday and maldacena used the ads / cft correspondence to map the problem of gluon scattering to that of strings moving in anti - de - sitter space @xcite . the ads background is weakly curved in this limit , relative to the string scale . another way of saying this is that the two - dimensional string world - sheet stretches over a large area , relative to the scale on which strings fluctuate , allowing a semi - classical expansion to be used at large @xmath44 . in the leading term in the expansion , different helicity configurations are not distinguished . the amplitude is given simply by @xmath45 , where @xmath46 is the classical action that minimizes the area ( at least for scattering configurations with a euclidean interpretation ) . using a @xmath47-duality transformation in string theory , alday and maldacena showed that the boundary conditions for the world - sheet are closed polygons , where each edge is a light - like segment given by the momentum @xmath48 of the @xmath49 gluon . the solution to the minimal area problem was found explicitly for four - gluon scattering . more recently , in order to handle more complicated kinematical configurations than the four - point case , the integrability of the string sigma - model action @xcite has been exploited . the minimal area problem has been solved by mapping it to a system of equations identical to those of the thermodynamical bethe ansatz @xcite . the first computation at strong coupling was for the four - gluon amplitude , and matched precisely the prediction of the abdk / bds ansatz @xcite . but it also suggested a weak - coupling correspondence between scattering amplitudes and wilson loops for closed light - like polygons , which was rapidly established for a variety of cases , initially for the simplest case of mhv scattering amplitudes @xcite . the wilson - loop correspondence has been very important both conceptually and technically . conceptually , it is still not entirely clear why it happens at all at weak coupling . technically , given its existence , it allows amplitudes to be computed in terms of wilson line integrals , which are generally somewhat more tractable than feynman loop integrals . some recent conceptual advances have come from considering certain correlation functions with close - to - light - like separations @xcite . the relative simplicity of the wilson line integrals for the six - point ( hexagon ) case allowed the remainder function @xmath43 to be evaluated analytically in terms of goncharov polylogarithms @xcite . the expression in ref . @xcite was then simplified considerably , to a few lines involving classical polylogarithms @xmath50 , using properties of the symbol operation which captures all of the essential analytic behavior of a multi - variable function @xcite . these results open the door to the possibility of finding simple analytic results for more general kinematic configurations . for some configurations in which the external momenta are restricted to two space - time dimensions , two - loop analytic results are already available beyond six external legs @xcite . there is certainly the prospect of more multi - loop analytic results on the horizon , for both wilson loops and scattering amplitudes in planar @xmath0 super - yang - mills theory . gravity seems to be a completely separate force from gauge theory . it has a spin two force carrier instead of spin one , no color degrees of freedom , and a dimensionful coupling constant . nevertheless , the two theories are intimately connected . the ads / cft correspondence @xcite relates the two , holographically , as a _ weak - strong _ duality . as mentioned earlier , strong - coupling scattering of gluons has an alternate description in terms of strings moving in a weakly - curved five - dimensional gravitational background . on the other hand , there is also a _ weak - weak _ duality of some kind between gravity and gauge theory , directly in four dimensions : it is possible to write perturbative scattering amplitudes for gravitons in terms of `` double copies '' of gluon amplitudes . the original examples of such relations are due to kawai , lewellen and tye ( klt ) @xcite , who found them by first deriving relations between closed and open string theory tree amplitudes , whose low - energy limits give tree - level relations in field - theory . graviton amplitudes are represented in terms of sums of products of pairs of gluon amplitudes with different color orderings . the klt relations hold not only for pure - graviton scattering , in terms of pure - gauge theory , but also for @xmath17 supergravity ( or any subsector of it ) in terms of amplitudes for @xmath0 sym . more recently it has been recognized that other double - copy formulas exist @xcite . the new formulas are simpler in some sense than the klt relations ; they involve the direct squares of gauge - theory components , whereas the klt relations employ multiple permutations of full gauge - theory amplitudes . as reviewed in this issue by carrasco and johansson @xcite , these relations , combined with generalized unitarity , have important consequences for multi - loop amplitudes in @xmath17 supergravity , because they map the problem to a much simpler one , the corresponding multi - loop amplitudes for ( non - planar ) @xmath0 sym . in the next two sections we discuss a few of the basic tools used to organize and illuminate the color and kinematic structure of amplitudes , which underlay the discussions in many of the other articles in this special issue . in this section we describe some common conventions for organizing the color structure of su@xmath8 gauge theory amplitudes , which are used elsewhere in this special issue . most of the articles implicitly discuss color - ordered partial amplitudes , which are particularly convenient in the large-@xmath12 limit in which planar diagrams dominate . color - ordered amplitudes emerge from a `` trace - based '' color decomposition ( as also reviewed in _ e.g. _ refs . @xcite ) . however , in some cases , in particular for describing subleading - color terms in amplitudes , and color - kinematic duality relations @xcite , decompositions based on the su@xmath8 structure constants are more useful . in general , we consider two different su@xmath8 representations for the external states : * the adjoint representation , for the gluon and any of its superpartners ( _ i.e. _ for all states in @xmath0 sym ) . adjoint color indices are denoted by @xmath51 . * the fundamental ( defining ) representation @xmath12 , and its conjugate representation @xmath52 , for quarks and anti - quarks respectively . fundamental color indices are denoted by @xmath53 , and anti - fundamental @xmath52 indices by @xmath54 . the generators of su@xmath8 in the fundamental representation are traceless hermitian @xmath55 matrices , denoted by @xmath56 . it is conventional when discussing helicity amplitudes to normalize the generators by @xmath57 , in order to avoid a proliferation of @xmath58 s . in qcd , the group theory ( color ) factors in feynman diagrams are of two types : @xmath56 for the gluon - quark - antiquark vertex , and the su@xmath8 structure constants @xmath59 ( or products of @xmath59 s ) for all other vertices . ( it is also convenient to normalize the structure constants differently , using @xmath60 , where @xmath59 is the standard , textbook normalization . ) the color factors can be represented diagrammatically @xcite using feynman - diagram notation , as in . lines with arrows ( quarks ) carry fundamental representation indices ; curly lines ( gluons ) carry adjoint indices . the adjoint representation is also a bi - fundamental ( @xmath61 ) representation from which the singlet , or trace component , has been projected out , as in the second equation for @xmath62 in the figure . this relation also gives rise to the su@xmath8 fierz identity , ( t^a)_i_1^ _ 1 ( t^a)_i_2^ _ 2 = _ i_1^ _ 2 _ i_2^ _ 1 - 1n_c _ i_1^ _ 1 _ i_2^ _ 2 , [ sunfierz ] which allows one to simplify products of traces of @xmath63 s . 4.8 truein color - ordering @xcite is related to the t hooft double - line formalism @xcite . we begin by rewriting the color factors entirely in terms of the @xmath63 generators , using the relation ^abc if^abc= ( [ t^a , t^b ] t^c ) , [ removefabc ] which follows from the definition of the structure constants , and is depicted graphically in . after inserting repeatedly into the color factor for a typical feynman diagram , one obtains a large number of traces of the generic form @xmath64 . if the amplitude has external quark legs , then there will also be strings of @xmath63 s terminated by fundamental indices , of the form @xmath65 , one for each external quark - antiquark pair . the number of traces can be reduced considerably by repeated use of the su@xmath8 fierz identity ( [ sunfierz ] ) . in the case of tree - level amplitudes with external states in the adjoint representation , such as @xmath14-gluon amplitudes , sketches how the color factors all may be reduced to a single trace , @xmath66 , for some permutation @xmath67 of the @xmath14 gluons . this reduction leads to the _ trace - based color decomposition _ for @xmath14-gluon tree amplitudes , -20 mm a^_n(\{k_i , h_i , a_i } ) = g^n-2 -1.3 mm _ s_n / z_n -1.3 mm ( t^a_(1)t^a_(n ) ) a_n^((1^h_1), ,(n^h_n ) ) . [ treegluecolor ] here @xmath68 is the full amplitude , with dependence on the external gluon momenta @xmath48 , @xmath69 , helicities @xmath70 , and adjoint indices @xmath71 . in qcd , the gauge coupling @xmath5 is related to the strong coupling by @xmath72 . the _ partial amplitudes _ ( or _ primitive _ or _ color - stripped amplitudes _ ) @xmath73 have had all the color factors removed , but contain all the kinematic information . cyclic permutations of the arguments of a partial amplitude , denoted by @xmath74 , leave it invariant , because the associated trace is invariant under these operations . however , all @xmath75 non - cyclic permutations , or orderings , of the partial amplitude appear in . these permutations are denoted by @xmath76 . 5.0 truein looking again at the way the different trace factors arise in , one sees that the partial amplitude @xmath77 only receives contributions from tree - level feynman diagrams that can be drawn on a plane , in which the cyclic ordering of the external legs , @xmath78 , matches the ordering of the arguments in @xmath79 . therefore each partial amplitude can only have singularities in momentum invariants formed by squaring color - adjacent sums of momenta , such as @xmath80 , @xmath81 , _ etc . _ , where all indices are defined modulo @xmath14 . in this way , the color decomposition ( [ treegluecolor ] ) disentangles the kinematic complexity of the full amplitude @xmath82 . similarly , tree amplitudes with two external quarks and @xmath83 gluons can be reduced to single strings of @xmath63 matrices , ^_n(q_1,g_3, ,g_n-1,|q_n ) & = & g^n-2 -1.3 mm _ s_n-2 -1.3 mm ( t^a_(2)t^a_(n-1))_i_1^ _ n + & & 1.3 cm a_n^(1_q,(2), ,(n-1),n_|q ) , [ treequarkgluecolor ] where we have omitted the helicity labels , and numbers without subscripts in the argument of @xmath79 refer to gluons . in this case there are @xmath84 terms , corresponding to all possible gluon orderings between the quarks . color decompositions for amplitudes containing additional external quark pairs have also been described @xcite . we note that the partial amplitudes appearing in could also be used to describe the scattering of two fermions with different color quantum numbers , and @xmath83 gluons . for example , if the fermions are gluinos in the adjoint representation , then the color decomposition has the form of , but the partial amplitudes , with two external fermions , are of the type given in . this is easy to see if the two fermions are color - adjacent ; essentially all one needs to do to go between the two cases is remove one color line running between the two gluinos . however , the other cyclic orderings can also be obtained from @xmath85 , using the kleiss - kuijf relations discussed in the next subsection . this property illustrates the universality of the color - ordered primitive amplitudes alluded to in the introduction . many articles in this issue consider loop - level amplitudes in which all of the external states are gluons in the adjoint representation . the general color decomposition here is similar to , except that multiple color traces are now possible . roughly speaking , a loop of gluons carries a fundamental and an anti - fundamental index around the loop with it , and an external gluon can attach to either index . the number of traces at @xmath39 loops is equal to @xmath86 . at one loop , the full color decomposition for @xmath14 gluons is @xcite -24 mm a^_n(\{k_i , h_i , a_i } ) & = & g^n . [ loopgluecolor ] here @xmath87 are the partial amplitudes , @xmath74 and @xmath88 are the subsets of @xmath89 that leave the corresponding single and double trace structures invariant , and @xmath90 is the greatest integer less than or equal to @xmath91 . the formula is more complicated than at tree level , but only by the need to keep track of more distinct trace structures and their various symmetries . if there are @xmath92 flavors of quarks in the loop , it is easy to see that they only contribute to the single - trace coefficient @xmath93 in , and with a weight @xmath94 relative to the adjoint gluons . ( similar color decompositions are available for one - loop amplitudes with an external quark pair @xcite . ) when one constructs the color - summed cross section , as is usually required for qcd applications , the contribution of the double - trace coefficients , @xmath87 for @xmath95 , is suppressed by a power of @xmath96 with respect to that of @xmath93 . ( it also turns out that the @xmath87 are not really independent , but can be computed as a sum over permutations of the @xmath93 @xcite . ) thus simplifies a lot in the large-@xmath12 limit . the factor of @xmath97 simply comes from those terms in which all external gluons attach to the same fundamental index line , leaving the trace of the identity matrix for the untouched line . correspondingly , if we only want the leading terms in the large-@xmath12 limit at @xmath39 loops , we can write the compact color decomposition , & & -1.5 cm a^_n(\{k_i , h_i , a_i})|_leadingcolor + & & -0.5 cm = g^n-2 ( g^2 n_c)^l _ s_n / z_n ( t^a_(1)t^a_(n ) ) a_n^(l)((1^h_1), ,(n^h_n ) ) , [ lloopleadingcolor ] where we have dropped the `` ; 1 '' index on the leading - color partial amplitude . the t hooft coupling @xmath10 emerges naturally in this limit . the articles in this issue about multi - loop amplitudes in planar @xmath0 sym are generally concerned with the @xmath98 in . because each loop integration typically brings a factor of @xmath99 ( or @xmath100 when using dimensional regularization in @xmath22 ) , in some cases the precise normalizations of the @xmath98 will differ by this factor , as well as possibly other conventional factors . there is another type of color decomposition that is actually more useful than the above trace - based decomposition for addressing certain issues , and particularly for working beyond the large-@xmath12 approximation . we could call this approach the @xmath101-based decomposition , because it goes back to the original color factors built out of the structure constants @xmath59 . ( now we assume that all states are in the adjoint representation . ) the diagrammatic representation of a general @xmath101-based color factor at tree level has the structure of a cubic tree graph , with vertices given by @xmath102 s and propagators given by @xmath62 s . however , these color factors are not all independent , due to the color jacobi identity . this identity can be written in different ways . one way , ^dae ^bce - ^dbe ^ace = ^abe ^cde , [ colorjacobiidentity ] which is also depicted in , corresponds to the fact that the structure constants are also the su@xmath8 generators in the adjoint representation , @xmath103 , so that their commutator gives @xmath104 contracted with @xmath105 . 4.0 truein at tree level , one can use the identity ( [ colorjacobiidentity ] ) repeatedly to transform all @xmath101-based color factors into a `` multi - peripheral '' form in which two selected gluons , say 1 and @xmath14 , are always at the end of a long chain of structure constants , and the other @xmath83 gluons are emitted from along the ladder @xcite , -20 mm a_n^(\{a_i } ) & = & g^n-2 _ s_n-2 ^a_1 a_(2 ) x_1 ^x_1 a_(3 ) x_2 ^x_n-3 a_(n-1 ) a_n + & & a_n^(1,(2), ,(n-1),n ) + & = & g^n-2 _ s_n-2 ( f^a_(2 ) f^a_(n-1 ) ) _ a_1 a_n a_n^(1,(2), ,(n-1),n ) . [ gluondecompnew ] notice that this representation is identical in form to the decomposition for two quarks and @xmath83 gluons in , except for the representation used for the su@xmath8 generator matrices . although correspond to the cases of fundamental and adjoint representations , respectively , the same type of decomposition clearly holds for two external matter fields in an arbitrary su@xmath8 representation . the fact that the coefficients of the @xmath101-based decomposition ( [ gluondecompnew ] ) are identical to the color - ordered partial amplitudes @xmath79 appearing in the trace - based decomposition can be demonstrated by contracting both representations with @xmath29 and keeping only the leading terms at large @xmath12 @xcite . note that there are only @xmath84 terms in the @xmath101-based color decomposition , in contrast to the @xmath75 terms in the trace - based decomposition ( [ treegluecolor ] ) . therefore there are identities between the @xmath79 , which are group - theoretical in nature generalizations of certain @xmath106 decoupling identities @xcite . these identities can be used to put any two legs next to each other , for example legs 1 and @xmath14 : a_n^(1,\{},n,\ { } ) = ( -1)^n _ _ \{}\{^t } a_n^(1,(\{}\{^t}),n ) , [ kkequation ] where @xmath107 and @xmath108 are two sets of external gluons , whose union is @xmath109 . also , @xmath110 is the number of elements in the set @xmath108 , the set @xmath111 is @xmath108 with the ordering reversed , and @xmath112 is the set of ordered permutations , or `` mergings '' , of the two sets @xmath107 and @xmath111 that preserve the ordering of the @xmath113 within @xmath107 and of the @xmath114 within @xmath111 , while allowing for all possible relative orderings of the @xmath113 with respect to the @xmath114 . these relations are known as the kleiss - kuijf relations @xcite . they can be derived from by expanding out the @xmath102 factors using , identifying the coefficients of the trace structures , and comparing them with the original trace - based decomposition ( [ treegluecolor ] ) @xcite . in fact , it has been realized recently that there are not even @xmath84 independent color - ordered amplitudes @xmath79 , but only @xmath115 @xcite , as reviewed in this issue by carrasco and johansson @xcite . the additional linear relations between @xmath79 s are not purely group - theoretical in nature , but also involve kinematical factors . they allow one to put any _ three _ external gluon legs next to each other . for example , for legs 1 , 2 and @xmath14 , the identities take the form , -1.0 cm a_n^((1),(2), ,(n ) ) = _ s_n-3 k^ ( ) _ a_n^(1,2,(3), ,(n-1),n ) , [ bcj3 ] where @xmath116 and @xmath117 are permutations and @xmath118 are kinematic - dependent ( but not state - dependent ) coefficients @xcite . these relations were first proved using open string theory @xcite , using a contour deformation argument @xcite similar to that used in deriving the klt relations @xcite between gravity and gauge theory amplitudes . they have also been proven directly in field theory , using a recursive argument @xcite . although expresses the consequences of certain color - kinematic identities in the trace basis , the identities themselves are more naturally stated in an @xmath101-based approach . for this purpose it is useful to _ not _ transform the @xmath101-based color factors into multi - peripheral form , but leave the color decomposition free , in terms of the set of all cubic graphs @xmath119 , writing _ n^= g^n-2 _ i_3 . [ bcjtreerep ] the color factors @xmath120 are products of structure constants for each vertex @xmath121 of the @xmath49 graph , contracted together along the propagators , @xmath122 . the factors of @xmath123 are scalar propagators , one for each internal line in the graph , while the @xmath124 are kinematical numerator factors . consider a triplet of graphs in @xmath119 , which are identical except for a region from which four lines emanate . within this region they differ according to the three ways of joining four legs with two cubic vertices , shown in . call the three graphs @xmath125 , @xmath126 and @xmath127 , and let the color factors @xmath128 be normalized ( signed ) so that they obey the jacobi identity ( [ colorjacobiidentity ] ) as @xmath129 . then the statement of color - kinematic duality @xcite is that the associated numerator factors should be related by @xmath130 , for every such triplet of graphs . a simple and intuitive argument for these relations has been provided based on the heterotic string @xcite . the existence of such a representation with _ @xmath124 ( _ i.e. _ polynomials in the momenta ) has been checked in various examples , not only at tree level , but also through three loops for @xmath0 sym amplitudes @xcite . as reviewed here by carrasco and johansson @xcite , the existence of a representation for gauge theory amplitudes satisfying @xmath130 for each triplet of jacobi - related cubic graphs has important implications . large classes of numerator factors are related to each other , and gravitational amplitudes are easily constructed from gauge theory ones . now that we have described two different ways to organize the color quantum numbers for scattering amplitudes , let us turn to kinematical issues , including convenient choices of external states and kinematical variables . in qcd , the helicities of massless quarks are conserved by their chirality - preserving interactions with gluons . hence it is natural to use a helicity basis for the quark amplitudes . in four - component notation , the spinors for the external states are taken to be @xmath131 for a quark with momentum @xmath132 , and @xmath133 for an anti - quark . however , in the massless limit these spinors can be chosen to be equal to each other , @xmath134 , and we can use two - component weyl spinors as well . we will see that helicity amplitudes for external gluons can be built from the same objects . we want to consider amplitudes with @xmath14 different momenta @xmath48 , @xmath69 , and to do so in a way that respects crossing symmetry . thus we take all the momenta to be outgoing , so that momentum conservation reads @xmath135 . in a realistic process with some incoming particles , the physical momentum for each of the incoming particles is simply the _ negative _ of the corresponding @xmath48 . also , the physical helicity @xmath136 is the negative of the one by which we label the amplitude . ( the spin @xmath137 does not reverse under crossing , but @xmath138 does . ) the two - component spinor for the @xmath49 state , if it is an outgoing fermion with helicity @xmath139 , has several notations @xcite : ( _ i ) _ & & |i^+ |k_i^+ [ u_+(k_i ) ] _ = [ v_-(k_i ) ] _ , + ( _ i ) _ & & |i^- |k_i^- [ u_-(k_i ) ] _ = [ v_+(k_i ) ] _ . similarly , the conjugate spinors are ( _ i)^ & & i^+| k_i^+| [ ] ^= [ ] ^ , + ( _ i)^ & & i^-| k_i^-| [ ] ^ = [ ] ^. two - component indices are raised and lowered with the two - dimensional antisymmetric tensors @xmath140 , @xmath141 , _ etc . _ the spinor inner products come from contracting spinors for different momenta with these tensors , .l & = & ^ ( _ j)_(_l ) _ = j^- | l^+ = u_+(k_l ) , [ spadef ] + .l & = & ^ ( _ j ) _ _ = j^+ | l^- = u_-(k_l ) . [ spbdef ] spinor products are antisymmetric under exchange of labels , @xmath142 , @xmath143 , and @xmath144 . lorentz vectors can be written as bi - spinors , or @xmath145 matrices , by contracting them with the pauli matrices . a massless vector written in this way factorizes into the product of two massless spinors , as ( _ i ) _ k_i^ ( _ ) _ = u_+(k_i ) = ( _ i)_(_i ) _ . [ masslesskfact ] this factorization plays a role in the bcfw complex - momentum shift , which is best described in terms of the @xmath44 and @xmath146 variables @xcite , as is done elsewhere in this issue @xcite . amplitudes with external gluons can also be described in terms of the @xmath147 and @xmath148 variables , thanks to the spinor - helicity formalism @xcite . the polarization vector @xmath149 for an outgoing massless vector particle with momentum @xmath132 and helicity @xmath150 , is required to be transverse to @xmath132 , @xmath151 . in the spinor - helicity formalism , @xmath152 is also chosen to be transverse to another massless momentum @xmath153 , called the _ reference momentum _ ( which should not be parallel to @xmath132 but is otherwise arbitrary ) . note that @xmath132 and @xmath153 span a two - dimensional subspace of four - dimensional space - time . because @xmath154 , the two orthogonal directions to this subspace are spanned by @xmath155 and @xmath156 . the spinor - helicity polarization vectors live in this subspace ; indeed , they are proportional to these two vectors , and are normalized by the condition that @xmath157 : _ ^(k , q ) = . [ spinorheldef ] as a bi - spinor , @xmath158 is given by -1 cm [ ^+(k_i , q_i ) ] _ = , _ = - . [ spinorheldef2 ] if each @xmath159 is chosen to be another momentum in the process , say @xmath160 , then it is clear from the general form of the feynman rules ( using also ) that the full amplitude can be built entirely out of the spinor products @xmath161 and @xmath162 , for @xmath163 . what are the basic properties of the spinor products which make them so convenient variables for helicity amplitudes ? first of all , for real momenta , the two types of spinor products are complex conjugates of each other , @xmath164 . also , for either real or complex momenta they `` square '' to give the ordinary dot products of momenta , .j.l = = 2 k_jk_l = s_jl . [ sfromspinors ] ( is written in the convention usually used in the qcd literature , but the reader should be aware that another convention exists , in which the sign of @xmath162 is reversed so that @xmath165 . ) for real momenta , the conjugation property and imply that .l = e^i_jl , .l = e^-i_jl , [ realmomsqrt ] where @xmath166 is a phase this phase shifts as @xmath160 and @xmath167 are rotated in azimuthal angle around the axis corresponding to their sum , @xmath168 . one way of seeing the utility of the spinor products for helicity amplitudes is to examine the limits in which two particles become collinear , @xmath169 . ( a ) shows that the typical behavior of amplitudes in a massless scalar field theory is @xmath170 , coming just from the scalar propagator . however , in massless gauge theory , there are numerator factors in the feynman diagrams which lessen this divergence . more physically , the factorization in the collinear limit should be onto a physical gluon state with helicity @xmath171 . however , the sum of the two external helicities must be @xmath172 or 0 . therefore there is at least a @xmath171 mismatch of the spin angular momentum along the @xmath173 axis , as illustrated in ( b ) . the spin angular - momentum mismatch requires some orbital angular momentum , which in turn causes a suppression of the magnitude of the amplitude , from @xmath174 to @xmath175 . it also dictates a phase shift under azimuthal rotation of @xmath160 and @xmath167 about the @xmath173 axis . both of these properties are captured by the spinor products , making them ideal for describing helicity amplitudes . 4.5 truein the simplest nonvanishing tree - level gluon amplitudes are the parke - taylor or mhv amplitudes @xcite , a_n^(1^+, ,l^-, ,m^-, ,n^+ ) = i .m^4 1.22.3.1 . [ mhvpt ] here exactly two gluons , @xmath176 and @xmath177 , have negative helicity , and the remaining @xmath83 gluons have positive helicity . the amplitudes with zero or one negative gluon helicity vanish by supersymmetry ward identities @xcite . the only denominator factors in these amplitudes are the spinor products for color - adjacent pairs of momenta , @xmath178 . they capture such universal collinear limits @xcite as , & & -1 cm a_n^(1^+, ,l^-, ,m^-, ,(n-1)^+,n^+)|_n-1n [ collimitexample ] + & ~ & a_n-1^(1^+, ,l^-, ,m^-, ,p^+ ) , where legs @xmath179 and @xmath14 are becoming parallel , with @xmath180 , @xmath181 and @xmath182 . another way to see why the spinor products are so useful is to consider the special case @xmath183 for the mhv amplitudes , a_3^(1 ^ -,2 ^ -,3^+ ) = i .2 ^ 4 1.22.33.1 . [ mmp ] one might think that the scattering of three massless particles is singular , because all of the @xmath184 vanish since they are equal to the squared momentum of the remaining particle . indeed , for real momenta , implies that all the spinor products vanish too , and there is no way to build a nonvanishing and nonsingular amplitude . however , for complex momenta , the complex - conjugation relation @xmath164 no longer holds , and it is possible for half the spinor products to be nonvanishing , while the other half vanish . specifically , we can choose the three negative - helicity two - component spinors to be proportional @xcite , _ 1^ _ 2^ _ 3 ^ . [ spbvanish ] then according to we have @xmath185 . however , the other three spinor products , @xmath186 , @xmath187 and @xmath188 , are allowed to be nonzero . ( this is consistent with the vanishing of , which still holds for complex momenta . ) for this choice of complex kinematics , the mhv three - point amplitude ( [ mmp ] ) is nonvanishing and nonsingular , while the conjugate `` @xmath189 '' amplitude , a_3^(1^+,2^+,3 ^ - ) = -i 1.2 ^ 41.22.33.1 , [ ppm ] vanishes . in contrast , for the conjugate kinematics with _ 1^ _ 2^ _ 3 ^ , [ spavanish ] vanishes while is nonvanishing and nonsingular . the three - point amplitudes ( [ mmp ] ) and ( [ ppm ] ) , as well as related amplitudes with two matter particles and one gluon , are very important in massless gauge theory . the bcfw relations build all tree - level scattering amplitudes recursively from them , using factorization onto multi - particle poles , as well as onto complex - momentum versions of the collinear poles shown in . of course feynman diagrams can also be used to build amplitudes from three- and four - point vertices . however , feynman vertices are typically evaluated with off - shell lines emanating from them , which makes them gauge - dependent . in contrast , are on shell , albeit for complex momenta , and therefore they are fully gauge invariant . in the recursive construction of @xmath14-gluon amplitudes , the four - gluon vertex is never needed , essentially because it is related to the three - vertex by gauge transformations . another reason why the three - point amplitudes are important is for building up loop amplitudes through generalized unitarity , as reviewed in this issue by ita @xcite and britto @xcite at one loop , and by bern and huang @xcite and carrasco and johansson @xcite at the multi - loop level . typically , generalized unitarity cuts place many propagators on shell , so they very often contain at least one three - point tree amplitude . finally , the relations between gravity and gauge theory tree amplitudes are the simplest of all for the three - point case , where they can be written as an exact square , m_3^(1,2,3 ) = [ a_3^(1,2,3 ) ] ^2 , [ klt3 ] ignoring overall coupling - constant factors . for planar ( large @xmath12 ) gauge theory amplitudes which includes all tree - level amplitudes new sets of kinematic variables have proved very useful for identifying new structures and symmetries . because the external legs have a definite cyclic order in the planar case , they carve the plane into sectors . to each such sector , one can assign a vertex @xmath190 for a dual graph , as shown in ( a ) . lines connecting vertices of the dual graph cross lines of the original graph . the latter lines carry momenta . differences between @xmath191 s are computed according to the net momentum flowing through the lines of the original graph . thus in ( a ) , adjacent sectors are separated by @xmath192 , or more generally , x_i , j^x_i^- x_j^= k_j^+ k_j+1^+ + k_i-1^. [ sectorvariable ] the @xmath191 are often called _ dual coordinates _ ( or sometimes region or sector variables ) , but they are essentially momenta , not coordinates . in a planar loop graph we can assign additional @xmath191 variables to the faces inside each loop , corresponding to the independent loop momenta . ( b ) shows a two - loop example . here the thick lines of the dual graph that are shown each cross ( and therefore correspond to ) one propagator of the planar double box integral . 4.5 truein more precisely , differences of the @xmath191 are momenta . the @xmath191 themselves are not constrained by momentum conservation . they are constrained by on - shell conditions , @xmath193 . however , unlike momenta , it is possible to invert the @xmath191 s , according to x_i^ , x_i , j^2 . [ dualinversion ] remarkably , this inversion is a symmetry of both integrands and amplitudes in planar @xmath0 sym . the integrands are functions of the combinations @xmath194 , which are manifestly invariant under lorentz transformations and translations of the @xmath191 s . ( translation invariance is guaranteed simply because the amplitude depends on momenta , which are differences of @xmath191 s . ) therefore the inversion ( [ dualinversion ] ) combines with poincar invariance to generate the conformal group so(4,2 ) . this group acts on the dual variables @xmath191 , not on space - time coordinates ; hence the symmetry is referred to as dual conformal invariance . at the loop level , dual conformal invariance can be spoiled by the infrared regulator needed for on - shell amplitudes . a loop - integration measure in four dimensions , @xmath195 , transforms under an inversion ( [ dualinversion ] ) as , @xmath196 . in @xmath197 this factor precisely cancels terms from inverting the associated propagators in graphs such as the one in ( b ) . however , in dimensional regularization , we have @xmath198 , and the invariance is lost due to extra factors in the transformation of the loop integration measure . as discussed in detail in this issue by henn @xcite , for planar @xmath0 sym it is possible to retain dual conformal invariance at the loop level by using instead a higgs regulator @xcite , which generates particle masses that can also be thought of as extra components of the @xmath190 . although amplitude computations with the higgs regulator have not been pushed quite as far yet as in dimensional regularization , there are certain advantages to this approach . for example , in dimensional regularization , terms in lower - loop amplitudes that vanish as @xmath199 often have to be computed , because they can multiply pole terms , @xmath200 , in other amplitudes ( or in the squares of amplitudes needed to construct differential cross sections ) . the article by schabinger in this issue discusses such @xmath201 contributions @xcite . in contrast , with the higgs regulator , the singularities have the form of powers of logarithms , @xmath202 , and power - suppressed terms of the form @xmath203 can always be dropped , because they vanish faster than any power of a logarithm . how constraining is dual conformal symmetry ? in general it can only determine amplitudes up to arbitrary functions of the cross - ratios , u_ijkl , [ genu ] because these variables are invariant under the inversion ( [ dualinversion ] ) . however , the on - shell constraints @xmath204 imply that there are no non - trivial cross ratios for four- and five - particle scattering . at the six - point level there are three non - trivial cross ratios , utilizing the allowed @xmath205 and @xmath206 kinematic invariants : [ defu ] -1.8 cm u_1 = = , u_2 = = , u_3 = = . as mentioned earlier , the remainder function @xmath207 characterizing the two - loop mhv six - point amplitude in planar @xmath0 sym is a function of these three variables , and this function is now known analytically in terms of polylogarithms @xcite . in planar @xmath0 sym , dual conformal invariance combines with dual @xmath0 supersymmetry to generate dual superconformal symmetry @xcite . it is natural to consider super - amplitudes , which package sets of amplitudes together by using an @xmath0 on - shell superfield @xcite , [ super - wave ] -2.3 cm ( ) = g^+ + ^a g_a + ^a ^b _ ab + ^a^b^c _ abcd |g^d + ^a^b^c ^d _ abcd g^- . here @xmath208 are the @xmath209 helicity gluons , @xmath210 and @xmath211 the four flavors of @xmath212 helicity gluinos , and @xmath213 the six real @xmath214 helicity scalar states . integrations over the grassmann variables @xmath21 , @xmath215 , can be used to pick off the desired component amplitudes from the super - amplitudes , defined by [ super - amplitude ] a_n(_i ) a ( _ 1(_1 ) , , _ just as the @xmath191 bosonic variables automatically satisfy momentum conservation , @xmath216 , one can construct their superpartners @xmath217 , whose differences satisfy _ i^a - _ i+1^a = _ i^_i^a , [ dualsuper ] for @xmath69 . super - momentum conservation constitutes eight constraints : q^a _ i=1^n _ i^_i^a = 0 . [ supercons ] like ordinary momentum conservation , it is satisfied by virtue of the periodicity of the dual coordinates , @xmath218 , @xmath219 . the mhv super - amplitude generalizing is _ n^,mhv = i = i . [ mhvsuper ] it transforms covariantly under dual superconformal transformations @xcite which extend so(4,2 ) to the superalgebra psu(2,2@xmath2204 ) . furthermore , the solutions to the bcfw super - recursion relation for amplitudes with more negative helicities ( nmhv , nnmhv , _ etc . _ ) are given by the product of this super - amplitude with collections of dual superconformal invariants , generically denoted by @xmath221 @xcite . superconformal and dual superconformal symmetry are separate symmetries , which partly overlap . they close into a very large symmetry group called the yangian @xcite , as discussed in this issue by bargheer , beisert and loebbert @xcite and by drummond @xcite . the generic generator of the yangian is not local in the sense of being a sum of single - particle terms ( like @xmath132 and @xmath153 ) ; rather it is a sum of multi - particle operators . ( for the dual superconformal generators , the nonlocality is mild ; only the sum of two - particle operators appears . ) there are certain `` anomalies '' in the action of the yangian on amplitudes . at tree level , the anomalies only act at the boundary of the @xmath14-particle phase space , where two particles become collinear , as in @xcite . at loop - level they act in the full phase - space , due to singularities in the loop integration @xcite . however , it is possible to deform either the symmetry algebra or the representation in a suitable way so as to preserve the invariance @xcite . it will be very interesting to see whether the constraints from the yangian are sufficient to determine multi - loop amplitudes directly , without having to pass through the computation of loop integrals . another very recent development , in the same vein of using symmetries or dynamical principles to bypass standard loop ( or wilson line ) integrals , concerns polygonal wilson loops . one considers an operator - product - like expansion for the wilson loop , associated with a limit in which multiple momenta become collinear @xcite . this expansion can be carried out at both weak and strong coupling . at weak coupling it has been used to compute the discontinuity of the integrated two - loop amplitude , directly in terms of one - loop quantities @xcite . there has been a remarkable array of breakthroughs in our understanding of scattering amplitudes over the past few years , as described in the review articles in this special issue of journal of physics a. this article has represented an overview of many of these developments , as well as providing some basic material as an introduction to the remaining contributions . however , it certainly did not do justice to all of the directions being pursued currently . one of the other remarkable features of this field has been the interplay and cross - fertilization between formal developments and phenomenology . the parke - taylor tree amplitudes were found in the course of trying to understand patterns arising in the structure of five- and six - gluon tree - level amplitudes . a similar analysis of the structure of the five - gluon amplitude at one loop in qcd led to the development of the unitarity method , which has been indispensable for computing high - loop - order amplitudes in maximally supersymmetric gauge theory and gravity , but which has also produced new nlo qcd results for colliders @xcite . a `` radiation zero '' found over thirty years ago in the electroweak process @xmath222 @xcite was studied for a general gauge theory soon thereafter @xcite ; the relations found there were recognized much later as the four - point versions of a more general color - kinematics duality @xcite . wilson loops , used to characterize the effects of soft gluons in qcd , also turned out to be in exact correspondence to full mhv amplitudes in planar @xmath0 super - yang - mills theory . this special issue should be of interest , not only to experienced practitioners in the field , but also to newcomers who want to get started . as the history of the field shows , new developments can and will come from totally unanticipated angles . many of the new ideas in the future may well come to physicists whose interest in the remarkable aspects of scattering amplitudes was sparked , at least in part , by reading the articles assembled in this special issue . * acknowledgments * i am grateful to radu roiban , mark spradlin and anastasia volovich for inviting me to contribute this introductory article , and i thank radu roiban also for valuable comments on the manuscript . this work was supported by the us department of energy under contract de ac0276sf00515 . the figures were generated using jaxodraw @xcite , based on axodraw @xcite . 10 j. m. maldacena , adv . phys . * 2 * , 231 ( 1998 ) [ int . j. theor . phys . * 38 * , 1113 ( 1999 ) ] [ hep - th/9711200 ] ; + s. s. gubser , i. r. klebanov and a. m. polyakov , phys . b * 428 * , 105 ( 1998 ) [ hep - th/9802109 ] ; + o. aharony , s. s. gubser , j. m. maldacena , h. ooguri and y. oz , phys . rept . * 323 * , 183 ( 2000 ) [ hep - th/9905111 ] . l. f. alday and j. m. maldacena , jhep * 0706 * , 064 ( 2007 ) [ 0705.0303 [ hep - th ] ] . s. j. parke and t. r. taylor , phys . lett . * 56 * , 2459 ( 1986 ) . v. p. nair , phys . b * 214 * , 215 ( 1988 ) . m. l. mangano and s. j. parke , phys . * 200 * , 301 ( 1991 ) [ hep - th/0509223 ] . m. t. grisaru , h. n. pendleton and p. van nieuwenhuizen , phys . d * 15 * , 996 ( 1977 ) . m. t. grisaru and h. n. pendleton , nucl . b * 124 * , 81 ( 1977 ) . s. j. parke and t. r. taylor , phys . b * 157 * , 81 ( 1985 ) [ erratum - ibid . * 174b * , 465 ( 1986 ) ] ; + z. kunszt , nucl . phys . b * 271 * , 333 ( 1986 ) . h. elvang , d. z. freedman and m. kiermaier , jhep * 1010 * , 103 ( 2010 ) [ 0911.3169 [ hep - th ] ] . h. elvang , d. z. freedman and m. kiermaier , to appear in j. phys . a [ 1012.3401 [ hep - th ] ] . e. witten , commun . phys . * 252 * , 189 ( 2004 ) [ hep - th/0312171 ] . r. penrose , j. math . * 8 * , 345 ( 1967 ) . f. cachazo , p. svrek and e. witten , jhep 0409:006 ( 2004 ) [ hep - th/0403047 ] . a. brandhuber , b. spence and g. travaglini , to appear in j. phys . a [ 1103.3477 [ hep - th ] ] . t. adamo , m. bullimore , l. mason and d. skinner , to appear in j. phys . a [ 1104.2890 [ hep - th ] ] . r. britto , f. cachazo , b. feng and e. witten , phys . lett . * 94 * , 181602 ( 2005 ) [ hep - th/0501052 ] . r. britto , f. cachazo and b. feng , nucl . b * 715 * , 499 ( 2005 ) [ hep - th/0412308 ] . n. arkani - hamed , f. cachazo and j. kaplan , jhep * 1009 * , 016 ( 2010 ) [ 0808.1446 [ hep - th ] ] . a. brandhuber , p. heslop and g. travaglini , phys . d * 78 * , 125005 ( 2008 ) [ 0807.4097 [ hep - th ] ] . j. m. drummond and j. m. henn , jhep * 0904 * , 018 ( 2009 ) [ 0808.2475 [ hep - th ] ] . j. drummond , to appear in j. phys . a. z. bern , l. j. dixon , d. c. dunbar and d. a. kosower , nucl . b * 425 * , 217 ( 1994 ) [ hep - ph/9403226 ] . z. bern , l. j. dixon , d. c. dunbar and d. a. kosower , nucl . b * 435 * , 59 ( 1995 ) [ hep - ph/9409265 ] . r. britto , to appear in j. phys . a [ 1012.4493 [ hep - th ] ] . z. bern , l. j. dixon and d. a. kosower , nucl . b * 513 * , 3 ( 1998 ) [ hep - ph/9708239 ] . r. britto , f. cachazo and b. feng , nucl . b * 725 * , 275 ( 2005 ) [ hep - th/0412103 ] . n. e. j. bjerrum - bohr and p. vanhove , jhep * 0804 * , 065 ( 2008 ) [ 0802.0868 [ hep - th ] ] ; jhep * 0810 * , 006 ( 2008 ) [ 0805.3682 [ hep - th ] ] ; fortsch . phys . * 56 * , 824 ( 2008 ) [ 0806.1726 [ hep - th ] ] . z. bern , l. j. dixon and d. a. kosower , jhep * 0001 * , 027 ( 2000 ) [ hep - ph/0001001 ] ; + z. bern , a. de freitas and l. j. dixon , jhep * 0203 * , 018 ( 2002 ) [ hep - ph/0201161 ] . z. bern and y .- t . huang , to appear in j. phys . a [ 1103.1869 [ hep - th ] ] . g. ossola , c. g. papadopoulos and r. pittau , nucl . b * 763 * , 147 ( 2007 ) [ hep - ph/0609007 ] ; jhep * 0803 * , 042 ( 2008 ) [ 0711.3596 [ hep - ph ] ] . r. k. ellis , w. t. giele and z. kunszt , jhep * 0803 * , 003 ( 2008 ) [ 0708.2398 [ hep - ph ] ] . c. f. berger _ _ , phys . d * 78 * , 036003 ( 2008 ) [ 0803.4180 [ hep - ph ] ] . w. siegel , phys . lett . b * 84 * , 193 ( 1979 ) ; + d. m. capper , d. r. t. jones and p. van nieuwenhuizen , nucl . b * 167 * , 479 ( 1980 ) ; + l. v. avdeev and a. a. vladimirov , nucl . b * 219 * , 262 ( 1983 ) . z. bern and d. a. kosower , nucl . b * 379 * , 451 ( 1992 ) ; + z. bern , a. de freitas , l. dixon and h. l. wong , phys . rev . d * 66 * , 085002 ( 2002 ) [ hep - ph/0202271 ] . r. akhoury , phys . d * 19 * , 1250 ( 1979 ) ; + a. h. mueller , phys . d * 20 * , 2037 ( 1979 ) ; + j. c. collins , phys . d * 22 * , 1478 ( 1980 ) ; in _ perturbative qcd _ , ed . a. h. mueller , advanced series on directions in high energy physics , vol . 5 ( world scientific , singapore , 1989 ) [ hep - ph/0312336 ] ; + a. sen , phys . rev . d * 24 * , 3281 ( 1981 ) ; phys . d * 28 * , 860 ( 1983 ) . l. magnea and g. sterman , phys . d * 42 * , 4222 ( 1990 ) . s. catani , phys . b * 427 * , 161 ( 1998 ) [ hep - ph/9802439 ] . g. sterman and m. e. tejeda - yeomans , phys . b * 552 * , 48 ( 2003 ) [ hep - ph/0210130 ] . g. p. korchemsky , mod . lett . a * 4 * , 1257 ( 1989 ) ; + g. p. korchemsky and g. marchesini , nucl . b * 406 * , 225 ( 1993 ) [ hep - ph/9210281 ] . n. beisert , b. eden and m. staudacher , j. stat . mech . * 0701 * , p01021 ( 2007 ) [ hep - th/0610251 ] . z. bern , l. j. dixon and v. a. smirnov , phys . d * 72 * , 085001 ( 2005 ) [ hep - th/0505205 ] . j. bartels , l. n. lipatov and a. sabio vera , phys . d * 80 * , 045002 ( 2009 ) [ 0802.2065 [ hep - th ] ] ; eur . j. c * 65 * , 587 ( 2010 ) [ 0807.0894 [ hep - th ] ] . j. bartels , l. n. lipatov and a. prygarin , to appear in j. phys . a [ 1104.0816 [ hep - th ] ] . l. f. alday , j. m. henn , j. plefka and t. schuster , jhep * 1001 * , 077 ( 2010 ) [ 0908.0684 [ hep - th ] ] . j. m. henn , to appear in j. phys . a [ 1103.1016 [ hep - th ] ] . r. k. ellis , w. t. giele , z. kunszt and k. melnikov , nucl . b * 822 * , 270 ( 2009 ) [ 0806.3467 [ hep - ph ] ] . s. lal and s. raju , phys . d * 81 * , 105002 ( 2010 ) [ 0910.0930 [ hep - th ] ] . a. v. kotikov , l. n. lipatov , a. i. onishchenko and v. n. velizhanin , phys . b * 595 * , 521 ( 2004 ) [ erratum - ibid . b * 632 * , 754 ( 2006 ) ] [ hep - th/0404092 ] . z. bern , m. czakon , l. j. dixon , d. a. kosower and v. a. smirnov , phys . d * 75 * , 085010 ( 2007 ) [ hep - th/0610248 ] . s. g. naculich , h. nastase and h. j. schnitzer , jhep * 0811 * , 018 ( 2008 ) [ 0809.0376 [ hep - th ] ] . n. arkani - hamed , j. l. bourjaily , f. cachazo , s. caron - huot and j. trnka , jhep * 1101 * , 041 ( 2011 ) [ 1008.2958 [ hep - th ] ] . r. h. boels , jhep * 1011 * , 113 ( 2010 ) [ 1008.3101 [ hep - th ] ] . j. m. drummond , j. henn , g. p. korchemsky and e. sokatchev , nucl . b * 826 * , 337 ( 2010 ) [ 0712.1223 [ hep - th ] ] . c. anastasiou , z. bern , l. j. dixon and d. a. kosower , phys . lett . * 91 * , 251602 ( 2003 ) [ hep - th/0309040 ] . j. m. drummond , j. henn , g. p. korchemsky and e. sokatchev , nucl . b * 815 * , 142 ( 2009 ) [ 0803.1466 [ hep - th ] ] ; + z. bern , l. j. dixon , d. a. kosower , r. roiban , m. spradlin , c. vergu and a. volovich , phys . d * 78 * , 045007 ( 2008 ) [ 0803.1465 [ hep - th ] ] . i. bena , j. polchinski and r. roiban , phys . d * 69 * , 046002 ( 2004 ) [ hep - th/0305116 ] . l. f. alday , d. gaiotto and j. maldacena , 0911.4708 [ hep - th ] ; + l. f. alday , j. maldacena , a. sever and p. vieira , j. phys . a * 43 * , 485401 ( 2010 ) [ 1002.2459 [ hep - th ] ] . j. m. drummond , g. p. korchemsky and e. sokatchev , nucl . b * 795 * , 385 ( 2008 ) [ 0707.0243 [ hep - th ] ] ; + a. brandhuber , p. heslop and g. travaglini , nucl . b * 794 * , 231 ( 2008 ) [ 0707.1153 [ hep - th ] ] ; + j. m. drummond , j. henn , g. p. korchemsky and e. sokatchev , nucl . b * 795 * , 52 ( 2008 ) [ 0709.2368 [ hep - th ] ] . l. f. alday , b. eden , g. p. korchemsky , j. maldacena and e. sokatchev , 1007.3243 [ hep - th ] ; + b. eden , g. p. korchemsky and e. sokatchev , 1007.3246 [ hep - th ] ; + l. j. mason and d. skinner , jhep * 1012 * , 018 ( 2010 ) [ 1009.2225 [ hep - th ] ] ; + s. caron - huot , 1010.1167 [ hep - th ] ; + a. v. belitsky , g. p. korchemsky and e. sokatchev , 1103.3008 [ hep - th ] . v. del duca , c. duhr and v. a. smirnov , jhep * 1003 * , 099 ( 2010 ) [ 0911.5332 [ hep - ph ] ] ; jhep * 1005 * , 084 ( 2010 ) [ 1003.1702 [ hep - th ] ] . a. b. goncharov , m. spradlin , c. vergu and a. volovich , phys . lett . * 105 * , 151605 ( 2010 ) [ 1006.5703 [ hep - th ] ] . v. del duca , c. duhr and v. a. smirnov , jhep * 1009 * , 015 ( 2010 ) [ 1006.4127 [ hep - th ] ] ; + p. heslop and v. v. khoze , jhep * 1011 * , 035 ( 2010 ) [ 1007.1805 [ hep - th ] ] . h. kawai , d. c. lewellen and s .- h . h. tye , nucl . b * 269 * , 1 ( 1986 ) . z. bern , j. j. m. carrasco and h. johansson , phys . d * 78 * , 085011 ( 2008 ) [ 0805.3993 [ hep - ph ] ] . z. bern , j. j. m. carrasco and h. johansson , phys . * 105 * , 061602 ( 2010 ) [ 1004.0476 [ hep - th ] ] . l. j. dixon , in _ qcd & beyond : proceedings of tasi 95 _ , ed . d. e. soper ( world scientific , 1996 ) [ hep - ph/9601359 ] . p. cvitanovi , _ group theory _ ( nordita , 1984 ) . f. a. berends and w. giele , nucl . b * 294 * , 700 ( 1987 ) ; + m. l. mangano , s. j. parke and z. xu , nucl . phys . b * 298 * , 653 ( 1988 ) ; + m. l. mangano , nucl . b * 309 * , 461 ( 1988 ) . g. t hooft , nucl . b * 72 * , 461 ( 1974 ) ; nucl . b * 75 * , 461 ( 1974 ) . z. bern and d. a. kosower , nucl . b * 362 * , 389 ( 1991 ) . z. bern , l. j. dixon and d. a. kosower , nucl . b * 437 * , 259 ( 1995 ) [ hep - ph/9409393 ] . v. del duca , a. frizzo and f. maltoni , nucl . b * 568 * , 211 ( 2000 ) [ hep - ph/9909464 ] . v. del duca , l. j. dixon and f. maltoni , nucl . b * 571 * , 51 ( 2000 ) [ hep - ph/9910563 ] . r. kleiss and h. kuijf , nucl . b * 312 * , 616 ( 1989 ) . n. e. j. bjerrum - bohr , p. h. damgaard and p. vanhove , phys . lett . * 103 * , 161602 ( 2009 ) [ 0907.1425 [ hep - th ] ] ; + s. stieberger , 0907.2211 [ hep - th ] . e. plahte , nuovo cim . a * 66 * , 713 ( 1970 ) . b. feng , r. huang and y. jia , phys . b * 695 * , 350 ( 2011 ) [ 1004.3417 [ hep - th ] ] ; + y. x. chen , y. j. du and b. feng , jhep * 1102 * , 112 ( 2011 ) [ 1101.0009 [ hep - th ] ] . s .- h . h. tye and y. zhang , jhep * 1006 * , 071 ( 2010 ) [ 1003.1732 [ hep - th ] ] . f. a. berends , r. kleiss , p. de causmaecker , r. gastmans and t. t. wu , phys . b * 103 * , 124 ( 1981 ) ; + p. de causmaecker , r. gastmans , w. troost and t. t. wu , nucl . b * 206 * , 53 ( 1982 ) ; + r. kleiss and w. j. stirling , nucl . b * 262 * , 235 ( 1985 ) ; + j. f. gunion and z. kunszt , phys . b * 161 * , 333 ( 1985 ) ; + z. xu , d. h. zhang and l. chang , nucl . b * 291 * , 392 ( 1987 ) ; + r. gastmans and t. t. wu , _ the ubiquitous photon : helicity method for qed and qcd _ ( clarendon press , 1990 ) . n. berkovits and j. maldacena , jhep * 0809 * , 062 ( 2008 ) [ 0807.3196 [ hep - th ] ] . n. beisert , r. ricci , a. a. tseytlin and m. wolf , phys . d * 78 * , 126004 ( 2008 ) [ 0807.3228 [ hep - th ] ] . j. m. drummond , j. m. henn and j. plefka , jhep * 0905 * , 046 ( 2009 ) [ 0902.2987 [ hep - th ] ] . t. bargheer , n. beisert and f. loebbert , to appear in j. phys . a [ 1104.0700 [ hep - th ] ] . t. bargheer , n. beisert , w. galleas , f. loebbert and t. mcloughlin , jhep * 0911 * , 056 ( 2009 ) [ 0905.3738 [ hep - th ] ] . n. beisert , j. henn , t. mcloughlin and j. plefka , jhep * 1004 * , 085 ( 2010 ) [ 1002.1733 [ hep - th ] ] . l. f. alday , d. gaiotto , j. maldacena , a. sever and p. vieira , jhep * 1104 * , 088 ( 2011 ) [ 1006.2788 [ hep - th ] ] . d. gaiotto , j. maldacena , a. sever and p. vieira , 1102.0062 [ hep - th ] . k. o. mikaelian , m. a. samuel and d. sahdev , phys . rev . lett . * 43 * , 746 ( 1979 ) . d. zhu , phys . d * 22 * , 2266 ( 1980 ) ; + c. j. goebel , f. halzen and j. p. leveille , phys . rev . d * 23 * , 2682 ( 1981 ) . d. binosi and l. theussl , comput . commun . * 161 * , 76 ( 2004 ) [ hep - ph/0309015 ] ; + d. binosi , j. collins , c. kaufhold and l. theussl , comput . commun . * 180 * , 1709 ( 2009 ) [ 0811.4113 [ hep - ph ] ] . j. a. m. vermaseren , comput . commun . * 83 * , 45 ( 1994 ) .

the problem of a single hole in the @xmath1-@xmath2 model interacting also with the lattice degrees of freedom has recently attracted a notable interest in connection with the physical properties of the high-@xmath4 cuprates . @xcite in particular , in parent and strongly underdoped compounds , angle resolved photoemission spectroscopy ( arpes ) reveals a low energy peak whose dispersion is well described by the @xmath1-@xmath2 model , while its anomalously large broadness has been ascribed to incoherent multi - phonon shake - off processes.@xcite a similar interplay between electron - electron and electron - phonon interactions should in principle be reflected in the optical conductivity spectra . as a matter of fact , the most remarkable features observed in the underdoped region are an ubiquitous mid - infrared ( mir ) peak at @xmath5 ev , and a weaker peak around @xmath6 ev , the latter being more strongly doping and temperature dependent.@xcite several interpretations have been proposed for the origin of these features , including midgap or impurity states , @xcite charge / spin stripes,@xcite polaronic excitations,@xcite and the interaction with the antiferromagnetic background.@xcite the first proposal of the @xmath1-@xmath2 model as a suitable basis to discuss the optical spectra of the cuprates was advanced by zhang and rice @xcite who observed that the @xmath7 behavior@xcite of the optical absorption above @xmath5 ev could be naturally associated to the incoherent motion in an antiferromagnetic background . successively , the optical conductivity of the @xmath1-@xmath2 model has been investigated in detail using several techniques , such as exact - diagonalization,@xcite analytical approximations@xcite and dynamical mean - field theory ( dmft).@xcite however , and in spite of the above discussed relevance of the electron - phonon coupling , the optical conductivity of the @xmath1-@xmath2 model _ in the presence _ of the lattice degrees of freedom has not been thoroughly investigated . numerical calculations based on exact - diagonalization of finite clusters were employed for instance in ref . to evaluate @xmath0 . alternatively , the optical conductivity was calculated analytically in ref . based on a non - crossing born approximation , which is however unable to describe the polaron formation . in this paper we present results for the optical conductivity of a single hole in the holstein-@xmath1-@xmath2 model obtained in the framework of dynamical mean - field theory . one - hole spectral properties at zero temperature were discussed in a previous publication where antiferromagnetic correlations were shown to enhance the effects of the electron - phonon coupling.@xcite a similar result was found also in the antiferromagnetic phase of the holstein - hubbard model using dmft techniques.@xcite a serious drawback of dmft , which is obtained as the exact solution of the lattice problem in the limit of infinite dimensions , is that the magnetic background is treated in a classical way . this , together with the fact that trugman loops are negligible in infinite dimensions , prevents the possibility to account for coherent hole - propagation , which is related to the spin - flip fluctuations . as a consequence , no drude peak can be observed in the optical conductivity . on the other hand , the incoherent contributions of @xmath0 are mainly dominated by _ properties , such as the local electron - phonon scattering and spin - string excitations within the magnetic polaron , which are well captured by this approach.@xcite bearing the above limitations in mind , the aim of the present work is thus to focus on the incoherent part of the optical conductivity of a single - hole and to investigate in detail its features in the different physical regimes of the holstein-@xmath1-@xmath2 model . the dependence of the optical spectra on the microscopic parameters is analyzed with special attention to the intermediate coupling region , where the interplay between magnetic and lattice degrees of freedom is strongest . we show that a crucial role is played by the formation of the lattice polaron , which drives a transfer of spectral weight towards higher frequencies , opening a pseudogap at low frequencies . conversely , starting from the polaronic phase , two different mechanisms can be clearly identified as being responsible for the disappearance of the pseudogap : @xmath8 ) reducing the effective exchange energy scale suppresses the positive feedback of magnetism on polaron formation and can lead to a _ closing _ of the pseudogap as the system crosses back to the non - polaronic regime ; @xmath9 ) increasing the temperature , which does not alter the lattice / magnetic interplay , leads to a _ filling _ of the pseudogap more similar to what is expected in purely polaronic models . in the immediate vicinity of the polaron crossover , the spectra are characterized by a coexistence of a magnon peak at low frequency and a broad polaronic band at higher frequency , which closely resembles the experimental situation observed in the cuprates . on theoretical grounds , the definition of the optical conductivity of a single hole is a delicate matter which needs particular care . we provide an analytical derivation which generalizes the results of refs . to the holstein-@xmath1-@xmath2 model . this approach permits us to identify the role of the different one - particle properties on the optical conductivity . comparison with numerical data is also discussed , showing a good agreement between our findings and exact diagonalization results . the paper is organized as follows . in sec . [ s - model ] we discuss the exact solution in infinite dimensions for the one - hole green s function of the holstein-@xmath1-@xmath2 model at finite temperature . results for the one - particle spectral features are discussed in sec . [ s - spectral ] . an analytical expression for the optical conductivity @xmath0 is derived in sec . [ s - opt ] where we also investigate the different polaronic features and their dependence on the microscopic parameters . in sec . [ s - sc ] we present a further simplified expression for @xmath0 valid in the lattice polaron regime . the main results are briefly summarized in sec . [ s - summary ] where also the consequences of including spin fluctuations ( here neglected ) is also discussed . finally , a detailed derivation of the analytical expression for the one - particle spectral function and the optical conductivity is reported in the appendices . in the following we consider the case of a single hole in an antiferromagnetic ( af ) background interacting with local holstein phonons . using the linear spin - wave approximation@xcite and neglecting terms that vanish in the limit of large coordination number @xmath10 , we can write the hamiltonian as:@xcite @xmath11 here @xmath12 is the creation operator for boson spin defects , @xmath13 is the single spinless hole operator and @xmath14 represents the density of spin defects which is finite at nonzero temperature . note that in the thermodynamical limit the presence of a single hole does not affect the magnetic state , which can be thus evaluated ( in the @xmath10 limit ) in the absence of spin dynamics . the density of spin defects can be obtained from the magnetization @xmath15 via the curie - weiss equation : @xmath16 which defines a nel temperature @xmath17 . concerning the electron - lattice interaction , we shall mainly focus on the adiabatic regime @xmath18 which is relevant to the experimental systems of interest . in this regime , a dimensionless electron - phonon coupling constant can be defined as @xmath19 , the polaron energy in units of the hopping integral . an exact solution for the thermodynamical and the one - particle spectral properties of eq . ( [ ham ] ) at @xmath20 was obtained in ref . in terms of a continued fraction . in order to derive the optical conductivity , the one - particle green s function must be generalized to finite temperature , which involves the following steps : @xmath8 ) one has to allow for thermally excited phonons ; @xmath9 ) the presence of thermally excited spin defects requires the introduction of a `` spin - resolved '' green s function , to distinguish hole excitations created on sites with / without spin defects ; @xmath21 ) finally , the reduced magnetization introduces an effective exchange coupling @xmath22 . the details of a formal derivation of the one - particle propagator at finite temperature are reported in appendix [ app - green - t ] ; we summarize here the main results . following ref . , we define @xmath23 as the green s function for one hole created on a site _ in the absence _ of spin defects . a careful analysis ( see appendix [ app - green - t ] ) shows that the green s function @xmath24 for a hole created on a site with a spin defect is simply @xmath25 . in addition , at finite temperature the green s function @xmath26 itself is defined as a thermal average over the phonons : @xmath27 where @xmath28 represents the propagation of one hole created on a site with @xmath29 excited phonons and @xmath30 is the single - site phonon partition function @xmath31 . following ref . , we can derive a self - consistent expression for @xmath32 in terms of a continued fraction . we can write : @xmath33 where @xmath34 carrier motion and exchange interactions are taken into account by the bath propagator @xmath35 where the `` hopping '' self - energy is defined as @xmath36 . \label{sigma_t}\ ] ] finally , the spin - defect averaged green s function , which is the physically probed quantity in photoemission experiments , is obtained as : @xmath37 note that the factors @xmath38 and @xmath39 in front of @xmath40 and @xmath41 account for the probability of a site to be , respectively , free or populated by a spin defect . note also that the green s function @xmath42 appearing in eqs . ( [ bathg],[sigma_t ] ) is a phonon averaged quantity , so that the solution of eqs . ( [ g - gnn])-([gtot ] ) involves the simultaneous self - consistency of all of the @xmath43 , which is much more computationally expensive than the solution at zero temperature . it is easy to check that in the absence of electron - phonon interaction eq . ( [ gtot ] ) recovers the thermal green s function for the pure @xmath1-@xmath2 model as defined in eq . ( 2.12 ) by stumpf and logan.@xcite on the other hand , it should be stressed that the present solution , although it formally recovers the results for the pure holstein model in the limit @xmath44 ( paramagnetic case ) [ eqs . ( 40)-(42 ) of ref . ] , is still described by a purely local green s function @xmath45 due to the assumption of a classical ( although disordered ) spin background . the main drawback of this assumption is thus that no coherent dispersive peak is obtained in this framework even in the @xmath46 limit of the @xmath1-@xmath2 holstein model , and , consequently , no drude peak appears in the optical conductivity . notwithstanding this limitation the present approach , as we will show below , is still quite able to reproduce in more than qualitative agreement the incoherent features of both the one - particle green s function and of the optical conductivity . before discussing the optical conductivity , let us briefly present our results for the one - particle spectral function in the holstein-@xmath1-@xmath2 model at finite temperature . previous approaches have focused separately on thermal effects either in the pure holstein or in the pure @xmath1-@xmath2 model . the temperature evolution of the hole spectral function @xmath47 for the @xmath1-@xmath2 model in the infinite dimensional limit has been analyzed in ref . . at @xmath20 , it consists of a series of @xmath48-function magnon peaks whose distribution reflects the strength of the magnetic polaron : for sufficiently large @xmath49 the profile is rapidly decaying with energy ( reminiscent of the string picture of small magnetic polarons ) whereas for small @xmath49 it acquires a more symmetric shape , reducing to a semi - circular function in the limit @xmath50 . the effect of a non - zero temperature within this context is to broaden each @xmath48-function with a bandwidth @xmath51 which is ruled by the thermal spin defect probability @xmath39 . we shall term this effect the _ intrinsic magnetic broadening_. such intrinsic magnetic broadening @xmath51 is however exponentially small for @xmath52 , and it becomes significant only close to @xmath53 . a semi - circular shape is recovered also in the paramagnetic phase @xmath54 , where @xmath55 . as the electron - lattice coupling is turned on , each magnon peak splits into several sub - peaks spaced by @xmath56 , reflecting the dressing of the hole by phononic excitations . in this context the strength of the electron - phonon coupling rules not only the number of phonon satellites but also its spectral weight profile . just as in the pure holstein model , while the number of phonon peaks is quite small in the weak coupling regime , in the polaronic state a large number of phonon satellites appear with a characteristic gaussian profile . the envelope of the phononic fine structure has a spread which is governed by the energy associated with the lattice fluctuations : it is given by @xmath57 in the quantum limit , and increases @xmath58 as @xmath59 . @xcite in fig . we show the temperature evolution of a typical polaronic spectral function , at @xmath60 , @xmath61 and @xmath62 . throughout the paper , when not specified , we shall take the hopping matrix element @xmath1 as the energy unit . the temperatures considered here are @xmath63 , corresponding to @xmath64 . at the lowest temperature ( @xmath65 ) a phononic fine structure can be clearly seen , superimposed on the magnon peaks . the width of each phononic peak is due to the intrinsic magnetic broadening @xmath51 described in ref . . it is exponentially small at this temperature , so that a small lorentzian broadening @xmath66 has been introduced for clarity . on the other hand , the spread of the multiphonon structure gives rise to an overall width to the magnon peaks that is in good agreement with the expected value @xmath67 . upon increasing the temperature to @xmath68 , two different effects are visible . first , the width @xmath51 of each of the fine peaks increases due to the intrinsic magnetic broadening , leading to a much smoother curve ( this effect actually overcomes the small lorentzian broadening @xmath69 introduced previously ) . also , the overall spread of the multiphonon profiles increases due to the thermal phonon fluctuations , as expected for @xmath70 finally , at @xmath71 , the system is so close to the paramagnetic phase that neither the phonon peaks nor the magnon structure can be resolved . at this point , it is useful to comment about the effects of coherent hole propagation , that are implicitly neglected in our approach . these would induce a finite dispersion of order @xmath72 to the lowest energy magnon peak.@xcite it is clear that such dispersion would be visible only at sufficiently low temperatures and at moderate electron - phonon coupling strengths , when the energy scale @xmath2 is smaller than both the intrinsic magnetic broadening and the gaussian phonon spread , whereas in the opposite case it will presumably be hidden below a featureless background . these considerations give further support to the present dmft approach in the polaronic and/or high temperature regime , where neglecting the coherent hole propagation would not affect significantly the spectral properties . as we shall see below , this is even more true for what regards the finite - frequency optical conductivity , where any dispersive peak would be convolved in any case with high - energy featureless structures . the evaluation of the optical conductivity for a single hole is a delicate matter , which is only partly simplified in the context of dmft due to the absence of vertex corrections.@xcite a controlled procedure is derived in terms of an expansion of the inverse fugacity at finite temperature , performing the limit @xmath73 to enforce the thermodynamically vanishing particle density . we can thus define the optical conductivity per hole , @xmath74 , which is a finite quantity and which presents the same features as in the dilute ( but finite ) hole density limit . applying this formalism one derives a similar expression as obtained in refs . , here adapted to take into account the electron - phonon interaction . leaving once more the technical details in appendix [ app - sigma ] , we report here the main results . the optical conductivity per hole is expressed as @xmath75 , \label{sigma } \end{aligned}\ ] ] where @xmath76 and @xmath77 the last line defines a `` weighted spectral function '' , @xmath78 , which represents thermally excited states and plays an important role in the determination of the optical conductivity . let us remark that , although eqs . ( [ sigma])-([wdos ] ) are formally analogous to those for the pure @xmath1-@xmath2 model , the electron - phonon interaction appears implicitly in them in the evaluation of the local green s function @xmath42 . note also that , in the paramagnetic limit @xmath79 , eqs . ( [ sigma])-([wdos ] ) do not recover the results of the holstein model,@xcite because eq . ( [ sigma ] ) involves the convolution of two _ local _ rather than * k*-_dependent _ propagators . as discussed above , this is due to the classical treatment of the spin degrees of freedom which does not allow for coherent transport , so that no drude peak is recovered in the present analysis . this , however , has only a minor influence on the finite - frequency optical conductivity , which is dominated by local incoherent excitations . in order to assess the validity of the present treatment , we compare in fig . [ f - feshke ] the optical conductivity of the holstein-@xmath1-@xmath2 model in infinite dimensions , as described by eqs . ( [ sigma])-([wdos ] ) , with numerical calculations using lanczos diagonalization for a single hole in the 2d holstein-@xmath1-@xmath2 model on a @xmath80 cluster.@xcite for technical reasons ( see discussion below ) , dmft data are averaged with a gaussian filter of amplitude @xmath81 such that phonon resonances are still well separated . microscopic values are @xmath82 , @xmath62 , @xmath83 , and @xmath20 ( for ref . ) and @xmath84 for the present results . these values correspond to a case where the lattice / magnetic polaron is formed and incoherent contributions to the optical conductivity are indeed dominant . the good agreement of the overall shape confirms the feasibility of our analysis to investigate the finite frequency optical conductivity . as discussed in the previous section , the one - hole spectral function consists of a set of narrow bands whose width @xmath51 is controlled by the intrinsic magnetic broadening . this intrinsic width vanishes at low temperature due to the absence of spin - wave dispersion , making the direct evaluation of eqs . ( [ sigma])-([wdos ] ) quite hard to perform . special care is thus needed in order to calculate numerically the optical conductivity . we shall be mainly interested in the features of the optical conductivity that are related to the electron - lattice coupling . we shall therefore retain the details of the spectra on the scale of the phonon frequency @xmath56 . from the practical point of view , the primary object of our analysis will be a gaussian average of the optical conductivity @xmath0 with a gaussian filter of amplitude @xmath81.@xcite note that , even though the gaussian average preserves the total spectral weight of the original set of data , the accuracy of the final result will be limited by the finite frequency sampling @xmath85 . in explicit terms , if the spacing @xmath85 is larger than the intrinsic peak - width @xmath51 , the data sampling will probe the spectral features in a random way , yielding a highly inaccurate result for both the shape and spectral weight of the optical conductivity . this is shown in the left panels of fig . [ filtro ] where we plot the dependence of the ( gaussian averaged ) optical conductivity on the sampling spacing @xmath85 for different temperatures . at high temperature @xmath86 the intrinsic magnetic broadening @xmath51 is large enough so that both the shape and the total spectral weight of the optical conductivity are well captured even with a relatively large mesh ( @xmath87 ) . at lower temperature @xmath68 however @xmath51 becomes so small that a much finer mesh ( @xmath88 ) is needed in order to get accurate results . at @xmath89 , finally , no convergence is achieved even for the finest sampling mesh considered in this paper , @xmath90 ( in fig . [ filtro ] , for graphical reasons , we plot curves only up to @xmath88 ) . clearly , since the intrinsic peak - width @xmath51 vanishes exponentially at low @xmath91 , the problem evidenced here can not be solved by merely reducing the sampling mesh . to overcome this difficulty we add to the system a small uncorrelated disorder with semielliptic distribution of amplitude @xmath92,@xcite which is able to yield a finite peak - width even in the zero temperature limit . we choose @xmath93 to assure a sufficiently dense mesh for an accurate sampling . we then scale @xmath94 keeping fixed the ratio @xmath95 in order to approach the correct physical limit in the absence of disorder . results obtained with a finite @xmath92 are shown in the right panels of fig . [ filtro ] . no appreciable difference is visible at high temperature @xmath86 where the thermally driven magnetic broadening @xmath51 is large enough to guarantee the convergence even in the absence of disorder . on the other hand our procedure provides a clear improvement already at @xmath68 where the convergence as function of the sampling spacing is more easily achieved in the presence of disorder ( note that the converged results coincide with the results for the most dense mesh in the _ absence _ of disorder , showing that no spurious structures appear due to our scaling procedure ) . finally , for @xmath89 the disorder scaling procedure is the only way to guarantee convergence of both the shape and the total spectral weight of the optical conductivity . using the above described procedure , we shall now concentrate on the evolution of the optical conductivity in the adiabatic regime , fixing the ratio @xmath61 . this value is qualitatively representative of the cuprates , where the half - bandwidth @xmath96 ev and typical optical phonon frequencies @xmath97 mev . this regime is also the most interesting one from the theoretical point of view , since in this case the interplay between lattice and spin degrees of freedom has its most dramatic effects . [ fig : scanl ] shows the evolution of the optical absorption at fixed @xmath62 and low temperature @xmath98 , upon varying the electron - lattice coupling . at @xmath99 , the result is very reminiscent of the spectra calculated for the pure @xmath1-@xmath2 model in ref . . it consists of a series of magnetic peaks , dominated by the sharp single - magnon peak located at @xmath100 and rapidly decaying at higher frequency . in this weak - coupling regime @xmath99 , the electron - phonon interaction simply gives rise to a multi - phonon fine - structure with a gaussian profile . each magnon peak acquires thus a phonon - driven width without modifying however the overall distribution of spectral weight . the main effect of increasing the electron - lattice coupling is a progressive shift of the spectral weight towards higher frequencies . this is an evidence of the formation of the lattice polaron , which occurs through a gradual crossover in the presence of a finite @xmath56 . increasing @xmath3 also modifies the shape of the low - energy absorption edge , converting the sharp magnon - peak at @xmath101 into a smoother gaussian lineshape , typical of polaronic absorption . in the strong coupling regime ( @xmath102 ) , characteristic of a small lattice polaron , the position of the maximum in the optical conductivity is expected to scale linearly as @xmath103 . this is in good qualitative agreement with our data reported in fig . [ fig : scanl ] where however the effects of a finite hopping integral @xmath1 ( ref . ) and of the @xmath7 behavior at high frequency ( see ref . ) result in a slight redshift of the maximum of the polaronic structure . the evolution of the optical conductivity with @xmath3 can be understood by analyzing the building blocks of eqs . ( [ rozz],[wdos ] ) . in fig . [ fig : dostot](a ) , we report both the spectral function @xmath104 and the weighted spectral function @xmath105 for two typical values @xmath99 and @xmath60 . in the first case ( dark red curve ) , there is no energy separation between the single - hole excitations in @xmath104 and the thermally excited states in @xmath105 . the low - frequency gap in the optical absorption seen in fig . [ fig : scanl ] arises due to the explicit shift of the spectral function by the quantity @xmath106 in eq . ( [ sigma ] ) , representing the energy cost to create one spin - defect as the hole hops in the af background . it is now interesting to compare these features with the results for @xmath60 [ light green curve in fig . [ fig : dostot](a ) ] . as we can see , the spectral functions @xmath107 for @xmath60 and @xmath99 are qualitatively similar , the only major difference being the increased number of phonon satellites involved in each magnetic peak , reflecting the increased number of phonons in the polaron cloud . in order to understand the modification of the optical conductivity we thus focus on the weighted spectral function @xmath105 . the latter undergoes a much more drastic change reflecting in an explicit way the signature of polaron formation . of particular relevance is the shift of @xmath105 to much more negative energies which characterizes the lattice trapping . note also that the magnetic peaks that are clearly visible at @xmath99 are completely washed out at @xmath60 , merging into a broad polaronic - like spectrum centered at higher `` binding '' energies . this change in the nature of the thermally excited states is at the origin of both the opening of a polaronic pseudogap and of the smoothening of the features observed in @xmath0 . a qualitatively similar evolution is observed upon varying @xmath2 , which gives a clear illustration of the positive interplay between magnetic and lattice polaron effects . this is shown in fig . [ fig : scanj ] where we report the optical conductivity for different values of @xmath2 at constant @xmath60 and at the same temperature @xmath108 as in fig . [ fig : scanl ] ( note that this corresponds to different @xmath109 as @xmath53 scales with @xmath2 ) . remarkably , even though @xmath3 is kept constant , reducing the magnetic exchange from the initial value @xmath110 leads to a gradual loss of the lattice polaronic features and to the undressing of the hole from multi - phonon excitations . this results in a shift of spectral weight from the polaronic peak to a magnon peak at lower frequency . the magnon peak is already visible as a shoulder in the data of fig . [ fig : scanj ] at @xmath111 and @xmath112 , and it emerges more clearly at lower values of @xmath2 . at @xmath113 , in particular , both the magnetic peak and the broad polaronic band are visible in the absorption spectrum . the featureless nature of the absorption curves at low @xmath49 can be ascribed to the increasing disorder of the af environment as @xmath2 diminishes and @xmath114 increases.@xcite once again , the comparison of @xmath105 and @xmath104 for @xmath62 [ light green curve in [ fig : dostot](a ) ] and for @xmath115 [ dark blue curve in [ fig : dostot](b ) ] allows to visualize in a simple way the loss of the polaronic features . even in this case the most relevant quantity is the weighted spectral function @xmath105 whose main excitations , for @xmath116 , are shifted to much less negative energies than for @xmath62 , closing the gap between @xmath105 and @xmath104 . note in addition that , although less evident than at @xmath62 , a magnetic peak in the spectral function @xmath104 is still visible even for @xmath115 . the convolution of @xmath104 with @xmath105 gives rise to the small magnetic peak at @xmath117 observed in the optical conductivity . in fig . [ fig : scant ] we report the evolution of the optical spectra versus temperature in the polaronic regime at @xmath60 , @xmath61 and @xmath118 . although fig . [ fig : scant ] can look at a first glance quite similar to fig . [ fig : scanl ] , no shift of the peak of @xmath0 is observed here . instead , there is a progressive _ filling _ of the low frequency gap with temperature that is quite similar to what is observed in the pure holstein model.@xcite this can be pointed out again by the comparative study of @xmath104 with @xmath105 . for @xmath71 [ light brown curve in [ fig : dostot](b ) ] the weighted spectral function @xmath78 still presents an extended peak at negative energy @xmath119 , signalizing that the lattice polaron is not completely destroyed . however , the broadening of @xmath78 is now significantly enhanced as @xmath120 . this feature , along with a similar broadening of the spectral function @xmath121 due to the approaching of the paramagnetic limit @xmath122 ( see sec . [ s - spectral ] ) , leads to a significant overlap of the two spectral functions and to a continuous filling of the gap in the optical conductivity as the temperature is increased . in the previous appendix we have derived an exact expression for the green s function of a single hole in the holstein-@xmath1-@xmath2 model in infinite dimensions . here we investigate within the same framework the optical conductivity per hole , @xmath0 , in the zero density limit . to this aim we provide an alternative derivation with respect to ref . . the main advantage of the present approach is to deal at the same level with both the spin and phonon degrees of freedom , allowing thus for an immediate generalization of the @xmath1-@xmath2 model to the holstein-@xmath1-@xmath2 model . as a result we obtain a final expression of the optical conductivity as a functional of the one - hole green s function which is formally similar to the one of ref . but where the local green s function takes into account the electron - phonon interaction . the formal way to derive the optical conductivity @xmath0 for a single charge , is to consider the limit @xmath203 , where @xmath204 , @xmath205 are quantities defined in the grand - canonical ensemble @xmath206 , and where @xmath207 it the total number of charges . the limit @xmath208 is enforced by expanding to lowest order in terms of the fugacity @xmath209 ( @xmath73 ) . in this limit only the subsector @xmath210 survives both in @xmath204 and in @xmath205 . let us first consider the hole density , which we can write as : @xmath211 where @xmath212 is a complete set of eigenstates with eigenvalues @xmath213 of one single hole ( subspace @xmath210 ) . here , since the state @xmath212 must contain one hole at site @xmath8 , the hole - number operator is simply defined as @xmath214 without any spin defect . inserting now into eq . ( [ ndef ] ) a complete set of eigenstates @xmath215 in the subspace @xmath216 ( no hole ) and a @xmath48-function , we obtain @xmath217 . \label{n1 } \end{aligned}\ ] ] note that only the @xmath215 which do not contain any spin defect at site @xmath8 would contribute to eq . ( [ n1 ] ) , otherwise , after the hole @xmath13 creation , we would end up with a state with one hole and one spin defect present , which is forbidden in the hilbert space . we have thus : @xmath218 , \label{n2 } \end{aligned}\ ] ] where @xmath219 is the statistical probability to have a site with no spin defect and where now only the @xmath215 with no spin defect at the site @xmath8 are selected . the square bracket in eq . ( [ n2 ] ) is just the local spectral function @xmath220 , namely @xmath221 [ remind that @xmath222 , so that we have @xmath223 let us turn now to the optical conductivity or , more precisely , to the current - current response function @xmath224 which is related to @xmath204 through the relation @xmath225 . according to the previous argumentation , in the limit @xmath226 we can limit our analysis to the @xmath210 subspace and write : @xmath227 where @xmath228 is the imaginary time in the matsubara space and @xmath229 is the current operator to be defined below . the current density operator can be written as @xmath233 , where we are summing explicitly on all possible directions ( this prescription compensates for the well - known vanishing of the current - current response in infinite dimensions@xcite ) . as discussed in refs . and as appearing in eq . ( [ ham ] ) , @xmath234 if there is a spin defect on the site @xmath180 ( @xmath235 ) , while @xmath236 if there is _ no _ spin defect on the site @xmath180 ( @xmath237 ) . due to the classical nature of the magnetic background , it is easy to realize that a retraceable path approximation is enforced also in the current - current response function just as in the one - particle green s function . we obtain thus : @xmath238 let us summarize the physical meaning of eq . ( [ jj5 ] ) . the eigenstate @xmath241 contains one hole at the site @xmath180 , and eq . ( [ jj5 ] ) describes the hopping of the hole to site @xmath8 . if no spin defect is initially present at @xmath8 ( @xmath145 ) , this process involves the creation of a spin defect at @xmath180 , and hence @xmath149 , @xmath242 . on the other hand , in the alternative case where a spin defect is initially present at @xmath8 ( @xmath146 ) , the hopping process destroys the spin defect at the site @xmath8 , so that @xmath243 and @xmath244 . let us now ask ourselves the following question : how much does the one - hole state @xmath241 differ from the `` free - like '' state @xmath246 in the absence of holes ? it is clear that in the @xmath182 case only the phonon - spin configuration on the site @xmath180 is affected by the presence of the hole , whereas all the other sites would be unaffected . in the presence of hole dynamics ( @xmath247 ) , however , all the other sites are in principle affected . we remind also that , because of the classical magnetic background , the hole dynamics obeys a retraceable path approximation just as in a bethe lattice , as depicted in fig . [ f - sketch ] . let us consider now a given specific link of two nearest neighbor sites , @xmath248 . in a true bethe lattice , the whole system can be divided in two subspaces , ( 1 ) and ( 2 ) , connected by the hopping term @xmath249 . this means that , to leading order , the presence of one hole on the site @xmath180 would not affect the subspace ( 1 ) . similar considerations hold true for a generic lattice under the retraceable path conditions : additional links between the two subspaces ( dashed lines in fig . [ f - sketch ] ) , will contribute only to @xmath250 and they can be neglected in the @xmath251 limit . for practical purposes we can thus split the total hamiltonian as @xmath252 , where @xmath253 accounts for the phonon - spin degrees of freedom of the subspace ( 1 ) ( not including @xmath180 ) , while @xmath254 contains the phonon - spin degrees of freedom of the subspace ( 2 ) ( not including @xmath8 ) . in a similar way the eigenstates @xmath241 , @xmath246 can be written ( in the leading order of a @xmath255 expansion ) as @xmath256 and @xmath257 . employing these results , and noting that @xmath258 , @xmath259 , @xmath260 , we have thus : where @xmath262 , @xmath263 are the partition functions of the corresponding subspaces and @xmath264 denotes a sum over the @xmath265 nearest neighbors of the site @xmath8 . performing the analytical continuation @xmath266 , and introducing once more appropriate @xmath48-functions , we end up thus with : @xmath267 } { 4 z \omega } \int d\omega \mbox{e}^{-\beta \omega } \left\ { \sum_{\langle j \rangle_i,\alpha_1,\gamma_1 } \frac{\mbox{e}^{-\beta e_{\gamma_1}}}{z^{\rm t}_1 } \left| \left\langle \alpha_1 \left| h_j^\dagger a_j \right| \gamma_1 \right\rangle \right|^2 \delta(\omega - e_{\alpha_1}+e_{\gamma_1 } ) \right\ } \nonumber\\ & & \times \left\ { \sum_{i,\alpha_2,\lambda_2 } \frac{\mbox{e}^{-\beta e_{\alpha_2}}}{z^{\rm t}_1 } \left|\left\langle \alpha_2 \left| h_i \right|\lambda_2\right\rangle\right|^2 \delta(\omega+\omega - e_{\lambda_2}+e_{\alpha_2 } ) \right\ } \nonumber\\ & = & \frac{t^2\pi n \mbox{e}^{\beta \mu } \left[1-\mbox{e}^{-\beta\omega}\right ] } { 4\omega } p_0 p_1\int d\omega \mbox{e}^{-\beta \omega } \bar{\rho}_1(\omega)\bar{\rho}_0(\omega+\omega ) , \label{pi0 } \end{aligned}\ ] ] the same derivation can be now employed for the case when a spin defect is present on the site @xmath8 . after few straightforward calculations , we obtain : @xmath270 } { 4\omega } p_0 p_1 \nonumber\\ & & \times \int d\omega \mbox{e}^{-\beta \omega } \bar{\rho}_0(\omega)\bar{\rho}_1(\omega+\omega ) . \label{pi1 } \end{aligned}\ ] ] summing the two contributions ( [ pi0 ] ) , ( [ pi1 ] ) , and reminding @xmath271 , @xmath272 , after a change of variable we obtain : @xmath273p_0 } { 4\omega } \int d\omega \mbox{e}^{-\beta \omega}\bar{\rho}(\omega ) \nonumber\\ & & \times % \int dz % \mbox{e}^{-\beta z}\bar{\rho}(z ) \left[p_1\bar{\rho}(\omega+\omega+\tilde{j}/2 ) + p_0 \bar{\rho}(\omega+\omega-\tilde{j}/2)\right ] , \label{pit } \end{aligned}\ ] ] where we made use also of the relation @xmath274 . finally , dividing eq . ( [ pit ] ) by ( [ nfin ] ) we obtain the dimensionless quantity eq . ( [ sigma ] ) . note that , dividing eq . ( [ pit ] ) by ( [ nfin ] ) , the common factors @xmath275 cancel out , so that the limit @xmath276 is well defined . shen , f. ronning , d.h . lee , n.j.c . ingle , w. meevasana , f. baumberger , a. damascelli , n.p . armitage , l.l . miller , y. kohsaka , m. azuma , m. takano , h. takagi , and z .- x . shen , phys . lett . * 93 * , 267002 ( 2004 ) . for a review of early works see e. dagotto , rev . phys . * 66 * , 763 ( 1994 ) . see also later papers as : j. jakli and p. prelovek , phys . b * 52 * , 6903 ( 1995 ) ; m.m . zemlji and p. prelovek , phys . b * 72 * , 075108 ( 20055 ) . in the previous section we have discussed the technical difficulties involved in the calculation of the optical conductivity . even adopting the proposed scaling procedure ( cf . [ tech ] ) , the computational cost can still be quite demanding , especially in the strong coupling lattice polaron regime where the number of phonons to be taken into account scales as @xmath123 . in this section we present a simple analytical formulation of the optical conductivity which is valid precisely in the adiabatic ( @xmath124 ) small polaron regime and which involves almost no numerical effort . the basis of this simplified formulation is the polaronic nature of the weighted spectral function @xmath105 pointed out in fig . [ fig : dostot ] . we have seen indeed that at moderate values of the electron - lattice coupling , just above the polaron crossover , the function @xmath105 becomes essentially independent of the exchange term @xmath2 . a closer look shows that in the lattice polaron regime the spectral function @xmath105 can be well approximated with the result of the atomic limit @xmath125,@xcite @xmath126,\ ] ] where @xmath127 . this function for the case @xmath60 , @xmath62 , @xmath61 , @xmath89 is shown in fig . [ fig : dostot]a compared with the full numerical solution for the same parameters . the good agreement can be ascribed to the atomic nature of the thermally induced excitations , which do not hybridize with the hopping continuum described by the self - energy eq . ( [ sigma_t ] ) ( an equivalent result was already pointed out in ref . ) . ( [ rhowsc ] ) provides thus a simple analytical expression for @xmath78 which can be employed in eq . ( [ sigma ] ) for the evaluation of the optical conductivity . a simple analytical approximation can also be derived for the one - hole spectral function @xmath107 involved in eq . ( [ sigma ] ) . at this level , we are mainly interested in the overall shape of the optical conductivity @xmath0 , disregarding the multi - phonon fine structure that will be anyway washed out at large @xmath3 as shown in fig . [ fig : scanl ] . in this perspective we can consider formally the limit @xmath128 and use the approach devised in refs . as well as in ref . to treat the adiabatic limit . in particular , the sum over the discrete energy levels in eq . ( [ g - gnn ] ) can be replaced by an integration over gaussianly distributed local random energies @xmath129 which account for the thermal / quantum fluctuations of the phonon field . the local propagator thus becomes@xcite @xmath130 where @xmath131 is a gaussian function with the same variance as in eq . ( [ rhowsc ] ) . self - consistency is achieved by employing eqs . ( [ bathg])-([sigma_t ] ) . the comparison between the approximate formula for @xmath107 valid in the polaronic regime and the full numerical solution is also shown in fig . [ fig : dostot]a for @xmath60 , @xmath62 , @xmath61 , @xmath89 . note that , although the bath propagator @xmath132 does not depend explicitly on the local random energies @xmath129 , it still depends parametrically on @xmath3 and @xmath2 through the self - consistency conditions eqs . ( [ bathg])-([sigma_t ] ) , so that it still retains the relevant lattice and magnetic spectral structures . once the approximate analytical expressions for the spectral functions @xmath104 and @xmath105 are provided , we can now simply evaluate the optical conductivity by using eq . ( [ sigma ] ) . let us stress that the numerical solution of the approximate expressions for @xmath104 and @xmath105 , defined in eqs . ( [ rhowsc])-([eq : avprop ] ) requires a remarkably smaller computational cost than the full numerical solution , especially in the lattice polaronic regime where eqs . ( [ rhowsc])-([eq : avprop ] ) are valid and where the computational cost of the full numerical solution is highest . the comparison of the approximate formula with the full numerical solution as function of the electron - phonon coupling constant @xmath3 is illustrated in fig . [ fig : sc ] . the agreement is excellent in the small polaron regime @xmath133 and it is still quite satisfactory even for moderate electron - phonon coupling @xmath134 . in this work , we have provided an analytical treatment for the optical conductivity @xmath0 of one hole in the holstein-@xmath1-@xmath2 model at finite temperature , in the limit of infinite dimensions . in this context a dynamical mean - field solution can be derived where the local self - consistent problem is solved exactly . due to intrinsic limitations enforced in infinite dimensions , we are not able to account for the coherent propagation of a single hole in the antiferromagnetic background . nevertheless , we have shown how the incoherent optical processes , related to local excitations ( einstein phonons and local spin - flips ) are well described in our approach , as confirmed by the comparison with numerical results obtained by lanczos diagonalization . our main aim has been to investigate the incoherent features of @xmath0 in an intermediate coupling region where the positive interplay between the magnetic and lattice degrees of freedom is more relevant , sustaining the formation of a spin - lattice polaron . in this context we have studied the evolution of polaronic features in the optical conductivity as function of the different microscopic parameters , as the electron - phonon coupling @xmath3 , the temperature @xmath91 and the effective exchange energy @xmath135 . we remind that @xmath135 does not represent the bare exchange energy but rather a mean - field - like weiss exchange coupling which depends on the local magnetization @xmath136 . in the cuprates , for example , this parameter can be tuned by varying the hole doping starting from the parent af phase . we have shown that the role of the electron - phonon coupling is twofold . on one hand it rules the formation of the lattice polaron , changing the incoherent part of @xmath0 from a typical spin - polaron spectrum , characterized by magnetic peaks and by an optical gap at @xmath137 , to a broad lattice - polaron - like shape located at higher frequency , from which the magnetic peaks are essentially washed out . on the other hand it also tunes the amount of quantum lattice fluctuations , reflected in the emergence of multi - phonon satellite peaks . this fine structure survives also when the lattice polaron is destroyed at small @xmath3 and it gives rise to an effective broadening of the magnetic peaks which can be much larger than the intrinsic broadening driven by the thermal magnetic fluctuations . the present approach allows us to distinguish between two different mechanisms leading to a suppression of the polaronic pseudogap : @xmath8 ) at intermediate values of the electron - phonon coupling , the reduction of the effective exchange energy @xmath135 leads to a shift of spectral weight from high energy lattice polaronic features to low energy magnetic excitations . this results in a _ closing _ of the pseudogap as the hole undresses from its lattice polaron cloud ; @xmath9 ) conversely , increasing the temperature within the polaronic regime , gives rise to a _ filling _ of the pseudogap . both mechanisms are actually observed in the optical spectra of the underdoped cuprates.@xcite as a final remark , we briefly comment on the robustness of our results in physical systems where quantum spin fluctuations are present , allowing for coherent motion of the holes . as discussed above , one of the main effects is the emergence of a dispersive pole with small spectral weight in the one - particle spectral function . in the optical spectra , this gives rise to a drude - like low - frequency response , but it does not affect the high - frequency incoherent part ( moreover the coherent spectral weight is strongly reduced when the electron - lattice coupling is turned on ) . regarding the high - frequency part , the intrinsic dispersion of the spin fluctuations is itself expected to smear the magnetic peaks . this would affect our results only in the weak electron - phonon coupling regime , where no lattice polaron is formed . in this case an intrinsic broadening of the magnetic peaks in the optical spectra due to the spin - fluctuation dispersion should be considered for a quantitative analysis . on the other hand , the effects of the dispersion of spin - fluctuations are expected to be barely visible in the lattice polaron regime , where the smearing due to the multi - phonon satellite structure around each magnetic peak dominates . in this appendix we provide a detailed derivation of the green s function of one hole in the holstein-@xmath1-@xmath2 model at finite temperature in the infinite dimensional limit . a formal derivation for the pure @xmath1-@xmath2 model was discussed in ref . , while the derivation for the full holstein-@xmath1-@xmath2 model at @xmath20 was provided in ref . . on this ground , here we limit ourselves to the derivation of the finite temperature self - consistent equations in terms of continued fractions . let us start by writing the hamiltonian in eq . ( [ ham ] ) as @xmath138 , where @xmath139 represents the non - local hopping term while @xmath140 contains all the other , purely local , contributions . we also define the green s function as : where @xmath142 denotes the state with @xmath143 phonons and @xmath144 spin defects ( @xmath145 : no spin defect , @xmath146 : spin defect ) on the site @xmath8 and with a generic set @xmath147 of phonons and @xmath148 of spin defects on all the other sites . with these notations @xmath149 when @xmath145 and @xmath150 when @xmath146 . in addition , @xmath151 is the total partition function of the system in the absence of holes , @xmath152 being the number of sites ; in this case @xmath139 does not contribute and @xmath151 factorizes into a phonon and a spin part as @xmath153 . each of them can be in addition factorized with respect to the site index , e.g. @xmath154 . similar considerations hold true for the exponential terms @xmath155 which apply on the states with no holes on the right side of eq . ( [ gdef ] ) . reminding @xmath156 , we can write , after few straightforward steps , in the fourier space : @xmath157 } } { z^{\rm ph}_i z^{\rm spin}_i } \sum_{\{n\}_{j \neq i } , \{s\}_{j \neq i } } \frac{\mbox{e}^{-\beta \sum_{j \neq i}[n_j \omega_0 + s_j \tilde{j}/2 ] } } { z^{\rm ph}_{j \neq i } z^{\rm spin}_{j \neq i } } \nonumber\\ & & \times \left\langle \{s\}_{j \neq i},\{n\}_{j \neq i};s_i , n_i \right| c_i^\dagger \frac{1}{\omega+n_i\omega_0+s_i\tilde{j}/2-h } c_i \left|n_i , s_i;\{n\}_{j \neq i},\{s\}_{j \neq i } \right\rangle . \label{gdefw } \end{aligned}\ ] ] this can be rewritten as @xmath158 where @xmath159 } } { z^{\rm ph}_{j \neq i } z^{\rm spin}_{j \neq i } } \nonumber\\ & & \times \left\langle \{s\}_{j \neq i},\{n\}_{j \neq i};s_i , n_i \right| c_i^\dagger \frac{1}{\omega+s_i\tilde{j}/2-h } c_i \left|n_i , s_i;\{n\}_{j \neq i},\{s\}_{j \neq i } \right\rangle . \label{gnn - def } \end{aligned}\ ] ] similarly , the factor @xmath160 defines the local population of spin defects which can be evaluated within the mean - field theory enforced by the infinite dimensional limit , namely @xmath161 , @xmath162 . we have thus : @xmath163 with @xmath164 } } { z^{\rm ph}_{j \neq i } z^{\rm spin}_{j \neq i } } \left\langle \{s\}_{j \neq i},\{n\}_{j \neq i};s_i , n_i \right| c_i^\dagger \frac{1}{\omega+s_i\tilde{j}/2-h } c_i \left|n_i , s_i;\{n\}_{j \neq i},\{s\}_{j \neq i } \right\rangle . \end{aligned}\ ] ] finally , using the definition of @xmath165 , one can see that @xmath166 , and we can write @xmath167 where @xmath168 } } { z^{\rm ph}_{j \neq i } z^{\rm spin}_{j \neq i } } \left\langle \{s\}_{j \neq i},\{n\}_{j \neq i};n_i \right| \frac{1}{\omega - h } \left|n_i;\{n\}_{j \neq i},\{s\}_{j \neq i } \right\rangle . \label{bargnn - def } \end{aligned}\ ] ] we can write eq . ( [ bargnn - def ] ) using the short - hand notation : @xmath172 where @xmath173 , and where the operator @xmath174 } } { z^{\rm ph}_{j \neq i } z^{\rm spin}_{j \neq i } } \nonumber\\ & & \hspace{-2cm}\times \left\langle \{s\}_{j \neq i},\{n\}_{j \neq i}\right| \ldots \left|\{n\}_{j \neq i},\{s\}_{j \neq i } \right\rangle , \end{aligned}\ ] ] denotes the average over the spin and phonon configurations on all the sites but @xmath8 . the operator @xmath175 represents a direct generalization of the quantity @xmath176 introduced by stumpf and logan in ref . to include the phonon degrees of freedom . it is also convenient to note that ( for @xmath177 ) : @xmath178 } } { z^{\rm ph}_j z^{\rm spin}_j } \nonumber\\ & & \hspace{-2cm}\times \left\langle s_j , n_j , n_i\right| \ldots \left|n_i , n_j , s_j\right\rangle , \label{pop } \end{aligned}\ ] ] where @xmath179 is defined in similar way as @xmath175 as the average over the spin and phonon configurations on all the sites except @xmath8 and @xmath180 . having introduced the necessary definitions , from now on we can follow the derivation in ref . properly adapted to the finite temperature case . in particular we can introduce the local green s function @xmath181 defined as the atomic @xmath182 limit of eq . ( [ bargnn ] ) . note that in the atomic limit @xmath183 so that @xmath184 we can also generalize eq . ( [ bargnn ] ) for off - diagonal local phonon matrix elements : @xmath185 whose analytical expression will be provided in eq . ( [ glocnm ] ) . we can now employ the standard relation @xmath186 which , on a classical spin background , gives rise to the retraceable path approximation . we have thus @xmath187 where @xmath188 represents the dynamics of the hole after hopping on the neighboring site @xmath180 . since , in the leading term of a @xmath189 expansion , the hopping process and the further dynamics of the hole do not involve the phonon degrees of freedom on site @xmath8 , such evolution occurs _ in the presence _ of @xmath190 phonons on the site @xmath8 , which are reflected in an shift of the frequency argument , @xmath191 . note that @xmath192 still contains the full thermal average on the phonons at site @xmath180 , so that it is related to the thermally averaged green s function defined in eq . ( [ g - gnn - app ] ) . in addition , @xmath192 will depend also on the initial spin configuration @xmath193 at site @xmath180 : hopping to a site @xmath180 free of spin defects will _ create _ a spin defect on the site @xmath8 , while hopping to a site @xmath8 with a spin defect will restore the initial magnetic background at site @xmath8 _ destroying _ a spin defect at @xmath8 . using these considerations we have thus : @xmath194 \nonumber\\ & & \hspace{-14 mm } = \frac{t^2}{4 } \left [ ( 1-x)\bar{g}_j(\omega-\tilde{j}/2 ) + x \bar{g}_j(\omega+\tilde{j}/2 ) \right ] , \label{sigma_t - app } \end{aligned}\ ] ] and we can write eq . ( [ recursion ] ) in the compact form : @xmath195 where we have dropped the unnecessary site indices and we have added the index @xmath196 to denote a self - energy term arising from the hopping processes . we can indeed employ now the usual procedure to obtain the electron - phonon self - energy in terms of a continued fraction . in particular , in addition to the diagonal elements of eq . ( [ glocnn ] ) , we specify also the non - diagonal ones which read : @xmath197^{nm}= \left[\omega - n\omega_0\right]\delta_{n , m}+g x^{n , m}. \label{glocnm } \end{aligned}\ ] ]

a fundamental tool to study the evolution of active galactic nuclei ( agns ) is their luminosity function ( lf ) and its evolution with redshift . this enables us to probe the accretion history of supermassive black holes ( smbhs ) which , in turn , set constraints on galaxy evolution , as smbhs are believed to be ubiquitous and tightly linked with all stellar bulges and spheroids today ( e.g. ferrarese & ford 2005 and references therein ) . moreover constraining the agn lf is essential to understand the agn contribution to the total emissivity background in different bands , their environmental impact , their clustering properties and the role agns had , if any , in the cosmological re - ionization of hydrogen and helium . x - ray observations carried out with _ chandra _ , _ asca _ , _ heao-1 _ are statistically complete in x - ray to a level of @xmath11 up to @xmath12 ( e.g. la franca _ et al . these observations have assessed that agns statistically follow a luminosity - dependent density evolution ( e.g. ueda et al . 2003 ) , in which lower luminous agns peak at @xmath13 , while higher luminous ones , usually optically bright , peak at higher redshifts , @xmath14 ( e.g. osmer 2004 ) . however how far in time can we track agn evolution ? the picture is still not yet complete at higher redshifts , where x - ray measurements of the agn lf are lacking . on the optical side the sdss team ( richard _ et al . _ 2006 ) , and earlier works by kennefick _ ( 1995 ) , combo-17 ( wolf _ et al . _ 2003 ) have probed the bright tail of the optical agn lf up to @xmath15 down to magnitudes of @xmath16 showing that a simple power - law can easily fit the data . the goods team ( cristiani _ et al . _ 2004 ; fontanot _ et al . _ 2006a ) has been able to set constraints on the shape of the optical lf down to @xmath17 and up to @xmath18 , showing evidence for a flattening at the faint - end . ( 2004 ; f04 from here on ) have instead probed the very luminous tail of the @xmath0 optical lf up to @xmath19 . here we use existing observations of optically selected agn and _ chandra _ deep fields to determine constraints on the redshift 6 shape of both the optical and x - ray lf . we will in particular show that x - ray observations allow us to probe the @xmath0 lf to @xmath20 magnitudes fainter than the deepest optical observations . in section [ sec|method ] we describe our method , present the constraints on the faint end lf in section [ sec|results ] , their implications in sections [ sec|implic ] and [ sec|reionization ] , and conclude in section [ concl ] . we use concordance cosmological parameters @xmath21 , @xmath22 , @xmath23 , @xmath24 km s@xmath25 mpc@xmath25 , @xmath26 , and @xmath27 ( spergel _ et al . in the following we will always refer to the total population of active galactic nuclei as agns . we will call `` qsos '' ( or `` quasars '' ) the type i , optical agns , and type ii agns those which have a column density above @xmath28 . the latter kind of sources are often very faint or even missed in optical surveys , but still detected in x - rays . several deep x - ray and optical surveys have now probed several portions of the agn luminosity function at high redshifts , however often finding a scarce number of faint sources . in this paper we will show that the paucity of sources at these faint fluxes gives the most up - to - date stringent constraints on the knee and the faint - end slope of the qso lf and give hints to the overall agn distribution at @xmath0 . similar studies , though with looser upper limits , have been carried out by sharp _ ( 2004 ) , willott _ ( 2005 ) , wyithe ( 2004 ) and dijkstra & wyithe ( 2006 ) . to convert x - ray to optical rest - frame luminosities we will use the equation ( see richards _ _ 2005 ) @xmath29 between monochromatic luminosities and a constant photon index @xmath30 to compute the broad - band x - ray luminosities . many authors ( e.g. vignali _ et al . _ 2003 ) have found that the spectral index @xmath31 depends significantly on luminosity . in particular , the latest results by steffen _ ( 2006 ) , who have collected a large number of optically and x - ray selected agns at several redshifts , claim a significant correlation of the spectral index with luminosity and a marginal one ( 1.1@xmath32 ) with reshift , i.e. @xmath33 where @xmath34 is the monochromatic luminosity at @xmath35 , in units of @xmath36 . in the following we will use results from several optical / nir surveys carried out on different areas . to compare the luminosities probed by different observations , we will first need to convert the broadband z@xmath37 , where most of the observations are carried out , to the apparent magnitude at wavelength @xmath38 , i.e. ab@xmath39 , and finally convert the latter to the rest - frame luminosities m@xmath40 . we convert z@xmath37 to ab@xmath39 magnitude following the spectrum described in cristiani & vio ( 1990 ) and updated in cristiani _ et al . _ ( 2004 ) . to convert then to absolute magnitudes m@xmath40 , we note that in the ab system @xmath41 , where the monochromatic flux @xmath42 is in units of @xmath43 , and the absolute magnitude is m@xmath44 , where l@xmath45 is the luminosity at rest frame frequency @xmath46 , in units of @xmath47 . using @xmath48 leads to ( see also stern _ 2000b ) @xmath49 and ( fan _ et al . _ 2001 ) @xmath50 where @xmath51 is the absolute b magnitude calibrated for a @xmath52 ( where we set @xmath53 ) spectrum and the factor 0.12 comes from the zero point difference between the ab and vega - based system ( fan _ et al . _ 2001 ) . to constrain the shape of the @xmath0 agn lf , following willott _ ( 2005 ) , we first define a double power - law model lf of the type @xmath54 where @xmath55 , with @xmath56 , is the average redshift of the f04 lf . with such a parametrization we assume that the high - redshift agn lf evolves with a pure density evolution ( pde ) beyond redshift of 5 at the same rate as the lf in fan _ ( 2001 ) . following the observations by f04 , @xmath57 has been calibrated to match the cumulative number density of @xmath58 for sources with @xmath59 at @xmath60 ( these are the limit and number density for @xmath61 , @xmath62 , @xmath63 and have been converted to our cosmology ) . the bright - end slope and the break magnitude have been fixed to @xmath64 and @xmath65 respectively , as this provides a very good match to the cool et al . 2006 observations ( see below ) . the faint - end slope @xmath66 has been left variable and constrained by matching the observed number density of sources in each survey . at @xmath18 the lf in equation ( [ eq|modellf ] ) is in agreement with the cumulative number density determined by fan _ ( 2001 ) . thus given a model lf we integrate it within the effective volume of the survey using the following equation ( see e.g. fan _ et al . _ 2001 ) @xmath67 and tune the faint - end slope @xmath66 in equation ( [ eq|modellf ] ) to match the observed number of sources @xmath68 in the survey . @xmath69 refers to the redshift dependent absolute luminosity , corresponding to the minimum optical apparent magnitude ( or the minimum flux ) considered . we will in general assume a constant selection function p(l , z , sed ) for each survey , as indicated by the authors or derived from their papers . in the following we will always use the form in equation ( [ eq|modellf ] ) for @xmath70 , unless in some specified cases , in which we will adopt the fan et al . ( 2001 ) lf . the latter is a single power - law function valid in the range @xmath71 , in which we let the slope @xmath72 to be variable , while we keep the redshift evolution and normalization at @xmath73 as in fan _ ( 2001 ; both @xmath73 and normalization have been converted to our cosmology , which is required at each redshift bin during the evolution ) . a significant improvement with respect to previous works has been to fix the knee of the lf in equation ( [ eq|modellf ] ) to the value recently inferred by cool _ these authors discovered three objects down to z@xmath37=21.93 , ( i.e @xmath74 at z@xmath75 using equation ( [ eq|mzmbm1450 ] ) ) in the agn and galaxy evolution survey ( ages ) spectroscopic observations of the noao deep wide - field survey ( ndwfs ) . they covered 7.7 deg@xmath76 of the ndwfs field with an average completeness level of 54% down to @xmath77 mag . these authors claim their observations agree with the f04 lf with a slope of @xmath78 and a redshift evolution as in fan et al . ( 2001 ) and in equation ( [ eq|modellf ] ) . using equation ( [ eq|integrallf ] ) , we find @xmath79 sources with z@xmath80 at @xmath81 , in very good agreement with observations and with the authors . the integration was pursued up to @xmath82 , as beyond this limit the optical selection of quasars becomes impossible because the ly@xmath83 line redshifs out of the z@xmath37-band ( see cool _ et al . _ 2006 ) . following a similar procedure we find significant evidence for a flattening of the qso lf below @xmath74 . recently , mahabal _ ( 2005 , m05 hereafter ) in the field around the @xmath84 quasar sdss j1148 + 5251 discovered a quasar at redshift @xmath85 with luminosity of @xmath86 , i.e. z@xmath87 . this quasar is the faintest known optically selected quasar at @xmath88 , allowing determination of the faint end of the quasar lf to fainter luminosities . the authors claim the source , identified in a @xmath89 deg@xmath76 survey , has a surface density consistent with the wolf _ ( 2003 ) lf extrapolated through a pde within their effective volume , which follows the @xmath90-band drop - out criterion of stern _ et al . _ ( 2000 ) , i.e. @xmath91 . we have checked that adopting the latter determination for the effective volume , we recover the results of the authors only if an average @xmath92 selection detection probability is used . however using the wolf _ ( 2003 ) lf , which is defined only for @xmath93 , needs an extrapolation to higher redshifts . moreover the functional form in equation ( [ eq|modellf ] ) has been calibrated on the data by cool _ ( 2006 ) at @xmath94 . we therefore replace equation ( [ eq|modellf ] ) with the fan _ ( 2001 ) qso lf , more appropriate for the redshift bin probed by the @xmath90-band drop - out technique . keeping the same value for the selection function we find that the detection of one object in the effective volume probed by m05 , is consistent with a number density at @xmath86 of @xmath95\sim -6.70 $ ] at the 99% confidence level ( cl ; with @xmath96 ) , down to @xmath95\sim -6.95 $ ] at the 90% cl ( with @xmath97 ; see table [ table|survey ] ; we plot the m05 data in figure [ fig|lfz6 ] shifted by -0.15 magnitudes for avoiding overcrowding with willott et al.s data ) . their data , shown as solid triangles in figure ( [ fig|lfz6 ] ) , are consistent with extrapolating equation ( [ eq|modellf ] ) to fainter optical magnitudes using @xmath98 and @xmath99 at the 99% and 90% cl respectively ( see table [ table|survey ] ) . note that these are conservative estimates . extending the computation of the lf down to the limiting magnitude of the m05 survey ( z@xmath100 , @xmath101 ) or using a higher completeness function would yield even lower limits on @xmath66 . ( 2005 ) undertook a multicolor quasar optical imaging with the canada - france - hawaii telescope . they found a lack of quasars down to a limiting magnitude of z@xmath10223.35 ( i.e. @xmath103 at @xmath104 ; see figure [ fig|lfz6 ] ) . the few detected quasar candidates had been subsequently rejected as being identified with low - mass stars in the near - infrared follow - up imaging . using equations ( [ eq|modellf ] ) and ( [ eq|integrallf ] ) , we find that the willott _ et al . _ ( 2005 ) results are consistent with a faint - end slope of @xmath105 at 99% cl , down to @xmath106 at the 90% cl . we use the selection function computed from simulated colors for high - redshift qsos in the redshift range @xmath107 ( see figure 2 in willott _ et al . this is the most conservative selection function from willott _ ( 2005 ) , determined for very extreme red color cuts . using a less conservative estimate ( solid line in their figure 2 ) our results drop to @xmath108 at 99% cl , down to @xmath2 at the 90% cl . the above findings are shown in figure [ fig|lfz6 ] . the gray area shows the data with its @xmath109-error bars by f04 for the @xmath110 lf , while the solid dots show the extrapolation of the f04 lf to faint luminosities assuming a single slope of @xmath78 . the presence of a significant downturn in the qso lf for magnitudes below @xmath74 is evident . in particular the m05 observations are significantly stringent in constraining the faint - end slope . ccccccc mahabal et al . & & 1.7 & 1.9 & 2.3 & 2.5 sq . 28 % + willott et al . & & 2.27 & 2.57 & 2.95 & 3.32 sq . variable + barger et al . ( single source ) & & 2.32 & 2.38 & 2.48 & 160 arcmin@xmath76 & 50 % + [ table|survey ] in figure [ fig|lfz6 ] we also show the interesting result by stern _ et al . _ ( 2000 ) . these authors have discovered a quasar at @xmath111 from deep , multicolor , ground - based observations covering 74 arcmin@xmath76 with z@xmath112 . to estimate their average selection function we redo the calculations in stern _ _ ( 2000 ) and integrate down to @xmath113 ( in their cosmology ) the boyle _ ( 1991 ) @xmath114 quasar lf , evolved as in schmidt _ the effective volume probed is again defined by the @xmath90-band drop - out as in m05 ( stern _ et al . we find that an average completeness of p@xmath115 recovers their quoted @xmath1160.15 quasars in the field of view . as for m05 , we then use the more up - to - date fan _ ( 2001 ) lf , more suitable for the redshift range probed by this survey , and equation ( [ eq|integrallf ] ) . even in this case we let the slope to be variable , @xmath117 in the specific case , while we keep the redshift evolution and normalization at @xmath73 as in fan _ ( 2001 ) . we integrate down to @xmath118 ( see table 1 in stern _ et al . _ 2000 ) , instead of using the measured z@xmath112 . this is because the authors find their source to have unusual spectral features for which they derive a special continuum model which is significantly different from the average spectral template used in this paper . the detection of one source in their field of view is consistent with a comoving density of @xmath119\sim -5.66\ , { \rm mpc^{-3}\ , mag^{-1}}$ ] , lower than the f04 extrapolation at the @xmath120 80% cl ( see figure [ fig|lfz6 ] ) . the number density we find still shows evidence for a turnover in the qso lf expressed in equation ( [ eq|modellf ] ) , with @xmath3 , though less significant than the above results , in agreement with the authors . we now turn to the analysis of x - ray data . ( 2003 ; see also cowie _ et al . _ 2003 ) have used deep multicolor optical data to search for @xmath81 agns in the 2 ms x - ray _ chandra _ deep field - north exposures . they found 500 sources in their field of view out of which 423 have z@xmath121 . color analysis was carried out on all the sources , while 249 sources have spectroscopically confirmed redshifts . of these , only one object was found at @xmath122 and none at @xmath123 . they found a faint agn at @xmath124 with z@xmath125 , corresponding to an absolute magnitude of @xmath126 , i.e. nearly a magnitude fainter than the m05 source , showing the power of x - ray observations . ( 2003 ) claim that none of the spectroscopically unidentified sources with z@xmath121 is red enough to have @xmath81 . with the aid of the @xmath127 method the authors estimated the lf in the comoving volume @xmath128 to be consistent with the extrapolation of the fan _ ( 2001 ) lf with a slope of @xmath129 in the luminosity bin @xmath130 ( in @xmath131 ) . here we proceed with a similar calculation but using equation ( [ eq|integrallf ] ) and extrapolating the fan _ ( 2001 ) lf at redshifts beyond @xmath18 , in the same manner we computed the number density for stern _ et al . _ ( 2000 ) . we find that the number of observed sources down to z@xmath132 integrated between @xmath133 , with @xmath134 and a completeness of p@xmath135 ( cowie _ et al . _ 2002 ) is @xmath136 , in very good agreement with observations . we considered an area of 160 arcmin@xmath76 , corresponding to the solid angle where the hard x - ray flux limit has a signal - to - noise ratio of s / n=3 ( see figure 19 in alexander _ the x - ray flux considered here is @xmath137 ( alexander _ et al . _ 2003 ; barger _ this flux limit is the minimum level of x - ray flux which corresponds to an optical magnitude brighter than z@xmath138 ( the optical detection limit in barger _ ( 2003 ) ) within the whole effective volume considered . this is required as all the 500 sources in the cdfn have optical colors . we find a number density of @xmath139 @xmath140 @xmath141 $ ] @xmath142 corresponding to @xmath143 @xmath144 ( equation ( [ eq|aoxvslandz ] ) ) , a factor of @xmath145 higher , but broadly consistent , with the results by barger _ ( 2003 ; plotted in figure [ fig|lfz6 ] as an open circle with its 99% cl error bars ) . indeed deep x - ray observations effectively allow us to probe even fainter optical luminosities . ( 2002 ; brandt _ et al . _ 2002 ) detected three @xmath0 quasars discovered by fan _ ( 2001b ) with _ chandra _ with exposure times of less than 10 ksec . the objects are detected in x - rays up to 55 kev in rest frame , demonstrating the incredible sensitivity of _ chandra _ to detect faint sources . the faintest quasar in their sample , sdss 1030@xmath1460524 , at @xmath147 was detected in about 8 ksec with an observed flux of @xmath148 . the _ chandra _ deep field observations have exposure time of 2 msec , so could detect down to @xmath149 times fainter sources with a flux limit of @xmath150 . this however would be the sensitivity limit at the aim point , close to the aim point sensitivity reported by alexander _ therefore we expect the _ chandra _ observations to be sensitive in detecting sources down the minimum flux considered , i.e. , we can extend the computation using equations ( [ eq|modellf ] ) and ( [ eq|integrallf ] ) down to the limit of @xmath151 ( i.e. down to z@xmath152 at the average redshift of @xmath153 ) . in figure [ fig|lfz6 ] and table [ table|survey ] we report our findings . using a completeness of about 50% , as computed by cowie _ ( 2002 ) for faint fluxes ( notice that the use of an higher completeness level would strengthen our results ) , we find that the detection of a single source in the _ chandra _ deep field in the range @xmath128 , is consistent with a faint slope of @xmath154 at the 99% cl , down to @xmath155 at the 90% cl ( shown as solid squares ) . ( 2003 ) claim that none of the unidentified sources is red enough to be at @xmath123 . they however also point out that up to nine out of the optically faint spectroscopically unidentified sources with z@xmath156 , could still possibly lie in the range @xmath128 . therefore assuming these sources to rely in the same luminosity bin as the single @xmath124 source , they find an upper limit to their comoving lf about an order of magnitude higher than their previous estimate . following the same approach undertaken for the case of a single source , we apply equations ( [ eq|modellf ] ) and ( [ eq|integrallf ] ) down to @xmath151 and using a completeness of @xmath157 , proper for faint fluxes ( cowie _ et al . we consider all the nine sources not to be heavily obscured , i.e. they are type i agn . we find that the comoving number density of the lf to be @xmath158\sim -4.52 \ , ( { \rm mpc^{-3}\ , mag^{-1}})$ ] , using a slope @xmath159 ( shown as an open square with its 99% cl in figure [ fig|lfz6 ] ) . our estimate is higher but still consistent with the one by barger _ ( 2003 ) . to notice also that our estimate is also quite conservative , lower limits would be inferred extending the computation to lower magnitudes , close to the survey limit ( z@xmath138 ) . the 1-@xmath32 uncertainties in equation ( [ eq|aoxlxlo ] ) produce uncertainties of @xmath160 magn in the final computation of the optical @xmath161 magnitude , as shown in figure [ fig|lfz6 ] . we also notice that our conclusions are quite insensitive to the actual fit used for the optical - to - x - ray flux ratios . for example , similar results are also obtained using the relation between luminosity in the rest - frame @xmath90 band , @xmath162 , and the hard x - ray luminosity @xmath163 calibrated by la franca _ et al . _ ( 2005 ; see their equation ( 4 ) ) for type i agns . the vignali _ ( 2003 ) fit instead of equation ( [ eq|aoxlxlo ] ) would yield even fainter optical magnitudes for the same x - ray flux and number density , thus strengthening our results for a presence of a flattening at the faint - end of the qso lf . we conclude this section noting that all the data are consistent with a break in the qso lf at @xmath164 . the faint - end slope of the _ optical _ type i agn lf , is consistent with being on average @xmath165 , at the 90% cl and @xmath166 , at the 99% cl ( see figure [ fig|lfz6 ] ) . our overall findings are consistent with the fan _ ( 2001 ) lf with @xmath167 extrapolated to @xmath104 ( shown with `` plus '' signs in figure [ fig|lfz6 ] ) . this clearly shows how the high - redshift lf is more consistent with a pde beyond @xmath18 . ( 2006 ) have recently found from a sample of sdss qsos , that the high - redshift qso lf bright slope flattens at @xmath168 from @xmath169 down to @xmath170 at @xmath18 . we suggest instead that a double power - law , steeper at bright magnitudes and shallower at fainter ones , is more indicative for the qso lf at @xmath0 . our results on the optical lf are instead in good agreement with those by fontanot _ ( 2006a ) , who compiled the qso lf from a sample of goods and sdss qsos in the range @xmath171 . they find significant evidence for a bright - end slope of @xmath172 , a break in the lf at @xmath173 and a strong flattening below this limit . in the previous section we have considered the detected sources in the cdfn as type i agns . the single high - redshift source detected with spectroscopic redshift of @xmath124 can in fact be safely considered as a type i agn . converting its measured hard x - ray flux @xmath174 to optical magnitudes ( using equation ( [ eq|aoxvslandz ] ) ) we get z@xmath175 , very close to its actual magnitude of z@xmath176 . however some of the nine optically faint spectroscopically unidentified objects found by barger _ ( 2003 ) could be considered moderately obscured sources , with very faint optical fluxes ( though still detectable ) . in this section we give constraints on the faint - end shape for the agn lf , including the cdfn spectroscopically unidentified sources assuming that a certain fraction of them are type ii agns lying at @xmath0 . indeed the @xmath177 band corresponds to the @xmath178 intrinsic band at @xmath179 , which is not much affected by absorption for column densities @xmath180 . here we notice however that we do not expect the detection of any very obscured , compton - thick source in the cdfn , with @xmath181 . from wilman & fabian ( 1999 ) we have taken the spectrum of two typical agns with column densities @xmath182 and @xmath183 , derived from a monte - carlo realization which includes compton down - scattering which reduces the intensity even in the highest bins of energy . we then computed the ratio between absorbed and unabsorbed luminosities ( the former is the actual _ observed _ flux ) for the various spectra in the @xmath179 redshifted , restframe energy hard band @xmath184 ( the least absorbed band of the cdfn ) . we find that the ratio @xmath185 is @xmath186 for @xmath182 and @xmath9 for @xmath183 ( to notice for comparison that an agn with @xmath187 has a ratio of @xmath188 ) . a typical source with intrinsic luminosity @xmath189 ( fainter luminosities would be below the x - ray flux limit ) in the observed frame will in turn produce observed fluxes @xmath190 and @xmath191 for respectively the @xmath182 and @xmath183 case . therefore sources with @xmath192 @xmath193 and @xmath194 , would be hardly visible in the cdfn field of view ( we remind that the flux limit is @xmath195 ) . on the other hand moderately obscured sources ( with @xmath196 ) would definitely be visible at these high redshifts , and some of them might have been already discovered in the deep _ chandra _ fields , among the optically faint barger _ ( 2003 ) sample . other possible indications of the presence of type ii sources in the high redshift universe also come from koekemoer _ et al . _ ( 2004 ) . from the 2 msec hubble deep field - north and the 1 msec cdfs , these authors have robustly detected a sample of seven sources with extreme optical / x - ray flux ratios @xmath19710 at a flux limit above @xmath198 . although these kind of sources are generally rare , if they are agns lying at @xmath123 such that their ly@xmath83 emission has been redshifted out of the z@xmath37 bandpass , then they would definitely be high - redshift obscured sources . we can not directly make a prediction with their data as we are converting to optical luminosities and these are special sources , strong outliers with respect to equation ( [ eq|aoxvslandz ] ) ( also a conversion to bolometric luminosity would be difficult in this case ) . in this section we still adopt the same x - ray @xmath199-correction as for unabsorbed sources with @xmath30 to convert flux to luminosities . we still use equation ( [ eq|aoxvslandz ] ) , which holds for type i agns , for converting x - ray fluxes to intrinsic optical luminosities for type ii agns , assuming that unification models are on average valid . first we consider _ all _ the sources detected by barger et al . ( 2003 ) as optically faint type ii agns lying at @xmath81 . we then integrate equation ( [ eq|modellf ] ) in the redshift range @xmath133 above a flux limit of @xmath200 corresponding to an effective area of 40 arcmin@xmath76 ( see figure 19 , bottom panel in alexander et al . the adopted flux limit corresponds to apparent magnitudes z@xmath201 at @xmath202 ( equation ( [ eq|aoxvslandz ] ) ) . using @xmath203 we obtain @xmath204 sources in the field of view , corresponding to a comoving density of @xmath205\sim -3.85 \ , { \rm mpc^{-3}\ , mag^{-1}}$ ] . the results are shown with an open square with its 99% confidence level in figure [ fig|lfz6obscu ] with its error bar on the optical luminosity when using equation ( [ eq|aoxvslandz ] ) . if the nine sources are confirmed to be type ii agns at @xmath0 , this would show evidence for an extraordinary contribution of type ii agns at the faint end of the agn lf . assuming all these faint sources to lie at @xmath0 would imply a fraction obscured - to - unobscured agns of order 9 . x - ray surveys at lower redshifts found much more moderate numbers for this ratio ( e.g. ueda _ et al . _ 2003 , la franca _ et al . _ 2005 , gilli _ et al . _ 2006 ) . in fact the above scenario is an extreme one , as most probably many of these optically faint sources probably belong to lower redshifts ( see discussions in barger _ _ 2003 and cowie _ et al . therefore we more safely assume that up to 3 sources , out of the 9 detected , i.e. three times the surface density of the type i lf ( made up of only the single @xmath124 source ) , are actual contributors to @xmath0 agn lf with @xmath206 , following the results by la franca et al . ( 2005 ) , who find an almost flat distribution of the fraction of sources per logarithm bin of @xmath207 . in this second case we get a comoving density of @xmath205\sim -4.3 \ , { \rm mpc^{-3}\ , mag^{-1}}$ ] , shown as a solid square in figure [ fig|lfz6obscu ] with its 99% cl error bars . we then compare our agn lf estimate with what would be obtained by multiplying the average qso lf ( with a faint slope of @xmath208 ; section [ sec|typeilf ] ) by a factor @xmath209 , where @xmath90 is the ratio between type ii and type i agns , as calibrated by several groups for agns detected at lower redshifts . for computing @xmath90 we also include sources up to @xmath210 . we use the results by ueda _ ( 2003 ) , la franca _ et al . _ ( 2005 ) and gilli _ et al . _ as specified above , we consider type ii agns all those sources with @xmath211 . therefore by definition the number ratio type ii - to - type i will be given by @xmath212/n[\log ( n_h/{\rm cm^{-2}})<22]$ ] . for ueda et al . ( 2003 ) and la franca et al . ( 2005 ) we use their quoted distributions for the fraction of objects up to @xmath213 . gilli _ et al . _ ( 2006 ) developed an updated model for the synthesis of the x - ray background , for which they adopted a different definition for the number ratio between obscured and unobscured agns , i.e. @xmath214/n[\log ( n_h/{\rm cm^{-2}})<21]$ ] . according to their model this ratio is redshift - independent , luminosity - dependent and , translated to optical magnitudes , it reads as @xmath215 with @xmath216 , @xmath217 and @xmath218 , i.e. @xmath219 , where @xmath220 and @xmath221 are the @xmath222 $ ] @xmath223 $ ] ratio at the low- and high - luminosity regimes respectively , and @xmath218 is the characteristic luminosity dividing the two regimes ( gilli _ et al . _ 2006 ) . to include also compton - thick sources with column densities up to @xmath224 , we consider double this ratio ( see figure 7 in gilli _ _ 2006 ) . the resulting agn lf from the gilli et al . ( 2006 ) model is shown with a dashed line in figure [ fig|lfz6obscu ] . in the same figure we also plot the results of using @xmath90 as calibrated from ueda _ et al . _ ( 2003 ; dot - dashed line ) and la franca _ et al . _ ( 2005 ; long - dashed line ) . all the models for the distribution of sources as a function of @xmath207 considered provide similar estimates for the total agn lf , and are consistent with the extrapolated data at the faint end , when using 3 sources as proxy for the @xmath0 agn population within the effective area . we remind here that our estimate for the faint - end agn lf could easily be underestimated , as the cdfn can not detect compton - thick sources ( those with @xmath225 ) . to notice that all the models also provide a significant number of sources at the bright end of the agn lf in excess of the qso lf . due to the small volume probed by the cdfn we however can not provide any kind of constraint on the actual contribution of intrinsically very luminous and more rare obscured sources . anyway in the following we will consider as our best estimate of the total agn lf at @xmath0 , the qso lf multiplied by ( 1+@xmath226 ) , where @xmath226 is the type ii - to - type i ratio as extracted from ueda et al . ( 2003 ) . in the following section we will compare theoretical models with such a determination of the agn lf the high redshift agn lf is a very sensitive tool to probe the connections between supermassive black holes ( smbhs ) and their host dark matter ( dm ) halos . in this section we discuss possible physical implications for smbh evolution from our estimated agn lf . we compute the _ total _ agn lf from the formation rates of dm halos simply as ( e.g. , mahmood _ et al . _ 2005 ) @xmath227 here @xmath228 is the formation rate of dm halos with mass @xmath229 at redshift @xmath230 , massive and virialized enough to host a galaxy / agn ; @xmath231 in this model is the average timescale during which an agn is active ( but not necessarily visible ) . we neglect any dependence on mass for this timescale and just set it equal to @xmath232 . @xmath233 is the smbh mass producing the luminosity @xmath234 . in the following for simplicity we will use extended press - schechter ( eps ) theory to compute @xmath228 and we will consider only halos @xmath235 . the limiting mass of @xmath236 as host for the most massive galaxies and most luminous quasars has been constrained through clustering ( e.g. croom _ et al . _ 2005 ) , statistical ( e.g. shankar _ et al . _ 2006 ) , and semi - analytical ( e.g. granato _ et al . _ 2004 ) techniques . most probably the eps theory is a rather poor approximation for describing the statistics of high redshifts agn host halos , moreover a more physical treatment of the baryonic processes is required when comparing with observations ( e.g. lapi _ et al . _ 2006 , fontanot _ et al . _ 2006b ) . however here we neglect such issues and focus on the most suitable relation @xmath237 , among those suggested in the literature , which better compares with observations within the very basic model expressed in equation ( [ eq|philz ] ) . we therefore consider different recipes for the jacobian relation in equation ( [ eq|philz ] ) relating dm and smbh mass . we first discuss the wyithe & loeb ( 2003 ; see also mahmood _ et al . _ 2005 ) model which connects the smbh and dm halo masses as @xmath238 , derived from a basic agn feedback - constrained model . we shall refer to this model as model a in the rest of the paper . using model a , we calculate the @xmath104 lf , shown as a solid line in figure [ fig|lfz6models1 ] . this model is consistent with the agn lf estimated in section [ sec|typeiilf ] . several other groups have also formulated theoretical models for describing smbh evolution in the context of cosmological theories ( e.g. haenelt & rees 1993 , cavaliere & vittorini 2000 , vittorini _ et al . _ 2005 , hopkins _ here we adopt a simple version of one of these models , viz . the model by kauffmann & haenelt ( 2000 ; see also croton 2005 ) who parameterize the link between smbh mass and halo mass as : @xmath239^{-1}\ , m_{\rm cold}\ , , \label{eq|mbhmhalokh}\ ] ] where @xmath240 represents the merger mass ratio between the central and satellite galaxies , i.e. @xmath241 , @xmath242 is the virial velocity of the agn host halo and @xmath243 is the amount of cold baryonic matter ( which will eventually form stars ) . the equation ( [ eq|mbhmhalokh ] ) prescription with @xmath244 , consisting of only major mergers , is our model b. we then consider three cases in which the cold fraction is @xmath245 , @xmath246 , and @xmath247 . the resulting lfs are plotted in figure [ fig|lfz6models2 ] . the first case , in which the amount of baryons per halo is close to the cosmological value , lies well above the observed lf . the second case fits the bright - end lf , assuming a baryonic fraction of a few percent , consistent with results by shankar _ it is interesting to note that a strong decrease of baryonic mass fraction with decreasing halo mass is needed to reproduce the agn lf , down to only 0.5% at the very faint end . we now test a model based on ferrarese _ et al . _ ( 2002 ; see also shankar _ et al . we call this model c , and is the most empirical way to track the @xmath237 relation . we relate the smbh mass to bulge velocity dispersion using the @xmath248 relation . the @xmath32 is then connected to the circular velocity of the host galaxy by using the empirical @xmath249 relation ( from baes _ et al . the circular velocity is then scaled to the virial velocity assuming a navarro , frenk & white ( 1996 ; nfw ) profile and a concentration parameter @xmath250 from bullock _ finally the virial velocity is converted to halo mass using the virial theorem , @xmath251 ( we use the relations in loeb & peebles 2003 given for each @xmath252 ) , which is computed at the moment of the virialization of the host dm halo ( i.e. we set @xmath253 ) . we also add a scatter of @xmath254 in the resulting @xmath237 relation , as empirically suggested by ferrarese ( 2002 ) and theoretically supported by lapi _ our result reads as @xmath255 which is close to the one calibrated locally through statistical arguments by shankar _ ( 2006 ) and to the one obtained by granato _ ( 2004 ) and lapi _ et al . _ ( 2006 ) in semi - analytical models . one uncertainty in the above computations is the radius at which one measures the circular velocity . ferrarese ( 2002 ) uses @xmath256 ( also used in equation [ eq|mbhmhaloc ] ) , where @xmath220 is the scale radius of the nfw dm profile . in figure [ fig|lfz6models3 ] we compare the predictions of model c with the calibrated high-@xmath230 agn lf . the model matches the data when @xmath257 and it has the advantage of being empirically - based , the only assumption here is that the @xmath248 and @xmath258 relations are almost constant in time . our findings therefore differ from those in willott _ ( 2005b ) , who attempted a similar calculation and found that the number of dm halos at @xmath0 falls short by several orders of magnitude in accounting for the number of high - redshift agns . their conclusions were based on the use of the virial theorem calibrated at @xmath259 , which inevitably leads to significantly underestimating the bh mass ( and therefore agn luminosity ) for a given halo . our findings are consistent with @xmath260 . the latter relation in turn has been used by cirasuolo _ ( 2005 ) to show that an impressive fit to the dispersion velocity function ( sheth et _ et al . _ 2003 ; shankar _ et al . _ 2004 ) is recovered when integrating the dm formation rates @xmath261 and setting @xmath262 ( see also loeb & peebles 2003 and shankar _ et al . _ 2006 ) . the simple approach undertaken in this section shows that a , redshift - dependent , super - linear relation of the type @xmath263 , with @xmath264 , as the ones proposed in model a and c , is best suited to represent the agn lf determined in section [ sec|typeiilf ] . alternative scenarios would require a very strong variation of baryonic content and/or agn activity timescale with halo mass . in this section we compute the possible contribution of agns and stars to the reionization of the universe . in the following we will consider only type i agns as contributors of the reionization along our line of sight . the transition from a neutral to a fully ionized inter galactic medium ( igm ) is usually statistically described by a differential equation for the time evolution of the volume filling factor of the medium @xmath265 . the latter quantity quantifies the level of the igm _ porosity _ created by the hii , heii and heiii ionization regions around radiative sources such as qsos and galaxies . when the clumping factor of the medium is high , @xmath266 , which is always a good approximation at high redshifts ( the simulations by gnedin & ostriker 1997 give @xmath267 ) , following madau _ ( 1999 ) one can approximate @xmath268 , where @xmath269 is the volume - averaged mean recombination rate , @xmath270 is the total rate of ionizing photons and @xmath271 is the mean hydrogen density of the expanding igm . the condition for reionization will then read @xmath272 , which translates into a total required ionizing photon rate ( madau _ et al . _ 1999 ; fan _ et al . _ 2001 ) : @xmath273 we compare such a rate with those from qsos and galaxies at similar redshifts . we start computing the photoionization rate from qsos , which is given by ( e.g. madau _ et al . _ 1999 , stiavelli _ et al . _ 2004 ) @xmath274 where @xmath275 is the planck s constant , @xmath276 is the frequency at the lyman limit ( 912 , i.e. one rydberg ) . the volume emissivity of the total population of sources is given as @xmath277 where @xmath278 is in @xmath279 . the integration in equation ( [ eq|enuagn ] ) is usually performed up to 4 rydbergs , as photons more energetic than that are preferentially absorbed by he atoms . anyway we have checked that extending the computation to @xmath280 rydbergs for example , increases our final results by just @xmath281 . @xmath282 in equation ( [ eq|enu ] ) is instead the minimum luminosity probed by the survey , which we fix to @xmath283 , following the results in section [ sec|typeilf ] . we thus find that the agn contribution to reionization at redshift @xmath0 integrating down to @xmath283 , varies from @xmath284 to @xmath285 , using a faint - end slope @xmath66 of -2.2 , -2.8 respectively . these results are rather independent of the details for the agn spectrum . for the computations above we have used the spectrum described in schirber & bullock ( 2003 ) , however using , for example , the three - power law spectrum adopted by madau _ ( 1999 ) , reduces the agn contribution by @xmath286 . we follow meiksin ( 2005 ) for computing the photoionization rate from galaxies @xmath287 here @xmath288 is the production rate of ionizing photons per solar mass of stars formed , @xmath289 is the star formation rate ( sfr ) in units of @xmath290 , and @xmath291 is the average fraction of ionizing photons which can freely escape the star - forming regions into the igm . the photons per solar mass has been taken from smith _ et al . _ ( 2002 ; see also meiksin 2005 ) for a salpeter initial mass function ( imf ) and a metallicity of 20% the solar . the sfr was instead recently derived by richard _ ( 2006 ) to be @xmath292 from a deep survey of lensing clusters aimed at constraining the abundance of star - forming galaxies at redshift @xmath293 . we then take results from the recent paper by shapley _ ( 2006 ) , who have constrained the escape fraction in the ionizing lyman continuum from a sample of 14 @xmath294 star - forming galaxies , finding an average @xmath295 , and we use the average value of @xmath296 . in the following we neglect the diffuse emission from the igm itself which however could add an additional photoionization rate ( up to 50% , e.g. meiksin & madau 1993 ) over the direct one from the sources . we find that galaxies can contribute about @xmath297 at @xmath104 to reionization . we now take a step forward and integrate the contribution to reionization from both type of sources within the range of the _ chandra _ deep fields , i.e. @xmath298 , and compare the inferred number of ionizing photons with the number of available neutral hydrogen atoms at @xmath0 , @xmath299 ( see stiavelli _ we find that the contribution from agns to grow up to @xmath300 while the one from galaxies being about @xmath301 , keeping the sfr constant in this redshift bin , as suggested by observations . the pde evolution of the agn lf for @xmath302 makes the contribution from agns grow more than the one from galaxies . the photons produced by galaxies reach an average contribution to reionization of @xmath303 , implying a minor contribution from qsos , a conclusion in very good agreement with previous calculations ( e.g. meiksin 2005 , fan _ et al . _ 2001 , barger _ et al . _ 2003 ) . however we also notice that bouwens _ ( 2005 ) from the deep nicmos fields in the hubble ultra deep fields ( hudf ) have detected 5 objects in 5.7 sq . deg . , inferring a sfr about a factor of @xmath145 below the one found by richard _ this would drop the galaxy contribution down to @xmath304 . similar points were also recently addressed by mannucci _ ( 2006 ) who claim that the high - redshift sfr could fall short of completely reionizing the universe . they however also point out that the bouwens _ et al . _ ( 2005 ) estimate should be further increased by a factor of 2 for dust extinction and/or accounting for luminosity evolution in the estimate . using therefore the bouwens _ et al . _ ( 2005 ) result in equation ( [ eq|enugalx ] ) would still leave room for a significant contribution from qsos . other ways to fill the possible `` missing '' photons from galaxies could be to alter significantly the photon luminosity from massive stars , invoking extra igm diffuse light , or including other photoionizing sources ( such as population iii stars and/or miniquasars powered by intermediate - mass bhs ) . however the contribution from the latter sources must be @xmath305 at the most . in a recent paper , cool _ et al . _ . ( 2006 ) found three qsos at high redshifts ( @xmath306 ) down to magnitudes of @xmath307 . the detection of these quasars is roughly consistent with the extrapolation of the bright steep slope of the quasar lf by f04 with @xmath308 . at lower magnitudes not many qsos are observed , as recently concluded by the optical / nir surveys conducted by m05 , stern _ ( 2005 ) and willott _ et al . _ ( 2005 ) . in this paper we show that such _ non_-detection of qsos implies a significant downturn in the optical qso lf at lower magnitudes with respect to those probed by cool _ moreover we show that the _ chandra _ deep field observations probe about @xmath309 magnitudes deeper than the optical ones ( down to @xmath310 ) . the inferred faint end co - moving number density is lower by at least an order of magnitude with respect to the extrapolation of the f04 lf , at a high significance level . if a fraction of the nine _ chandra _ deep field sources are type ii agns detected at x - ray fluxes @xmath311 and with very faint optical magnitudes ( z@xmath156 ) , this would place a constraint on the shape of the type i plus type ii agns at faint magnitudes , being consistent with a slope close to the extrapolation of the f04 lf , i.e. @xmath312 . this estimate of the agn lf is however representative of only the sources with @xmath313 , as more obscured sources would be below the cdfn flux limit . x - ray surveys at lower redshifts ( e.g. la franca et al . 2005 ) have shown that an almost flat distribution in the number of sources is observed up to column densities of @xmath313 . we can therefore safely assume that only 3 , out of the nine optically faint sources to be agns at @xmath0 . this yields a total agn lf consistent with updated synthesis models for the xrbg . we then compare this result with predictions from three simple physically and empirically motivated basic models for agn evolution in dm halos . we generally find that a redshift - dependent relation of the kind @xmath314 , with @xmath264 , produces models more consistent with the agn lf . such relations were also found from independent statistical studies by shankar _ ( 2006 ) and rotation curve extrapolations in spirals and ellipticals ( ferrarese 2002 ) . alternative models , such as varying the amount of cold gas fraction or agn active timescale in smaller halos , would require extreme fine tuning and a large variation in the parameters with dark matter halo mass . here we also notice that if we reduce the agn duty - cycle just to the visible optical phase , i.e. substituting @xmath231 with @xmath315 in equation ( [ eq|modellf ] ) , the match between models and the type i agn lf derived in section [ sec|typeilf ] , would yield similar results on the type of @xmath237 relation . finally we have shown that agns can account for at most @xmath316 of the total reionization of the universe at redshifts around @xmath317 . however we also notice that a strong caveat on the actual role of galaxies in reionizing the universe , is set by the normalization of the cosmological sfr at high redshifts , which is still uncertain by factors of about @xmath318 . we acknowledge c. s. kochanek , p. s. osmer , d. weinberg and p. martini for helpful discussions . fs also acknowledges s. cristiani and f. fontanot for sharing data on the quasar template spectrum . the authors are highly indebted to the referee for his / her very useful comments and suggestions . this work was supported in part by nasa grant grt000001640 . alexander , d. m. , et al . 2003 , aj , 126 , 539 baes , m. , buyle , p. , hau , g. k. t. , & dejonghe , h. 2003 , mnras , 341 , l44 barger , a. j. , et al . 2003 , aj , 126 , 632 boyle , b. j. , et al . 1991 , in asp conf . 21 , the space distribution of quasars , ed . d. crampton ( san francisco : asp ) , 191 brandt , w. n. et al . 2002 , apj , 569 , l5 bullock , j. s. , kolatt , t. s. , rachel , y. s. , somerville , s. , kravtsov , a. v. , klypin , a. a. , primack , j. r. , & dekel , a. 2001 , mnras , 321 , 559 cavaliere , a. , & vittorini , v. 2000 , apj , 543 , 599 cimatti , a. , et al . 2002 , a&a , 392 , 395 cirasuolo , m. , shankar , f. , granato , g. l. , de zotti , g. , & danese , l. 2005 , apj , 629 , 816 cool , r. j. , kochanek , c. s. , eisenstein , d. j. , stern , d. , brand , k. , brown , m. j. i. , dey , a. , eisenhardt , p. r. , fan , x. , gonzalez , a. h. , green , r. f. , jannuzi , b. t. , mckenzie , e. h. , rieke , g. h. , rieke , m. , soifer , b. t. , spinrad , h. , & elston , r. j. 2006 , aj , in press , astroph/0605030 cowie , l. l. , et al . 1995 , nature , 377 , 603 cowie , l. l. , et al . 2002 , apj , 566 , l5 cowie , l. l. , et al . 2003 , apj , 584 , l57 cristiani , s. , & vio , r. 1990 , a&a , 227 , 385 cristiani , s. , et al . 2004 , apj , 600 , l119 croom , s. m. , et al . 2005 , mnras , 356 , 415 croton , d. j. 2005 , astroph/0512375 dijkstra , m. , & wyithe , j. s. b. 2006 , astroph/0606334 fan , x. , et al . 2001 , aj , 121 , 54 fan , x. , et al . 2001b , aj , 122 , 2833 fan , x. , et al . 2004 , apj , 128 , 515 ferrarese , l. 2002 , apj , 578 , 90 ferrarese , f. , & ford , h. c. 2005 , ssrv , 116 , 523 fiore , f. , et al . 2003 , a&a , 409 , 79 fontanot , f. , cristiani , s. , monaco , p. , nonino , m. , vanzella , e. , brandt , w. n. , grazian , a. , & mao , j. 2006a , a&a , accepted , astroph/0608664 fontanot , f. s. , monaco , p. , cristiani , s. , & tozzi , p. 2006b , mnras , accepted , astroph/0609823 gehrels , n. 1986 , apj , 303 , 336 gilli , r. , comastri , a. , & hasinger , g. 2006 , a&a , accepted , astroph/0610939 granato , g. l. , silva , l. , monaco , p. , panuzzo , p. , salucci , p. , de zotti , g. , & danese , l. 2001 , mnras , 324 , 757 granato , g. l. , de zotti , g. , silva , l. , bressan , a. , & danese , l. 2004 , apj , 600 , 580 haehnelt , m. g. , rees , m.j . 1993 , mnras , 263 , 168 hopkins , p. f. , hernquist , l. , cox , t. j. , di matteo , t. , robertson , b. , & springel , v. 2005 , apjs , in press ( astroph/0506398 ) kauffmann , g. , & haehnelt , m. 2000 , mnras , 311 , 576 kennefick , j. d. , djorgovski , s. g. , & de carvalho , r. r. 1995 , aj , 110 , 2553 la franca , f. , et al . 2005 , apj , 635 , 864 lapi , a. , shankar , f. , mao , j. , granato , g. l. , silva , l. , de zotti , g. , & danese , l. apj , in press , astroph/0603819 loeb , a. , & peebles , p. j. e. 2003 , apj , 589 , 29 mahabal , a. , stern , d. , bogosavljevic , m. , djorgovski , s.g . , & thompson , d. 2005 , apj , 634 , l9 mahmood , a. , devriendt , j. e. g. , & silk , j. 2005 , mnras , 359 , 1363 mannucci , f. , buttery , h. , maiolino , r. , marconi , a. , & pozzetti , l. 2006 , astroph/0607143 mathur , s. , wilkes , b. j. , & ghosh , h. 2002 , apj , 570 , l5 monaco , p. , salucci , p. , & danese , l. 2000 , mnras , 311 , 279 navarro , j. f. , frenk , c. s. , & white , s. d. m. 1997 , apj , 490 , 493 ( nfw ) osmer , p. s. 2004 , in ho l.c . , ed . , carnegie observatories astrophysics series , vol . 1 , coevolution of black holes and galaxies . cambridge univ . press , cambridge , p. 324 pei , y. c. 1995 , apj , 438 , 623 richards , g. t. , et al . 2005 , mnras , 360 , 839 richards , g. t. , et al . 2006 , aj , in press ( astroph/0601434 ) shankar , f. , salucci , p. , granato , g. l. , de zotti , g. , & danese , l. 2004 , mnras , 354 , 1020 shankar , f. , lapi , a. , salucci , p. , de zotti , g. , & danese , l. 2006 , apj , 643 , 14 shapley , a. e. , steidel , c. c. , pettini , m. , adelberg , k. l. , & erb , d. k. 2006 , apj in press , astroph/0606635 sharp , r. g. , crampton , d. , hook , i. m. , & mcmahon , r. g. 2004 , mnras , 350 , 449 sheth , r. k. , & tormen , g. 2002 , mnras , 329 , 61l schirber , m. , & bullock , j. s. 2003 , apj , 584 , 110 spergel , d. n. , et al . 2003 , apjs , 148 , 175 springel , v. , et al . 2005 , nature , 435 , 629 schmidt , m , schneider , d. p. , & gunn , j. e. 1995 , aj , 110 , 68 steffen , a. t. , strateva , i. , brandt , w. n. , alexander , d. m. , koekemoer , a. m. , lehmer , b. d. , schneider , d. p , & vignali , c. 2006 , aj , 131 , 2826 stern , d. , et al . 2000 , apj , 533 , l75 stern , d. , et al . 2000b , aj , 119 , 1526 stiavelli , m. , fall , s. m. , & panagia , n. 2004 , apj , 600 , 508 tezca , m. , et al . 2004 , apj , 605 , l109 ueda , y. , akiyama , m. , ohta , k. , & miyaji , t. 2003 , apj , 598 , 886 vignali , c. , brandt , w. n. , & schneider , d.p . 2003 , aj , 125 , 433 vittorini , v. , shankar , f. , & cavaliere , a. 2005 , mnras , 363 , 1376 willott , c. j. , delfosse , x. , forveille , t. , delorme , p. , & gwyn , s. d. j. 2005 , apj , 633 , 630 willott , c. j. , et al . 2005b , apj , 626 , 657 wilman , r. j. , & fabian , a. c. 1999 , apj , 309 , 862 wolf , c. , et al . 2003 , a&a , 408 , 499 wyithe , j. s. b. 2004 , mnras , 351 , 1266 wyithe , j. s. b. , & loeb , a. 2003 , apj , 595 , 614

the diffuse spectrum of ultra - high energy ( uhe ) cosmic rays ( cr ) is expected to consist of a superposition of fluxes from many individual point - sources distributed throughout the universe . the absence of significant event clustering across the sky sets a lower limit on the local source density or , equivalently , a lower limit on the number of sources that effectively contribute to the spectrum . typically this number is very large and the uhe cr spectrum is expected to reflect the average contribution of these sources . it is hence common practice for theoretical studies to approximate the distribution of cr sources via a spatially homogeneous and isotropic emission rate density that reproduces the average source spectrum . this treatment greatly simplifies the study of uhe crs and can reproduce various spectral features from a simple power - law source spectrum . the energy spectrum of cosmic rays follows a simple power - law over many energy decades . small variations of the spectral index can be interpreted either as a transition between cr populations or as an imprint of cr propagation effects . the _ ankle _ a hardening of the spectrum at @xmath1 ev could be formed naturally by the superposition of two power - law fluxes and serves as a candidate of the transition between galactic heavy nuclei and extra - galactic cosmic ray protons @xcite . it has also been advocated that this feature could be well reproduced by a proton - dominated power - law spectrum , where the ankle is formed as a _ dip _ in the spectrum from the energy loss of protons via bethe - heitler pair production @xcite . in this case extra - galactic protons would already have started to dominate the spectrum beyond the _ 2nd knee _ , a feature which corresponds to a slight _ softening _ of the spectrum at @xmath2 . proton - dominance beyond the ankle is ultimately limited by the onset of photopion production on the cosmic microwave background ( cmb ) , whereas dominance of a heavy composition is restricted by nucleus photodisintegration through the giant dipole resonance ( gdr ) @xcite the so called _ greisen - zatsepin - kuzmin _ ( gzk ) suppression at around @xmath3 ev . indeed , a flux suppression in this energy region has been observed in the hires and auger data with a high statistical significance @xcite . as noted elsewhere @xcite , secondary neutrinos and @xmath4-rays of these hadronic interactions can serve as additional discriminators between various cr models . in order to collect the elusive events above @xmath5 ( which present an integrated flux of less than 1 event per km@xmath6 per steradian and per century ) observatories with large apertures and long exposure time are needed . today , the leading role in cr is played by ground based facilities that cover vast areas with particle detectors overlooked by fluorescence telescopes . the largest is the pierre auger observatory , with a surface detector array of @xmath7 water cherenkov tanks covering @xmath8 km@xmath6 , which accumulates annually about @xmath9 of exposure @xcite . the more recently constructed telescope array ( ta ) covers 700 km@xmath6 with 507 scintillator detectors @xcite , which should accumulate annually about @xmath10 of exposure . in the near future , the jem - euso mission will orbit the earth on board the international space station at an altitude of of about 400 km . whilst in the `` nadir '' mode , the remote - sensing space instrument ( with @xmath11 field of view ) will monitor an area of approximately @xmath12 , recording video clips of fast uv flashes by sensing the fluorescence light produced through charged particle interactions . this innovative pathfinder mission will observe approximately @xmath13 annually @xcite , a factor of 10 above auger . in this paper we elaborate on the question as to what extent the spectral information in the gzk region can be used to discriminate between different cr source composition models . due to the strength of the gzk mechanism the spectrum in this region is dominated by ( and requires the presence of ) local sources @xcite . in this case the flux from a few cr sources can significantly fluctuate from a homogeneous distribution that is typically assumed in cr flux predictions @xcite . in contrast to poisson fluctuations in the gzk region @xcite the manifestations of ensemble fluctuations persist in the limit of large event statistics . we will quantify these stochastic fluctuations in the following utilizing an analytic solution to the flux of cr nuclei derived in refs . @xcite . we will start in sec . [ sec : propagation ] with a brief review of the propagation of cr nuclei and the calculation of the mean observed fluxes . in sec . [ sec : variation ] these results will be used to derive an analytic approximation of the flux and mean mass variations due to the distribution of sources . we summarize our findings in sec . [ sec : conclusion ] . since cosmic rays are subject to deflections in galactic and inter - galactic magnetic fields the observed cr events do not point directly back to their sources . the identification of the cr sources is hence experimentally challenging and has so far proved inconclusive . there is a general consensus that the sources which are responsible of the uhe cr spectrum are of extra - galactic origin . these sources are expected to follow a spatially homogeneous distribution and the mean ( ensemble - averaged ) flux of uhe crs ( of type @xmath14 ) follows a set of ( boltzmann ) continuity equations of the form : @xmath15 together with the friedman - lematre equations describing the cosmic expansion rate @xmath16 as a function of red - shift @xmath17 . we follow the usual cosmological _ concordance model _ dominated by a cosmological constant with @xmath18 and a ( cold ) matter component , @xmath19 where , normalized to its value today of @xmath20 kms@xmath21mpc@xmath21 @xcite . the time - dependence of the red - shift can be expressed via @xmath22 . the first and second terms on the r.h.s . of eq . ( [ eq : diff0 ] ) describe , respectively , red - shift and other continuous energy losses ( cel ) with rate @xmath23 . the third and fourth terms describe more general interactions involving particle losses ( @xmath24 anything ) with total interaction rate @xmath25 , and particle generation of the form @xmath26 with differential interaction rate @xmath27 . the last term on the r.h.s . corresponds to the emission rate of crs of type @xmath14 per co - moving volume , depending on the emission rate @xmath28 per source and their density @xmath29 . the two main reactions of uhe cr nuclei during their cosmic evolution are photodisintegration @xcite and bethe - heitler pair production @xcite with the cosmic radiation background . in addition to the dominant contribution of the cmb we also include the infra - red / optical background from ref . @xcite in our calculation of interaction and energy loss rates . photodisintegration is dominated by the gdr with main branches @xmath30 and @xmath31 where @xmath32 indicates a proton or neutron @xcite . the gdr peak in the rest frame of the nucleus lies at at about @xmath33 mev for one - nucleon emission , corresponding to @xmath34 ev in the cosmic frame with photon energies @xmath35 mev . at energies below 10 mev there exist typically a number of discrete excitation levels that can become significant for low mass nuclei . above 30 mev , where the photon wavelength becomes comparable or smaller than the size of the nucleus , the photon interacts via substructures of the nucleus . out of these the interaction with quasi - deuterons is typically most dominant and forms a plateau of the cross section up to the photopion production threshold at @xmath36 mev . bethe - heitler pair production can be treated as a continuous energy loss process with rate @xmath37 , where @xmath38 is the energy loss rate of protons @xcite . the ( differential ) photodisintegration rate @xmath39 ( @xmath40 ) is discussed in more detail in ref . @xcite . the evolution of the spectra proceeds very rapidly on cosmic time scales and the diffuse flux of secondary nuclei , @xmath41 , looks generally quite different from the initial source injection spectrum @xcite . the reaction network of nuclei depend in general on a large number of stable or long - lived isotopes . if the life - time of an isotope is much shorter than its photodisintegration rate it can be effectively replaced by its long - lived decay products in the network ( [ eq : diff0 ] ) . typically , neutron - rich isotopes @xmath42-decay to a stable or long - lived nucleus with the same mass number . in most cases there is only one stable nucleus per mass number below @xmath43fe with the exception of the pairs @xmath44cr/@xmath44fe , @xmath45ca/@xmath45ti , @xmath46ar/@xmath46ca and @xmath47s/@xmath47ar . we follow here the approach of puget , stecker and bredekamp ( psb ) @xcite and consider only a single nucleus per mass number @xmath48 in the decay chain of primary iron @xmath43fe . this psb - chain of nuclei linked by one - nucleon losses is discussed in more detail in ref . @xcite . note that the set of boltzmann eqs . ( [ eq : diff0 ] ) does not take into account the deflection of charged cr nuclei during their propagation through magnetic fields . magnetic scattering can be viewed as a diffusion process that depends on the particle s larmor radius @xmath49 and the coherence length of the magnetic field . at energies where the diffusion length becomes larger than the distance to the nearest cr source the cr spectrum is expected to be suppressed . it has been speculated that for particularly strong inter - galactic magnetic fields of strength @xmath50 ng and coherence length of @xmath50 mpc ( see _ e.g. _ ref . @xcite ) , the diffusive propagation of cr protons can start to affect the spectrum below about @xmath51 ev @xcite . for heavier nuclei diffusive propagation can in principle remain important up to higher energies due to the dependence @xmath52 @xcite . the results of this paper assume that the contribution of inter - galactic or galactic magnetic fields can be neglected for the calculation of the uhe cr spectrum . as an example we show in fig . [ fig1 ] the solution of eq . ( [ eq : diff0 ] ) for an iron source model using a power - law spectral emission rate @xmath53 with @xmath54 and @xmath55 ev . this model is motivated by a previous study @xcite and reproduces the auger data above the ankle within systematic uncertainties . the dashed black line corresponds to source contributions above @xmath56 mpc extending up to redshift @xmath57 where no red - shift evolution of the emission is assumed , _ i.e. _ @xmath58 . the solid black line marks the local contribution up to 1 gpc calculated with the same method . this local contribution where red - shift scaling of energies and interaction rates is suppressed can be approximated by an analytic solution which is shown as the green solid line in the plot . we will discuss this method in the next section . gev and power index @xmath54 . the upper solid ( green ) line shows the result of the analytic approximation ( [ eq : meann ] ) for a homogeneous distribution of sources @xmath59 with @xmath60 mpc and @xmath61 gpc . the lower solid ( black ) line is the corresponding numerical calculation based on eqs . ( [ eq : diff0 ] ) including redshift scaling of interaction rates and energies . the relative difference between these calculations can be traced back to the onset of redshift scaling for propagation distances of the order of gpc ( see main text ) . the dashed black line shows a model including sources up to @xmath57 . also shown is recent data from hires @xcite , auger @xcite and the telescope array ( ta ) @xcite . ] photodisintegration and photopion interactions of nuclei happen on timescales much shorter than the hubble scale . since we are interested in the variation of the average flux and mass composition from the distribution of local cr sources we will neglect redshift scalings in the following . in this case the green s function of the boltzmann equations ( [ eq : diff0 ] ) can be expressed in a simple analytic form , as discussed in refs . @xcite . it is convenient to study the flux of nuclei with mass number @xmath48 in terms of the energy per _ nucleon _ @xmath62 that is conserved under photodisintegration . introducing the cr flux @xmath63 per nucleon energy bin @xmath14 with bin - width @xmath64 ( centered at a nucleon energy @xmath65 ) and corresponding emission rates @xmath66 we find an analytic solution of the form @xmath67 the sum on the r.h.s . runs over all possible production paths @xmath68 of a cr nucleus with final mass number @xmath48 and nucleon energy bin @xmath14 . each of the @xmath69 elements @xmath70 of the path @xmath68 consists of a doublet @xmath71 denoting the mass number @xmath72 and nucleon energy bin @xmath73 of intermediate nuclei . the first element @xmath74 of the paths corresponds to the nucleus emitted from a source at a rate @xmath75 ( possibly equal to zero depending on source composition ) and the last element is fixed at @xmath76 . each path is weighted by a set of @xmath69 dimensionless amplitudes @xmath77 that are independent of the source distance @xmath78 . for more details we refer to ref . @xcite and appendix [ sec : analytic ] . we now want to study the statistical mean and variation of the aggregated flux of @xmath79 local cr sources denoted by @xmath80 as well as its corresponding mass composition . herein we assume that the probability distribution function ( pdf ) for local ( @xmath81 ) sources is flat in euclidean space . let @xmath79 sources be distributed between redshift @xmath82 and @xmath83 . the number of sources can then be expressed via the ( local ) source density @xmath84 as @xmath85 . the pdf of a single source is then given by @xmath86 note that the local density @xmath84 is limited by auto - correlation studies of uhe cr events ; the lower @xmath84 the larger the average emission rate of the sources and the greater the chance of local event clusters across the sky . a local source density of @xmath87 mpc@xmath88 is consistent with the absence of `` repeaters '' in cr data @xcite . values as low as @xmath89 mpc@xmath88 are still marginally consistent with auto - correlation studies of uhe cr nuclei @xcite . we will consider these two cases as fiducial values in the following . following ref . @xcite the ensemble - average of a quantity @xmath90 depending on the distance of the @xmath79 sources can then be expressed as @xmath91 the ensemble - average of the local flux of cr nuclei using eqs . ( [ eq : nai ] ) and ( [ eq : fullsol ] ) is then simply @xmath92 using the abbreviation ( [ eq : a ] ) we can then write this in an analytic form as @xmath93 where we define @xmath94 from the experimental point of view the interesting quantities are the ensemble - averaged total flux of nuclei @xmath95 and mean mass number @xmath96 at the highest cr energies @xmath97 . the mean total flux is simply given by eq . ( [ eq : mean ] ) as is a short - hand notation for @xmath98 with @xmath99 . ] @xmath100 we now return to the example of an iron source model shown in fig . the solid green line in fig . [ fig1 ] shows the result of the analytic approximation ( [ eq : mean ] ) . this approximation agrees with the numerical result ( solid black line ) within a factor two or better depending on the energy . the relative difference is expected from the onset of redshift scaling for propagation distances of the order of gpc ( @xmath101 ) . at low energies this introduces a relative upward shift of the flux of @xmath102% for a source model with @xmath54 and @xmath58 . this agrees well with the result of the calculations . in addition , threshold effects that lead to breaks in the spectrum scale with redshift as @xmath103 and are shifted to lower energies by up to @xmath104% . this effect can also be noticed by the relative position of the break in fig . [ fig1 ] . this example illustrates the limitations of the analytic solution for the calculation of large scale ( early - time ) contributions to the cr flux . however , the analytic approximation provides a convenient description of the nuclei cascades that happen on small scales and depend on the local source distribution . in particular , it enables us to study ensemble - variations of the flux . defining the residual @xmath105 , _ etc . _ , we have as usual @xmath106 . using eqs . ( [ eq : fullsol ] ) and ( [ eq : nai ] ) we can then write the variation of the cr flux in an explicit analytic form as @xmath107 where @xmath108 is defined as the expression @xmath109 note that the last term in eq . ( [ eq : var ] ) is sometimes omitted since the number of sources @xmath79 is expected to be large , but we keep it in our calculations . based on these definitions we can express the relative variation of the total flux via the two - point density perturbations ( [ eq : var ] ) as @xmath110 with eq . ( [ eq : meann ] ) we can also _ define _ the mean mass number as @xmath111 note that eq . ( [ eq : meana ] ) is in the strict sense not the ensemble - average but serves as a first order estimator . for small fluctuations around the mean value we can approximate the relative variation of the mean mass number ( [ eq : meana ] ) via the two - point correlation function ( [ eq : var ] ) as @xmath112 in figure [ fig2 ] we show contour plots of the relative error of the flux ( red ; left plots ) and mass composition ( blue ; right plots ) for the case of iron sources with different model parameters indicated above the plots . the axes show the observed cr energy vs. the exponential cutoff energy @xmath113 . for the calculation we introduced logarithmic energy bins of @xmath114 and smoothed the result with a gaussian kernel to account for an experimental resolution of @xmath115 . this procedure smoothes out features in the relative variance that are beyond the experimental resolution and result from the rapid nuclei transitions in the gzk region in combination with the relative cr energy shift with the mass of the daughter nuclei . the results do not strongly depend on the spectral index as can be seen for the cases @xmath54 and @xmath116 with otherwise equal parameters that are shown in the plots of the first two rows . this set of plots assumes a local distribution between 10 mpc and 1 gpc . the relative errors are significantly reduced as we increase the distance to the closest source to 100 mpc as shown in the plots of the last row in fig . however , this is only marginally reproducing the uhe cr spectrum as pointed out in ref . note that the green lines in the plots of the top row mark the contribution for the iron source model with @xmath55 ev considered in fig . [ fig1 ] . one can also notice from the lower plots of fig . [ fig2 ] that the relative error of the average mass composition is below 1% for cr energies below @xmath117 ev . this energy marks the end of the energy region where cr event statistics allow an inference of the mass composition from cr data . note that all the plots in fig . [ fig2 ] show the case of a local source density of @xmath87 mpc@xmath88 and levels increase as @xmath118 . hence , even for a density @xmath89 mpc@xmath88 , still marginally consistent with the data , the ensemble fluctuation on the average mass composition probed by present generation experiments may be safely neglected . so far we have considered the case that @xmath119 valid at the highest cr energies where we can neglect cosmological contributions of the sources and treat the problem as effectively local . in the opposite case , @xmath120 , it is also possible to give an analytic expression of the statistical variation of the cr flux . we assume that @xmath79 sources are isotropically distributed between redshift @xmath121 and @xmath122 with comoving density @xmath123 , and local density @xmath84 . the number of sources is then given by @xmath124 where the comoving volume is @xmath125 with comoving distance ( in a flat universe ) @xmath126 . the pdf of a single source as a function of redshift is then a simple generalization of eq . ( [ eq : prob1 ] ) , @xmath127 as long as only adiabatic redshift scaling is involved the flux of a single source is given by @xmath128 . assuming @xmath129 we then obtain a relative variation of @xmath130 ^ 2}{\left[\int{\rm d } z\,p(z)(1+z)^{-\gamma}/d^2_c(z)\right]^2}-\frac{1}{n_s}\,.\ ] ] we show values for @xmath79 and @xmath131 in table [ tab1 ] . obviously , the statistical variation of the average mass number is negligible in this regime . .estimated source number and adiabatic variation defined in eqs . ( [ eq : ns ] ) and ( [ eq : adisig ] ) . we show results for a power - law cosmological evolution as @xmath132 with @xmath133 and for the star formation rate ( sfr ) according to ref . @xcite ( with cutoff @xmath134 ) . in all cases we assume a local density of @xmath87 mpc@xmath88 . [ cols="^,^,^,^,^,^",options="header " , ] the two limiting behaviors of the relative flux variation ( [ eq : locsig ] ) and ( [ eq : adisig ] ) motivates the following treatment of the overall relative flux variation . using the numerical result of the set of boltzmann eqs . ( [ eq : diff0 ] ) we can calculate the average flux for the local ( @xmath135 gpc ) , and global source distribution . this corresponds to the solid and dashed lines , respectively , in fig . the flux variation can then be approximated via the superposition @xmath136 where @xmath137 is calculated from the analytic approximation ( [ eq : locsig ] ) , @xmath138 the adiabatic limit ( [ eq : adisig ] ) for @xmath139 gpc ( @xmath140 ) and @xmath141 from the numerical evaluation . in the left plot of fig . [ fig3 ] we show the result of this procedure for the previous example of iron sources shown in fig . [ fig1 ] . the solid line ( corresponding to the dashed line in fig . [ fig1 ] ) shows the contribution from the cosmological distribution of iron sources . the shaded areas show the flux within its variation based on eq . ( [ eq : combinedsig ] ) for a local source density @xmath87 mpc@xmath88 ( dark gray ) and @xmath89 mpc@xmath88 ( light gray ) . note that the relative size of the error bands increases as @xmath118 . the right plot of fig . [ fig3 ] shows the corresponding result of a proton source model with similar source parameters . the amplitude of the ensemble fluctuation is comparable to the case of iron and does not serve as a direct measure of the source composition . however , the spectral feature of the gzk - suppression is significantly different to the case of the iron model . in particular , the iron source model experiences a much steeper suppression at the upper end of the uhe cr spectrum . this poses a challenge to account for the most extreme uhe events like the @xmath142 ev fly s eye event @xcite , if sources are too distant @xcite . candidate nearby sources of heavy nuclei include starburst galaxies @xcite and ultra - fast spinning newly - born pulsars @xcite . our method provides a tool to distinguish ensemble fluctuations from spectral features of the the average source contribution on a statistical basis . proposed future space - based cr observatories can reach integrated exposures of @xmath143 @xcite . this is a factor of about 100 larger than the integrated exposure reached by present ground - based air shower arrays . the statistical error of uhe cr measurements can thus be improved by a factor 10 . with such a resolution the observed uhe cr spectrum can show a significant deviation from the ensemble mean due to the discreteness of close - by source contributions . for instance , the spectrum could exhibit `` spectral wiggles '' beyond the gzk suppression . our method describes a way to quantify the bin - to - bin amplitude of these modulations for the case of a ultra - high energy cosmic ray nuclei . in this paper we have studied the ensemble fluctuations of the mean flux and average mass number of uhe cr nuclei from the distribution of sources . we have derived an analytic expression for the relative errors which applies to the cr data at the highest energies dominated by the local sub - gpc ( low redshift ) source distribution . for lower energies another analytic approximation can be derived through the use of adiabatic scaling of the source contributions . this method can be easily generalized to the case of ensemble fluctuations due to the source emission parameters such as the spectral index and the maximal energy . as an illustration , we applied these results to a fiducial iron source model for which a homogeneous distribution of sources had previously been found to successfully reproduce the cr data within systematic uncertainties . for the case of a local source density @xmath87 mpc@xmath88 , the resultant ensemble fluctuations of the average mass composition on top of the ensemble mean were found to exist at a tolerable level for the analysis of present generation cr composition data below @xmath117 ev ( see right panels of fig.[fig2 ] ) . the relative ensemble fluctuation around the mean flux increases with energy and rises above the level of 10% at about @xmath144 ( see left panels in fig . [ fig2 ] and left plot in fig.[fig3 ] ) . this flux variation is beyond the sensitivity of present day cr observatories as indicated by the range of the dark gray shaded bands in fig . [ fig3 ] in comparison to the statistical errors of cr data . the amplitude of the flux variation is similar in the case of proton models as shown in the right plot of fig . however , the level of ensemble fluctuations for present generation spectral studies of the gzk suppression are potentially not ignorable if a smaller source density of @xmath89 mpc@xmath88 is assumed , which still remains marginally consistent with angular correlation studies assuming heavy nuclei . this is shown as the light gray shaded bands in fig . [ fig3 ] . unless the actual source densities are much larger than those considered here , next generation experiments reaching accumulated exposures of @xmath143 should be sufficiently sensitive to potentially discern these fluctuations . for instance , the surface detector array of the auger observatory observed 25 events between @xmath145 ev to @xmath146 ev with an integrated exposure of @xmath147 @xcite . with an almost fifty times larger integrated exposure of future observatories the relative poisson error at this energy should drop below 3% and hence would be smaller than the ensemble fluctuation , @xmath148 , even for a local density of @xmath87 mpc@xmath88 , see fig . [ fig2 ] . we have seen that ensemble fluctuations of the uhe cr spectrum increase as we decrease the minimal distance @xmath82 to the extra - galactic source population , while keeping the local source density constant . this behavior is of course expected since the relative abundance of local sources is small @xmath149 and hence more susceptible for fluctuation , but with a higher flux weight @xmath150 compared to more distant populations . in fact , the ensemble variance formally diverges at _ all _ energies as we set @xmath151 . this is a well - known effect in studies of galactic ensemble fluctuations @xcite . the minimal distance @xmath82 hence serves here as a regulator . even for a moderate minimal distance of 10mpc and a source density of @xmath152 mpc@xmath88 the ensemble fluctuation can reach large values @xmath153 in the gzk region as indicated in the left column of fig . this is an indication that very few or just one local source can entirely dominate the spectrum at these energies . some authors have considered even the extreme case that a local source like the radio galaxy centaurus a at a distance of 3 - 4 mpc can be responsible for the _ entire _ uhe cr spectrum above the ankle @xcite . we have checked that for @xmath154 and @xmath87 mpc@xmath88 the ensemble fluctuation below @xmath155 ev are less than 10% and this particular ensemble realization of extra - galactic sources to appear by chance is hence unlikely within our setup . in addition , the missing anisotropy of uhe cr events challenges this interpretation . possible caveats to this result could be large ensemble - fluctuations of intrinsic source properties that have been omitted in this study ( but could easily be included ) and/or strong deflections in inter - galactic magnetic fields @xcite ( see further below ) . all calculations in this study consider the continuous and isotropic emission of cr sources . however , it is easy to extend the discussion for the case of episodal or anisotropic emission . if @xmath156 is the typical timescale of the source emission and @xmath157 the total experimental observation time we can define an effective local source density as @xmath158 this compensates the reduced continuous - equivalent emission rate @xmath159 of the single sources averaged over the duration of the experiment . note that the emission rate density @xmath160 that is fixed by the observed cr spectrum is independent of this rescaling . an analogous argument can be given for an anisotropic cr emitter with a preferred local ( but not global ) emission direction . this applies to jet - like source emission like in blazars or gamma - ray bursts . if the emission is concentrated in a cone @xmath161 we can again define a reduced effective local density @xmath162 that compensates the enhanced isotropic - equivalent emission rate @xmath163 . we can hence account for time - variable or anisotropic sources by rescaling the local source density . these two effects work in opposite directions . time variability increases the number of sources contributing to the spectrum and lowers the amplitude of the ensemble fluctuation by @xmath164 . on the other hand , anisotropy decreases the source number and increases the fluctuation by @xmath165 . note , however , that autocorrelation studies of uhe cr constrain the number of sources and limit the effective source density @xmath166 . hence , the lower limit of @xmath167 used in this analysis does still apply . in this context it is interesting to consider the effect of magnetic fields on the propagation of uhe crs . this has been so far neglected in this study . as mentioned before , a magnetic field ( regular or turbulent ) has no effect on a spatially homogeneous distribution of sources . however , we break homogeneity by introducing a minimal distance @xmath82 in our calculation . it was shown for the case of protons sources that diffusion in intergalactic magnetic fields can have a significant effect on the extra - galactic cr spectrum at lower energies . the limited distance to the closest source introduces a low energy suppression , sometimes called `` anti - gzk - cutoff '' @xcite . in this case , ensemble - fluctuations at the low energies close the cr ankle can increase dramatically beyond the adiabatic value ( [ eq : adisig ] ) considered in this paper . on top of these spectral variations there is also an effect on the the effective source density @xmath166 . with magnetic diffusion the time - scale @xmath156 and emission cone @xmath161 are expected to disperse . eventually , for strong diffusion the effective local density @xmath166 is simply the true density @xmath84 . since the diffusion coefficient does in general depend on the rigidity of the cr nuclei this effect would introduce an additional energy dependence of the variation . in conclusion , future space - based observatories with colossal exposures , @xmath143 , will provide the required large statistics at the high - energy end of the cr spectrum , allowing identification of ensemble fluctuation from the gzk suppression features on a statistical basis . in combination with information on the arrival - direction distribution of crs and on the secondary fluxes of @xmath4-rays and neutrinos these spectral features can provide a coherent picture for an _ indirect _ determination of the uhe cr nuclear composition @xcite and will naturally complement the current _ direct _ measurements of extensive air shower observables . we would like to thank tom paul for valuable comments on the manuscript . ma acknowledges support by a john bahcall fellowship for neutrino astronomy of the wisconsin icecube particle astrophysics center ( wipac ) and by the u.s . national science foundation ( nsf ) under grants opp-0236449 and phy-0236449 . laa is supported by the nsf under career grant phy-1053663 , the national aeronautics and space administration ( nasa ) grant 11-apra11 - 0058 , and the uwm research growth initiative . amt acknowledges support by a schroedinger fellowship at dublin institute for advanced study . any opinions , findings , and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of nasa or nsf . the boltzmann equations ( [ eq : diff0 ] ) describe the time - evolution of an isotropic flux of crs from a homogenous distribution of isotropic cr emitters . in the following we want to study the contribution of a single close - by ( @xmath81 ) source that is continuously emitting cr nuclei . as a first step we can hence neglect the redshift loss term @xmath168 in eqs . ( [ eq : diff0 ] ) and regard @xmath169 as the local density of cr particles . for the case of an inhomogeneous distribution of sources we have to add a cr convection term in eqs . ( [ eq : diff0 ] ) , that replaces the time - derivative for continuous emission . since we are not interested in the direction of the source we can imagine sitting at a center of a sphere with radius @xmath170 with surface emission rate spectrum @xmath171 @xcite . the convection term in this spherically symmetric setup reduces then to @xmath172 . between an initial configuration ( blue dot ) and a final configuration ( magenta dot ) including one - nucleon losses ( red solid arrows ) , two - nucleon losses ( red dotted arrows ) and continuous energy loss ( green horizontal arrows ) . for the exact analytic solution ( [ eq : fullsol ] ) all possible transition chains of this type are taken into account . ( figure adopted from ref . @xcite.),height=192 ] the secondary nuclei produced via photodisintegration carry approximately the same lorentz factor as the initial nucleus . as already discussed in section [ sec : variation ] it is hence convenient to express the energy of a nucleus with mass number @xmath48 as @xmath173 where @xmath174 denotes the energy _ per nucleon_. the differential interaction rate in the set of eqs . ( [ eq : diff0 ] ) corresponding to the production of a nucleus with mass number @xmath72 and energy @xmath175 from a nucleus with mass number @xmath176 and energy @xmath177 can be approximated as @xmath178 where @xmath179 is the partial width of the transition . finally , introducing the binned cr flux @xmath63 , and corresponding emission rates , @xmath180 we can describe the point - source flux as a solution of the compact set of equations @xmath181 where we define the rates : @xmath182 note that eqs . ( [ eq : diff1 ] ) holds for nuclei heavier than beryllium . we can easily compensate for the process @xmath183be @xmath184 @xmath185he + @xmath185he + n of the psb chain by re - defining @xmath186 for @xmath187 and @xmath188 for other nuclei . similarly , the chains with nucleons as a final particle are re - weighted by the corresponding multiplicity in the case of intermediate nuclei lighter than @xmath183be ( see ref . @xcite for further details ) . in ref . @xcite it was shown that the general solution of eq . ( [ eq : diff1 ] ) can be written in the form of eq . ( [ eq : fullsol ] ) . the first sum in eq . ( [ eq : fullsol ] ) runs over all possible production chains @xmath68 of nuclei with mass @xmath48 in the nucleon energy bin @xmath14 . the intermediate configurations @xmath70 of this chain are parametrized by the tuple @xmath71 denoting the mass number @xmath72 and nucleon energy bin @xmath73 . the links of the chain correspond to the partial width @xmath189 . the dimensionless amplitudes in eq . ( [ eq : fullsol ] ) are defined as @xmath190 the total width @xmath191 is simply the total rate including photo - disintegration and cel terms . these amplitudes satisfy the identity @xmath192 independent of the path @xmath68 ( see refs . @xcite ) . an example of a chain is shown in fig . the elements @xmath70 of the chain correspond to intermediate configurations of mass number and energy bin on a grid . the first element @xmath74 ( blue dot ) corresponds to the cr particle emitted at the source ; the last element @xmath193 ( magenta dot ) is the observed ( final ) configuration @xmath194 . the example shows one - nucleon disintegration ( solid red line ) , two - nucleon disintegration ( dashed red line ) and continuous energy loss ( green lines ) . in general , the number of possible paths is quite large and the computation of eq . ( [ eq : fullsol ] ) numerically expensive . for the calculation in this study we have applied two more approximations to the solution ( [ eq : fullsol ] ) to reduce the number of terms . firstly , we reduced the cel ( last two terms on the r.h.s . ( [ eq : diff1 ] ) ) to the effective loss term @xmath195 this introduces a relative error of the order @xmath196 , which is small in the relevant energy region of @xmath197 ev . and , secondly , we considered only the one - nucleon loss chain of the psb approximation . it was shown in refs . @xcite that this is a good approximation for cr nuclei with a large mass number , which is the focus of this analysis . j. linsley , _ proceedings of icrc 1963 , jaipur , india _ , ( commercial printing press , ltd . , bombay , india , 1964 ) , pp . c. t. hill and d. n. schramm , phys . d * 31 * , 564 ( 1985 ) . m. hillas , phys . a * 24 * , 677 ( 1967 ) . v. berezinsky , a. z. gazizov and s. i. grigorieva , phys . d * 74 * , 043005 ( 2006 ) [ arxiv : hep - ph/0204357 ] . k. greisen , phys . rev . lett . * 16 * , 748 ( 1966 ) . g. t. zatsepin and v. a. kuzmin , jetp lett . * 4 * , 78 ( 1966 ) [ pisma zh . * 4 * , 114 ( 1966 ) ] . r. abbasi _ et al . _ [ hires collaboration ] , phys . lett . * 100 * , 101101 ( 2008 ) [ arxiv : astro - ph/0703099 ] . j. abraham _ et al . _ [ pierre auger collaboration ] , phys . rev . * 101 * , 061101 ( 2008 ) [ arxiv:0806.4302 [ astro - ph ] ] . j. abraham _ et al . _ [ pierre auger collaboration ] , phys . b * 685 * , 239 ( 2010 ) [ arxiv:1002.1975 [ astro-ph.he ] ] . m. ahlers , l. a. anchordoqui and s. sarkar , phys . d * 79 * , 083009 ( 2009 ) [ arxiv:0902.3993 [ astro-ph.he ] ] . d. hooper , a. m. taylor and s. sarkar , astropart . phys . * 34 * , 340 ( 2011 ) [ arxiv:1007.1306 [ astro-ph.he ] ] . j. abraham _ et al . _ [ pierre auger collaboration ] , nucl . instrum . a * 613 * , 29 ( 2010 ) [ arxiv:1111.6764 [ astro-ph.im ] ] . t. abu - zayyad _ et al . _ [ telescope array collaboration ] , nucl . instrum . a * 689 * , 87 ( 2012 ) [ arxiv:1201.4964 [ astro-ph.im ] ] . j. h. adams jr . _ _ , [ arxiv:1203.3451 [ astro-ph.im ] ] . a. m. taylor , m. ahlers and f. a. aharonian , phys . d * 84 * , 105007 ( 2011 ) [ arxiv:1107.2055 [ astro-ph.he ] ] . j. miralda - escude and e. waxman , astrophys . j. * 462 * , l59 ( 1996 ) [ arxiv : astro - ph/9601012 ] . l. a. anchordoqui , m. t. dova , l. n. epele and j. d. swain , phys . d * 57 * , 7103 ( 1998 ) [ arxiv : astro - ph/9708082 ] . g. a. medina - tanco , astrophys . j. * 510 * , l91 ( 1999 ) [ arxiv : astro - ph/9810366 ] . r. aloisio , d. boncioli and d. boncioli , astropart . phys . * 35 * , 152 ( 2011 ) [ arxiv:1002.4134 [ astro-ph.he ] ] . d. de marco , p. blasi and a. v. olinto , astropart . * 20 * , 53 ( 2003 ) [ arxiv : astro - ph/0301497 ] . d. hooper , s. sarkar and a. m. taylor , phys . d * 77 * , 103007 ( 2008 ) [ arxiv:0802.1538 [ astro - ph ] ] . m. ahlers and a. m. taylor , phys . d * 82 * , 123005 ( 2010 ) [ arxiv:1010.3019 [ astro-ph.he ] ] . j. beringer _ [ particle data group ] , phys . d * 86 * , 010001 ( 2012 ) . f. w. stecker , phys . * 180 * , 1264 ( 1969 ) . j. l. puget , f. w. stecker and j. h. bredekamp , astrophys . j. * 205 * , 638 ( 1976 ) . l. n. epele and e. roulet , jhep * 9810 * , 009 ( 1998 ) [ arxiv : astro - ph/9808104 ] . j. p. rachen , _ interaction processes and statistical properties of the propagation of cosmic - rays in photon backgrounds _ , phd thesis of the bonn university , 1996 . e. khan _ et al . _ , astropart . * 23 * , 191 ( 2005 ) [ arxiv : astro - ph/0412109 ] . d. allard , e. parizot , e. khan , s. goriely and a. v. olinto , astron . astrophys . * 443 * , l29 ( 2005 ) [ arxiv : astro - ph/0505566 ] . s. goriely , s. hilaire and a. j. koning , astron . astrophys . * 487 * , 767 ( 2008 ) [ arxiv:0806.2239 [ astro - ph ] ] . g. r. blumenthal , phys . d * 1 * , 1596 ( 1970 ) . a. franceschini , g. rodighiero and m. vaccari , astron . astrophys . * 487 * , 837 ( 2008 ) [ arxiv:0805.1841 [ astro - ph ] ] . kronberg , rept . phys . * 57 * , 325 ( 1994 ) . r. aloisio and v. s. berezinsky , astrophys . j. * 625 * , 249 ( 2005 ) [ arxiv : astro - ph/0412578 ] . d. harari , s. mollerach and e. roulet , jhep * 9908 * , 022 ( 1999 ) [ arxiv : astro - ph/9906309 ] . t. abu - zayyad _ et al . _ [ telescope array collaboration ] , [ arxiv:1205.5067 [ astro-ph.he ] ] . e. waxman , k. b. fisher and t. piran , astrophys . j. * 483 * , 1 ( 1997 ) [ arxiv : astro - ph/9604005 ] . t. kashti and e. waxman , jcap * 0805 * , 006 ( 2008 ) [ arxiv:0801.4516 [ astro - ph ] ] . h. takami , s. inoue and t. yamamoto , astropart . phys . * 35 * , 767 ( 2012 ) [ arxiv:1202.2874 [ astro-ph.he ] ] . m. a. lee , astrophys . j. * 229 * , 424 ( 1979 ) . a. m. hopkins and j. f. beacom , astrophys . j. * 651 * , 142 ( 2006 ) [ arxiv : astro - ph/0601463 ] . h. yuksel , m. d. kistler , j. f. beacom and a. m. hopkins , astrophys . j. * 683 * , l5 ( 2008 ) [ arxiv:0804.4008 [ astro - ph ] ] . d. j. bird , astrophys . j. * 441 * , 144 ( 1995 ) [ arxiv : astro - ph/9410067 ] . l. a. anchordoqui , g. e. romero and j. a. combi , phys . d * 60 * , 103001 ( 1999 ) [ arxiv : astro - ph/9903145 ] . p. blasi , r. i. epstein and a. v. olinto , astrophys . j. * 533 * , l123 ( 2000 ) [ arxiv : astro - ph/9912240 ] . k. fang , k. kotera and a. v. olinto , astrophys . j. * 750 * , 118 ( 2012 ) [ arxiv:1201.5197 [ astro-ph.he ] ] . a. santangelo and a. petrolini , new j. phys . * 11 * , 065010 ( 2009 ) [ arxiv:0909.5370 [ astro-ph.he ] ] . p. abreu _ et al . _ [ pierre auger collaboration ] , [ arxiv:1107.4809 [ astro-ph.he ] ] . p. blasi and e. amato , jcap * 1201 * , 010 ( 2012 ) [ arxiv:1105.4521 [ astro-ph.he ] ] . g. bernard , t. delahaye , p. salati and r. taillet , astron . astrophys . * 544 * , a92 ( 2012 ) [ arxiv:1204.6289 [ astro-ph.he ] ] . p. l. biermann and v. de souza , astrophys . j. * 746 * , 72 ( 2012 ) [ arxiv:1106.0625 [ astro-ph.he ] ] . l. a. anchordoqui , h. goldberg and t. j. weiler , phys . d * 84 * , 067301 ( 2011 ) [ arxiv:1103.0536 [ astro-ph.he ] ] . h. yuksel , t. stanev , m. d. kistler and p. p. kronberg , astrophys . j. * 758 * , 16 ( 2012 ) [ arxiv:1203.3197 [ astro-ph.he ] ] . m. ahlers , l. a. anchordoqui and a. m. taylor , in preparation

devices in which organic molecules form the active element are taking an increasingly important place in solid state physics . these are the building blocks of the field of molecular electronics , whose potential applications underpin a vast number of technological areas . novel components for high - performance computers @xcite , highly sensitive chemical sensors @xcite , and disposable and wearable devices are among the many proposed applications . in addition the same set - up can be used for medical purposes , for instance in the detection of viruses @xcite or in the construction of a rapid and reliable protocol for dna sequencing @xcite . most of the progress originates from the recently achieved ability to construct single molecule junctions and to measure their transport properties . the experimental strategies for fabricating such devices include mechanically controllable break junctions @xcite , scanning tunneling spectroscopy @xcite , lithographically fabricated nanoelectrodes @xcite and colloid solutions @xcite . unfortunately most of these techniques are actually `` blind '' in the sense that the geometrical configuration of the device at the atomic scale is not known . for this reason it should not be surprising that different experiments for the same molecule yield conductances differing by of up to three orders of magnitude @xcite . in view of all these uncertainties _ ab initio _ quantum transport schemes for calculating the electrical response of a molecular device become crucial . the most common computational scheme for evaluating the electronic transport of a molecular device combines scattering theory in the form of the non - equilibrium green s function ( negf ) formalism @xcite , with an electronic structure method , the most widely used being density - functional theory ( dft ) @xcite . alternatives are time - dependent dft @xcite or many - body methods @xcite , which at present however are more computationally involved and are at an earlier stage of development as routinely used schemes . unfortunately for several molecules negf combined with dft appears to disagree with the experimental results . in particular it seems to systematically overestimate the current flowing across a device , even if one bears in mind the uncertainty over the detailed geometry of the junction . in the prototypical case of benzenedithiol ( bdt ) sandwiched between two gold electrodes , the conductance for the most probable contact geometry calculated with negf and dft is higher than that of _ any _ of the experiments by at least one order of magnitude @xcite . in this work we demonstrate that most of the errors originate from using local or semi - local exchange and correlation potentials in dft . this shortfall however can be corrected by applying an atomic self - interaction correction ( asic)@xcite scheme to the local functionals , leading to a dramatic improvement of the agreement between the calculated results and those obtained from experiments . the reason for this improvement is rooted in the ability of asic to predict the correct ionization potential of molecules , and hence to correctly describe the band alignment between the molecular orbitals and the fermi level , @xmath2 , of the metallic electrodes . here we extend the results previously published for bdt attached to the gold hollow site of the ( 111 ) surface @xcite to new molecules such as benzenedimethanethiol ( bdmt ) and biphenyldithiol ( bpd ) . we also carry out a thorough investigation of the effects of changing the anchoring geometry between the molecule and the electrodes , demonstrating that , for the case of bdt attached to gold , the asic calculated @xmath0-@xmath1 curves are rather stable with respect to geometry changes . this explains the relatively narrow distribution of conductance histograms found in experiments @xcite . the problem we wish to solve is that of calculating the two - probe @xmath0-@xmath1 curve of a molecule sandwiched in between two metallic electrodes . the negf scheme partitions such system into three regions , respectively the two current / voltage electrodes ( leads ) and a middle region called the scattering region ( sr ) ( see figure [ fig1 ] ) . the sr includes the molecule and a portion of the leads , which is extended enough for the charge density calculated at the most external atomic layer to resemble that of the bulk electrodes . the leads , assumed to be a periodic crystal in the direction of the transport , are kept at different chemical potentials @xmath3 where @xmath1 is the applied potential bias and @xmath4 is the electron charge . the sr is described by a hamiltonian @xmath5 , which is used to construct the non - equilibrium green s function , @xmath6 @xmath7^{-1}\ : , \label{negfmx}\ ] ] where @xmath8 are the self - energies for the leads . @xmath6 enters in a self - consistent procedure to calculate the density matrix , @xmath9 , of the sr , and hence the two probe - current of the device @xcite . the non - equilibrium @xmath9 is calculated following the negf prescription as @xmath10 g^{\dag}(e ) de \ : , \label{chdens}\ ] ] where @xmath11 $ ] . in practice , this integral is performed by splitting it into two parts@xcite : an equilibrium part which can be integrated along a contour in the complex plane , and a non - equilibrium part which has to be integrated along the real energy axis but which only contributes around @xmath2 . the non - equilibrium charge density @xmath9 is then used to calculate a new hamiltonian for the scattering region @xmath12 $ ] ( where it is assumed that the hamiltonian has some functional dependence on @xmath9 ) . this procedure is repeated self - consistently until the density matrix converges . finally , the converged green s function can be used to calculate the current @xmath0 through the device : @xmath13 ( f(e , \mu_\mathrm{1 } ) - f(e , \mu_\mathrm{2 } ) ) de \ : . \label{curint}\ ] ] this is effectively the integral between @xmath14 and @xmath15 ( the bias window ) of the transmission coefficients @xmath16 $ ] . in brief the negf scheme calculates the non - equilibrium scattering potential of the device . thus the transmission coefficient @xmath17 is simply a superposition of resonances located at the molecular single - particle energy levels , including a possible shift and broadening due to the interaction with the leads . it follows that the molecular levels close to the electrode fermi level , i.e. the highest highest occupied molecular orbital ( homo ) and the lowest unoccupied molecular orbital ( lumo ) , provide the dominant contribution to the current . it is therefore crucial to employ an electronic structure theory capable of describing the position of these two levels accurately . the negf scheme is general and not related to a particular electronic structure theory . however , since the transmission occurs through single - particle states , the associated electronic structure theory should meet several requirements @xcite . first of all , the single - particle levels must closely resemble the physical removal energies of the system , i.e. the energy levels of the molecule should line up correctly with the @xmath2 of the metals forming the electrodes . secondly , the theory should work at both integer and fractional occupation , ensuring the correct response of the molecular orbitals to changes in occupation due to the applied potential bias . finally , the matrix elements describing the interaction between the leads and the molecule should be calculated accurately avoiding the use of phenomenological parameters or fitting procedures . the kohn - sham ( ks ) form @xcite of dft is certainly the most widely used _ ab initio _ electronic structure theory associated with the negf method . in the ks scheme the dft problem of finding the ground state charge density is mapped onto that of solving a system of non - interacting single - particle schr " odinger equations @xmath18 \psi^{\sigma}_n({\bf r } ) = \epsilon^{\sigma}_n \psi^{\sigma}_n({\bf r } ) \ : , \label{kseqn}\ ] ] where @xmath19 is the hartree ( classical coulomb ) potential and @xmath20 is the exchange - correlation ( xc ) potential . the exact form of @xmath20 is unknown , so it is typically approximated by local functionals such as the local density approximation @xcite ( lda ) or the generalized gradient approximation @xcite ( gga ) . therefore the ks scheme is only a convenient prescription for minimizing the energy functional , and as such the individual ks eigenvalues do not necessarily correspond to physical energy levels of the system . the dft observables such as the total energy and the charge density are in fact only integral quantities respectively of the ks eigenvalues and eigenvectors . however , in solid state theory it is common to associate the ks energies with the system band structure . this , although it is not justified at the fundamental level , is supported by the practical evidence that ks bands represent a good first approximation of the true energies of a system , particularly for metals . however , there is a remarkable exception to the non - physical nature of the ks eigenvalues . in fact , the ks energy of the homo level ( @xmath21 ) can be rigorously associated with the negative of the ionization potential @xmath0 @xcite . similarly the homo ks energy for the negatively charged system can be associated with the chemical affinity @xmath22 . this suggests that , at least for moderate bias where the transport is through the homo , ks theory can be used effectively for transport calculations @xcite . unfortunately , standard local and semi - local functionals completely misplace @xmath21 , which is often nowhere near @xmath23 . this is particularly problematic for organic molecules , for which the lda @xmath21 is typically off from the experimental value of @xmath0 by approximately 4 ev @xcite . most of the failures of lda and gga can be traced back to the self - interaction ( si ) problem , i.e. the spurious interaction of an electron with the hartree and exchange - correlation ( xc ) potentials generated by itself @xcite . in the case of hartree - fock theory the self - hartree energy @xmath24 $ ] cancels exactly the self - xc energy @xmath25 $ ] @xmath26 + u[\rho^{\sigma}_n ] = 0 \ : , \label{exxhar}\ ] ] where we introduce the orbital density @xmath27 . however , this cancellation is incomplete for both lda and gga . the resulting ks potential therefore appears too repulsive and the eigenstates of a molecule are pushed to higher energies . when this is transferred to the transport problem , the position of the peaks in the transmission coefficient results are in the wrong place . in particular it is likely that peaks will erroneously move close to the fermi level of the electrodes , generating a large conductivity at zero bias . since the position of the peaks is ultimately determined by the interplay between the hopping probabilities from the electrodes to the molecule with the charging energy of the molecule itself , this failure is somewhat reminiscent of the inability of local and semi - local functionals to describe mott - hubbard insulators . perdew and zunger suggested @xcite that the self - interaction correction ( sic ) @xmath28 for an occupied ks orbital @xmath29 can be calculated from the sum of the self - hartree and self - xc energies @xmath30 + u[\rho^{\sigma}_n ] \ : . \label{siclda}\ ] ] note that , although here we consider the case of sic - lda only , the same procedure can be readily applied to any approximated functionals . the sic - lda xc energy @xmath31 is thus obtained by subtraction @xmath32 = e^\mathrm{lda}_\mathrm{xc}[\rho^{\uparrow } , \rho^{\downarrow } ] - \sum^\mathrm{occ.}_{n \sigma } \delta^\mathrm{sic}_n \ : , \label{sicen}\ ] ] and the sic - lda xc potential @xmath33 , to be subtracted from @xmath34 , is simply given by @xmath35 ; { \bf r})+v_\mathrm{xc}^{\mathrm{lda},\sigma}([\rho_n^\uparrow,0 ] ; { \bf r } ) \ ; , \label{sicpot}\end{aligned}\ ] ] with @xmath36 ; { \bf r})=\int\mathrm{d}^3{\bf r}^\prime\frac{\rho({\bf r}^\prime)}{|{\bf r}-{\bf r}^\prime|}\ ; , \\ v_\mathrm{xc}^{\mathrm{lda},\sigma}([\rho^\uparrow,\rho^\downarrow ] ; { \bf r})=\frac{\delta}{\delta \rho^\sigma({\bf r } ) } e_\mathrm{xc}^\mathrm{lda}[\rho^\uparrow,\rho^\downarrow]\;.\end{aligned}\ ] ] the problem of finding the energy minimum is complicated by the fact that @xmath31 is not invariant under unitary rotations of the occupied ks orbitals , which instead leave @xmath9 invariant . this creates problem to the standard ks scheme since the theory becomes size - inconsistent . in order to avoid such a complication modern sic theory introduces a second set of orbitals @xmath37 related to the canonical ks orbitals @xmath29 by a unitary transformation @xmath38 @xmath39 the functional can then be minimized by varying both the ks orbitals and the unitary transformation @xmath38 , leading to the system of ks - like equations : @xmath40 \psi^{\sigma}_n({\bf r } ) = \epsilon^{\sigma , \mathrm{sic}}_n \psi^{\sigma}_n({\bf r } ) \ : , \label{siceqn}\ ] ] where @xmath41 is the lda hamiltonian , and the sic potential , @xmath42 , is given by : @xmath43 where @xmath44 is the projector @xmath45 . the numerical implementation of the full perdew - zunger sic scheme in typical solid state codes is cumbersome since the theory is orbital dependent and the energy minimization can not follow the standard ks scheme @xcite . when this is applied to transport there is the additional complication that the ks orbitals are never individually available and moreover that one has to deal with an intrinsically non - local potential . a drastic simplification of the problem can be obtained by using a recently implemented atomic approximation to the sic scheme , which we call the asic @xcite . this is based on the pseudo - sic approximation , originally proposed by vogel and coworkers @xcite , and later extended to non - integer occupation by filippetti and spaldin @xcite . the main idea is that of replacing the @xmath46 orbitals with atomic like functions , which are not calculated self - consistently . thus the si correction becomes atomic - like and no information is needed other than the charge density and the asic projectors @xcite . asic has been demonstrated to produce single - particle energy levels which match the experimental molecular removal energies quite well @xcite . in particular , for several different molecules investigated , @xmath47 , and the @xmath21 for negatively charged molecules is close to the molecular affinity @xcite . as an example , asic places @xmath21 for 1,2-bdt at -8.47 ev to compare with the lda value of -4.89 ev and the experimental ionization potential @xmath488.5 ev @xcite . asic is the scheme that we adopt in this work . local and semi - local functionals are affected by another fundamental problem , i.e. the lack of the derivative discontinuity ( dd ) in the dft energy . this is the discontinuity at integer occupation in the derivative of the total energy of a system @xmath49 with respect to its occupation @xmath50 @xcite . from janak s theorem it follows that there should be also a discontinuity in @xmath51 when going from @xmath50 to ( @xmath52 ) with @xmath53 . in a previous work @xcite , we have investigated the consequences of the absence of the dd in local xc functionals on the electronic transport of organic molecules . we have demonstrated that the dd can strongly affect the @xmath0-@xmath1 characteristics of metal - molecule - metal junctions when the coupling between the molecule and the metal is relatively weak . this would be the case , for instance , where molecules bind to adatoms on the metal surface . however , we have also shown that in the case of strong coupling the dd had little effect on the transport . the self - interaction error is largely responsible for the disappearance of the dd in local and semi - local functionals . it follows that sic methods restore the dd at least in part . this last aspect is deeply rooted in the orbital minimization needed to extract the sic potential , which unfortunately is not captured by our asic approximation . therefore the main feature of our asic methodology is simply the correction of the ionization potentials of the molecules @xcite , thus yielding a quantitatively realistic energy level alignment . hence , the calculations described in this paper investigate a second important aspect of the sic beside that of the dd . this is the quantitative description of the band alignment between the metal fermi energy and the molecular levels . in particular , for most of the paper we investigate molecules attached to the gold fcc ( 111 ) hollow sites , for which the coupling is expected to be strong and the effects due to the dd small . we have numerically implemented the asic method @xcite in the localized atomic orbital code siesta @xcite , which is the dft platform for our transport code _ smeagol _ @xcite , and carried out calculations for the prototypical au / benzene / au molecular devices . the asic corrections are not applied to the gold atoms , as si errors for metals are small @xcite . unless otherwise specified , we use a double zeta polarized basis set @xcite for carbon and sulfur @xmath54 and @xmath55 orbitals , double zeta for the 1@xmath54 orbitals of hydrogen and 6@xmath54-only double zeta for gold . the mesh cut - off is 200 ry and we consider 500 real and 80 complex energy points for integrating the green s function . calculations were also performed using double zeta 6@xmath54 and single zeta 5@xmath56 and 6@xmath55 orbitals for the gold to investigate the effect of an enlarged basis on the transport . results for calculations using both basis sets are shown , with the 6@xmath54-only basis used unless otherwise indicated . for geometry optimizations and total energy calculations , the extended 5@xmath566@xmath546@xmath55 basis set is employed , as the 6@xmath54-only does not yield enough accuracy . in calculating the @xmath0-@xmath1 curves , the potential bias was incremented in steps of 0.1 volts . we compare the electronic transport properties for three different molecules calculated either by using lda or asic . these are 1,4-benzenedithiol ( 1,4-bdt ) , benzenedimethanethiol ( bdmt ) , and biphenyldithiol ( bpd ) . the electronic transport through these molecules have already been investigated at length both experimentally and computationally @xcite . thus these offer us a unique benchmark for our calculations . in all cases , the molecule is attached via the sulfur atoms to fcc ( 111 ) gold electrodes on each side . in the case of bdt , we look at several different anchoring geometries . it is worth pointing out that although the exact anchoring geometry encountered in breaking junction experiments is unknown , the resulting conductance histograms are relatively narrow suggesting an intrinsic stability in the measurements @xcite . this may be an indication that similar anchoring configuration are highly probable in the breaking process , or alternatively that several anchoring geometries yield the same @xmath0-@xmath1 characteristic . one of our goals is to distinguish between these two possibilities . the first device which we consider is 1,4-benzenedithiol ( 1,4-bdt ) attached to two fcc ( 111 ) gold surfaces . the anchoring geometries for the sulfur atoms investigated are the hollow site , the top site , the bridge site , ( see figure [ fig3 ] ) , as well as asymmetric configurations where the s ions are attached to an adatom on one side and to the hollow site on the other . we also examine the effect of altering the angle which the molecule makes with the metal surface , and of varying the distance between the sulfur atom and the surface ( i.e. varying the strength of the coupling between the molecule and the metal ) . finally , we investigate the effect of hydrogenating the thiol groups . as explained previously the actual experimental contact geometry is unknown . electronic structure calculations indicate that the lowest energy configuration for the molecule on the ( 111 ) surface occurs when the s atom attaches to the hollow site , although other calculations suggest that the bridge site configuration has a lower energy @xcite . recent x - ray standing wave measurements suggest that molecules in monolayers prefer to attach to adatoms on the metal surface @xcite . it is also worth noticing that in breaking junctions the actual geometry might be different from any equilibrium geometries , since the system is likely to be under strain . hence , a thorough analysis of this system should comprise many different configurations . we start by attaching bdt at the au ( 111 ) hollow site ( see figures [ fig3 ] and [ fig4 ] ) . the distance of the sulfur atom from the plane of the gold surface ( the `` sulfur - surface separation '' , @xmath57 , shown in figure [ fig3]a ) is optimized ( using the 5@xmath566@xmath546@xmath55 basis set for gold ) to a value of 1.9 . this corresponds to a distance of 2.53 between the sulfur atoms and the nearest gold atoms on the surface in good agreement with previous calculations @xcite . the @xmath0-@xmath1 characteristic , the orbital resolved density of states ( dos ) and the conductance as a function of energy , @xmath58 , are presented in figure [ fig5 ] . the conductance is plotted in units of @xmath59 , so that it is equivalent to the transmission coefficients @xmath60 . lda results are shown in the left panels , while the asic ones are on the right . from the dos it is clear that the effect of asic is that of shifting the occupied orbitals downwards in energy relatively to the @xmath2 of gold . the asic homo - lumo gap is considerably larger than that calculated with lda and most importantly in the case of asic there is little dos originating from the molecule at @xmath2 . this has profound effects on the electron transmission . the lda peaks of @xmath17 arising from occupied orbitals are shifted downwards in energy and away from @xmath2 . at variance from lda ( figure [ fig5]c ) , where @xmath61 is dominated by a resonance at @xmath62 , the asic transmission ( figure [ fig5]d ) is through the bdt gap and therefore it is tunneling - like . this results in a drastic reduction of the low - bias current when going from lda to asic ( figure [ fig5]e ) . the asic - calculated conductance at zero bias is now about 0.06@xmath63 ( @xmath64 ) , compared to 0.23@xmath63 for lda . a conductance of 0.06@xmath63 is much closer to the value of 0.011@xmath63 obtained by xiao et . @xcite and is actually lower than values 0.09 - 0.14@xmath63 obtained by tsutsui et . @xcite . without altering the anchoring geometry , the basis set on the gold atoms is then changed to include the 5@xmath56 and 6@xmath55 orbitals . the relative orbital resolved dos , transmission coefficients and @xmath0-@xmath1 curves are presented in figure [ fig6 ] for both lda and asic . clearly the enriched basis set does not have a large effect on the electronic transport , particularly at low bias . as can be seen from figure [ fig6]e , the @xmath0-@xmath1 curves calculated with the 6@xmath54-only basis set for gold are approximately the same as that calculated with the 5@xmath566@xmath546@xmath55 basis set up to about 1 volt for both the lda and asic . differences appear only at high bias ( @xmath65 volt ) and are due to the presence of au @xmath56 electrons in an energy range comprising the bias window . however the zero - bias conductances , from the transmission plots in panels ( c ) and ( d ) of figure [ fig6 ] , are very similar to those for the 6@xmath54-only basis set with a value of 0.21@xmath63 for lda and a value of 0.06@xmath63 for asic . this demonstrates that the 6@xmath54-only basis set should give reasonably reliable results for electronic transport at low bias . next , the transport properties of the system are calculated for the hollow site at different @xmath57 , as well as for the equilibrium distance @xmath66 but different angles of the molecule with respect to the direction of transport . the results are reported in figure [ fig7 ] . in general asic seems to be much less sensitive to changes in the anchoring geometry than lda , particularly for the case of bond stretching . for instance , the zero bias conductance values calculated with lda are 0.16 @xmath63 , 0.23 @xmath63 , 0.32 @xmath63 and 0.77 @xmath63 for @xmath57 values of 1.8 , 1.9 , 2.1 and 2.5 respectively . for the same @xmath57 asic returns 0.05 @xmath63 , 0.06 @xmath63 , 0.07 @xmath63 , and 0.14 @xmath63 . the stability of the asic @xmath0-@xmath1 curve with respect the details of the anchoring geometry of the hollow site is an interesting feature , since it is suggesting that in breaking junction experiments several different anchoring configurations may yield similar @xmath0-@xmath1 curves . this is consistent with the relatively narrow peaks measured in the typical conductance histograms @xcite . lda does not seem to have this property and in fact the zero - bias conductance increases quite drastically when @xmath57 is increased . while counterintuitive , this is consistent with previous results @xcite , and it is due to the realignment of the homo of the molecule to @xmath2 of gold . such a feature is investigated in more detail in figure [ fig8 ] where we report the zero - bias @xmath17 for different @xmath57 . when @xmath57 is increased , the transmission peaks corresponding to molecular orbitals become narrower , as expected , due to the weakened molecule - lead coupling . however , charge transfer between the molecule and the metal is also affected , so that the charge on the actual molecule slightly increases as @xmath57 gets larger . this extra charge produces an electrostatic shift of the homo level towards the au fermi energy . as a consequence the narrowing of the resonance peak at the homo level is compensated by the upwards shift of the same peak , resulting in an enhancement of the zero - bias conductance . when asic is used this feature is strongly suppressed . in fact , the asic calculated transmission peaks are narrower and further away from @xmath2 that their lda counterparts . thus , although also for asic the molecular homo realigns with respect to the au fermi level , this is is not enough to produce high transmission at reasonable @xmath57 . note that a similar realignment is expected for weaker binding anchoring geometries which we will investigate in the remainder of this section ( see figures [ fig9]b , [ fig11 ] , [ fig12 ] and [ fig14 ] ) the second contact geometry investigated is that where the s atom is connected to the bridge site on the fcc ( 111 ) au surface at both of the electrodes ( figure [ fig3]b ) . total energy dft calculations suggest that this configuration ( @xmath67 ) , has a lower energy than that of the hollow site @xcite . we calculate a lda zero bias conductance for the bridge site of 0.1 @xmath63 , which is lower than the value of 0.23 @xmath63 obtained for the hollow site . this can be seen from the @xmath0-@xmath1 curves presented in figure [ fig9]a . interestingly this lower conductance found for the bridge site with respect to the hollow site is a low - bias feature and the two @xmath0-@xmath1 curves matches closely for @xmath68 volt . the effect of asic on the transmission of the bridge site configuration is rather similar to that on the hollow site . also in this case the molecular homo - lumo gap opens and the current gets suppressed . the zero bias conductance for the bridge site is calculated to be 0.06 @xmath63 , the same value found for the hollow site . hence , whether the molecule is anchored to the hollow site or to the bridge site makes relatively little difference to the low - bias transport properties as obtained with asic . this further confirms that our asic calculations are certainly more compatible than their lda counterparts with the relative stability of the experimental conductance histograms . the top site of the au fcc ( 111 ) surface ( figure [ fig3]b ) is the next anchoring site to be investigated . the equilibrium distance @xmath57 is 2.39 and the calculated @xmath0-@xmath1 curves are presented in figure [ fig9]b . this time both lda and asic present @xmath0-@xmath1 characteristics with much higher conductance than that associated with the hollow site . this is particularly dramatic for lda for which the zero - bias conductance goes from 0.23 @xmath63 for the hollow to 0.65 @xmath63 for the top site . for asic the increase is only twofold with the zero bias conductance going from 0.06 @xmath63 to 0.12 @xmath63 . also for this case the general increase in current can be correlated with the larger au - s bond - length of the top site compared to the hollow site . this has the same effect found when analyzing the hollow site under strain and it is due to the realignment of the homo of bdt to @xmath2 of gold . it is important to point out that the top site is less energetically favorable than both the bridge and the hollow sites . although it can still play a role in breaking junctions , we believe that this may be only of secondary importance and therefore would not have a considerable influence on the typical conduction histograms . in actual breaking junctions , the electrodes are under strain and so the thiol groups are less likely to attach to a regular surface site . for instance in figure [ fig10 ] we present the case where one electrode connects to the molecule via the hollow site ( @xmath66 ) and the other via a gold adatom , with a distance of 2.39 between the sulfur and the adatom . recent x - ray standing wave experiments @xcite demonstrate that gold adatoms may be the most favorable sites for the thiol groups in self - assembled molecular monolayer . this therefore may be a likely configuration for the bottom electrode in a stm - break junction . the orbital resolved dos , transmission coefficients and @xmath0-@xmath1 curves for this system are presented in figure [ fig11 ] . this configuration shows the largest difference between the conductance calculated with lda and that calculated with asic . the lda conductance is 0.32 @xmath63 , whereas when asic is applied this drops by a order of magnitude to 0.03 @xmath63 . because of the weaker interaction between the @xmath69 orbitals of the molecule and the gold surface , the molecular orbitals in the dos ( figure [ fig11]a and figure [ fig11]b ) and hence the peaks in the transmission coefficients ( figure [ fig11]c and figure [ fig11]d ) are considerably narrower than for the case of the hollow site . these narrow levels are however closer to @xmath2 , resulting in the relatively high conductance at low bias . this originates from the same mechanism producing a change in conductance when @xmath57 is increased ( see figure [ fig8 ] ) . at variance with the case of stretched hollow bonding geometry , this time the asic does not shift the conductance peaks enough to bring them close to the fermi level of gold , and the conductance remains rather small . as the peaks in the transmission are rather narrow we checked our results against the choice of basis set . in figure [ fig12 ] we present the orbital resolved dos , transmission coefficients and @xmath0-@xmath1 curves for the enlarged 5@xmath566@xmath546@xmath55 basis , to be put in perspective with those calculated with 6@xmath54 only in figure [ fig11 ] . the most notable difference is the appearance of transmission away from @xmath2 , in particular at low energies . this is connected to the gold @xmath56 electrons , which are neglected in the small basis . importantly however the low - energy features around @xmath2 are only slightly affected with the zero - bias conductance assuming values of 0.47 @xmath63 and 0.05@xmath70 for lda and asic respectively . this confirms the feasibility of using a restricted basis set for the @xmath0-@xmath1 curve at low bias . finally , we investigate two more configurations . the first considers adatoms as the bonding site at both sides of the junction ( figure [ fig13 ] ) , while the second investigates hollow sites where the two hydrogens of the thiol groups are not dissociated in forming thiolate as in all the other cases studied ( figure [ fig15 ] ) . in the first case we expect a rather weak bond , which is confirmed by the sharp peaks in transmission shown in figure [ fig14 ] . the lda low - bias conductance , calculated with the 5@xmath566@xmath546@xmath55 basis for gold , is 0.43 @xmath63 , which drops to 0.19 @xmath63 when asic is applied . in this case the molecular homo is pinned at @xmath2 and the transmission remains large . interestingly this is a situation similar to the one we discussed in our previous work in the context of tight - binding hamiltonian @xcite . in this case the suppression of the conductance may further originate from the derivative discontinuity of the potential , which may move the peak in the conductance away from @xmath2 . since this is not described by the asic scheme , we believe that the current in this case may be erroneously large . the results for the case where bdt is attached to the hollow sites but hydrogen atoms remain attached to thiol anchoring groups are presented in figure [ fig16 ] . the total energy for this situation is 1.47 ev higher than that obtained when the end groups are thiolate and the remaining two h atoms form a h@xmath71 molecule . as can be seen from the dos in panels ( a ) and ( b ) and the transmission coefficients in panels ( c ) and ( d ) , the transport is now through the lumo of the system , the energy of which is lowered only slightly by asic . hence , the asic zero - bias conductance of 0.09 @xmath63 is higher than the value of 0.06 @xmath63 calculated with lda . however it is important to note that the small , but significant , downshift of the lumo state caused by asic is only an artifact of our atomic approximation to sic . as an empty state , the sic for the lumo is not defined and no corrections should be applied . the second device studied consists of a benzenedimethanethiol ( bdmt ) molecule attached to two gold fcc ( 111 ) surfaces . two different isomers of are investigated ; in the first the sulfur atoms of the thiol groups lie in the same plane of the benzene ring ( figure [ fig17]a ) , while in the second they are aligned out of plane ( figure [ fig17]b ) . in both cases they anchor the molecule to the hollow site of the gold fcc ( 111 ) surface . the orbital resolved dos , transmission coefficients and @xmath0-@xmath1 curves for the first isomer are presented in figure [ fig18 ] for both lda and asic . in this case , as one can see from the orbital resolved density of states ( figures [ fig18]a and [ fig18]b ) , the homo - lumo gap is much larger than that of bdt . the asic again has the effect of shifting the occupied orbitals downwards in energy , but this has little effect on the transport since the resonances in the transmission coefficients due the homo and lumo lie well outside the bias region investigated . therefore , asic increases the homo - lumo gap but the actual transmission coefficient around @xmath2 does not change much from its lda value . the asic conductance at zero bias is calculated to be 0.004@xmath63 , compared to 0.006@xmath63 obtained with lda only . both these values are one order of magnitude larger than the experimental value of 0.0006@xmath63 obtained by xiao et . it also follows that the @xmath0-@xmath1 curves ( figures [ fig18]e and [ fig18]f ) obtained with lda and asic respectively are quite similar . the @xmath0-@xmath1 curve calculated for the second isomer is presented in figure [ fig19 ] . also in this case the effects of asic are minimal due to the large homo - lumo gap . the conductance at zero bias is calculated to be 0.015@xmath63 with lda , compared to the value of 0.013@xmath63 for asic . interestingly this is about a factor 3 larger than that calculated for the first isomer . in general , in contrast to the case of bdt , for bdmt asic does not appear to improve the agreement between theory and experiments . the fact that the fermi level of au is pinned in the middle of the large homo - lumo gap of bdmt , which is already well described by lda , makes the asic corrections rather marginal to transport at moderate voltages . given the fact that the geometry of the anchoring is not known and that we have only investigated one case , we may conclude that the disagreement between theory and experiments may simply lie in the details of the anchoring geometry . the third and final system investigated is that of biphenyldithiol ( bpd ) attached to two gold surfaces as shown in figure [ fig20 ] . once again we consider only the gold fcc ( 111 ) hollow site as the anchoring site . the planes of the two benzene rings are usually tilted with respect to each other by an angle known as the torsion angle , that in this case is calculated to be 37@xmath72 . the orbital resolved dos , transmission coefficients and @xmath0-@xmath1 curves for this system are presented in figure [ fig21 ] for both lda and asic . once again , asic has the effect of lowering the energy of the occupied molecular orbitals , as can be seen from the dos ( panels ( a ) and ( b ) ) . this in turn results in opening up the conductance gap in the transmission as shown in panels ( c ) and ( d ) of figure [ fig21 ] . for this molecule , the homo is near @xmath2 as in the case of bdt , giving a lda conductance of 0.07@xmath63 at zero bias . when asic is applied , the homo is shifted downwards in energy and outside the bias window for voltages up to 2 volt . in this case the asic conductance drops to 0.018@xmath63 at @xmath2 and the whole low bias @xmath0-@xmath1 curve shows rather smaller current with respect to its lda counterpart . the optimum torsion angle is @xmath73@xcite . however , this may fluctuate due to temperature or when the molecule is under strain in a breaking junction . panel ( e ) of figure [ fig21 ] shows the @xmath0-@xmath1 curves calculated for the equilibrium torsion angle of @xmath73 , whereas panel ( f ) shows the result for the case when the benzene rings are in the same plane ( i.e. when the torsion angle is @xmath74 ) . reducing the torsion angle causes an increase in the transmission since the overlap between the @xmath69 orbitals is increased . the conductance at zero bias for a torsion angle of @xmath74 is 0.09@xmath63 with lda only , and 0.024@xmath63 when asic is applied . these results show that asic has an effect on bpd similar to the one it has on bdt , i.e. it shifts the homo downwards in energy and reduces the zero - bias conductance . in this case however even our asic results differ from the experimental data of dadosh et . @xcite by several orders of magnitude . since we are essentially in a tunneling situation , the magnitude of the current is severely dependent on the tunneling matrix elements , which in turn are rather sensitive to the details of the anchoring geometry . although the hollow site is believed to be the preferential site , such a large discrepancy between theory and experiments could be attributed to the contact geometry . in view of the many contradictory results for the transport properties of molecular junctions , first principles computational tools are becoming increasingly important . these , however , also suffer from some uncertainty in the results , rooted in the approximations used for computing the junction electronic structure . in this work , we have addressed the problem of the self - interaction error in dft and we demonstrate how a simple self - interaction correction scheme can be used to improve the agreement between theory and experiments . in particular , we have considered the problem of correctly reproducing the molecular ionization potential as the negative of the single particle kohn - sham homo energy . our asic scheme is capable of achieving such a description , and for most of the molecules investigated this has the effect of moving conductance resonances away from the fermi level of the electrodes . thus , molecular junctions with zero bias resonant transport through the homo become insulating when calculated with asic . we then investigated transport through bdt attached to the gold ( 111 ) surfaces in several different anchoring geometries . in general , we have found a systematic reduction of the low bias current when asic is applied . most importantly , we have demonstrated that a number of anchoring geometries all yield similar low bias conductivities , which in most cases are rather insensitive to bond stretching or bond angle changes . this suggests that several anchoring geometries might contribute to the peaks in the typical experimental conductance histograms , thus providing a natural explanation for their stability . nevertheless , even with asic , some problems still remain . in general , if the first molecular orbital to enter the bias window as the bias increases is the lumo , then our method potentially fails . in fact , asic only partially restores the dft derivative discontinuity , and erroneously @xmath75 . since the size of the derivative discontinuity is unknown , it is difficult at this time to evaluate the impact of such an error . a second problem is connected to the accuracy of the calculation of the wave - functions . when no molecular levels appear at the fermi level , the transport is essentially tunneling like , and therefore becomes crucially dependent on the accuracy of the calculated hopping integrals between the molecule and the electrodes . asic does not drastically improve the quality of the lda wave - function and thus one might expect some errors in the tunneling transmission coefficients . additionally , we do not correct the surface atoms of the electrodes for si , so it is expected that our asic results in general produce overextended orbitals and hence a systematically overestimated current . finally , it is worth noting that asic still overestimates the polarizability of molecules @xcite , with a quantitatively incorrect prediction of the response exchange and correlation field . the use of xc potentials constructed from exact charge densities @xcite which correct both the molecule and the metallic surfaces may offer a solution to these problems . however , for some of the molecules investigated the disagreement between theory and experiments is still of several orders of magnitude . this is difficult to associate with any systematic error in the theory , and might be connected to the specific contact geometry . in conclusion , negf combined with the asic electronic structure method offers a description of transport through molecules which is considerably improved with respect to standard negf lda / gga . since this is also a computationally undemanding scheme , it can be employed for large systems , and therefore can become an important tool for systematically predicting the performance of molecular junctions . we thank alex reily rocha , chaitanya pemmaraju , kieron burke and alessio filippetti for useful discussions . this work is funded by science foundation of ireland ( grant sfi02/in1/i175 ) . computational resources have been provided by the hea iitac project managed by the trinity center for high performance computing and by ichec . + j. lagerqvist , m. zwolak and m. di ventra , nano letters * 6 * , 779 ( 2006 ) ; m. zwolak and m. di ventra , nano letters * 5 * , 421 ( 2005 ) ; r. zikic , p.s . krstic , x .- zhang , m. fuentes - cabrera , j. wells and x. zhao , phys . e * 74 * , 011919 ( 2006 ) . hettler , w. wenzel , m.r . wegewijs and h. schoeller , phys . lett . * 90 * , 076805 ( 2003 ) ; p. delaney and j.c . greer phys . * 93 * , 036805 ( 2004 ) ; b. muralidharan , a.w . ghosh , and s. datta , phys . b , * 73 * , 155410 ( 2006 ) ; s .- h . ke , h.u . baranger , and w. yang , j. chem . phys . * 126 * , 201102 ( 2007 ) ; k.s . thygesen and a. rubio , j. chem . * 126 * , 091101 ( 2007 ) . a rigorous demonstration that ks orbitals can be used as single particle states for electronic transport is still not available . for an extensive discussion see m. koentopp , k. burke and f. evers , phys . b rapid comm . * 73 * , 121403 ( 2006 ) ; g. stefanucci , s. kurth , e.k.u . gross , and a. rubio , cond - mat/0607333 ; g. stefanucci , phys . rev . b * 75 * , 195115 ( 2007 ) .

theoretical study of the space - time evolution of matter in high energy nuclear collisions has a long history since the early pioneering works of fermi and landau employing thermodynamics and fluid mechanics @xcite . more recent works @xcite have been motivated by the prospect of studying a new form of matter experimentally by means of very energetic collisions of heavy nuclei as those currently underway at brookhaven with relativistic heavy - ion collider ( rhic ) and planned at cern with large hadron collider ( lhc ) . the data taken from the rhic experiments has shown existence of a strong anisotropic ( elliptic ) collective flow of hadrons in non - central collisions @xcite , indicating early local equilibration of dense matter produced by the collision @xcite . this supports the hydrodynamic picture of matter evolution and gives us a hope to learn information of the equation of state of dense matter from systematic analyses of the data . much attention has been paid to the early stage of matter evolution where a dense plasma of unconfined quarks and gluons has been expected to be formed through some complex non - equilibrium processes @xcite . one anticipates also a breakdown of the hydrodynamic behavior of the system when it is diluted sufficiently and the collision time exceeds the characteristic time scale of expansion . this stage of the matter evolution is usually refered to as the freezeout stage . the aim of this work is to present a kinetic theory which is designed to describe the freezeout stage of the expanding hadron gas . the freezeout of the expanding hadronic matter would proceed in several steps ; chemical freezeout of the relative abandunce of hadron spieces may occur before the kinetic freezeout of the momentum distribution of hadrons . observed relative abandunce of hadrons fit very nicely to a simple picture that the chemical freezeout occurs at a certain temperature and baryon chemical potential @xcite . some of the other observables , such as two particle momentum correlations , the pion analogue of the hanbury brown and twiss two photon intensity interferometry , may be very sensitive to the dynamics of the final kinetic freezeout as indicated by the hbt puzzle " found in the recent data analysis @xcite . if the initial state of the dense matter formed in the collision is a plasma of deconfined quarks and gluons , freezeout of the color degrees of freedom should proceed these processes when the quark - gluon plasma hadronizes . chiral symmmetry @xcite , an approximate global symmetry of qcd which becomes exact in the limit of vanishing quark masses and is considered to be broken spontaneously in the qcd vacuum , may also play an important role in the freezeout dynamics . as a dense hadronic matter formed by ultrarelativistic nuclear collision is diluted by the expansion , one expects that the system undergoes a phase transition associated with the spontaneous breakdown of the chiral symmetry which is restored temporalily after the collision by the formation of a quark - gluon plasma . as the quark - gluon plasma hadronizes and the system turns into the confining phase , the system would gradually develop a vacuum chiral condensate and remaining excitaions would expand in the influence of the growing chiral condensate . this physical picture has been elaborated in terms of a classical equation of motion for the chiral condensate ; effects of excitations were described in terms of statistical fluctuations in the classical fields @xcite . the fluctuating condensate described by classical pion field has been termed disoriented chiral condensate ( dcc ) , emphasizing the symmetry aspect of the problem @xcite . what was missing in these classical treatments of the meson fields is the existence of particle excitations in addition to the condensate . inclusion of particle excitations requires the quantization of the fields . the effect of the quantum fluctuations of meson fields has been studied by tsue , vautherin and one of the present authors ( tm ) @xcite by the functional schrdinger picture formalism @xcite . the physical picture we have described is very similar to what happens when the dilute gas of magnetically trapped atoms of alkali metals cools down by evapolation @xcite . some of the atoms condense into the lowest single particle level in the trapping external potential forming a bose - einstein condensate . the dynamics of such a system may be described by the coupled equations of motion for the condensate , or the gross - pitaevskii equation @xcite , and the kinetic equation , or the boltzmann - vlasov equations , for the phase space distribution of the excitations in the presence of the condensate @xcite . in this paper we will show that a similar set of equations can be derived for a system of interacting mesons described by the relativistic quantum field theory by the mean field approximation . this approximation corresponds to a neglect of all correlations in the system @xcite . we will make no attempt in this work to justify this approximation and leave it as an open problem for further study . in this context we note that field theoretical derivation of a boltzmann - type kinetic equation has been given by many others @xcite . most of these works focuses on the derivation of the collision terms . the present work is distinguished from these works in the emphasis of the role of the mean field in the evolution of the system coupled with the quasi - particle excitations in the same spirit as in @xcite . we omit the effect of the quasi - particle collisions in this work . this procedure may be reasonable , at least as a first step , for describing dynamical aspect of the freeze - out process : even in the absence of collisions , interactions between the quasi - particles and the evolving condensate would affect the final particle distribution . but the present approach is not adequate , however , for the early thermalization problem , where the collision terms play essential role @xcite . we adopt the standard heisenberg picture instead of the functional schrdinger picture used in @xcite since we found it more straightforward to see the connection to familiar semi - classical kinetic equation . in this first of a series of papers we shall use one - component real scalar field model interacting via a @xmath0 self - interaction term in order to concentrate on the presentation of the basic features of the theory . analysis with multi - component scalar model with @xmath1 symmetry will be deferred for the forthcoming paper @xcite . in the next section we formulate a quantum kinetic theory for quantum scalar meson field , starting from the heisenberg equations for quantum field of one - component real scalar . the mean field approximation replaces the products of quantum fields by the products of classical fields and the statistical average of bilinear forms of the quantum fluctuations . the latter is expressed in terms of the wigner functions which is reduced to the single particle distribution function in the classical boltzmann equation . our theory contains another forms of the wigner functions which have no classical counter parts and arises due to the coherent pair creation and annihilation processes in non - uniform systems . only in uniform systems , these _ off - diagonal _ components of the fluctuation can be eliminated by suitable redefinition of the particle mass . appearance of the off - diagonal wigner functions is reminiscent of the anomalous propagators in the microscopic theory of superconductivity @xcite . similar structure also appears in the theory of bose - einstein condensate @xcite . some details of the mean field calculation is given in appendix . in section 3 , we apply our method to uniform systems and show that each mode characterized by the particle momentum obeys non - linear forced oscillatory motion . in section 4 we show that in equilibrium these kinetic equations are reduced to a gap equation @xcite which determine the equilibrium amplitude of the condensate and the mass parameter as a function of the temperature . in this paper , we ignore the effect of the divergent vacuum polarizations which requires a subtle renormalization procedure in the mean field approximation @xcite . the solution exhibits characteristic features of the first order phase transition . in section 5 , we study slowly varying non - uniform systems . taking long wavelength approximation , the equations of motion of the diagonal components of the wigner functions are reduced to a vlasov equation in a form generalized by landau for quasi - particles excitations in quantum fermi liquid @xcite with the quasi - particle energy given in the mean field approximation . non - vanishing off - diagonal components of the wigner functions generate extra terms in the vlasov equation which may be interpreted as particle source and sink terms . in section 6 , we compute the dispersion relations of excitations of the system near equilibrium . solving the coupled kinetic equations by linearizing the equations with respect to small oscillatory deviations from equilibrium solution we obtain dispersion relation of the excitation modes in the system near equilibrium . we find the continuum of quasi - particle excitations in the entire space - like energy - momentum region in addition to the continuum in time - like region due to the ( thermally induced ) pair creation . we found that in the low temperature phase the meson pole shifts due to the coupling to the quasi - particle continua . the effective meson mass vanishes at the edge of the spinodal instability line of the first order transition . meson mode has been studied first by hatsuda and kunihiro as a precursory phenomenon of the chiral phase transition using the nambu - jona - lasinio model with quark fields.@xcite ] we also found that the coupling of the off - diagonal components of the wigner function plays an important role to prevent appearance of undamped tachyonic sound mode which propagates with a velocity greater than that of light . a short summary of the paper is given in section 7 with remarks on the remaining problems . in this section we derive quantum kinetic equations which describes the time evolution of the meson condensate coupled with mesonic quasi - particle excitations . we use the natural unit @xmath2 throughout this paper . we first take a simple model of a self - interacting real scalar field in the heisenberg picture . the hamiltonian is given by @xmath3 \right]\ ] ] where @xmath4 = \frac{1}{2 } m^2 { { \hat \phi}}^2 + \frac{\lambda}{4 ! } { { \hat \phi}}^4 .\ ] ] here the scalar field @xmath5 and its canonical conjugate momentum field @xmath6 are quantized by the equal - time commutation relations : @xmath7 & = & i \delta ( { { \mathbf r}}- { { \mathbf r } } ' ) \\ \left [ { { \hat \phi } } ( { { \mathbf r } } , t ) , { { \hat \phi}}({{\mathbf r } } ' , t ) \right ] & = & [ { { \hat \pi } } ( { { \mathbf r } } , t ) , { { \hat \pi } } ( { { \mathbf r } } ' , t ) ] = 0 \ ] ] the heisenberg equation of motion of the quantum field @xmath8 is given by @xmath9 = { { \hat \pi } } ( { { \mathbf r } } , t ) \ ] ] while the equation of motion of the canonical conjugate field @xmath10 becomes @xmath11 = ( \nabla^2 - m^2 ) { { \hat \phi}}({{\mathbf r } } , t ) - \frac{1}{3 ! } \lambda { { \hat \phi}}^3 ( { { \mathbf r } } , t ) \ ] ] elimination of the field momentum @xmath6 from these equations yields a modified klein - gordon equation for the quantum scalar field @xmath5 : @xmath12 where @xmath13 . we are interested in the time evolution of the system described by the density operator @xmath14 where @xmath15 are a set of normalized wave functions and @xmath16 is the probability distribution for a mixed state described by this density matrix so that it satisfies @xmath17 this density matrix can be also expressed in terms of some complete set @xmath18 of the wave functions of our hilbert - fock space as @xmath19 where @xmath20 all physical information of the system to be described are contained in a specific form of the density matrix . in thermodynamic equilibrium , the density matrix is given by @xmath21 with @xmath22 = e^{\beta f ( t ) } \ ] ] where @xmath23 gives the helmholtz free energy of the system at temperature @xmath24 . in the heisenberg picture the density matrix is time - independent since the wave functions are time - independent ; all time dependence arises from the time dependence of an operator : @xmath25\ ] ] for example , we define the classical condensate fields by the statistical average of the quantum fields , @xmath26 in the following we take a gaussian ansatz for the density matrix : @xmath27 and @xmath28 where @xmath29 and @xmath30 are shifted field operators defined by @xmath31 the shifted field operators obey the same equal - time commutation relations as the original fields : @xmath32 & = & i \delta ( { { \mathbf r}}- { { \mathbf r } } ' ) \label{q1 } \\ \left [ { { \tilde \phi } } ( { { \mathbf r } } , t ) , { { \tilde \phi}}({{\mathbf r } } ' , t ) \right ] & = & [ { { \tilde \pi } } ( { { \mathbf r } } , t ) , { { \tilde \pi } } ( { { \mathbf r } } ' , t ) ] = 0 \label{q2 } \end{aligned}\ ] ] we will show that this choice of the density matrix is a non - equilibrium generalization of the hartree approximation in equilibrium . it has been alternatively introduced in the functional schrdinger representation @xcite we introduce the one - particle wigner function by @xmath33 here the particle creation and annihilation operators may be defined in terms of the fourier transforms of the shifted fields @xmath34 as @xmath35 \\ a^\dagger_{{\mathbf p}}(t ) & = & \frac{1}{\sqrt{2 { { \omega_{{\mathbf p } } } } } } \left [ { { \omega_{{\mathbf p}}}}{{\tilde \phi}}_{{\mathbf p}}(t ) - i { { \tilde \pi}}_{- { { \mathbf p } } } ( t ) \right ] \end{aligned}\ ] ] with @xmath36 these relations are rewritten as : @xmath37 \\ { { \tilde \pi}}({{\mathbf r } } , t ) & = & i \sum_{{\mathbf p}}e^{i { { \mathbf p}}\cdot { { \mathbf r } } } \sqrt{\frac{{{\omega_{{\mathbf p}}}}}{2 } } \left [ a_{{\mathbf p}}^\dagger ( t ) - a_{-{{\mathbf p } } } ( t ) \right]\end{aligned}\ ] ] the quantization rules ( [ q1 ] , [ q2 ] ) are transcribed to @xmath38 & = & \delta_{{{\mathbf p } } , { { \mathbf p } } ' } \\ \left [ a_{{\mathbf p}}(t ) , a_{{{\mathbf p } } ' } \right ( t ) ] & = & [ a_{{\mathbf p}}^\dagger ( t ) , a_{{{\mathbf p}}'}^\dagger ( t ) ] = 0 .\end{aligned}\ ] ] with which we may interpret @xmath39 ( @xmath40 ) as annihilation ( creation ) operator of particle excitation " with momentum @xmath41 . we also introduce the following other forms of the wigner functions : @xmath42 these wigner functions are not independent but are related to each other . the complex conjugate of the wigner functions are given by @xmath43 where the asterisk ( @xmath44 ) stands for the complex conjugate . the commutation relations imply also that @xmath45 and @xmath46 are even functions of @xmath41 @xmath47 and @xmath48 the four wigner functions may be grouped together to form a matrix form of the wigner function @xmath49 the appearance of the off - diagonal " components of the wigner functions is reminiscent of the anomalous propagators in the bcs theory of superconductivity which arises due to the presence of fermion pair condensate @xcite . note that our definition of the matrix components of the wigner function is slightly different from those for the propagators . we write the fourier transforms of the wigner functions as @xmath50 the ( [ fstar ] ) and ( [ fbarstar ] ) imply that @xmath51 and @xmath52 are real functions and are related to each other by @xmath53 while ( [ gstar ] ) implies that @xmath54 and @xmath55 are complex conjugate to each other : @xmath56 here we have chosen the particle mass " @xmath57 to be different from the mass parameter @xmath58 in the original hamiltonian . physical particle mass for interacting fields is generally different from the mass parameter in the hamiltonian or lagrangian due to the effect of interaction , e.g. renormalization with or without the spontaneous symmetry breaking . it may also depend on the physical conditions described by the statistical average with the density matrix @xmath59 . since we are interested in non - equilibrium time - evolution of the system where the physical particle mass may not have a definite meaning in the intermediate states , we consider here @xmath57 just as a parameter to be chosen at our discretion , for an appropriate choice of the initial conditions specified by the gaussian density matrix . a different choice of this mass parameter would give different definition for particle excitations " ; the wigner functions thus depend on the particular choice of the mass parameter . suppose we take a different particle mass @xmath60 to define the particle creation and annihilation operators @xmath61 \\ b^\dagger_{{\mathbf p}}(t ) & = & \frac{1}{\sqrt{2 { { \omega_{{\mathbf p } } } } ' } } \left [ { { \omega_{{\mathbf p } } } } ' { { \tilde \phi}}_{{\mathbf p}}(t ) - i { { \tilde \pi}}_{- { { \mathbf p } } } ( t ) \right ] \end{aligned}\ ] ] with @xmath62 these new particle creation and annihilation operators should also obey the commutation relations , @xmath63 & = & \delta_{{{\mathbf p } } , { { \mathbf p } } ' } \\ \left [ b_{{\mathbf p}}(t ) , b_{{{\mathbf p } } ' } \right ( t ) ] & = & [ b_{{\mathbf p}}^\dagger ( t ) , b_{{{\mathbf p}}'}^\dagger ( t ) ] = 0 .\end{aligned}\ ] ] so that they are related to the original ones by the bogoliubov transformation : @xmath64 where the real parameter @xmath65 is determined by requiring that they describe the same fields : @xmath66 = \sum_{{\mathbf p}}\frac{e^{i { { \mathbf p}}\cdot { { \mathbf r } } } } { \sqrt{2 \omega'_{{\mathbf p } } } } \cdot \left [ b_{{\mathbf p}}(t ) + b_{-{{\mathbf p}}}^\dagger ( t ) \right ] \nonumber \\\end{aligned}\ ] ] and this gives @xmath67 or @xmath68 the new wigner functions defined by replacing the creation and annihilation by the new ones are related to the original wigner functions by : @xmath69 where @xmath70 with @xmath71 for a small change of the mass parameter @xmath72 the wigner function will change by @xmath73 in particular , @xmath74 in uniform equilibrium system the mass parameter may be chosen to `` diagonalize '' the one - body mean field hamiltonian . as will be shown later , this procedure will lead to the well - known gap equation , a self - consistency condition to determine @xmath57 . if the system is slowly changing in time , one may still use such procedures adjusted to slowly varying quasi - equiliblium conditions , introducing a time dependent effective mass as a dynamical parameter to describe such adiabatic process . for such calculations , the relation ( [ df ] ) may be used to describe the adiabatic change of the wigner functions by the change of the mass parameter . in more general non - equilibrium situations as we expect to encounter at the freeze - out stage of expanding matter , however , there may be no such appropriate condition to determine the mass parameter . the situation could even be worse : if the system goes through unstable state with respect to small fluctuation of the field , then the adiabatically determined mass parameter would become pure imaginary , reflecting the extremum , instead of the minimum , of the effective potential . in such case we may keep the value of @xmath57 at a certain real value reflecting initial conditions . the instability would then show up as appearance of a growing solution to our kinetic equations ; which would eventually be stabilized by the non - linear interaction . to extract the physical information , such as the particle distribution in the final asymptotic state , we should use the wigner function defined with the physical particle mass in the vacuum . but these asymptotic physical wigner functions may be calculated from the wigner functions with a different choice of the mass parameter by the relation ( [ rotation ] ) . the equation of motion of the classical mean field @xmath75 is obtained by taking the quantum statistical average of the field equation ( [ kg ] ) . with the gaussian ansatz for the density matrix , we find @xmath76\ ] ] this equation corresponds to the non - linear schdinger equation ( also called the gross - pitaevskii equation @xcite ) in the theory of bose - einstein condensates . so we may call this equation _ non - linear klein - gordon equation_. the non - linearity arises due to the self - interaction of the classical field @xmath75 ( condensate ) and also due to the interaction with fluctuations @xmath77 which also depends on @xmath75 implicitly . the latter may be interpreted as due to particle excitations " , since the fluctuation can be expressed by the wigner functions as @xmath78 \label{phi2}\end{aligned}\ ] ] the time - evolution of the classical mean field @xmath79 is thus coupled with the time - evolution of the wigner functions . to derive the equation of motion of the wigner functions , we need to compute the time - derivative of the bilinear forms of the operators @xmath80 and @xmath81 which in turn requires computation of the commutators of these operators with the hamiltonian . we decompose the original hamiltonian as @xmath82 where @xmath83 is the classical hamiltonian obtained from @xmath84 by replacing the quantum fields by their classical expectation values and @xmath85 contain the @xmath86-thrth power of the quantum fluctuation @xmath5 ( or @xmath39 and @xmath87 ) . a straightforward calculation yields @xmath88 \\ h_2 & = & \int d { { \mathbf r}}\left [ \frac{1}{2 } { { \tilde \pi}}^2 + \frac{1}{2 } ( \nabla { { \tilde \phi}})^2 + \frac{1}{2 } m^2 { { \tilde \phi}}^2 ( { { \mathbf r } } , t ) + \frac{\lambda}{4 } \phi_c^2 ( { { \mathbf r } } , t ) { { \tilde \phi}}^2 ( { { \mathbf r } } , t ) \right ] \\ h_3 & = & \frac{\lambda}{3 ! } \int d { { \mathbf r}}\phi_c ( { { \mathbf r } } , t ) { { \tilde \phi}}^3 ( { { \mathbf r } } , t ) \\ h_4 & = & \frac{\lambda}{4 ! } \int d { { \mathbf r}}{{\tilde \phi}}^4 ( { { \mathbf r } } , t ) \end{aligned}\ ] ] the commutators of bilinear forms of @xmath80 and @xmath81 with @xmath89 vanish and the commutators with @xmath90 would give either a linear term or the third power of the fluctuation , both of which may vanish when taking the average with the gaussian density matrix . what remain to be computed are then the commutators with @xmath91 and with @xmath92 . they will give either the bilinear form of @xmath80 and @xmath87 or the fourth power of the fluctuations . the gaussian average of the resultant equations of motion of the bilinear field operators would give the desired equations of motion of the wigner functions . details of this computation is given in appendix a. the resultant equation of motion of the wigner functions may be obtained more easily by introducing the mean field hamiltonian defined by @xmath93 \nonumber \\ & = & \int d { { \mathbf r}}\left [ \frac{1}{2 } { { \tilde \pi}}^2 + \frac{1}{2 } ( \nabla { { \tilde \phi}})^2 + \frac{1}{2 } \mu^2 { { \tilde \phi}}^2 + \frac{1}{2 } { { \delta \pi } } ( { { \mathbf r } } , t ) { { \tilde \phi}}^2 \right ] \end{aligned}\ ] ] where @xmath94 and @xmath95 in the momentum representation this mean field hamiltonian may be written as @xmath96 where @xmath97 the commutator of a bilinear operator product of @xmath39 and @xmath87 with this mean - field hamiltonian is given by @xmath98 & = & - ( \omega_{{{\mathbf p}}_1 } - \omega_{{{\mathbf p}}_2 } ) a^\dagger_{{{\mathbf p}}_1 } a_{{{\mathbf p}}_2 } - \sum_{{\mathbf q}}{{\delta \pi}}_{{\mathbf q}}\cdot \frac { ( a_{-{{\mathbf p}}_1-{{\mathbf q } } } + a_{{{\mathbf p}}_1 + { { \mathbf q}}}^\dagger ) a_{{{\mathbf p}}_2 } } { \sqrt{2 \omega_{{{\mathbf p}}_1 + { { \mathbf q } } } } \sqrt{2\omega_{{{\mathbf p}}_2 } } } \nonumber \\ & & \qquad + \sum_{{\mathbf q}}{{\delta \pi}}_{{\mathbf q}}\cdot \frac{a_{{{\mathbf p}}_1}^\dagger ( a_{{{\mathbf p}}_2 - { { \mathbf q } } } + a_{- { { \mathbf p}}_2 + { { \mathbf q}}}^\dagger ) } { \sqrt{2 \omega_{{{\mathbf p}}_1 } } \sqrt{2\omega_{{{\mathbf p}}_2 - { { \mathbf q } } } } } \ ] ] we show in appendix that the quantum statistical average of this commutator with the gaussian density matrix gives precisely the same result for the same statistical average of the commutator with the original hamiltonian : @xmath99 { \rangle}= { \langle } [ a^\dagger_{{{\mathbf p}}_1 } a_{{{\mathbf p}}_2 } , h ] { \rangle}\ ] ] therefore one can compute the equations of motion of the wigner functions using this effective hamiltonian , @xmath100 { \rangle}= { \langle } [ a^\dagger_{{{\mathbf p}}+ { { \mathbf k}}/2 } a_{{{\mathbf p}}- { { \mathbf k}}/2 } , h_{\rm mf } ] { \rangle}\ ] ] using this we find , @xmath101 equations of motion of other three wigner functions @xmath102 , @xmath103 , @xmath104 can be also computed from the commution relations of the product operators @xmath105 , @xmath106 , @xmath107 , with the mean field hamiltonian @xmath108 , respectively : we obtain @xmath109 and the equation of motion of @xmath110 can be obtained from ( [ eomf ] ) by the substitution @xmath111 : @xmath112 these equations form a closed system of coupled differential equations with the non - linear klein - gordon equation ( [ nlkg ] ) which may be rewritten as @xmath113 these four equations of motion of the wigner function may be combined into a single matrix form as @xmath114 for a uniform system , we expect that the classical mean field and the self - energy become functions only of time : @xmath115 thus the non - linear klein - gordon equation ( [ nlkg ] ) becomes @xmath116 and the mean - field hamiltonian is reduced to @xmath117 in this case the wigner functions contain non - vanishing components only for the diagonal elements ( @xmath118 ) so that they may be written as @xmath119 then the equations of motion of the wigner functions become @xmath120 these coupled equations have a time - independent solution of the form @xmath121 with @xmath122 thus if we define the particle mass @xmath57 so as to satisfy @xmath123 , then two components of the wigner functions @xmath124 and @xmath125 vanish for all @xmath41 . in this case , @xmath126 may be interpreted as the momentum distribution of physical particle excitations . to find time - dependent solutions , we write @xmath127 and rewrite the equations as @xmath128 these equations are to be solved together with the non - linear klein - gordon equation ( [ nlkg0 ] ) with @xmath129\ ] ] from these equations we can derive second order differential equation : @xmath130 with @xmath131\ ] ] we observe that each of these coupled equations looks like an equation of a forced oscillator . for example the frequency of the oscillator @xmath132 is given by @xmath133 while the frequency of the oscillator @xmath134 is given by @xmath135 so that @xmath134 oscillates much rapidly than @xmath132 . the coupling of these oscillators may produce an interesting interesting effects which may be studied by numerical integration . in statistical equilibrium , we have the bose distribution @xmath136 together with @xmath137 where @xmath138 is the inverse temperature . the condition of vanishing @xmath139 implies from ( [ self ] ) @xmath140 where the thermal fluctuation of the quantum field is given by the relation ( [ phi2 ] ) as @xmath141 on the other hand , the condensate amplitude @xmath142 in equilibrium should also satisfy the static non - linear klein - gordon equation , @xmath143\ ] ] from these two conditions , we find , for @xmath144 , @xmath145 and @xmath146 we note that , since the thermal distribution @xmath147 depends on the mass ( gap ) parameter @xmath57 through @xmath148 , the equation ( [ gap ] ) determines the mass parameter @xmath57 self - consistently as a function of the temperature . this equation is called the _ gap equation _ @xcite . as the temperature increases the mass gap @xmath57 and the condensate amplitude @xmath142 decreases and vanishes at high temperature . to find the behavior of the mass parameter as a function of the temperature we can use the following formula @xcite for the integral of the bose distribution function : @xmath149 where the dimensionless function @xmath150 is given as @xmath151 ( @xmath152 is euler s number ) . the gap equation ( [ gap ] ) has a solution @xmath153 at @xmath154 determined by @xmath155 however the solution exhibits the behavior of the first order transition due to the non - analytic behavior of the function @xmath156 @xcite . this is a generic feature of the mean field approximation @xcite which may be a theoretical artifact and may not survive the inclusion of correlations missing in the mean field approximation . an improvement of the mean field approximation has been proposed in @xcite and later it has been shown to lead to the second order transition in two - loop approximations @xcite . for @xmath157 ( dashed line ) and @xmath158 ( solid line ) ] in fig . 1 we plot @xmath142 as determined by ( [ phi0 ] ) as a function of temperature . it shows a behavior characteristic of the first order phase transition : there is a region @xmath159 where there are three solutions of ( [ phi0 ] ) , one at @xmath160 and other two at @xmath161 as well as @xmath162 . one expects that two solutions @xmath163 and @xmath163 correspond to two local minima of the effective potential @xmath164 while the other one @xmath165 corresponds to the local maximum of @xmath166 . indeed we can show this explicitly by constructing the effective potential @xmath167 by demanding that the local extremal condition @xmath168 coincides with the condition ( [ phi0 ] ) . @xmath169 the integration of the first two terms are trivial . beside irrelevant integration constant , which we may choose to be zero , we obtain @xmath170 to carry out the remaining integral over @xmath171 we demand that the mass parameter @xmath57 depends on @xmath171 through the relation @xmath172 which appears to be equivalent to the condition ( [ mu20 ] ) , but we assume that this relation holds not only for the equilibrium value of @xmath171 , namely at @xmath173 , but also for any value of @xmath171 . noting @xmath174 which results from ( [ mu2 ] ) , we transform the integration variable from @xmath171 to @xmath175 : @xmath176 remaining integral over @xmath175 can be carried out as @xmath177 = - \sum_{{\mathbf p}}\frac{{{\mathbf p}}^2}{3 \omega_{{\mathbf p } } } f_{\rm eq . } ( { { \mathbf p}})\ ] ] where in deriving the final expression we performed integration by part in @xmath178 and omitted again the irrelevant integration constant . to sum up , we find the following expression for the effective potential : @xmath179 where @xmath171 dependence of the last two terms are implicitly given from the @xmath171-dependence of @xmath57 . we note that this result coincides , in the neglect of the renormalization effects due to the meson mass shift , with the effective potential obtained by amelino - camelia and pi @xcite using the cornwall - jackiw - tombolis ( cjt ) composite operator effective potential formalism @xcite , we note that our mass parameter @xmath57 corresponds to the variational mass parameter @xmath180 of the cjt potential . the third term may be interpreted as the pressure of an ideal gas of the quasi - particles obeying the dispersion @xmath181 . the effective potential given by the formula ( [ veff ] ) takes in general a complex value since @xmath175 becomes negative at low temperature and at small value of @xmath171 . at low temperatures the potential develops a second local minimum at non - zero value of @xmath171 , which becomes the absolute minimum below certain temperature @xmath182 where the first order transition takes place in equilibrium and the order parameter @xmath142 jumps discontinuously from @xmath183 to @xmath142 as the temperature is lowered in the presence of inhomogeneity the wigner functions acquire non - vanishing elements with @xmath184 . if we assume that this inhomogeneity are due to the long - wavelength fluctuations in the system , @xmath185 in this case we can obtain a familiar form of the vlasov equation @xcite from the equation of motion of the wigner function by the procedure usually referred to as the gradient expansion @xcite . to show this we make the following approximations : @xmath186 and @xmath187 we also make use of the taylor expansion of the wigner functions : @xmath188 then the equations of motion ( [ eomf ] ) of the wigner function @xmath189 becomes @xmath190 on the right hand side , the first term and the second term are disguised forms of the drift term and the vlasov term respectively . to obtain more familiar form we make the fourier transforms ( 42 ) and ( 43 ) of the wigner functions . then we find @xmath191 now it is clear that the first term is the drift term which describes the change of the particle position by the drift with the velocity @xmath192 . the second term can be interpreted as the vlasov term which represents the change of particle momentum due to the continuous acceleration by the velocity dependent equivalent potential , @xmath193 acting on the particle . the third term appears to be the correction to the drift term due to the local change of particle mass @xmath194 which causes change in particle velocity . other three terms are associated with the other components of the wigner function and has no counter parts in non - relativistic vlasov equation . noting that @xmath195 and @xmath196 the above equation may be rewritten in a more compact form : @xmath197 where @xmath198 and @xmath199 the quantity @xmath200 plays the same role as the quasi - particle energy which appears in the kinetic equation of landau s fermi - liquid theory @xcite . we note that because of the relation ( [ gstar - gbar ] ) , @xmath201 is a pure imaginary function , while @xmath202 , @xmath203 , @xmath204 , and @xmath200 are all real functions . in the long wavelength approximation , the equation of motion of @xmath45 becomes @xmath205 which may be rewritten for the fourier transforms in a compact form as @xmath206 where @xmath207 . thus the kinetic equation for @xmath54 does not look like a vlasov equation ; it takes a form of a simple oscillator equation with the oscillator frequency @xmath208 exposed to the external perturbation created by the particle distribution @xmath209 . similar equation is derived for @xmath210 which is the complex conjugate of @xmath54 . taking the complex conjugate of the eq . ( [ g - eq ] ) , we obtain the kinetic equation for @xmath211 : @xmath212 adding or subtracting ( [ g - eq ] ) and ( [ gbar - eq ] ) , we find for the real function @xmath213 and the pure imaginary function @xmath214 @xmath215 in the long wavelength approximation , we also have @xmath216 \nonumber \\ & = & \sum_{{\mathbf p}}\frac{1}{2\omega_{{\mathbf p } } } \left [ f_+ ( { { \mathbf p } } , { { \mathbf r } } , t ) + g_+ ( { { \mathbf p } } , { { \mathbf r } } , t ) \right ] \end{aligned}\ ] ] which appears in the non - linear klein - gordon equation ( [ nlkg ] ) for the condensate and in the self energy term ( [ pi ] ) in the vlasov equation . we now apply our coupled kinetic equations , consisting of the non - linear klein - gordon equation and the vlasov equations , to find a dispersion relation of the excitations in the system near equilibrium . the dispersion relations of the excitations of the finite temperature system may be computed using the propagator ( two - time ) formalism , either with real - times ( schwinger - keldysh ) formalism @xcite or with imaginary - times ( matsubara ) formalism with subsequent analytic continuation @xcite . although such calculations have been performed by various authors@xcitenumerical results have been presented focusing on the time - like region of the spectral functions . we study with our kinetic theory the entire range of the @xmath217 plane including the space - like excitations of the system . as has been shown by one of the present authors @xcite the calculation of the zero sound mode in the degenerate fermi liquid using the relativistic landau kinetic equation@xcite , similar to ours , reproduces the same result as the propagator theory in the long wavelength limits . we thus expect that our result may also reproduce the long wavelength behaviors of the excitations obtained from the propagator theory . for this purpose we assume that the distribution functions consist of uniform equilibrium term and a small deviation from it : @xmath218 where @xmath219 we also assume the mean meson field consists of uniform equilibrium term and a deviation from it @xmath220 where @xmath142 is determined by the static non - linear klein - gordon equation ( [ phi0 ] ) together with the solution of the gap equation ( [ gap ] ) for the mass gap @xmath57 . linearization of the non - linear klein - gordon equation with respect to @xmath221 yields @xmath222 \nonumber \\\end{aligned}\ ] ] where the thermal fluctuation of the quantum field in equilibrium is given by @xmath141 and the deviation @xmath223 is expressed in term of the small deviation of the distribution function from equilibrium value . @xmath224 \nonumber \\ & = & \sum_{{\mathbf p}}\frac{1}{2\omega_{{\mathbf p } } } \left [ \delta f ( { { \mathbf p } } , { { \mathbf r } } , t ) + \delta f ( - { { \mathbf p } } , { { \mathbf r } } , t ) + \delta g ( { { \mathbf p } } , { { \mathbf r } } , t ) + \delta g^ * ( { { \mathbf p } } , { { \mathbf r } } , t ) \right ] \nonumber \\\end{aligned}\ ] ] where in deriving the last line we have used @xmath225 which follow from ( [ fbar ] ) and ( [ gbar ] ) respectively . by the linearization with respect to the small increments , @xmath226 , @xmath227 , the vlasov equations become @xmath228 the right hand side of the vlasov equation ( [ vlasov1 ] ) vanishes because @xmath229 vanishes in equilibrium . the fluctuation in the mean field potential @xmath230 which appears in these equations is related to the fluctuations of the condensate amplitude and the distribution functions @xmath231 the linearized klein - gordon equation ( [ linkg ] ) and the linearized vlasov equations ( [ lin - vlasov1 ] ) and ( [ lin - vlasov2 ] ) , supplimented by the constitutive relations " ( [ deltau ] ) and ( [ deltaphi2 ] ) , form a closed set of equations to determine a small fluctuation propagating the system in equilibrium . noting that @xmath232 and @xmath233 are real functions , we seek a solution in the form : @xmath234 where we have introduced the landau prescription @xcite , @xmath235 with a positive infinitesimally small constant @xmath236 , to set an adiabatic switching - on of the fluctuation , namely @xmath237 , @xmath238 all vanish slowly as @xmath239 . the constitutive relations for the fluctuations ( [ deltaphi2 ] ) and the mean field potential ( [ deltau ] ) now read @xmath240 and @xmath241 e^ { i ( { { \mathbf k}}\cdot { { \mathbf r}}- \omega_+ t ) } \nonumber \\ & & \quad + \frac{\lambda}{2\omega_{{\mathbf p } } } \left [ 2 \phi^*_0 \delta \phi^ * + \sum_{{\mathbf p}}\frac{1}{2\omega_{{\mathbf p } } } \left ( \delta f^*_{{\mathbf p}}+ \delta f^ * _ { - { { \mathbf p } } } + \delta g^*_{{\mathbf p}}+ \delta g_{-{{\mathbf p } } } \right ) \right ] e^ { - i ( { { \mathbf k}}\cdot { { \mathbf r}}- \omega_- t ) } \nonumber \\\end{aligned}\ ] ] respectively . using these relations , we find from the linearized klein - gordon equation ( [ linkg ] ) , @xmath242 \delta \phi = - \frac{1}{2 } \lambda \phi_0 \sum_{{\mathbf p}}\frac{1}{2\omega_{{\mathbf p } } } \left [ \delta f_{{\mathbf p}}+ \delta f_{-{{\mathbf p } } } + \delta g_{{\mathbf p}}+ \delta g^*_{-{{\mathbf p } } } \right ] , \ ] ] and its complex conjugate relation , where we have used the equilibrium relation @xmath243 also , from the linearized vlasov equations we obtain @xmath244 \nonumber \\ & & \qquad \qquad \times \left ( 1 + f_{\rm eq . } ( { { \mathbf p } } ) \right ) f_{\rm eq . } ( { { \mathbf p } } ) \nonumber \\ \label{deltaf}\end{aligned}\ ] ] and @xmath245 f_{\rm eq . } ( { { \mathbf p } } ) \nonumber \\ \label{deltag}\end{aligned}\ ] ] where we have used @xmath246 using ( [ deltaphi ] ) , we eliminate @xmath247 in ( [ deltaf ] ) and ( [ deltag ] ) and find @xmath248 where we have introduced the notation @xmath249 for the deviation of the fluctuation . solving ( [ df ] ) and ( [ dg ] ) for @xmath250 and @xmath251 respectively , and inserting the results into ( [ f ] ) we obtain the relation @xmath252 where @xmath253 ~. \nonumber \\\end{aligned}\ ] ] applying the prescription , @xmath254 with @xmath255 implying that the principal part should be taken in integration , we find that @xmath256 consists of the real and imaginary parts : @xmath257 the real part is given by @xmath258 with @xmath259 ~ , \nonumber \\\end{aligned}\ ] ] while the imaginary part is given by @xmath260 with @xmath261 \nonumber \\ \end{aligned}\ ] ] where the integration can be carried out analytically , yielding @xmath262 for @xmath263 . the values of this function for @xmath264 can be found by noting that it is an odd function of @xmath265 . the first term in ( [ phi2 ] ) corresponds to the space - like ( scattering ) continuum with @xmath266 . while the second term corresponds to the continuum of thermally induced pair creation / annihilation of mesons with energy @xmath267 , the function @xmath268 ( and @xmath269 ) has non - vanishing supports in the regions on @xmath217 plane as indicated by the shaded areas in fig . 2 . we plotted in fig . 3 @xmath270 and @xmath268 as functions of @xmath271 at @xmath272 for two different values of @xmath273 . and @xmath268 have non - vanishing value is shown by shaded areas ( two - quasi - particle continua ) . the solid hyperbola corresponding to the meson poles ( @xmath274 ) on which @xmath269 has a @xmath275-function singularity in the low temperature phase . the shaded area in the time - like region below the curve @xmath276 is only an artifact of the long wavelength approximation . ] and @xmath277 as a function of @xmath271 at @xmath278 and @xmath279 ( dashed line ) , @xmath280 ( solid line ) . both @xmath281 and @xmath282 vanish at @xmath283 . @xmath282 also vanishes at the pair creation threshold @xmath284 ( in the long wavelength approximation ) while @xmath281 has a cusp at this point.,title="fig : " ] and @xmath277 as a function of @xmath271 at @xmath278 and @xmath279 ( dashed line ) , @xmath280 ( solid line ) . both @xmath281 and @xmath282 vanish at @xmath283 . @xmath282 also vanishes at the pair creation threshold @xmath284 ( in the long wavelength approximation ) while @xmath281 has a cusp at this point.,title="fig : " ] the condition @xmath285 which follows from ( [ omegaf ] ) determines the dispersion relation of a possible long wavelength collective excitation of the system . before examining the solutions of the dispersion relation ( [ dispersion ] ) . we note that at zero temperature , namely in the absence of quasi - particle excitations , the only solution of our coupled kinetic equations is the one which satisfies the linearized klein - gordon equation ( [ linkg ] ) or ( [ deltaphi ] ) with @xmath286 and @xmath287 . this gives a simple meson pole @xmath288 which appears in the time - like region ( @xmath289 ) . there is no space - like mode of excitations in the absence of quasi - particle excitations due to the mass gap ( @xmath290 ) for the excitations of the vacuum . the situation is different in the case of the ordinary bose - einstein condensate ( bec ) which possesses , even at zero temperature , a low - lying acoustic mode ( bogoliubov phonon ) in the mean - field approximation to repulsive two - body contact interaction@xcite . the ordinary bec contains space - like quasi - particle excitations even at absolute zero temperature which corresponds to the excitation of one of the zero energy particles forming the condensate . the bogoliubov phonon is a collective excitation of such space - like quasi - particle excitation modes modified by the repulsive mean - field interaction . at finite non - zero temperature , the system contains continuum of the space - like excitations for all @xmath291 satisfying @xmath292 as signified by the non - vanishing value of @xmath293 in addition to the continuum in the time - like region @xmath294 . ( see fig . 2 ) any solution of ( [ dispersion ] ) in the space - like region therefore is subject to the collisionless dissipation known as the landau damping@xcite . we plot the function @xmath295 in fig . 4 as a function of @xmath265 at fixed value of @xmath272 . in the low temperature phase , it contains a singularity at the position of the mesonic pole at @xmath296 while this singularity disappears in the high temperature phase . in the presence of the condensate , a disturbance in the quasi - particle distribution may be absorbed into an excitation of the condensate which then propagates with the mesonic dispersion relation and is converted back again to the quasi - particle excitations . this coupling between the condensate and the quasiparticle excitations generates a collective mode in the low temperature phase.he due to the excitation of the condensate coupled with quasi - particle excitations . ] in the high temperature phase , the quasi - particle excitations couple each other only through their direct interaction , hence no mixing with single mesonic pole . the high temperature behavior of @xmath297 is the same as that of the function @xmath281 : it is negative at @xmath298 and increases with @xmath265 and reaches a positive maximum slightly below @xmath283 . in the low temperature phase , @xmath295 reverts the sign for @xmath265 below this singularity . one may interpret that effective coupling strength @xmath299 changes its sign below the meson pole . note that here we have used the relation @xmath300 ( 111 ) for the condensate amplitude . we note that @xmath301 always vanishes at @xmath283 since @xmath302 . the cusp at @xmath284 appears at the threshold of the two quasiparticle creation . the condition ( [ dispersion ] ) is fulfilled only near the meson pole in the low temperature phase and it gives a shift of the mesonic excitation spectrum . we could not find any additional solution satisfying ( [ dispersion ] ) . we like to note here that without the coupling to the fluctuations to pair creation or annihilation , we would not have gotten the first term in the function @xmath270 and then @xmath303 would have become 1 at @xmath304 creating a undamped tachyonic sound mode . hence the off - diaginal wigner functions plays an important role in making our framework consistent with causality . as a function of @xmath271 at fixed @xmath305,title="fig : " ] as a function of @xmath271 at fixed @xmath305,title="fig : " ] we introduce the `` response function '' defined by @xmath306 = - \frac { \omega_2 ( \omega , \mu ) } { \left[1 - \omega_1(\omega , \mu ) \right]^2 + \left [ \omega_2 ( \omega , \mu ) \right]^2}\ ] ] it is plotted in fig . 5 as a function of @xmath265 and compared with the `` bare '' response function given by @xmath307 the response function in the space - like momentum region gives the dynamic form factor of the cross section of the scattering of particles coupled to the excitations of the system @xcite , while its time - like component may give the rate of pair annihilation of the quasi - particles . we note that in the low temperature phase the response function changes its sign below the meson pole due to the sign change of the effective coupling ( [ lambda ] ) . we observe some enhancement ( depletion ) of the strength in the time - like pair annihilation near threshold and small depletion ( enhancement ) in the space - like region in the low ( high ) temperature phase . ( [ response ] ) ( solid lines ) at @xmath278 . for comparison the bare response function @xmath308 is also shown ( dashed lines ) . note that in the low temperature phase @xmath309 changes sign in the space - like region and at the meson pole ( @xmath310 ) due to the sign change of the effective coupling strength ( [ lambda]).,title="fig : " ] ( [ response ] ) ( solid lines ) at @xmath278 . for comparison the bare response function @xmath308 is also shown ( dashed lines ) . note that in the low temperature phase @xmath309 changes sign in the space - like region and at the meson pole ( @xmath310 ) due to the sign change of the effective coupling strength ( [ lambda]).,title="fig : " ] as we have noted , in the low temperature phase with non - vanishing condensate amplitude @xmath142 , the mesonic excitations couple with the quasi - particle excitations and this give the additional shift of the meson mass from the one obtained by the gap equation . the shift of the meson pole from the bare spectrum @xmath311 may be computed by the condition @xmath312 which in the long wavelength limit ( @xmath313 ) gives a solution at @xmath314 . the shift of the meson mass @xmath315 is determined by @xmath316 where we have used @xmath300 . in the case of @xmath317 this may be solved approximately @xmath318 the equation ( [ mass - shift ] ) may be solved by iteration starting from the first approximation ( [ mass - shift0 ] ) by inserting it for @xmath319 in the integrand on the right hand side . we plot in fig . 6 the meson mass shift determined by this method . the meson mass becomes smaller as @xmath320 increases and eventually becomes zero at the temperature which satisfies the condition @xmath321 the vanishing of the effective mass of mesonic excitation may be identified as the onset of instability in the metastable low temperature phase by small fluctuation ( spinodal decomposition ) . the same instability occurs when one approaches to @xmath322 from high temperature phase . we note that the temperature @xmath323 at the spinodal point given by ( [ spinodal ] ) does not coincide with the `` backbending '' temperature @xmath324 where @xmath325 diverges . the latter temperature is determined by the condition @xmath326 which can be obtained from the gap equation ( [ gap ] ) . we found that @xmath324 is slightly above @xmath323 and the unstable region appears associated with the lower solutions @xmath57 of the gap equation . these unstable solutions thus appear only in the region which is not easily accessible and may well be considered as another pathology of the mean field approximation . in the case of the second order transition of the landau - type . such behavior has been obtained in @xcite by a temperature - dependent loop expansion method formulated in @xcite . ] ) is the mass of the excitation determined by ( [ mass - shift ] ) and the dashed line is the solution of the gap equation evaluated at @xmath32710 . the slopes of two curves diverge at the temperature @xmath328 while @xmath60 vanishes at @xmath329 . ] in this paper , we have developed a kinetic theory for a system of interacting quantum fields in the mean field approximation taking into account the existence of quasi - particle excitations . we have obtained a set of coupled equations of motion , one for the meson condensate in the form of non - linear klein - gordon equation which contains extra non - linearlity due to the particle excitations . the equations of motion for quasi - particle excitations are described in terms of the wigner functions , which reduce to a semiclassical vlasov equation for one - particle distribution function with a modification due to the coherent pair - creation and pair - annihilation expressed by the off - diagonal components in the wigner function . these off - diagonal components may be eliminated by a suitable bogoliubov transformation of the particle creation and annihilation operators for a uniform , time - independent system . however , they remain non - vanishing in general non - uniform , time - dependent systems . we have shown that in equilibrium these equations are reduced to a gap equation in the hartree approximation . this implies that our kinetic equations are natural extension of the hartree approximation to the non - equilibrium situation . it is well known , however , that in this approximation the phase transition becomes first order . this is a generic feature of the mean field approximation @xcite which persists also for multi - components scalar fields with continuous @xmath1 symmetry for arbitrary finite integer @xmath330 . in the present work , we studied also the excitations spectrum in the system near equilibrium . in the high temperature phase there is no collective excitation mode in the system besides the two continua of the quasi - particle excitations in the entire space - like energy - momentum region and the time - like region above the pair creation threshold . in the low temperature phase we found that coupling of the meson pole to the quasi - particle continua give rise to the shift of the meson mass which becomes zero at the edge of the spinodal instability line of the first order transition . the spinodal point appears deep inside the meta - stable region of the first order transition which may well be an artifact of the mean field approximation . we found that the coupling to the off - diagonal components of the wigner functions should be properly included to avoid the appearance of the undamped tachyonic mode . in this work we used a single component real scalar field model which possesses only discrete symmetry . it is straightforward to extend the present analysis to models with continuous symmetry such as the sigma model with @xmath1 symmetry . basic features of the present analysis is preserved in such extension . it is known however that the goldstone theorem is apparently violated in this approximation for a system with continuous symmetry . we will show in the forthcoming paper @xcite , that the missing nambu - goldstone mode may be retrieved in the collective excitations of the system . our coupled kinetic equations may be solved for an arbitrary initial conditions . we plan to study the freezeout dynamics with these equations with more realistic interactions . it would be interesting to see in particular how much flow is generated by the acceleration by the mean field as the vacuum condensate is restored . 0.5 cm * acknowledgment * we thank gordon baym , hirotsugu fujii , tetsuo hatsuda , tetsufumi hirano , osamu morimatsu and koichi ohta for helpful discussions . we are indebted to gordon baym for correcting a sign error in the original manuscript and to tetsuo hatsuda and osama morimatsu for calling our attention to some works on the spectral functions at finite temperature which prompted us to extend our analyses of the excitation spectrum in section 6 . we are also grateful to francois gelis , kazunori itakura , yusuke kato , and tetsuro nikuni for their interests in this work . this work has been supported by the grants - in - aid of the japanese ministry of education , culture , sports , science and technology no . 13440067 and 19540269 . here we present some details of the computation of the equation of motion of the wigner functions and show that with the gaussian ansatz for the density matrix the result is equivalent to what we obtain from the mean - field hamiltonian ( [ hmf ] ) . we like to compute the time - derivative of the operator product @xmath331 which appears in the definition of @xmath332 with @xmath333 and @xmath334 . @xmath335 a_{{{\mathbf p}}_2 } + a^\dagger_{{{\mathbf p}}_1 } [ a_{{{\mathbf p}}_2 } , h ] \end{aligned}\ ] ] the commutators @xmath336 $ ] is decomposed into a sum of the commutators with @xmath85 ( @xmath337 ) among which the commutators with @xmath89 and @xmath90 do not survive the average with the gaussian density matrix since they only contain the odd power of the field operators . we only need to compute the commutators with @xmath91 and @xmath92 . to compute the commutator with @xmath91 it is convenient to rewrite @xmath91 as @xmath338 we then find @xmath339 & = & \omega_{{{\mathbf p } } } a_{{\mathbf p}}(t ) + \frac{\lambda}{4 } \int d { { \mathbf r}}\phi_c^2 ( { { \mathbf r } } , t ) [ a_{{\mathbf p}}(t ) , { { \tilde \phi}}^2 ( { { \mathbf r } } , t ) ] \nonumber \\ & = & \omega_{{{\mathbf p } } } a_{{\mathbf p}}(t ) + \frac{\lambda}{2 } \int d { { \mathbf r}}\phi_c^2 ( { { \mathbf r } } , t ) \frac{e^{- i { { \mathbf p}}\cdot { { \mathbf r}}}}{\sqrt{2 \omega_{{{\mathbf p } } } } } { { \tilde \phi}}({{\mathbf r } } , t ) \\ \left [ a_{{\mathbf p}}^\dagger ( t ) , h_2 \right ] & = & - \omega_{{{\mathbf p } } } a_{{\mathbf p}}^\dagger ( t ) + \frac{\lambda}{4 } \int d { { \mathbf r}}\phi_c^2 ( { { \mathbf r } } , t ) [ a_{{\mathbf p}}^\dagger ( t ) , { { \tilde \phi}}^2 ( { { \mathbf r } } , t ) ] \nonumber \\ & = & - \omega_{{{\mathbf p } } } a_{{\mathbf p}}(t ) - \frac{\lambda}{2 } \int d { { \mathbf r}}\phi_c^2 ( { { \mathbf r } } , t ) \frac{e^ { i { { \mathbf p}}\cdot { { \mathbf r}}}}{\sqrt{2 \omega_{{{\mathbf p } } } } } { { \tilde \phi}}({{\mathbf r } } , t ) \end{aligned}\ ] ] where we have used the following formulae : @xmath340 & = & n { { \tilde \phi}}^{n-1 } ( { { \mathbf r } } , t ) [ a_{{\mathbf p}}^\dagger , { { \tilde \phi}}({{\mathbf r}},t ) ] = - n { { \tilde \phi}}^{n-1 } ( { { \mathbf r } } , t ) \frac{e^ { i { { \mathbf p}}\cdot { { \mathbf r}}}}{\sqrt{2 \omega_{{{\mathbf p } } } } } \\ \left [ a_{{\mathbf p } } , { { \tilde \phi}}^n ( { { \mathbf r } } , t ) \right ] & = & n \phi^{n-1 } ( { { \mathbf r}},t ) [ a_{{\mathbf p } } , { { \tilde \phi}}({{\mathbf r}},t ) ] = n \phi^{n-1 } ( { { \mathbf r}},t ) \frac{e^{- i { { \mathbf p}}\cdot { { \mathbf r}}}}{\sqrt{2 \omega_{{{\mathbf p } } } } } \end{aligned}\ ] ] with these results we obtain @xmath341 & = & - ( \omega_{{{\mathbf p}}_1 } - \omega_{{{\mathbf p}}_2 } ) a^\dagger_{{{\mathbf p}}_1 } a_{{{\mathbf p}}_2 } \nonumber \\ & & \quad - \frac{\lambda}{2 } \int d { { \mathbf r}}\phi_c^2 ( { { \mathbf r } } , t ) { { \tilde \phi}}({{\mathbf r } } , t ) \frac{e^ { i { { \mathbf p}}_1 \cdot { { \mathbf r}}}}{\sqrt{2 \omega_{{{\mathbf p}}_1 } } } a_{{{\mathbf p}}_2 } \nonumber \\ & & \quad \qquad + \frac{\lambda}{2 } \int d { { \mathbf r}}\phi_c^2 ( { { \mathbf r } } , t ) { { \tilde \phi}}({{\mathbf r } } , t ) \frac{e^{- i { { \mathbf p}}_2 \cdot { { \mathbf r}}}}{\sqrt{2 \omega_{{{\mathbf p}}_2 } } } a_{{{\mathbf p}}_1}^\dagger \end{aligned}\ ] ] the gaussian average of this equation gives @xmath342 { \rangle } & = & - ( \omega_{{{\mathbf p}}_1 } - \omega_{{{\mathbf p}}_2 } ) { \langle}a^\dagger_{{{\mathbf p}}_1 } a_{{{\mathbf p}}_2 } { \rangle}\nonumber \\ & & \quad - \frac{\lambda}{2 } \int d { { \mathbf r}}\phi_c^2 ( { { \mathbf r } } , t ) \frac{e^ { i { { \mathbf p}}_1 \cdot { { \mathbf r}}}}{\sqrt{2 \omega_{{{\mathbf p}}_1 } } } { \langle}{{\tilde \phi}}({{\mathbf r } } , t ) a_{{{\mathbf p}}_2 } { \rangle}\nonumber \\ & & \quad \qquad + \frac{\lambda}{2 } \int d { { \mathbf r}}\phi_c^2 ( { { \mathbf r } } , t ) \frac{e^{- i { { \mathbf p}}_2 \cdot { { \mathbf r}}}}{\sqrt{2 \omega_{{{\mathbf p}}_2 } } } { \langle}{{\tilde \phi}}({{\mathbf r } } , t ) a_{{{\mathbf p}}_1}^\dagger { \rangle}\end{aligned}\ ] ] next we compute the commutators with @xmath92 . we first compute @xmath343 & = & \frac{\lambda}{3 ! } \int d { { \mathbf r}}{{\tilde \phi}}^3 ( { { \mathbf r } } , t ) \frac{e^ { i { { \mathbf p}}\cdot { { \mathbf r}}}}{\sqrt{2 \omega_{{{\mathbf p } } } } } a_{{\mathbf p}}(t ) \\ \left [ a_{{\mathbf p}}^\dagger ( t ) , h_4 \right ] & = & - \frac{\lambda}{3 ! } \int d { { \mathbf r}}{{\tilde \phi}}^3 ( { { \mathbf r } } , t ) \frac{e^ { i { { \mathbf p}}\cdot { { \mathbf r}}}}{\sqrt{2 \omega_{{{\mathbf p } } } } } a_{{\mathbf p}}(t ) \end{aligned}\ ] ] we then use these results to obtain @xmath344 & = & - \frac{\lambda}{3 ! } \int d { { \mathbf r}}{{\tilde \phi}}^3 ( { { \mathbf r } } , t ) \frac{e^ { i { { \mathbf p}}_1 \cdot { { \mathbf r}}}}{\sqrt{2 \omega_{{{\mathbf p}}_1 } } } a_{{{\mathbf p}}_2 } \nonumber \\ & & \qquad + \frac{\lambda}{3 ! } \int d { { \mathbf r}}{{\tilde \phi}}^3 ( { { \mathbf r } } , t ) \frac{e^{- i { { \mathbf p}}_2 \cdot { { \mathbf r}}}}{\sqrt{2 \omega_{{{\mathbf p}}_2 } } } a_{{{\mathbf p}}_1}^\dagger \end{aligned}\ ] ] taking the average with the gaussian density matrix , we find @xmath345 { \rangle } & = & - \frac{\lambda}{2 } \int d { { \mathbf r}}{\langle}{{\tilde \phi}}^2 ( { { \mathbf r } } , t ) { \rangle}\frac{e^ { i { { \mathbf p}}_1 \cdot { { \mathbf r}}}}{\sqrt{2 \omega_{{{\mathbf p}}_1 } } } { \langle}{{\tilde \phi}}({{\mathbf r } } , t ) a_{{{\mathbf p}}_2 } { \rangle}\nonumber \\ & & \qquad + \frac{\lambda}{2 } \int d { { \mathbf r}}{\langle}{{\tilde \phi}}^2 ( { { \mathbf r } } , t ) { \rangle}\frac{e^{- i { { \mathbf p}}_2 \cdot { { \mathbf r}}}}{\sqrt{2 \omega_{{{\mathbf p}}_2 } } } { \langle}{{\tilde \phi}}({{\mathbf r } } , t ) a_{{{\mathbf p}}_1}^\dagger { \rangle}\end{aligned}\ ] ] we observe the similarity between the last two terms of the commutator with @xmath91 and the commutator with @xmath92 after gaussian average . these terms can be combined by introducing the self - energy function , @xmath346 integration over the space coordinate @xmath347 give a delta function @xmath348 and @xmath349 . performing the integral over @xmath41 yields @xmath350 { \rangle } & = & - ( \omega_{{{\mathbf p}}_1 } - \omega_{{{\mathbf p}}_2 } ) { \langle}a^\dagger_{{{\mathbf p}}_1 } a_{{{\mathbf p}}_2 } { \rangle}\nonumber \\ & & \quad - \frac{\lambda}{4 } \sum_{{\mathbf q}}\frac{\pi_{{\mathbf q}}}{\sqrt{\omega_{{{\mathbf p}}_1 + { { \mathbf q}}}\omega_{{{\mathbf p}}_2 } } } ( { \langle}a_{-{{\mathbf p}}_1-{{\mathbf q } } } a_{{{\mathbf p}}_2 } { \rangle}+ { \langle}a_{{{\mathbf p}}_1 + { { \mathbf q}}}^\dagger a_{{{\mathbf p}}_2 } { \rangle } ) \nonumber \\ & & \qquad + \frac{\lambda}{4 } \sum_{{\mathbf q}}\frac{\pi_{{\mathbf q}}}{\sqrt { \omega_{{{\mathbf p}}_1 } \omega_{p_2 - { { \mathbf q } } } } } ( { \langle}a_{{{\mathbf p}}_2 - { { \mathbf q } } } a_{{{\mathbf p}}_1}^\dagger { \rangle}+ { \langle}a_{- { { \mathbf p}}_2 + { { \mathbf q}}}^\dagger a_{{{\mathbf p}}_1}^\dagger { \rangle } ) \nonumber \\\end{aligned}\ ] ] this result coincides with the average of the commutator with the mean field hamiltonian @xmath108 defined by ( [ hmf ] ) . @xmath351 { \rangle}= { \langle } [ a^\dagger_{{{\mathbf p}}_1 } a_{{{\mathbf p}}_2 } , h_{\rm mf } ] { \rangle}\ ] ] commutators of the four bilinear operator products with the mean - field hamiltonian are listed below : @xmath98 & = & - ( \omega_{{{\mathbf p}}_1 } - \omega_{{{\mathbf p}}_2 } ) a^\dagger_{{{\mathbf p}}_1 } a_{{{\mathbf p}}_2 } \nonumber \\ & & \quad - \frac{1}{2 } \sum_{{\mathbf q}}\frac{\pi_{{\mathbf q}}}{\sqrt{\omega_{{{\mathbf p}}_1 + { { \mathbf q}}}\omega_{{{\mathbf p}}_2 } } } ( a_{-{{\mathbf p}}_1-{{\mathbf q } } } a_{{{\mathbf p}}_2 } + a_{{{\mathbf p}}_1 + { { \mathbf q}}}^\dagger a_{{{\mathbf p}}_2 } ) \nonumber \\ & & \qquad + \frac{1}{2 } \sum_{{\mathbf q}}\frac{\pi_{{\mathbf q}}}{\sqrt { \omega_{{{\mathbf p}}_1 } \omega_{{{\mathbf p}}_2 - { { \mathbf q } } } } } ( a_{{{\mathbf p}}_1}^\dagger a_{{{\mathbf p}}_2 - { { \mathbf q } } } + a_{{{\mathbf p}}_1}^\dagger a_{- { { \mathbf p}}_2 + { { \mathbf q}}}^\dagger ) \nonumber \\ \\ \left [ a_{{{\mathbf p}}_1 } a_{{{\mathbf p}}_2 } , h_{\rm mf } \right ] & = & ( \omega_{{{\mathbf p}}_1 } + \omega_{{{\mathbf p}}_2 } ) a_{{{\mathbf p}}_1 } a_{{{\mathbf p}}_2 } \nonumber \\ & & \quad + \frac{1}{2 } \sum_{{\mathbf q}}\frac{\pi_{{\mathbf q}}}{\sqrt{\omega_{{{\mathbf p}}_1 + { { \mathbf q}}}\omega_{{{\mathbf p}}_2 } } } ( a_{{{\mathbf p}}_1-{{\mathbf q } } } + a_{-{{\mathbf p}}_1 + { { \mathbf q}}}^\dagger ) a_{{{\mathbf p}}_2 } \nonumber \\ & & \qquad + \frac{1}{2 } \sum_{{\mathbf q}}\frac{\pi_{{\mathbf q}}}{\sqrt { \omega_{{{\mathbf p}}_1 } \omega_{{{\mathbf p}}_2 - { { \mathbf q } } } } } a_{{{\mathbf p}}_1 } ( a_{{{\mathbf p}}_2 - { { \mathbf q } } } + a_{- { { \mathbf p}}_2 + { { \mathbf q}}}^\dagger ) \nonumber \\ \\ \left [ a^\dagger_{{{\mathbf p}}_1 } a^\dagger_{{{\mathbf p}}_2 } , h_{\rm mf } \right ] & = & - ( \omega_{{{\mathbf p}}_1 } + \omega_{{{\mathbf p}}_2 } ) a^\dagger_{{{\mathbf p}}_1 } a_{{{\mathbf p}}_2 } \nonumber \\ & & \quad - \frac{1}{2 } \sum_{{\mathbf q}}\frac{\pi_{{\mathbf q}}}{\sqrt{\omega_{{{\mathbf p}}_1 + { { \mathbf q}}}\omega_{{{\mathbf p}}_2 } } } ( a_{-{{\mathbf p}}_1-{{\mathbf q } } } a_{{{\mathbf p}}_2 } + a_{{{\mathbf p}}_1 + { { \mathbf q}}}^\dagger a_{{{\mathbf p}}_2 } ) \nonumber \\ & & \qquad - \frac{1}{2 } \sum_{{\mathbf q}}\frac{\pi_{{\mathbf q}}}{\sqrt { \omega_{{{\mathbf p}}_1 } \omega_{{{\mathbf p}}_2 - { { \mathbf q } } } } } ( a_{{{\mathbf p}}_2 - { { \mathbf q } } } a_{{{\mathbf p}}_1}^\dagger + a_{- { { \mathbf p}}_2 + { { \mathbf q}}}^\dagger a_{{{\mathbf p}}_1}^\dagger ) \nonumber \\ \\ \left [ a_{{{\mathbf p}}_1 } a^\dagger_{{{\mathbf p}}_2 } , h_{\rm mf } \right ] & = & ( \omega_{{{\mathbf p}}_1 } - \omega_{{{\mathbf p}}_2 } ) a_{{{\mathbf p}}_1 } a^\dagger_{{{\mathbf p}}_2 } \nonumber \\ & & \quad + \frac{1}{2 } \sum_{{\mathbf q}}\frac{\pi_{{\mathbf q}}}{\sqrt{\omega_{{{\mathbf p}}_1 + { { \mathbf q}}}\omega_{{{\mathbf p}}_2 } } } ( a_{{{\mathbf p}}_1-{{\mathbf q } } } + a_{- { { \mathbf p}}_1 + { { \mathbf q}}}^\dagger ) a^\dagger_{{{\mathbf p}}_2 } \nonumber \\ & & \qquad - \frac{1}{2 } \sum_{{\mathbf q}}\frac{\pi_{{\mathbf q}}}{\sqrt { \omega_{{{\mathbf p}}_1 } \omega_{{{\mathbf p}}_2 - { { \mathbf q } } } } } a_{{{\mathbf p}}_1}^\dagger ( a_{{{\mathbf p}}_2 - { { \mathbf q } } } + a_{- { { \mathbf p}}_2 + { { \mathbf q}}}^\dagger ) \nonumber \\\end{aligned}\ ] ] # 1ann . phys . * # 1 * # 1phys . rep . * # 1 * # 1act . phys . pol . * # 1 * # 1prog . phys . * # 1 * # 1prog . phys . supplement*#1 * # 1phys . rev . * # 1 * # 1phys . rev . * d#1 * # 1phys . rev . * c#1 * # 1phys . rev . * a#1 * # 1phys . * # 1 * # 1phys . # 1 * # 1rev . phys.*#1 * # 1nucl . phys . * # 1 * # 1nuovo cimento * # 1 * # 1sov . # 1 * # 1eur . j. * a#1 * # 1coll . a#1 * j. d. bjorken , _ phys . rev . _ * d27 * , 865 ( 1983 ) ; g. baym , b. l. friman , j .- blaizot , m. soyeur , w. czy , _ nucl . _ * a407 * , 541 ( 1983 ) ; k. kajantie and l. mclerran , , 261 ( 1983 ) ; m. gyulassy , t. matsui , _ phys . rev . _ * d29 * , 419 ( 1984 )

recently , the atlas @xcite and cms @xcite collaborations have both reported an excess in the diphoton channel near @xmath0 gev . with an integrated luminosity of 3.2 @xmath2 , the atlas collaboration has observed a local @xmath3 excess at a diphoton invariant mass of around 747 gev , assuming a narrow width resonance . for a wider width resonance , the signal significance increases to @xmath4 with a preferred width of about 45 gev . with an integrated luminosity of 2.6 @xmath2 , the cms collaboration has also observed a diphoton excess with a local significance of @xmath5 at invariant mass of around 760 gev . assuming a decay width of around 45 gev , the significance reduces to @xmath6 in this case . the corresponding excesses in the cross section can be roughly estimated as @xmath7 @xcite . while this is well below the threshold to claim a discovery , this excess could be the first signal of physics beyond the standard model . as such , it is worthwhile to consider possible models of new physics which may explain the excess . indeed , many groups have proposed such possible explanations @xcite perhaps the simplest explanation for the excess is the addition of a sm singlet scalar with a mass near @xmath0 gev along with additional vector - like multiplets of colored particles . with this set - up , the singlet may be produced via gluon fusion with the vector - like particles appearing in loops . similarly , the singlet may decay to diphotons . however , in order to preserve gauge coupling unification as in supersymmetric versions of the sm , these vector - like states should come in complete multiplets of @xmath1 . moreover , to preserve unification and avoid landau poles , the types and numbers of @xmath1 multiplets is restricted @xcite . such light vector - like multiplets are often found in models constructed within the framework of string theory@xcite . indeed , vector - like states are generically present in intersecting / magnetized d - brane models on orientifold backgrounds @xcite . one such model satisfying all global consistency conditions has been constructed from intersecting / magnetized d - branes within the context of type ii orientifold compactifications @xcite on a @xmath8 background . this model corresponds to the mssm with three generations of quarks and leptons as well as a single pair of higgs fields . the model contains a minimal amount of exotic matter , which may be decoupled from the low - energy sector . in addition , the tree - level gauge couplings are automatically unified at the string scale @xcite . finally , the yukawa couplings are allowed by global @xmath9 symmetries , and it is possible to obtain correct masses and mixings for quarks and charged leptons . thus , this is a phenomenologically interesting model worthy of detailed study . in the following , we briefly summarize the intersecting d - brane model under study , which is a variation of the model discussed above . it is shown that vector - like quarks are present in the model , and that these states may appear in complete multiplets of @xmath1 so that gauge coupling unification may be maintained . furthermore , it is shown that there are sm singlets in the model which have yukawa couplings to the vector - like quarks and leptons . it should be emphasized that this is a non - trivial result as these fields carry global @xmath9 charges , under which the yukawa coupling must be neutral . finally , we show that the diphoton excesses can be accomodated in the model . a phenomenologically interesting intersecting d - brane model has been studied in refs . @xcite . a variation of this model with a different hidden sector was also studied in refs . type iia orientifold string compactifications with intersecting d - branes ( and their type iib duals with magnetized d - branes ) have provided exciting geometric tools with which the mssm may be engineered . while this approach may not allow a first - principles understanding of why the sm gauge groups and associated matter content arises , it may allow a deeper insight into how the finer phenomenological details of the sm may emerge . in short , d6-branes in type iia fill ( 3 + 1)-dimensional minkowski spacetime and wrap 3-cycles in the compactified manifold , such that a stack of @xmath10 branes generates a gauge group u(@xmath10 ) [ or u(@xmath11 ) in the case of @xmath12 in its world volume . in general , the 3-cycles wrapped by the stacks of d6-branes intersect multiple times in the internal space , resulting in a chiral fermion in the bifundamental representation localized at the intersection between different stacks @xmath13 and @xmath14 . the multiplicity of such fermions is then given by the number of times the 3-cycles intersect . each stack of d6-branes @xmath13 may intersect the orientifold images of other stacks @xmath15 , also resulting in fermions in bifundamental representations . each stack may also intersect its own image @xmath16 , resulting in chiral fermions in the symmetric and antisymmetric representations . non - chiral matter may also be present between stacks of d - branes which do not intersect on one two - torus . a zero intersection number between two stacks of branes implies that the branes are parallel on at least one torus . at such kind of intersection additional non - chiral ( vector - like ) multiplet pairs from @xmath17 , @xmath18 , and @xmath19 can arise . global consistency of the model requires certain constraints to be satisfied , namely , ramond - ramond ( r - r ) tadpole cancellation and the preservation of @xmath20 supersymmetry . in particular , the conditions for preserving @xmath20 supersymmetry fixes the complex structure parameters . .d6-brane configurations and intersection numbers for a three - family pati - salam model on a type - iia @xmath21 orientifold , with a tilted third two - torus . the complete gauge symmetry is @xmath22_{\rm observable } \times [ { \rm u}(2 ) \times { \rm usp}(2)^2]_{\rm hidden}$ ] and @xmath20 supersymmetry is preserved for @xmath23 , @xmath24 , @xmath25 . [ cols="^,^,^,^,^,^,^,^,^,^,^,^,^ " , ] vector - like matter appears in intersecting / magnetized d - brane models on toroidal orientifolds between stacks of d - branes which do not intersect . the mass of such vector - like states depends upon the separation between the stacks of d - branes in the internal space . as such , it is generically massive . only stacks of d - branes which are directly on top of one another have massless vector - like states between them . in the model studied in section ii , the toroidal orientifold consist of a six - torus which is factorizable , @xmath26 . if two stacks of d - branes are parallel on one two - torus , then vector - like matter appears in the bifundamental representation of the gauge groups within the world - volume of each stack . if the two stacks are not separated on the two torus on which they are parallel , the vector - like multiplets are massless however , these states become massive if the stacks are separated . the most straightforward way to obtain the diphoton excesses is with a sm singlet scalar with a mass @xmath27 gev coupled to vector - like quarks . the coupling to vector - like quarks is necessary in order to produce the scalar via gluon fusion and to allow the decay of the scalar into diphotons . the requisite vector - like quarks are indeed present in the model . in particular , the vector - like quarks and leptons in the @xmath28 and @xmath29 sector fill @xmath30 and @xmath31 spinorial representations of so(10 ) , or equivalently @xmath32 of su(5 ) . for example , @xmath33 it is well - known that gauge coupling unification may be preserved at the 1-loop level if the extra matter comes in complete representations of @xmath1 . however , at the 2 or 3-loop level , a landau pole may appear . this restricts the number of @xmath1 multiplets which may remain light to either one @xmath34 or three copies of @xmath35 . in addition , any number of sm singlets may be present in the light spectrum . generically , there are many more vector - like states in the spectrum , as can be seen from tables [ vector - likespectrummssm1],[vector - likespectrummssm2 ] , and [ vector - likespectrummssm3 ] . thus , in order to preserve gauge coupling unification , many of these states must obtain string - scale masses , while simultaneously maintaining light masses for either of the two cases stated above . recalling that vector - like matter appears between stacks of d - branes which are parallel on one two - torus , the masses of these vector - like states depend on the separation between the stacks on the two - torus on which they are parallel . so , it is possible to give string - scale masses to some of the vector - like states by separating the two stacks of d - branes on the two - torus on which they are parallel and where these vector - like states are localized . is it possible to choose the position of the stacks of d - branes so that only the @xmath36 or @xmath37 fields remain light ? the answer to this question is yes . as an example , let us consider just the fields in the @xmath28 and @xmath29 sectors , where it should be noted that these stacks of d - branes are parallel on the third two - torus . thus , for example the fields @xmath38 may become massive by separating stacks @xmath39 and @xmath15 on the third torus . similarly , the fields @xmath40 may become massive by separating stacks @xmath39 and @xmath41 , while @xmath42 become massive if stacks @xmath43 and @xmath44 are separated on the third two - torus . the fields @xmath45 may remain massless if stacks @xmath39 and @xmath44 overlap on the third two - torus , and the same is true for @xmath46 if stacks @xmath43 and @xmath15 overlap on the third two - torus . thus with this configuration , only the fields in the @xmath47 and @xmath48 ( with @xmath49 ) representations of eq . [ abvector ] remain light . in addition , there are additional vector - like quarks and leptons in the spectrum in the @xmath50 and @xmath51 sectors , as shown in tables [ vector - likespectrummssm2 ] and [ vector - likespectrummssm3 ] . these fields may also become massive by separating the relevant stacks of d - branes on the two - torus where they are parallel . for example , stacks @xmath39 and @xmath43 are parallel with stacks @xmath52 and @xmath53 on the third two - torus , and so the vector - like matter in these sectors may be eliminated if these stacks are separated on the third two - torus . similarly , stacks @xmath39 and @xmath43 are parallel with stacks @xmath54 and @xmath55 on the second two - torus , and the vector - like matter in this sector may be eliminated by displacing these stacks on the second two - torus . on the other hand , it must also be kept in mind that splitting the stacks in this fashion may affect the higgs sectors , which are also vector - like . for example the higgs fields @xmath56 and @xmath57 arise from vector - like matter in the @xmath58 and @xmath59 sectors respectively . stackss @xmath39 and @xmath44 must overlap to keep @xmath60 light while stacks @xmath39 and @xmath15 must be separated to give masses to the fields @xmath61 ) such that the light vector - like fields may be placed in @xmath36 representations . however , this implies that stacks @xmath14 and @xmath62 must be separated and thus that the higgs field @xmath57 is not present in the light spectrum . clealy , then it is not possible to eliminate the fields in the @xmath63 while keeping the fields in the @xmath64 without also eliminating some of the higgs field . note that this constraint only applies to those fields in the @xmath28 and @xmath29 sectors . in the next section , we shall also consider the vector - like matter in the @xmath50 and @xmath51 sectors . we shall find that it is possible to have just vector - like matter in the @xmath64 representation without eliminating some of the higgs fields as a by - product . we have seen in the previous section that the model contains vector - like quarks and leptons in the @xmath28 and @xmath29 sectors which may be grouped into complete representations of @xmath1 . furthermore , by displacing the stacks of d - branes , it is possible to eliminate some of these fields from the spectrum so that only three copies of @xmath65 remain light . in this way , gauge coupling unification may be maintained while also avoiding the landau pole problem . however , upon inspection , none of these vector - like quarks and leptons appear to have yukawa couplings with any of the singlet fields present in the model . therefore , they may not be involved in producing the diphoton excesses . however , this is not the case if we examine the vector - like quarks and leptons present in the @xmath50 and @xmath51 sectors . let us turn our attention to the sm singlet fields @xmath66 and @xmath67 shown in table [ hiddenspectrum ] . these fields have yukawa couplings with the vector - like quarks in the @xmath50 and @xmath51 sectors : @xmath68 where @xmath69 and @xmath70 are near the string scale @xmath71 as discussed in section i. in addition , there are yuakwa couplings between these singlets and some of the vector - like doublets : @xmath72 finally , there are additional yukawa couplings between @xmath66 and @xmath67 and the singlet fields in the @xmath73 and @xmath74 sectors : @xmath75 thus , the required couplings of the sm singlet fields to vector - like quarks are present so that the singlet fields may be produced by gluon fusion and decay to diphotons . in addition , the singlet fields have additional couplings to other doublet and singlet fields . note that the vector - like quarks involved in these couplings combined with additional vector - like matter may be placed in complete representations of @xmath1 by replacing the right - handed vector - like quarks , leptons , and singlets in eq . [ abvector ] with some of the quarks , leptons and singlets of tables [ vector - likespectrummssm2 ] and [ vector - likespectrummssm3 ] . for example , making the interchanges @xmath76 we have @xmath77 where @xmath78 , and making the interchanges @xmath79 we have @xmath80 in order to have just three @xmath65 multiplets + additional singlets in the light spectrum , the fields grouped into the @xmath81 multiplets must become massive , as well as any additional vector - like states . to eliminate the vectorlike fields in the @xmath28 and @xmath29 sectors from the light spectrum , we must require that stacks @xmath39 and @xmath43 are separated from stacks @xmath15 , @xmath41 , and @xmath44 on the third two - torus . futhermore , to eliminate the fields in the above @xmath37 , we require that stack @xmath39 be separated from stack @xmath52 on the third two - torus and from @xmath54 on the second two - torus . let us also require that stack @xmath43 be separated from both stacks @xmath52 , @xmath54 , @xmath53 and @xmath55 . to keep the fields in the above @xmath65 representations light , we require that stack @xmath39 overlap with stack @xmath53 on the third two - torus and with stack @xmath55 on the second two torus . in addition , we require that stack @xmath14 overlap stacks @xmath82 , @xmath41 , @xmath62 , and @xmath44 as well as stacks @xmath54 . this configuration assures that the higgs fields @xmath83 and @xmath84 are present in the spectrum . in addition , stack @xmath14 may not overlap stack @xmath54 since stack @xmath39 overlaps stack @xmath52 . since stack @xmath39 and @xmath15 are separated , this implies that stacks @xmath14 and @xmath54 are also separated . then , the fields in the light spectrum consist of the following fields with quantum numbers under the @xmath85 gauge symmetry shown : @xmath86 plus additional singlets . as shown in eq . [ vectorquarksingletyukawa ] , the vector - like quarks @xmath87 and @xmath88 have yukawa couplings with the singlet field @xmath89 . thus , this singlet field may be produced via loops involving these vector - like quarks as well as decay to diphotons via gluon fusion . in addition , the singlet field s has couplings to the doublets @xmath90 to which it may also decay . using the notation of eq . [ lightreps ] , the superpotential for the extra vector - like states and the singlet @xmath91 is @xmath92 the corresponding supersymmetry breaking soft terms are @xmath93 @xmath94&@xmath95&@xmath96&@xmath97&@xmath98&@xmath99&@xmath100&@xmath101&@xmath102&@xmath103&@xmath104&@xmath105&@xmath106&@xmath107&@xmath108&@xmath109&@xmath110&@xmath111&@xmath112 + @xmath113&@xmath114&@xmath115&@xmath116&@xmath117&@xmath118&@xmath119&@xmath120&@xmath121&@xmath122&@xmath123&@xmath124&@xmath125&@xmath126&@xmath127&@xmath128&@xmath129&@xmath130&@xmath131 + @xmath132&@xmath133&@xmath134&@xmath135&@xmath136&@xmath137&@xmath138&@xmath120&@xmath139&@xmath140&@xmath141&@xmath142&@xmath143&@xmath144&@xmath145&@xmath146&@xmath147&@xmath148&@xmath149 + @xmath113&@xmath150&@xmath151&@xmath152&@xmath153&@xmath154&@xmath148&@xmath120&@xmath121&@xmath155&@xmath156&@xmath157&@xmath158&@xmath159&@xmath160&@xmath161&@xmath162&@xmath163&@xmath164 + @xmath165&@xmath166&@xmath134&@xmath135&@xmath167&@xmath168&@xmath169&@xmath120&@xmath170&@xmath171&@xmath172&@xmath173&@xmath174&@xmath175&@xmath176&@xmath177&@xmath178&@xmath179&@xmath180 + @xmath181&@xmath133&@xmath151&@xmath182&@xmath183&@xmath184&@xmath169&@xmath120&@xmath185&@xmath140&@xmath186&@xmath187&@xmath143&@xmath188&@xmath189&@xmath146&@xmath190&@xmath191&@xmath192 + @xmath193&@xmath194&@xmath195&@xmath135&@xmath167&@xmath196&@xmath119&@xmath120&@xmath197&@xmath198&@xmath199&@xmath200&@xmath201&@xmath144&@xmath202&@xmath203&@xmath204&@xmath205&@xmath206 + @xmath193&@xmath114&@xmath207&@xmath135&@xmath208&@xmath209&@xmath119&@xmath120&@xmath197&@xmath122&@xmath210&@xmath211&@xmath125&@xmath212&@xmath213&@xmath128&@xmath214&@xmath215&@xmath216 + [ tab:5gev ] @xmath94&@xmath95&@xmath96&@xmath97&@xmath98&@xmath99&@xmath100&@xmath101&@xmath102&@xmath103&@xmath104&@xmath105&@xmath106&@xmath107&@xmath108&@xmath109&@xmath110&@xmath111&@xmath112 + @xmath113&@xmath114&@xmath134&@xmath217&@xmath218&@xmath133&@xmath169&@xmath120&@xmath121&@xmath122&@xmath219&@xmath220&@xmath125&@xmath221&@xmath222&@xmath223&@xmath224&@xmath225&@xmath119 + @xmath226&@xmath114&@xmath227&@xmath182&@xmath228&@xmath229&@xmath119&@xmath120&@xmath230&@xmath122&@xmath231&@xmath232&@xmath125&@xmath233&@xmath234&@xmath223&@xmath235&@xmath236&@xmath237 + @xmath238&@xmath114&@xmath195&@xmath239&@xmath240&@xmath241&@xmath169&@xmath120&@xmath242&@xmath122&@xmath243&@xmath244&@xmath125&@xmath245&@xmath199&@xmath223&@xmath246&@xmath236&@xmath247 + @xmath193&@xmath150&@xmath140&@xmath135&@xmath248&@xmath249&@xmath169&@xmath120&@xmath197&@xmath155&@xmath250&@xmath251&@xmath158&@xmath221&@xmath252&@xmath253&@xmath224&@xmath254&@xmath255 + @xmath256&@xmath150&@xmath257&@xmath217&@xmath258&@xmath259&@xmath119&@xmath120&@xmath260&@xmath155&@xmath261&@xmath262&@xmath158&@xmath245&@xmath263&@xmath253&@xmath224&@xmath264&@xmath191 + @xmath226&@xmath150&@xmath227&@xmath239&@xmath228&@xmath265&@xmath169&@xmath120&@xmath230&@xmath155&@xmath266&@xmath267&@xmath158&@xmath245&@xmath268&@xmath253&@xmath269&@xmath270&@xmath271 + @xmath272&@xmath150&@xmath273&@xmath274&@xmath275&@xmath276&@xmath169&@xmath120&@xmath277&@xmath155&@xmath278&@xmath279&@xmath158&@xmath280&@xmath281&@xmath253&@xmath282&@xmath283&@xmath284 + [ tab:45gev ] the 750 gev diphoton production cross - sections observed by the cms collaboration are @xmath285 at @xmath286 tev @xciteand @xmath287 at @xmath288 tev @xcite , while the atlas collaboration observed @xmath289 at @xmath286 tev @xcite and @xmath290 at @xmath288 tev @xcite . replicating the strategy of ref . @xcite , we constrain the total decay width to @xmath291 gev . to reproduce the observed production cross - sections , we constrain the model using @xmath292 . the effective loop - level couplings amongst the standard model gauge bosons and scalar @xmath91 are given by @xmath293 \label{lagrangian}\ ] ] where @xmath294 and @xmath295 are the photon and gluon field strength tensors , respectively , with @xmath296 . the effective operators are represented by @xmath297 and @xmath298 , which are written as @xmath299 \label{kem}\ ] ] @xmath300 \label{k3}\ ] ] where @xmath301 , @xmath302 , and @xmath303 , and the functions @xmath304 and @xmath305 are expressed as @xmath306 \\ & & f_{\widetilde{f } } = \chi \left [ -1 + \chi f(\chi ) \right ] \ \label{feq}\end{aligned}\ ] ] with the function @xmath307 denoted by @xmath308 the triangle loop functions @xmath309 are defined here as @xmath310 & \mbox{if } \chi \geq 1 \\ -\frac{1}{4 } \left [ \ln \frac{1 + \sqrt{1 - \chi}}{1 - \sqrt{1 - \chi } } - i \pi \right]^2 & \mbox{if } \chi < 1 . \end{cases } \label{loop_function}\end{aligned}\ ] ] the diphoton and digluon decay widths in @xmath311-@xmath1 are computed from @xmath312 @xmath313 the diphoton production cross - section is calculated from @xmath314 where k is the qcd k - factor , @xmath315 is the total decay width , @xmath316 is the proton - proton center of mass energy , and @xmath317 is the dimensionless partonic integral computed for an @xmath318 gev resonance , yielding @xmath319 at @xmath286 tev and @xmath320 at @xmath288 tev @xcite . we use the gluon fusion k - factor of 1.98 . we construct our intersecting d - brane model with the ( @xmath321,@xmath322 ) and ( @xmath323,@xmath324 ) vector - like particles , implementing three copies of the ( @xmath325,@xmath326 ) . for the calculations , we decompose the @xmath327 multiplet into its @xmath328 and @xmath329 components . given the null @xmath330 electric charge @xmath331 , no constraints can be placed on @xmath332 , @xmath333 , or @xmath334 in the model via the production cross - section calculations , so for simplicity we set @xmath335 and @xmath336 when computing the decay of the scalar @xmath91 directly to the @xmath330 multiplet . the multiplets @xmath337 and @xmath328 participate in the @xmath338 loop diagrams and only @xmath337 in the @xmath339 loops , hence there are 8 free parameters in the effective operators @xmath297 and @xmath298 consisting of the yukawa couplings @xmath340 , trilinear a term couplings @xmath341 , fermionic component masses @xmath342 , and scalar component masses @xmath343 . in total , there are 10 parameters to compute : @xmath344 though the supersymmetry breaking soft terms @xmath345 and @xmath346 can be trivially computed from the fermionic and scalar components using the following relations @xmath347 the freedom on the 8 free - parameters engenders a large d - brane model parameter space . we treat the fermionic component of the vector - like particle and its soft supersymmetry breaking term independently , such that @xmath348 . recent constraints at the lhc on vector - like @xmath349 quarks @xcite provide lower limits of around 735 gev for the @xmath321 multiplet , allowing a light @xmath350 multiplet , which contributes to invisible branching fractions when @xmath351 gev . for the large yukawa couplings @xmath352 and @xmath101 at the unification scale , there exist the quasi focus points with @xmath353 and @xmath354 at low energy , respectively @xcite . constraints are placed on the @xmath355 terms of @xmath356 and @xmath357 to preclude premature breaking of the @xmath358 gauge symmetry . we provide some sample benchmark points in table [ tab:5gev ] and table [ tab:45gev ] to illustrate that the model can accommodate the observed diphoton cross - section . note that the cross - sections @xmath111 and @xmath112 in tables [ tab:5gev ] - [ tab:45gev ] display a gain of 4.65 from 8 tev to 13 tev . we use values of the coupling constants at the @xmath359 scale in our calculations of @xmath360 and @xmath361 . the diphoton and digluon decay modes are only a small portion of the total decay width , as shown in tables [ tab:5gev ] - [ tab:45gev ] . these practical constraints on the diphoton and digluon decay widths allocate the majority of the total width to the other decay channels , for instance , to dark matter invisibly and/or @xmath362 and @xmath363 , kinematically allowed for both @xmath95 and @xmath332 less than 375 gev , which is reflected in the @xmath106 decay width and @xmath364 branching ratio in tables [ tab:5gev ] - [ tab:45gev ] . specifically , the decay width for the decay into pairs of fermions @xmath365 is given by @xmath366 where we take @xmath367 and @xmath368 in the calculations . we have studied the diphoton excesses near @xmath0 gev recently reported by the atlas and cms collaborations within the context of a phenomenologically interesting intersecting / magnetized d - brane model on a toroidal orientifold . we have shown that the model contains a sm singlet scalar as well as vector - like quarks and leptons . in addition , we have shown that the singlet scalar has yukawa couplings with vector - like quarks and leptons such that it may be produced in proton - proton collisions via gluon fusion as well as decay to diphotons through loops involving the vector - like quarks . moreover , the required vector - like quarks and leptons may appear in complete @xmath1 multiplets so that gauge coupling unification may be maintained . in particuar , we showed that we may have three copies of @xmath65 representations of @xmath1 in the light spectrum which are present in the model . finally , we showed that the diphoton excesses observed by the atlas and cms collaborations may be accommodated . it should be emphasized that we have obtained these results within the context of a complete , globally consistent string model . this particular model has many interesting phenomenological features such as automatic gauge coupling unification , realistic yukawa mass matrices for quarks and leptons , and mimimal exotic matter . it is of note that the singlet fields required to explain the diphoton signal arise from this extra matter . in addition , the required vector - like quarks are naturally present in the spectrum . finally , the yukawa couplings between the singlet fields and the vector - like quarks are allowed by the global symmetries arising from @xmath9 factors whose gauge bosons become heavy at the string scale via the green - schwarz mechanism , a result which is completely non - trivial . an interesting question is whether or not is it possible to obtain supersymmetry partner spectra from the model which take into account the light vector - like matter . as mentioned earlier , the vector - like quarks and leptons have yukawa couplings with the higgs fields in the model , and thus may raise the higgs mass by up to a few gev . this may alleviate the problem with electroweak fine - tuning . in addition , the vector - like matter affects the rge running of the soft masses . finally , the soft supersymmetry breaking masses may be calculated at the string scale in the model . thus , it would be very interesting to study the possible supersymmetry partner spectra obtainable in the model including the extra vector - like matter . we plan to study this in future work . k. harigaya and y. nomura , arxiv:1512.04850 [ hep - ph ] ; y. mambrini , g. arcadi and a. djouadi , arxiv:1512.04913 [ hep - ph ] ; m. backovic , a. mariotti and d. redigolo , arxiv:1512.04917 [ hep - ph ] ; a. angelescu , a. djouadi and g. moreau , arxiv:1512.04921 [ hep - ph ] ; y. nakai , r. sato and k. tobioka , arxiv:1512.04924 [ hep - ph ] ; s. knapen , t. melia , m. papucci and k. zurek , arxiv:1512.04928 [ hep - ph ] ; d. buttazzo , a. greljo and d. marzocca , arxiv:1512.04929 [ hep - ph ] ; a. pilaftsis , arxiv:1512.04931 [ hep - ph ] ; r. franceschini _ et al . _ , arxiv:1512.04933 [ hep - ph ] ; s. di chiara , l. marzola and m. raidal , arxiv:1512.04939 [ hep - ph ] . s. d. mcdermott , p. meade and h. ramani , arxiv:1512.05326 [ hep - ph ] ; r. benbrik , c. h. chen and t. nomura , arxiv:1512.06028 [ hep - ph ] ; j. ellis , s. a. r. ellis , j. quevillon , v. sanz and t. you , arxiv:1512.05327 [ hep - ph ] ; m. low , a. tesi and l. t. wang , arxiv:1512.05328 [ hep - ph ] ; b. bellazzini , r. franceschini , f. sala and j. serra , arxiv:1512.05330 [ hep - ph ] ; r. s. gupta , s. jger , y. kats , g. perez and e. stamou , arxiv:1512.05332 [ hep - ph ] ; c. petersson and r. torre , arxiv:1512.05333 [ hep - ph ] ; e. molinaro , f. sannino and n. vignaroli , arxiv:1512.05334 [ hep - ph ] ; q. h. cao , y. liu , k. p. xie , b. yan and d. m. zhang , arxiv:1512.05542 [ hep - ph ] ; s. matsuzaki and k. yamawaki , arxiv:1512.05564 [ hep - ph ] ; a. kobakhidze , f. wang , l. wu , j. m. yang and m. zhang , arxiv:1512.05585 [ hep - ph ] ; r. martinez , f. ochoa and c. f. sierra , arxiv:1512.05617 [ hep - ph ] ; p. cox , a. d. medina , t. s. ray and a. spray , arxiv:1512.05618 [ hep - ph ] ; d. becirevic , e. bertuzzo , o. sumensari and r. z. funchal , arxiv:1512.05623 [ hep - ph ] ; j. m. no , v. sanz and j. setford , arxiv:1512.05700 [ hep - ph ] ; s. v. demidov and d. s. gorbunov , arxiv:1512.05723 [ hep - ph ] ; w. chao , r. huo and j. h. yu , arxiv:1512.05738 [ hep - ph ] ; s. fichet , g. von gersdorff and c. royon , arxiv:1512.05751 [ hep - ph ] ; d. curtin and c. b. verhaaren , arxiv:1512.05753 [ hep - ph ] ; l. bian , n. chen , d. liu and j. shu , arxiv:1512.05759 [ hep - ph ] ; j. chakrabortty , a. choudhury , p. ghosh , s. mondal and t. srivastava , arxiv:1512.05767 [ hep - ph ] ; a. ahmed , b. m. dillon , b. grzadkowski , j. f. gunion and y. jiang , arxiv:1512.05771 [ hep - ph ] ; c. csaki , j. hubisz and j. terning , arxiv:1512.05776 [ hep - ph ] ; d. aloni , k. blum , a. dery , a. efrati and y. nir , arxiv:1512.05778 [ hep - ph ] ; y. bai , j. berger and r. lu , arxiv:1512.05779 [ hep - ph ] ; e. gabrielli , k. kannike , b. mele , m. raidal , c. spethmann and h. veerme , arxiv:1512.05961 [ hep - ph ] ; j. s. kim , j. reuter , k. rolbiecki and r. r. de austri , arxiv:1512.06083 [ hep - ph ] ; a. alves , a. g. dias and k. sinha , arxiv:1512.06091 [ hep - ph ] ; e. megias , o. pujolas and m. quiros , arxiv:1512.06106 [ hep - ph ] ; l. m. carpenter , r. colburn and j. goodman , arxiv:1512.06107 [ hep - ph ] ; j. bernon and c. smith , arxiv:1512.06113 [ hep - ph ] ; w. chao , arxiv:1512.06297 [ hep - ph ] ; m. t. arun and p. saha , arxiv:1512.06335 [ hep - ph ] ; c. han , h. m. lee , m. park and v. sanz , arxiv:1512.06376 [ hep - ph ] ; s. chang , arxiv:1512.06426 [ hep - ph ] ; i. chakraborty and a. kundu , arxiv:1512.06508 [ hep - ph ] ; h. han , s. wang and s. zheng , arxiv:1512.06562 [ hep - ph ] ; x. f. han and l. wang , arxiv:1512.06587 [ hep - ph ] ; f. wang , l. wu , j. m. yang and m. zhang , arxiv:1512.06715 [ hep - ph ] ; j. cao , c. han , l. shang , w. su , j. m. yang and y. zhang , arxiv:1512.06728 [ hep - ph ] ; f. p. huang , c. s. li , z. l. liu and y. wang , arxiv:1512.06732 [ hep - ph ] ; j. j. heckman , arxiv:1512.06773 [ hep - ph ] ; x. j. bi , q. f. xiang , p. f. yin and z. h. yu , arxiv:1512.06787 [ hep - ph ] ; j. s. kim , k. rolbiecki and r. r. de austri , arxiv:1512.06797 [ hep - ph ] ; j. m. cline and z. liu , arxiv:1512.06827 [ hep - ph ] ; m. chala , m. duerr , f. kahlhoefer and k. schmidt - hoberg , arxiv:1512.06833 [ hep - ph ] ; s. m. boucenna , s. morisi and a. vicente , arxiv:1512.06878 [ hep - ph ] ; p. s. b. dev and d. teresi , arxiv:1512.07243 [ hep - ph ] ; j. de blas , j. santiago and r. vega - morales , arxiv:1512.07229 [ hep - ph ] ; c. w. murphy , arxiv:1512.06976 [ hep - ph ] ; u. k. dey , s. mohanty and g. tomar , arxiv:1512.07212 [ hep - ph ] ; g. m. pelaggi , a. strumia and e. vigiani , arxiv:1512.07225 [ hep - ph ] ; w. c. huang , y. l. s. tsai and t. c. yuan , arxiv:1512.07268 [ hep - ph ] ; q. h. cao , s. l. chen and p. h. gu , arxiv:1512.07541 [ hep - ph ] ; s. chakraborty , a. chakraborty and s. raychaudhuri , arxiv:1512.07527 [ hep - ph ] ; w. altmannshofer , j. galloway , s. gori , a. l. kagan , a. martin and j. zupan , arxiv:1512.07616 [ hep - ph ] ; m. cveti , j. halverson and p. langacker , arxiv:1512.07622 [ hep - ph ] ; k. das and s. k. rai , arxiv:1512.07789 [ hep - ph ] ; k. cheung , p. ko , j. s. lee , j. park and p. y. tseng , arxiv:1512.07853 [ hep - ph ] ; j. liu , x. p. wang and w. xue , arxiv:1512.07885 [ hep - ph ] ; j. zhang and s. zhou , arxiv:1512.07889 [ hep - ph ] ; g. li , y. n. mao , y. l. tang , c. zhang , y. zhou and s. h. zhu , arxiv:1512.08255 [ hep - ph ] ; m. son and a. urbano , arxiv:1512.08307 [ hep - ph ] ; h. an , c. cheung and y. zhang , arxiv:1512.08378 [ hep - ph ] ; f. wang , w. wang , l. wu , j. m. yang and m. zhang , arxiv:1512.08434 [ hep - ph ] ; q. h. cao , y. liu , k. p. xie , b. yan and d. m. zhang , arxiv:1512.08441 [ hep - ph ] ; j. gao , h. zhang and h. x. zhu , arxiv:1512.08478 [ hep - ph ] ; x. j. bi _ et al . _ , arxiv:1512.08497 [ hep - ph ] . f. goertz , j. f. kamenik , a. katz and m. nardecchia , arxiv:1512.08500 [ hep - ph ] ; p. s. b. dev , r. n. mohapatra and y. zhang , arxiv:1512.08507 [ hep - ph ] ; l. a. anchordoqui , i. antoniadis , h. goldberg , x. huang , d. lust and t. r. taylor , phys . b * 755 * , 312 ( 2016 ) doi:10.1016/j.physletb.2016.02.024 [ arxiv:1512.08502 [ hep - ph ] ] . y. l. tang and s. h. zhu , arxiv:1512.08323 [ hep - ph ] ; j. cao , f. wang and y. zhang , arxiv:1512.08392 [ hep - ph ] ; c. cai , z. h. yu and h. h. zhang , arxiv:1512.08440 [ hep - ph ] ; w. chao , arxiv:1512.08484 [ hep - ph ] ; n. bizot , s. davidson , m. frigerio and j .- l . kneur , arxiv:1512.08508 [ hep - ph ] ; l. e. ibanez and v. martin - lozano , arxiv:1512.08777 [ hep - ph ] ; y. hamada , t. noumi , s. sun and g. shiu , arxiv:1512.08984 [ hep - ph ] ; s. k. kang and j. song , arxiv:1512.08963 [ hep - ph ] ; s. kanemura , k. nishiwaki , h. okada , y. orikasa , s. c. park and r. watanabe , arxiv:1512.09048 [ hep - ph ] ; y. jiang , y. y. li and t. liu , arxiv:1512.09127 [ hep - ph ] ; k. kaneta , s. kang and h. s. lee , arxiv:1512.09129 [ hep - ph ] ; l. marzola , a. racioppi , m. raidal , f. r. urban and h. veerme , arxiv:1512.09136 [ hep - ph ] . a. dasgupta , m. mitra and d. borah , arxiv:1512.09202 [ hep - ph ] . w. chao , arxiv:1601.00633 [ hep - ph ] . w. chao , arxiv:1601.04678 [ hep - ph ] . k. ghorbani and h. ghorbani , arxiv:1601.00602 [ hep - ph ] . u. danielsson , r. enberg , g. ingelman and t. mandal , arxiv:1601.00624 [ hep - ph ] . l. j. hall , k. harigaya and y. nomura , arxiv:1512.07904 [ hep - ph ] ; k. m. patel and p. sharma , arxiv:1512.07468 [ hep - ph ] . r. ding , l. huang , t. li and b. zhu , arxiv:1512.06560 [ hep - ph ] . b. c. allanach , p. s. b. dev , s. a. renner and k. sakurai , arxiv:1512.07645 [ hep - ph ] . u. aydemir and t. mandal , scalars in @xmath370 grand unification , arxiv:1601.06761 [ hep - ph ] . m. berkooz , m. r. douglas and r. g. leigh , nucl . b * 480 * , 265 ( 1996 ) . l. e. ibanez , f. marchesano and r. rabadan , jhep * 0111 * , 002 ( 2001 ) . r. blumenhagen , b. kors , d. lust and t. ott , nucl . b * 616 * , 3 ( 2001 ) . m. cveti , g. shiu and a. m. uranga , phys . rev . lett . * 87 * , 201801 ( 2001 ) ; m. cveti , g. shiu and a. m. uranga , nucl . b * 615 * , 3 ( 2001 ) . c. m. chen , t. li and d. v. nanopoulos , nucl . b * 740 * , 79 ( 2006 ) [ hep - th/0601064 ] . c. m. chen , t. li and d. v. nanopoulos , nucl . b * 751 * , 260 ( 2006 ) [ hep - th/0604107 ] . c. m. chen , t. li , v. e. mayes and d. v. nanopoulos , phys . b * 665 * , 267 ( 2008 ) [ arxiv : hep - th/0703280 ] . c. m. chen , t. li , v. e. mayes and d. v. nanopoulos , phys . d * 77 * , 125023 ( 2008 ) [ arxiv:0711.0396 [ hep - ph ] ] . j. a. maxin , v. e. mayes and d. v. nanopoulos , phys . d * 84 * , 106009 ( 2011 ) [ arxiv:1108.0887 [ hep - ph ] ] . e. cremmer , s. ferrara , c. kounnas and d. v. nanopoulos , phys . b * 133 * , 61 ( 1983 ) . m. dine , n. seiberg , and e. witten , nucl . b * 289 * ( 1987 ) 589 ; j. attick , l. dixon , and a. sen , nucl . b * 292 * ( 1987 ) 109 ; m. dine , i. ichinose , and n. seiberg , nucl . b * 293 * ( 1987 ) 253 . v. khachatryan _ et al . _ [ cms collaboration ] , phys . b * 750 * , 494 ( 2015 ) [ arxiv:1506.02301 [ hep - ex ] ] . g. aad _ et al . _ [ atlas collaboration ] , phys . lett . * 113 * , no . 17 , 171801 ( 2014 ) [ arxiv:1407.6583 [ hep - ex ] ] .

studying the properties of resolved stellar populations through color - magnitude diagrams ( cmds ) can provide important constraints on stellar evolution and on the formation and evolution of stellar systems . for example , the apparent magnitude of the horizontal branch ( hb ) stars in a population , or the magnitude and color of the main sequence turn - off ( msto ) , constrain or allow determination of the distance to and age and metallicity of the population . the most detailed and most objectively quantifiable information can be obtained by a numerical cmd fitting analysis ( see e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) . over the last seven years , the sloan digital sky survey ( sdss , * ? * ; * ? ? ? * ) , has performed a large wide - field photometric and spectroscopic survey in the northern galactic cap . because sdss , unlike previous all - sky surveys , is a ccd - based survey , the photometric depth , accuracy , and homogeneity achieved is unprecedented over such a large area . sdss data have proven to be of great value for the study of the galactic stellar halo and nearby stellar systems with resolved populations ( e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? they have also led to the discovery of a host of previously unknown , faint local group members @xcite . the study of dwarf galaxies and the milky way halo can provide important clues to the formation and assembly of the local group of galaxies and its individual constituents , as well as test cosmological theories . in the cold dark matter ( cdm ) paradigm , an abundance of subhaloes should be present around milky way type galaxies . the observed number of satellites is , however , one or two orders of magnitudes lower than what cdm theory predicts @xcite . the recently discovered satellites partially alleviate this discrepancy , but if the cdm predictions are accurate , many more new satellite galaxies are to be discovered . another prediction of cdm theories is that galaxies are formed hierarchically , i.e. built up by accreting smaller systems ( e.g. * ? ? ? * ; * ? ? ? * ) . among the population of satellite galaxies surrounding the mw , some should have been formed within the mw halo while others should have a different origin studying the stellar populations , masses and abundance patterns of these satellites can provide important constraints for galaxy formation and cdm theory . our goals for this paper are two - fold . first , we adapt the cmd - fitting software match @xcite for use on sdss data , making use of the theoretical isochrones provided by @xcite , and describe several methods of applying this software to the data . a set of globular clusters taken from the sdss data is used to test both the algorithmic approach and the theoretical isochrones . second , we apply match to the sdss data for the new milky way satellites ( that were recently discovered using the same sdss data ) , to provide a uniform analysis of these objects and constrain their population parameters . in section [ sec : methods ] we describe the data and the different cmd - fitting methods . the tests on globular clusters are done in section [ sec : gctests ] and in section [ sec : newsats ] we analyze the properties of the new galactic satellites . we conclude and summarize our results in section [ sec : summary ] . the sdss is a large photometric and spectroscopic survey , primarily covering the north galactic cap . as of the data release 5 ( dr5 , * ? ? ? * ) , sdss photometry covers almost a quarter of the sky in five passbands ( @xmath0 , @xmath1 , @xmath2 , @xmath3 , and @xmath4 ; * ? ? ? * ; * ? ? ? * ; * ? ? ? for the work presented in this paper we use photometry in the two most sensitive bands , namely @xmath1 and @xmath2 . a fully automated data reduction pipeline produces accurate astrometric and photometric measurements for all detected objects @xcite . photometric accuracy is on the order of 2% down to @xmath522.5 and @xmath622 @xcite , and comparison with deeper surveys has shown that the data are complete to @xmath622 . we use stellar data extracted from the sdss archive with artifacts removed and we apply the extinction corrections from @xcite provided by dr5 . in order for the cmd fitting with match to work , a model describing the completeness and the photometric accuracy as a function of magnitude is needed . completeness has been checked for @xmath2-band by comparing detection of stars and galaxies by sdss with the deeper combo-17 survey . for stellar sources the completeness is 100% to @xmath2=22 and then drops to 0% at r=23.5 . since the stellar classification of combo-17 starts to become incomplete at approximately one magnitude fainter than in sdss , this completeness is not extremely accurate , but amply sufficient for our purposes . the photometric errors given by the automated reduction pipeline have been found to be very accurate @xcite . comparison of the photometric accuracy in @xmath1 and @xmath2 as a function of apparent magnitude shows similar behavior , but in @xmath1 the rise in the uncertainties occurs some 0.5 magnitudes fainter . for the @xmath1-band completeness we assume similar behavior as in @xmath2 , but also shifted half a magnitude fainter . using this information we create artificial star test files that are given as input to our cmd fitting code . from this , the code creates the photometric accuracy and completeness model with which the theoretical population models are then convolved . figure [ fig : fakedetails ] shows the completeness and average photometric errors of our artificial star files for the @xmath1- and @xmath2-band separately . even though differences in , for example , seeing cause differences in photometric accuracy and completeness between different parts of the sky , we use this same average model in the remainder of this paper . several software packages have been developed for quantitative and algorithmic cmd ( or hess diagram ) fitting ( e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? the strength of these methods lies in the fact that all photometric information from imaging in two bands can be used simultaneously in a robust and numerically well - defined way . the software package we use is match , which was written and successfully used for the cmd analysis of globular clusters and dwarf galaxies @xcite . the software works by converting an observed cmd into a hess diagram ( a two - dimensional histogram of stellar density as a function of color and magnitude ; * ? ? ? * ) and comparing it with synthetic hess diagrams of model populations . to create these synthetic hess diagrams , theoretical isochrones are convolved with a model of the photometric accuracy and completeness . the use of hess diagrams enables a pixel - by - pixel comparison of observations to models and a maximum - likelihood technique is used to find the best - fitting single model , or best linear combination of models . to account for contamination by fore- and background stars , a `` control field '' cmd needs to be provided . match will then use the control field hess diagram as an extra ` model population ' that can be scaled up or down . the resulting output will be the best - fitting linear combination of population models plus the control field . to construct star formation histories ( sfhs ) and age - metallicity relations ( amrs ) , a set of theoretical isochrones is needed that spans a sufficient range of ages , masses and metallicities . when match was written , the only published set of isochrones with full age and metallicity coverage available was provided by @xcite . since we are specifically interested in the application to sdss data , we need isochrones in the sdss photometric system . the only set available to date is one from @xcite and has the same parameter space coverage . some worries exist about the accuracy of the isochrone calibration ( e.g. * ? ? ? * ) , which was limited to three systems available in the first data release ( dr1 ) of sdss . this unavoidably adds an extra source of uncertainty to all cmd fitting results . one of the main motivations for the various tests that are described in this paper is to see whether reasonable results can be obtained using the current @xcite isochrones . we use match for slightly different purposes than originally described by @xcite ; the new `` modes '' are described briefly below . match was originally developed to solve for the sfh and amr of a stellar system for which all stars can be assumed to be at the same distance , such as globular clusters and dwarf galaxies . in this original mode of match the distance and the foreground extinction to all stars are fixed a priori for each fit . from a set of age bins and metallicity bins , the combination that best fits the observed cmd is found , resulting in an estimate of the sfh and amr . by doing the same fit with different values of the foreground extinction and distance , these two latter parameters can also be determined by comparing the goodness - of - fit of the individual fits . in principle , parameters like the steepness of the imf and the binary fraction can be constrained in the same fashion , although in most cases the data quality and the degeneracies between different parameters will prevent the determination of all of these parameters simultaneously . analyzing the parameter range of acceptable fits also gives robust error estimates for the sfh and amr ; this approach we refer to as match s `` standard mode '' . to apply cmd fitting techniques to nearby and extended stellar systems ( for example substructure within the milky way as in @xcite ) , the assumption that all stars are practically equidistant need not hold . to fit populations at differing distances simultaneously we have implemented a new way in which to run match that keeps the same number of free parameters , or parameter dimensions , to be explored . while in the standard mode the distance is fixed and age and metallicity are independent variables , here the distance becomes a free parameter , but age and metallicity are being limited via a specified age metallicity relation ( amr ) . for any given amr there is only one `` population parameter '' , e.g. age with a certain metallicity linked to each age bin . of course , different age - metallicity relations can be explored , potentially covering a wide range of possibilities . it seems sensible to choose an amr that is appropriate for the studied stellar population on astrophysical grounds , such as the expected or known metallicity evolution . for example , for the study of structures in the galactic thin disk one would consider a different amr than for the study of a metal - poor dwarf galaxy . however , within this new mode of running match the choice of amr is completely flexible and a flat amr ( no dependence of metallicity on age ) or even an amr where metallicity decreases with time is possible . when studying low - luminosity or low surface brightness contrast systems , such as some of the recently discovered dsphs , or systems at low galactic latitudes , for example substructures in the milky way , the contamination of the cmds of these systems by fore- and background stars will be very high . if the ratio of ` member ' stars to field stars becomes low , the s / n of the cmd features to be fit consequently becomes very low . in such cases , all free parameters used to constrain distance , age , and metallicity in the match mode of sections [ subsec : sfhamr ] and [ subsec : distsol ] can no longer be constrained by the data alone . to test a more restricted set of hypotheses in the case of low s / n systems we have developped a single component ( sc ) fitting mode for match . in this mode , template stellar population models are created that consist of a single component with given , fixed distance , foreground extinction , age range and metallicity range . such a sc template is fit to the data together with a control field . since the template consists of a single component , the only free parameters during the fit are the absolute scaling of ( i.e. the number of stars in ) the template and the control field . for each individual sc ( i.e. for each combination of distance , foreground extinction , age range and metallicity range ) , a value of the goodness - of - fit is obtained . comparing the goodness - of - fit values for all the sc s we can then find the best - fitting templates and thus constrain the overall properties of the system . the quality of each fit is expressed by the goodness - of - fit parameter q , determined using the maximum - likelihood technique described in @xcite . this statistic is meant to measure how likely it is that the observed distribution of stars would result from a random drawing from the fitted model . in the sc fitting scheme we know that our fits will not be perfect since an sc model will not be a perfect representation of a real stellar population , but we are primarily interested in a likelihood - ratio test for a range of sc models . the expected variance in the parameter @xmath7 for a certain cmd can be calculated from poisson statistics and also obtained from monte carlo simulations using random drawings from the models . all fits with a value of @xmath7 less than 1 @xmath8 away from that of the best - fit sc model can be considered statistically acceptable , under the assumption that the best fit is a reasonable representation of the data . taking all these fits together , error estimates on the fit parameters can be determined from the encountered spread of the parameter values . this approach of using relative values of @xmath7 should only be used in cases where the `` best fit '' is a reasonable approximation of the data . in cases where there are several very different prominent stellar populations this will not be the case and results will break down . studying different parts of the cmd separately can help to circumvent such problems , as is done in @xcite and also in section [ subsec : nsscfits ] of this paper for the dwarf galaxy leo t. the same cmd fitting techniques can also be used to detect or verify , faint stellar over - densities in large datasets like that of the sdss survey . in case of an overdensity at a specific distance , the stars will not only cluster around a certain spatial position , but also along a certain locus in the cmd . the method we have developped works by fitting the cmd of a small target patch of sky both with a control cmd obtained from a large area around the target patch and with the same control field plus a simple stellar population model . if no over - density is present , the fit with the control cmd should be good and no significant improvement in fit quality will result by including an extra stellar population model . however , if there is an over - density present and the parameters of the population model , such as distance , age and metallicity , are close enough to those of the over - density , the fit including the model will result in a significant improvement in the fit quality . by doing this for a range of models the presence of a stellar over - density can be assessed by looking at the difference in the fit quality parameter , @xmath9 , between the control cmd fit and the control cmd plus model fit . moreover , by comparing the values of @xmath9 for different models , a rough estimate of the properties of the overdensity can be obtained at the same time . an important prerequisite for this method is of course that the over - density should contribute a larger fraction of the total number of stars to the target cmd than to the control cmd , and ideally should be contained completely within the target patch . this means that over - densities will be detected with the highest s / n if their size matches the size of the target patch . cccc 1 & 1013 & @xmath101.5 & 14.5 + 2 & 1013 & @xmath101.7 & 14.5 + & 22.5 & @xmath101.3 & 14.5 + 3 & 1013 & @xmath101.5 & 20.5 + 4 & 1013 & @xmath101.7 & 20.5 + & 22.5 & @xmath101.3 & 20.5 + [ tab : fakegcprops ] before turning to real data it is useful to get a feel for the degree of precision that we can expect cmd fits to achieve . to do this , we create a few realistic artificial cmds and apply sfh fitting , sc fitting and distance fitting methods to these to try and recover their properties . using the theoretical isochrones from @xcite , a salpeter initial mass function @xcite , and our model for the photometric errors and completeness of the sdss data , we generate artificial stellar populations , to which field stars taken from the sdss data are added . in total we create four cmds , shown in figure [ fig : fakecmds ] . the number of stars in each cmd is chosen to be the same , to make for a fair comparison . cmds 1 and 3 contain a simple , single burst stellar population , located at different distances , so that either the ms and msto or the rgb and hb are prominent features . artificial cmds 2 and 4 contain two different age and metallicity populations , again with the cmds corresponding to different distances . the detailed parameters of the artificial populations are listed in table [ tab : fakegcprops ] . from the sdss archives all stellar sources in a 5@xmath115 patch of sky close to the globular cluster m13 were extracted and a small fraction used to create the field star contamination in these cmds . the remaining sources will be used as a `` control field '' population when fitting these background contaminated cmds . we recover the sfhs of the artificial cmds by running match with nine age bins and twelve metallicity bins . the age bins are defined in log @xmath12 because the isochrones are spaced more evenly in log @xmath12 than in @xmath12 , with the oldest running from 10.0 to 10.2 ( 10 gyr to 15.8 gyr ) and the youngest from 8.6 to 8.8 ( 400 myr to 630 myr ) . the metallicity range of [ fe / h]@xmath132.4 to 0.0 is covered by twelve bins of 0.2 dex width . thus , the artificial cmds are fit with 108 model cmds plus the `` control field '' stars , for a total of 109 partial cmds . the best fitting linear combination of model cmds will form the resulting sfh and metallicity evolution . to get a sense of the uncertainties on the sfh , each cmd is fit thrice , each time with a slightly different assumed distance . the distance moduli used are 14.3 , 14.4 and 14.5 mag . the resulting sfhs and metallicities are presented in figure [ fig : fakesfhamr ] as the relative star formation rate ( sfr ) and as [ fe / h ] as function of the age of the stars . for the sc fits models are created for a grid of parameters with a distance modulus range of 2.0 with 0.1 magnitude intervals centered on the the actual distance modulus ( table [ tab : fakegcprops ] ) . ages used run from 9.0 to 10.2 in log @xmath12 with bin widths of 0.1 dex and metallicities go from [ fe / h]@xmath132.1 to 0.0 in steps of 0.1 dex and bin widths of 0.2 dex . each model plus the `` control field '' is fit to an artificial cmd , after which the resulting goodness - of - fit values are compared . the resulting contour plots showing the best - fitting parameter combinations are shown in figure [ fig : fakescfit ] . using the new distance solution option of match we solve for the sfh of the artificial cmds again while fixing the metallicity to a single , uniform value . the same age bins are used as for the sfh fits described above , and a distance range of 3.0 is probed with 0.1 magnitude resolution . including the `` control field '' , 271 partial cmds are used in these fits . the fits are done at three different values of the metallicity : [ fe / h]@xmath131.6 , @xmath101.5 , and @xmath101.4 to probe the uncertainties . figure [ fig : fakesfhdist ] shows the resulting sfhs in relative sfrs and distance determinations . the sfhs in figures [ fig : fakesfhamr ] and [ fig : fakesfhdist ] show that the software succesfully recovers the single burst in cmds 1 and 2 , and the two distinct episodes of star formation in cmds 3 and 4 . some bleeding of star formation in adjacent bins will occur even when the fitted age bins exactly match the input sfh @xcite . therefore it is to be expected that this also happens here , since the duration of the input star forming episodes do not match any age bins perfectly . it should be noted that the error bars in the sfhs represent the range of sfrs encountered in each age bin within the three fits on which the sfhs are based , which means that the errors between adjacent bins are correllated . for example , in the sfh of cmd 4 in figure [ fig : fakesfhamr ] the two bins at ages of 2 to 4 gyr are not independent ; when the sfr in one of them is low , it is high in the other . that the age found for the young population in cmd 4 in figure [ fig : fakesfhdist ] is the natural result of the metallicities in the fit being lower than the actual metallicity of the younger population . in all cases , the average metallicities and distances found in figures [ fig : fakesfhamr ] and [ fig : fakesfhdist ] agree well with the input values . turning to the sc fits in figure [ fig : fakescfit ] , we see that for cmds 1 and 3 , the input properties ( indicated by stars ) are recovered very well . because there is a degeneracy between distance , age and metallicity when only the ms and msto are seen , the properties of cmd 1 are , however , constrained more poorly than for cmd 3 . since the sc fits use a single model population , the results for cmds with multiple populations , such as cmds 2 and 4 , can not be expected to properly recover the properties of all stars . rather , the population representing the majority of stars in the cmd will have the strongest weight . this is clearly visible in the sc fitting results for cmd 4 , that recover the properties of the older population , that dominates the star counts in the rgb and hb . for cmd 2 , however , the upper ms and msto of the younger population has such high s / n that this population does influence the sc fits significantly and the favored age is relatively young ( @xmath144 gyr ) . this shows that for high s / n cmds with a complex stellar population make - up , the sc fits is not well suited ; they were , however , developed for application to low s / n cmds . what distinguishes these artificial cmds from the real data , is that the artificial stellar populations are based on the exact same isochrones , imf , and photometric error and completeness model as used by match . inaccuracies in any of these will degrade the fit results one obtains for real data compared to the results for the artificial data . lccccc m 13 & 7.7 , 7.5@xmath15 & 14.38 , 14.43 & @xmath101.54 & 12@xmath15 & 0.019 + m 15 & 10.3 , 10.0@xmath16 & 15.37 , 15.0@xmath16 & @xmath102.26 & 13.2@xmath16 & 0.11 + m 92 & 8.2 & 14.6 & @xmath102.28 & 14@xmath17 & 0.025 + ngc 2419 & 84.2 , 98.6@xmath18 & 19.63 , 19.97@xmath18 & @xmath102.12 & 14@xmath18 & 0.060 + ngc 6229 & 30.4 & 17.44 & @xmath101.43 , @xmath101.1@xmath19 & @xmath1411.5@xmath20 & 0.023 + pal 14 & 73.9 & 19.34 & @xmath101.52 & @xmath1410@xmath21 & 0.035 + [ tab : gcproperties ] globular clusters are ideal test cases for the sdss implementation of match since in most of them their stars have been found to be of a single age and nearly single metallicity . here we will use six gcs present in the sdss data to test how well match , using the isochrones from @xcite , can recover their known population properties . the systems in question , , , , , , and , cover a large range in distance , background contrast ( i.e. ratio of member stars versus field stars ) and have metallicities between [ fe / h]@xmath22 and @xmath23 dex . for the following tests , all stars within a 5@xmath115 field centered on each gc were obtained from the sdss catalog archive server ( cas ) . it should be noted that in the cases of m15 , m92 and ngc 2419 these fields were not completely covered by the dr5 footprint . flags were used to exclude objects with poor photometry and all stars were individually corrected for foreground extinction using the correction factors provided by dr5 and based on the maps by @xcite . any contamination of the samples of stars obtained by misclassified galaxies or other contaminants do not pose problems for our analysis because of the use of control fields that should contain the same populations of these contaminants . `` control field '' cmds were constructed from all stars outside a radius of 0.5 from the center of each gc . `` target fields '' were constructed by selecting stars within a certain radius from the center . the radii used for m13 , m15 , m92 , ngc 2419 , ngc 6229 , and pal 14 are 0.25 , 0.2 , 0.25 , 0.15 , 0.15 , and 0.1 , respectively . these radii are similar to the tidal radii of the gcs but are not strictly related to them , since in all clusters and especially in the densest ( m13 and m15 ) the sdss photometry in the central parts is missing due to the high degree of crowding . with the radii used , a significant number of field stars is present in each target cmd , so that match has a good handle on the proper scaling of the control cmd . the sdss cmds of the gcs are shown in figure [ fig : gccmds ] and their previously measured parameters ( distance , age , [ fe / h ] ) are listed in table [ tab : gcproperties ] ; the mean reddening @xmath24 in magnitudes according to the @xcite maps in the target area for each gc is also listed in the table . the gcs in the three top panels in figure [ fig : gccmds ] have metallicities of around [ fe / h]@xmath25 dex , and the gcs in the lower panels are more metal - poor ( [ fe / h]@xmath26 ) . for the more nearby systems ( m13 , m15 , m92 and ngc 6229 ) the main cmd features that will drive the fits are the main sequence ( ms ) and main sequence turn off ( msto ) ; for the more distant systems ( ngc 2419 and pal 14 ) only the red giant branch ( rgb ) and horizontal branch ( hb ) are observed . pal 14 in particular represents the case of a very sparsely populated system with high contamination from field stars . in the following subsections , we discuss the results of applying match to these gcs in the four different modes described in section [ sec : modes ] . in fitting the cmds of these objects we will use a color range of @xmath27 , and limit the magnitude range to @xmath28 and @xmath29 ; the color cut includes the msto and the tip of the rgb , but excludes a large fraction of the numerous low - mass foreground stars . for all the fits in this section we will assume a slope for the stellar initial mass function that is close to salpeter ; the binary fraction is assumed to be 0.1 , in line with recent observational and theoretical evidence for low binary fractions in globulars ( e.g. , * ? ? ? * ; * ? ? ? to measure the sfh and amr of the clusters we run match fits with nine age bins and twelve metallicity bins . because the isochrones are spaced more evenly in log @xmath12 than in @xmath12 , the age bins are defined to have equal width in log @xmath12 , with the oldest running from 10.0 to 10.2 ( 10 gyr to 15.8 gyr ) and the youngest from 8.6 to 8.8 ( 400 myr to 630 myr ) . the metallicity bins cover the range from [ fe / h]@xmath132.4 to 0.0 in twelve bins of 0.2 dex width . together with the control cmd this yields 109 model cmds for each gc . the fitting software then finds the linear combination of these model cmds that best fits the observed cmd . each object is fit several times with slightly different fixed distances to determine the impact of distance errors on the recovered solutions . in the cases of m15 and ngc 2419 , where values from the literature for the distance modulus are quite discrepant , we run fits for distance moduli of 15.0 to 15.4 ( m15 ) and 19.6 to 20.0 ( ngc 2419 ) in 0.1 magnitude steps . for the other clusters we run three fits each , one at the literature value and at 0.1 magnitude further and nearer . figure [ fig : gcsfh ] shows the results of these fits , with the fractional sfr as a function of time in the left column , and the metallicity as a function of time in the right column . metallicities are only plotted for age bins with a significant sfr . the plotted sfr for a given age bin is the average of the values given by the fits at the different distances , with the error bars showing the complete range of sfr values found . the metallicity in a bin is the average of the fits weighted by the sfr , and the error bars correspond to the standard deviation . looking at the three intermediate metallicity gcs in the top half of figure [ fig : gcsfh ] we see that in all cases match infers a purely old ( @xmath307 gyr ) population , as expected . for m13 and ngc 6229 practically all stars are found to be older than 10 gyrs ; some bleeding into the second oldest age bin is always likely to occur , cf . section [ subsec : fakegc ] and @xcite . in the case of pal 14 the sfr is found to be constant over the two oldest age bins , but this only confirms the age estimate of @xmath1410 gyr @xcite , which implies that star formation would likely have been spread over these two bins . the intermediate metallicities of m13 , ngc 6229 and pal 14 are reproduced well and our recovered values are within 0.1 to 0.2 dex of the literature values . for ngc 6229 our fits favor the slightly lower metallicity from @xcite rather than the higher value from @xcite . turning to the bottom half of figure [ fig : gcsfh ] with the three metal - poor clusters , we see that here also the results are consistent with a purely old population , although the sfhs of m15 and ngc 2419 show a non - zero sfr in a few younger age bins . in the case of m15 this is due to the blue horizontal branch ( bhb ) stars that are not properly reproduced by the theoretical isochrones , causing the code to fit them as a younger turn - off . this demonstrates that cmd fitting results always need to be looked at in detail in order to understand their implications . the age distribution in ngc 2419 is not very well constrained because no msto is present in the cmd . the metallicities of m15 and m92 are recovered well for the old stars . while the estimated metallicity of @xmath101.9 for ngc 2419 is on the high side , for these metal - poor gcs the recovered metallicities are within 0.2 dex of the literature values . as described in section [ subsec : scfit ] above , in the single component ( sc ) fitting mode an individual model cmd is fit to the data together with the control cmd , rather than a whole set of models . the results therefore provide information on the characteristic properties of the dominant stellar population in the cmd . we sample the 3d parameter space of distance , age and metallicity by using one of nine age bins , one of 21 distance bins , and one of 23 metallicity bins . the age bins have a width of 0.1 log - years , going from 10.1 - 10.2 log - years ( 12.5 gyr to 15.8 gyr ) to 9.7 - 9.8 log - years ( 5 gyr to 6.3 gyr ) in steps of 0.05 log - years . this age range more than suffices for these gcs . we use a distance modulus range of two magnitudes centered on the literature value with a resolution of 0.1 magnitude . metallicity bins are centered at [ fe / h]@xmath132.2 to 0.0 with an inter - bin distance of 0.1 dex and a width of 0.2 dex . after matching each of the 4347 sc options to each gc and determining the expected variance in the goodness - of - fit parameter @xmath7 , we take the best fit as our measurement and compare the other fits with respect to these best values . the resulting likelihood contour plots for metallicity , distance and age measurements are shown in figure [ fig : gcscfit ] . note that in these contour plots , the third parameter is unconstrained , rather than fixed at its known value . if the third parameter would be kept fixed , the contours would shrink significantly . in general the agreement with the literature values ( indicated with asterisks in figure [ fig : gcscfit ] ) is very good . the small offset for m13 is most likely due to inaccuracies in the isochrones , which illustrates the added uncertainty in cmd fitting results that depend on a single set of theoretical isochrones . our distance modulus measurement from the sc fits for m15 lies exactly in between the two literature values , while for ngc 2419 our results clearly favor the smaller distance value from @xcite . from the two metallicity determinations in the literature for ngc 6229 , the fit results agree significantly better with the lower metallicity of [ fe / h]@xmath311.4 dex . in some cases , it looks as if the peak of the likelihood lies outside of the parameter space probed , at lower metallicities or older ages . unfortunately , these limits of the parameter space are dictated by the parameter range of the available isochrones . also shown in the right column of figure [ fig : gcscfit ] are the observed cmds of the gcs , with contours representing the model populations . in most cases we used the literature parameters for these models , except for m15 , where we chose a distance modulus of 15.2 , the mean of the two different available values . for ngc 6229 and ngc 2419 we use the literature metallicities and distance moduli that best correspond to our results . the agreement of model and data is generally acceptable , although inaccuracies , especially in the sub - giant branch , can be seen in m13 and m15 . important to note is the influence of the degeneracies between age , distance , and metallicity . for the nearby systems , m13 , m15 and m92 , where the fit is driven by the ms , there is a clear distance - metallicity covariance ; the left - most panels in figure [ fig : gcscfit ] show that the distance estimate increases with increasing metallicity . for the distant gc pal 14 , where the fit is driven by the rgb , the covariance is the other way around . another important , yet not wholly surprising , point is that higher s / n in the cmd features results in much tighter contours , illustrated by the fact that m13 , m15 and ngc 2419 give the best constrained results . ngc 2419 has the additional advantage that the well - populated hb provides a very strong constraint on the distance , making this system the best - constrained of the lot . as mentioned in section [ subsec : distsol ] , we explore fits for different distances and different ages , with the metallicity linked to the age . the result of each fit is a sfh with a ( potentially different ) distance for each age bin ; differing distances for different sub - populations are not astrophysically relevant for globular clusters , but serve as a test case . for fitting the gcs it seems reasonable to use the same metallicity in all age bins . we use the same nine age bins as in section [ subsec : gcsfh ] . in distance space we use a distance modulus range of three magnitudes centered on the known distance of each system with a resolution of 0.1 magnitudes , giving a total of 31 distance bins . for each gc we carry out three fits with different metallicities , centered on the literature values ( see table [ tab : gcproperties ] ) with offsets of 0.1 dex . for ngc 6229 we assume the lower metallicity value [ fe / h]@xmath131.43 , which is favored by our sfh and single component fitting results . in figure [ fig : gcdfit ] we present the age - distance distributions that result from these fits . compared to the fits in section [ subsec : gcsfh ] the number of free parameters per fit is the same , with now metallicity fixed instead of distance . thus , in general similar quality sfhs should be expected , although not necessarily exactly the same . while in the `` standard '' sfh fits the star formation within an age bin can be spread over several metallicity bins , it can now be spread over several distance bins . these spread will translate differently into the color - magnitude plane , and therefore can result in differences between the sfhs in figures [ fig : gcsfh ] and [ fig : gcdfit ] . as in figure [ fig : gcsfh ] all gcs are found to have a single peak in their sfr in the oldest age bin , in some cases with some bleeding into the adjacent bin . the extension of the sfh of pal 14 to more recent times has large uncertainties and is not significant . it should be taken into account that the exact age of this system is difficult to constrain with sdss data , since the msto is not present . interesting to see is that m13 and m92 show almost no bleeding into the second oldest age bin here , while they did in figure [ fig : gcsfh ] ; for m15 and ngc 6229 the exact opposite is the case . apparently , the former are fit better with some distance spread , while the latter benefit more from a metallicity spread . this does not necessarily reflect an actual distance or metallicity spread , but rather shows that inaccuracies in the isochrones are sometimes better accounted for one way and sometimes another way . the distances we find for the oldest age bin , and therefore for the majority of the stars , show very good agreement with the literature values . comparing figures [ fig : gcsfh ] and [ fig : gcdfit ] shows that the sfhs obtained when fixing either the distance or the metallicity are very similar . the main motivation for enabling this distance fitting mode was , however , not to analyze single - distance systems , but to be able to cope with cases where stars are located at different distances ( cf . canis major , * ? ? ? * ) . to demonstrate this application , we combine the cmds of m13 and ngc 6229 , resulting in a cmd containing two populations with similar metallicities and ages , but located at significantly different distances . the cmds of m13 , ngc 6229 and the combined cmd are shown in panels a ) , b ) and c ) of figure [ fig : m13ngc6229 ] . we run distance solution fits to these cmds in the same way as before , with the same age bins , but now using a larger distance modulus range , from 12.0 to 19.0 magnitudes with a resolution of 0.2 magnitudes . three fits are run , for uniform metallicities of [ fe / h]@xmath131.4 , @xmath101.5 , and @xmath101.6 . we include also a control cmd in the fit which is the control cmd of m 13 ; since m 13 and ngc 6229 are only 10 degrees apart and at similar galactic latitudes , this gives good results . the averaged best - fitting distributions of stars over the grid of age and distance bins are shown in panels d ) , e ) and f ) of figure [ fig : m13ngc6229 ] . both ages and distances of the two populations are recovered extremely well , with all stars in the oldest age bin ( @xmath3010 gyr ) , m13 stars centered at a distance modulus of 14.4 and ngc 6229 stars at a distance modulus of 17.4 . moreover , the result for the combined cmd is almost exactly the sum of the results for the individual cmds , clearly illustrating the power of match to decompose more complex cmds . the final application we describe and test here is aimed at algorithmically detecting stellar populations which are confined to a limited distance and area on the sky , i.e. that form `` overdensities '' . of course , the gcs used in this section are very obvious stellar overdensities and should be detected at very high s / n . following the approach described in section [ subsec : detect ] we define a grid of 20@xmath1120 boxes in each 5@xmath115 field that was obtained from the sdss catalog server , with the central box centered on the gc . the cmd of each box is fit with the control field cmd and with the control field plus a simple population model . for each gc we use a model with the appropriate distance , age and metallicity . figure [ fig : gcareas ] shows the difference in the goodness - of - fit parameter @xmath7 between the two fits for areas around each gc . note that for m15 and m92 , the 5@xmath115 field is not completely observed by sdss . clearly , all gcs produce high peaks in @xmath9 . since @xmath9 is roughly proportional to the number of stars attributed to the population model in the model fit , the objects with the largest numbers of stars produce the strongest detections . only in the plot for pal 14 does the vertical scale allow the random variations in @xmath9 to be visible in the area around the object . the s / n of these detections should be measured with respect to this random noise . using the 216 boxes around the central 9 boxes we obtain an estimate of the rms in @xmath9 from random variations in the cmds . negative values of @xmath9 are suppressed because a negative number of stars in a population is unphysical , which has to be taken into account when calculating the rms . even pal 14 , a rather sparsely populated and distant cluster , is detected with high s / n , as @xmath32 . in summary , section [ sec : gctests ] has illustrated the robustness and precision with which match can interpret high - quality cmds in a variety of circumstances . over the past few years the importance of deep wide - field surveys for the study of the stellar components of the milky way galaxy has been demonstrated by 2mass ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) and the sdss . specifically , the sdss has enabled a detailed picture of the galactic stellar halo ( e.g. , * ? ? ? * ; * ? ? ? * ) and led to the discovery of at least fourteen previously unknown dwarf galaxies and globulars : ( umai , * ? ? ? * ) , ( wil1 , * ? ? ? * ) , ( cvni , * ? ? ? * ) , ( booi , * ? ? ? * ) , ( umaii , * ? ? ? * ) , , , , , ( com , cvnii , her , leoiv , seg1 , * ? ? ? * ) , @xcite , ( booii , * ? ? ? * ) , and two very low - luminosity globulars and @xcite . while these discoveries are important for our understanding of the so - called `` missing satellite problem '' ( e.g. , * * ; * ? ? ? * ) , these new systems might also shed new light on the mechanisms of star formation in low - mass halos with small baryon content . why these systems were only discovered with sdss data can be explained by their low surface brightness , as shown by @xcite . apart from the promise that more systems of even lower surface brightness might be waiting to be discovered by future generations of deep imaging surveys , this fact raises intriguing questions . do these systems show that star formation can occur under conditions generally considered to be incapable of supporting such activity , or have these systems evolved to their current state through , for example , tidal interactions with the galaxy ? a thorough and systematic study of the properties of these objects is needed to understand their formation and evolution and the possible cosmological implications . several groups have been working on photometric and spectroscopic follow - up of the new mw companions in order to constrain their properties . a compilation of these results is given in table [ tab : newprops ] . although individual measurements of properties might be less precise in some cases , the advantage of studying these objects using the sdss photometry is that this highly uniform dataset allows a systematic approach to the complete set of objects and fair comparisons between them . in the following , we apply our single component ( sc ) fitting techniques to constrain the overall distances , metallicities and ages of the objects listed in table [ tab : newprops ] . we also constrain the sfh of a subset of objects , since the majority contain too few stars to make this a worthwhile exercise . the objects we use are boo , cvn i , and uma i because of their relatively well - populated cmds , and leo t because of its clearly complex star formation history . for all the fits in this section we assume a salpeter - like initial mass function and a binary fraction of 0.5 , similar to the value for disk stars @xcite . lccccc bootes i & 60@xmath336 & 18.9@xmath330.2 , 19.1@xmath330.1@xmath15 , 19.0@xmath330.1@xmath16 & @xmath14 - 2 , @xmath14 - 2.5@xmath34 , @xmath14 - 2.1@xmath35 & 12.8@xmath330.7 & -5.8@xmath330.5 + bootes ii & 60@xmath3310 & 18.9@xmath330.5 & @xmath14 - 2 & 4@xmath332 & -3@xmath331 + canes venatici i & 220@xmath3320 & 21.75@xmath330.20 & @xmath14 - 2 , -2.5 -1.5@xmath18 , -2.0@xmath35 , -2.09@xmath19 & 8.5@xmath330.5 & -7.9@xmath330.5 + canes venatici ii & 150@xmath3315 & 20.9@xmath330.2 & -2.3@xmath19 & @xmath143.2 & -4.8@xmath330.6 + coma berenices & 44@xmath334 & 18.2@xmath330.2 & @xmath14 - 2 , -2.00@xmath19 & @xmath145.5 & -3.7@xmath330.6 + hercules & 140@xmath3313 & 20.7@xmath330.2 & -2.27@xmath19 & @xmath148.2 & -6.0@xmath330.6 + leo iv & 160@xmath3315 & 21.0@xmath330.2 & -2.3@xmath19 & @xmath143.4 & -5.1@xmath330.6 + leo t & @xmath14420 & @xmath1423.1 & @xmath14 - 1.6 , -2.3@xmath19 & @xmath141.4 & @xmath14 - 7.1 + segue 1 & 23@xmath332 & 16.8@xmath330.2 & - & @xmath144.5 & -3.0@xmath330.6 + ursa major i & @xmath14100 & @xmath1420 & @xmath14 - 2 , -2.2@xmath35 , -2.1@xmath19 & @xmath147.8 & @xmath14 - 6.5 + ursa major ii & 32@xmath335 & 17.5@xmath330.3 & @xmath14 - 1.3 , -1.6@xmath35 , -2.0@xmath19 & @xmath1412 & -3.8@xmath330.6 + willman 1 & 38@xmath337 & 17.9@xmath330.4 & @xmath14 - 1.3 , -1.6@xmath35 & @xmath141.9 & @xmath14 - 2.5 + [ tab : newprops ] as for the gcs we obtained data in 5@xmath115 fields around each object from the sdss cas as outlined in section [ sec : gctests ] . for the target cmds we used the stars within one half - light radius ( @xmath36 , table [ tab : newprops ] ) from the center ( with a minimum of 3 ) , with all stars outside a radius of 60 used for the control cmds . figure [ fig : nscmds ] shows the target cmds for the twelve objects . in all cases except leo t a color range of @xmath27 is used for the cmd fits . the population parameter space we sample consists of 12 age bins , 24 metallicity bins and 31 distance bins . the age bins have a 0.1 log - year width and separations of 0.1 log - year , with the oldest running from 10.110.2 log - years ( 12.5 gyr to 15.8 gyr ) and the youngest from 9.09.1 ( 1 gyr to 1.26 gyr ) . metallicities run from [ fe / h]=@xmath102.3 to 0.0 in steps of 0.1 dex , and each metallicity bin has a width of 0.2 dex . the range in distance modulus probed is 3 magnitudes wide , with a resolution of 0.1 magnitudes , and is centered on the literature value for each object . for leo t we use a slightly different approach . the group of blue stars in the cmd is too bright and does not have the right morphology to be interpreted as hb stars belonging to the same population as the rgb stars . this means there are two populations present in two different parts of the cmd , with the rgb corresponding to the old population and the stars at @xmath370.3 to a much younger population . as demonstrated in section [ subsec : fakegc ] , in such a case the sc fits will give very poor results . however , since the two populations in leo t are only present on one side of the @xmath38 line , they can be easily separated . thus , to determine the properties of the two populations separately we divide the leo t cmd into a red ( -1.0@xmath390.3 ) and a blue ( 0.3@xmath391.5 ) part and fit these separately . the red part of the cmd is fit with the same parameter space as the other objects . for the blue part the same metallicities and distance moduli are probed , but a younger set of age bins is used . here we use nine bins , the four oldest of which have a width of 0.1 log - years centered on 9.05 , 9.15 , 9.25 and 9.35 ; the five youngest have a width of 0.2 log - years centered on 8.1 , 8.3 , 8.5 , 8.7 and 8.9 log - years . in figures [ fig : nsscfit1 ] and [ fig : nsscfit2 ] the sc fitting results for the new mw satellites , except leo t , are presented . the right - most panels show the cmd of each object again , with the best - fit sc model overplotted as contours . monte carlo runs were used to determine the values of @xmath8 . except for boo i and cvn i , solid contours are shown for the case of a fixed metallicity or age , where such a measurement is available in the literature . figure [ fig : leotscfit ] shows the same results for the red and blue part of the leo t cmd . the fact that low s / n hinders a robust determination of the population properties provides a serious problem for the majority of the targets . although for most objects a finite region of parameter space lies within 1 @xmath8 from the best fit ( the white area in the contour plots ) , reasonable measurements of any individual parameter can only be obtained for boo i , cvn i , cvn ii , com and her . especially for boo ii , leo iv and leo t , parameters are very poorly constrained , and @xmath7 space shows a complicated morphology with multiple minima . to constrain the sfhs of boo i , cvn i , leo t and uma ii we follow the same approach as before for the gcs . a grid composed of nine age bins and twelve metallicity bins is used , the age bins going from 10.2 to 8.4 log - years with an 0.2 log - year width , and the metallicities from [ fe / h]=-2.4 to 0.0 with a 0.2 dex width . a fit is done at three different distance moduli , 0.1 magnitude apart , with the middle value equal to the literature value . the resulting sfhs and amrs are plotted in figure [ fig : nssfh ] . as with the gcs , we also test our overdensity detection scheme , described in section [ subsec : detect ] , on these new milky way satellites . the 5@xmath115 field centered on each object is divided in a square grid of 225 20@xmath1120 boxes , the cmds of which are fit both with a control field and with the control field plus a single component population model . as the control field the whole 5@xmath115 field is used . for the sc model we consider two cases . first we use a typical old population model , with age range 10.0 to 10.1 log - year and [ fe / h]@xmath40 at the best - fit distance obtained with the sc fits from section [ subsec : nsscfits ] . second we use a model corresponding to the best - fit properties obtained for each object . the former case serves to simulate a realistic search of all sdss data for unknown stellar overdensities . when performing such a search , an old , metal - poor model would be a likely choice since most mw satellites are dominated by such populations . another factor that would influence the results of a search for unknown stellar overdensities is the size of the box used . as a test , we also did the same exercise using different box sizes for boo i. in addition to the 20@xmath1120 boxes , we also used 40@xmath1140 boxes and 1@xmath111boxes . to enable the use of these coarser grids , we retrieved a larger , 10@xmath1110 field from the sdss dr5 archive centered on boo i. in figure [ fig : nsarea ] we present the results of the fits with the fine - tuned stellar population models to the twelve objects . each panel shows the improvement in the goodness - of - fit , @xmath41 , as function of the position in the field centered on a satellite . figure [ fig : booiareas ] shows the resulting @xmath41 for the three different box sizes in a 10@xmath1110 field centered on boo i. the significances of all detections , expressed in @xmath42 , are listed in table [ tab : detsig ] . lcc boo i@xmath15 & 106.9 & 121.6 , 58.2 , 27.8 + boo ii & 0.86 & 0.93 + cvn i & 447.7 & 589.4 + cvn ii & 2.61 & 13.3 + com & 57.5 & 53.9 + her & 15.2 & 28.2 + leo iv & 13.8 & 38.2 + leo t & 5.8 & 11.3 + seg 1 & 9.9 & 7.8 + uma i & 12.9 & 11.2 + uma ii & 1156.2 & 865.3 + wil 1 & 48.8 & 47.8 + [ tab : detsig ] we have performed a systematic and uniform numerical cmd analysis of the recently discovered mw satellites . how well the properties of an individual object can be determined depends on the number of member stars present in the cmd and the observed cmd features . for some objects , this type of analysis is hard - pressed to provide strong statistical constraints due to the sparseness of their cmds . in table [ tab : newresults ] we summarize the results of our single component and star formation history fits . the distance moduli , metallicities and ages listed are the unweighted averages of all sc fits that lie within 1 @xmath8 of the best sc fit and are therefore considered to be acceptable fits . the errors are the standard deviations of the parameter values of these fits , combined with an added systematic uncertainty of 0.1 dex in [ fe / h ] and 0.1 mag in distance modulus to account for isochrone uncertainties . in cases where these properties can not be sufficiently constrained we list the metallicities and ages we get when assuming the distance modulus from the literature , as this is usually the best - determined property . the star formation history results , where available , are summarized in table [ tab : newresults ] by the main star formation episodes . below follows an object - by - object discussion of these and the star formation history results . _ boo i._ this dwarf spheroidal is relatively nearby at @xmath1460 kpc and an rgb , hb and msto are all present in the sdss data , albeit sparsely sampled . the overall properties can therefore be determined with relatively good accuracy , and the star formation history is the best - constrained of all . especially the distance of boo i is very well constrained thanks to the hb which is detected at high s / n due to the absence of field stars at @xmath430 . the msto and slope of the rgb provide constraints on the age and metallicity , respectively . our distance , metallicity and age measurements agree very well with the findings of other authors based on spectroscopy and deeper imaging data . the star formation history shows one significant epoch of star formation in the oldest ( @xmath44 gyr ) bin , with no other bins containing a significant sfr . considering the ability of these fits to distinguish between a single and a multiple component system , demonstrated by the fits to the artificial cmds in section [ subsec : fakegc ] , we conclude , despite the sparseness of the cmd , that boo i very probably only contains old stars . thus , our results are consistent with the picture of a purely old , metal - poor stellar population . _ boo ii._ the number of boo ii member stars in the sdss data is very small and traces only the rgb . unfortunately , the resulting very low s / n combined with the degeneracies between the different parameters make it impossible for match to constrain the stellar population parameters in a statistically significant way . although @xcite show that the sdss cmd of boo ii is _ consistent _ with boo ii having similar metallicity , age , and distance as boo i , the sdss data alone do not support any significant constraints on any of these properties . the only way to get such constraints is by obtaining deeper follow - up data . _ cvn i._ being the brightest of the twelve objects analyzed here , cvn i has a well - populated rgb and hb . this helps to increase the robustness of the constraints we obtain from the sc fits . although at the edge of the sensitivity limits of sdss , the hb enables the determination of the distance . moreover , the extent of the hb to very blue colors requires the stars in cvn i to be both old and metal - poor . this constraint on the metallicity is further enhanced by the slope of the rgb . together , the extended hb and rgb thus can provide robust constraints on the properties of the dominant stellar population . the distance modulus we find is slightly smaller than that found in the literature , but not significantly so . the metallicity of [ fe / h]=-1.8@xmath330.2 falls nicely within the range of metallicities observed by @xcite . although an age of 14@xmath332 gyr fits the cmd best , the star formation history shows two peaks . the oldest star forming episode in the 10 - 16 gyr bin is dominant in terms of the number of stars and drives the sc fits , but the peak in the 2.5 - 4 gyr bin is equally significant . for the old stars the metallicity is found to be close to [ fe / h]=-2 , while the younger stars have metallicities around [ fe / h]=-1.5 . it should be noted that the young population found here is not caused by the algorithm attempting to fit bhb stars , as is the case for m15 in figure [ fig : gcsfh ] . stars belonging to this young population only appear in the regions of the cmd where the rgb and the red part of the hb are located . the presence of two populations is therefore purely based on the rgb morphology and the density of stars along the hb . this detection of an old and intermediate age population with these metallicities confirms the findings of @xcite and @xcite . furthermore , @xcite note the presence of a carbon star very close to the center of cvn i and its heliocentric radial velocity of 36@xmath3320 km / s is consistent with the radial velocities that @xcite and @xcite measure for cvn i , making it is likely that this star belongs to the cvn i dwarf . since carbon stars are often of intermediate age , this is another indication of the presence of an intermediate age population in this system . _ cvn ii._ although very faint and distant , our distance measurement for cvn ii is very accurate and in very good agreement with the value from @xcite due to the presence of hb stars . from the extension of the hb to very blue colors ( @xmath45 ) we can infer that the stellar population is both old and metal - poor . despite the sparseness of the cmd we thus get relatively good constraints on these parameters and we find a metallicity of [ fe / h]@xmath14 - 2 and a very high age . cvn ii thus appears to be a very old and metal - poor object . _ com._ an absence of a clear hb makes constraining the distance of com difficult and leads to a strong degeneracy between age , distance and metallicity . similar degeneracies occur for artificial cmd 1 in section [ subsec : fakegc ] and the gc ngc 6229 ( see figures [ fig : fakescfit ] and [ fig : gcsfh ] ) , even though their cmds are much less sparsely populated . the rgb and msto alone do not allow a robust determination of the individual population parameters . as a consequence , the distance modulus we recover has rather large error bars but is consistent with the findings of @xcite . the sc fits favor an ancient population ( @xmath44 gyr ) and a metallicity of [ fe / h]@xmath312 , but also here the uncertainties are large . _ her._ in the case of her the contamination of the cmd with fore- and background stars is severe , which is due to its location at lower galactic latitudes compared to the other objects . this high contamination means a low s / n of the cmd features of her , apart from the hb , which is in the relatively clean part of the cmd at @xmath46 . our distance modulus of 20.4@xmath330.2 , which largely is determined by the hb , is somewhat lower than estimated by @xcite , but is consistent within the error bars . this object too is consistent with an ancient and metal - poor ( [ fe / h]@xmath14 - 2 ) stellar population . even though the region of parameter space that lies within 1 @xmath8 of the best fit values is well constrained , the complete parameter space lies within 2 @xmath8 , which is due to the low s / n of the cmd . these results based only on sdss data are also consistent with the results obtained by applying the same methods to much deeper @xmath47 photometry from the large binocular telescope @xcite . in fact , the uncertainties on the values of distance , metallicity and age from the deeper data are very similar to the 1@xmath8 values we obtain here . lcccr bootes i ( booi ) & 18.8@xmath330.2 & -2.2@xmath330.2 & 14@xmath332 & @xmath1410 - 16 gyr + bootes ii ( booii ) & - & - & - & - + canes venatici i ( cvni ) & 21.5@xmath330.2 & -1.8@xmath330.2 & 14@xmath332 & @xmath142 - 4 gyr , @xmath1410 - 16 gyr + canes venatici ii ( cvnii ) & 20.8@xmath330.2 & -2.1@xmath330.3 & 14@xmath332 & - + coma berenices ( com ) & 18.4@xmath330.4 & -1.9@xmath48 & 11@xmath49 & - + hercules ( her ) & 20.4@xmath330.2 & -2.1@xmath330.2 & 14@xmath333 & - + leo iv ( leoiv ) & - & _ -2.1@xmath330.3 _ & _ 14@xmath332 _ & - + leo t ( leot ) & 23.1@xmath330.6 & -1.6@xmath330.6 & - & @xmath50 gyr , @xmath146 - 10 gyr + segue 1 ( seg1 ) & - & _ -1.6@xmath330.5 _ & _ 11@xmath334 _ & - + ursa major i ( umai ) & - & _ -1.8@xmath330.4 _ & _ 10@xmath335 _ & - + ursa major ii ( umaii ) & - & _ -1.5@xmath330.5 _ & _ 11@xmath334 _ & @xmath142 - 6 gyr , @xmath1410 - 16 gyr + willman 1 ( wil1 ) & 17.9@xmath330.4 & -1.6@xmath330.5 & 10@xmath335 & - + [ tab : newresults ] _ leo iv._ the cmd of leo iv is so sparse that a statistically significant determination of the population properties is not possible based on the sc fits alone . the best fit is given by a population of @xmath1414 gyr old stars with [ fe / h]=@xmath102.2 at a distance modulus of 20.8 , but a large number of models gives a goodness - of - fit that is only marginally worse . assuming a distance modulus of 21.0 @xcite , gives the solid contours in the age - metallicity panel for leo iv in figure [ fig : nsscfit2 ] . using this prior , the sc fits allow the age and metallicity to be constrained to 14@xmath332 gyr and [ fe / h]@xmath132.1 @xmath330.3 , respectively . _ leo t._ this object is very interesting and differs from the others analyzed here , in that it is the least luminous system known which shows signs of on - going star formation and has hi gas associated with it @xcite . because of the presence of two clearly very different populations the sc fitting method is not very well suited . even when fitting the two populations separately , by dividing the cmd in two distinct parts as we have done , the results are poor , due to the small numbers of stars in each part . taking the results of the two populations together we find a distance modulus of 23.1@xmath330.6 magnitudes . both the old and young populations are best fit with low metallicities , [ fe / h]@xmath51 - 1.5 . the star formation history we find ( figure [ fig : nssfh ] ) clearly shows two peaks , confirming the presence of two distinct populations . the age of the old population , not well - constrained with the sc fits , is put in the 6.5 - 10 gyr age bin . comparing this with the sfhs of the gcs in figure [ fig : gcsfh ] and the other sfhs in figure [ fig : nssfh ] , where in all cases the peak of star formation is in the oldest age bin , this implies that the old population in leo t is on average younger than the old population in these other objects . the young stars are clearly found in the three youngest age bins , where they are fit as young ( @xmath50 gyr ) blue loop stars . @xcite interpreted the nature of these stars similarly . _ seg1._ in seg 1 only a ms and msto are populated in the very sparse cmd . as with the artificial cmd 1 in section [ subsec : fakegc ] and the gcs m13 , m15 and m92 , where the sc fits also solely relie on the ms and msto , strong degeneracies between age , distance and metallicity are expected . indeed , the contour plots for seg 1 in figure [ fig : nsscfit2 ] show a similar morphology , but due to the sparseness of the cmd with even poorer constraints . this makes any determination of the population properties of seg 1 , based solely on the sc fits , impossible . even when using the distance modulus from @xcite as a prior , the age - metallicity degeneracy prevents strong constraints on these parameters , resulting in very large error bars on the values for seg 1 listed in table [ tab : newresults ] . _ uma i._ this object also has a very sparsely populated cmd . an rgb is readily seen , with some probable hb stars that , however , are too few to produce strong statistical constraints on the distance . because of this and the usual degeneracies , the significance of the sc fit results is very low , causing all parameters to be poorly constrained . assuming a distance modulus of 20 @xcite , our results are consistent with an old ( @xmath1410 gyr ) and metal - poor population , but with large error bars . _ uma ii._ although the msto of uma ii is well - populated , the scarcity of stars in other parts of the cmd provides problems for this object . there are some rc stars , but the s / n of this cmd feature is not high enough to make for a reliable distance determination . added to that there are strong degeneracies between the different solutions ; similar degeneracies are seen in section [ sec : gctests ] for the gc ngc 6229 where also the msto drives the sc fit results . our results are , however , consistent with previous findings . with a prior on the distance modulus , @xmath5217.5 , the fits agree with the conclusion of @xcite that uma ii is not very metal - poor , but seems to have a more intermediate metallicity of [ fe / h]@xmath531.5 , although the error bars are large ( see table [ tab : newresults ] ) . the spectroscopic metallicity measurement from @xcite is consistent with our value , but the one from @xcite is more metal - poor , but still consistent within the errors . while the sc fits imply that the majority of stars is likely old , @xmath1411 gyr , the sfh ( figure [ fig : nssfh ] ) hints at more complexity . according to the sfh results , there is a population of very old ( @xmath3010 gyr ) and metal - poor ( [ fe / h]@xmath54 - 2 ) , as well as a younger ( @xmath145 gyr ) , close to solar metallicity population . since the sc fits measure the average properties , this is consistent with the intermediate metallicity we find there . @xcite report that the upper main sequence of uma ii in their deeper subaru data is wider than expected just from observational errors , one explanation for which could be the presence of a mix of populations . also in our fit , the double peak results from the width of the msto , which is less well - fit by a single - age population . the exact ages and metallicities of the populations should be studied with deeper and spectroscopic data , since the sdss cmd does not support such detailed conclusions . however , taking our sfh fit and these other results together it does seem likely that uma ii , like cvn i , is a system with a complex star formation history , which might also be ( part of ) the explanation for the discrepancy between the two available spectroscopic metallicity measurements . _ wil1._ for this system only the msto is observed , which is also sparsely populated . this leads to a strongly degenerate solution space , similar to the cases of uma ii and seg 1 . the distance can be determined slightly better here , but the range of distance moduli for which fits within 1@xmath8 from the best fit can be obtained is still very large : 17.2@xmath5518.5 . using the distance modulus from @xcite ( m - m=17.9 ) as a prior , the fit results imply an old population ( @xmath1410 gyr ) with an intermediate to low metallicity ( [ fe / h]@xmath14 - 1.6 ) , but also here the error bars on these numbers are large . figure [ fig : nssummary ] summarizes our results from the match fits to the cmds for the sample of new milky way satellites . in the left panel we plot our metallicity estimates for all objects versus their absolute magnitudes , together with the spectroscopic metallicities from @xcite and @xcite . especially for the fainter objects our metallicity estimates are very poorly constrained ( figures [ fig : nsscfit1 ] and [ fig : nsscfit2 ] ) and because of the available isochrones our metallicity range is restricted to [ fe / h]@xmath132.4 at the metal - poor end . in light of this , our photometric results agree remarkably well with the spectroscopic values . still , care has to be taken when drawing conclusions from these metallicity estimates . we can say at best that we find no indication for a corellation between metallicity and the luminosity of the objects , which is observed for brighter dwarf spheroidals ( e.g. , * ? ? ? * ; * ? ? ? the mean age of the stars versus absolute magnitude is shown in the middle panel of figure [ fig : nssummary ] . for objects in which the sfh fits showed evidence for two populations , both populations are plotted separately . the right panel shows the inferred ages of the satellites versus distance modulus , to probe possible environmental influences . looking at the middle panel of figure [ fig : nssummary ] there is no clear evidence for a luminosity dependence of the ability of a dwarf galaxy to form stars recently . although the two most luminous objects in this sample , leo t and cvn i , both have young or intermediate age populations , the much fainter uma ii does as well . while the environment might play a role , the right panel of figure [ fig : nssummary ] shows no obvious relation between age or the presence of multiple star formation epochs with distance from the sun ( since these objects are at high galactic latitude and at large distances , the heliocentric and galactocentric distances are not significantly different ) . this is contrary to the trend suggested by @xcite for the brighter local group dwarf spheroidals , in which systems with a larger fraction of young or intermediate age stars are systematically more distant . since most of these recently detected objects are intrinsically faint and/or distant , they are of course much harder to detect by any overdensity detection scheme than the gcs . the improvement in the goodness - of - fit , @xmath41 , between a fit with a control field only and a control field plus sc model is roughly proportional to the number of stars in the cmd that can be fit by the sc model . this means that the objects with sparsely populated cmds are the most difficult to detect . comparing the overdensity detection results in figure [ fig : nsarea ] with the cmds of the objects ( figure [ fig : nscmds ] ) confirms this , as boo , cvn i and uma ii give by far the highest @xmath41 . similarly , the rms noise in @xmath41 increases with increasing numbers of field stars in the cmds , and at the low s / n end the noise level starts to play an important role . for example , comparing the panels for boo ii and cvn ii in figure [ fig : nsarea ] , which have the same @xmath41 scale , reveals very different noise levels . this can be traced back to the positions of these objects on the sky , with bootes ii lying behind the sagittarius stream and cvn ii in a much ` cleaner ' region of the halo . the sagittarius stream also produces a high noise level around segue 1 . while most objects produce clear peaks in @xmath41 , the detectability of several is rather low . for example , cvn ii , leo t , segue 1 and uma i have a @xmath41 only a factor of a few higher than the most prominent noise peaks ; boo ii is not clearly detected at all . when inspecting the values in table [ tab : detsig ] , it turns out that the use of a fine - tuned population model as compared to the generic old , metal - poor model has a varied effect . in some cases the detection significance is improved , in some cases the change has little effect and in a few cases the detectability is actually decreased . in these last cases , even though the peak @xmath41 is the same to or higher for the fine - tuned model , the change in the rms causes the detection to become less significant . this indicates that , while sampling a range of population models would be preferable when doing an overdensity search , a relatively small grid of models will not severely compromise the outcome . to test the significance of the @xmath42 measure , we tested the overdensity detection scheme on a presumably ` empty ' patch of sdss sky with 170@xmath56ra@xmath56200and 40@xmath56dec@xmath5665 . this patch of sky was divided into 20@xmath1120 boxes with some overlap , giving a total of 7088 boxes , equalling an effective search area of almost 800 square degrees . a control field ( the cmd from the surounding 5@xmath115area ) was fit to the cmd of each box , as well as this control field plus a typical old stellar population model with an age bin of 10.0 to 10.1 log - year and [ fe / h]=-2 . this was done seven times , with the stellar population model shifted to different distance moduli : 17.0 , 18.0 , 19.0 , 20.0 , 21.0 , 22.0 and 23.0 magnitudes . thus , for each box seven values of @xmath42 are obtained , where the rms is obtained from the boxes in the surounding 5@xmath115 . the methodology closely resembles the approach used for the mw satellites . figure [ fig : pksig ] shows the histogram of all @xmath42 values for all boxes ( solid line ) , as well as for two individual distance moduli , 18.0 ( dotted ) and 22.0 ( dashed ) mag . all three histograms show the same distribution , implying that the noise behaves similarly for all distance moduli . the tail of the distribution extends to @xmath57 . an inspection of the locations with values higher than 7 reveals no obvious overdensities , implying that this tail is indeed caused by noise . in order to obtain a clean sample of stellar overdensity detections , a @xmath42 threshold of @xmath149 seems appropriate . this means that when using a sufficiently sampled grid of population models and distances , all new mw satellites would be detected , with the exception of boo ii ( cf . table [ tab : detsig ] ) . the results of the fits to boo i with different box sizes , listed in table [ tab : detsig ] and shown in figure [ fig : booiareas ] , illustrate the effect of the spatial resolution of the cmd grid . here , the detectability is also strongly affected by the rms of the background noise . since the @xmath41 due to random variations in the field population increases with increasing numbers of field stars , the rms is higher for larger box sizes , favoring the smallest boxes used here . the fact that the @xmath58 decreases roughly by a factor of 2 as we move to larger box sizes is caused by the number of background stars increasing by a factor 4 , while the number of boo i stars in the central box does not change . it is also interesting to note that here the noise is dominated by the sagittarius stream , which runs accross the lower left corner of the panels in figure [ fig : booiareas ] . the significance of the detections in table [ tab : detsig ] can be boosted by decreasing the box sizes , especially for the objects with the smallest ( projected ) size . we have shown that application of match @xcite to hess diagrams or cmds of sdss data successfully recovers the known stellar population properties of globular clusters ( section [ sec : gctests ] ) , within the limitations of the data . this gives confidence that despite inaccuracies in the theoretical isochrones in isolated parts of the cmd , the overall cmd - fitting technique still provides good and trustworthy results . apart from the standard star formation history fitting method , we have also demonstrated three other ways of applying match to sdss data : + 1 ) identifying the best fitting ` single component ' stellar population , a mode useful for cmds of very faint or distant objects with limited s / n in their cmd ; + 2 ) fitting the line - of - sight distribution of stars to keep the fit s degrees of freedom constant , we fix the metallicity at each age bin ; and + 3 ) algorithmic detection of objects , i.e. of a localized stellar population overdensity in a particular ( unknown ) direction and at a particular ( unknown ) distance . this is done by comparing the difference in the goodness - of - fit between a cmd - fit to a control field and a cmd - fit to a control field plus model population . these tests show that the application of cmd - fitting techniques to wide - field survey data is a very promising endeavor in the light of sdss and other surveys that will start yielding data over the next few years . further improvements in software and theoretical isochrones can only increase the use and accuracy of such work . on the basis of these tests and the application to known globular clusters , we have applied our single component fitting technique to twelve recently discovered milky way satellites . the sparseness of several of the cmds of these objects inhibits a robust determination of the population parameters and the best fit solutions are often not unique . in order to obtain good constraints at least some cmd features such as the msto , hb and rgb should be observed with sufficiently high s / n . for the least luminous objects this is inherently problematic because of their lack of ( evolved ) stars . for only four of the satellites , the cmds had sufficiently high s / n to warrant a star formation history fit . the results of the population fits to the new milky way satellites , summarized in table [ tab : newresults ] , are consistent with the individual discovery and follow - up studies . when looking at the sample as a whole , we find that the majority of these objects are likely dominated by old ( @xmath59 10 gyr ) and metal - poor ( [ fe / h]@xmath14 - 2 ) stellar populations . for boo , cvn ii , com , and her the single component fits converge to this picture , and for boo the full fit of a star formation history also confirms this . seg 1 , uma i , uma ii , and wil 1 are not well constrained by the sc fits , but using the distances from the literature as priors , they are also consistent with similar ages and metallicities , although slightly higher metallicities ( [ fe / h]@xmath311.6 ) can not be excluded . the single component fits show cvn i to be old as well , but not extremely metal - poor . however , the sfh reveals the presence of both an old , metal - poor , as well as a younger ( @xmath513 gyr ) and more metal - rich population . this is consistent with the findings of @xcite , who detect two kinematically distinct populations in cvn i which differ in age and metallicity , and the detection of a young component by @xcite in deeper photometry . the sfh of uma ii also shows a peak at an age of @xmath145 gyr , which supports suggestion of multiple populations in this object by @xcite , who find the msto width of uma ii to be too large to be explained by photometric errors alone . finally , in leo t the presence of two distinct populations is readily seen in the cmd and reproduced in the star formation history . the rgb seen in this system belongs most likely to an intermediate age ( @xmath145 - 10 gyr ) population with a metallicity of [ fe / h]@xmath14 - 1.5 to -2 , while the blue stars in the cmd are most likely very young ( @xmath561 gyr ) blue loop stars with similar metallicities . the objects studied here are all intrinsically faint and sparse , and in some cases the population parameters can be constrained only poorly . however , their star formation histories do not appear uniform , nor do all of them appear to be exclusively old and metal poor . we thank matthew coleman for helpful discussions and the anonymous referee for his useful comments that helped to clarify this paper . j.t.a.d.j . acknowledges support from dfg priority program 1177 . d.b.z . was supported by a pparc - funded rolling grant position . thanks the deutsche forschungsgemeinschaft for their support through the emmy noether program . v.b . was supported by a pparc fellowship . funding for the sdss and sdss - ii has been provided by the alfred p. sloan foundation , the participating institutions , the national science foundation , the u.s . department of energy , the national aeronautics and space administration , the japanese monbukagakusho , the max planck society , and the higher education funding council for england . the sdss web site is http://www.sdss.org/. the sdss is managed by the astrophysical research consortium for the participating institutions . the participating institutions are the american museum of natural history , astrophysical institute potsdam , university of basel , university of cambridge , case western reserve university , university of chicago , drexel university , fermilab , the institute for advanced study , the japan participation group , johns hopkins university , the joint institute for nuclear astrophysics , the kavli institute for particle astrophysics and cosmology , the korean scientist group , the chinese academy of sciences ( lamost ) , los alamos national laboratory , the max - planck - institute for astronomy ( mpia ) , the max - planck - institute for astrophysics ( mpa ) , new mexico state university , ohio state university , university of pittsburgh , university of portsmouth , princeton university , the united states naval observatory , and the university of washington . aparicio , a. , gallart , c. & bertelli , g. 1997 , , 114 , 680 , j. , catelan , m. , ferraro , f. r. , spassova , n. , buonanno , r. , iannicola , g. , richtler , t. , & sweigart , a. v. 1999 , , 343 , 813 clem , j. l. 2006 , phd th . , univ . of victoria dallora , m. , et al . 2006 , , 653 , l109 de jong , j. t. a. , butler , d. j. , rix , h - w . , dolphin , & a. e. , martnez - delgado , d. 2007 , , accepted , astro - ph/0701140 dolphin , a. e. 1997 . , 2 , 397 dolphin , a. e. 2001 , , 332 , 91 duquennoy , a. & mayor , m. 1991 , , 248 , 485 fukugita , m. , ichikawa , t. , gunn , j. e. , doi , m. , shimasaku , k. , & schneider , d. p. 1996 , , 111 , 1748 gallart , c. , aparicio , a. , bertelli , g. , & chiosi , c. 1996 , , 112 , 1950 , l. , bertelli , g. , bressan , a. , chiosi , c. , groenewegen , m. a. t. , marigo , p. , salasnich , b. , & weiss , a. 2002 , , 391 , 195 girardi , l. , grebel , e. k. , odenkirchen , m. , & chiosi , c. 2004 , , 422 , 205 gunn , j.e . 1998 , , 116 , 3040 gunn , j.e . 2006 , , 131 , 2332 harris , w. e. 1996 , , 112 , 1487 harris , w. e. , et al . 1997 , , 114 , 1030 harris , j. & zaritsky , d. 2001 , , 136 , 25 hernandez , x. , gilmore , g. & valls - gabaud , d. 2000 , , 317 , 831 hogg , d.w . , finkbeiner , d.p . , schlegel , d.j . , & gunn , j.e . 2001 , aj , 122 , 2129 holtzman , j. a. et al . 1999 , , 118 , 2262 ibata , r. a. , chapman , s. , irwin , m. j. lewis , g. , & martin , n. f. 2006 , , 373 , l70 ivezi , . 2004 , an , 325 , 583 kroupa , p. , tout , c. a. & gilmore , g. 1993 , , 262 , 545 lupton , r. , gunn , j. , & szalay , a. 1999 , , 118 , 1406 muoz , r. r , carlin , j. l. , frinchaboy , p. m. , nidever , d. l. , majewski , s. r. , & patterson , r. j. 2006 , , 650 , l51 olsen , k. a. g. 1999 , , 117 , 2244 sarajedini , a. 1997 , , 113 , 682 schlegel , d. , finkbeiner , d. & davis , m. 1998 , , 500 , 525 smith , j. a. , et al . 2002 , , 123 , 2121 stoughton , c. et al . 2002 , aj , 123 , 485 tolstoy , e. & saha , a. 1996 , , 462 , 672 york , et al . , 2000 , , 120 , 1579

a pattern of chemical reactions frequently encountered in cell biology is one where @xmath0 phosphate groups are attached to a protein by reactions catalysed by one enzyme @xmath2 ( a kinase ) and removed again by reactions catalysed by another enzyme @xmath3 ( a phosphatase ) . this is sometimes called a multiple futile cycle . an introduction to this type of biological system and how it can be modelled mathematically using ordinary differential equations ( ode ) depending on parameters can be found in @xcite . it is proved in @xcite that for a @xmath0-fold futile cycle , @xmath4 , this system of ode exhibits multistationarity for certain values of the parameters , i.e. that there exist several different stationary solutions . upper and lower bounds for the number of stationary solutions as a function of @xmath0 were also proved in @xcite . for @xmath5 there is a unique stationary solution and it is globally asymptotically stable @xcite . the results which follow concern the case @xmath1 of this system , the dual futile cycle . in that case the maximal number of stationary solutions for any values of the parameters is three . note for comparison that in the case @xmath6 while the results of @xcite only guarantee that the maximal number of stationary solutions is between three and five it was recently shown in @xcite that the upper bound is sharp . in other words , for @xmath6 there are parameter values for which there exist five stationary solutions . beyond the question of the number of stationary solutions it is of great interest to obtain information about their stability . this allows conclusions about the significance of the stationary solutions for the dynamics of more general solutions of the system . it is of particular interest to know whether there exist more than one stable stationary solution for fixed values of the parameters . this phenomenon is called bistability . ( in ode describing chemical systems there are often preferred invariant affine subspaces , the stoichiometric compatibility classes . when talking about fixed parameters it is understood that the stoichiometric compatibility class has been fixed . ) in @xcite it was concluded using numerical and heuristic approaches that there is bistability in the dual futile cycle . to the authors knowledge there is no rigorous and purely analytical proof of this statement in the literature . the main result of the present paper is a proof of this type . in @xcite and @xcite the reactions are modelled using a standard michaelis - menten scheme for the catalysis of each reaction and mass action kinetics for the elementary reactions involved . the resulting system will be called the mm - ma system ( michaelis - menten via mass action ) in what follows and it is the system of principal interest here . it is possible , via a quasistationarity assumption of michaelis - menten type , to pass formally to a smaller system , called the mm system ( michaelis - menten ) in what follows . the question of bistability in the latter system has been studied by methods which are partly numerical and heuristic in @xcite . the strategy used in what follows is to first give a rigorous analytical proof of bistability in the mm system using bifurcation theory . it is shown that there is a generic cusp bifurcation and this implies bistability by well - known methods @xcite . then bistability is concluded for the mm - ma system with the help of geometric singular perturbation theory @xcite which gives control over the limiting process from the mm - ma system to the mm system and can be thought of as a far - reaching generalization of earlier work of tikhonov ( see @xcite , sect . 39 ) and hoppensteadt @xcite . the relevance of this type of result to the quasistationary approximation was pointed out in @xcite . a similar strategy has been used in @xcite to prove that generic solutions of the system describing the dual futile cycle converge to stationary solutions . the paper is organized as follows . in the next section the basic equations of the model studied in the paper are explained . [ techniques ] is a concise introduction to some of the main mathematical tools used . in sect . [ bistabmm ] bistability is proved for the michaelis - menten system and in sect . [ bistabma ] this is used to obtain a corresponding result for the full system . in sect . [ further ] some directions in which this research could be extended are indicated . these concern the mapk cascade . in particular it is discussed how michaelis - menten reduction can be applied in that case and an explicit reduced system is presented . consider a chemical system consisting of a protein @xmath7 and the substances @xmath8 and @xmath9 obtained by attaching one or two phosphate groups to @xmath7 . the reactions which attach phosphate groups are catalysed by an enzyme @xmath2 and those which remove phosphate groups by an enzyme @xmath3 . it is assumed that the enzyme @xmath2 can only add one phosphate group before releasing the substrate . this is what is called distributive phosphorylation in contrast to processive phosphorylation where more than one phosphate is added during one encounter between the enzyme and its substrate . it is also assumed that the phosphate groups are added at binding sites in a certain order . this is called sequential phosphorylation . it is assumed that dephosphorylation by @xmath3 has corresponding properties and that phosphate groups are removed in the reverse order to that in which they are added . these assumptions about the nature of the phosphorylation and dephosphorylation processes are common in modelling approaches in the literature . adopting them , the chemical reactions we are modelling can be written in the following form where the label @xmath10 denotes the reaction constant of reaction @xmath11 . @xmath12^{k_1 } & ye \ar@/^1pc/[l]^{k_2 } \ar[r]^{k_3 } & yp+e \\ & & \\ yp+e \ar@/^1pc/[r]^{k_4 } & ype\ar@/^1pc/[l]^{k_5 } \ar[r]^{k_6 } & ypp+e\\ & & \\ ypp+f \ar@/^1pc/[r]^{k_7 } & yppf \ar@/^1pc/[l]^{k_8 } \ar[r]^{k_9 } & yp+f\\ & & \\ yp+f \ar@/^1pc/[r]^{k_{10 } } & ypf\ar@/^1pc/[l]^{k_{11 } } \ar[r]^{k_{12 } } & y+f } \ ] ] an important biological example is the case where @xmath7 is the extracellular signal - regulated kinase ( erk ) , @xmath2 is the mapk / erk kinase ( mek ) and @xmath3 is mapk phosphatase 3 ( mkp3 ) . this is part of a mitogen activated protein kinase ( mapk ) cascade , a pattern of chemical reactions of key significance in molecular biology . some experimental results on this system are summarized in @xcite and are as follows . both mek and mkp3 act in a distributive fashion . dephosphorylation by mkp3 is sequential but phosphorylation by mek has a random component - one order of phosphorylation predominates but the other also occurs . there is some discussion in @xcite of the effects of this fact on the dynamics . in what follows the analysis is restricted to the case where both @xmath2 and @xmath3 act sequentially . it is of interest to note that if distributive phosphorylation is replaced by processive phosphorylation for one or both of the enzymes in this model then the existence of more than one stationary solution , and in particular bistability , can be ruled out @xcite . the concentration of a substance @xmath13 is denoted by @xmath14 . if mass action kinetics are imposed then the assumptions made up to this point uniquely determine the evolution equations for the concentrations of the substances taking part in the reactions . these are the free substrates @xmath7 , @xmath8 and @xmath9 , the free enzymes @xmath2 and @xmath3 and the substrate - enzyme complexes @xmath15 , @xmath16 , @xmath17 and @xmath18 . they satisfy the following system of nine ordinary differential equations : @xmath19 the reaction constants @xmath10 are positive numbers . it should be noted that apart from the differences in notation these equations are identical to those in @xcite and @xcite . let @xmath20 these quantities , which are the total concentrations of the enzymes and of the substrate , are conserved under the evolution . thus @xmath21 and @xmath22 can be expressed in terms of @xmath23 and @xmath24 and the concentrations of their complexes . in a similar way , @xmath25 can be expressed in terms of @xmath26 , @xmath27 , @xmath28 and the concentrations of the complexes involving @xmath7 . it is thus possible to discard the evolution equations ( [ dfc2 ] ) , ( [ dfc8 ] ) and ( [ dfc9 ] ) and reduce the number of equations from nine to six . the notation @xmath25 should now be thought of as an abbreviation for the expression of this quantity in terms of @xmath26 and the unknowns in the remaining evolution equations . the six equations depend on the three parameters @xmath23 , @xmath24 and @xmath26 . if a solution of the system of nine equations is given , a solution of the system of six equations with certain values of the parameters is obtained . conversely , given a solution of the system of six equations and a choice of the constants @xmath23 , @xmath24 and @xmath26 a solution of the set of nine equations can be obtained . in order to ensure that the latter is positive the constants @xmath23 , @xmath24 and @xmath26 must be chosen sufficiently large . next we want to pass to the michaelis - menten limit . this can be done by introducing rescaled variables . if @xmath13 is the concentration of a free enzyme or that of a substrate - enzyme complex let @xmath29 . let @xmath30 and similarly for @xmath3 . the smallness of @xmath31 amounts to the enzyme concentrations being small compared to the concentrations of the substrates . finally , let @xmath32 . for @xmath31 small , the system exhibits fast - slow dynamics . the two different time scales are reflected in the time variables @xmath33 and @xmath34 . the transformed equations are @xmath35 note that equations ( [ dfc2 ] ) , ( [ dfc8 ] ) and ( [ dfc9 ] ) could also be rescaled in a similar way so that in total a system of nine equations depending on @xmath36 is obtained whose solutions for any @xmath31 are in one to one correspondence with the solutions of the system ( [ dfc1])-([dfc9 ] ) . setting @xmath37 in equations ( [ rescaled4])-([rescaled7 ] ) gives @xmath38 adding these equations in pairs gives @xmath39\tilde x_{e}\label{erel},\\ & & \tilde f=\left[1+\frac{k_{10}}{k_{11}+k_{12}}x_{yp } + \frac{k_7}{k_8+k_9 } x_{ypp}\right]\tilde x_f.\label{frel}\end{aligned}\ ] ] using ( [ stat1])-([stat4 ] ) the evolution equations ( [ rescaled1])-([rescaled3 ] ) can be rewritten as @xmath40 note that @xmath41 . we define @xmath42 to emphasize the dependence on @xmath36 . then for @xmath37 the relation @xmath43 holds so that setting @xmath44 this is a closed system of equations . summing up , the equations are of the form @xmath45 where , using ( [ erel ] ) and ( [ frel ] ) , @xmath46 and @xmath47 the conditions for a stationary solution are @xmath48 and @xmath49 . from now on the simplifying assumption will be made that @xmath50 call the common value of these quantities @xmath51 . the simplifying assumption ( [ common ] ) means that the ratios @xmath52 , @xmath53 , between the constants of the reactions producing and consuming the substrate - enzyme complexes during phosphorylation are equal . note that ( [ common ] ) is equivalent to a set of relations between the @xmath10 and @xmath51 . if @xmath54 and @xmath55 are given then @xmath56 , @xmath57 , @xmath58 and @xmath59 are determined by the @xmath60 and @xmath51 . there are then four degrees of freedom in choosing the other @xmath10 so as to satisfy the other relations . thus if values @xmath60 , @xmath51 , @xmath54 and @xmath55 are given they can be realized by many choices of @xmath10 . in this section some ideas from bifurcation theory and geometric singular perturbation theory will be reviewed . bifurcation theory for ordinary differential equations concerns parameter - dependent systems @xmath61 where the variable @xmath62 and the parameter @xmath63 belong to subsets of euclidean spaces . under some circumstances the flows of the equations for different values of @xmath63 are related to each other by a diffeomorphism of the space with coordinates @xmath62 . in that case the qualitative nature of the flow does not change with @xmath63 . this is for instance true in a neighbourhood of a point @xmath64 and a parameter value @xmath65 for which @xmath66 ( flow - box theorem ) , or in a neighbourhood of a hyperbolic equilibrium ( hartman - grobman theorem ) . if the flows can not be related in this way for @xmath63 close to some value @xmath65 then it is said that a bifurcation occurs at @xmath65 . for example at an equilibrium @xmath64 ( where of course @xmath67 ) , a bifurcation occurs iff the linearisation @xmath68 admits eigenvalues with vanishing real part . in this case a non - trivial centre manifold exists . the aim of bifurcation theory is to obtain insights into the qualitative changes in the flow in this kind of situation . this can often provide valuable information on the dynamics of the system for parameter values close to @xmath65 , for instance the number and stability of stationary solutions . this information is obtained when certain criteria are satisfied . typical examples of relevant conditions are whether certain combinations of derivatives of @xmath69 at @xmath70 are zero or non - zero or the signs of such combinations . when studying a bifurcation at a stationary point @xmath64 it is often possible to reduce the dimension of the space of unknowns @xmath62 using centre manifold theory . a detailed discussion of this can be found in chapter 5 of @xcite . the essential qualitative features occur on the centre manifold whose dimension is equal to the dimension of the space of generalized eigenvectors of the linearization @xmath71 corresponding to purely imaginary eigenvalues . to make this more concrete we consider the example of greatest importance in what follows , the cusp bifurcation . suppose that the dynamical system is one - dimensional , remembering that this may have resulted from a system of higher dimension by centre manifold reduction . suppose that @xmath63 has two components @xmath72 and @xmath73 . by choosing coordinates appropriately it can be assumed that the bifurcation occurs at the point @xmath74 . the system is said to exhibit a generic cusp bifurcation if @xmath75 where the subscripts denote partial derivatives with respect to the corresponding variables . if only the first three of these conditions are satisfied it might be said loosely that there is a cusp bifurcation but this does not exclude the possibility that extra degeneracies might occur . these can be ruled out by imposing the last two conditions and this is what is meant by the term generic here . when the one - dimensional system arises from a system of higher dimension by centre manifold reduction the diagnostic conditions for a cusp bifurcation can be reexpressed in terms of algebraic conditions on derivatives of the original system , as explained in chapter 8 of @xcite . this will be seen in more detail in the example of the dual futile cycle in sect . [ bistabmm ] . the significance of the cusp for the results of this paper is that when a bifurcation of this type with @xmath76 occurs at some point there are nearby parameter values for which there exist two stable stationary solutions , and one which is unstable . the case @xmath77 gives one stable and two unstable stationary solutions and is not relevant for what follows . the name cusp bifurcation comes from the following . figure [ cuspfig ] shows @xmath78 , where the horizontal plane is the plane of parameters @xmath79 , and the vertical direction the one dimensional variable @xmath62 . bifurcation diagram for a generic cusp bifurcation : the unstable branch of the surface of equilibria is shaded , as well as the region in the parameter plane with multiple equilibria.,scaledwidth=100.0% ] the set of equilibria forms a surface in this bifurcation diagram . near a cusp bifurcation , this surface shows a fold . the line delimiting the folding region projected into the horizontal parameter plane forms a cusp , although the surface in @xmath78 is smooth . for a given parameter in the cusp region , there are two stable equilibria and one unstable one . furthermore there are heteroclinic orbits connecting the unstable equilibrium to the two stable ones . next some remarks will be made on geometric singular perturbation theory ( gspt ) . this theory was developed by fenichel @xcite and section 3 of that paper provides an introduction to some of the main ideas involved . the standard situation in which this theory is applied is for a system of ode of the form @xmath80 where the prime denotes the derivative with respect to a time coordinate @xmath34 . here @xmath69 and @xmath81 are smooth functions and @xmath36 is a parameter . the aim is to understand the qualitative behaviour of solutions of this system in the limit @xmath82 where the system ( [ gspt ] ) reads @xmath83 the system ( [ gspt0 ] ) is called the reduced system . the dynamics of the reduced system may provide useful information about the qualitative behaviour of solutions of the system ( [ gspt ] ) with @xmath36 small and non - zero . the system ( [ gspt ] ) is singular at @xmath37 since the time derivative of @xmath84 is multiplied by a factor which vanishes there . this means that regular perturbation theory can not be applied . note that the system ( [ rescaled1])-([rescaled7 ] ) is of this general form . suppose that in the above system @xmath62 is a point of @xmath85 and @xmath84 a point of @xmath86 . let @xmath32 for @xmath31 and denote the derivative with respect to @xmath33 by a dot . transforming the equations to the time coordinate @xmath33 and adding the equation @xmath87 gives the following system of @xmath88 equations @xmath89 call system ( [ extsys ] ) the extended system . this extends smoothly to @xmath37 to @xmath90 assume that the set defined by the equation @xmath91 is equivalent to @xmath92 for a smooth function @xmath93 . in particular it is a manifold @xmath94 . this manifold consists of stationary points of the extended system ( [ extsys ] ) at @xmath37 . assume further that the linearization of system ( [ extsys ] ) at any point of @xmath94 has no purely imaginary eigenvalues other than the zero eigenvalues arising from the fact that @xmath94 consists of stationary solutions and the fact that @xmath36 is a conserved quantity . call these eigenvalues corresponding to eigenvectors transverse to @xmath94 the transverse eigenvalues . they correspond to the eigenvalues of the linearization of the system @xmath95 with @xmath62 held fixed . this leads to the following [ transeigen ] for a point @xmath96 , we define the transverse eigenvalues at @xmath62 as the eigenvalues of the linearisation @xmath97 , under the assumption that none of them is purely imaginary . a centre manifold @xmath98 of the extended system ( [ extsys ] ) at any point of @xmath94 contains @xmath94 . it was shown in @xcite that there exists a manifold @xmath98 which is a centre manifold for all points of @xmath94 close to a given point . the manifold @xmath98 is sometimes called a slow manifold . the restriction of the extended system ( [ extsys ] ) to the slow manifold @xmath98 can be interpreted as a system of @xmath99 equations depending on the parameter @xmath36 . remarkably , rewriting this system in terms of the time coordinate @xmath34 gives a system which depends on @xmath36 in a regular fashion , even at @xmath37 . for @xmath37 it coincides with the reduced system ( [ gspt0 ] ) . for @xmath31 call this the perturbed reduced system . the reduction theorem of shoshitaishvili ( see @xcite , theorem 5.4 ) shows that the qualitative behaviour of solutions of the extended system ( [ extsys ] ) near a point of @xmath94 is determined in a simple way by the dynamics on @xmath98 . in fact it is topologically equivalent to the product of the dynamics on @xmath98 with a standard saddle . what this means in practise is that dynamical features of the reduced system ( [ gspt0 ] ) are inherited by the full system ( [ gspt ] ) for @xmath36 small . in the case of the central example of this paper the real parts of the transverse eigenvalues of the linearization at points of @xmath94 defined in [ transeigen ] are all negative . thus the standard saddle mentioned above is a hyperbolic sink in this case . this implies that bistability of the mm system ( which is the reduced system in this case ) is inherited by the perturbed reduced system . it then follows from the theorem of shoshitaishvili that bistability is inherited by the mm - ma system . in other situations , as will be explained further in sect . [ further ] , the existence of periodic solutions of the reduced system is inherited by the full system . the mm system can be written as @xmath100 it has been observed in @xcite that if certain restrictions are imposed on the parameters it is possible to find an explicit stationary solution of equations ( [ mm1])-([mm2 ] ) with interesting properties . consider a stationary solution that we call @xmath101 , with starred coordinates @xmath102 and the corresponding quantity @xmath103 . we look for an equilibrium @xmath101 satisfying @xmath104 , a condition which is satisfied by the special solutions considered in @xcite . then the denominators in the expressions for the evolution equations are all equal and the equations for stationary solutions imply that @xmath105 it follows that @xmath106 let @xmath107 . the linearization of the system at the stationary point @xmath101 is @xmath108 times @xmath109 & -a_1bx - a_2 n \\ -a_3 n - ba_4x & -(a_3+a_4)+b[-a_4(\tilde y - x)-a_3x ] \end{array } } \right]\ ] ] where @xmath110 . the equations @xmath111 and @xmath112 can be used to eliminate @xmath113 and @xmath114 in favour of @xmath115 and @xmath116 . after simplification and multiplication by an overall factor @xmath62 the matrix ( [ rescaledmatrix ] ) becomes @xmath117.\ ] ] now we look for a stationary point @xmath101 where a bifurcation takes place , so that the above matrix should admit a zero eigenvalue . the determinant of this matrix vanishes precisely when @xmath118 there are two cases according to the relative signs of the quantities inside the absolute values . setting @xmath119 and @xmath120 we get the alternative conditions @xmath121 and @xmath122 . the case @xmath123 can be discarded since it corresponds to a vanishing concentration of substrate . the case @xmath124 is also not relevant since it corresponds to a vanishing concentration of @xmath25 . the case @xmath125 leads to a negative concentration of @xmath25 . consider now the remaining case @xmath126 . a computation shows that @xmath127 now @xmath128 we have a bifurcation if and only this last relation is satisfied . a necessary and sufficient condition that it can be satisfied for parameters with @xmath129 positive is that @xmath130 . in that case @xmath131 . we summarize the conditions necessary and sufficient for the existence of the stationary point denoted by @xmath101 with coordinates given by ( [ expstat ] ) : @xmath132 substituting the bifurcation condition into the linearization @xmath108 times ( [ rescaledmatrix ] ) simplifies it to @xmath133.\ ] ] this matrix has rank one with the right eigenvector corresponding to the zero eigenvalue being the transpose of @xmath134 $ ] and the left eigenvector @xmath135 $ ] . the other eigenvalue has the same sign as the trace and is negative . next it will be shown that there is a non - degenerate cusp bifurcation at the point @xmath101 . the defining conditions for a bifurcation of this type are given by ( [ cusp ] ) in the case of a one - dimensional system . as discussed in sect . [ techniques ] a non - degenerate cusp bifurcation in higher dimensional dynamical systems is defined by relating the given system to the one - dimensional case by means of centre manifold reduction . the essential dynamics is determined by that on the centre manifold which is one - dimensional in this case . the point @xmath101 in the mm system has a one - dimensional centre manifold and so these techniques are applicable . the reduction theorem makes transparent what is happening on an abstract level . on the other hand the concrete calculations which are needed to verify the presence of a non - degenerate cusp bifurcation by obtaining a suitable approximation to the centre manifold and analysing the dynamics on that manifold are hard to keep track of . here we follow an approach to organizing these calculations explained in section 8.7 of @xcite . it is necessary to calculate certain combinations of derivatives of the right hand side of the mm system at the point @xmath101 with respect to the unknowns and the parameters . for the cusp bifurcation it is necessary to vary two parameters . we choose these to be @xmath113 and @xmath114 and hold all other parameters in the system fixed . it is convenient to introduce the notation @xmath136 , @xmath137 for the right hand sides of the equations of the mm system ( [ mm1])-([mm2 ] ) and @xmath138 , @xmath137 for the coordinates @xmath27 and @xmath28 . the components of the right and left eigenvectors of the linearization will be denoted by @xmath139 and @xmath140 , respectively . for derivatives we use the following abbreviations @xmath141 now higher derivatives of the coefficients in the system mm will be examined . for this it is useful to consider derivatives of rational functions more generally . each @xmath136 is a sum of rational functions where numerator and denominator are linear in the arguments . if @xmath69 and @xmath81 are smooth functions of two variables with @xmath81 non - vanishing then @xmath142 when @xmath69 is linear the term with @xmath143 vanishes . now @xmath144 when @xmath81 is linear the first term vanishes . thus when both @xmath69 and @xmath81 are linear we get @xmath145 contracting with @xmath146 ( i.e. multiplying by this expression and summing over repeated indices ) gives @xmath147 here and in what follows we use the summation convention - sums over repeated indices are implicitly assumed . consider the quantities @xmath148 , evaluated at @xmath101 . they are given by @xmath149,\\ & & x^2{}_{,jk}r^jr^k=2n^{-3}[a_3b^2x_{yp}+a_4b(-n+bx_{ypp})].\end{aligned}\ ] ] after some substitutions this gives @xmath150 = 4n^{-3}a_2 b^2x_{yp},\\ & & x^2{}_{,jk}r^jr^k=\frac23 n^{-3}a_3[3b^2x_{yp}+b(b\tilde y-2 ) ] = 4n^{-3}a_3 b^2x_{yp}.\end{aligned}\ ] ] it follows that @xmath151 and this is one of the conditions for a cusp bifurcation . the next step is to examine the third order derivatives . note first that @xmath152 in the given situation @xmath153 and @xmath154 it follows that @xmath155 hence @xmath156 in the one - dimensional case one of the conditions for a cusp is that the third derivative of the right hand side does not vanish . in higher dimensions the analogue of the third derivative is not just the expression in ( [ thirdder ] ) . there is an extra correction given on p. 374 of @xcite . to compute this we need a vector whose image under the linearization at @xmath101 is the vector with components @xmath148 . call it z. this is @xmath157^t$ ] . the crucial quantity is @xmath158 to evaluate this we first calculate the second derivatives of @xmath159 and @xmath160 , which are @xmath161\ ] ] and @xmath162.\ ] ] contracting with @xmath163 corresponds to taking the differences of the columns of these matrices . the results are @xmath164\ ] ] and @xmath165.\ ] ] eliminating @xmath113 and @xmath114 in terms of @xmath115 and @xmath116 as has been done in previous arguments and expressing everything in terms of @xmath129 gives @xmath166 , \ \ \ n^{-3}\left[\frac23 a_3b(b\tilde y+1)\ \ \ \frac23 a_3b(b\tilde y+1)\right].\ ] ] contracting this with @xmath167^t$ ] and @xmath168 gives @xmath169 . hence @xmath170 the crucial quantity is strictly negative and so another of the conditions for the cusp bifurcation is satisfied . it remains to examine the derivatives with respect to parameters . recall that the parameters to be varied are @xmath171 and @xmath172 . the derivatives with respect to the two parameters of the right hand sides of the two mm equations are @xmath173 contracting these with respect to the left eigenvector @xmath168 gives @xmath174 differentiating the vector field in the direction of the right eigenvector @xmath163 gives @xmath175 ^ 2 } , \frac{a_4(1+bx_{yp})+a_3bx_{yp}}{[1+b(x_{yp}+x_{ypp})]^2}\right).\ ] ] the derivatives of this with respect to the two parameters are @xmath176 ^ 2 } , 0\right),\ \ \ \left(0,\frac{(1+bx_{yp})}{[1+b(x_{yp}+x_{ypp})]^2 } \right).\ ] ] contracting these with respect to the left eigenvector gives @xmath177 ^ 2},\ \ \ \frac{-a_3(1+bx_{yp})}{[1+b(x_{yp}+x_{ypp})]^2}.\ ] ] it follows that @xmath178l_il_jr^k\nonumber\\ & & = -(a_2 ^ 2+a_3 ^ 2)n^{-3}x_y(1+bx_{yp}).\end{aligned}\ ] ] this is the last diagnostic quantity for the cusp bifurcation and it is seen to be non - zero . this means that all conditions needed for the proof that there is a generic cusp bifurcation at @xmath101 have been verified . [ bistabmm ] there exist values of the parameters @xmath60 , @xmath51 and @xmath26 for which the mm system - has three stationary solutions , two of which are locally asymptotically stable and one of which is a saddle of morse index one . heteroclinic orbits connect the saddle to both stable equilibria . it has been shown that when the conditions ( [ cond1 ] ) and ( [ cond2 ] ) are assumed the system exhibits a generic cusp bifurcation . it follows from the analysis in section 8.7 of @xcite that centre manifold reduction leads to a one - dimensional system satisfying the hypotheses of theorem 8.1 of @xcite . this implies that near the bifurcation point the dynamics on the centre manifold is topologically equivalent to the model system for the cusp bifurcation . combining this with theorem 5.4 of @xcite and using the fact that the non - zero eigenvalue of the linearization at the bifurcation point is negative gives the statement about the qualitative properties of the stationary solutions . in this section it is shown that theorem [ bistabmm ] , which asserts bistability for the mm system , can be used to show an analogous result for the mm - ma system . the existence of three stationary solutions for the dual futile cycle with suitable parameter values was proved in @xcite . the aim here is to obtain information about the stability of these solutions . [ bistabmm - ma ] there exist values of the parameters @xmath10 and values of the conserved quantities @xmath23 , @xmath24 and @xmath26 for which the system ( [ dfc1])-([dfc9 ] ) has three stationary solutions , two of which are stable and one of which is a saddle of morse index one . their coordinates are compatible with the given values of the conserved quantities . furthermore heteroclinic orbits connect the saddle to both stable equilibria . information about stability is to be propagated from the mm system to the mm - ma system by using a result showing that solutions of the mm - ma system approximate solutions of the mm system for @xmath36 small . for this it is useful to introduce a condensed notation for the equations ( [ rescaled1])-([rescaled7 ] ) with fixed values of @xmath26 , @xmath54 and @xmath55 . let @xmath179 and @xmath180 . here the quantities @xmath181 and @xmath182 are the coordinates of the stationary solution @xmath101 given by ( [ expstat ] ) and the other starred quantities are those obtained from @xmath181 and @xmath182 by substituting them into - . then the evolution equations for the variables @xmath62 and @xmath84 in the time scale @xmath183 are of the form ( [ gspt ] ) . the dependence on @xmath36 of the right hand sides of these equations arises because @xmath26 depends on @xmath36 . for @xmath37 the second equation of the reduced system ( [ gspt0 ] ) becomes an algebraic equation which can be solved in the form @xmath92 so that @xmath184 . substituting this into the first equation of the reduced system ( [ gspt0 ] ) gives an equation of the form @xmath185 , which is the mm - system ( [ mm1])-([mm2 ] ) . the conditions @xmath186 and @xmath187 hold . the limit of these equations for @xmath82 can be analysed using the results of @xcite as mentioned in sect . [ techniques ] . recall the form of the extended system in terms of the time variable @xmath188 : @xmath189 at each of the points of the manifold @xmath94 defined by the vanishing of @xmath190 the rank of the matrix which defines the linearization of the system is four . this is equal to the rank of the matrix defining the linearization of the equation @xmath191 for fixed values of the variables @xmath62 . in this case the calculation is particularly simple since the right hand side of the equation is an affine function of the unknowns . the matrix is the direct sum of two diagonal @xmath192 blocks . each of these blocks has positive determinant and negative trace and thus all eigenvalues have negative real parts . thus in particular at each of the points of @xmath94 there is a three - dimensional centre manifold . it follows from theorem 9.1 of @xcite that there is a three - dimensional invariant manifold which is a centre manifold for all of these points . it can be chosen to be @xmath193 for any finite @xmath194 and it can be written in the form @xmath195 , where @xmath196 . this manifold can be coordinatized by @xmath62 and @xmath36 and the restriction of the system to the manifold can be represented in the form @xmath197 since it has been shown that the system for @xmath37 has three hyperbolic stationary solutions it follows that the system for small positive values of @xmath36 also has at least three stationary solutions . the heteroclinics connecting the unstable equilibrium to the stable one also persist for small values of @xmath36 . the three stationary solutions are also equilibria of the full system and it follows from the theorem of shoshitaishvili that two of them are stable and the other is a saddle of morse index one connecting to both stable equilbria . there are no more stationary solutions due to the results of @xcite . for a fixed value of @xmath31 and using the values of @xmath26 , @xmath54 and @xmath55 positive quantities @xmath198 , @xmath199 and @xmath200 can be determined which define positive stationary solutions of the rescaled version of ( [ dfc1])-([dfc9 ] ) . undoing the rescaling gives solutions of the original system which all have the same values of the conserved quantities . this completes the proof of theorem [ bistabmm - ma ] . since several conditions are imposed on the parameters of the system , we want to summarize here the remaining degrees of freedom . we already saw that the simplifying condition , given @xmath113 , @xmath115 , @xmath116 , @xmath114 , @xmath51 , @xmath201 and @xmath202 , leaves 4 degrees of freedom for the coefficients @xmath10 . the equilibrium condition is a condition on the parameters @xmath113 , @xmath115 , @xmath116 and @xmath114 and leave us three degrees of freedom for them : given @xmath113 , @xmath115 , @xmath116 and @xmath114 we have @xmath203 . putting this into the bifurcation condition gives @xmath204 , which is only a slight restriction to the freedom of chosing @xmath116 . hence , the parameters @xmath51 , @xmath113 , @xmath204 , @xmath201 , @xmath202 and for example @xmath205 , @xmath206 , @xmath207 and @xmath208 can be chosen freely and then all other parameters in the system ( the other @xmath10 , @xmath115 and @xmath209 ) are determined uniquely by the conditions , and . in particular , we can choose @xmath113 and @xmath114 in such a way that the system is indeed in the cusp region where bistability holds . we can also look at those conditions from the point of view of fixed reaction constants @xmath10 . we saw that we can see the simplifying condition as a comparison between reaction constants of production and consumption of substrate - enzyme complexes during phosphorylation . once the simplifying assumption has been made , the equilibrium condition translates to @xmath210 the quantities of enzymes @xmath211 have to be tuned accordingly . together with bifurcation condition , equation leads to the inequality @xmath212 the quotients @xmath213 and @xmath214 again compare reaction constants : @xmath213 compares the constants of the reactions @xmath215^{k_3 } & yp+e & , & ypf \ar[r]^{k_{12 } } & y+f , } \ ] ] and the second quotient the analogous reactions for the double phosphorylated form @xmath9 . in order to understand the biological meaning of this inequality it may be useful to state its concrete interpretation in terms of the reactions . the relative rate of production of the phosphorylated and dephosphorylated form is greater for the first cycle than for the second one . furthermore the total quantity of @xmath7 has to be tuned accordingly . an important cellular signalling system is the mapk cascade . it contains three layers , each of which is a multiple phosphorylation loop of the type of the multiple futile cycle . they are linked by the fact that the fully phosphorylated form of the protein which is the substrate in one layer is the kinase for the next layer . @xmath216^{1}&&mkkk - p\ar@/^1pc/[ll]^{2 } \ar[dl ] \ar[dr]&&&&\\ & & & & & & \\ mkk\ar@/^2pc/[rr]^{3 } & & mkk - p\ar@/^2pc/[rr]^{4 } \ar@/^1pc/[ll]^{6 } & & mkk - pp\ar@/^1pc/[ll]^{5 } \ar[dl ] \ar[dr ] & & \\ & & & & & & \\ & & mapk\ar@/^2pc/[rr]^{7}&&mapk - p\ar@/^2pc/[rr]^{8 } \ar@/^1pc/[ll]^{10}&&mapk - pp\ar@/^1pc/[ll]^{9 } } \ ] ] note that we later focus on the two upper layers of this cascade and use notation @xmath217 and @xmath7 for the substrates @xmath218 and @xmath219 respectively . a fundamental study , both theoretical and experimental , of this system was carried out by huang and ferrell @xcite . they modelled it by a system of ode using mass action kinetics . numerical and heuristic evidence has been found indicating that this system has periodic solutions @xcite . it may be that there is also chaotic behaviour in solutions of this system @xcite . here we will concentrate on the issue of oscillations . to what extent can the ideas used to prove bistability here be applied to give an analytical proof of oscillatory behaviour in the mapk cascade and can michaelis - menten reduction help ? mapk cascades are often embedded in positive or negative feedback loops and it was suggested in @xcite that negative feedback could lead to oscillations in this context . a model was introduced where the mapk cascade is in a negative feedback loop belonging to a more complex network of chemical reactions taking place in the cell . in this model a feedback loop between a phosphorylated form of one protein and a substrate in a layer above is induced by chemical reactions external to the cascade . these other reactions are not modelled in detail - instead a generic form is chosen for the coupling . in the paper @xcite a michaelis - menten - like description of the cascade was used but the assumptions leading to the equations used there were not stated . an insightful discussion of the modelling issues involved has been given in @xcite . the equations in @xcite with feedback loop are of the type called phenomenological in @xcite . an equation with mass action kinetics like the mm - ma model considered above is called mechanistic. finally a model like the mm model considered above is referred to as reduced mechanistic. we will adopt the terminology of @xcite in what follows . in a phenomenological model reduced equations for the individual reactions in a network are used and this may fail to capture interactions between the different reactions . the difference between phenomenological and reduced mechanistic models will now be discussed in some detail in the case of the dual futile cycle . a model for the dual futile cycle can be extracted from the feedback model of @xcite as follows . we take the second layer of the cascade and set the concentration of the phosphorylated form of the protein in the first layer , @xmath220 $ ] in the notation of @xcite , to a constant value . this gives equations for the quantities @xmath221 $ ] and @xmath222 $ ] corresponding to our @xmath223 and @xmath28 . for example , the equation for the first of these is of the form @xmath224}{d\tau}=v_6-v_3\ ] ] where @xmath225)[mkk]}{k_3+[mkk]}.\ ] ] this quantity @xmath226 corresponds to @xmath227 in our notation and the key difference is that it does not depend on @xmath228 $ ] while our @xmath227 does depend on @xmath25 . in this case the model extracted from @xcite is the phenomenological one while our model is the reduced mechanistic one . returning to the question of oscillations in the mapk cascade , note first that according to @xcite oscillations are already found numerically in a truncated version of the huang - ferrell model with only two layers , the first involving only one phosphorylation and the second involving two . for simplicity we now concentrate on that truncated model . more precisely , the reactions involved are the following : @xmath229^{k_1 } & xe \ar@/^1pc/[l]^{k_2 } \ar[r]^{k_3 } & xp+e \\ & & \\ xp+f \ar@/^1pc/[r]^{k_1 } & xp.f \ar@/^1pc/[l]^{k_2 } \ar[r]^{k_3 } & x+f } \\ \text{2nd layer : } & \xymatrix { y+xp \ar@/^1pc/[r]^{k_4 } & y.xp \ar@/^1pc/[l]^{k_5 } \ar[r]^{k_6 } & yp+xp \\ & & \\ yp+xp \ar@/^1pc/[r]^{k_7 } & yp.xp\ar@/^1pc/[l]^{k_8 } \ar[r]^{k_9 } & ypp+xp\\ & & \\ ypp+g \ar@/^1pc/[r]^{k_{10 } } & ypp.g \ar@/^1pc/[l]^{k_{11 } } \ar[r]^{k_{12 } } & yp+g\\ & & \\ yp+f \ar@/^1pc/[r]^{k_{13 } } & yp.g\ar@/^1pc/[l]^{k_{14 } } \ar[r]^{k_{15 } } & y+g } \end{aligned}\ ] ] michaelis - menten reduction is more subtle in this case due to the fact that a single substance , @xmath230 , is both a substrate in one reaction and an enzyme in others . this problem has been analysed in @xcite and @xcite . in fact these papers treat more general sequences of coupled multiple phosphorylation loops . the numerical results of @xcite lead to the conclusion that the periodic solutions of the mechanistic model seen in @xcite are also present in the reduced mechanistic model . they also show that the reduced mechanistic model often gives more accurate results than a phenomenological model even when the parameter @xmath36 is not very small . while the reduction procedure of @xcite and @xcite is close to being a direct generalization of the reduction to the mm system done above to the case of the truncated huang - ferrell model it is not exactly the same and it needs to be modified slightly to give a limit for @xmath82 which is well - behaved from the point of view of gspt . this modification will now be carried out . the notation is chosen to fit that in the rest of the paper . the proteins in the first and second layers are denoted by @xmath217 and @xmath7 respectively . the kinase in the first layer is denoted by @xmath2 and the phosphatases in the first and second layers by @xmath3 and @xmath231 . the basic unknowns are the concentrations of @xmath217 , @xmath230 , @xmath7 , @xmath8 , @xmath9 , @xmath232 , @xmath233 , @xmath234 , @xmath235 , @xmath236 , @xmath237 , @xmath2 , @xmath3 and @xmath231 . the conservation laws for the three enzymes @xmath2 , @xmath3 and @xmath231 and the two substrates @xmath217 and @xmath7 can be used to eliminate five of the variables . a key idea in @xcite is to introduce a new variable @xmath238 , which leads to a cancellation in one of the evolution equations . new variables are introduced by rescaling in a way similar to that done for the dual futile cycle above . they are defined by the relations @xmath239 , @xmath240 , @xmath241 , @xmath242 , @xmath243 , @xmath244 , @xmath245 , @xmath246 , @xmath247 , @xmath248 , @xmath32 . in addition , and this is the difference to what was done in @xcite , two of the reaction constants are rescaled , @xmath249 and @xmath250 . these are the reaction constants for the binding of the enzymes @xmath2 and @xmath3 to their substrates . the result is the system @xmath251 it is then possible to pass to the limit @xmath82 . carrying out steps analogous to those done in the case of the dual futile cycle in a routine way gives @xmath252 the constants in this system can be expressed explicitly in terms of the parameters in the full system . to get a closed system the conservation laws for the total amounts of the substrates @xmath217 and @xmath7 must be used . this system may be compared with one extracted from the system of @xcite in a way similar to that done for the dual futile cycle above . this time we take the equations for the substances in the first two layers and remove the explicit feedback . this is done by setting the concentrations @xmath253 $ ] , @xmath254 $ ] and @xmath255 $ ] to zero . then the equations for those quantities are satisfied identically and the dependence of the first reaction rate @xmath227 in @xcite on @xmath255 $ ] disappears . in the resulting system the evolution equations for the second layer substrates @xmath221 $ ] and @xmath222 $ ] are independent of the variables involving the third layer substrate @xmath253 $ ] . this is in contrast to the system ( [ mapkmm1])-([mapkmm3 ] ) where there is an intrinsic negative feedback loop . for if the right hand sides of these equations are denoted by @xmath256 then @xmath257 while @xmath258 . multiplying the two different signs of these derivatives gives the claimed negative feedback . ( for background on the defintion and significance of feedback loops the reader is referred to @xcite . ) in the equations extracted from the system of @xcite , on the other hand , we have @xmath259=0 $ ] and there is no intrinsic feedback . according to common heuristics the negative feedback loop in ( [ mapkmm1])-([mapkmm3 ] ) can be seen as a possible source of oscillations . the interaction not captured in the phenomenological description is what is called sequestration . in certain circumstances a lot of the substance @xmath230 is bound to its substrates @xmath7 and @xmath8 and so is not available to take part in the reaction converting it back to @xmath217 in the first layer . the equations ( [ mapkmm1])-([mapkmm3 ] ) are a reduced system for the truncated huang - ferrell model in the sense of gspt . the transverse eigenvalues ( as defined in [ transeigen ] ) have negative real parts . in contrast to the system arising for the dual futile cycle the right hand side of the equations entering the computation of the transverse eigenvalues is no longer affine . however the only nonlinear terms are in the equation for @xmath260 and the derivatives of these terms do not influence the eigenvalues . all the other terms are either diagonal or belong to @xmath192 blocks which are similar to those already treated in the case of the dual futile cycle . hence if it were possible to show that ( [ mapkmm1])-([mapkmm3 ] ) has a hyperbolic stable periodic solution then it would follow immediately that for @xmath36 sufficiently small the full system also has a periodic solution . this is because hyperbolic stable periodic solutions persist under regular perturbations of a dynamical system . thus the task of proving the existence of the periodic solutions of the truncated huang - ferrell model found numerically in @xcite reduces to proving an analogous statement for the reduced system . since the former has dimension nine , even after exploiting all conservation laws , and the latter has dimension three , this could be a significant advantage . the question of existence of periodic solutions of the reduced system will be investigated elsewhere . another interesting question concerns the dependence of the results of theorems [ bistabmm ] and [ bistabmm - ma ] on the modelling by mass - action kinetics . a crucial argument in the present paper consists in verifying the conditions for a cusp bifurcation to take place . it is already technical for the mass - action kinetics where equilibria and derivatives can be computed explicitly . on the one hand , such computations seem difficult with a more general model for the kinetics of the chemical reactions . but on the other hand , the fact that the evolution equations have a polynomial reaction rates is not essential . models with more general right hand side could lead to similar results . furthermore , we conjecture that the assumption is not a necessary condition , but only a convenience to simplify the computations . angeli , d. and sontag , e. d. 2006 translation - invariant monotone systems and a global convergence result for enzymatic futile cycles . rwa 9 , 128140 . conradi , c. , saez - rodriguez , j. , gilles , e .- d . and raisch , j. 2005 using chemical reaction network theory to discard a kinetic mechanism hypothesis . systems biology 152 : 243248 . fenichel , n. 1979 geometric singular perturbation theory for ordinary differential equations . 31 , 5398 . flockerzi , d. , holstein , k. and conradi , c. 2013 @xmath0-site phosphorylation systems with @xmath261 steady states . preprint arxiv:1312.4774 . heineken , f. g. , tsuchiya , h. m. and aris , r. 1967 on the mathematical status of the pseudo - steady state hypothesis of biochemical kinetics . biosci . 1 , 95 - 113 . hoppensteadt , f. c. 1966 singular perturbations on the infinite interval . 123 , 521535 . huang , c .- y . f. and ferrell , j. e. 1996 ultrasensitivity in the mitogen - activated protein kinase cascade . usa 93 , 1007810083 . kholodenko , b. n. 2000 negative feedback and ultrasensitivity can bring about oscillations in the mitogen activated protein kinase cascades . j. biochem . 267 , 15831588 . kuznetsov , y. a. 2010 elements of applied bifurcation theory . springer , berlin . markevich , n. i. , hoek , j. b. and kholodenko , b. n. 2004 signaling switches and bistability arising from multisite phosphorylation in protein kinase cascades . j. cell biol . 164 , 353359 . ortega , f. , garcs , j. l. , mas , f. , kholodenko , b. n. and cascante , m. 2006 bistability from double phosphorylation in signal transduction . febs j. 273 , 39153926 . qiao , l. , nachbar , r. b. , kevrekidis , i. g. and shvartsman , s. y. 2007 bistability and oscillations in the huang - ferrell model of mapk signaling . plos comp . 3 , 18191826 . sontag , e. 2007 monotone and near monotone biochemical networks . synth . biol . ventura , a. c. , sepulchre , j .- a . merajver , s. d. 2008 a hidden feedback in signalling cascades is revealed . plos comp . biol . ventura , a. c. and sepulchre , j .- a . 2013 intrinsic feedbacks in mapk signalling cascades lead to bistability and oscillations . acta biotheor . 61 , 5978 . wang , l. and sontag . e. d. 2008 on the number of steady states in a multiple futile cycle . 57 , 2952 . wang , l. and sontag . e. d. 2008 singularly perturbed monotone systems and an application to double phosphorylation cycles . j. nonlin . 18 , 527550 . wasow , w. 1965 asymptotic expansions for ordinary differential equations . dover , new york . zumsande , m. and gross , t. 2010 bifurcations and chaos in the mapk signalling cascade . 265 , 481491 .

causality is a topic which nowadays receives much attention . statisticians , epidemiologists , biostatisticians , social scientists , computer [ especially those in artificial intelligence , see , e.g. , @xcite ( @xcite ) ] , econometricians and philosophers are investigating questions like `` what would have happened if '' and `` what would happen if . '' this article discusses estimating the effect of a time - varying treatment . as a recurring example , this article focuses on the effect of a medical treatment which is adapted to a patient s state during the course of time . large observational studies have become widely used in medical research when data from randomized experiments are not available . randomized clinical trials are often expensive , impractical , and sometimes unfeasible for ethical reasons because treatment is withheld from some patients regardless of medical considerations . also , in some instances , exploratory investigations using nonexperimental data are used before conducting a randomized trial . in observational studies there is no pre - specified treatment protocol . data are collected on patient characteristics and treatments in the course of the normal interaction between patients and doctors . obviously , it is considerably more difficult to draw correct causal conclusions from observational data than from a randomized experiment . the main reason is the so - called confounding by indication or selection bias . for example , doctors may prescribe more medication to patients who are relatively unhealthy . thus , association between medication dose and health outcomes may arise not only from the treatment effect but also from the way the treatment was assigned . if this confounding by indication only takes place at the start of the treatment , one can condition on initial patient characteristics or covariates , such as blood pressure or number of white blood cells , in order to remove the effect of the confounding , and get meaningful estimates of the treatment effect . linear regression , logistic regression or cox regression can be used for this purpose . however , estimating treatment effects is more difficult if treatment decisions after the start of the treatment are adapted to the state of the patients in subsequent periods . treatment might be influenced by a patient s state in the past , which was influenced by treatment decisions before ; thus , simply conditioning on a patient s state in the past means disregarding information on the effect of past treatment . in such a case , even the well - known time - dependent cox model does not answer the question of whether , or how , treatment affects the outcome of interest . the time - dependent cox model studies the rate at which some event of interest happens ( e.g. , the patient dying ) , given past treatment- and covariate history . however , under time - dependent confounding , past covariate values may have been influenced by previous treatment . the net effect of treatment can thus not be derived from just this rate ; see , for example , @xcite , @xcite or @xcite . structural nested models , proposed in @xcite , lok , gill , van der vaart and robins ( @xcite ) and robins ( @xcite ) to solve practical problems in epidemiology and biostatistics , effectively overcome these difficulties and estimate the effect of time - varying treatments . the main assumption underlying these models is that all information the doctors used to make treatment decisions , and which is predictive of the patient s prognosis with respect to the final outcome , is available for analysis . this assumption of `` no unmeasured confounding '' makes it possible to distinguish between treatment effect and selection bias . what data have to be collected to satisfy this assumption of no unmeasured confounding is for subject matter experts to decide . all of the past treatment- and covariate information which both ( i ) influences a doctor s treatment decisions and ( ii ) is relevant for a patient s prognosis with respect to the outcome of interest , has to be recorded . in section 5 of @xcite and in section 8.1 of robins , rotnitzky and scharfstein ( @xcite ) , a sensitivity analysis methodology for estimation of structural nested models is developed that does not assume no unmeasured confounders . beyond treatment and covariates , the data requirements also include the measure of an outcome of interest ; for example , survival time , time to clinical aids or cd4 count after the treatment period . lok et al . ( @xcite ) study structural nested models in discrete time . these models assume that changes in the values of the covariates and treatment decisions take place at finitely many deterministic times , which are the same for all patients and known in advance . lok et al . ( @xcite ) also assume that covariates and treatment take values in a discrete space . they indicate why it is reasonable to expect consistency and asymptotic normality in discrete time , and they refer to @xcite for the proofs . @xcite generalize lok et al . ( @xcite ) to covariates and treatment taking values in @xmath0 . in this article we consider structural nested models in continuous time , proposed in robins ( @xcite ) . structural nested models in continuous time allow for both changes in the values of the covariates and treatment decisions to take place at arbitrary times for different patients . as noted in @xcite , structural nested models in continuous time assume that a short duration of treatment has only a small effect on the distribution of the outcome of interest . the effect of the treatment on an individual patient may be large , but then the probability of such effect has to be small for any particular short duration of treatment ( see page 7 , bottom ) . this article provides a conceptual framework and mathematical formalization of these practical methods , solving important outstanding problems and contributing to the causality discussion , especially for the time ordered and continuous time case . in particular , this article proves the conjectures in @xcite that structural nested models in continuous time lead to estimators which are both consistent and asymptotically normal . the proof simplifies considerably for structural nested models in discrete time ( see our discussion , section [ discu ] ) . this article also proves that a test related to the score test can be used to investigate whether treatment affects the outcome of interest without specifying a model for the treatment effect . the setting to which structural nested models in continuous time apply is as follows . the outcome of interest , from now on called @xmath1 , is a continuous real variable . for example , the survival time of a patient , time to clinical aids , or cd4 count after the treatment period . we wish to estimate the effect of treatment on the outcome @xmath1 . there is some fixed time interval @xmath2 $ ] , with @xmath3 a finite time , during which treatment and patient characteristics are observed for each patient . we suppose that after time @xmath3 treatment is stopped or switched to some kind of baseline treatment . in this article we assume that there is no censoring , and that the outcome @xmath1 is observed for every patient in the study . see , for example , @xcite , hernn et al . ( @xcite ) and @xcite for ideas about dealing with censoring . we denote the probability space by @xmath4 . the covariate process describes the course of the patient characteristics , for example , the course of the blood pressure and the white blood cell count . we assume that a realization of this covariate process is a function from @xmath2 $ ] to @xmath5 , and that such a _ sample path _ is continuous from the right with limits from the left ( cadlag ) . the covariates which _ must _ be included are those which both ( i ) influence a doctor s treatment decisions _ and _ ( ii ) possibly predict a patient s prognosis with respect to the outcome of interest . if such covariates would not be observed , the assumption of no unmeasured confounding , mentioned in the introduction , will not hold . for the moment consider one single patient . we write @xmath6 for the covariate- _ and _ treatment values at time @xmath7 . we assume that @xmath6 takes values in @xmath8 , and that @xmath9 is measurable for each @xmath10 $ ] . moreover , we assume that @xmath11 , seen as a function on @xmath2 $ ] , is cadlag . we write @xmath12 for the covariate- and treatment history until time @xmath7 , and @xmath13 for the space of cadlag functions from @xmath14 $ ] to @xmath8 in which @xmath15 takes it values . similarly , we write @xmath16 for the whole covariate- and treatment history of the patient on the interval @xmath2 $ ] , and @xmath17 for the space in which @xmath16 takes its values . in this article we choose the projection @xmath18-algebra as the @xmath18-algebra on @xmath13 and @xmath17 ; measurability of @xmath19 for each @xmath20 is then equivalent with measurability of the random variable @xmath15 . for technical reasons , we include in @xmath11 a counter of the number of jump times of the measured treatment- and covariate process . we suppose that observations on different patients are independent . structural nested models are models for relations between so - called counterfactuals . consider for a moment just one patient . in reality this patient received a certain treatment and had final outcome @xmath1 . if his or her actual treatment had been stopped at time t , the patient s final outcome would possibly have been different . the outcome he or she would have had in that case we call @xmath21 . of course , @xmath21 is generally not observed , because the patient s actual treatment after @xmath7 is usually different from no treatment ; it is a counterfactual outcome . instead of stopping treatment , one can also consider switching to some kind of baseline treatment , for example , standard treatment . figure [ figuur1 ] illustrates the nature of counterfactual outcomes . we suppose that all counterfactual outcomes @xmath21 , for @xmath10 $ ] and for all patients , are random variables on the probability space @xmath4 . to formalize the assumption of no unmeasured confounding , consider the history of a particular patient . decisions of the doctors at time @xmath7 may be based , in part , on recorded information on the state of the patient and treatment before @xmath7 , that is , on @xmath22 , but not on other features predicting the outcome of the patient . in particular , given @xmath23 , changes of treatment at time @xmath7 should be independent of @xmath21 , the outcome of the patient in case he or she would not have been treated after time @xmath7 , given @xmath23 . note that @xmath21 is an indication of the prognosis of the patient which does not depend on treatment decisions at or after time @xmath7 , since it is the counterfactual outcome which we would have observed if treatment would have been stopped at time @xmath7 . only if treatment would have no effect , the observed outcome @xmath1 could play this role . this is why robins assumption of no unmeasured confounding demands the independence , given @xmath23 , of treatment decisions at time @xmath7 and @xmath21 . similar conditions , though without time - dependence , can be found in , for example , @xcite . the statement `` changes of treatment at @xmath7 should be independent of @xmath21 , the outcome of the patient in case he or she would not have been treated after time @xmath7 , given @xmath23 '' is not a formal statement : it includes conditioning of null events ( since the probability that treatment changes at @xmath7 may be @xmath24 for every fixed @xmath7 ) on null events ( @xmath23 ) . to overcome this difficulty , we assume that the treatment process can be represented by or generates a ( possibly multivariate ) counting process @xmath25 . for instance , @xmath26 registers the number of treatment changes until time @xmath7 and/or the number of times treatment reached a certain level until time @xmath7 . a counting process constructed this way may serve as @xmath25 in the following . more about counting processes can be found in , for example , andersen et al . ( @xcite ) . we assume that the treatment process @xmath25 has an intensity process . formally , such an intensity process @xmath27 is a predictable process such that @xmath28 is a martingale . the intensity @xmath27 with respect to @xmath29 can be interpreted as the rate at which the counting process @xmath25 jumps given the past treatment- and covariate history @xmath23 . [ int ] @xmath25 has an intensity process @xmath27 on @xmath2 $ ] with respect to the filtration @xmath29 . this intensity process satisfies the following regularity conditions : @xmath30 is bounded by a constant which does not depend on @xmath31 , @xmath27 is continuous from the left . according to this assumption , @xmath32 is a martingale on @xmath2 $ ] with respect to the filtration @xmath29 . since most counting process martingale theory deals with filtrations @xmath33 which satisfy the usual conditions ( @xmath34 contains all null sets and @xmath35 ) , we mention that , under , @xmath36 is also a martingale with respect to @xmath37 , the usual augmentation of @xmath29 . this follows from lemma 67.10 in @xcite , since @xmath38 is cadlag . often , @xmath25 will be chosen to count the number of events of a certain type concerning the treatment process ( e.g. , the number of times treatment changed ) . at @xmath3 , the time the study ends , treatment is stopped or switched to baseline treatment , so a natural choice of @xmath25 will often jump at @xmath3 with positive probability . however , jumps of @xmath25 at @xmath3 are not useful for estimation , and we wish to avoid modeling jumps of @xmath25 at @xmath3 . therefore , we assume that , with probability @xmath39 , @xmath25 does not jump at @xmath3 , and if a natural choice of @xmath25 _ does _ jump at @xmath3 with positive probability , then we just adapt it , only at @xmath3 , so that it does not jump there . we also make the following assumption . [ ycadlag ] @xmath40 is a cadlag process . within this framework , the assumption of no unmeasured confounding could be operationalized as follows . the rate at which the counting process @xmath25 jumps given past treatment- and covariate history is also the rate at which @xmath25 jumps given past treatment- and covariate history and @xmath41 . that is , the following : [ intconf ] the intensity process @xmath27 of @xmath25 with respect to @xmath29 is also an intensity process of @xmath25 with respect to @xmath42 . this can be interpreted as conditional independence ( given @xmath23 ) of treatment decisions at time @xmath7 and @xmath43 . this assumption is stronger than just conditional independence of treatment decisions at @xmath7 and @xmath21 as assumed in robins ( @xcite ) . however , also @xmath44 for @xmath45 is an indication of the patient s prognosis upon which treatment decisions at time @xmath7 ( @xmath46 ) should not depend . assumption [ intconf ] allows us to use the usual counting processes framework . under assumption [ intconf ] , @xmath47}\lambda(s)\ , ds$ ] is a martingale also with respect to @xmath42 and its usual augmentation @xmath48 . this formalization of the assumption of no unmeasured confounding in terms of compensators with respect to the filtration @xmath42 is a novel feature of this paper relative to the previous literature on structural nested models . robins et al . ( @xcite ) , @xcite and @xcite use a cox model for initiation and/or changes in treatment . however , none of them formalized the assumption of no unmeasured confounders in terms of compensators with respect to @xmath42 . as a consequence , they could not use the extensive theory on counting process martingales to show the asymptotics of their estimators , which then remained without proof . structural nested models in continuous time model distributional relations between @xmath21 and @xmath49 , for @xmath50 small , through a so - called infinitesimal shift - function @xmath51 . write @xmath52 for the cumulative distribution function and @xmath53 for its generalized inverse @xmath54 then the infinitesimal shift - function @xmath51 is defined as @xmath55 the ( right - hand ) derivative of the quantile quantile transform which moves quantiles of the distribution of @xmath21 to quantiles of the distribution of @xmath49 ( @xmath56 ) , given the covariate- and treatment history until time @xmath7 , @xmath15 . notice that for differentiability of @xmath57 with respect to @xmath58 we need that a short duration of treatment has only a small effect on the distribution of the outcome of interest ( see page 3 , second paragraph ) , since @xmath59 must be equal to @xmath60 . [ daexa ] robins , blevins , ritter and wulfsohn ( @xcite ) describe an aids clinical trial to study the effect of azt treatment on survival in hiv - infected subjects . embedded within this trial was an essentially uncontrolled observational study of the effect of prophylaxis therapy for pcp on survival . pcp , pneumocystis carinii pneumonia , is an opportunistic infection that afflicts aids patients . the aim of robins et al . ( @xcite ) was to study the effect of this prophylaxis therapy on survival . thus , the outcome of interest @xmath1 is the survival time , and the treatment under study is prophylaxis for pcp . once treatment with prophylaxis for pcp started , it was never stopped . suppose that @xmath61 then ( see section [ cmim ] for details ) , for @xmath62 , withholding treatment from @xmath7 onward leads to ( with @xmath63 meaning `` is distributed as '' ) @xmath64\\[-8pt ] & = & e^{\psi}\cdot \mathit{dur}(t , y)+1\cdot \bigl(y - t-\mathit{dur}(t , y)\bigr)\qquad \mbox{given } \overline{z}_t,\nonumber\end{aligned}\ ] ] with @xmath65 the duration of treatment in the interval @xmath66 . thus , treated residual survival time ( @xmath7 until @xmath1 ) is multiplied by @xmath67 by withholding treatment ; compare this with accelerated failure time models , see , for example , @xcite . this multiplication factor @xmath67 should be interpreted in a distributional way . one of the models studied in robins et al . ( @xcite ) assumes that ( [ dasim ] ) is true even with @xmath63 replaced by @xmath68 ( though only for @xmath69 ) . notice that supposing ( [ dasim ] ) to be true with @xmath63 replaced by @xmath68 would be much stronger . [ daexa2 ] consider the situation from example [ daexa ] again . in another model mentioned in robins , blevins , ritter and wulfsohn ( @xcite ) the factor with which treated residual survival time is multiplied when treatment is withheld can depend on the azt treatment the patient received and whether or not the patient had a history of pcp prior to start of pcp prophylaxis . since this was a clinical trial for azt treatment , the azt treatment was described by a single variable @xmath70 indicating the treatment arm the patient was randomized to ( @xmath70 is @xmath24 or @xmath39 ) . whether or not the patient had a pcp history prior to start of prophylaxis is described by an indicator variable @xmath71 . @xmath71 equals @xmath39 if the patient had pcp before or at @xmath7 _ and _ before prophylaxis treatment started ; otherwise @xmath71 equals @xmath24 . if @xmath72 then ( see section [ cmim ] for details ) withholding prophylaxis treatment from @xmath7 onward leads to @xmath73 for @xmath62 . [ prior ] following @xcite , again consider survival as the outcome of interest . suppose that it is known that treatment received at time @xmath7 only affects survival for patients destined to die by time @xmath74 if they would receive no further treatment . an example would be a setting in which failure is death from an infectious disease , the treatment is a preventive antibiotic treatment which is of no benefit unless the subject is already infected and , if death occurs , it always does within five weeks from the time of initial unrecorded subclinical infection . in that case , as remarked in @xcite , the natural restriction on @xmath51 is that @xmath75 more biostatistical examples of models for @xmath51 can be found in , for example , @xcite , witteman et al . ( @xcite ) , @xcite and keiding et al . ( @xcite ) . @xmath76 can be interpreted as the infinitesimal effect on the outcome @xmath1 of the treatment actually given in the time - interval @xmath77 ( relative to baseline treatment ) . to be more precise , from the definition of @xmath51 we have @xmath78 in figure [ dfig ] this is sketched . @xmath79 in the picture is the @xmath80th quantile of the distribution of @xmath21 given @xmath15 . for @xmath50 , the @xmath80th quantile of the distribution of @xmath49 given @xmath15 is @xmath81 . thus , to shift from quantiles of the distribution of @xmath21 to the distribution of @xmath49 given @xmath15 ( @xmath50 ) is approximately the same as to just add @xmath82 to those quantiles . for example , if @xmath83 lies to the right of @xmath84 for @xmath50 , then treatment between @xmath7 and @xmath85 increases the outcome ( in distribution ) , and @xmath86 is greater than @xmath24 . . ] consider again this interpretation of @xmath51 as the infinitesimal effect of treatment given in @xmath77 . if the outcome of interest is survival , then @xmath76 should be zero if @xmath15 indicates the patient is dead at time @xmath7 . indeed , in that case @xmath83 and @xmath84 should be almost surely the same for every @xmath56 , since withholding treatment after death does not change the survival time . thus , @xmath87 is constant in @xmath58 for @xmath56 and , therefore , @xmath88 . however , this reasoning is not precise because of the complication of null sets . we will therefore just formally define @xmath76 to be zero if the outcome of interest is survival and @xmath15 indicates the patient is dead at time @xmath7 . it can be shown that @xmath89 if treatment does not affect the outcome of interest , as was conjectured in @xcite . to be more precise , @xcite shows that , for example , @xmath89 if and only if , for every @xmath50 and @xmath7 , @xmath49 has the same distribution as @xmath21 given @xmath15 . that is , @xmath89 if and only if `` at any time @xmath7 , whatever patient characteristics are selected at that time ( @xmath15 ) , stopping ` treatment as given ' at some fixed time after @xmath7 would not change the distribution of the outcome in patients with these patient characteristics . '' in the rest of this article @xmath90 will always indicate a correctly specified parametric model for @xmath51 , with @xmath89 if @xmath91 . define @xmath92 as the continuous solution to the differential equation @xmath93 with final condition @xmath94 , the observed outcome ( see figure [ dtfig ] ) . then @xmath92 mimics @xmath21 in the sense that it has the same distribution as @xmath21 given @xmath15 . this rather surprising result was conjectured in @xcite and proved in lok ( @xcite ) . to prove this result , we need the following consistency assumption . to the differential equation @xmath95 with final condition @xmath94 in case the outcome is survival time . ] [ inst ] @xmath96 has the same distribution as @xmath1 given @xmath97 . notice that since by assumption no treatment was given after time @xmath3 and since treatment is right - continuous , there is no difference in treatment between @xmath96 and @xmath1 . we suppose that this assumption holds , and we also suppose that a short duration of treatment has only a small effect on the distribution of the outcome of interest ( @xmath98 ) . under these assumptions and regularity conditions only , lok ( @xcite ) proved that indeed equation ( [ xdef ] ) has a unique solution for every @xmath99 , and that this solution @xmath92 mimics @xmath21 in the sense that @xmath92 has the same distribution as @xmath21 given @xmath15 ( see appendix [ xtap ] ) . throughout this article we will assume that this result holds true . [ daexai ] survival of aids patients ( continuation of example ) . suppose that @xmath100 then @xmath101 if @xmath102 , and @xmath103 for @xmath104 . suppose now that one has a correctly specified parametric model for the infinitesimal shift - function @xmath51 , @xmath90 . then one can calculate `` @xmath105 , '' the solution to @xmath106 with final condition @xmath107 . for the true @xmath108 , @xmath105 has the same distribution as @xmath21 , the outcome with treatment stopped at @xmath7 , even given all patient - information at time @xmath7 , @xmath15 . so instead of the unobservable @xmath21 s , we have the observable @xmath109 s which for the true @xmath108 mimic the @xmath21 s . although we do not know the true @xmath108 , this result turns out to be very useful , both for estimating @xmath108 ( sections [ ees ] and [ asnorm ] ) and for testing [ section [ test ] ; notice that when testing whether treatment affects the outcome ( i.e. , whether @xmath89 ) , @xmath110 can simply be calculated from the data ( @xmath111 ) under the null hypothesis of no treatment effect ] . in this appendix we present conditions under which @xmath92 mimics @xmath21 in the sense that it has the same distribution as @xmath21 given @xmath15 . this result is used heavily in this article . for the proofs , which are lengthy and use discretization , we refer to lok ( @xcite ) . section [ xtapt ] deals with survival outcomes , section [ xtapy ] with other outcomes . survival outcomes require a different set of assumptions , as will become clear below . the conditions here are somewhat more restrictive than the ones in lok ( @xcite ) , but they are simpler . this section contains a sufficient set of regularity conditions to have existence and uniqueness of a solution @xmath92 to ( [ xdef ] ) , @xmath507 with final condition @xmath94 , the observed outcome ( see figure [ dtfig ] ) . furthermore , together with assumption [ inst ] ( consistency ) , they imply that @xmath92 has the same distribution as @xmath21 given @xmath15 . the regularity conditions below should be read as the following : there exist conditional distribution functions @xmath83 such that all these assumptions are satisfied . they can be relaxed to @xmath58 in a neighborhood of @xmath24 , if this neighborhood does not depend on @xmath16 . we only consider @xmath56 , so the derivative with respect to @xmath58 at @xmath508 is always the right - hand derivative . [ sup ] ( _ support _ ) . there exist finite numbers @xmath509 and @xmath510 such that all @xmath83 have the same bounded support @xmath511 $ ] . all @xmath512 have a continuous nonzero density @xmath513 on @xmath514 $ ] . there exists an @xmath216 such that @xmath515 for all @xmath514 $ ] , @xmath99 and @xmath10 $ ] . ( _ smoothness _ ) . for every @xmath99 , @xmath516 is differentiable with respect to @xmath7 , @xmath79 and @xmath58 with continuous derivatives on @xmath511\times[t_1,t_2)\times { \mathbb r}$ ] if @xmath11 does not jump in @xmath239 , with a continuous extension to @xmath511\times[t_1,t_2]\times [ 0,\infty)$ ] . the derivatives of @xmath512 with respect to @xmath79 and @xmath58 are bounded by constants @xmath315 and @xmath317 , respectively . @xmath517 and @xmath518 have derivatives with respect to @xmath79 which are bounded by constants @xmath519 and @xmath520 , respectively . the support conditions may be restrictive for certain applications . nevertheless , most real - life situations can be approximated this way , since @xmath509 and @xmath510 are unrestricted and @xmath216 is unrestricted , too . although the support conditions may well be stronger than necessary , they simplify the analysis considerably and , for that reason , they are adopted here . the smoothness conditions allow for nonsmoothness where the covariate- and treatment process @xmath11 jumps . this is important , since if the covariate- and treatment process @xmath11 jumps , this can lead to a different prognosis for the patient and thus to nonsmoothness of the functions concerned . [ thm1 ] suppose that regularity condition is satisfied . then @xmath76 exists . furthermore , for every @xmath99 , there exists exactly one continuous solution @xmath92 to @xmath95 with final condition @xmath94 . if also assumption ( consistency ) is satisfied and there are no more than finitely many times @xmath7 for which the probability that the covariate- and treatment process jumps at @xmath7 is greater than @xmath24 , then this @xmath92 has the same distribution as @xmath21 given @xmath15 for all @xmath10 $ ] . for a proof we refer to lok ( @xcite ) . this section contains a sufficient set of regularity conditions to have existence and uniqueness of a solution @xmath92 to equation ( [ xdef ] ) , @xmath95 with final condition @xmath94 , the observed outcome ( see figure [ dtfig ] ) . furthermore , together with assumption [ inst ] ( consistency ) and assumptions [ ttcons ] and [ ttinst ] below , they imply that @xmath92 has the same distribution as @xmath21 given @xmath15 . the conditions here are natural conditions if the outcome of interest @xmath1 is a survival time . as compared to section [ xtapy ] , we make two extra assumptions . the first is a consistency assumption , stating that stopping treatment after death does not change the survival time . the second assumption states that there is no instantaneous effect of treatment at the time the patient died [ notice that the difference between @xmath521 , the outcome with treatment stopped at the survival time @xmath1 , and @xmath1 is in treatment _ at _ @xmath1 ] . [ ttcons ] @xmath523 on @xmath524 . [ ttinst ] @xmath525 under these assumptions , treatment in the future does not cause or prevent death at present or before : [ conslem ] under assumptions and : for all @xmath56 : @xmath526 on @xmath527 . for all @xmath528 with @xmath529 and @xmath530 . for a proof we refer to @xcite . if the outcome is survival , the support condition in assumption [ xtapy ] , saying that all @xmath83 have the same bounded support @xmath511 $ ] , will not hold . the reason for this is as follows . @xmath15 includes the covariate - measurements and treatment until time @xmath7 . if covariates and treatment were measured at time @xmath7 , it can not be avoided to include in @xmath15 whether or not a patient was alive at time @xmath7 . given that a patient is dead at time @xmath7 and given his or her survival time , the distribution of this survival time can not have the fixed support @xmath511 $ ] , which is independent of @xmath7 . also , given that a patient is alive at time @xmath7 , this is hardly ever the case ; one often expects that @xmath7 is the left limit of the support . thus , in case the outcome is survival , the support condition for theorem [ thm1 ] has to be slightly changed . [ tsup ] there exists a finite number @xmath531 such that : for every @xmath99 and @xmath7 with @xmath102 , all @xmath83 for @xmath56 have support @xmath532 $ ] . for every @xmath99 and @xmath7 with @xmath102 , all @xmath512 for @xmath56 have a continuous nonzero density @xmath513 on @xmath533 $ ] . there exists a number @xmath216 such that , for all @xmath99 and @xmath7 with @xmath102 , @xmath534 for @xmath535 $ ] . next we look at the differentiability conditions in assumption [ sup ] . it does not seem reasonable to assume that @xmath512 is continuously differentiable with respect to @xmath58 and @xmath79 on @xmath536 $ ] since , for @xmath529 , @xmath537 [ lemma [ conslem](b ) ] . therefore , the derivative of @xmath512 with respect to @xmath58 is likely not to exist at @xmath538 ( and is equal to zero for @xmath539 ) . also , the derivative of @xmath512 with respect to @xmath79 may not exist at @xmath538 , because of the different treatment before and after @xmath85 . for survival outcomes , we replace the smoothness conditions of assumption [ sup ] by the following : [ tsmooth ] for every @xmath99 : if @xmath11 does not jump in @xmath239 and @xmath540 , the restriction of @xmath541 to @xmath542\times[t_1,t_2)\times { \mathbb r}_{\geq 0}\dvtx y\geq t+h\}$ ] is @xmath543 in @xmath528 . the derivatives of @xmath512 with respect to @xmath79 and @xmath58 are bounded by constants @xmath315 and @xmath317 , respectively , for @xmath533 $ ] . @xmath517 and @xmath518 have derivatives with respect to @xmath79 which are bounded by constants @xmath519 and @xmath520 , respectively , for @xmath533 $ ] . the smoothness condition above concentrates on @xmath544 . for @xmath545 we can choose @xmath537 because of lemma [ conslem](b ) . because of assumption [ inst ] ( consistency ) , @xmath546 has the same support as @xmath547 , so @xmath546 has support @xmath532 $ ] if @xmath102 [ assumption [ tsup](a ) ] . assume the following [ ytsup ] for all @xmath99 and @xmath7 with @xmath102 , @xmath548 is continuous and strictly increasing on its support @xmath532 $ ] . [ tthm1 ] suppose that regularity conditions and are satisfied . then @xmath76 exists . furthermore , for every @xmath99 , there exists exactly one continuous solution @xmath92 to @xmath95 with final condition @xmath94 . if also assumptions and ( consistency and no instantaneous treatment effect at time of death ) are satisfied , then this @xmath92 has the same distribution as @xmath21 given @xmath15 for all @xmath10 $ ] . for a proof we refer to lok ( @xcite ) . previous applications of structural nested models [ see , e.g. , robins et al . ( @xcite ) , @xcite ( @xcite ) , witteman et al . ( @xcite ) and keiding et al . ( @xcite ) ] have assumed the so - called local rank preservation condition . local rank preservation states that @xmath21 is a local solution to ( [ xdef ] ) . however , if @xmath21 is locally a solution to ( [ xdef ] ) , it is usually also globally a solution to ( [ xdef ] ) ; see , for example , theorem [ isun ] in the . hence , if one knew the parameter @xmath108 , every @xmath21 would be a deterministic function of the observed data . deterministic dependence of counterfactuals on the observed data is a very strong condition , which , though untestable , is generally considered implausible . the previous literature [ see , e.g. , @xcite ( @xcite ) , robins et al . ( @xcite ) , @xcite ( @xcite ) and keiding et al . ( @xcite ) ] acknowledged this problem , and conjectured that the assumption of local rank preservation could be relaxed in continuous time [ since it is known that the assumption of local rank preservation can be relaxed for structural nested models in discrete time ; this was pointed out by @xcite ( @xcite ) , and lok et al . ( @xcite ) provided a proof ] . see @xcite for a more elaborate discussion . the following example describes a setting where the assumption of local rank preservation is implausible . in the situation of example [ daexa ] , consider the following thought experiment . suppose that two patients had the same covariate history until time @xmath7 , and both received the same constant treatment to prevent pcp until time @xmath7 ( equal @xmath15 ) . suppose , furthermore , that both patients received no treatment after time @xmath7 , that they did not have pcp before time @xmath7 and both died at the same time @xmath112 ( for both , @xmath113 ) . possibly , the first patient would have had pcp at some time @xmath45 and would have died from it before @xmath114 in case he or she would not have been treated . possibly , the other patient would not have had pcp in case he or she would not have been treated , and would have died at the same time @xmath112 as without treatment . thus , it is easy to imagine that these patients would have had different outcomes under no treatment ( different @xmath115 ) . however , the assumption of local rank preservation excludes this possibility . local rank preservation is a very strong condition , for which structural nested models have previously been attacked . in fact , this article shows that the assumption of local rank preservation is not needed for structural nested models . however , proofs would be much easier under rank preservation ; for details , see the remarks before the proofs of theorems [ thec ] and [ mgc ] . see also @xcite for a more informal reasoning . to estimate the infinitesimal shift - function @xmath51 , @xcite proposes to use a ( semi-)parametric model to predict future treatment ( @xmath25 in our case ) on the basis of past treatment- and covariate history @xmath23 . this may seem odd , since prediction of treatment is not what we are interested in . however , we will show that such a model to predict treatment changes can indeed be a tool to get unbiased estimating equations for the parameter @xmath108 in the model for @xmath51 . moreover , often doctors may have a better understanding , at least qualitatively , about how decisions about treatment were made than about the effect of the treatment . in what follows we will assume that @xmath116 is a correctly specified parametric model for the intensity @xmath30 of @xmath25 . recall from section [ nounsec ] that , under no unmeasured confounding ( assumption [ intconf ] ) , @xmath21 does not contain information about treatment changes given past treatment- and covariate history @xmath23 . since @xmath92 has the same distribution as @xmath21 given @xmath15 , one could expect that also @xmath92 does not contain information about treatment changes given @xmath23 . unfortunately , this reasoning is not precise : we have to somehow deal with null sets since the probability that treatment changes at @xmath7 given past covariate- and treatment history is often equal to @xmath24 for each @xmath7 . in section [ mgs ] we will show how this can be dealt with . in the current section we present a class of unbiased estimating equations for @xmath117 and @xmath108 . these will be used for the proof in the next section , but they are also of interest in their own . in section [ mgs ] we will see that these estimating equations are in fact martingales , for the true parameters @xmath118 and @xmath119 . recall from section [ nounsec ] that , under no unmeasured confounding , we have the martingale @xmath47}\lambda(s)\,ds$ ] with respect to the filtration @xmath42 and its usual augmentation . from this martingale we can construct a whole family of martingales . if @xmath120 is a @xmath48-predictable process , then , under regularity conditions , @xmath121 is a martingale with respect to @xmath48 . for a more formal statement , we first make sure that @xmath120 is predictable . we put the following restriction on the functions @xmath122 we consider here : [ hreg ] when in this section we consider functions @xmath122 from @xmath123 , we assume that they are measurable and satisfy the following : @xmath122 is bounded by a constant which does not depend on @xmath7 and @xmath16 , for all @xmath124 $ ] , @xmath125 and @xmath99 , @xmath126 when @xmath127 and @xmath128 . for such @xmath122 , @xmath120 is a @xmath48-predictable process : [ hpred ] suppose that @xmath40 is cadlag assumption . then @xmath129 is a @xmath48-predictable process for any @xmath122 satisfying restriction . proof @xmath120 is adapted . it is also left - continuous : because @xmath40 is cadlag , @xmath130 exists . thus , we come to the following lemma : [ lemec ] under assumptions bounded intensity process @xmath40 cadlag and [ intconf ] ( no unmeasured confounding ) , @xmath131 is a martingale on @xmath2 $ ] with respect to @xmath48 for all @xmath122 satisfying restriction . proof @xmath132 is a martingale on @xmath2 $ ] with respect to @xmath48 , because of assumption [ intconf ] . it is of integrable variation . because of lemma [ hpred ] , @xmath133 is a @xmath48-predictable process . it is also bounded [ restriction [ hreg](a ) ] . hence,@xmath134 is an integral of a bounded predictable process with respect to a martingale of integrable variation , and , therefore , a @xmath48-martingale . to construct unbiased estimating equations for @xmath135 , we need to assume that the probability that @xmath136 and @xmath40 jump at the same time is zero . this assumption is a formalization of the assumption of no instantaneous treatment effect as proposed in @xcite , which can be seen as follows . given @xmath23 and @xmath137 , @xmath25 jumps at @xmath7 with rate @xmath27 ( assumption [ intconf ] , no unmeasured confounding ) . @xmath40 is a cadlag process ( assumption [ ycadlag ] ) , which thus for every @xmath99 jumps at most countably many times on the finite time interval @xmath2 $ ] . therefore , if @xmath40 and @xmath25 would jump at the same time with positive probability , this would imply a dependence of these jumps ; the obvious interpretation of this dependence would be that a change of treatment instantaneously affects the outcome of interest . [ jump ] the probability that there exists a @xmath7 such that @xmath136 and @xmath40 both jump at time @xmath7 is @xmath24 . notice that this excludes estimation of the effect of point exposures . for example , if treatment is surgery or another point exposure given at some time @xmath7 , the outcome under `` treatment stopped at time @xmath7 '' will typically jump at time @xmath7 if treatment affects the outcome of interest , at the same time as the treatment itself . however , this assumption does not exclude the possibility that the outcome differs depending on whether a patient is treated or not at a certain point in time . for example , @xmath138 and @xmath139 may be different when a virus is contacted at time @xmath7 . the model in this article can accommodate differences between @xmath138 and @xmath139 , as long as the probability that the observed treatment changes is @xmath24 at that precise time . or , in more generality , as long as the probability that @xmath25 jumps at the same time is @xmath24 . this was previously noticed in @xcite , section 8 . the estimating procedures in this article do not deal with instantaneous treatment effects . suppose that the above conditions hold and that @xmath140 for @xmath10 $ ] ( see section [ cmim ] ) . then if @xmath141 and @xmath116 are correctly specified ( parametric ) models for @xmath51 and @xmath30 , respectively , each choice of @xmath122 satisfying restriction [ hreg ] leads to an unbiased estimating equation for both the parameter of interest @xmath108 and the ( nuisance ) parameter @xmath117 : [ thec ] suppose that assumptions bounded intensity process@xmath40 cadlag ( no unmeasured confounding ) and ( no instantaneous treatment effect ) are satisfied . suppose also that , for every @xmath10 $ ] , @xmath92 has the same distribution as @xmath21 given @xmath15 . then @xmath142 for each @xmath122 satisfying restriction . thus , if @xmath90 and @xmath116 are correctly specified parametric models for @xmath51 and @xmath30 , respectively , @xmath143 with @xmath144 the empirical measure @xmath145 , is an unbiased estimating equation for @xmath146 , for each @xmath122 satisfying restriction . @xmath122 here is allowed to depend on @xmath108 and @xmath117 , as long as it satisfies restriction for @xmath146 . as before , @xmath109 here is the continuous solution of ( [ xpsi ] ) , @xmath147 with boundary condition @xmath107 . moreover , as before , we assume that for all @xmath90 we have existence and uniqueness of such solutions ; theorem [ isun ] in the appendix provides sufficient conditions for that . under local rank preservation ( see section [ loc ] ) , @xmath148 for each @xmath7 . in that case theorem [ thec ] follows immediately from lemma [ lemec ] . however , as argued in section [ loc ] , local rank preservation is generally considered implausible . proof of theorem [ thec ] we have to show that @xmath149 has expectation zero for all @xmath122 satisfying restriction [ hreg ] . to do that , we prove that it has the same expectation as @xmath150 which has expectation zero because of lemma [ lemec ] . we will first show that the terms with @xmath151 have the same expectation , that is , @xmath152 after that we show that the terms with @xmath153 have the same expectation , that is , @xmath154 as we will see below , ( [ firste ] ) and ( [ seconde ] ) have to be proved separately , since we do not have or expect that @xmath155 for @xmath45 ; we only have this for @xmath156 . therefore , the approximations below have to be chosen carefully . at first we prove ( [ firste ] ) , by approximating these sums and showing that the approximations have the same expectation . next we show that the approximations converge and that ( [ firste ] ) follows with lebesgue s dominated convergence theorem . define @xmath157 , @xmath158 , etc . , the jump times of the counting process @xmath25 in the interval @xmath2 $ ] . they are measurable [ e.g. , because of @xcite ( @xcite ) , lemma 74.4 ] . note that the number of jumps in @xmath2 $ ] is almost surely finite because @xmath25 is integrable ( it has a bounded intensity process ) . in the following read @xmath159 if there is no @xmath160th jump of @xmath25 in the interval @xmath2 $ ] . next split up the interval @xmath2 $ ] in intervals of equal length : for @xmath161 fixed , put @xmath162 , @xmath163 . fix @xmath164 for the moment . the right - hand side of equation ( [ firste ] ) is harder to approximate than the left - hand side , both because @xmath21 does not need to be continuous in @xmath7 while @xmath92 does and because knowing @xmath21 and @xmath15 does not imply knowing @xmath44 for @xmath45 and we do not have or expect @xmath155 for @xmath45 . the approximations we choose are @xmath165\\[-8pt ] \qquad & \approx & \sum_{j=1}^\infty\sum_{k=0}^{k-1 } 1_{(\tau_k,\tau_{k+1}]}(t_j ) h_{\tau_{k}}\bigl(y^{(\tau_{k+1})},\overline{z}_{\tau_{k}-}\bigr)\nonumber\end{aligned}\ ] ] and @xmath166\\[-8pt ] \qquad & \approx & \sum_{j=1}^\infty\sum_{k=0}^{k-1 } 1_{(\tau_k,\tau_{k+1}]}(t_j ) h_{\tau_{k}}\bigl(x(\tau_{k+1}),\overline{z}_{\tau_{k}-}\bigr).\nonumber\end{aligned}\ ] ] to show that these approximations have the same expectation , we use that @xmath167 . therefore , also @xmath168}(t_j ) h_{\tau_{k}}(x(\tau_{k+1}),\overline{z}_{\tau_{k-}})\sim 1_{(\tau_k,\tau_{k+1}]}(t_j ) h_{\tau_{k}}\bigl(y^{(\tau_{k+1})},\overline{z}_{\tau_{k-}}\bigr)\ ] ] [ notice that @xmath169}(t_j)$ ] is a function of @xmath170 . hence , the expectation of each of the terms on the right - hand side of ( [ apy ] ) is equal to the expectation of the corresponding term on the right - hand side of ( [ apx ] ) . since @xmath122 is bounded [ restriction [ hreg](a ) ] and the expected number of jump times @xmath171 is finite ( @xmath25 is integrable ) , this implies that the expectation of the right hand - side of equation ( [ apy ] ) is equal to the expectation of the right - hand side of equation ( [ apx ] ) . equation ( [ firste ] ) follows if the expectation of the approximations in ( [ apy ] ) and ( [ apx ] ) converges to the right - hand side and left - hand side of equation ( [ firste ] ) , respectively . this convergence is harder to show for ( [ apy ] ) than for ( [ apx ] ) , since @xmath40 may jump [ while @xmath172 does not , by construction ] . fix @xmath160 for a moment . define @xmath173 and @xmath174 as the grid points such that @xmath175 $ ] . as @xmath176 , @xmath177 , so that since @xmath178 is cadlag , @xmath179 . moreover , as @xmath180 , @xmath181 , so that because of restriction [ hreg ] on @xmath58 , @xmath182 . combining this for all @xmath160 leads to @xmath183}(t_j ) h_{\tau_{k}}\bigl(y^{(\tau_{k+1})},\overline{z}_{\tau_{k-}}\bigr ) \rightarrow \sum_{j=1}^\infty h_{t_j}\bigl(y^{(t_{j})},\overline{z}_{t_j-}\bigr)\ ] ] as @xmath184 for every @xmath31 for which the number of jumps of @xmath25 is finite , so for almost every @xmath99 . @xmath122 is bounded [ restriction [ hreg](a ) ] and the left - hand side is bounded by the number of jumps of @xmath25 times this bound . the expectation of that is finite because @xmath25 is integrable . thus , lebesgue s dominated convergence theorem can be applied , and @xmath185}(t_j ) h_{\tau_{k}}\bigl(y^{(\tau_{k+1})},\overline{z}_{\tau_{k}-}\bigr)\biggr ) \rightarrow e\biggl(\sum_{j=1}^\infty h_{t_{j}}\bigl(y^{(t_{j})},\overline{z}_{t_j-}\bigr)\biggr)\ ] ] as @xmath184 . because with probability one @xmath40 and @xmath25 do not jump at the same time ( assumption [ jump ] of no instantaneous treatment effect ) , @xmath186 so that we can replace @xmath187 by @xmath188 on the right - hand side of ( [ naarplus ] ) . therefore , indeed , the expectation of the approximation in ( [ apy ] ) converges to the expectation of the left - hand side of ( [ apy ] ) . the same reasoning shows this for ( [ apx ] ) . here less caution is necessary since @xmath92 is continuous in @xmath7 . that concludes the proof of equation ( [ firste ] ) . next we prove ( [ seconde ] ) , also by approximation . here , too , we show that the approximations have the same expectation and that ( [ seconde ] ) follows with lebesgue s dominated convergence theorem . divide the interval @xmath2 $ ] as above . the approximations we choose here are @xmath189 and @xmath190 because @xmath191 and @xmath192 are a measurable function of @xmath193 ( assumption [ int ] , bounded intensity process ) , the expectation of each of the terms in ( [ apy2 ] ) is equal to the expectation of the corresponding term in ( [ apx2 ] ) . thus , the expectations of these approximations are equal . equation ( [ seconde ] ) follows if the expectation of the approximations in ( [ apy2 ] ) and ( [ apx2 ] ) converge to the right - hand side and left - hand side of equation ( [ seconde ] ) , respectively . this convergence is also harder to show for ( [ apy2 ] ) than for ( [ apx2 ] ) because of possible discontinuities of @xmath40 . first notice that as @xmath184 , for @xmath7 fixed , @xmath194}(t ) h_{\tau_k}\bigl(y^{(\tau_k)},\overline{z}_{\tau_k-}\bigr ) \lambda(\tau_k)\rightarrow h_{t}\bigl(y^{(t-)},\overline{z}_{t-}\bigr ) \lambda(t)\ ] ] for every @xmath99 fixed and for every @xmath195 : @xmath40 has limits from the left ( assumption [ ycadlag ] ) , so that as @xmath196 , restriction [ hreg](b ) on @xmath58 can be used , and @xmath30 is continuous from the left [ assumption [ int](b ) ] . taking integrals and applying lebesgue s dominated convergence theorem [ @xmath122 and @xmath30 are bounded because of restriction [ hreg](a ) and assumption [ int](a ) , resp . ] leads to @xmath197 for every @xmath99 . as both @xmath58 and @xmath30 are bounded , lebesgue s dominated convergence theorem guarantees that indeed the expectation of the approximation in ( [ apy2 ] ) converges to the expectation of the left - hand side of ( [ apy2 ] ) . the same reasoning shows this for ( [ apx2 ] ) , which concludes the proof of equation ( [ seconde ] ) and theorem [ thec ] . @xcite shows that if the rest of the conditions in this section are satisfied , assumption [ jump ] ( treatment does not instantaneously affect the outcome of interest ) is a necessary condition for theorem [ thec ] . [ wei ] consider the setting of examples [ daexa ] and [ daexa2 ] and define @xmath198 if prophylaxis treatment started at or before time @xmath7 and @xmath24 otherwise . suppose that initiation of prophylaxis treatment can be correctly modeled with the time - dependent weibull proportional hazards model @xmath199 where @xmath200 equals @xmath39 if the patient had pcp before time @xmath7 and @xmath24 otherwise , and @xmath201 and @xmath202 are greater than zero [ for more about the weibull proportional hazards model and its applications see , e.g. , @xcite ( @xcite ) ] . if the patient died before @xmath7 or prophylaxis treatment already started before , the patient is not `` at risk '' for initiation of treatment and , thus , @xmath30 equals @xmath24 . then the ( partial ) score equations for estimation of @xmath203 are @xmath204 such estimating equations can also be written down for the model including @xmath205 , @xmath206 @xcite proposes to estimate the parameters in a model like this by choosing those parameters @xmath207 which maximize the likelihood when @xmath208 is considered fixed and known , and for which @xmath209 : for the true @xmath108 , @xmath210 does not contribute to the model for treatment changes ( under no unmeasured confounding ) . to make the connection with the estimators in the current article , notice that this leads to the same estimators as the ones that solve the estimating equations arising from the likelihood when @xmath208 is considered fixed and known , with @xmath211 put to zero . more precise , since we know that the true @xmath211 is equal to @xmath24 , we put @xmath211 equal to @xmath24 and get the estimating equations @xmath212 for the parameter @xmath108 ( and thus also for @xmath51 ) and the ( nuisance ) parameters @xmath203 . these estimating equations are of the form of theorem [ thec ] , @xmath213 but the function @xmath122 here is not bounded and @xmath30 need not be bounded ( if @xmath214 ) , so unbiasedness does not follow immediately from theorem [ thec ] . however , we could restrict the interval @xmath2 $ ] to @xmath215 $ ] for @xmath216 ( to assure that @xmath30 is bounded ) and @xmath217 can be approximated by the bounded functions @xmath218 ( @xmath219 ) ( to make @xmath122 bounded ) , which all lead to unbiased estimating equations because of theorem [ thec ] . the above estimating equations are then also unbiased because of lebesgue s dominated convergence theorem [ the dominating function is integrable since @xmath220 which is finite since @xmath221 . under the model for @xmath51 of example [ daexa ] , @xmath222 if the patient did not die before time @xmath7 . in that case these are five unbiased estimating equations for five unknown parameters . if the parameter @xmath108 is of dimension greater than @xmath39 , more unbiased estimating equations can be constructed by adding more terms of the form @xmath223 . we show that , under no unmeasured confounding , just as @xmath21 , @xmath92 does not predict treatment changes , given past treatment- and covariate history @xmath23 . we could hope for that since @xmath224 given @xmath15 ( see section [ cmim ] ) . the formal statement is ( compare with assumption [ intconf ] , no unmeasured confounding ) the following : the intensity process @xmath27 of @xmath25 with respect to @xmath29 is also the intensity process of @xmath25 with respect to @xmath225 . then @xmath226 is also a martingale with respect to @xmath225 . that will be useful later when we study the behavior of estimators @xmath227 and @xmath228 which are constructed with estimating equations of the form of theorem [ thec ] , @xmath229 . for example , we can use the fact that usually @xmath230}h(s)\,dm(s)$ ] is a martingale if @xmath38 is a martingale and @xmath231 a predictable process ; a sufficient condition for this is that @xmath232 . before going on , we first clarify why @xmath233 is indeed a filtration . for @xmath45 , @xmath234 is a deterministic ( though unknown ) function of @xmath235 ( i.e. , if solutions to the differential equation with @xmath51 are unique ; see , e.g. , theorem [ isun ] in the ) . similarly , for @xmath45 , @xmath92 is a deterministic function of @xmath236 . in the rest of this article we will assume that these functions are measurable functions on @xmath237 ( sufficient conditions for that are that the infinitesimal shift - function @xmath51 satisfies regularity assumption [ dreg ] below and that for each @xmath99 , @xmath11 only jumps finitely many times ; see appendix [ smtb ] , lemma [ mtb ] ) . thus , @xmath238 we will use the filtration @xmath233 below , keeping in mind that it is indeed a filtration and satisfies equation ( [ fil ] ) . in the rest of this section we assume that the infinitesimal shift - function @xmath51 satisfies the following regularity condition : [ dreg ] ( _ continuity between the jump times of _ @xmath11 ) . if @xmath11 does not jump in @xmath239 , then @xmath76 is continuous in @xmath240 on @xmath241 and can be continuously extended to @xmath242 $ ] . ( _ boundedness _ ) . for each @xmath99 , there exists a constant @xmath243 such that @xmath244 for all @xmath10 $ ] and all @xmath79 . ( _ lipschitz continuity _ ) . for each @xmath99 , there exist constants @xmath245 and @xmath246 with @xmath247 for all @xmath10 $ ] and all @xmath248 and @xmath249 if @xmath45 and @xmath11 does not jump in @xmath250 $ ] . most regularity conditions on @xmath51 here are satisfied for the @xmath51 s from appendix [ xtap ] [ see also lok ( @xcite ) ] . only the second lipschitz condition is extra . the lipschitz conditions are satisfied , for example , if , in between the jump times of @xmath11 , @xmath51 is continuously differentiable with respect to @xmath79 and @xmath7 with derivatives which are bounded for every fixed @xmath99 . the next theorem states that @xmath38 is indeed also a martingale with respect to @xmath225 : [ mgc ] suppose that the conditions of theorem hold : assumptions ( bounded intensity process ) , ( @xmath40 cadlag ) , ( no unmeasured confounding ) , ( no instantaneous treatment effect ) and for every @xmath10 $ ] , @xmath92 has the same distribution as @xmath21 given @xmath15 . suppose , furthermore , that for each @xmath99 , @xmath11 jumps at most finitely many times , and that @xmath51 satisfies regularity condition [ dreg ] . then the intensity process @xmath27 of @xmath25 with respect to @xmath29 is also the intensity process of @xmath25 with respect to the filtration @xmath225 . recall that in section [ cmim ] we already mentioned that , under regularity conditions , @xmath92 mimics @xmath21 in the sense that it has the same distribution as @xmath21 given @xmath15 . under local rank preservation ( see section [ loc ] ) , @xmath148 for each @xmath7 . in that case theorem [ mgc ] would be the same as the assumption of no unmeasured confounding [ intconf ] . however , as argued in section [ loc ] , local rank preservation is generally considered implausible . proof of theorem [ mgc ] because of assumption [ int ] , @xmath251 is predictable with respect to @xmath29 , so then it is also predictable with respect to the larger filtration @xmath225 . we still need to prove that @xmath38 is a martingale with respect to @xmath225 . since a cadlag martingale with respect to some filtration is also a martingale with respect to its usual augmentation [ see @xcite ( @xcite ) , lemma 67.10 ] , it suffices to prove that @xmath38 is a martingale with respect to @xmath233 . thus we need to prove that , for @xmath252 , @xmath253 = 0.\ ] ] this is not immediate , since we do not have or expect that @xmath254 if @xmath255 . by the definition of conditional expectation , the above is the same as @xmath256 for all @xmath257 . because of theorem 34.1 in @xcite , it is sufficient to consider @xmath258 s forming a @xmath259-system generating @xmath260 . with @xmath261 the @xmath18-algebra on @xmath262 , @xmath263 is such a @xmath259-system : it is closed under the formation of finite intersections and generates @xmath260 . therefore , we only consider @xmath258 s of this form . we prove ( [ enul ] ) for any @xmath264 . let @xmath265 be any approximation of @xmath266 which is continuous for every fixed @xmath267 , with @xmath268 for every @xmath269 as @xmath270 and @xmath271 for all @xmath269 and @xmath267 . then @xmath272}\ , dm(t)\biggr)\\ & & \qquad = e\int 1_{(t_1,t_2]}(t)\ , 1_a(\overline{z}_{t_1 } ) 1_{(x_1,x_2)}(x(t_1))\ , dm(t)\\ & & \qquad = e \int 1_{(t_1,t_2]}(t)\ , 1_a(\overline{z}_{t_1 } ) \lim_{n\rightarrow \infty } 1^{(n)}_{(x_1,x_2)}(x(t_1))\ , dm(t)\\ & & \qquad = e\lim_{n\rightarrow \infty } \int 1_{(t_1,t_2]}(t)\ , 1_a(\overline{z}_{t_1 } ) 1^{(n)}_{(x_1,x_2)}(x(t_1 ) ) \bigl(dn(t)-\lambda(t)\,dt\bigr)\\ & & \qquad=\lim_{n\rightarrow \infty } e \int 1_{(t_1,t_2]}(t)\ , 1_a(\overline{z}_{t_1 } ) 1^{(n)}_{(x_1,x_2)}(x(t_1))\,dm(t).\end{aligned}\ ] ] the last two equalities follow from lebesgue s dominated convergence theorem [ the prior to last equality since , for @xmath273 fixed , the integral is bounded since @xmath25 is finite and @xmath30 is bounded ; the last equality since the integrals are all bounded by @xmath274 , whose expectation is bounded by @xmath275 times the upper bound of @xmath30 ] . equation ( [ enul ] ) and the result of the theorem would follow from theorem [ thec ] if @xmath277}(t)\ , 1_a(\overline{z}_{t_1 } ) 1^{(n)}_{(x_1,x_2)}(x(t_1 ) ) = h^{(n)}_t(x(t),\overline{z}_{t-})\ ] ] for some @xmath278 from @xmath279 satisfying restriction [ hreg ] for each fixed @xmath267 . in principle , this seems possible , since @xmath280 is a function of @xmath92 and @xmath23 for every @xmath281 . indeed , under the conditions above on @xmath51 and @xmath11 , it is possible to find such an @xmath278 , as follows . write @xmath282 for the solution of the differential equation @xmath283 with ( final or initial , depending on @xmath7 ) condition @xmath284 . existence and uniqueness of @xmath282 on @xmath2 $ ] for every fixed @xmath99 follows from theorem [ isun ] in appendix [ difeqap ] . in this notation , @xmath285}(t)\ , 1_a(\overline{z}_{t_1 } ) 1^{(n)}_{(x_1,x_2)}(x(t_1))\nonumber\\ & & \qquad=1_{(t_1,t_2]}(t)\ , 1_a(\overline{z}_{t_1 } ) 1^{(n)}_{(x_1,x_2)}(x(t_1;t , x(t ) ) ) = h^{(n)}_t(x(t),\overline{z}_{t-})\end{aligned}\ ] ] with @xmath286}(t)\ , 1_a(\overline{z}_{t_1 } ) 1^{(n)}_{(x_1,x_2)}(x(t_1;t , y)).\ ] ] we have to show that ( [ ht ] ) satisfies restriction [ hreg ] . first we show that , for @xmath7 fixed , @xmath287 is measurable . from ( [ ht ] ) we see that this is the case if @xmath288 is measurable , which follows immediately from lemma [ mtb ] . restriction [ hreg](a ) is immediate , since @xmath278 is bounded by 1 . for restriction [ hreg](b ) , we have to prove that , for all @xmath99 , @xmath289 when @xmath127 and @xmath128 . fix @xmath99 . we consider three different kinds of @xmath290 . if @xmath291 and @xmath128 , @xmath292 , so that the convergence follows immediately . if @xmath293 and @xmath128 , eventually @xmath294 , so that the convergence also follows immediately . if @xmath295 $ ] , convergence of the first two factors is immediate . for the last factor , we need differential equation theory . @xmath296 is continuous . thus , to prove that the last factor in equation ( [ ht ] ) converges , it suffices to show that @xmath297 as @xmath128 and @xmath127 . fix @xmath99 . for @xmath7 close enough to @xmath290 , we compare the solution of the differential equation with final condition @xmath79 at @xmath7 with the solution of the differential equation with final condition @xmath298 at @xmath290 ; we look at the value of the solution at the time point @xmath299 before both @xmath7 and @xmath290 . first , notice that because of existence and uniqueness of solutions ( theorem [ isun ] ) , the solution of the differential equation with final condition @xmath79 at @xmath7 takes a unique value @xmath300 at @xmath290 . since @xmath269 is differentiable with respect to its first argument with derivative @xmath51 and @xmath51 is bounded by @xmath243 [ assumption [ dreg](b ) ] , @xmath301 is not far from @xmath79 if @xmath7 is not far from @xmath290 : @xmath302 next , notice that , again because of existence and uniqueness of solutions , the value at @xmath299 of the solution of the differential equation with final condition @xmath79 at @xmath7 is the same as the value at @xmath299 of the solution of the differential equation with final condition @xmath300 at @xmath290 . this observation implies that @xmath303 for the first inequality , we use corollary [ difbc1 ] and assumption [ dreg ] [ notice that possible jumps of @xmath51 at the jump times of @xmath11 do not matter here since one can split up the interval , so if , e.g. , there is just one jump at @xmath304 $ ] , one gets a factor @xmath305 etc . ( a formal proof can be given with induction since , with @xmath99 still fixed , there are only finitely many jumps of @xmath11 ) ] . for the last inequality , we use equation ( [ yyt ] ) . if @xmath127 and @xmath128 , the bound in equation ( [ xafs ] ) converges to @xmath24 for every fixed @xmath99 . thus , indeed , if @xmath127 and @xmath128 , @xmath306 converges to @xmath307 . this finishes the proof . the estimating equations for @xmath308 from section [ ees ] were all of the form @xmath309 . in the current section we choose the dimension of @xmath310 the same as the dimension of @xmath308 . estimating equations of this form are well known . theorem [ consas ] below is an example of asymptotic theory in the setting of this article , with conditions in terms of @xmath58 and the intensity process @xmath30 . notice , however , that these conditions are in fact stronger than necessary . for more theory about these types of estimating equations and less restrictive conditions , see @xcite , chapter 5 . in particular , conditions could be weakened by considering the estimating equations as a whole instead of looking at @xmath58 and @xmath30 separately ( see , e.g. , example [ weilem ] ) . we only consider smooth @xmath311 : [ hregpc ] the functions @xmath312 are measurable and : every component of @xmath313 satisfies restriction [ hreg ] . @xmath314 is bounded by a constant @xmath315 not depending on @xmath117 , @xmath108 , @xmath7 , @xmath79 and @xmath99 . for each @xmath10 $ ] and @xmath99 , @xmath316 is continuous . there exists a neighborhood of @xmath135 such that , for each @xmath10 $ ] and @xmath99 , @xmath314 is continuously differentiable with respect to @xmath117 , @xmath108 and @xmath79 , and these derivatives are all bounded by a constant @xmath317 not depending on @xmath117 , @xmath108 , @xmath7 , @xmath79 and @xmath99 . every component of @xmath318 satisfies restriction [ hreg ] . [ consas ] suppose that assumptions ( bounded intensity process ) , [ @xmath178 cadlag ] , ( no unmeasured confounding ) and no instantaneous treatment effect are satisfied . suppose also that , for every @xmath10 $ ] , @xmath92 has the same distribution as @xmath21 given @xmath15 . from theorem we know that , for @xmath58 satisfying restriction , @xmath135 is a zero of @xmath319 suppose now that @xmath135 is the only zero . suppose , furthermore , that we know that @xmath320 with @xmath321 compact , that @xmath322 is continuous for each @xmath7 and bounded by a constant @xmath323 which does not depend on @xmath324 , and that @xmath325 is continuous for each @xmath7 . then any sequence of ( almost ) zeros @xmath326 of @xmath327 that is , any sequence of estimators @xmath326 such that @xmath328 converges in probability to zero , is a consistent estimator for @xmath135 for each @xmath122 satisfying restriction . suppose , moreover , that @xmath329 is differentiable with the respect to @xmath117 with derivative bounded by a constant @xmath330 in a neighborhood of @xmath118 , and @xmath331 is differentiable with respect to @xmath108 with the derivative bounded by a constant @xmath332 in a neighborhood of @xmath119 . then for each @xmath58 satisfying restriction [ hregpc ] there is a neighborhood of @xmath135 such that @xmath333 is continuously differentiable with respect to @xmath308 . suppose , moreover , that the matrix @xmath334 is nonsingular . then there exists a sequence of ( almost ) zeros @xmath326 to @xmath335 . furthermore , any such sequence is asymptotically normal : @xmath336 with @xmath337 the matrix above and , with @xmath338 , @xmath339 proof consistency follows from theorem 5.9 of @xcite . repeatedly applying lebesgue s dominated convergence theorem shows that our conditions imply the conditions of the first and second paragraph after this theorem 5.9 . the existence of ( almost ) zeros follows from @xcite , section 3.9 , problem 9 , whose solution is practically given by the hint below it . this problem 9 states that if @xmath340 is a homeomorphism of a neighborhood of @xmath341 onto a neighborhood of @xmath342 , then every continuous @xmath343 for which @xmath344 is sufficiently small has at least one zero . in our case @xmath345 is continuously differentiable in a neighborhood of @xmath135 by restriction [ hregpc](d ) and the assumptions on @xmath116 and @xmath105 , under which differentiation and integration can be exchanged ( twice ) . the derivative of this @xmath346 at @xmath135 is nonsingular by assumption , and hence , @xmath310 is a homeomorphism of a neighborhood of @xmath341 onto a neighborhood of @xmath347 . @xmath335 is continuous in @xmath308 and close enough to @xmath346 for large @xmath267 with probability approaching @xmath39 because of the second paragraph below theorem 5.9 in @xcite . hence , @xmath335 has a zero with probability approaching @xmath39 . asymptotic normality follows from theorem 5.21 of @xcite , as follows . define @xmath348 , which is continuously differentiable with respect to @xmath308 in a neighborhood @xmath349 of @xmath135 under our conditions . making @xmath349 smaller so that all boundedness conditions hold on @xmath349 , we define @xmath350 , which is bounded by @xmath351 , a constant plus a constant times @xmath352 . this @xmath353 is square integrable since @xmath352 is square integrable : it is well known that counting processes with bounded intensity processes are square integrable [ it follows from , e.g. , proposition ii.4.1 of andersen et al . ( @xcite ) ] . for the same reason , @xmath354 . the remaining conditions of theorem 5.21 from @xcite were checked before , so , indeed , @xmath326 is asymptotically normal with asymptotic covariance matrix ( [ asvar ] ) . the asymptotic variance ( [ asvar ] ) is often estimated by replacing @xmath135 by their estimates and @xmath355 by @xmath144 . thus , confidence intervals for @xmath119 can be constructed . also , tests for whether @xmath119 has a specific value can be constructed that way . for more about testing , see section [ test ] . one can often simplify the expression for the asymptotic variance in equation ( [ asvar ] ) using corollary [ casvar ] below . we use the following lemma : [ caslem ] suppose that the conditions of theorem hold : assumptions bounded intensity process [ @xmath40 cadlag ] , no unmeasured confoundingno instantaneous treatment effect , for every @xmath10 $ ] , @xmath92 has the same distribution as @xmath21 given @xmath15 ( see section ) , for each @xmath99 , @xmath11 jumps at most finitely many times , and @xmath51 satisfies regularity condition . write @xmath356 and @xmath338 . then for @xmath357 every component of which satisfies restriction @xmath358 if , furthermore , @xmath359 is a correctly specified model for @xmath30 such that @xmath360 exists and @xmath141 is a correctly specified model for @xmath51 such that , for each @xmath7 , @xmath109 is differentiable with respect to @xmath108 at @xmath361 , then , for @xmath362 satisfying restriction [ hregpc ] , @xmath363 if the left- or right - hand side exists and @xmath364 if the left- or right - hand side exists , where @xmath365 . proof for the first statement , we use counting process theory from andersen et al . ( @xcite ) , chapter 2 . if @xmath366 is a martingale , @xmath367 ( if it exists ) is defined as a predictable process such that @xmath368 is a ( local ) martingale . if @xmath369 is another martingale , @xmath370 ( if it exists ) is defined as a predictable process such that @xmath371 is a ( local ) martingale . @xmath367 is called the predictable variation process of @xmath366 and @xmath370 is called the predictable covariation process of @xmath366 and @xmath369 . for vector - valued @xmath366 , @xmath367 is defined as a predictable process such that @xmath372 is a ( local ) martingale . hence , it is a matrix with @xmath373 at the @xmath374th row , @xmath160th column . as shown in theorem [ mgc ] , @xmath132 is a martingale with respect to the filtration @xmath225 . counting process martingales like this have compensators : @xmath375 [ see , e.g. , proposition ii.4.1 in andersen et al . ( @xcite ) ] . moreover , if @xmath376 and @xmath377 are ( locally ) bounded @xmath225-predictable processes with values in @xmath378 , then @xmath379 exists and @xmath380 [ proposition ii.4.1 or ( 2.4.9 ) in andersen et al . ( @xcite ) ] . because @xmath381 satisfies restriction [ hreg ] , @xmath382 is a bounded @xmath225-predictable process ( proof just as in lemma [ hpred ] ) . therefore , the theory above leads to @xmath383 for the second statement , notice that , under the conditions of the lemma , @xmath384 and the expectation of the second term here is equal to zero because of theorem [ thec ] . for the third statement , notice that , under the conditions of the lemma , @xmath385 because of the chain rule , and the expectation of the second term is equal to zero because of theorem [ thec ] . this lemma simplifies the asymptotic variance formula of the estimators in equation ( [ asvar ] ) : [ casvar ] suppose that the conditions of theorem hold : assumptions ( bounded intensity process ) , [ @xmath40 cadlag ] , no unmeasured confounding ( no instantaneous treatment effect ) , for every @xmath10 $ ] , @xmath92 has the same distribution as @xmath21 given @xmath15 ( see section ) , for each @xmath99 , @xmath11 jumps at most finitely many times , and @xmath51 satisfies regularity condition . suppose also that @xmath359 is a correctly specified model for @xmath30 such that @xmath360 exists and that @xmath141 is a correctly specified model for @xmath51 such that , for each @xmath7 , @xmath109 is differentiable with respect to @xmath108 at @xmath361 . then if @xmath386 and @xmath362 satisfies restriction [ hregpc ] , the asymptotic variance ( [ asvar ] ) is equal to @xmath387 with @xmath388 and @xmath389 with @xmath390 and , with @xmath365 , @xmath391 we conclude this section with an example to see the machinery work in practice . notice that the boundedness conditions of theorem [ consas ] are somewhat too restrictive for this example , but that the results hold true under these weaker restrictions , too . [ ] [ weilem ] consider the setting of example , and suppose that the assumptions of section [ mgs ] are satisfied . in example @xmath392 where at risk means at risk for initiation of prophylaxis treatment . @xmath109 is the solution to the differential equation @xmath393 with final condition @xmath394 and @xmath395 , so @xmath396 in example [ wei ] we already saw that @xmath397 is a zero of @xmath398 suppose now that @xmath397 is the only zero . suppose , furthermore , that the survival time @xmath1 takes values in a compact space @xmath399\subset\mathbb{r}$ ] and that we know that @xmath400 , with @xmath401 , @xmath402 , @xmath403 and @xmath404 all four compact ( note that this implies that @xmath201 and @xmath202 are bounded away from @xmath24 ) . then any sequence of ( almost ) zeros @xmath405 of @xmath406 that is , any sequence of estimators @xmath405 such that @xmath407 converges in probability to zero , is a consistent estimator for @xmath397 . moreover , @xmath389 as in corollary [ casvar ] exists , and @xmath408 and @xmath409 is a five - dimensional vector with zeros in the first four positions and @xmath410 in the fifth , with @xmath411 if this @xmath337 is a nonsingular matrix , then there exists a sequence of ( almost ) zeros @xmath405 of ( [ weiboh ] ) . furthermore , any such sequence is asymptotically normal : @xmath412 with @xmath337 the matrix above and @xmath413 moreover , @xmath227 and @xmath228 are asymptotically independent . the asymptotic independence here turns out to be no coincidence ; see lok ( @xcite ) . proof of lemma [ weilem ] the findings here are similar to the findings in theorem [ consas ] , but the boundedness conditions fail to hold here . consistency follows from @xcite , theorem 5.9 . existence of a sequence of ( almost ) zeros follows from @xcite , section 3.9 , problem 9 , whose solution is practically given by the hint below it . asymptotic normality follows from @xcite , theorem 5.21 . the asymptotic variance equals @xmath414 because of corollary [ casvar ] ( @xmath30 and @xmath58 are not bounded here , but we can restrict the interval to @xmath215 $ ] for @xmath216 and let @xmath415 ) . we leave checking the conditions of these theorems to the reader [ or see @xcite ( @xcite ) , section 7.7 ] . asymptotic independence of @xmath227 and @xmath228 follows by direct calculation , after noticing that @xmath416 with @xmath417 and @xmath418 define @xmath419 this concludes the proof : @xmath420 we show that one can often test whether treatment affects the outcome of interest without specifying a model @xmath141 for @xmath51 . this was conjectured , but not proved , in @xcite . if one does not have to specify a model for @xmath51 in order to test whether treatment affects the outcome , false conclusions caused by misspecification of the model for @xmath51 can be avoided . under the null hypothesis of the no treatment effect , @xmath89 and @xmath421 ( see section [ model ] ) . if there is no unmeasured confounding , @xmath92 does not predict @xmath26 given @xmath23 ( see theorem [ mgc ] ) . hence , if there is no unmeasured confounding and no treatment effect , adding the _ observed outcome _ @xmath1 to the prediction model for treatment effect should not help the prediction . this idea , presented in @xcite for the case of local rank preservation ( see section [ loc ] ) , can be proven to be correct as follows . technically , the tests in this section are similar to the score test [ for more about the score test see , e.g. , @xcite ( @xcite ) ] . suppose that the conditions of section [ mgs ] are satisfied , and that we have a correctly specified parametric model @xmath359 for @xmath30 . define @xmath422 with @xmath423 satisfying the regularity condition restriction [ hreg ] . the key idea of this procedure is that if treatment does not affect the outcome , @xmath89 , so @xmath421 , and @xmath424 has expectation zero because of theorem [ thec ] . since @xmath118 is unknown , we base the test on the limiting behavior under @xmath89 of @xmath425 , where @xmath227 is an estimator of the nuisance parameter @xmath118 . we will show that if @xmath89 , @xmath425 converges to a normal random variable with expectation zero , which leads to a test for whether @xmath89 in the usual way . the nuisance parameter @xmath118 will be estimated using some set of estimating equations @xmath426 with @xmath427 , @xmath428 differentiable in @xmath117 and @xmath429 . a natural choice would be a maximum ( partial ) likelihood estimator for @xmath118 . we suppose throughout this section that the resulting estimator @xmath227 is consistent and asymptotically normal with @xmath430 as will usually follow from , for example , @xcite , theorem 5.21 . if @xmath116 and @xmath431 are sufficiently smooth , @xmath432 is differentiable with respect to @xmath117 , and a taylor expansion around @xmath118 leads to @xmath433 with @xmath434 the derivative of @xmath435 with respect to @xmath117 and @xmath436 between @xmath118 and @xmath227 . since @xmath227 converges in probability to @xmath118 , so does @xmath436 . therefore , usually , @xmath437 . sufficient conditions under which this holds are given in appendix [ gcap ] , lemma [ pgc ] . because of ( [ thetaas ] ) and the central limit theorem , @xmath438 converges in distribution . therefore , an application of slutzky s lemma leads to @xmath439 with @xmath440 . if @xmath89 , @xmath421 , so that theorem [ thec ] implies that also the expectation of @xmath441 is equal to zero . therefore , the central limit theorem can be applied on the vector with @xmath442 on the right - hand side ; it converges to a normal random variable with expectation zero . because of the continuous mapping theorem [ see , e.g. , @xcite ( @xcite ) , chapter 18 ] , @xmath425 then converges to a normal random variable with expectation zero , too . calculation of its limiting covariance matrix is standard [ see , e.g.,@xcite ( @xcite ) , chapter 18 ] . to save space , we omit that calculation here . if desirable , one can use theorem [ thec ] and lemma [ caslem ] to simplify the expression . notice that a test for whether @xmath443 for any specific @xmath444 can be constructed in exactly the same way . if we have a correctly specified model @xmath141 for @xmath51 , this thus also leads to a confidence region for @xmath119 in the usual way , using the duality between testing and confidence regions : include those @xmath108 for which the null hypothesis @xmath445 is not rejected . the proof of consistency and asymptotic normality of the estimators presented in this article applies to continuous - time structural nested models . a similar proof applies to structural nested models in discrete time ( when covariates are only measured at finitely many fixed times @xmath446 , which are the same for all patients and known in advance ) . @xcite argue without proof that consistency and asymptotic normality should hold for discrete time structural nested models under reasonable assumptions ; the proof is completed with the current article , as follows . it is easy to see that in discrete time , @xmath447 is the compensator of @xmath25 with respect to @xmath29 [ see , e.g. , @xcite ( @xcite ) , section 7.4 ] . the assumption of no unmeasured confounding can be formalized as @xmath448 since @xmath449 given @xmath193 , this implies that also @xmath450 the discrete - time counterparts of theorem [ thec ] and theorem [ mgc ] follow immediately [ see @xcite ( @xcite ) ] . consistency and asymptotic normality follow in the same way as for continuous time models . the tests for treatment effect in this article can be carried out without specifying a model for treatment effect ; that is , no model for @xmath51 is needed . this is an important feature of the tests because it allows one to avoid false conclusions caused by misspecification of the model for @xmath51 . in practice , it may be hard to specify a correct parametric model for the infinitesimal shift - function , @xmath51 . thus , it is good that specification of a model for @xmath51 is not needed to test for treatment effect . the estimators in this article require the correct specification of a model for treatment effect and of a model for prediction of treatment changes . for the discrete - time setting , @xcite has recently proposed estimators which are doubly robust . doubly robust estimators are consistent and asymptotically normal if ( i ) the model for prediction of treatment changes ( @xmath30 in the current article ) is correctly specified or if ( ii ) a regression model of a blipped down outcome [ @xmath109 in the current article ] on past treatment- and covariate history @xmath23 is correctly specified . in any case , the model for treatment effect ( @xmath90 in the current article ) has to be well specified . in this article estimation started with the specification of a model for the infinitesimal shift - function , @xmath51 . interpretation of results may be easier when one starts with a model like ( [ dasim ] ) , @xmath451 , given @xmath452 . here , @xmath65 is the duration of treatment in the interval @xmath66 . the main results in the current article apply also to this model . the proofs for theorems [ thec ] and [ consas ] do not depend on @xmath110 being the solution to @xmath453 . the proof of theorem [ mgc ] does depend on @xmath453 , but it simplifies considerably if ( [ dasim ] ) or ( [ dasim2 ] ) is used as a starting point . let us show this for ( [ dasim ] ) . define @xmath454 . using the first part of the proof of theorem [ mgc ] , for @xmath455 , @xmath285}(t ) 1_{a}(\overline{z}_{t_1})1^{(n)}_{(x_1,x_2)}(\tilde{x}(t_1))\\ & & \qquad=1_{(t_1,t_2]}(t ) 1_{a}(\overline{z}_{t_1})1^{(n)}_{(x_1,x_2)}\bigl(\tilde{x}(t)+\mathit{dur}(t_1,t ) ( -1+e^\psi)\bigr)\\ & & \qquad = h_t^{(n)}(\tilde{x}(t),\overline{z}_{t-}),\end{aligned}\ ] ] where @xmath456 is the duration of treatment in the interval @xmath457 . @xmath458 is measurable if the duration of treatment until @xmath7 is in included in @xmath6 . moreover , if @xmath459 and @xmath127 , then @xmath460 . hence , it follows immediately that @xmath458 satisfies restriction [ hreg ] , which concludes the proof . depending on the specific application , it will be more appropriate to start with @xmath461 given @xmath452 or with a model for @xmath51 ; see , for example , example [ prior ] . in the previous literature [ see , e.g. , robins et al . ( @xcite ) , @xcite , witteman et al . ( @xcite ) , keiding et al . ( @xcite ) and hernn et al . ( @xcite ) ] applications have been carried out under the assumption of local rank preservation , where @xmath148 for each @xmath7 . as pointed out in these papers , and as discussed in section [ loc ] , the assumption of local rank preservation is generally considered implausible . this article relaxes the assumption of local rank preservation in structural nested models . the estimators and tests applied in the previous literature are specific cases of the estimators and tests studied in this article , with the only difference that some of the estimators in the previous literature allow for censoring of @xmath1 . aside from the issue of censoring , this article provides a mathematical foundation behind previous estimators , relaxes the specification of the counterfactual outcomes as deterministic variables , and allows for a distributional interpretation of the estimators . @xcite conjectures that one can often use standard software to test whether treatment affects the outcome of interest ( without specifying a model @xmath90 for @xmath51 ) , and to estimate @xmath108 . lok ( @xcite ) shows that both testing and estimation can also be considered from a partial likelihood point of view . as shown in lok ( @xcite ) , this approach leads to a subclass of the estimators and tests studied in this article which can indeed be calculated with standard software . example [ wei ] is a specific case of that . the possibility to use standard software may be a good reason to choose these estimators in practice . see @xcite and lok ( @xcite ) for a more elaborate discussion . the approach adopted in the current article leads to a large class of estimators and tests . when treatment and covariates change at finitely many fixed times only , robins ( @xcite ) proposes , without proof , an optimal procedure for survival and nonsurvival outcomes , respectively . the optimal choice of estimators or tests under the framework of this article is another intriguing topic for future research . the current article assumes a parametric model @xmath116 for the prediction of treatment changes . in practice , applications have often used a semi - parametric cox model for @xmath116 . @xcite shows that specifying @xmath116 using a semiparametric cox model leads to unbiased estimating equations , which just as in this article are martingales for the true parameters . consistency and asymptotic normality of the resulting estimators for @xmath51 still remain to be shown and constitute interesting topics for future research . in many applications , observations are censored . @xcite and hernn et al . ( @xcite ) have proposed methods to deal with censoring that could potentially be adapted to the results in this paper . for @xmath51 of the form of example [ daexa ] , @xcite includes proofs with censoring due to the fact that the study ends , so - called administrative censoring , using ideas from these previous papers . the estimators in this paper depend on solving a differential equation for each observation . in the examples , these equations are simple enough to be solved analytically . if that is not possible , these equations should be solved numerically . it might be worth investigating how a small contamination of the solution to the differential equation @xmath92 affects the estimates of treatment effect . i conclude with a discussion of the assumptions used in this article . the most important assumption in this article is the assumption of no unmeasured confounding ( assumption [ intconf ] ) . as discussed before , this assumption is valid if all information has been recorded which both ( i ) predicts treatment decisions and ( ii ) is an independent risk factor for the outcome of interest . the validity of the assumption of no unmeasured confounding can not be tested statistically , and depends on the quality of the recorded information . therefore , it is for subject matter experts to decide about the plausibility of the assumption of no unmeasured confounding . second , we only estimate the effect of treatment for which a short duration of treatment has only a small effect on the distribution of the outcome of interest . the effect of the treatment on an individual patient may be large , as long as the probability of such an effect is small for any small duration of treatment . third , the assumption of no instantaneous treatment effect ( assumption [ jump ] ) is also restrictive : it excludes the estimation of the effect of treatments that have instantaneous effects , such as surgery or other point exposures . the remaining assumptions in this article are mostly benign . the assumption that the covariate- and treatment process can be represented by a cadlag process is generally accepted for most medical situations [ see , e.g. , andersen et al . ( @xcite ) ] . the functions @xmath122 and @xmath90 can be chosen such that the regularity conditions on these functions are satisfied . even if @xmath122 is not bounded , it can often be approximated by bounded functions , and results may follow by a simple application of lebesgue s dominated convergence theorem [ see , e.g. , example [ wei ] ] . the same is true for the boundedness condition ( assumption [ int ] ) of the intensity process @xmath30 . assumption [ ycadlag ] that the counterfactual process @xmath21 is cadlag is impossible to verify , but it is a plausible and convenient regularity condition . [ isun ] suppose that a function @xmath76 satisfies the following : ( ) . if @xmath11 does not jump in @xmath239 , then @xmath76 is continuous in @xmath240 on @xmath241 and can be continuously extended to @xmath242 $ ] . ( ) . for each @xmath99 , there exists a constant @xmath462 such that @xmath463 for all @xmath10 $ ] and all @xmath248 . suppose , furthermore , that , for each @xmath99 , there are no more than finitely many jump times of @xmath11 . then , for each @xmath124 $ ] and @xmath125 , there is a unique continuous solution @xmath464 to @xmath465 with boundary condition @xmath466 and this solution is defined on the whole interval @xmath2 $ ] . this theorem follows from well - known results about differential equations ; see , for example , @xcite , chapter 2 . for the next theorem , we also refer to @xcite , chapter 2 . it is a consequence of gronwall s lemma . [ difb ] suppose that @xmath467 is an open or closed interval in @xmath468 , @xmath469 is continuous and @xmath470 is continuous , and suppose that @xmath471 for all @xmath472 and @xmath473 . then , for every @xmath474 and @xmath125 , there is a unique solution @xmath475 of @xmath476 with @xmath477 , and this solution is defined for all @xmath472 . if @xmath478 is continuous and @xmath479 is a solution of @xmath480 , then @xmath481 for all @xmath482 with @xmath483 . in @xcite the interval is always an open interval , but as is generally known , this can be overcome by extending both @xmath484 and @xmath310 outside the closed interval @xmath467 by taking the values at the boundary of @xmath467 . this preserves the lipschitz- and continuity conditions . existence and uniqueness on all of finitely many intervals implies global existence and uniqueness . we have a differential equation with end condition at @xmath3 , so we are interested in @xmath485 with @xmath486 : [ difbc1 ] suppose that the conditions of theorem are satisfied . then , for every @xmath474 and @xmath487 , there is a unique solution @xmath475 of @xmath476 with @xmath477 , and this solution is defined for all @xmath472 . if @xmath478 is continuous and @xmath479 is a solution of @xmath480 , then @xmath488 for all @xmath485 with @xmath486 . proof put @xmath489 . then @xmath490 where @xmath491 . thus , @xmath489 is a solution of the differential equation @xmath492 with boundary condition @xmath493 . define also @xmath494 . applying theorem [ difb ] on @xmath301 concludes the proof , as follows : @xmath495 with @xmath496 notice that , because of equation ( [ lc2 ] ) , @xmath497 with @xmath498 . hence , theorem [ difb ] implies that @xmath499 for the first term , we do a change of variables ; @xmath201 from @xmath24 to @xmath7 , put @xmath500 ; @xmath501 . @xmath502 ; @xmath503 from @xmath504 to @xmath505 . we conclude that the first term is equal to @xmath506 for the second term , similar changes of variables can be done , resulting in corollary [ difbc1 ] . in most of this article we assume that the function which maps @xmath549 to @xmath550 , with @xmath551 , is a measurable function on @xmath123 , with the projection @xmath18-algebra on @xmath552 ( see section [ sets ] ) . moreover , we sometimes assume that the function which maps @xmath553 to @xmath92 , with @xmath551 , is a measurable function on @xmath554 . in this appendix we give sufficient conditions for this . if these two functions are measurable , @xmath233 is a filtration , and , moreover , @xmath233 is the same as @xmath555 [ see equation ( [ fil ] ) in section [ mgs ] ] . [ mtb ] suppose that @xmath51 satisfies regularity assumption and that , for each @xmath99 , @xmath11 jumps at most finitely many times . then the function which maps @xmath549 to @xmath550 , with @xmath551 , is a measurable function from @xmath123 to @xmath378 . also , the function which maps @xmath553 to @xmath92 , with @xmath551 , is a measurable function from @xmath123 to @xmath378 . for the proof of this result , which is quite technical since many results on differential equations are nonconstructive , we refer to @xcite . the proof uses the idea behind euler s forward method to approximate the solution to the differential equation . the following lemma is a worked - out case of theory from @xcite , chapter 19 . [ pgc ] let @xmath556 be i.i.d . random variables with values in a measurable space @xmath557 . let @xmath558 be a collection of measurable functions from @xmath557 to @xmath559 indexed by a subset @xmath560 which contains an open neighborhood @xmath561 of @xmath118 . suppose that @xmath562 is continuous on @xmath561 for every @xmath563 . suppose also that there exists a measurable function @xmath52 on @xmath557 such that @xmath564 for every @xmath565 and such that @xmath566 exists . then if @xmath227 converges in probability to @xmath118 , @xmath567 where @xmath144 indicates the empirical distribution of @xmath568 . proof notice that @xmath569 we show that both terms converge to zero in probability . choose @xmath570 compact and such that it contains an open neighborhood of @xmath118 . example 19.8 from @xcite implies that , under the conditions above , @xmath571 since @xmath572 and @xmath573 contain an open neighborhood of @xmath118 , this implies that the first term converges in probability to zero . for the second term , notice that , on @xmath561 , @xmath574 is continuous in @xmath117 and that each of the components of @xmath575 is bounded by the integrable function @xmath52 , so that lebesgue s dominated convergence theorem implies that @xmath576 is continuous in @xmath117 on @xmath561 . thus , since @xmath572 , also the second term converges in probability to zero . an earlier version of this article was a chapter of my ph.d . dissertation at the free university of amsterdam . i am indebted to richard gill and aad van der vaart for their support , insight and encouragement on this project . i also thank james robins for fruitful discussions . i thank both referees for helpful suggestions . besides that , i thank the netherlands organization for scientific research ( nwo ) for a scholarship . hernn , m. a. , cole , s. r. , margolick , j. , cohen , m. and robins j. m. ( 2005 ) . structural accelerated failure time models for survival analysis in studies with time - varying treatments . _ pharmacoepidemiology and drug safety _ * 14 * 477491 . keiding , n. , filiberti , m. , esbjerg , s. , robins , j. m. and jacobsen , n. ( 1999 ) . the graft versus leukemia effect after bone marrow transplantation : a case study using structural nested failure time models . _ biometrics _ * 55 * 2328 . lok , j. j. , gill , r. d. , van der vaart , a. w. and robins , j. m. ( 2004 ) . estimating the causal effect of a time - varying treatment on time - to - event using structural nested failure time models . _ statist . neerlandica _ * 58 * 271295 . mark , s. d. and robins , j. m. ( 1993 ) . estimating the causal effect of smoking cessation in the presence of confounding factors using a rank preserving structural failure time model . _ statistics in medicine _ * 12 * 16051628 . robins , j. m. ( 1989 ) . the analysis of randomized and non - randomized aids treatment trials using a new approach to causal inference in longitudinal studies . in _ health service research methodology : a focus on aids _ ( l. sechrest , h. freeman and a. bailey , eds . ) 113159 . nchsr , u.s . public health service , washington , dc . robins , j. m. ( 1993 ) . analytic methods for hiv treatment and cofactor effects . in _ methodological issues of aids behavioral research _ ( d. g. ostrow and r. kessler , eds . ) 213287 . plenum press , new york . robins , j. m. ( 2000 ) . robust estimation in sequentially ignorable missing data and causal inference models . in _ proceedings of the american statistical association section on bayesian statistical science _ * 1999 * 610 . robins , j. m. , blevins , d. , ritter , g. and wulfsohn , m. ( 1992 ) . g - estimation of the effect of prophylaxis therapy for pneumocystis carinii pneumonia on the survival of aids patients . _ epidemiology _ * 3 * 319336 . robins , j. m. , rotnitzky , a. and scharfstein , d. o. ( 2000 ) . sensitivity analysis for selection bias and unmeasured confounding in missing data and causal inference models . in _ statistical models in epidemiology the environment and clinical trials _ ( m. e. halloran and d. berry , eds . ) springer , new york . robins , j. m. and wasserman , l. ( 1997 ) . estimation of effects of sequential treatments by reparameterizing directed acyclic graphs . in _ proceedings of the thirteenth conference on uncertainty in artificial intelligence _ ( d. geiger and p. shenoy , eds . ) 409420 . morgan kaufmann , san francisco , ca . witteman , j. c. m. , dagostino , r. b. , stijnen , t. , kannel , w. b. , cobb , j. c. , de ridder , m. a. j. , hofman , a. and robins , j. m. ( 1998 ) . g - estimation of causal effects : isolated systolic hypertension and cardiovascular death in the framingham study . _ american j. epidemiology _ * 148 * 390401 .

partly motivated by our increased capability in recent years to perform accurate observations on the sky , we have turned our attention to investigating the impact of quantum effects on the evolution of the universe . in pursuit of this goal , quantum field theory in cosmological spacetimes continues to be an increasingly important field of research . among other things , it investigates how quantum fluctuations affect the background spacetime in perturbative quantum gravity , a process also known as quantum backreaction . a line of research deals for example with the backreaction of quantum fields whose spectrum is nearly flat . examples of these fields are the minimally coupled massless scalar and the graviton . consequently , these fields are expected to yield a substantial backreaction @xcite . also , fermions in curved spacetimes and in particular maximally symmetric spacetimes were investigated ( see e.g. : @xcite ) . another approach to this question is concerned with calculating the possible effect of the trace anomaly on the background spacetime . some authors argue that this effect , particularly in relation with the cosmological constant problem , could be significant @xcite , however also see @xcite . an essential element to study quantum effects in curved spaces is the propagator . due to its high degree of symmetry , de sitter spacetime proves to be ideally suited for calculating various quantum effects . moreover , results are immediately applicable to inflation , cosmologists favourite paradigm for a brief exponentially fast expansion in the early universe . chernikov and tagirov calculated the scalar propagator in de sitter spacetime @xcite , also see @xcite . the vector propagator was constructed by @xcite and the graviton propagator received contributions from @xcite . candelas and raine calculated the fermion propagator @xcite . however , de sitter spacetime suffers from several drawbacks . firstly , pure de sitter spacetime is never realised in nature . since de sitter spacetime corresponds to a globally constant hubble parameter @xmath4 , this is in reality never attained . secondly , de sitter spacetime is non - dynamical . it is therefore inconsistent to study backreaction effects while at the same time assuming a constant background . we should allow the background to change to encompass all possible backreaction effects consistently . therefore , it is necessary to consider propagators in more general spacetimes . the scalar and graviton propagator in quasi de sitter spacetime were calculated by @xcite and subsequently generalised to friedmann - lematre - robertson - walker or flrw spacetimes with constant deceleration by @xcite . in this paper we calculate the fermion propagator in flrw spacetimes with constant deceleration . we should touch upon an important issue . in the massive case we need two additional constraints to be satisfied in order to solve for the propagator : [ assumptions ] @xmath5 recall that all interesting epochs in the evolution of our universe satisfy the first constraint . in the matter era we have @xmath6 , in the radiation dominated epoch we find @xmath7 . it also serves as an approximation to both inflation and the current dark energy dominated epoch when @xmath8 . note that @xmath9 coincides with the slow - roll parameter of inflation and it is straightforwardly related to the somewhat more familiar deceleration parameter @xmath10 . hence ( [ epsilon ] ) is equivalent to requiring a constant deceleration parameter . the second assumption ( [ mhassumption ] ) is required to analytically derive the propagator in the massive case . in yukawa theory , the mass of the fermion is generated by a scalar field for which we can achieve @xmath11 in several cases to which we will turn shortly . we then calculate the one loop effective action induced by fermions using our propagator . the one loop backreaction arises from integrating out a free , quadratic fermion field and , using dimensional regularisation , this generates a correction to the ( classical ) friedmann equations . the hubble parameter occurring in this effective action is now a dynamical quantity . consequently , the backreaction of these fermions on the background spacetime can thus be analysed dynamically . also , when the fermions are coupled to a scalar field , this opens up the possibility to study the impact of the fermions on the evolution of scalar fields . this paper is outlined as follows . in section [ fermions in flrw spacetimes ] we review the basic theory required to study fermions in curved spaces and we establish our notation . sections [ flrw fermion propagator : the massless case ] and [ flrw fermion propagator : the massive case ] are devoted to deriving the fermionic propagator in flrw spacetimes in the massless and massive case , respectively . finally , in section [ one loop effective action ] we apply our propagator to calculate the one loop effective action . fermions are in general @xmath12 dimensional curved spacetimes described by the action @xcite : @xmath13 - m \bar{\psi}\psi\right\ } , \ ] ] where the dirac matrices @xmath14 satisfy the following anti - commutation relations : @xmath15 variation with respect to @xmath16 yields the equation of motion the fermion field @xmath17 satisfies : @xmath18 we will make use of the vierbein formalism which can be thought of as a transformation of the metric tensor to a locally flat minkowski metric : @xmath19 where @xmath20 is the minkowski metric . we use greek letters to run over spacetime indices whereas we use latin letters either for the tangent space indices or for spinor indices . it will be clear from the context which interpretation we are using . the minkowski metric and the ( flat space ) dirac matrices @xmath21 are spacetime independent . spacetime and tangent space indices are raised or lowered by making use of the full metric or minkowski metric , respectively . the covariant derivative acting on a dirac spinor is defined as : @xmath22 where the spin connection is given by : @xmath23 \,,\ ] ] such that @xmath24 . let us specialise to flat flrw spacetimes in which the metric is given by @xmath25 . here , @xmath26 is the scale factor of the universe in conformal time defined by @xmath27 . in flrw spacetimes the vierbeins are a function of conformal time only : @xmath28 we thus find : @xmath29 this useful identity relates covariant derivatives to partial derivatives in the tangent space . since we work in the chirality representation , the flat space dirac matrices are given by : @xmath30 where @xmath31 is the @xmath32 identity matrix and : @xmath33 and where the @xmath34 denote the pauli matrices . ) assuming @xmath35 . near the minimum of the potential ( [ phiproptoh ] ) , assumption ( [ mhassumption ] ) is satisfied . [ fig : potential ] ] _ _ in yukawa theory the mass of the fermion appearing in equation ( [ fermionaction ] ) is time dependent and generated by a scalar field . let us thus briefly consider the complex system of a fermion whose mass is generated by a homogeneous scalar field @xmath36 , both of which are coupled to gravity . the action is : @xmath37 \right\}\label{actiontotal } . \phantom{1}\end{aligned}\ ] ] here @xmath38 the yukawa coupling constant . by comparing equations ( [ actiontotal ] ) and ( [ fermionaction ] ) we can thus identify the mass of the fermion as : @xmath39 furthermore , the ( classical ) evolution of the homogeneous background field @xmath36 is governed by : @xmath40 where @xmath41 is the potential : @xmath42 here , @xmath43 denotes some non - minimal coupling to gravity and @xmath44 is the ricci scalar . in flrw spacetimes in @xmath45 we have @xmath46 , where the hubble parameter is given by @xmath47 . dots denote derivatives with respect to cosmic time and dashes correspond to conformal time derivatives . assuming @xmath48 , the solution of ( [ eomphi ] ) is : @xmath49}\,.\ ] ] clearly , @xmath11 and assumption ( [ mhassumption ] ) is satisfied . moreover , observe we need @xmath50 . the potential is depicted in figure [ fig : potential ] . given a potential of the form ( [ potential ] ) , we identify some interpretations leading to this scenario . in the early universe the higgs field satisfies @xmath51 and hence ( [ phiproptoh ] ) . a quintessence field has @xmath52 and ( [ phiproptoh ] ) is also satisfied in the late universe . finally , we can infer that a massless minimally coupled scalar field is also allowed in the regime where @xmath53 or @xmath54 . we work in the approximation where @xmath55 , which is not equivalent to the standard slow - roll approximation of inflation , where one introduces the slow - roll parameters @xmath9 and @xmath56 . these parameters are allowed to vary slowly in time such that @xmath57 and @xmath58 are higher order in the slow - roll parameters and can be neglected when one is interested in the leading order behaviour only . when @xmath9 is allowed to vary in time , equation ( [ phiproptoh ] ) generalises to @xmath59 , where : @xmath60 } + \mathcal{o}\left(\dot{\alpha}\right ) \,.\ ] ] the leading order slow - roll approximation to this equation corresponds to : @xmath61 } \,.\ ] ] note the equation above depends on @xmath9 , whereas it does not depend on @xmath56 . depending on which pole prescription one uses in the spirit of the schwinger - keldysh formalism ( see e.g. @xcite ) , one can obtain different propagators such as the anti - time ordered , time ordered or wightman propagators . we are primarily interested in the time ordered or feynman fermion propagator , defined by : @xmath62 the symbols @xmath63 and @xmath64 are the spinor indices and @xmath65 denotes the state of the system . also note the minus sign in front of the second @xmath66-function . the fermion propagator at tree level satisfies : @xmath67 is_{\mathrm{f}}^{ab } ( x,\tilde{x } ) = i \delta^{\scriptscriptstyle{d}}\!(x-\tilde{x } ) \mathbb{i}^{ab}\,,\ ] ] and a likewise equation when @xmath68 and @xmath69 are interchanged . we already discussed that flrw spacetimes can be characterised by the scale factor @xmath63 as a function of either cosmic or conformal time . there are however a few more important properties of this spacetime that are worth mentioning here . an important relation that we will use throughout this manuscript is : @xmath70 this equation is equivalent to assumption ( [ epsilon ] ) , see @xcite . let us define some relevant geometrical functions : [ y ] @xmath71 which all vanish near the lightcone . here , @xmath72 in ( [ y ] ) refers to the feynman or time ordered pole prescription of the feynman pole prescription is unrelated to the `` slow roll '' parameter @xmath9 in flrw spacetimes . ] . in de sitter spacetime , i.e. : when @xmath73 , the de sitter invariant function @xmath74 is related to the geodesic length @xmath75 as @xmath76 . massless fermions in flrw spacetimes are not difficult to deal with . by making use of equation ( [ covariantderivative2 ] ) the propagator for massless ( conformal ) fermions should satisfy : @xmath77 let us recall the following identity : @xmath78 we can thus immediately infer the solution of equation ( [ fermionpropmassless ] ) : @xmath79 here , @xmath80 . this completes the calculation for the propagator of massless fermions in flrw spacetimes in @xmath12 dimensions . this propagator is valid in any flrw spacetime and assumption ( [ epsilon ] ) can be relaxed . because massless fermions are conformal in any dimension , their propagator is much easier to calculate than the massless scalar propagator @xcite . a massive fermion is not conformal and hence its propagator in general contains some complicated mass dependence ) is much more complicated than for scalar fields . ] . we firstly solve for the fermionic mode functions . we generalise the approach outlined in @xcite to incorporate flrw spacetimes with constant @xmath9 . using these mode functions , we then return to position space to construct the feynman propagator . let us firstly define the rescaled fermionic spinor : @xmath81 keeping an eye on equation ( [ covariantderivative2 ] ) one can easily check that this factor is chosen such to conveniently transform the covariant derivative into a partial derivative . equation of motion ( [ fermioneom ] ) thus reads : @xmath82 the canonical momentum associated with @xmath83 follows from promoting the variation of ( [ fermionaction ] ) with respect to @xmath84 to an operator . according to the usual moves in quantum field theory , we impose anti - commutation relations between @xmath83 and its associated canonical momentum : @xmath85 with the other anti - commutators vanishing . by making use of the rescaling in equation ( [ fermionrescaled ] ) , appreciate that the anti - commutation relations above simplify : [ anticommutation2 ] @xmath86 therefore we expand the rescaled spinors @xmath87 and @xmath88 in creation and annihilation operators as follows : [ creationannihilationexpansion ] @xmath89 the expansion above merits a few remarks . firstly , @xmath90 and @xmath91 are the fermion and anti - fermion annihilation operators of helicity @xmath92 , respectively , in the usual sense : @xmath93 . the helicity @xmath92 , i.e. : the spin in the direction of motion , can be either @xmath94 or @xmath95 in units of @xmath96 . let us for simplicity return to the @xmath45 setting we are familiar with . we will generalise the following considerations shortly to arbitrary dimensions . in the equation above @xmath97 is a 4-spinor of helicity @xmath92 . we decompose the 4-spinor @xmath98 into a direct product of chirality and helicity 2-spinors : @xmath99 here , @xmath100 and @xmath101 are left- and right - handed 1-spinors of helicity @xmath92 , respectively . furthermore , @xmath102 is the helicity 2-eigenspinor : @xmath103 where @xmath104 is a shorthand notation for the pauli matrices . furthermore note that @xmath105 . for future convenience , let us explicitly derive the two vectors @xmath102 . we write : @xmath106 and keep in mind that : @xmath107 it is a trivial exercise to solve for the two eigenvectors of @xmath108 , which correctly normalised to unity read : [ helicity1 ] @xmath109 moreover , the helicity eigenstates are mutually orthogonal : @xmath110 finally , we have tacitly ignored the expansion of the anti - particle contribution @xmath111 in chirality and helicity spinors in equation ( [ creationannihilationexpansion ] ) . we will return to this subtlety shortly . our chirality and helicity decomposition of the spinor degrees of freedom in equation ( [ fermiondecompose ] ) also works in @xmath12 spacetime dimensions . once a particular representation of the gamma matrices has been found , the projection operator : @xmath112 splits the @xmath113-spinor degrees of freedom in two equal contributions of definite chirality , i.e. : a left- and right - handed @xmath114-spinor ( see e.g. @xcite ) . here : @xmath115 where @xmath116 is fixed by requiring that @xmath117 is a proper projection operator : @xmath118 this yields : @xmath119 \,.\ ] ] in @xmath45 , we recognise the familiar @xmath120 . an helicity operator that commutes with the projection operator can be defined in fourier space as : @xmath121 where @xmath122 is the levi - civita tensor in @xmath123 dimensions . the exponential phase ensures that @xmath108 is hermitian . finally , the gamma function has been inserted to account for the number of different permutations of gamma matrices . hence , @xmath124 and @xmath125 are . note the definition of the helicity operator is consistent with ( [ helicity ] ) . most importantly , the chirality and helicity decomposition as proposed in equation ( [ fermiondecompose ] ) carries through for any dimension . when an explicit form of @xmath102 is required , we return to @xmath45 and use equation ( [ helicity1 ] ) . we subsequently generalise the result of such calculations by analytically extending it to arbitrary @xmath12 . in textbooks on quantum field theory , such as @xcite , one expands the fermion in momentum space in spin eigenstates , whereas we expand in helicity eigenstates . this results in a different normalisation requirement on the spinors in momentum space and consequently the standard textbook results can not be straightforwardly copied . the reason for expanding in helicity eigenstates is that in curved spacetimes helicity is more convenient to work with rather than spin . spin is a conserved quantity , i.e. : an appropriate quantum number , in static backgrounds , whereas helicity is conserved in time dependent but spatially homogeneous backgrounds @xcite . one can easily show that the helicity operator constructed above in ( [ helicityd ] ) commutes with the kinetic operator in wigner space for spatially homogeneous correlators . we normalise the spinors using two conditions : a consistent canonical quantisation and orthogonality of particle and anti - particle spinors . we thus impose the usual anti - commutation relations between creation and annihilation operators : [ anticommutation3 ] @xmath126 and all other anti - commutators vanish identically . now ( [ anticommutation3 ] ) is consistent with ( [ anticommutation2 ] ) if and only if : @xmath127 note that this is a condition on @xmath128 matrix elements in spinor space , i.e. : for each @xmath63 and @xmath64 the above equality has to be satisfied . secondly , we require that the particle and anti - particle spinors are mutually orthogonal : @xmath129 this restricts 4 matrix elements in helicity space , i.e. : for each @xmath130 the above equality needs to be obeyed . peskin and schroeder @xcite impose normalisation conditions on both particle and anti - particle spinors in spin space . the corresponding conditions in helicity space are only partially satisfied . both particle and anti - particle spinors of different helicity are trivially orthogonal by construction ( [ kvectorsorthogonal ] ) . however , we do not relate the amplitude of equal helicity states for particles and anti - particles to the mass of the spinor because helicity eigenstates for a particle at rest can not be constructed . we will now solve the dirac equation . let us insert equation ( [ creationannihilationexpansion ] ) into the dirac equation ( [ fermioneom2 ] ) . making use of our chirality and helicity decomposition ( [ fermiondecompose ] ) transforms the dirac equation to a first order coupled system of differential equations : [ diraceq1 ] @xmath131 the arguments of the functions in the equation above have been omitted for notational convenience . the left- and right - handed spinors however are now a function of the magnitude of the fourier mode @xmath132 only , because the differential equation is invariant under @xmath133 . standard model fermions are chiral , in the sense that @xmath134 . for simplicity , we take one mass @xmath135 only . the following derivation can easily be generalised when standard model fermions are considered by making use of the @xmath117 projectors in equation ( [ p+- ] ) . we now return to the expansion in chirality and helicity spinors of the anti - particle contribution @xmath111 , which we did not need so far . since the dirac equation is a linear differential equation , the solution for @xmath111 can not contain any new degrees of freedom once we completely solved for @xmath98 . in other words : the form of @xmath111 is dictated once we have completely specified @xmath98 . we expand @xmath111 slightly differently : @xmath136 note we flipped the position of the left- and right - handed spinors compared to ( [ fermiondecompose ] ) . consequently , the resulting equations of motion for both @xmath137 and @xmath138 are identical to ( [ diraceq1 ] ) . alternatively , we could have expanded differently and flipped helicity , i.e. : send @xmath139 in equation ( [ fermiondecompose ] ) . because the mass enforces mixing between different chirality states , labelling these states differently has no physical relevance . we return to equation of motion ( [ diraceq1 ] ) and define : @xmath140 this transforms equation ( [ diraceq1 ] ) to : [ diraceq2 ] @xmath141 we exploit assumption ( [ epsilon ] ) , which is equivalent to ( [ epsilonconstant ] ) . inserting this expression into the equation above yields : [ diraceq3 ] @xmath142 recall that the mass of a fermion in yukawa theory is time dependent in flrw spacetimes . there does not exist a rescaling of the functions @xmath143 that removes the mass dependence in the equation of motion above . even when one assumes that the mass @xmath144 is time independent , one can not find such a rescaling . it is particularly simple to choose @xmath145 constant as following from equation ( [ phiproptoh ] ) and for convenience we define : @xmath146 we can proceed and decouple the two linear differential equations to find : @xmath147 the solutions are given by : @xmath148 where @xmath149 are two normalisation constants that still need to be determined and @xmath150 is the hankel function of the first kind of order : @xmath151 this implies : @xmath152 the hankel function of the second kind also solves equation ( [ diraceq4 ] ) . in the infinite asymptotic past , equivalent to the deep uv , the fermions become effectively massless for @xmath153 and the distinction between @xmath154 and @xmath155 vanishes as a consequence . therefore we require that the vacuum mode functions in this regime equal the standard conformal vacuum solutions : [ uvlimit ] @xmath156 this excludes the hankel function of the second kind for the moment from contributing . away from the vacuum , we should of course also incorporate the second solution to ( [ diraceq4 ] ) to allow for mode mixing . note that the uv limits above yield the familiar uv behaviour for the left- and right - handed spinors ( e.g. : @xcite ) : [ uvlimit2 ] @xmath157 when solutions ( [ diraceq5sols ] ) are substituted in ( [ diraceq3a ] ) one finds : @xmath158 we are free to fix the remaining coefficient @xmath159 by the following condition : @xmath160 we postpone the motivation for this requirement to section [ normalisation of the mode functions ] . for the moment , let us simply normalise accordingly for convenience . we thus find : @xmath161 where we have chosen the unobservable phase factor in accordance with the uv limit ( [ uvlimit ] ) . we use some well - known identities involving hankel functions which we list for convenience in appendix [ properties of hankel functions ] . concluding , we find that the vacuum solutions to equation ( [ diraceq3 ] ) neglecting derivatives of @xmath9 and treating the quotient @xmath145 as time independent are given by : [ diraceq6sols ] @xmath162 as we send @xmath163 and approach de sitter space , our solution is in accordance with @xcite , as can be seen from ( [ order ] ) . in this solution we have tacitly assumed that @xmath164 such that @xmath165 , equivalent to an accelerated expansion . if however @xmath166 or @xmath167 , which corresponds to a decelerating universe , one should simply use relations ( [ hankelprops8][hankelprops9 ] ) in appendix [ properties of hankel functions ] . the case @xmath168 requires special attention , see e.g. @xcite for the analogous scalar propagator case . let us now turn our attention to finding the solution for the second mode function corresponding to the excited state . the procedure is completely analogous : we again let @xmath169 and match to the appropriate uv expansion involving the second solution @xmath170 . we normalise the solution according to ( [ normalisation2 ] ) and choose the phase factor in accordance with the corresponding uv limit . we can thus construct the total solution by linear superposition : [ diraceqtotalsol1 ] @xmath171 where : [ diraceqtotalsol2 ] @xmath172 note that we have omitted all @xmath132 dependence in ( [ diraceqtotalsol1 ] ) for notational convenience . note finally that the two fundamental solutions ( [ diraceqtotalsol2 ] ) in @xmath173 are not complex conjugates of each other , unlike for example for the scalar field case . however , the second solution @xmath174 can be obtained from complex conjugation of the fundamental solutions comprising @xmath175 , as can clearly be seen in equation ( [ diraceqtotalsol1 ] ) above . for the moment , we leave the normalisation constants @xmath176 and @xmath177 undetermined . appreciate we are nevertheless free to normalise the fundamental solutions to unity . of course we will determine the normalisation constants shortly by applying our analysis performed in subsection [ spinorial normalisation conditions ] . but first , we need to solve for the anti - particle mode functions . let us stress again that the other spinor @xmath178 can not contain any new degrees of freedom because dirac s equation is linear . we only have to determine the connection with @xmath97 . when we expand according to ( [ fermiondecompose2 ] ) the coupled linear differential equations @xmath137 and @xmath138 obey are identical to ( [ diraceq1 ] ) . therefore , we can analogously define : @xmath179 we are rewarded for our extensive discussion of chirality and helicity . the solutions of @xmath180 are identical to ( [ diraceqtotalsol1 ] ) , where we should only denote the coefficients in this solution differently : [ diraceqtotalsol5 ] @xmath181 the functions appearing in this equation are given by ( [ diraceqtotalsol2 ] ) . we need to determine how the coefficients @xmath176 and @xmath177 of @xmath182 relate to @xmath183 and @xmath184 of @xmath180 . moreover , we need to find the correct normalisation condition for @xmath176 and @xmath177 . inspired by our discussion in [ spinorial normalisation conditions ] , we impose two normalisation conditions on the spinors : a consistent canonical quantisation ( [ consistentquantisation ] ) and orthogonality between particle and anti - particle spinors ( [ orthogonality ] ) . after quite some work the , naively , 10 constraint equations in @xmath45 in ( [ consistentquantisation ] ) yield the following 3 conditions on our coefficients : [ consistentquantisation2 ] @xmath185 the orthogonality condition ( [ orthogonality ] ) yields one more condition : @xmath186 the solution is : [ consistentquantisation3 ] @xmath187 the last line relates the phases of our normalisation constants , e.g. : @xmath188 denotes the phase of @xmath183 . moreover , from ( [ consistentquantisation2 ] ) we can derive : @xmath189 ) results in a change of the right - hand side of equations ( [ consistentquantisation2a ] ) , ( [ consistentquantisation3d ] ) and ( [ consistentquantisation3e ] ) . normalising the fundamental solutions to unity as in ( [ normalisation2 ] ) is particularly convenient as it allows for a particle interpretation of @xmath190 and @xmath191 . the normalisation constants provide us with three physical degrees of freedom . of course @xmath192 is physical which determines @xmath193 through ( [ consistentquantisation3d ] ) . hence , @xmath194 is physical just as in the scalar field case . since @xmath195 is a phase that can be removed without physical consequences , the phase relation ( [ consistentquantisation3c ] ) determines @xmath188 in terms of the third physical phase @xmath196 . note finally that one of the two phase relations ( [ consistentquantisation2c ] ) and ( [ consistentquantisation2d ] ) is redundant in this part of the analysis . a final remark is in order . to derive ( [ consistentquantisation2 ] ) we needed the explicit form of @xmath102 in equation ( [ helicity1 ] ) , which is only valid in @xmath45 . however , we analytically continue this result to arbitrary dimension and one naturally finds ( [ consistentquantisation2 ] ) . using the solutions for @xmath197 and @xmath198 given in ( [ diraceqtotalsol1 ] ) and ( [ diraceqtotalsol5 ] ) respectively , we recall relations ( [ uplusminus ] ) and ( [ nuplusminus ] ) to find : [ summary1 ] @xmath199 \phantom{1111 } \label{summary1a } \\ \chi_{\mathrm{\scriptscriptstyle{r}},h}(k,\eta ) & = & \frac{1}{\sqrt{2}}\left [ c\ { \mathcal{h } + h \tilde{\mathcal{h}}^{\ast } \ } + d\ { \tilde{\mathcal{h } } - h \mathcal{h}^{\ast } \ } \right ] \label{summary1b } , \phantom{1111}\end{aligned}\ ] ] and moreover : @xmath200\phantom{1111 } \label{summary1c } \\ \nu_{\mathrm{\scriptscriptstyle{r}},h}(k,\eta ) & = & \frac{1}{\sqrt{2}}\left [ f\ { \mathcal{h } + h \tilde{\mathcal{h}}^{\ast } \ } + g\ { \tilde{\mathcal{h } } - h \mathcal{h}^{\ast } \}\right ] , \phantom{1111}\label{summary1d}\end{aligned}\ ] ] where the fundamental solution are given in equation ( [ diraceqtotalsol2 ] ) . the normalisation conditions ( [ consistentquantisation3 ] ) determine @xmath183 and @xmath184 in terms of @xmath176 and @xmath177 . moreover , we have @xmath201 . having discussed the fermionic mode functions in some detail , we can turn our attention to solving the feynman propagator for fermions . we insert the rescaled spinors , expanded in terms of creation and annihilation operators , into the formal definition of the feynman propagator ( [ feynmanpropagator ] ) , which yields : @xmath202 where @xmath203 and @xmath80 . the reader can easily see that we have to make use of the solutions ( [ summary1 ] ) above to calculate the 16 matrix elements of the propagator . let us outline this rather cumbersome calculation in seven steps . we recall the form of e.g. the massless fermionic propagator ( [ fermionpropmassless2 ] ) and realise we have to extract an operator acting on @xmath68 , rather than on @xmath69 . the reader can easily see that the 16 matrix elements of the propagator contain products of two hankel functions . some of these products involve hankel functions at the same order , but others encompass hankel functions where the order differs . since we can only easily integrate products of hankel functions of equal order we need to transform precisely the latter products by making use of identities ( [ hankelprops ] ) , e.g. : @xmath204 where @xmath205 or @xmath206 applies when @xmath144 equals 1 or 2 , respectively . clearly , the `` extracted operator '' is a function of @xmath56 . we rewrite the hankel functions in terms of macdonald functions . by making use of @xcite we find : [ hankelmacdonald ] @xmath207 where @xmath208 is a macdonald function . the emerging matrix structure of equation ( [ feynmanpropagator2 ] ) is most easily uncovered by evaluating its @xmath32-matrix constituents successively and writing the result in terms of pauli matrices . recalling equations ( [ kvector1 ] ) and ( [ kvector2 ] ) we derive : @xmath209 where the @xmath206 or @xmath205 sign applies depending on whether one deals with the @xmath210 or @xmath211 contribution in ( [ feynmanpropagator2 ] ) , respectively . the last step is possible because only the exponent depends on @xmath212 . note finally that the index @xmath213 in equation ( [ convertingderivatives ] ) is supposed to run over ( flat ) spatial indices exclusively . the @xmath214 independent contribution to the propagator can thus be pulled out of the fourier integral . these matrices simplify considerably when we consider the following identities : @xmath215 clearly , we recognise the complex conjugate of the dirac operator and the @xmath216 structure we are familiar with from the de sitter propagator . let us present an intermediate result that already captures much of the final structure of our propagator . for brevity , let us only consider the contribution to the propagator arising from the first @xmath66-function in the propagator , i.e. : when @xmath217 . this part of the feynman propagator reads : @xmath218 \nonumber , \end{aligned}\ ] ] where : @xmath219 and where : @xmath220 the contribution to the propagator when @xmath221 is given by the same expression , where we only need to replace the normalisation constants @xmath176 and @xmath177 by @xmath183 and @xmath184 , respectively . the complicated structure of the propagator in spinor space has dramatically simplified by extracting the appropriate operators in position space , which could already be expected from the propagator in de sitter spacetime . we need to perform the fourier integrals over the functions @xmath222 and @xmath223 next . we use : @xmath224 which is valid for any function @xmath225 that depends solely on @xmath226 . here , @xmath227 is a bessel function of the first kind . the reader can easily verify that all the fourier integrals we need to perform are of the following form : @xmath228 where the arguments of the macdonald functions are purely imaginary : [ fourierintegral2 ] @xmath229 we perform these integrals in appendix [ fourier transforming hankel functions ] by making use of @xcite . the result of this integral can be expressed in terms of the gauss hypergeometric function @xmath230 . the @xmath66-functions in the propagator in combination with the arguments of the hypergeometric functions allow us to rewrite these arguments in terms of the @xmath231-functions ( [ y ] ) we introduced in section [ properties of flrw spacetimes ] . a well - known fact in minkowski or de sitter spacetime extends to our analysis in flrw spacetimes : the small real contribution needed to make the fourier integrals convergent determines the @xmath232 pole prescription of the various contributions to the propagator . the final result for the constant @xmath9 and constant @xmath233 flrw fermion propagator reads : @xmath234 \,,\ ] ] where : @xmath235 where the @xmath231-functions are given in equation ( [ y ] ) and @xmath233 in ( [ zeta ] ) . moreover , to write the propagator in terms of the @xmath231-functions we introduced the following notation : @xmath236 one can think of @xmath237 as an antipodal coordinate ( see e.g. : @xcite ) . note that we extract an operator in ( [ fermionpropagator ] ) that does not depend on @xmath69 . however , there is nothing special about the @xmath68 leg of the propagator . keeping remark 1 above in mind , we know that we could have equally well written the propagator ( [ fermionpropagator ] ) in terms of an operator acting on @xmath69 by replacing : @xmath238 this propagator would then have satisfied the dirac equation ( [ fermionprop ] ) in terms of @xmath69 . as a first check of this result , note that if we send @xmath239 and @xmath240 , the massive fermionic flrw propagator ( [ fermionpropagator ] ) correctly yields the massless fermionic flrw propagator earlier derived in equation ( [ fermionpropmassless2 ] ) . again , in order to derive this result we needed the explicit form of @xmath102 given by ( [ helicity1 ] ) . the operator in ( [ fermionpropagator ] ) has thus been constructed in @xmath45 . however , we can analytically continue again to arbitrary dimensions . finally , let us study the ir behaviour of our propagator . note that the coincidence propagator ( in fourier space ) is ir finite due to the pauli exclusion principle . in position space , note that the ir limit of all the y - functions in equation ( [ y ] ) is : @xmath241 where the plus or minus sign applies for infinite spacelike or timelike separation , respectively . if we recall the following well - known relation : @xmath242 we can easily extract the leading order ir behaviour of the propagator in equation ( [ fermionpropagator2 ] ) : @xmath243 note that for the spacelike ir , all y - functions behave identically and the result above can be simplified further . for the timelike ir , the @xmath232-prescription in equation ( [ y ] ) dictates how to take the powers of the y - functions . of course , one has to insert equation ( [ fermionpropagator2ir ] ) above into ( [ fermionpropagator ] ) to obtain the full propagator in the ir limit . the fermionic propagator ( [ fermionpropagator ] ) is analogous to the scalar and graviton propagator in universes with constant deceleration parameter @xmath0 derived in @xcite . unlike in the case of scalars and gravitons , where quite generically the bunch - davies vacuum is ir divergent , the fermionic propagator is ir finite due to the pauli exclusion principle which forbids an accumulation of fermions in the deep ir . the ir behaviour for the scalar and graviton are important , as it can comprise secular effects ( that grow as a power in time ) in for example @xmath244 ( see : @xcite ) and @xmath245 ( see : @xcite ) . mathematically speaking , pauli blocking is enforced by the imaginary @xmath246 in equation ( [ diraceq4 ] ) , which ultimately carries through in the complex indices of the hypergeometric functions in the propagator ( [ fermionpropagator2 ] ) . having derived the fermionic flrw propagator , let us verify whether it correctly reduces to the fermionic de sitter propagator , which is a well - known result in the literature . the fermion propagator in de sitter spacetime reads @xcite : @xmath247 \nonumber,\end{aligned}\ ] ] where the superscript @xmath248 denotes de sitter spacetime and where : @xmath249 here , @xmath231 is shorthand for @xmath250 . if we send @xmath163 and set @xmath251 and @xmath252 , equation ( [ fermionpropagator ] ) above correctly reproduces the de sitter result . moreover , using : @xmath253 note that the massless limit of ( [ fermionpropmassdesitter ] ) indeed corresponds to ( [ fermionpropmassless2 ] ) in de sitter space . we conclude that the fermionic propagator in flrw spacetimes correctly reduces to the known cases in existing literature . as a simple application for the propagator derived in section [ the feynman propagator ] , we calculate the one loop contribution to the effective action . this is important to study the impact of fermions on the evolution of the background spacetime and of scalar fields coupled to these fermions . we renormalise using the minimal subtraction dimensional renormalisation technique . the one loop effective action formally reads : @xmath254 \\ & = & \int^{m}\ ! \mathrm{d}\bar{m } \ , \mathrm{tr } \left [ \sqrt{-g } \ , i s_{\mathrm{f}}^{ab}(x,\tilde{x})\right ] \nonumber\,.\end{aligned}\ ] ] here , the trace is both over spatial and spinorial indices . we have been able to evaluate the logarithm at the expense of losing all mass independent contributions to the effective action . the effective action for a massless fermion is completely fixed by the trace anomaly which this field is known to exhibit ( see for example @xcite ) . the trace anomaly can be generated from a finite non - local effective action ( see e.g. : @xcite ) . it can alternatively be generated from an infinite but local effective action @xcite . we will henceforth neglect all contributions from the trace anomaly to the effective action in this paper because it has already been extensively discussed in the literature . we treat the covariant derivative according to ( [ covariantderivative2 ] ) and we can easily expand the hypergeometric functions in equation ( [ fermionpropagator2 ] ) at coincidence . in dimensional regularisation , the @xmath12 dependent powers in this expansion do not contribute . these are generated for the @xmath255 and @xmath256 contributions to ( [ fermionpropagator2 ] ) by the first term in : @xmath257 moreover , when we trace over spacetime indices note there is no contribution at all from the derivative term hitting one of the hypergeometric functions : @xmath258 here , @xmath259 is the integer valued coefficient of the taylor expansion of a hypergeometric function . a similar identity holds for @xmath256 . likewise , for @xmath260 or @xmath261 , we have e.g. : @xmath262 note that both @xmath260 and @xmath261 depend on the antipodal coordinate @xmath237 in equation ( [ fermionpropagator ] ) . at coincidence , we also use : @xmath263 the result is : @xmath264 let us compare this result with the known calculations in the literature in de sitter spacetime ( @xmath265 ) and in the vacuum ( @xmath266 ) . note the two omissions in the candelas and raine effective action @xcite . when tracing over spinor indices , we have : @xmath267 and this trace does not , as is apparent from @xcite , equal @xmath268 because a spinor in @xmath12 spacetime dimensions has @xmath269 degrees of freedom ( see e.g. : @xcite ) . indeed , this has already been noted by @xcite . moreover , the effective action in @xcite misses a factor of 2 stemming from the separate particle and anti - particle contributions , which has been corrected for in e.g. @xcite . the reader can easily verify that ( [ effectiveactionfullresult ] ) , taking these errors into account agrees with @xcite in de sitter spacetime and in the vacuum . in the spirit of dimensional regularisation , we now expand around @xmath45 . we can use the familiar result : @xmath270 + \mathcal{o}(\tilde{\epsilon}^{2})\,,\ ] ] where @xmath271 and where @xmath272 is the digamma function defined by @xmath273 . anticipating the form of the counterterms we will add shortly , we introduce a scale @xmath274 by : @xmath275 \nonumber , \end{aligned}\ ] ] we we neglected @xmath276 contributions . finally , we recall the familiar result : @xmath277 where @xmath278 is the euler mascheroni constant . we thus arrive at : @xmath279 \right ) \label{effectiveactionexpansiond4 } \\ & & \ ! + \ ! \int^{m}\!\!\mathrm{d}\tilde{m}\ ! \left(\tilde{m}^{3}\!+\!\tilde{m}h^{2}(1\!-\!\epsilon)^{2 } \right)\!\left [ \psi(2\!-\!i\tilde{\zeta})\!+\!\psi(2\!+\!i\tilde{\zeta})\ ! \right]\ ! \bigg]\ ! \nonumber \\ & & \ ! + 2 h^{3}(1-\epsilon)^{3 } \mathrm{im}\left(c d^{\ast}\right ) \ ! \mathrm{d}\tilde{m } \ , \gamma(2+i \tilde{\zeta } ) \gamma(2-i\tilde{\zeta})\bigg\ } \nonumber \!.\end{aligned}\ ] ] here , @xmath280 denotes the imaginary part of @xmath281 . the one loop backreaction contains a divergence when @xmath45 . however , appreciate that the contribution to the effective action multiplying the mixed coefficients is finite in @xmath45 . anticipating the renormalisation procedure in the next section , we need the tree level , i.e. : classical , friedmann equations of motion responsible for driving the expansion of the universe . the system of interest is thus given by : @xmath282 \right\ } \ , . \nonumber\end{aligned}\ ] ] here , @xmath283 represents the rescaled newton constant and @xmath248 denotes the cosmological constant . the reader can easily recognise the usual einstein - hilbert action on the first line . we introduce a new scalar field @xmath284 , responsible for the dynamics of the universe . by an appropriate choice of the potential @xmath285 , a scalar field can mimic any mixture of fluids relevant for the evolution of our universe ( see e.g. @xcite ) . the last two lines in equation ( [ actiontotalgravity ] ) above are identical to ( [ actiontotal ] ) and contain the fermion field and the scalar field @xmath286 generating the mass of the fermion through a yukawa coupling , as in ( [ fermionmass ] ) . we write : [ backgroundfields1 ] @xmath287 where we assume the background fields @xmath288 and @xmath289 to be homogeneous . moreover , we assume that , such that @xmath290 drives the dynamics of the universe as stated above . the classical friedmann equations of motion are : [ treelevelfriedmann1 ] @xmath291 and moreover , the classical scalar field equation of motion reads : @xmath292 we can thus derive the following identities : [ treelevelfriedmann2 ] @xmath293 here , we have used assumption ( [ epsilon ] ) . in order to derive equation ( [ effectiveactionexpansiond4 ] ) , we have assumed a constant deceleration @xmath9 , as in equation ( [ epsilon ] ) . we now promote this constant to a dynamical quantity : @xmath294 we will motivate this step shortly . the effective action ( [ effectiveactionexpansiond4 ] ) now contains divergences in both @xmath295 and @xmath296 that we ought to cancel by an appropriate counterterm action . it is not possible to identify local covariant counterterms of curvature invariants only that remove the singularities in both of these quantities at the level of the effective action . we can use the tree level equations of motion derived above in ( [ treelevelfriedmann2 ] ) to renormalise the effective action , which is first order in @xmath96 . by making use of an infinitesimal field redefinition , one can show the following : substituting the zeroth order equation of motion into the first order contribution to the action yields , up to one loop , the same equation of motion as directly varying the action . we thus add the following counterterm action : @xmath297 the form of the counterterm action is unique , which can be shown by making use of dimensional analysis and by requiring that only two time derivatives can act on @xmath288 or @xmath298 for stability . these conditions limit us to the counterterms above and possibly @xmath299 . however , one can easily verify that the latter term does not have the correct form required to cancel the divergences in ( [ effectiveactionexpansiond4 ] ) . we recall relation ( [ fermionmass ] ) . we need to expand the various terms in the counterterm action above around @xmath45 , by making use of @xmath300 and equation ( [ treelevelfriedmann2 ] ) : [ counterterms1 ] @xmath301 where we neglected all @xmath302 contributions in the equations above . we can now easily solve for the coefficients in the counterterm action : [ counterterms2 ] @xmath303 where the divergent coefficients in @xmath45 are fixed to cancel the divergences occurring in ( [ effectiveactionexpansiond4 ] ) , and where e.g. @xmath304 is a finite but arbitrary coefficient of the counterterm action . the renormalised one loop effective action thus reads : @xmath305 \right ) - \frac{1}{12}m^{2}h^{2}(7 - 5\epsilon ) \nonumber \\ & & \qquad\qquad\qquad\qquad\qquad\quad + \int^{m}\!\mathrm{d}\tilde{m } \left(\tilde{m}h^{2}(1-\epsilon)^{2 } + \tilde{m}^{3 } \right ) \left [ \psi(2-i\tilde{\zeta})+\psi(2+i\tilde{\zeta } ) \right ] \bigg ] \label{effectiveactionrenormalised } \\ & & \qquad\qquad\quad + 2 h^{3}(1-\epsilon)^{3 } \mathrm{im } \left ( c d^{\ast } \right ) \int^{m}\ ! \mathrm{d}\tilde{m } \ , \gamma(2+i \tilde{\zeta } ) \gamma(2-i\tilde{\zeta})\bigg\ } \nonumber \\ & & + \int \mathrm{d}^{4}\!x \sqrt{-g } \big(c_{1}^{\mathrm{f } } \phi^{2 } r + c_{2}^{\mathrm{f } } \phi^{2}\frac{\partial^{2}v } { \partial\varphi^{2}}(\varphi_{0 } ) - c_{3}^{\mathrm{f } } \kappa \phi^{2 } g^{\mu\nu}\partial_{\mu}\varphi \partial_{\nu}\varphi + c_{4}^{\mathrm{f } } \phi^{4 } \big ) \nonumber\,.\end{aligned}\ ] ] the effective action above comprises the one loop backreaction of fermions . this term gives rise to additional physical corrections to the ( classical ) friedmann equations , thus obtaining the so - called quantum corrected friedmann equations . we will study its effect on the evolution of our universe and its impact on the evolution of the scalar field to which the fermions are coupled in a future publication . in this equation we have neglected all contributions to the one loop effective action arising from the trace anomaly . if we expand equation ( [ effectiveactionrenormalised ] ) to first order in @xmath9 , we can find out how large the first order correction to the de sitter result is . we have : @xmath306\right\ } - \frac{1}{2}m^{4 } \label{effectiveactionrenormalisedfirstorder } \\ & & \quad + \int^{m}\!\mathrm{d}\tilde{m } i \left(\tilde{m}^{2}h + \frac{\tilde{m}^{4}}{h } \right ) \big [ \psi'(2+i\tilde{m}/h)-\psi'(2-i\tilde{m}/h ) \big]-2\tilde{m}h^{2}\big [ \psi(2+i\tilde{m}/h)+\psi(2-i\tilde{m}/h ) \big ] \bigg ] \nonumber \\ & & \quad + 2 h^{3 } \mathrm{im } \left ( c d^{\ast } \right ) \int^{m}\ ! \mathrm{d}\tilde{m } \ , \gamma(2+i \tilde{m}/h ) \gamma(2-i\tilde{m}/h ) \left[i\frac{\tilde{m}}{h}\big ( \psi(2+i\tilde{m}/h)-\psi(2-i\tilde{m}/h ) \big)-3 \right ] \bigg\ } \nonumber + \mathcal{o}\left(\epsilon^{2}\right),\end{aligned}\ ] ] where the prime denotes a derivative with respect to the argument , and the superscript @xmath248 represent the well - known de sitter result , which trivially follows from ( [ effectiveactionrenormalised ] ) by letting @xmath265 . let us at this point however already consider the @xmath307 limit , to extract the physical behaviour of the effective potential resulting from ( [ effectiveactionrenormalised ] ) . note firstly that : @xmath308 where @xmath309 denotes the one loop effective potential for @xmath286 . in order to evaluate the integrals over @xmath310 occurring in equation ( [ effectiveactionrenormalised ] ) , we need the following asymptotic expansions ( @xmath311 ) : @xmath312 and also : @xmath313 & = & \ln [ \zeta^{2 } ] + \frac{13}{6}\frac{1}{\zeta^{2 } } \\ & & - \frac{119}{60}\frac{1}{\zeta^{4 } } + \mathcal{o}\left(\frac{1}{\zeta^{6}}\right ) \nonumber \,.\end{aligned}\ ] ] keeping the leading order terms of the effective potential ( [ effectivepotential1 ] ) in this limit only , we find from ( [ effectiveactionrenormalised ] ) and ( [ limitbigmass1][limitbigmass2 ] ) : @xmath314 \\ & & \!+ \frac{\mathrm{im}(c d^{\ast})}{\pi^{2 } } g_{\mathrm{\scriptscriptstyle{y}}}^{3 } \phi^{3 } h(1\!-\!\epsilon ) \exp\left[{-\pi \frac{g_{\mathrm{\scriptscriptstyle{y } } } equation ( [ fermionmass ] ) has been used . without studying the dynamics resulting from this asymptotic effective potential , let us briefly consider its stability properties . clearly , this potential is unstable in the vacuum ( @xmath266 ) because the effective potential is unbounded from below in this case . this wild behaviour has already been recognised in de sitter spacetime @xcite . an interesting phenomenon can be observed precisely when @xmath315 . this is realised for example in a thermal state with a temperature @xmath316 much larger than the uv cutoff of the theory . in this case , the asymptotic effective potential is exponentially suppressed and its ( small ) contributions can either be positive or negative . let us make a few general remarks . consider quantum field theories in de sitter spacetime . the hubble parameter @xmath4 in de sitter spacetime is a constant . an effective action for fermions or scalars in de sitter spacetime contains divergences for example proportional to a power of @xmath4 ( see e.g. : @xcite ) . the same statement holds for divergences in the stress energy tensor in for example the trace anomaly literature ( see e.g. : @xcite ) . at this point , one can interpret the results differently : 1 . the hubble parameter can quite conservatively be treated as a _ constant_. the argument for this case is simple : throughout the calculation one simply assumed a constant hubble parameter . the hubble parameter is promoted to a _ dynamical _ quantity : @xmath317 a reason for supporting this case would be the following : when @xmath8 one would find the de sitter result as a leading order contribution in each time interval where the hubble parameter varies only adiabatically slowly , a spacetime also known as locally de sitter spacetime . depending on which of the two interpretations one adheres to , one would renormalise differently . if one advocates the first point of view , all terms of e.g. an effective action merely contribute to the renormalised cosmological constant . if , however , one prefers the second interpretation , one adds ( dynamical ) curvature invariants in order to cancel these divergences in a covariant manner . we observe the following . in the effective action ( [ effectiveactionrenormalised ] ) above , the hubble parameter is a dynamical quantity . hence , covariant counterterms added to cancel uv divergences can and should be treated dynamically . this allows us to study the backreaction of fermions in a dynamical manner . albeit strictly speaking not allowed to relax the assumption @xmath318 in de sitter spacetime , we conclude that promoting the hubble parameter to a dynamical quantity @xmath319 has been a useful approach . when in our calculation we keep @xmath9 constant as assumed , result ( [ effectiveactionfullresult ] ) favours the second interpretation above . let us now return to our calculation to see what the above considerations can tell us about our effective action ( [ effectiveactionfullresult ] ) . we face the following two options : 1 . we treat @xmath9 as a _ constant_. 2 . we promote @xmath9 to a _ dynamical _ quantity , as already stated in ( [ epsilonpromotion ] ) . we could call this a spacetime with an almost constant deceleration . despite the fact that throughout the calculation we have assumed that @xmath9 is a constant , it will come as no surprise that we argue in favour of the latter option for two reasons : if one would allow @xmath9 to change , one would probably find our main result for the effective action before renormalisation ( [ effectiveactionfullresult ] ) in each time interval where @xmath9 changes only very slowly . secondly , the latter option is the generalisation of the confirmed de sitter case above . an obvious disadvantage of our choice is that we can only renormalise on - shell . consider the divergences in ( [ effectiveactionexpansiond4 ] ) . as already mentioned , it is not possible to identify local covariant counterterms of curvature invariants only that remove all singularities in @xmath295 and @xmath296 at the level of the effective action . therefore , we relied on an on - shell renormalisation technique and inserted the zeroth order ( tree level ) friedmann equations of motion into the effective action at first order in @xmath96 . finally , let us try to comfort the reader who would rather keep @xmath9 constant , but who would simultaneously like to understand how to improve on his final result for the renormalised effective action . in this case , the only appropriate counterterms to renormalise ( [ effectiveactionexpansiond4 ] ) available to us are the following : @xmath320 we can then renormalise the theory as usual would have to be considered separately , for @xmath321 in a radiation dominated universe . ] , with : [ countertermscoefficientsalt ] @xmath322 we could now study mode mixing . in their paper , tsamis and woodard study mode mixing for a massless minimally coupled scalar field in constant @xmath9 flrw spacetimes . the basic strategy is as follows : in each possibly infinitesimally small time interval @xmath9 can be well approximated by a constant . the mode functions in two adjacent time intervals can be related by imposing two conditions : ( i ) continuity of the mode functions and ( ii ) continuity of their first derivatives . hence , the mode functions at some later time @xmath56 can be obtained from the initial condition at @xmath323 by means of a transfer matrix . ] . one could simply allow @xmath9 and @xmath145 to vary in time and choose ( if necessary infinitesimally small ) time intervals in which @xmath9 and @xmath145 vary only slowly . subsequently , one can match the solutions in two neighbouring intervals and study mode mixing in the spirit of @xcite . in this case , this procedure would imply that we would have to renormalise for each time interval separately , for the coefficients in equation ( [ countertermscoefficientsalt ] ) depend on @xmath9 . this is not surprising : a sudden jump , even infinitesimally small , in for example @xmath145 is not a physical phenomenon and would call for non - local counterterms ( see e.g. : @xcite ) . in this paper , we have constructed the fermionic propagator in realistic flrw spacetimes with constant deceleration parameter @xmath0 . moreover , we assumed that @xmath145 is also constant , which can be realised in yukawa theory by means of a scalar field for which @xmath11 . we have derived the propagator both in the ( trivial ) massless case ( [ fermionpropmassless2 ] ) and in the massive case ( [ fermionpropagator ] ) . two pillars support our derivation . we split the fermionic degrees of freedom into a direct product of chirality and helicity eigenspinors . we normalise the spinors using a consistent canonical quantisation and we require orthogonality of particle and anti - particle spinors . moreover , we verify that upon sending @xmath9 to zero , we recover the known vacuum - to - vacuum de sitter results in the literature . we calculate the one loop effective action induced by fermions using our propagator . the one loop backreaction arises from integrating out a free , quadratic fermion field and , using dimensional regularisation , this generates a correction to the ( classical ) friedmann equations . this effective action allows for the first time for a dynamical interpretation of @xmath4 such that the study of dynamical backreaction either on the background spacetime or on scalar fields coupled to these fermions is within reach . * acknowledgements * jfk thanks gerben stavenga , shun - pei miao and timothy budd for useful suggestions . the authors thank gerard t hooft , gleb arutyunov and tomas janssen and acknowledge financial support from fom grant 07pr2522 and by utrecht university . in section [ particle mode functions ] we used the following well - known identities involving hankel functions : [ hankelprops ] @xmath324 the wronskian of two hankel functions reads : @xmath325 & = & -\frac{4 i}{\pi z } \label{hankelprops5}\,.\end{aligned}\ ] ] moreover , we made use of the following recurrence relation : @xmath326 consider the following integral equation : @xmath327 here , @xmath328 is a bessel function of the first kind and @xmath208 is a macdonald function . note that for all integrals in ( [ fourierintegral ] ) both @xmath63 and @xmath64 are purely imaginary . the requirements from @xcite are such that : we choose @xmath331 to make the integral convergent . we now use equation ( 6.578.10 ) from @xcite to find : @xmath332 where : @xmath333 here , @xmath334 represents the associated legendre function which we can rewrite in terms of a gauss hypergeometric function @xmath335 by means of equations ( 8.702 ) and ( 9.131.1 ) of @xcite : @xmath336 upon taking @xmath337 we find : @xmath338 \,.\ ] ] finally note that we can further simplify the argument of the hypergeometric function in equation ( [ i6 ] ) to : @xmath339 let us finally stress that we have introduced @xmath331 to make the integral convergent . it is however of significant physical importance for it is the same @xmath340 that dictates the @xmath232 pole prescription in the propagator ( see equations ( [ fermionpropagator][fermionpropagator2 ] ) and ( [ y ] ) ) . this should not come as a surprise : we observe the same behaviour in e.g. minkowski and de sitter spacetimes . 99 l. h. ford , quantum instability of de sitter space - time , phys . rev . d * 31 * ( 1985 ) 710 . i. antoniadis and e. mottola , graviton fluctuations in de sitter space , j. math . phys . * 32 * ( 1991 ) 1037 . n. c. tsamis and r. p. woodard , the structure of perturbative quantum gravity on a de sitter background , commun . * 162 * ( 1994 ) 217 . n. c. tsamis and r. p. woodard , quantum gravity slows inflation , nucl . b * 474 * ( 1996 ) 235 [ arxiv : hep - ph/9602315 ] . n. c. tsamis and r. p. woodard , one loop graviton self - energy in a locally de sitter background , phys . d * 54 * ( 1996 ) 2621 [ arxiv : hep - ph/9602317 ] . l. r. w. abramo , r. h. brandenberger and v. f. mukhanov , the energy - momentum tensor for cosmological perturbations , phys . d * 56 * ( 1997 ) 3248 [ arxiv : gr - qc/9704037 ] . f. finelli , g. marozzi , g. p. vacca and g. venturi , adiabatic regularization of the graviton stress - energy tensor in de sitter space - time , phys . d * 71 * ( 2005 ) 023522 [ arxiv : gr - qc/0407101 ] . t. prokopec , a solution to the cosmological constant problem , arxiv : gr - qc/0603088 . t. prokopec , n. c. tsamis and r. p. woodard , stochastic inflationary scalar electrodynamics , annals phys . * 323 * , 1324 ( 2008 ) [ arxiv:0707.0847 [ gr - qc ] ] . a. bilandzic and t. prokopec , quantum radiative corrections to slow - roll inflation , phys . d * 76 * ( 2007 ) 103507 [ arxiv:0704.1905 [ astro - ph ] ] . t. janssen and t. prokopec , a graviton propagator for inflation , class . * 25 * ( 2008 ) 055007 [ arxiv:0707.3919 [ gr - qc ] ] . t. janssen , s. p. miao and t. prokopec , graviton one - loop effective action and inflationary dynamics , arxiv:0807.0439 [ gr - qc ] . t. janssen and t. prokopec , implications of the graviton one - loop effective action on the dynamics of the universe , arxiv:0807.0447 [ gr - qc ] . t. m. janssen , s. p. miao , t. prokopec and r. p. woodard , infrared propagator corrections for constant deceleration , class . * 25 * ( 2008 ) 245013 [ arxiv:0808.2449 [ gr - qc ] ] . p. candelas and d. j. raine , general relativistic quantum field theory - an exactly soluble model , phys . d * 12 * ( 1975 ) 965 . b. allen and c. a. lutken , spinor two point functions in maximally symmetric spaces , commun . * 106 * ( 1986 ) 201 . r. camporesi , the spinor heat kernel in maximally symmetric spaces , commun . * 148 * ( 1992 ) 283 . miao and r. p. woodard , the fermion self - energy during inflation , class . * 23 * ( 2006 ) 1721 [ arxiv : gr - qc/0511140 ] . s. p. miao and r. p. woodard , leading log solution for inflationary yukawa , phys . d * 74 * ( 2006 ) 044019 [ arxiv : gr - qc/0602110 ] . e. t. tomboulis , dynamically adjusted cosmological constant and conformal anomalies , nucl . b * 329 * ( 1990 ) 410 . i. antoniadis and e. mottola , 4-d quantum gravity in the conformal sector , phys . d * 45 * ( 1992 ) 2013 . i. antoniadis , dynamics of the conformal factor in 4-d gravity , [ arxiv : hep - th/9211055 ] . i. antoniadis , p. o. mazur and e. mottola , fractal geometry of quantum spacetime at large scales , phys . b * 444 * ( 1998 ) 284 [ arxiv : hep - th/9808070 ] . h. salehi and y. bisabr , conformal anomaly and large scale gravitational coupling , int . j. theor . * 39 * ( 2000 ) 1241 [ arxiv : hep - th/0001095 ] . i. antoniadis , p. o. mazur and e. mottola , cosmological dark energy : prospects for a dynamical theory , new j. phys . * 9 * ( 2007 ) 11 [ arxiv : gr - qc/0612068 ] . j. f. koksma and t. prokopec , effect of the trace anomaly on the cosmological constant , phys . d * 78 * ( 2008 ) 023508 [ arxiv:0803.4000 [ gr - qc ] ] . n. a. chernikov and e. a. tagirov , quantum theory of scalar fields in de sitter space - time , annales poincare phys . a * 9 * ( 1968 ) 109 . b. allen , vacuum states in de sitter space , phys . d * 32 * ( 1985 ) 3136 . b. allen and a. folacci , the massless minimally coupled scalar field in de sitter space , phys . rev . d * 35 * ( 1987 ) 3771 . b. allen and t. jacobson , vector two point functions in maximally symmetric spaces , commun . * 103 * ( 1986 ) 669 . n. c. tsamis and r. p. woodard , a maximally symmetric vector propagator , j. math . * 48 * ( 2007 ) 052306 [ arxiv : gr - qc/0608069 ] . b. allen , the graviton propagator in de sitter space , phys . d * 34 * ( 1986 ) 3670 . b. allen and m. turyn , an evaluation of the graviton propagator in de sitter space , nucl . b * 292 * ( 1987 ) 813 . e. g. floratos , j. iliopoulos and t. n. tomaras , tree level scattering amplitudes in de sitter space diverge , phys . b * 197 * ( 1987 ) 373 . a. higuchi and s. s. kouris , the covariant graviton propagator in de sitter spacetime , class . grav . * 18 * ( 2001 ) 4317 [ arxiv : gr - qc/0107036 ] . n. d. birrell and p. c. w. davies , quantum fields in curved space , cambridge monographs on mathematical physics , cambridge university press ( 1982 ) . t. prokopec and e. puchwein , photon mass generation during inflation : de sitter invariant case , jcap * 0404 * ( 2004 ) 007 [ arxiv : astro - ph/0312274 ] . s. weinberg , quantum contributions to cosmological correlations , phys . d * 72 * ( 2005 ) 043514 [ arxiv : hep - th/0506236 ] . j. f. koksma , t. prokopec and g. i. rigopoulos , the scalar field kernel in cosmological spaces , class . * 25 * ( 2008 ) 125009 [ arxiv:0712.3685 [ gr - qc ] ] . i. i. cotaescu , polarized dirac fermions in de sitter spacetime , phys . rev . d * 65 * ( 2002 ) 084008 [ arxiv : hep - th/0109199 ] . b. garbrecht and t. prokopec , fermion mass generation in de sitter space , phys . d * 73 * ( 2006 ) 064036 [ arxiv : gr - qc/0602011 ] . b. de wit and j. smith , field theory in particle physics . 1 , north - holland personal library ( 1986 ) . m. e. peskin and d. v. schroeder , an introduction to quantum field theory , addison - wesley ( 1995 ) . c. itzykson and j. b. zuber , quantum field theory , mcgraw - hill ( 1980 ) . t. prokopec , m. g. schmidt and s. weinstock , transport equations for chiral fermions to order h - bar and electroweak baryogenesis , annals phys . * 314 * ( 2004 ) 208 [ arxiv : hep - ph/0312110 ] . i. s. gradshteyn and i. m. ryzhik , table of integrals , series , and products , academic press ( 1965 ) . a. p. prudnikov , yu . a. brychkov and o. i. marichev , integrals and series , volume 2 , special functions , gordon and breach science publishers ( 1986 ) . e. mottola , particle creation in de sitter space , phys . d * 31 * ( 1985 ) 754 . k. goldstein and d. a. lowe , a note on alpha - vacua and interacting field theory in de sitter space , nucl . b * 669 * , 325 ( 2003 ) [ arxiv : hep - th/0302050 ] . m. b. einhorn and f. larsen , squeezed states in the de sitter vacuum , phys . d * 68 * ( 2003 ) 064002 [ arxiv : hep - th/0305056 ] . t. m. janssen , s. p. miao , t. prokopec and r. p. woodard , the hubble effective potential , arxiv:0904.1151 [ gr - qc ] . t. m. janssen , s. p. miao , t. prokopec and r. p. woodard , paper in preparation . t. inagaki , s. mukaigawa and t. muta , a soluble model of four fermion interactions in de sitter space , phys . d * 52 * ( 1995 ) 4267 [ arxiv : hep - th/9505058 ] . t. inagaki , t. muta and s. d. odintsov , dynamical symmetry breaking in curved spacetime : four - fermion interactions , prog . * 127 * ( 1997 ) 93 [ arxiv : hep - th/9711084 ] . t. padmanabhan , accelerated expansion of the universe driven by tachyonic matter , phys . d * 66 * ( 2002 ) 021301 [ arxiv : hep - th/0204150 ] . n. c. tsamis and r. p. woodard , plane waves in a general robertson - walker background , class . grav . * 20 * ( 2003 ) 5205 [ arxiv : astro - ph/0206010 ] . a. vilenkin and l. h. ford , gravitational effects upon cosmological phase transitions , phys . d * 26 * ( 1982 ) 1231 .

quantum chromodynamics is a perturbative field theory for @xmath10 , corresponding to energies above @xmath11 mev . however , setting up a good perturbation expansion is more difficult than simply working order by order in @xmath12 . the difficulty is that when computing a quantity with more than one scale , logarithms of ratios of those scales appear which invalidate the naive perturbation expansion . for example , in @xmath0 collisions at center - of - mass energy @xmath4 , we might look for the distribution of jets as a function of the invariant mass @xmath13 of the jet . then the differential cross section will have a correction of the form @xmath14 . even for @xmath10 these large logarithms can dominate if @xmath13 is sufficiently small . luckily , these logarithms appear only in certain combinations even at higher order , so that all the terms of the form @xmath15 can be ( re)summed at once . however , understanding which terms will appear , how to resum them , and how to combine the resummed result with fixed order results can be quite complicated . it is the goal of this paper to show how it can be done using effective field theory techniques . in this paper , resummation and matching of event shapes are studied using soft - collinear effective theory ( scet ) @xcite . event shapes are observables which are sensitive to the overall distribution of final state particles , and therefore involve both short and long distance physics . we will consider mainly the event shape @xmath16 where the thrust @xmath17 is defined by @xmath18 summing over all momentum @xmath19-vectors @xmath20 in the event and maximizing over unit 3-vectors @xmath21 . in the threshold region near @xmath22 large logarithms of the form @xmath23 appear at fixed order in perturbation theory . the resummed result will be valid even if @xmath23 is large , as long as @xmath12 and @xmath24 are small . effective field theories provide a systematic approach to resummation . they separate out physics at a hard underlying scale @xmath4 from physics associated with a scale of interest @xmath25 and from even lower scales . at each scale a separate matching calculation is done which is independent of physics at asymptotically lower or higher energy . then the scale dependence is calculated using the renormalization group . in this way , large logarithms can not appear because no two largely separated scales are accessible to the theory at the same time . the soft - collinear effective theory works by separating the degrees of freedom of qcd into soft modes and collinear modes in different directions . the relevant scales are then associated with the observable of interest , such as @xmath26 , or with the large components of the collinear modes . the resummation of event shapes in @xmath0 is not currently of extreme importance phenomenologically . it nevertheless provides a clean arena ( as compared to hadron collisions ) to explore resummation and matching . the scet techniques and most of the formulas we discuss here were originally developed for @xmath27-physics , such as resummation in @xmath28 decays @xcite . they have also been applied to the study of deep - inelastic - scattering near @xmath29 @xcite and to the production of massive jets initiated by top quark decays @xcite . one convenient feature of effective field theories is factorization , which allows us to use objects , such as soft and jet functions , calculated in one process to study another . thus , most of the hard work required for calculation of the distributions we describe here can be extracted from the literature . nevertheless , there are certain features , in particular the nlo matching step , for which event shapes are uniquely illuminating . the breakdown of naive perturbation theory due to the appearance of large logarithms is independent of @xmath12 blowing up and of non - perturbative effects . to emphasize this point , and to simplify the resummed expressions , the running of @xmath12 will be simply turned off ( setting @xmath30 ) for most of this paper . working in the conformal limit removes one scale ( @xmath31 ) from the problem and thus clarifies which large logs are being resummed . it is not hard to turn @xmath32 back on , as will be shown in section [ sec : gen ] . also , we will be including all one - loop results but no two - loop results . thus , our expressions will not contain a complete next - to - leading log resummation , which should also resum the two - loop double logs . there are three steps involved in the calculation of event shapes in scet , associated with three separate scales . at the hard scale @xmath33 , qcd is matched onto scet by demanding parton level matrix elements in the two theories be the same . this can be done at leading order by matching onto an operator @xmath34 with two collinear fields , or at next - to - leading order by matching in addition to an operator @xmath35 with collinear fields in three directions @xcite . the threshold resummation only involves @xmath34 as the matrix elements of @xmath35 vanish near @xmath36 . ( an alternative approach to @xmath37 matching is described in @xcite ) . at the scale @xmath38 , the collinear fields freeze and can be removed from the theory by integrating them out . this results in the a jet function @xmath39 . finally , even though we are interested in a distribution at the scale @xmath40 , the soft degrees of freedom remain relevant down to a scale @xmath41 , after which they too can be integrated out of the theory . the fixed order result will have large logarithms of @xmath42 and @xmath43 , but in the resummed result all these logs are exponentiated and innocuous . before we present the factorization formula and calculate the thrust distribution in scet , we will review the way resummation of thrust is traditionally handled . there are a number of ways to resum event shapes @xcite . since the focus of this work is on comparisons to scet , we will confine our attention to the original approach of catani , trentadue , turnock and webber @xcite , which will be referred to as cttw throughout . moreover , most of the other approaches reduce to @xcite at next - to - leading fixed order in @xmath12 ( nlo ) and to leading log , so no significant loss of generality is sustained . we will find that in the two - jet limit scet also agrees with cttw to the order we are working , although the resummed expressions are not exactly the same . a more significant difference is in the nlo matching . a critical advantage of the effective theory approach is that the finite parts of loop amplitudes are automatically incorporated into the perturbative expressions . for example , the total nlo cross section for @xmath0 is reproduced by combining the finite parts of the hard , jet , and soft functions and a contribution from a finite integral over higher - order operators in scet . this does not necessarily entail less work than in calculating the total cross section through the traditional combination of real and virtual contributions . however , due to factorization , the infrared divergent contributions which are absorbed into the jet and soft functions are universal and thus the could potentially be used for many processes . in this section , we will review some basic results from qcd on thrust , and the resummed expressions presented in @xcite . to begin , consider the parton model description of @xmath0 annihilation . at order @xmath44 , the only process which contributes is @xmath45 . these two jets have no structure and hence the cross section is simply @xmath46 , where @xmath47 is leading order total @xmath48 annihilation cross section . at order @xmath49 , there are two @xmath50 diagrams which contribute @xmath51 where the photon line on the bottom is the @xmath52 current coming in . the differential cross section is @xmath53_{\mathrm{parton } } = \delta ( { s } ) \delta ( t ) + { \bar{\alpha}}\frac{{s}^2 + { t}^2 + 2 { u}}{{s}\,{t } } \label{sigparton}\ ] ] where we have defined @xmath54 and the reduced mandlestam variables are @xmath55 and @xmath56 with @xmath57 . now , for 3-parton events the thrust variable @xmath58 reduces to @xmath59 so that @xmath60_{\mathrm{parton } } & = & \delta ( \tau ) + { \bar{\alpha}}\left [ \frac{2 ( 3 \tau^2 - 3 \tau + 2)}{\tau ( 1 - \tau ) } \log \frac{1 - 2\tau}{\tau } - \frac{3 ( 1 - 3 \tau ) ( 1 + \tau)}{\tau } \right]\\ & = & \delta ( \tau ) + { \bar{\alpha}}\left [ \frac{- 4 \log \tau - 3}{\tau } \right ] + { \bar{\alpha}}d_{\mathrm{fin } } ( \tau ) . \label{partondc}\end{aligned}\ ] ] the second line is written to manifest the singularity structure . the remainder @xmath61 is finite as @xmath62 . instead of the differential thrust distribution , it is useful to work directly with the integrated quantity @xmath63 for small @xmath64 , @xmath65 here we see explicitly the large logarithms @xmath23 and @xmath66 which demand resummation . a useful theoretical trick , used in @xcite and in @xcite , is to employ a hemisphere mass definition . this greatly simplifies the factorization formula in the two - jet limit . the hemisphere momentum @xmath67 is defined to be the sum of the 4-momenta of all the particles in the hemisphere of the left ( right ) jet . then thrust reduces to @xmath68 , and we can calculate it from @xmath69 which we will see has a closed form expression . the traditional approach to resummation which we review here is due to catani et al . the basic idea is that terms of the form @xmath70 come from multiple real collinear and soft emissions . the kinematics of multiple collinear emissions can be modeled by parton branchings , with branching probabilities proportional to the altarelli - parisi splitting functions @xmath71 in the two - jet limit , the corrections due to soft effects can be modeled by imposing angular ordering . all of this follows from the coherent - branching algorithm developed in @xcite . in fact , more general methods are available for handling soft emissions ( for example , @xcite ) . but , to nlo for event shapes like thrust , the methods set up in @xcite are sufficient to compare to scet . in the two - jet limit , the factorized expression for the differential cross section in cttw is @xmath72_\mathrm{cttw } = { j_{\mathrm c } } ( q^2 , p_l^2 ) { j_{\mathrm c } } ( q^2 , p_r^2 ) . \label{factqcd}\ ] ] the subscript on @xmath73 refers to the 2-jet contribution . @xmath74 has a simple physical interpretation : it is the probability of finding a final state jet with invariant mass @xmath13 . thus , it satisfies @xmath75 @xmath4 is a scale associated with the hard process . we will see later that the equivalent jet function in scet has a different normalization . at leading order the jet is a massless parton and so @xmath76 . at next - to - leading order , the coherent branching algorithm allows one - angular ordered emission . so @xmath77 here @xmath78 is the energy fraction in the emission and @xmath79 . the angular ordering constraint for the coherence of soft emissions is implicit in the restriction @xmath80 ; for @xmath81 it cuts off the infrared divergences in the integral as @xmath82 . evaluating gives @xmath83_{\star}^{[p^2 , q^2 ] } + \cdots . \label{jfirst}\ ] ] thus , to first order in @xmath12 , @xmath84_\mathrm{cttw } = \delta ( \tau ) + { \bar{\alpha}}\left [ \frac{- 4 \log \tau - 3}{\tau } \right]_{\star}^{[\tau , 1 ] } , \label{fixedr}\ ] ] which reproduces the divergent as @xmath85 part of the the parton model result . here we have introduced the @xmath86 distribution ( alternatively called the r- or @xmath87-distribution ) , which is a generalization of the + -distribution . it is uniquely defined by the two conditions @xmath88^{[x , a]}_{\star } = f ( x)$ for $ x > 0$}\ ] ] @xmath89^{[x , a]}_{\star } g ( x ) = \int_0^a d x f ( x ) \left [ g ( x ) - g ( 0 ) \right]\ ] ] for clarity , we have added to the notation an explicit instance of the dependent variable @xmath90 in @xmath91_{\star}^{[x , a]}$ ] . a useful relation is @xmath92^{[x , a]}_{\star } = [ f ( x)]^{[x , b]}_{\star } + \delta ( x ) \int_a^b d x ' f ( x ' ) .\ ] ] this will be used extensively in the next section . the leading order resummation is preformed by iterating the angular - ordered emissions . this leads to an integro - differential equation for @xmath93 , similar to the evolution equation for parton - distribution functions . the equation is solved in an integrated form , and the resummed expression for the integrated thrust in the two jet limit is @xmath94_\mathrm{cttw } = \exp \left [ - 2 { \bar{\alpha}}\log^2 \tau - 3{\bar{\alpha}}\log \tau \right ] \frac{e^{- 2 \gamma_e \eta}}{\gamma [ 2 \eta+1 ] } , \label{qcdresum}\ ] ] where @xmath95 this equation is a combination of expressions in @xcite taken with @xmath30 . expanding @xmath96 to order @xmath12 the fixed order result is reproduced . this expression is resummed in the sense that @xmath97 as @xmath85 , in contrast to the fixed order result for @xmath98 or the parton model expression which diverge as @xmath62 . having reviewed the way the resummed thrust distribution is traditionally calculated , we now turn to the equivalent calculation in scet . we will see that there are a number of advantages of this effective field theory treatment . instead of a classical treatment , where multiple real emissions are summed at the level of the cross section , the entire resummation is done in scet through the renormalization group . this makes explicit the various scales in the problem , and allows greater freedom to choose those scales to minimize the large logarithms in the distributions of interest . the idea behind scet is to separate out the quarks and gluons of qcd into soft and collinear degrees of freedom . a collinear field is associated with a light - like direction @xmath99 . the component of its momentum in that direction , @xmath100 , must be much larger than any of the other momentum components . all the qcd degrees of freedom which could change this momentum have been integrated out , so @xmath101 appears as a fixed label . for example , a collinear quark is written as @xmath102 with a fixed @xmath101 . in addition to collinear quarks and gluons , there are soft quarks and gluons . these soft fields can only interact with each other or transfer momentum to the soft components @xmath103 of a collinear field . the lagrangian of scet has a separate gauge invariance associated with soft gluons and gluons in each collinear direction . it is useful to maintain this gauge invariance explicitly in the operators of the theory with the use of wilson lines . for example , a two - jet operator is @xmath104 where @xmath105 @xmath106 the collinear wilson lines @xmath107 maintain collinear gauge invariance and the soft wilson lines @xmath108 maintain soft gauge invariance . the starting point in the effective theory approach to event shapes is again factorization in the two - jet limit . in scet , the event - shape distributions near the endpoint come from matrix elements of @xmath34 @xcite . in terms of the hemisphere masses @xmath109 and @xmath110 defined above , factorization implies @xcite @xmath111_\mathrm{scet}= |c_h ( \mu)|^2 \int { \mathrm{d}}k_l { \mathrm{d}}k_r j ( p_l^2 - q k_l , \mu ) j ( p_r^2 - q k_r , \mu ) s ( k_l , k_r , \mu ) . \label{factorform}\ ] ] the hard function @xmath112 , the jet functions @xmath113 and the soft function @xmath114 will be defined below . the form of this factorized expression has a physical explanation . each jet function @xmath115 comes from one of the collinear quarks . it represents , like @xmath74 , the probability for producing a jet of invariant mass @xmath13 from that collinear field ( the precise relation to @xmath74 is explored below ) . recall that the large component of the momentum of a collinear field @xmath116 is fixed . since , in the two - jet limit , the jets are back to back with center of mass energy @xmath4 , we must have @xmath117 . the small component @xmath118 can vary . if there are no soft effects , then the hemisphere mass is simply @xmath119 . however , as the factorization formula implies , the collinear jet can give up some soft momentum to the soft qcd background , leading to @xmath120 . the hemisphere mass is unchanged by this emission , but now @xmath121 , which explains the form of . the function @xmath122 is a hard function , which comes from integrating out hard modes of qcd in matching to scet . demanding @xmath123 for states with two external quarks lets us calculate @xmath122 order by order in perturbation theory . the computation entails taking the difference between qcd and scet graphs , such as @xmath124 where the @xmath125 refers to an insertion of @xmath126 , and the right diagram is only a representative contribution ( for details , see @xcite ) . the difference is finite because the infrared divergences in qcd and scet are the same and the uv divergences are removed with counterterms . at one - loop the matching gives @xcite @xmath127 with @xmath128 , @xmath129 and @xmath130 . the anomalous dimension and renormalization group evolution of @xmath112 can be found from the uv divergences of the same 1-loop graphs . the rg equation is @xmath131 with solution @xmath132 \left ( \frac{q^2}{\mu_h^2}\right)^{-\gamma_h \log\frac{\mu^2}{\mu_h^2 } } .\ ] ] to first order in @xmath12 , the @xmath133 dependence drops out and is reproduced . the natural matching scale @xmath134 is manifest in this expression . the jet function @xmath115 is the imaginary part of the propagator of a collinear quark . it is defined by @xcite @xmath135 .\ ] ] it can be thought of as the spectral density for a jet of collimated particles interacting with a soft qcd background . the jet function can be calculated order - by - order in perturbation theory through the discontinuity of conventional feynman diagrams , such as @xmath136 some of the infrared divergences in these graphs are cut off because the collinear quark is taken to have invariant mass @xmath13 . these calculations have been done in the context of @xmath28 @xcite and for deep - inelastic scattering @xcite and the jet function is known to two loops @xcite . due to factorization , the same jet function applies in these processes and in @xmath0 annihilation . to first order in @xmath12 , the result is @xmath137 + \left [ \frac{\gamma_j \log \frac{p^2}{\mu^2 } + \gamma_j}{p^2 } \right]^{[p^2 , \mu^2]}_{\star } , \label{jetfixed}\ ] ] with @xmath138 , @xmath139 and @xmath140 . the renormalization group evolution of the jet function , in contrast to that of the hard function , is non - local in @xmath13 @xmath141 j ( p^2 , \mu ) + 2\gamma_j \int_0^{p^2 } d q^2 \frac{j ( p^2 , \mu ) - j ( q^2 , \mu)}{p^2 - q^2 } .\ ] ] this is similar to the altarelli - parisi equation for the evolution of the parton distribution functions ( pdfs ) . in contrast to pdfs , however , the jet function @xmath115 is perturbative as long as @xmath142 . simplification is achieved through use of the laplace transform @xcite , @xmath143 whereby the evolution becomes local in @xmath144 @xmath145 now the rge can be solved like that of @xmath122 : @xmath146 \left ( \nu e^{\gamma_e } \mu_j^2 \right)^{-\eta_j } \tilde{j}(\nu,\mu_j)\ ] ] with @xmath147 finally , the inverse laplace transform produces @xcite @xmath148 \tilde{j } \left ( \partial_{\eta_j } \right ) \left [ \frac{1}{p^2 } \left ( \frac{p^2}{\mu_j^2 } \right)^{\eta_j } \right]_{\star}^{[p^2 , \mu_j^2 ] } \frac{e^{- \gamma_e \eta_j}}{\gamma [ \eta_j]},\ ] ] where @xmath149 the functional dependence on @xmath150 in @xmath151 comes from a functional dependence on @xmath152 in the fixed order expression for the jet function at the matching scale @xmath153 . finally , we have to calculate the soft function @xmath154 . it is defined though matrix elements of soft wilson lines @xcite @xmath155 unlike the jet function , the soft function is often evaluated at scales @xmath156 , where it is non - perturbative . it can , in principle , be measured experimentally and then be evolved perturbatively , like the pdfs . or it can be modeled @xcite . however , in our case , since we are taking @xmath12 to be small and fixed , it can be calculated in perturbation theory . even if @xmath157 is non - perturbative , the perturbative calculation is useful because the uv divergences dictate the anomalous dimensions and hence the evolution equation . the soft function , being completely defined in terms of wilson lines , can be calculated in qcd through diagrams such as @xmath158 where the kinked lines are wilson lines in the @xmath159 and @xmath160 directions . to order @xmath12 , @xmath161 can be written as a product @xmath162 where @xcite @xmath163 + \left [ \frac{\gamma_s \log \frac{k}{\mu } + \gamma_s}{k } \right]^{[k , \mu]}_\star \label{sfirst } , \ ] ] with @xmath164 , @xmath165 and @xmath166 . the evolution of the soft function involves the same kind of non - local equation as for the jet function . this can be seen from examining its anomalous dimension directly , or from rg invariance of the convolution appearing in the factorization formula . the solution is therefore similar . explicitly , @xmath167 \tilde{s } ( \partial_{\eta_s } ) \left [ \frac{1}{k } \left ( \frac{k}{\mu_s } \right)^{\eta_s } \right]_{\star}^{[k , \mu_s ] } \frac{e^{- \gamma_e \eta_s}}{\gamma [ \eta_s ] } , \ ] ] where @xmath168 and @xmath169 expanding to first order in @xmath12 reproduces . now we have all the ingredients appearing in the the factorization formula . at order @xmath12 , the result is @xmath170_\mathrm{scet } = \delta ( p^2_l ) \delta ( p_r^2 ) \left [ 1 + { \bar{\alpha}}\left ( - 1 + \frac{\pi^2}{3 } \right ) \right ] \nonumber\ ] ] @xmath171^{[p^2_l , q^2]}_\star + { \bar{\alpha}}\delta ( p_l^2 ) \left [ \frac{2 \log \frac{q^2}{p^2_r } - \frac{3}{2}}{p^2_r } \right]^{[p^2_r , q^2]}_\star . \label{dplpr}\ ] ] note that the dependence on all the scales @xmath172 and @xmath173 has completely canceled . this is a non - trivial result which requires three relations among the six anomalous dimensions : @xmath174 and is a strong consistency check on the entire formalism . s can be understood on more general grounds by relating the @xmath175s to a universal cusp anomalous dimension @xcite . ] already scet can be compared to cttw . the differential distribution in scet , eq . , matches the cttw distribution , eq . , when the jet functions are expanded to first order using eq . . the scet expression has an additional finite piece ( the @xmath176 term ) which comes from loop graphs which do not enter into the traditional formulation . still working at leading order in @xmath12 , scet produces a simple form for the thrust distribution near @xmath177 : @xmath178_\mathrm{scet } = \delta ( \tau ) \left [ 1 + { \bar{\alpha}}\left ( - 1 + \frac{\pi^2}{3 } \right ) \right ] + { \bar{\alpha}}\left [ \frac{- 4 \log \tau - 3}{\tau } \right]^{[\tau , 1]}_\star . \label{scetsig2}\ ] ] this reproduces the leading behavior for small @xmath64 of the both the parton model expression , eq . , and the resummed expression with the functions @xmath74 expanded to first order , eq . . to get the resummed thrust distribution from scet , we need to calculate a couple of convolution integrals . first , the soft and jet functions must be combined into @xmath179 second , the @xmath180 functions must be integrated against the event shape . for thrust , we need @xmath181 both of these convolutions can be evaluated by performing the laplace transform . note that the function names can be misleading . here , @xmath182 ( and not just the scet jet function @xmath115 ) is playing the role of @xmath183 from the cttw formulation . thus , the cttw jet function is a combination of the soft and jet functions in scet , as expected because the coherent branching algorithm used to derive it incorporates both soft and collinear effects . solving the @xmath180 function with the laplace transform techniques gives @xmath184 \exp \left [ \frac{\gamma_s}{2 } \log^2 \frac{\mu}{\mu_s } - \gamma_s \log \frac{\mu}{\mu_s } \right ] \left ( \frac{q \mu_s}{\mu_j^2 } \right)^{-\eta_s}\ ] ] @xmath185 \left [ \frac{1}{p^2 } \left ( \frac{p^2}{\mu_j^2 } \right)^{\eta_k } \right]_{\star}^{[p^2 , \mu_j^2 ] } \frac{e^{- \gamma_e \eta_k}}{\gamma [ \eta_k ] } , \label{resid}\ ] ] where @xmath186 and @xmath187 note the residual logarithms of @xmath188 in line which are not resummed . these imply that we can not choose @xmath189 . instead , the natural scales ( those which remove the large logs ) should satisfy @xmath190 . then , we can evolve either the jet function from @xmath191 down to @xmath173 or the soft function from @xmath173 up to @xmath191 but there is not a single scale which can minimize all the large logs . finally , to evaluate the convolution for the thrust integral , we use the laplace transform of the @xmath192-function @xmath193 combining this with the expression for @xmath194 above , the scet prediction for thrust in the 2-jet limit is @xmath94_\mathrm{scet } = |c_h ( \mu_h ) |^2 \exp \left [ 2 { \bar{\alpha}}\log^2 \frac{\mu_h^2}{\mu_j^2 } + 3 { \bar{\alpha}}\log \frac{\mu_h^2}{\mu_j^2 } - { \bar{\alpha}}\log^2 \frac{\mu_h^2}{\mu_s^2 } \right ] \left ( \frac{q \mu_s}{\mu_j^2 } \right)^{- 4 { \bar{\alpha}}\log \frac{\mu_h^2}{\mu_s^2 } } \nonumber\ ] ] @xmath195 ^ 2 \left ( \frac{q^2 \tau}{\mu_j^2 } \right)^{2 \eta_k } \frac{e^{- 2 \gamma_e \eta_k}}{\gamma [ 2 \eta_k + 1 ] } .\ ] ] note that this expression is explicitly independent of @xmath87 . now , choosing the natural scales @xmath196 , @xmath197 , and @xmath198 to remove the large logs , the thrust distribution becomes @xmath94_\mathrm{scet } = \exp \left [ - 2 { \bar{\alpha}}\log^2 \tau - 3 { \bar{\alpha}}\log \tau \right ] \tilde{r } ( \partial_{\eta } ) \frac{e^{- 2 \gamma_e \eta}}{\gamma [ 2 \eta + 1]}\ ] ] with @xmath199 and @xmath200 as in eq . . this expression would be identical to the cttw expression , eq . , if @xmath201 . recall that the function @xmath202 comes from 1-loop matching in scet , which turns into boundary conditions for the renormalization group evolution . in the approach of cttw the boundary condition is simply that @xmath203 . nevertheless , the effect of @xmath204 to order @xmath12 is only to provide a finite constant , as can been seen from comparing equations and that is , there is no contribution of order @xmath205 in the difference , only of order @xmath206 and higher . beyond order @xmath12 , the @xmath207 does change the @xmath205 dependence . but since @xmath208 these terms are subleading to the dominant @xmath209 and can get corrections from higher - loop effects . for example , at two - loops , a term of the form @xmath210 is generated . thus , scet and cttw agree to first order in @xmath12 and for the resummation of the leading large logarithms , which is the order to we have been working . now let us look at how fixed order results and resummation are combined . there are two elements to this : ( 1 ) matching to hard emissions to get the differential distribution correct away from the two - jet region ( i.e. away from @xmath22 ) ; and ( 2 ) including finite parts of loops to reproduce fixed order inclusive results . let us begin with the matching procedure described in @xcite . to match to the hard emissions at order @xmath12 we need the parton model differential cross section from equation ( [ partondc ] ) . the divergent part is already contained in the two jet contribution , so the remainder is @xmath211 @xmath212 - 2 \log^2 ( 1 - \tau ) + 3 ( 1 - 2 \tau ) \log ( 1 - 2 \tau ) - 4 \mathrm{li}_2 ( \frac{\tau}{1 - \tau } ) + 6\tau + \frac{9}{2 } \tau^2 .\ ] ] the final result for the matched integrated thrust distribution from @xcite is @xmath213_{\mathrm{cttw } } = ( 1 + \sigma_1 ) \left\ { \left ( 1 + c_1 \right ) \exp \left [ - 2 { \bar{\alpha}}\log^2 \tau - 3{\bar{\alpha}}\log \tau \right ] \frac{e^{- 2\gamma_e \eta}}{\gamma [ 2\eta+1 ] } + { \bar{\alpha}}{d_\mathrm{fin}}(\tau ) \right\ } , \label{matchedqcd}\ ] ] where @xmath214 is the nlo contribution to the total cross section and @xmath215 is chosen so that @xmath216 ( recall that @xmath217 at order @xmath12 ) . in the qcd approach the factor @xmath218 must be be determined by a independent calculation of the total rate . this involves combining the real and virtual contributions at order @xmath12 using an appropriate infrared regulator . as emphasized in @xcite there is arbitrariness in the matching because the hard emissions ( in @xmath219 ) are fixed order but the two - jet contribution is resummed . for example , to the same accuracy @xmath220 could be multiplied by the exponential . now , let us turn to the matching in scet . to include thrust distributions away from the endpoint , we can either attempt to add power corrections to scet , or we can match to higher order operators as described in @xcite . matching is much simpler . to perform the matching , we add new operators @xmath221 where the @xmath222 are the additional terms coming from wilson lines necessary for gauge invariance . @xmath223 is a collinear gluon in direction @xmath224 . the matching demands that @xmath225 where the subscript means the matching is done at the hard scale @xmath133 . there is some arbitrariness in the definition of the matrix elements in scet due to reparameterization invariance . the matrix element of a collinear quark @xmath102 on a qcd quark state @xmath226 is only defined up to its soft momentum component @xmath227 . moreover , a basis for summing over directions @xmath159 for @xmath126 and @xmath228 , @xmath224 , and @xmath229 for @xmath37 must be chosen as well . a certain convention was described in @xcite for resolving these ambiguities and others are possible . in any case , while different conventions may shift the contributions from @xmath34 and @xmath35 in , the sum is parameterization invariant . thus , independently of the convention we have @xmath230_{\mathrm{scet}}^{\mu_h } = [ { \mathrm{d}}\sigma]_{\mathrm{parton}},\ ] ] where @xmath231_{\mathrm{parton}}\sim \langle \mathrm{qcd } \rangle^2 $ ] is the tree - level parton model cross section , as shown in eq . . now , we already know that @xmath232 gives to first order in @xmath12 and that the parton model distribution at order @xmath12 is . thus , with obvious implicit phase space factors , @xmath233 so , at leading order , the contribution from the sum of @xmath234 and the interference between @xmath34 and @xmath35 to the differential cross section is unambiguous . the running of @xmath35 could also be included even though it does not resum any large logs for the event shapes under consideration . with running , the matrix elements of @xmath34 and @xmath35 would appear with separate renormalization kernels , and so the final differential cross section would end up depending on the conventions chosen . the ambiguity could be resolved by a careful higher order treatment , but for the purposes of comparing to the cttw prediction for thrust , we will simply not renormalize the finite terms . then , @xmath213_{\mathrm{scet } } = \exp \left [ - 2 { \bar{\alpha}}\log^2 \tau - 3 { \bar{\alpha}}\log \tau \right ] \tilde{r } ( \partial_{\eta } ) \frac{e^{-2 \gamma_e \eta}}{\gamma [ 1 + 2\eta ] } + { \bar{\alpha}}{d_\mathrm{fin}}(\tau ) . \label{matchedscet}\ ] ] with @xmath235 and @xmath205 as in eq . and @xmath219 in . the total cross section in scet is given @xmath236_\mathrm{scet } = \sigma_0\left[1 + { \bar{\alpha}}\left ( - 1 + \frac{\pi^2}{3 } \right ) + { \bar{\alpha}}{d_\mathrm{fin } } ( \frac{1}{3 } ) \right ] = \sigma_0\left [ 1 + \frac{\alpha_s}{\pi}\right].\ ] ] this is the correct total @xmath0 total cross section to first order in @xmath12 ! let us review the contributions that go into the cross section . first , at the hard scale @xmath134 , there is the finite part of the loop matching to qcd , @xmath237 . next , at the scale @xmath238 where we integrate out the collinear fields , the jet functions give @xmath239 . at the seesaw scale @xmath240 , the soft function gives @xmath241 , and finally the finite part of the real emission integral , away from @xmath22 gives @xmath242 . in producing the total cross section , only the soft and jet contributions are infrared divergent . however , their convolution , which appears in the function @xmath243 is infrared finite . the hard matching and the hard emissions are ir finite by themselves . thus , the total cross section can be understood as a combination of a process dependent ir finite hard part and universal but ir regulator dependent contributions from soft and collinear emissions . the scet thrust distribution is compared to the cttw thrust distribution in figure [ fig : dt ] . the plot is of @xmath244 with @xmath12 fixed at @xmath245 . a more careful rendition of the differential thrust distribution would take the derivative of @xmath246 before assigning the matching scale @xmath247 , as is done below . however , the effect is higher order , and so we just plot @xmath248 directly . matched resummed differential thrust distribution in scet and in the standard approach of cttw with fixed coupling @xmath249,scaledwidth=95.0% ] in this section , some simple generalizations are described . the above results were derived assuming @xmath12 to be constant in order to emphasize the resummation of sudakov logarithms in contrast to large logarithms associated with the scale @xmath31 . now , it is shown how the results can be modified with running @xmath12 . also , the scet prediction for another event shape , the jet mass @xmath250 , is given . it is straightforward to allow @xmath12 to run . including 1-loop running , the effect is to modify the single and double logs in the following way @xcite : @xmath251\ ] ] @xmath252 for example , the differential thrust distribution for @xmath253 with running @xmath12 becomes @xmath254_{\mathrm{scet } } \!\!\!\!&=&\ ! \frac{1}{\tau } \exp\left [ 4 s(q , q \tau ) + 6 a(q , q \tau ) - 8 s(q\sqrt{\tau},q \tau ) -6 a(q \sqrt{\tau},q \tau)\right ] \tilde{r } ( \partial_{\eta } ) \frac{e^{- 2 \gamma_e \eta}}{\gamma [ 2 \eta ] } \nonumber \\ & & + \frac{2 \alpha_s}{3 \pi } d_{\mathrm{fin } } ( \tau ) , \label{scetfinal}\end{aligned}\ ] ] where @xmath255 @xmath256 is given in eq . and @xmath257 @xmath258 \partial_{\eta}^2 - \frac{3}{2 } \alpha_s(q\sqrt{\tau } ) \partial_{\eta } \label{r2 } \right\ } .\ ] ] this is the same function @xmath202 as in eq . , but with the @xmath12 factors evaluated at the appropriate matching scales . at this point , one would hope to compare to data . however , besides the obvious shortcoming of not containing the full nll resummation ( it does not include effects of the two - loop cusp anomalous dimension ) , this parton - level expression does not include important non - perturbative effects . due to the running of @xmath12 , the perturbative expression breaks down when the soft scale is of order @xmath31 , that is , when @xmath259 , as can be seen explicitly in . in fact , even for significantly larger values of thrust power corrections of order @xmath260 become quantitatively important , at least at lep energies . this problem has been approached elsewhere using scet @xcite and with other techniques @xcite . other event shapes can be studied the same way as thrust . for example , consider the heavy jet mass @xmath250 defined by @xmath261 in this case , the matching scales are @xmath134 , @xmath262 and @xmath263 and scet gives for @xmath264 @xmath265_\mathrm{scet } & = & \frac{2}{\rho } |1 + c_h |^2 \exp\left [ 4 s(q , q \rho ) + 6 a(q , q \rho ) - 8 s(q\sqrt{\rho},q \rho ) - 6 a(q \sqrt{\rho},q \rho ) \right ] \nonumber\\ & & \times \left [ \tilde{k } ( \partial_{\eta } ) \frac{e^{- \gamma_e \eta}}{\gamma [ \eta ] } \right ] \left [ \tilde{k } ( \partial_{\eta } ) \frac{e^{- \gamma_e \eta}}{\gamma [ \eta + 1 ] } \right ] + \frac{2 \alpha_s}{3 \pi } d_{\mathrm{fin } } ( \rho ) , \end{aligned}\ ] ] where @xmath266 this formula agrees with the jet mass distribution in @xcite to leading log and first order in @xmath12 . the same function @xmath267 appears for jet mass and for thrust because to order @xmath12 in the parton model , @xmath268 . we have investigated how to combine resummation with next - to - leading order matching of event shapes in the original approach of @xcite ( cttw ) and using scet . in the cttw formulation , real emissions from collinear splitting functions are used and various kinematical features associated with soft emission are combined to derive a differential cross section . the cross section factorizes into the product of two jet functions . resummation is done by solving a differential equation for the jet functions in terms of the physical scales @xmath13 and @xmath269 of the event . in contrast , scet factorizes the event shape distribution into a contribution from hard , jet , and soft functions . these functions are matched at the scales @xmath196 , @xmath270 and @xmath271 respectively and resummation is done through renormalization group evolution . the resummed thrust distribution in scet and cttw have been compared , and found to agree to next - to - leading order in @xmath12 and for leading - log resummation . the resummation of thrust in scet brings to light a number of features of the process not obvious in cttw formulation . for example , the appearance of the seesaw scale @xmath272 as the natural matching scale for the soft function follows from the kinematics of the scet factorization theorem . of course , the existence of this scale has been known for a long time from qcd , but in the effective field theory approach this scale just drops out of the factorized expression . thus , for more complicated processes , it is reasonable to expect a similar transparency for the matching scales , which may facilitate subleading log resummation . in fact , two - loop , three - loop , and some four - loop anomalous dimensions for various quantities are already available @xcite , and so subleading log resummation appears quite feasible . the biggest impediment to using these more accurate resummed results in a comparison to data is that power corrections of order @xmath260 have an important quantitative effect on event shapes . however , these corrections should modify only the soft function while higher order resummed expressions for the the hard and jet functions will remain valid . thus , the effective theory should be able to weave together the perturbative and non - perturbative calculations . a new result of this paper is the demonstration that inclusive quantities , such as the total cross section for @xmath0 can be calculated in a new way using scet . instead of taking the full differential @xmath159 + 1 body cross section and combining with the one - loop @xmath159-body cross section , scet combines finite parts of loops of soft and jet functions with a hard matching calculation and a finite integral over hard emissions . the soft and jet functions depend on the infrared regulator , but their convolution , and the hard function , do not . for @xmath0 annihilation at nlo , this may not be so impressive , but the procedure promises to apply to more complicated processes , perhaps even some for which nlo results are not available . it would also be very interesting to explore this mechanism at nnlo or to work with hadronic processes where the singularity structure is more complicated . i would like to thank t. becher and s. fleming for many useful discussions . i also benefited from the hospitality of the les houches ecole de physique and discussions with many of its participants . this work was supported in part by the national science foundation under grant nsf - phy-0401513 and by the johns hopkins theoretical interdisciplinary physics and astronomy ceneter . c. w. bauer , s. fleming and m. e. luke , phys . d * 63 * , 014006 ( 2001 ) [ arxiv : hep - ph/0005275 ] . c. w. bauer , s. fleming , d. pirjol and i. w. stewart , phys . d * 63 * , 114020 ( 2001 ) [ arxiv : hep - ph/0011336 ] . c. w. bauer , d. pirjol and i. w. stewart , phys . d * 65 * , 054022 ( 2002 ) [ arxiv : hep - ph/0109045 ] . m. beneke , a. p. chapovsky , m. diehl and t. feldmann , nucl . b * 643 * , 431 ( 2002 ) [ arxiv : hep - ph/0206152 ] . c. w. bauer and a. v. manohar , phys . d * 70 * , 034024 ( 2004 ) [ arxiv : hep - ph/0312109 ] . s. w. bosch , b. o. lange , m. neubert and g. paz , nucl . b * 699 * , 335 ( 2004 ) [ arxiv : hep - ph/0402094 ] . m. neubert , eur . j. c * 40 * , 165 ( 2005 ) [ arxiv : hep - ph/0408179 ] . a. v. manohar , phys . 114019 ( 2003 ) [ arxiv : hep - ph/0309176 ] . t. becher and m. neubert , phys . * 97 * , 082001 ( 2006 ) [ arxiv : hep - ph/0605050 ] . t. becher , m. neubert and b. d. pecjak , jhep * 0701 * , 076 ( 2007 ) [ arxiv : hep - ph/0607228 ] . s. fleming , a. h. hoang , s. mantry and i. w. stewart , arxiv : hep - ph/0703207 . c. w. bauer and m. d. schwartz , arxiv : hep - ph/0607296 . c. w. bauer and m. d. schwartz , phys . lett . * 97 * , 142001 ( 2006 ) [ arxiv : hep - ph/0604065 ] . m. trott , phys . d * 75 * , 054011 ( 2007 ) [ arxiv : hep - ph/0608300 ] . s. catani , g. turnock , b. r. webber and l. trentadue , phys . b * 263 * , 491 ( 1991 ) . s. catani , m. l. mangano , p. nason and l. trentadue , nucl . b * 478 * , 273 ( 1996 ) [ arxiv : hep - ph/9604351 ] . y. l. dokshitzer , a. lucenti , g. marchesini and g. p. salam , jhep * 9801 * , 011 ( 1998 ) [ arxiv : hep - ph/9801324 ] . s. catani , b. r. webber and g. marchesini , nucl . b * 349 * , 635 ( 1991 ) . s. catani and m. h. seymour , nucl . b * 485 * , 291 ( 1997 ) [ erratum - ibid . b * 510 * , 503 ( 1998 ) ] [ arxiv : hep - ph/9605323 ] . c. w. bauer , a. v. manohar and m. b. wise , phys . lett . * 91 * , 122001 ( 2003 ) [ arxiv : hep - ph/0212255 ] . c. lee and g. sterman , phys . d * 75 * , 014022 ( 2007 ) [ arxiv : hep - ph/0611061 ] . t. becher and m. neubert , phys . b * 637 * , 251 ( 2006 ) [ arxiv : hep - ph/0603140 ] . g. p. korchemsky and g. marchesini , phys . b * 313 * , 433 ( 1993 ) . g. p. korchemsky and s. tafat , jhep * 0010 * , 010 ( 2000 ) [ arxiv : hep - ph/0007005 ] . g. p. korchemsky and g. sterman , nucl . b * 555 * , 335 ( 1999 ) [ arxiv : hep - ph/9902341 ] . g. p. korchemsky and a. v. radyushkin , nucl . b * 283 * , 342 ( 1987 ) . a. v. belitsky , phys . b * 442 * , 307 ( 1998 ) [ arxiv : hep - ph/9808389 ] . s. moch , j. a. m. vermaseren and a. vogt , nucl . b * 688 * , 101 ( 2004 ) [ arxiv : hep - ph/0403192 ] . t. becher and m. neubert , phys . b * 633 * , 739 ( 2006 ) [ arxiv : hep - ph/0512208 ] .

it is well established that the diffuse x - ray emission from clusters of galaxies implies the existence of a hot ( 10@xmath1 k ) , thermally radiating intracluster medium ( icm ) . in the dense centers of many clusters this gas is expected to cool radiatively in less than a hubble time . as pressure support decreases the gas will gravitate towards the bottom of the cluster potential well , a phenomenon that has come to be known as a `` cooling flow '' ( cf . fabian 1994 ) . such cooling flows can be detected by direct x - ray imaging , where they are seen as a surface brightness enhancement in the x - ray emission from a region typically @xmath0 100 - 200 kpc in radius , and by spectroscopic observations which reveal gas at cooler temperatures than the surrounding icm . the excess x - ray luminosity in the cooling flow region is typically 10@xmath2 ergs s@xmath3 , corresponding to roughly 10 - 50 % of the total cluster x - ray emission ( peres et al . the mass deposition rate is @xmath4 , where @xmath5 is the gas temperature at r@xmath6 , the cooling radius . estimated cooling flow rates in the centers of rich clusters vary from 30 - 300 m@xmath7 yr@xmath3 ( white , jones , & forman 1997 ) and in some cases reach mass deposition rates of 1000 m@xmath7 yr@xmath3 or more ( allen 2000 ) . a long - standing problem with the cooling flow scenario has been the fact that the large quantities of cool gas expected to accumulate from the flow have rarely been detected . searches for cool gas in atomic or molecular form have generally proven unsuccessful ( e.g. , heckman et al . 1989 ; jaffe 1990 ; mcnamara , bregman , & oconnell 1990 ; odea et al . 1994 ; voit & donahue 1995 ; jaffe & bremer 1997 ; donahue et al . 2000 ) , with reported upper limits two orders of magnitude smaller than the masses implied by x - ray mass deposition rates ( although see edge 2001 ) . while some support for the cooling flow picture is provided by low energy x - ray spectra of clusters which appear to require larger absorption than can be accounted for by foreground gas alone ( white et al . 1991 ; allen et al . 1993 ; irwin & sarazin 1995 ) , it now appears that the fraction of clusters that actually possess such excess absorption may be much less than previously thought ( arabadjis & bregman 2000 , hereafter ab2000 ; bhringer et al . furthermore , recent high - resolution x - ray spectra obtained with the xmm - newton and chandra satellites have failed to show expected emission lines from cooled gas at temperatures below a few kev ( fabian et al . 2001 ; peterson et al . 2001 ; tamura et al . 2001 ) and it has also been found that local isothermality matches the spectral data of `` cooling flow '' regions better than the continually declining temperature profile predicted by the standard cooling flow model ( e.g. molendi & pizzolato 2001 ; bhringer et al . a variety of mechanisms have been proposed to account for the lack of detectable cool gas in cooling flows ; these include reheating the gas , mixing , differential absorption , efficient conversion of cooling flow gas into low mass stars , inhomogeneous metallicity distributions , or disruption of cooling flows by recent subcluster mergers ( see recent discussions in fabian et al . 2001 and peterson et al . 2001 ) . as of yet , it is unclear whether any of these mechanisms are capable of resolving the aforementioned problems with the cooling flow model . one characteristic that is common to the mechanisms mentioned above is that they have mainly been introduced in an `` after the fact '' manner . that is , they assume the observed mass deposition rates are basically correct but that something is happening to the gas after ( or while ) it cools , be it reheating or conversion into low mass stars or some other process . perhaps a simpler way of addressing the problems of the cooling flow model are to look for ways to reduce the mass deposition rates from the beginning . one way to achieve this is to reduce the excess emission ( @xmath8 ) associated with cooling flows by attributing some ( or all ) of it to x - ray sources other than thermal bremsstrahlung or line emission from the icm . but are there sources luminous enough to reproduce @xmath8 , can they be found within galaxy clusters , and , most importantly , can they mimic hot cluster gas ? soon after x - ray emission was first detected from galaxy clusters , katz ( 1976 ) proposed that much of this emission might come from individual compact sources such as those known to exist in our own galaxy . however , the subsequent discovery of iron emission lines in x - ray spectra ( forman & jones 1982 , and references therein ) clearly demonstrated that thermal bremsstrahlung from a diffuse intracluster gas is the dominant source of x - ray emission in clusters , and interest in compact sources quickly waned . nevertheless , compact sources must surely contribute some fraction of the total x - ray emission from galaxy clusters . a growing body of evidence indicates that a substantial intergalactic population of stars , star clusters and stellar remnants is present in the cores of rich clusters , and some of these objects will be x - ray sources . for example , ferguson , tanvir , & von hippel ( 1998 ) showed the existence of intergalactic stars in the virgo cluster , and planetary nebulae associated with this population have also been detected ( e.g. , arnaboldi et al . 1996 ; ciardullo et al . 1998 ) . similarly , west et al . ( 1995 ) hypothesized the existence of a large population of intergalactic globular clusters in the cores of galaxy clusters based on observed galaxy - to - galaxy variations in globular cluster populations . furthermore , there is mounting evidence that many cluster galaxies have been destroyed over a hubble time , victims of collisions with other galaxies or disruption by the mean tidal field of the cluster potential in which they reside ( e.g. , thompson & gregory 1993 ; lopez - cruz et al . 1997 ; gregg & west 1998 ; moore et al . 1999 ; calcaneo - rodin et al . 2000 ) . as galaxies are disrupted they will spill their contents throughout intracluster space , including x - ray sources such as x - ray binaries , globular clusters , and black holes . such disruption is likely to be most frequent in the dense cluster core , and hence would be expected to produce a substantial pool of intracluster material , including compact x - ray sources , at the cluster center . any significant contribution by these compact sources to the x - ray flux will have to be taken into account when estimating the properties of icm , including temperatures , densities , and cooling flow mass deposition rates . an intracluster population of compact objects would have a different spectral signature than thermal emission from intracluster gas . for instance , the spectra of large spiral and elliptical galaxies reveal the presence of hard x - ray emission which is thought to be produced by either high mass x - ray binaries or accretion onto massive black holes and is adequately represented by power - laws having @xmath9 @xmath0 0.6 - 3.0 ( where @xmath10 ; colbert & mushotzky 1999 ; allen , di matteo , & fabian 2000 ) . luminous agn , on the other hand , are often fit by power - laws with @xmath9 @xmath0 1.6 - 2.0 ( nandra et al . indeed , the spectral signatures of such objects have already been identified in several x - ray clusters by allen et al . ( 2001 ) . using asca data , these authors found very luminous sources with power - law spectral indices characteristic of compact sources . the recent discovery of luminous _ extended _ non - thermal x - ray emission in several weak ( or non- ) cooling flow clusters ( rephaeli , gruber , & blanco 1999 ; fusco - femiano et al . 1999 ; 2000 ; 2001 ) suggests that one should also look for such emission in massive cooling flow clusters . at present , the origin of the diffuse non - thermal emission is still a matter of some debate . suggested mechanisms include inverse - compton scattering of cmb photons up to x - ray energies via a population of highly relativistic electrons , and non - thermal bremsstrahlung , likewise , by a population of relativistic electrons ( sarazin 2001 ; rephaeli 2001 ) . those highly relativistic electrons may have been accelerated by recent subcluster merger events which shocked the icm or may have been ejected from a central massive black hole . the discovery of x - ray holes at the positions of radio lobes of a central compact object in several massive cooling flow clusters ( e.g. bhringer et al . 1993 , mcnamara et al . 2000 ; fabian et al . 2000 ) certainly lends credence to the second hypothesis . recently , harris et al . ( 2000 ) ( chandra data ) and bhringer et al . ( 2001b ) ( xmm - newton data ) have even identified extended non - thermal emission directly associated with radio jets / lobes in 3c295 and m87 , respectively . the spectral signature of extended non - thermal emission is also different from that of thermal emission . to date , the extended non - thermal components of x - ray spectra of clusters have adequately been modeled by power - laws having @xmath11 ( fusco - femiano et al . 1999 ; 2000 ; 2001 ) . as with luminous compact sources , a proper deduction of cluster properties should explicitly include the effects of extended non - thermal emission in the analysis . in the present work we assess the case for a power - law component to the x - ray spectra of cluster centers , in particular those with cooling flows . we show that simultaneous fits of rosat pspcb and asca gis / sis spectra , which have been used to infer the presence of cooling flows in many clusters to date , are in fact consistent with a significant contribution from compact sources and/or extended non - thermal emission . using rosat hri data , we have placed limits on the contribution of any bright point sources within the central regions of these clusters and conclude that the power - law components are most likely produced by either a collection of many low - luminosity compact sources ( with @xmath12 ergs s@xmath3 ) or by diffuse non - thermal emission or by some combination of the two . we shall assume @xmath13 = 50 km s@xmath3 mpc@xmath3 and @xmath14 throughout . ten well - studied x - ray luminous galaxy clusters ( table 1 ) were chosen to span a wide range of cooling flow mass deposition rates . five of these have essentially no cooling flow ( ncf ) , while the other five cooling flow ( cf ) clusters have calculated mass deposition rates ranging from 189 to 645 solar masses per year . observations by the rosat position sensitive proportional counter ( pspcb ) and high resolution imager ( hri ) and the asca gas imaging spectrometer ( gis ) and solid - state imaging spectrometer ( sis ) were obtained from the heasarc public archives and reduced using ftools v5.01 . the combination of rosat and asca data provides good sensitivity to both soft and hard x - ray emission , spanning the total energy range 0.14 - 10.0 kev , and should provide a strong test of power - law components in the x - ray spectra of clusters . the rosat pspcb data were corrected for spatial and temporal gain fluctuations in the detector . x - ray spectra , spanning the energy range 0.14 - 2.04 kev , of all ten clusters were extracted from circular regions centered on the peak of the emission with a constant extraction radius ( r ) of 3 arcminutes . background spectra were selected from source - free circular regions of radius r. all spectra were corrected for the effects of cosmic rays in the detector ( snowden et al . 1992 ) , vignetting , spatial variations in detector efficiency , and deadtime . the background - subtracted spectra were then grouped to ensure a minimum of 20 counts per radial bin so that reliable @xmath15 statistics could be computed . the rosat hri data was reduced in a manner similar to the pspcb data . the hri data was used to place limits on the contribution of any point sources within the central regions of these clusters . unfortunately , the hri lacks spectral resolution and was not used in the main spectral analysis . a description of the analysis of hri data is given in section 5 . no hri data were available for a2063 or a3532 . the screened events from `` revision 2 '' processing of asca gis / sis data were used . additional cleaning procedures ( discussed in the asca abc guide ) were implemented , including gain and deadtime corrections . for the gis detectors ( gis2 and gis3 ) , x - ray spectra spanning the energy range 0.7 - 10.0 kev were extracted from the same region of sky as that of the rosat pspcb source spectra . for the sis detectors ( sis0 and sis1 ) , x - ray spectra spanning the range 2.0 - 10.0 kev were extracted , thus avoiding complications with the degrading efficiency of the sis detectors at low energies . background spectra for both the sis and gis detectors were extracted from the `` blank sky '' observations made during the performance testing stage of the mission . spectra were extracted from the same regions of the detectors as that of the source spectra ( thus minimizing position - dependent effects associated with background subtraction ) . the background - subtracted spectra were also grouped ( or binned ) to ensure @xmath15 statistics could be used . for the clusters a2142 and a3667 sis1 data were ignored because of an excess of hot and flickering pixels . reduced rosat pspcb and asca gis / sis spectra were fit simultaneously using models within the software package xspec v11 ( arnaud 1996 ) . each cluster was fit with a variety of models based on mekal thermal plasma routine ( mewe & gronenschild 1985 ; kaastra 1992 ; liedahl et al . 1995 ) , and/or redshifted power - laws . in addition , we used the popular cooling flow model comprising a thermal plasma ( mekal ) and cooling flow component implemented by the mkcflow routine within xspec ( mushotzky & szymkowiak 1988 ) . fits were obtained by varying the temperature ( @xmath16 ) and metallicity ( @xmath17 ) of thermal plasma component(s ) and by varying the photon index ( @xmath9 ) of any power - law components . photoelectric absorption ( balucinska - church & mccammon 1992 ) was applied to the spectral models through two separate methods : i ) by applying only one absorption component to the entire spectral model and fixing the hi column density ( @xmath18 ) at the known galactic value ( see table 1 ) and then letting it vary to see if the fit could be improved ; ii ) by applying one absorption component to the entire spectral model ( with @xmath18 fixed at the galactic value ) and a separate redshifted variable absorption component ( @xmath19 ) to any power - law components via the zphabs model in xspec . the second method allows for _ intrinsic _ absorption to be associated with any power - law components . spectra of nearby high mass x - ray binaries and accreting massive black holes provide strong evidence for heavy intrinsic absorption , with @xmath20 @xmath21 above the galactic column ( colbert & mushotzky 1999 ; iwasawa et al . 2000 ; akylas et al . 2001 ) . we therefore regard the addition of a free intrinsic absorption parameter as an important feature of our spectral analysis . henceforth we will use tp , pl , and cf to refer , respectively , to thermal plasma ( mekal ) , power - law , and mkcflow components . for example , in this notation tppl identifies a model having thermal plasma and power - law components . goodness of fit was judged by the reduced chi - square ( @xmath22 ) and by visual inspection of residual plots . when additional components were added to a spectral model , for example adding a power - law component to a thermal plasma component to make tppl , the corresponding f - statistic was calculated in order to judge whether the addition was reasonable . unless stated otherwise , the derived relative contributions of each of the components to the total flux refer to the energy range 0.14 - 10.0 kev stated , whereas the luminosities presented refer to the asca gis band only ( 0.7 - 10.0 kev ) , since the spectra were fit simultaneously and not combined . all luminosities presented were corrected for galactic absorption only . to check our data reduction procedures , spectra from each of the detectors were also fit individually and no significant deviations were found . we have also attempted to duplicate published results by arabadjis & bregman ( 2000 ) for three clusters derived using the same data ( rosat pspcb ) and the same software . the results , which refer to the inner 3 arcminute radius , are presented in table 2 . here , the hotter tp component is fixed at the white , jones , & forman ( 1997 ) value , @xmath23 , @xmath18 is fixed at the stark et al . ( 1992 ) value , and the energy range is 0.14 - 2.4 kev and 0.14 - 2.04 kev for ab2000 and our study , respectively . the comparison shows excellent agreement , with derived temperatures and mass deposition rates having the same values to within one standard deviation . we find that single - temperature tp models provide acceptable fits to the spectra of a119 , a2063 , a3266 , a3532 , and a3667 ( 1.00 @xmath24 @xmath22 @xmath24 1.17 ) with temperatures in very good agreement with previous studies ( white , jones , & forman 1997 ; markevitch et al . significant improvement is not usually obtained by varying the hydrogen column density from the known galactic value reported by dickey & lockman ( 1990 ) or stark et al . the derived metallicities range from @xmath25 which is typical of the intracluster medium ( allen & fabian 1997 ) . in general , both pure pl models with or without variable intrinsic absorption do not provide acceptable fits to the ncf cluster spectra with @xmath26 @xmath27 1.4 . this result is not unexpected since it leaves no room for thermal emission from the icm . with intrinsic absorption included , an acceptable fit is obtained for a3532 with @xmath28 , however . this cluster is located at the center of the shapley concentration and is thought to be undergoing a merger with a3528 . the diffuse non - thermal emission generated from such a merger might possibly account for the ability of the simple pl model to fit the spectrum of a3532 . unfortunately , no hri data exists for this cluster so we can not place limits on the contribution of any point sources within the central 3 arcminutes . 0.1 in 0.1 in in the best fitting tppl models , the pl component typically contributes 1 - 14% ( mean of 7 % ) when absorbed by the galactic column only or 10 - 25 % ( mean of 20 % ) when intrinsic absorption is included ( excluding a3532 which has an anomalously large pl component ) . we note that in virtually every case the inclusion of intrinsic absorption noticeably improves the quality of the fit , making it superior to even the simple tp models ( see table 3 ) . a mean f - statistic of @xmath29 ( @xmath27 95 % confidence ) was found when the pl component with intrinsic absorption was included in the fit indicating that its addition is , indeed , reasonable . the corresponding luminosities of intrinsically absorbed pl components range from @xmath30 erg s@xmath3 with a mean value of @xmath31 erg s@xmath3 ( corrected for galactic absorption only ) . photon indices for these components are typical of compact accretion - driven x - ray sources and extended non - thermal emission . interestingly enough , the amount intrinsic absorption ranges from @xmath32 @xmath21 which is typical of optically thick compact sources . we also note that the temperature of the icm tends to decrease significantly when a power - law component is included . this trend was previously reported by allen et al . ( 2001 ) when including a power - law component in their spectral models . figure 1 presents the best fit tppl model to the spectrum of a3667 . for clarity , only the rosat pspcb data are plotted . as expected , standard cooling flow ( tpcf ) models of our non - cooling flow sample yield best fits with @xmath33 = 0 m@xmath7 yr@xmath3 to within one sigma . the mass deposition rate for the modest cooling flow cluster a2063 , however , is 25 @xmath34 12 m@xmath7 yr@xmath3 . the cooler component in our tptp models typically contributes zero flux to within one sigma ( @xmath0 10 % for a2063 & a3667 ) . neither of these models provided fits significantly better than those of the standard tp models . in summary , we confirm earlier investigations which find that tp models provide satisfactory fits to rosat and asca spectra of the centers of non - cooling flow clusters . however , tppl models which include intrinsic absorption generally provide fits of superior quality , and predict that the power - law component contributes 20 percent of the total flux . the photon indices of this component are typical of accreting x - ray sources and/or diffuse non - thermal x - ray emission which usually range from @xmath35 . we have fit the spectra of the cf clusters with the same variety of models used in the previous section , but also with an additional model ; tptppl . this model represents the emission from the observed region as the sum of the hot icm , a diminished cooling flow ( second tp component ) , and intermixed compact sources and/or extended non - thermal emission characterized by the pl component . simple tp models generally provide fairly reasonable fits to the spectra with 1.1 @xmath36 @xmath22 @xmath36 1.25 . derived temperatures and metallicities are in good agreement with previous studies ( white , jones , & forman 1997 , allen & fabian 1997 ; sarazin & mcnamara 1997 ; markevitch et al . 1998 ; allen 2000 ; ab2000 , allen et al . 2001 ) . none of the fits were significantly improved by letting the value of @xmath18 vary . hence , we find no evidence for excess absorption in these clusters , which is also consistent with several other investigations ( sarazin 1996 ; allen & fabian 1997 ; sarazin & mcnamara 1997 ; allen 2000 ; ab2000 ) . as anticipated , pure pl models ( with or without intrinsic absorption ) do not provide acceptable fits to the x - ray spectra of any of the cf clusters . typically , @xmath26 @xmath27 1.5 for this model . our tppl models fit the spectra of all 5 cf clusters very well , with 1.0 @xmath36 @xmath22 @xmath36 1.2 ( see table 4 ) . with the exception of a2029 , which is best fit by simple tp models , the pl component typically contributes 5 - 30 % ( mean of 20 % ) of the total flux when absorbed only by the galactic column . the inclusion of intrinsic absorption significantly improves the quality of the fit and raises the contribution of the pl component to 9 - 47 % ( mean of 27 % ) . the mean f - statistic value for the addition of this power - law component is @xmath37 which suggests it is significant ( @xmath38 99 % confidence ) . the luminosity ( 0.7 - 10.0 kev ) of pl components in these four clusters span @xmath39 ergs s@xmath3 , with a mean value of @xmath40 ergs s@xmath3 . the photon indices of pl components are reasonably well constrained and range from 1.0 @xmath41 2.4 ( mean of 1.8 ) ( see table 4 ) . the implied intrinsic column densities acting on the pl component are rather high for three of the cf clusters ( a644 , a1689 , & a2597 ) and are consistent with high - mass compact sources . as with the ncf clusters , a lower temperature is found for the thermal component than is found from the simple tp models . tptp models were also implemented . the quality of the fits provided by these models is generally comparable to that of tppl models . calculation of the f - statistic suggests that the addition of the extra tp component is significant . thus , we confirm previous reports that multi - phase thermal plasmas fit the spectra of cf clusters much better than that of isothermal models . the temperatures and metallicites of the two thermal components are consistent with the results of previous studies ( white , jones , & forman 1997 , markevitch et al . 1998 ; allen 2000 ; ab2000 ; allen et al . 2001 ) . for the cooler component in particular , we find @xmath42 and a mean contribution of roughly 22 % of the total flux , which is typical of cooling flow clusters ( peres et al . 1998 ) . the standard cooling flow model ( tpcf ) provides adequate fits to the spectra of the cf clusters , as expected . derived mass deposition rates , temperatures , and metallicities are generally consistent with previous studies ( white , jones , & forman 1997 ; peres et al . 1998 ; allen et al . 1999 ; allen 2000 ; ab2000 ) . the best fitting tpcf model to the spectrum of a644 is presented in figure 2 . only the asca sis0 and gis2 data are plotted . a comparison of the tpcf and tppl models is presented in table 4 . it is clear that the tppl models fit the spectra of the 5 cf clusters just as well as standard tpcf models and without any additional complexity . 0.1 in 0.1 in finally , the tptppl model also fits the x - ray spectra of these clusters well , with 1.0 @xmath36 @xmath22 @xmath36 1.2 . when absorbed by the galactic column only , the pl component typically contributes 2 - 5 % of the total flux inside 3 arcminutes . the addition of intrinsic absorption on the pl component , again , improves the fit and the mean contribution of the pl component is raised to @xmath43 32 % . when compared to the tppl model , the mean f - statistic value is about 5 ( @xmath43 68 % confidence ) which suggests the addition of a second tp component only improves the fit marginally . the luminosity ( 0.7 - 10.0 kev ) of the pl component ranges from @xmath44 ergs s@xmath3 with a mean value of @xmath45 ergs s@xmath3 . the power - law components have photon indices of 1.8 @xmath41 3.3 with a mean of 2.3 . with the exception of a2029 ( which has virtually no contribution from the pl component ) , the intrinsic column densities on the pl component are @xmath46 @xmath21 . the temperature of the cooler tp component did not change significantly from the value derived from tptp models . however , it s contribution to the flux was generally lower than that of tptp models ( mean @xmath0 10 % ) . in summary , we find that in all cases the tppl and tptppl models fit the rosat and asca spectra of the five cf clusters just as well as the standard cooling flow model ( tpcf ) or the two - temperature thermal plasma ( tptp ) . the pl components of these models are quite luminous ( contributing between 27 - 32 % of the total flux ) and characterized by photon indices typical of compact accreting x - ray sources and/or extended non - thermal emission . this , then , is the main conclusion of our work : _ rosat and asca spectra can be satisfactorily modeled by attributing a significant part of the luminosity within the central regions of cf clusters to sources other than cooling gas . _ our analysis shows that luminous ( l@xmath47 @xmath0 10@xmath48 ergs s@xmath3 ) power - law components with @xmath49 might be present in the centers of rich clusters . including intrinsic absorption dramatically improves the quality of the fit ; in these circumstances , there is an apparent dichotomy in the importance of the pl components between ncf and cf clusters . not only is the typical pl component in the cf cluster more luminous by about a factor of 10 than that of its ncf counterpart , but it also contributes a larger fraction of the total x - ray flux within the central 3 arcminutes of its cluster ( between 27 - 32 % for cf clusters as opposed to 20 % for ncf clusters ) . this may be consistent with the fact that the _ cores _ of cf clusters are more often observed to be active ( i.e. radio loud ) , perhaps implying the presence of a central massive black hole(s ) and a surrounding diffuse plasma of relativistic electrons , than the cores of ncf clusters ( burns 1990 ) . it is also worth noting that the improvement of the fit to the spectra of cf clusters far exceeds the improvement of the fit to the spectra of ncf clusters when a power - law component is included in the spectral model . the most attractive feature of the tppl models is that one does not expect large quantities of atomic or molecular gas to be present in the centers of clusters , nor does one expect to see emission lines below a few kev , because cooling is not significant . this is consistent with current observations . on the other hand , evidence which apparently supports the existence of cooling flows ( e.g. excess absorption , detection of co line emission in extreme cf clusters ) may contradict this model . however , the success of the single - temperature plasma + power - law model in fitting the spectra of the central regions of rich clusters , regardless of cooling flow status , has also been reported by a number of authors using a variety of x - ray instruments , including markevitch & vikhlinin ( 1997 ) ( asca and rosat data ) , markevitch et al . ( 1998 ) ( asca data ) , rephaeli , gruber & blanco ( 1999 ) ( rxte data ) , guainazzi & molendi ( 1999 ) ( beppo - sax data ) , fusco - femiano et al . ( 1999 ; 2000 ; 2001 ) ( beppo - sax data ) , and recently by sambruna et al . ( 2000 ) who used chandra to uncover a previously _ hidden _ agn in hydra a with @xmath50 ergs s@xmath3 and @xmath51 . however , our study is the first that systematically addresses the problems of the standard cooling flow model by implementing the single - temperature plasma + power - law model . the tptppl models , which might represent a compromise between the standard cooling flow model and our favoured tppl models , also provide good fits to the spectra of the cf clusters . however , the addition of the second , cooler tp component only marginally improves the fit and at the expense of adding much more complexity ( and uncertainty ) to the model . furthermore , the predicted temperatures and relative contributions of the cooler tp component in the tptppl models suggest that emission lines below a few kev should be present , however they have failed to turn up in chandra and xmm spectra . finally , because the mean contribution of the cooler component is @xmath0 10 % , as compared with 20 % in the tptp models , this implies that mass deposition rates are only roughly reduced by a factor of two . using models similar to our tptppl models , allen et al . ( 2001 ) reached the same conclusion for a number of clusters in their sample . these reduced deposition rates are still far too large to be consistent with the small amounts of atomic and molecular gas found in cooling flow clusters . we expect that chandra and xmm - newton will place much tighter constraints on the parameters of the complex tptppl models and it will be interesting to see if the models can be reconciled with problems of the standard cooling flow model . assuming that a power - law component is , indeed , present in the spectra of these clusters , some of the possible candidates are : * seyfert - like agn could contribute most , if not all , of the pl component luminosity found in our study . these objects can be quite luminous and are often observed to have @xmath52 @xmath53 ergs s@xmath3 with power - law indices of 1.6 - 2.0 ( nandra et al . 1997 ) . in fact , allen et al . 2001 have suggested that agn are the likely candidate for the observed pl component in their sample . likewise , the luminosities and steepness of the pl components in our sample are generally consistent with emission from a central agn(s ) . * diffuse non - thermal emission , which has possibly been seen in the central regions of coma , a2256 , and a2199 . searches of those clusters using beppo - sax and rxte have yielded very luminous ( @xmath43 10@xmath48 ergs s@xmath3 ) sources which are described by power - laws with @xmath54 and may possibly be linked to power - law components observed in the asca band ( rephaeli , gruber , & blanco 1999 ; fusco - femiano et al . 1999 ; 2000 ; 2001 ) . unfortunately , x - ray spectra usually have to be extracted from regions _ much _ larger than that of the central galaxy and , thus , it is difficult to draw conclusions about the source of the power - law component . a vla survey of more than 200 nearby clusters found that only 29 had extended radio emission on mpc scales ( giovannini , tordi , & feretti 1999 ) perhaps implying that extended non - thermal emission is not that common in galaxy clusters . it should be noted , however , that the detection of true intracluster radio emission is exceptionally difficult , as one must remove bright radio galaxies from the analysis ( rephaeli 2001 ) . furthermore , the non - thermal emission may be associated with the radio galaxies themselves ( e.g. in the radio lobes ) . * massive black holes associated with elliptical and lenticular galaxies . the regions sampled in our study generally include a single giant elliptical or cd galaxy plus a number of large galaxies , probably e or s0 galaxies . colbert & mushotzky ( 1999 ) and allen , di matteo , & fabian ( 2000 ) have recently discovered hard power - law components in several nearby ge galaxies with @xmath55 ergs s@xmath3 with photon indices of 0.5 - 3.0 . such sources are typically less luminous than seyfert nuclei , perhaps as a result of low radiative efficiency from an advection - dominated accretion flow ( adaf ; narayan & yi ) . while several of these sources could , in principle , account for the pl component in the ncf clusters they would have to be scaled up by a factor of 100 to account for the emission in the cf clusters . faint galaxies , on the other hand , are much more numerous and many of them might harbor less massive central black holes . the local group elliptical m32 , whose central object has @xmath56 erg s@xmath3 ( loewenstein et al . 1998 ) offers a possible guide to the x - ray luminosity of faint cluster members . hence @xmath57 such galaxies , each with a supermassive black hole , would be required to account for the power - law component of the clusters studied here . on this basis , faint galaxies seem incapable of producing significant power - law emission inside the regions we have observed . we note , however , that if massive black holes originally belonging to bright or faint galaxies are now scattered throughout intracluster space ( i.e. no longer associated with optically visible galaxies ) as the result of past mergers or stripping both of these limits may be relaxed . * galactic x - ray binaries , having @xmath58 ( white et al . 1988 , verbunt et al . 1995 ) , and which in our galaxy are most commonly associated with globular clusters . intergalactic globular clusters , if similar to globular clusters in the milky way , would have to be present in large numbers to collectively account for @xmath59 erg s@xmath3 attributable to the pl component in our models . assuming a typical intergalactic globular contains a binary x - ray source with @xmath60 @xmath61 ergs s@xmath3 ( hertz & wood 1985 ) and the specific frequency of globulars in the central ge galaxy is @xmath62 ( harris 1991 ) then even a very bright central galaxy with m@xmath63 = -24 would have a total x - ray luminosity of @xmath64 ergs s@xmath3 , too low by a factor of @xmath65 . even the most optimistic models of intracluster globular clusters of west et al . ( 1995 ) predict @xmath66 such clusters , significantly less than is required to account for the entire pl component . hence , it appears that there are probably two or three sources capable of producing the luminosity of the pl components of our models : massive black holes associated ( or once associated ) with large elliptical galaxies , diffuse non - thermal emission , and agn . does existing data favor one type of source over the others ? to determine the luminosity of any point sources within the central 3 arcminutes we used rosat hri data . two different procedures were implemented : i ) spectra of clearly visible point sources were extracted from a 5 region centered on the source and grouped into one large energy bin ( 0.5 - 2.0 kev ) . the spectra were then fit within the xspec package using a simple power - law model with @xmath67 and absorbed by the galactic column only . the 0.5 - 2.0 kev flux was then converted into a 0.7 - 10.0 kev unabsorbed flux using the program w3pimms . if more than one point source was present the individual fluxes were summed . as an example , figure 3 is the hri observation of a3667 , smoothed with a 5 gaussian beam . there are at least two bright compact sources within the central 3 arcminutes of this cluster . ; ii ) to place limits on the contribution of point sources which are hidden in the bright diffuse background ( what we refer to as `` barely visible '' point sources ) we calculated the mean standard deviation of counts per pixel ( within 3 ) and then scaled this to a standard deviation per circular region of diameter = 5 ( comparable to the hri psf ) . the latter value was expressed as unabsorbed flux ( 0.7 - 10.0 kev ) using the program w3pimms and assuming a simple power - law spectral model with @xmath67 absorbed only by the galactic column listed in table 1 . we regard the result as a 1-sigma limit to the contribution of any point sources . 0.1 in 0.1 in the results of hri data analysis are presented in table 5 . one or more point sources are easily visible within the central 3 arcminutes of all three ncf clusters having hri observations . it is clear from a comparison with the luminosities of the observed pl component from the best fitting tppl models ( with intrinsic absorption ) , however , that the bright point sources can not be solely responsible for the power - law emission , as they contribute only @xmath43 5 % of the pl component luminosity . significant contributions are required from low - luminosity point sources and/or diffuse non - thermal emission . the third column in table 5 ( method ii ) presents the 3-sigma limit of `` barely visible '' point sources , suggesting that at least @xmath68 of them are needed ( in addition to the visible point sources ) to account for the entire power - law component . there are no visible point sources within the central 3 of the 5 cf clusters studied . this is almost certainly the result of the cf clusters being much more centrally peaked than the ncf clusters . at least @xmath68 `` barely visible '' point sources are required to account for the observed pl component in these clusters . perhaps another way to differentiate between compact sources and diffuse non - thermal emission is through the presence or absence of flux variability , since most luminous accreting x - ray sources are known to vary significantly on short time scales ( e.g. nandra et al . 1997 ) . however , because the pl component typically contributes only @xmath69 % this might make detection of variability difficult in clusters ( especially in the cf clusters ) . furthermore , if the pl component is produced by @xmath68 point sources , and the variations in each are random , it will make the variation in the collective luminosity quite small , perhaps rendering this method unusable . if compact or diffuse non - thermal sources provide significant emission from the centers of rich clusters , it will be necessary to revise a number of conclusions derived from x - ray data . fully exploring the consequences is beyond the scope of the present paper ; however , we will mention a few important issues . allen ( 2000 ) reports that a deprojection analysis of rosat hri data yields cooling flow mass deposition rates that agree reasonably well with those derived from asca spectra . a consistent re - analysis would require first subtracting the power - law contribution from the x - ray images , but of course the spatial distribution of the power - law emission must first be understood . the presence of emission lines from heavy ions originally led astronomers to accept thermal emission as the dominant mechanism in rich cluster icms . accretion - driven x - ray sources do exhibit line emission ( kallman , vrtilek , & kahn 1989 ; narayan & raymond 1999 ) , but the lines are produced under quite different physical conditions than those within the canonical hot icm . the implications for icm abundance estimates and enrichment scenarios will require careful study . likewise , previously derived temperatures of the icm will also be affected ( as demonstrated in table 3 ) . might emission from compact sources account for the excess hard x - radiation recently discovered ( xu et al . 1998 ; ikebe et al . 1997 ) in the centers of several clusters ? the next generation of x - ray detectors and satellites ( i.e. xmm and chandra ) will be needed to map the x - ray emission with adequate spatial resolution to address this and other questions raised by the present work . we have shown that rosat pspcb and asca gis2/gis3 spectra of ten x - ray luminous clusters permit a significant contribution from compact sources and/or diffuse non - thermal emission . although this conclusion holds irrespective of the presence of a cooling flow , we find that power - law components are especially luminous in clusters with massive cooling flows . our study indicates that inferred cooling flow mass deposition rates may be lowered substantially , thus ameliorating , or possibly resolving , the long - standing problem of the missing cool gas . existing x - ray observations of large elliptical galaxies , agn , and extended non - thermal emission suggest that any of these sources could account for the observed luminosity of the power - law component . however , the absence of extremely bright point sources in the rosat hri observations rules out the possibility that the power - law emission comes from a few extremely bright compact sources . more likely is that most of the power - law emission is produced by some combination of diffuse and low - luminosity compact sources , such as massive black holes . the deep imaging capabilities and superior spectral resolution of chandra and xmm should be able to place much tighter constraints on the relative contributions of power - law sources and thermal bremsstrahlung to cluster x - ray emission and , perhaps , even determine the source of the power - law emission . we thank the rosat and asca help teams , in particular jane turner , for advice with data reduction techniques . we also thank paul nandra , peter thomas , arif babul , ann gower , pat henry , and alastair edge for useful comments and suggestions . igm acknowledges support from nserc of canada ( usra & pgsa ) . gaw and mjw were supported by nserc research grants . mjw also acknowledges support from nsf grant ast-0071149 . akylas , a. , georgantopoulos , i. , & comastri , a. , 2001 , mnras , 324 , 521 allen , s.w . , 2000 , mnras , 315 , 269 allen , s.w . , di matteo , t. , & fabian , a.c . 2000 , mnras , 311 , 493 allen , s.w . , et al . , 1993 , mnras , 262 , 901 allen , s.w . , et al . , 2001 , mnras , 322 , 589 allen , s.w . , & fabian , a.c . , 1997 , mnras , 286 , 583 arabadjis , j.s . , & bregman , j.n . , 2000 , apj , 536 , 144 ( ab2000 ) arnaboldi , m. , et al . , 1996 , apj , 472 , 145 arnaud , k.a . , 1996 , in asp conf . ser . 101 , astronomical data analysis software and systems v , ed . jacoby , g. , & barnes , j. , ( san francisco : asp ) , 17 balucinska - church , m. , & mccammon , d. , 1992 , apj , 400 , 699 bardelli , s. , zucca , e. , & baldi , a. , 2001 , mnras , 320 , 387 bauer , f. , & sarazin , c.l . , 2000 , apj , 530 , 222 bazzano , a. , et al . , 1984 , apj , 279 , 515 bhringer , h. , et al . , 1993 , mnras , 264 , l25 bhringer , h. , et al . , 2001a , a&a , in press ( astro - ph/0111112 ) bhringer , h. , et al . , 2001b , a&a , 365 , l181 burns , j.0 . , 1990 , aj , 99 , 14 calcneo roldn , c. , et al . , 2000 , mnras , 314 , 324 ciardullo , r. , et al . , 1998 , aj , 118 , 488 colbert , e.j.m . , & mushotzky , r.f . , 1999 , apj , 519 , 89 dickey , j.m & lockman , f.j . 1990 , ara&a , 28 , 215 donahue , et al . , 2000 , apj , 545 , 670 edge , a.c . , 2001 , mnras , in press ( astro - ph/0106225 ) fabian , a.c . , 1994 , ara&a , 32 , 277 fabian , a.c . , et al mnras , 318 , 65 fabian , a.c . , mushotzky , r.f . , nulsen , p.e.j . , & peterson , j.r . , 2001 , mnras , 321 , l33 feretti , l. , et al . , 1999 , a&a , 344 , 472 ferguson , h.c . , tanvir , n.r . , & von hippel , t. , 1998 , nature , 391 , 461 flores , r.a . , quintana , h. , & way , m.j . , 2000 , apj , 532 , 206 forman , w. , & jones , c. , 1982 , ara&a , 20 , 547 fusco - femiano , r. , et al . , 1999 , apj , 513 , l21 fusco - femiano , r. , et al . , 2000 , apj , 534 , l7 fusco - femiano , r. , et al . , 2001 , apj , 552 , l97 gregg , m.d . , & west , m.j . , 1998 , nature , 396 , 549 giovannini g. , tordi , m. , & feretti , l. , 1999 , new ast . , 4 , 141 guainazzi , m. , & molendi , s. , 1999 , a&a , 351 , l19 harris , w. e. , 1991 , ara&a , 29 , 543 harris , d.e . et al . , 2000 , apj , 530 , l81 heckman , t.m . , baum , s.a . , van breugel , w.j.m . , & mccarthy , p.j . , 1989 , apj , 338 , 48 hertz , p. , & wood , k.s . , 1985 , apj , 290 , 171 ikebe , y. , et al . , 1997 , apj , 481 , 660 irwin , j.a . , & sarazin , c.l . , 1995 , apj , 455 , 497 iwasawa , k. , et al . , 2000 , mnras , 318 , 879 jaffe , w. , 1990 , a&a , 240 , 254 jaffe , w. , & bremer , m.n . , 1997 , mnras , 284 , l1 kaastra , j.s . , 1992 , an x - ray spectral code for optically thin plasmas ( internal sron - leiden report , updated version 2.0 ) kallman , t.r . , vrtilek , s.d . , & kahn , s.m . , 1989 , apj , 345 , 498 katz , j.i . , 1976 , apj , 207 , 25 liedahl , d.a . , osterheld , a.l . , & goldstein , w.h . , 1995 , apjl , 438 , 115 loewenstein , m. , hayashida , k. , toneri , t. , & davis , d.s . , 1998 , apj , 497 , 681 lopez - cruz , o. , et al . , 1997 , apj , 475 , l97 markevitch , m. , forman , w.r . , sarazin , c.l . , & vikhlinin , a. , 1998 , apj , 503 , 77 markevitch , m. , et al . , 2000 , apj , 541 , 542 markevitch , m. & vikhlinin , a. , 1997 , apj , 491 , 467 mcnamara , b.r . , bregman , j.n . , & oconnell , r.w . , 1990 , apj , 360 , 20 mcnamara , b.r . , et al . , 2000 , apj , 534 , l135 mewe , r. , gronenschild , e.h.b.m . , & van den oord , g.h.j . , 1985 , a&as , 62 , 197 molendi , s. , & de grandi , s. 1999 , a&a , 351 , l41 molendi , s. , & pizzolato , f. , 2001 , apj , in press ( astro - ph/0106552 ) moore , b. , lake , g. , quinn , t. , & stadel , j. , 1999 , mnras , 304 , 465 mushotzky , r.f . , & szymkowiak , a.e . , 1988 , in cooling flows in clusters and galaxies , ed . fabian , ( dordrecht : kluwer ) , 53 nandra , k. , et al . , apj , 477 , 602 narayan , r. , & raymond , j. , 1999 , apj , 515 , l69 narayan , r. , & yi , i. , 1995 , apj , 452 , 710 odea , c.p . , et al . , 1994 , apj , 422 , 467 oegerle , w.r . , & hill , j.m . , 1994 , aj , 107 , 857 peres , c.b . , et al . , 1998 , mnras , 298 , 416 peterson , j.r . , et al . , 2001 , a&a , 365 , l104 rephaeli , y. 2001 , astro - ph/0101363 rephaeli , y. , gruber , d.e . , & blanco , p. , 1999 , 511 , l21 sarazin , c.l . , 1996 , in galactic and cluster cooling flows , ed . n. soker , ( san francisco : publ . pacific ) , 115 , 172 sarazin , c.l . , 2001 , in merging processes in clusters of galaxies , ed . l. feretti , m. gioia , and g. giovannini ( dordrecht : kluwer ) , in press ( astro - ph/0105418 ) sambruna , r.m . , et al . , 2000 , apj , 532 , 91 . sarazin , c. l. , burns , j.o . , roettiger , k. , & mcnamara , b.r . , 1995 , apj , 447 , 559 . sarazin , c.l . , & mcnamara , b.r . , 1997 , apj , 480 , 203 snowden , s.l . , et al . , 1992 , apj , 393 , 819 stark , a.a . , et al . , 1992 , apjs , 79 , 77 tamura , t. , et al . , 2001 , a&a , 365 , l86 thompson , l.a . , & gregory , s.a . , 1993 , aj , 106 2197 verbunt , f. , bunk , w. , hasinger , g. , johnston , h.m . , 1995 , a&a , 300 , 732 voit , g.m . & donahue , m. , 1995 , apj , 452 , 164 west , m. , ct , p. , jones , c. , forman , w. , & marzke , r. , 1995 , apj , 453 , l77 white , d.a . , et al . , 1991 , mnras , 252 , 72 white , d.a . , jones , c. , & forman , w. , 1997 , mnras , 292 , 419 white , n.e . , stella , l. , & parmar , a.n . , 1988 , apj , 324 , 363 xu , h. , et al . , 1998 , apj , 500 , 738 ccccccll a119 & 0.0442 & 3.17 & ncf & 0 & 5.3 & 3 bright radio galaxies near center@xmath70 + a644 & 0.0704 & 8.00@xmath71 & cf & 189 & 6.5 & minor merger event@xmath72 + a1689 & 0.1832 & 1.81 & cf & 645 & 8.7 & + a2029 & 0.0773 & 3.14 & cf & 556 & 7.4 & non - thermal emission not detected@xmath73 + a2063 & 0.0353 & 2.90@xmath71 & ncf & 37 & 4.1 & subcluster merger?@xmath74 + a2142 & 0.0909 & 4.16 & cf & 350 & 11.4 & merger@xmath75 & hard x - ray emission detected@xmath76 + a2597 & 0.0852 & 2.49 & cf & 271 & 9.1 & powerful cd radio galaxy@xmath77 + a3266 & 0.0589 & 3.00@xmath71 & ncf & 0 & 6.2 & major merger@xmath78 + a3532 & 0.0554 & 5.98 & ncf & 0 & 4.7 & merger@xmath79 + a3667 & 0.0556 & 4.00@xmath71 & ncf & 0 & 7.1 & hard x - ray emission found@xmath80 + a119 & tp & - & - & 0.987 & & - & - & 1.011 + & tpcf & 0.08 @xmath34 @xmath81 & 0 @xmath34 3 & 0.994 & & 0.01 @xmath34 151 & 0.6 @xmath34 5 & 1.026 + a2029 & tptp & 1.29 @xmath34 0.15 & - & 1.592 & & 1.19 @xmath34 0.31 & - & 1.499 + & tpcf & 0.08 @xmath34 6.97 & 132 @xmath34 31 & 1.911 & & 0.03 @xmath34 4.82 & 170 @xmath34 43 & 1.772 + a2142 & tpcf & 0.08 @xmath34 7.41 & 188 @xmath34 43 & 1.190 & & 0.10 @xmath34 9.16 & 230 @xmath34 73 & 1.130 + a119 & tp & @xmath82 & & & & 467.4/446 + & tppl & @xmath83 & @xmath84 & 25 % & @xmath85 & 454.0/443 + a2063 & tp & @xmath86 & & & & 945.6/807 + & tppl & @xmath87 & @xmath88 & 8.5 % & @xmath89 & 933.3/804 + a3266 & tp & @xmath90 & & & & 1160.5/1117 + & tppl & @xmath91 & @xmath92 & 23 % & @xmath93 & 1158.0/1114 + a3532 & tp & @xmath94 & & & & 776.4/746 + & tppl & @xmath95 & @xmath96 & 87.5 % & @xmath97 & 756.0/743 + a3667 & tp & @xmath98 & & & & 809.0/791 + & tppl & @xmath99 & @xmath100 & 22 % & @xmath101 & 797.6/788 + a644 & tppl & & @xmath102 & 35 % & @xmath103 & 1855.7/1613 + & tpcf & @xmath104 & & & & 1841.2/1613 + a1689 & tppl & & @xmath105 & 9 % & @xmath106 & 1643.6/1409 + & tpcf & @xmath107 & & & & 1653.9/1409 + a2029 & tppl & & @xmath108 & 5 % & @xmath109 & 1460.3/1149 + & tpcf & @xmath110 & & & & 1462.1/1149 + a2142 & tppl & & @xmath111 & 47 % & @xmath112 & 1010.4/868 + & tpcf & @xmath113 & & & & 1013.2/868 + a2597 & tppl & & @xmath114 & 17 % & @xmath115 & 1305.1/1281 + & tpcf & @xmath116 & & & & 1302.3/1281 + a119 & @xmath117 & @xmath118 & @xmath119 + a644 & & @xmath120 & @xmath121 + a1689 & & @xmath122 & @xmath123 + a2029 & & @xmath124 & @xmath125 + a2142 & & @xmath126 & @xmath127 + a2597 & & @xmath128 & @xmath129 + a3266 & @xmath130 & @xmath131 & @xmath132 + a3667 & @xmath133 & @xmath134 & @xmath135 +

the perturbative nonrenormalizability of the theory of general relativity ( gr ) is one of the greatest obstacles against proposing quantum gravity theory that should be correct at all scales . the latter belongs to the yet - unsolved problems in the fundamental physics @xcite . weakness of the gravitational force is another difficulty preventing the quantum effects of gravity to be detectable at the currently - probed scales . the recent detection of the gravitational waves is strictly limited to pulsars in binary star systems @xcite . accordingly , the resulting gravitational waves become very great . at a gaussian fixed point , gr does not satisfy the perturbative renormalizability because the gravitational coupling ( @xmath0 ) is a dimensioned quantity ( @xmath1 ) and accordingly the ultraviolet ( uv ) divergences are uncontrollable . by introducing higher terms of the scalar curvature , the action shall be cured . on the other hand , this introduces an additional problem , for instance , to the uniqueness @xcite . from a minimal approach that will be elaborated below , which assures perturbatively renormalizable uv completion of gr as quantum field theory ( qft ) for @xmath2-dimensional metric field , one looses some properties of gr and creates additional degrees - of - freedom . such an approach benefits from the success of qft in describing all forces ( except gravity so - far ) and the standard model of the elementary particles . the scalar field theory , that was proposed by lifshitz in order to explain the quantum critical phenomena in condensed matter @xcite , inspired horava to propose a theory for the quantum gravity with an anisotropic scaling in uv . this new approach is thus known as horava - lifshitz gravity ( hlg ) @xcite , where the dynamical critical exponent ( @xmath3 ) makes the theory non - relativistic . in analogy to critical systems , the uv fixed point has been treated @xcite . the hlg approach assures causal dynamical triangulations @xcite , renormalization group approaches based on asymptotic safety @xcite and symmetries of gr . this approach was discussed in an extensive amount of literature @xcite , where problems such as , internal consistency and compatibility with observations are treated , as well . the new dimensionless coupling @xmath4 is assumed to approach unity in the infrared ( ir ) limit with the argumentation that at the hlg - charactering parameter @xmath5 , hlg theory ca nt be reduced to gr . for the sake of completeness , we mention that this is not always the case for all hlg models @xcite . recent astrophysical observations obviously indicate that the universe expands with an accelerating rate , probably due to dark energy @xcite . besides inflation , a series of symmetry - breaking phase transitions , where topological defects may have been formed @xcite , were conjecture to take place . on the other hand , the hlg - type approach @xcite proposes that the modification in gravity is responsible for the accelerated expansion of the universe . furthermore , various modified theories for gravity can be implemented @xcite . they successfully unify various inflation models with the late - time acceleration and cosmological observations . but when endorsing any proposal about ( non)singular big bang due to controversial quantum gravity approaches such as generalized uncertainty principle ( gup ) and modified dispersion relation ( mdr ) @xcite should be seen as unwarranted claims . the big bang singularity is a fundamental problem , which should be tackled by a good quantum gravity theory , such as hlg or even better . the present paper is organized as follows . the cosmological evolution governed by a simple version of hlg shall be introduced . the horava - lifshitz gravity is introduced in section [ sec : hlg ] . the field equations in friedmann - lemaitre - robertson - walker ( flrw ) cosmology are elaborated in section [ sec : fe ] . section [ sec : nondet ] and section [ sec : det ] are devoted to horava - lifshitz gravity with non - detailed and detailed balance conditions , respectively . for the different equations - of - state ( eos ) , different solutions for the resulting differential equations shall be outlined . an extensive comparison between scale factor in cosmological radiation , matter , @xmath6cdm , de sitter , chaplygin gas @xcite and quantum chromodynamical ( qcd ) eos shall be introduced . the results and discussion are given in section [ sec : disc ] . to section [ sec : conc ] the final conclusions are assigned . the basic assumptions of horava - lifshitz gravity are antisotropic uv scalings between minkowskian space and time @xmath7 where @xmath3 is the dynamical critical exponent given in uv , @xmath8 is constant by which the scaling is performed and to the critical exponent the value @xmath9 is assigned . because the anisotropic scaling implies a preferred time coordinate , the usual @xmath2-dimensional lorentzian metric ca nt be the only fundamental structure on a pseudo - riemannian ( lorentzian ) manifold . the original theory postulated @xmath3 to take the value assuring dimensionless gravitational coupling , which in turn guarantees power - counting renormalizability . at @xmath10 , the symmetries of gr are recovered . in order to single out a spacial ( temporal ) coordinate in a differentiable manifold , a codimension foliation on the manifold as a basic structure of the theory was proposed @xcite . the foliation structure fulfils local galilean invariance but implies impossible full - diffeomorphism invariance in gr . the latter means the existence of an undo one - to - one mapping @xmath11 and its inverse besides their differentiabilities . thus , gr may be considered as an emerging in an infrared fixed point @xcite . the hlg is a projectable approach minimizing the number of independent couplings in the potential and adopting an extra principle to contract the potential , i.e. detailed balance conditions . in constructing the theory , a new set of symmetries should be introduced . imposing invariance under foliation - preserving diffeomorphisms , @xmath12 and @xmath13 would define field content , which paradoxically not always the one in gr and would make the theory violates the lorentz invariance principle . on the other hand , for arnowitt , deser and misner [ adm ] formalism @xcite ( gr with global time foliation , @xmath14-dimensional metric of the spatial hypersurfaces gives all the dynamics ) , the metric is given as @xmath15 where @xmath16 , and @xmath17 ( @xmath18 ) being lapse function ( shift @xmath14-vector ) , which are the gauge fields of the diffeomorphism group . the lapse variable is taken to be just time - dependent so that the projectability condition holds and by using the foliation - preserving difeomorphisms @xmath19 can be fixed to unity . the variables in eq . ( [ adm1 ] ) are dynamical so that under the above scaling @xmath20 in terms of the metric eq . ( [ adm1 ] ) , the ricci scalar reads @xmath21 where @xmath22 , @xmath23 is the extrinsic curvature of spatial slices , @xmath24 is the spacial scalar curvature , and @xmath25 is a unit vector perpendicular to a hypersurface of constant time . @xmath26 is the covariant derivative on the spatial slice . the hlg action is decomposed of kinetic , potential and matter parts . * the kinetic action can be built with at most two time - derivatives of the metric . this is a universal and most general invariant term . the new dimensionless coupling @xmath4 , whose role in the theory is still a subject of debate , differentiates the hlg kinetic term from the gr one @xmath27 when combining this kinetic term with the line element and the anisotropic scaling of space and time , the scaling dimension of the gravitational coupling can be determined . * the potential action term @xcite ( gr contains only two derivatives ) is affected by the choice between projectable and non - projectable theories @xcite @xmath28 . \label{eq : termp}\end{aligned}\ ] ] the coefficients are given in just three parameters , @xmath29 , @xmath30 and @xmath31 @xcite @xmath32 where @xmath33 and @xmath30 are einstein coupling and cosmological constant , respectively . the lower - order variables in expression ( [ eq : termp ] ) ; the cosmological constant and the ricci scalar , match the ones in gr . * the matter part of the action @xcite is given as @xmath34 where @xmath35 is the lagrangian density of matter fields @xmath36 . in ref . @xcite , a simple scenario for the scale - invariant quantum fluctuations was introduced . this was based on hlg without an additional scalar degree - of - freedom . accordingly , it is believed that the inflation is self - consistently existing in the early universe . for this simple scenario , the detailed balance conditions were not necessary but the inflation scenario itself may be still or no longer needed , for instance , serious horizon problems remain unsolved , such as monopole and domain walls . they still require inflation with slow - roll conditions . later on many authors seem to disagree with ref . the inflation was studied in hlg without the projectability conditions @xcite . but , opposite to the proposal of ref @xcite , the linear scalar perturbations equations of the flrw universe are derived for a single scalar field . a master equation of the perturbations has been specified for a particular gauge . the power spectrum and the spectrum index of the comoving curvature perturbations have been determined . it was noticed that the perturbations remain scale - invariance , and hlg without the projectability conditions is consistent with all current cosmological observations . this is another solid support for adding scalar field(s ) . in framework of nonrelativistic hlg with the projectability conditions and an arbitrary coupling constant @xmath4 , the inflation has been studied @xcite . accordingly , the flrw universe without specifying the gauge is necessarily flat . but by adding a single scalar field , it was noticed that both metric and scalar field become strongly coupled and almost identically oscillating in sub - horizon regions . in super - horizon regions , the comoving curvature perturbation remains constant although the flrw perturbations become adiabatic . furthermore , the perturbations of the slow - roll parameters , for instance , both scalar and tensor are found scale - invariant . concrete tuning the coupling coefficients makes the spectrum index of the tensor perturbation identical as that in gr . but the ratio of scalar to tensor spectra can be similar to that from gr and seems to depend on the spatial higher - order derivative terms . furthermore , the emergence of finite - time future singularities has been studied in ref . it was found that such singularities can be cured by adding a higher - order spatial derivative term . recently , the general formulas for the inflationary power spectra of scalar and tensor are driven in the presence of a scalar field @xcite . we conclude that in simple scenarios for scale - invariant cosmological perturbations with hlg , the inflation is not necessarily guaranteed . for a cosmological context , we recall flrw metric , which is an exact solution of the einstein s field equations of gr , @xmath37 where @xmath38 is the scale factor and @xmath39 . accordingly , the homogeneous and isotropic metric is given as @xmath40,\end{aligned}\ ] ] with the curvature constant @xmath41 , @xmath42 and @xmath43 represents open , flat or closed universe , respectively . in hlg with non - detailed balance conditions and by varying @xmath19 and @xmath44 , the friedman equations can be extracted @xcite , @xmath45 , \label{13 } \\ ( 3 \lambda - 1 ) \left(\dot{h } + \frac{3}{2 } h^{2}\right ) & = & - \frac{\kappa^{4}}{4 } p - 3 \left[-\frac{\sigma}{6 } - k \frac{\gamma}{3 a^{2}}+ 2 k^{3 } \frac{9\ , \sigma_{1 } + 3\ , \sigma_{2 } + \sigma_{3}}{a^{6}}\right ] . \label{14}\end{aligned}\ ] ] by substituting with the variables given in eqs . ( [ eq : variables ] ) , ( [ 13 ] ) and ( [ 14 ] ) , the freidmann equations read @xmath46 - \dfrac{\kappa^4\ , \mu^2\ , \lambda_{\omega}^2}{16(3 \lambda -1)^2}\right\}\ ; a + \dfrac{k^2\ , \kappa^4\ , \mu^2}{32(3 \lambda - 1)^2 } a^{-3}. \label{21b}\end{aligned}\ ] ] based on equations of state ( eos ) describing the cosmological background geometry , this differential equation can be solved . from the continuity equation @xmath47 and with the eos @xmath48 , we get @xmath49 and @xmath50 , where @xmath51 - in natural units - is a dimensionless quantity . it is related to the speed of sound squared within the system of interest . then , eq . ( [ 21b ] ) can be expressed as @xmath52 equation ( [ 20 ] ) can be rewritten as & = & - h^2 - p - l + , [ 20pert ] where @xmath53 $ ] and @xmath54 $ ] . assuming perturbation in eq . ( [ 20pert ] ) leads to + h & = & - h^2 - 3 h h- p - l + - 2 . [ 20pert2 ] for an eos in which @xmath55 and for the first - order perturbation , we get h & = & - 3 h h - 2 . [ 20pert3 ] however from the continuity equation , eq . ( [ eq : rhodot ] ) , & & = -3 ( 1 + ) . then h & = & m h + n , + & = & p h + q , with @xmath56 , @xmath57 $ ] , @xmath58 , and @xmath59 . thus , h - ( m + + q ) h - ( - m + p n - q m ) h & = & 0 , for @xmath60 , the term with the first - order perturbation reads @xmath61 is apparently positive . this gives a stable solution irrespective to the third term in left - hand side . in describing the cosmic geometry , we wanted to analyse systematically the effects of implementing various equations - of - state . * when the cosmological geometry is dominantly filled with matter ( dust approximation ) , @xmath62 , eq . ( [ 19 ] ) can be rewritten as + @xmath63 + when assuming that the cosmological constant @xmath30 and einstein coupling @xmath64 are negligibly small , eq . ( [ eq : add0 ] ) gets the solution @xmath65 where @xmath66 $ ] and @xmath67 $ ] . deducing scale factor does not restrict the applicability of the present analysis . other cosmological quantities such as hubble and deceleration parameters are good examples . * for cosmological geometry filled with radiation ( perfect fluid approximation ) , @xmath68 , eq . ( [ 19 ] ) becomes @xmath69 which has the following solution @xmath70 where @xmath71 - k^2 \kappa^4 \mu^2/[32(3 \lambda - 1)^2]$ ] and @xmath72 $ ] and @xmath73 . * for cosmological geometry filled with cold dark - matter , @xmath74 , eq . ( [ 19 ] ) is written as , @xmath75 the solution looks similar to the one in eq . ( [ solsim ] ) @xmath76 * for cosmological geometry filled with vacuum ( dark ) energy ( cosmological constant ) , @xmath77 , then eq . ( [ 19 ] ) can be written as @xcite , @xmath78 then , the solution reads @xmath79 * for cosmological geometry filled with chaplygin gas @xcite , which is a hypothetical ( non - baryonic ) substance satisfying @xmath80 , where @xmath81 is positive and @xmath82 ranges between @xmath42 and @xmath43 , i.e. negative pressure , and does not cluster , gravitationally . it was named after sergey chaplygin ( 1869 - 1942 ) , and expresses energy of quantum vacuum , and gravity modification . it might signal extra - dimensions and is conjectured as a candidate for dark energy . the energy density and pressure , respectively , read @xmath83^{1/(1+\alpha ) } , \label{eq : chplp}\\ p & = & -a\ , \left[a+\frac{b}{a^{3(1+\alpha)}}\right]^{-\alpha/(1+\alpha ) } , \label{eq : chple}\end{aligned}\ ] ] where @xmath84 is an integration constant . + in flrw cosmology , the chaplygin equations of state , eqs . ( [ eq : chplp ] ) and ( [ eq : chple ] ) , are substituted in eqs . ( [ 19 ] ) and ( [ 21b ] ) to give @xmath85 \right.{\nonumber}\\ & & \left . - \dfrac{\kappa^4\ , \mu^2\ , \lambda_\omega^2}{16(3\lambda-1)^2}\right\}\ , a + \dfrac{k^2\ , \kappa^4\ , \mu^2}{32(3\lambda-1)^2}\ , a^{-3}. \label{eq : daachapa0}\end{aligned}\ ] ] + when ignoring the taylor terms with orders @xmath86 , then @xmath87 + the possible solution for this differential equations depends on the choices of @xmath81 and @xmath82 . here , we elaborate different cases : * * first for a modified chaplygin gas @xcite with @xmath88 and by assuming that @xmath89 , then eq . ( [ eq : daachapa1 ] ) is reduced to @xmath90 - \dfrac{\kappa^4\ , \mu^2\ , \lambda_{\omega}^2}{16(3\ , \lambda-1)^2}\right\}\ , a + \dfrac{k^2\ , \kappa^4\ , \mu^2}{32(3\lambda-1)^2}\ , a^{-3 } , \label{eq : addchapb1 } \\ \dot{a}^2 & = & \left[\dfrac{\kappa^2\ , a^{3/4}}{6(3\lambda-1 ) } - \dfrac{\kappa^4\ , \mu^2\ , \lambda_{\omega}^2}{16(3\ , \lambda-1)^2}\right ] a^2 + \left[\dfrac{\kappa^2\ , b a^{-1/4}}{8(3\lambda-1 ) } - \dfrac{k^2\ , \kappa^4\ , \mu^2}{16(3\lambda-1)^2 } \right ] a^{-2 } { \nonumber}\\ & + & \dfrac{\kappa^4\ , k \mu^2 \lambda_{\omega}^2}{8 ( 3 \lambda - 1)^2},\end{aligned}\ ] ] which has a solution @xmath91,\end{aligned}\ ] ] where @xmath92 , { \nonumber}\\ b_3 & = & \left[\dfrac{\kappa^2\ , b a^{-1/4}}{8(3\lambda-1 ) } - \dfrac{k^2\ , \kappa^4\ , \mu^2}{16(3\lambda-1)^2 } \right ] . { \nonumber}\end{aligned}\ ] ] * * for matter , the pressure vanishes or @xmath93 . ( [ eq : daachapa1 ] ) is reduced to @xmath94 which has a solution @xmath95 * for cosmological geometry filled with qcd matter , the eos can be deduced from the recent lattice qcd simulations @xcite , where @xmath96 ranges between @xmath97 and @xmath98 . for sake for completeness , we highlight the importance of precise estimation of eos from heavy - ion collisions @xcite . by substituting the deduced values of @xmath51 in eq . ( [ 21a ] ) , the resulting differential equation turns to be unsolvable . for simplicity , the first term in eq . ( [ 21a ] ) can be approximated , @xmath99\ ; a^{-3 } - \dfrac{\kappa^4 \mu^2 \lambda_w^2}{16(3 \lambda -1)^2}\ ; a , \\ \dot{a}^2 & \approx & \left [ \frac{\kappa^2\ , \rho_0}{16 ( 3 \lambda -1 ) } - \dfrac{k^2 \kappa^4 \mu^2}{32(3 \lambda - 1)^2}\right]\ ; a^{-2 } - \dfrac{\kappa^4 \mu^2 \lambda_w^2}{16(3 \lambda -1)^2}\ ; a^2 . \end{aligned}\ ] ] similar to eq . ( [ eq : awm1 ] ) , the solution reads @xmath100}{2\ , \sqrt{b_2}\ ; { \cal a}},\end{aligned}\ ] ] where @xmath101^{1/2}$ ] . the detailed balance condition is a technical trick proposed to reduce the couplings . furthermore , it was argued that hlg with detailed balance conditions has to be broken in order to enable the theory to be compatible with the observations @xcite . but , according to horava s remarks , its restrictive power remains useful @xcite . there is a possible connection between detailed balance conditions and the entropic origin of gravity @xcite . the detailed balance condition is often believed to be abandoned with the aim to obtain an ultraviolet stable scalar field in the theory . but because of its several attractive features , we wanted it to be implemented in constructing the potential instead of effective field theory . many authors wanted to improve it by adding extra term to the action of the theory that softly violates the detailed balance conditions @xcite as an attempt to assure a more realistic theory in its infrared - limit . it was shown @xcite that the properties of the extra scalar degrees - of - freedom are the reasons why detailed balance conditions do nt lead to gr in the infrared - limit . from perturbation study of detailed balance conditions @xcite , two different strong coupling problems have been identified . the first one explains why the gr solutions are typically not recovered . the second is not necessarily associated with detailed balance conditions but refers to breaking of diffeomorphism invariance , which is required for anisotropic scaling in uv . the authors of ref . @xcite claimed to uncover an additional mode in the detailed balance conditions satisfying an equation of motion that is of first - order in time derivative . but , the proposed mode leads to very fast exponential instabilities at short distances and becomes strongly coupled at an extremely low cutoff scale . for hl gravity with detailed balance conditions , the field equations read @xcite @xmath102 the continuity equations @xmath103 can be substituted in and the friedmann equation can be derived from the action @xmath104 where @xmath105 , \\ \ell_1 & = & \sqrt{g } n \left[\dfrac{\kappa^2 \mu^2 ( 1- 4\lambda)}{32(1 - 3\lambda ) } r^2 - \dfrac{\kappa^2}{2 w^4 } \left(c_{ij } - \dfrac{\mu w^2}{2 } r_{ij } \right ) \left ( c^{ij } - \dfrac{\mu w^2}{2 } r^{ij } \right ) \right ] , \end{aligned}\ ] ] and @xmath4 , @xmath64 , @xmath29 , @xmath51 and @xmath30 are constants , and @xmath106 being ricci scalar ( tensor ) . @xmath107 is the cotton tensor , @xmath108 . when comparing hlg @xmath109 to that of gr in adm formalism , then the speed of light and gravitational constant , respectively , reads @xmath110 when assuming loss of generality , @xmath111 . in solving eqs . ( [ 26 ] ) and ( [ 27 ] ) four cases are likely to occur : * if @xmath112 , and @xmath113 @xmath114 where @xmath82 is constant of integration ( arbitrary constant ) . * if @xmath115 , and @xmath116 @xmath117 * if @xmath118 @xmath119 * if @xmath115 , and @xmath113 @xmath120 with @xmath121 . the solutions of the differential equations ( [ 26 ] ) and ( [ 27 ] ) depend on the eos charactering the cosmological background geometry . this shall be elaborated in the section that follows . at finite cosmological constant , the friedmann equations read @xcite @xmath122.\end{aligned}\ ] ] with continuity equation , this equation becomes @xmath123 due to various equations - of - state , this differential equation can be solved . * at @xmath62 , eq . ( [ eq : bdhrgadd ] ) can be reduced to @xmath124 then , the solution reads @xmath125 where @xmath126 $ ] and @xmath127 $ ] . * at @xmath68 , we obtain @xmath128 which can be solved as follows . @xmath129 , \label{eq : dtl2}\end{aligned}\ ] ] where @xmath130 and @xmath131 + k^2/[\lambda_{\omega } ( 3\ , \lambda-1)]$ ] an d @xmath132 . * at @xmath74 , it results is @xmath133 , \label{eq : dtldt3}\end{aligned}\ ] ] with the solution @xmath134 , \label{eq : dtl3}\end{aligned}\ ] ] where @xmath135 $ ] . * at @xmath77 , we get @xmath136 with the solution @xmath137 , \label{eq : dtl4}\end{aligned}\ ] ] * for the generalized chaplygin gas @xcite , it becomes @xmath138 , \label{eq : dtl5}\end{aligned}\ ] ] where @xmath139 $ ] . + in a matter - dominated universe , i.e. @xmath140 or @xmath93 , then @xmath141 and @xmath142 similar to eq . ( [ eq : dtldt1 ] ) , @xmath30 is assumed to be negligibly small @xmath143 where @xmath144 $ ] . * for cosmological geometry filled with qcd matter , the eos can be deduced from the recent lattice qcd simulations @xcite . as given in fig . [ fig : peqcd ] , @xmath96 ranges between @xmath97 and @xmath98 . for simplicity , the second term in eq . ( [ eq : bdhrgadd ] ) can be approximated , @xmath145\ , a^{-3 } + \frac{\lambda_{\omega}}{3\ , \lambda-1}\ , a. \end{aligned}\ ] ] similar to eq . ( [ eq : dtl2 ] ) , the solution reads @xmath146}{2\ , \sqrt{d_2}\ ; { \cal b}_7 } , \label{eq : dtl2}\end{aligned}\ ] ] where @xmath147^{1/2}$ ] and @xmath130 and @xmath148 $ ] . based on studying a @xmath4-dependent version of the friedmann equations , it was concluded that hlg becomes an attractive gravity theory if the generalized wheeler - dewitt metric @xmath149 has an indefinite signature . but , it becomes a repulsive gravity theory when the metric gets a positive sign . these results are also found in our calculations . the various equations - of - state lead to different behavior of the scalar factor as an example about the cosmological quantities . in fig . [ fig:1 ] , the cosmic time ( @xmath150 ) is given in dependence on the scale factor ( @xmath151 ) . in these calculations all parameters are kept fixed , @xmath152 and @xmath153 , @xmath154 and @xmath155 . correspondingly , neither @xmath150 nor @xmath151 is given in physical units . we compare between hlg with non - detailed and detailed balance conditions and gr gravity for various equations - of - state . it is worthwhile to notice that flrw which is characterized by gr gravity is non - singular for four types of equations of state ; @xmath68 ( a ) , @xmath62 ( b ) , @xmath74 ( c ) and @xmath77 ( d ) . the singularity in the flrw cosmology , which can be characterized by hlg with non - detailed and detailed balance conditions , varies due to the equations of state . such conclusions are only valid for the given parameters . furthermore , the dependence of @xmath150 on @xmath151 apparently varies with the equation of state , as well . again , such conclusions are explicitly valid for the given parameters . in other words , absence or presence of singular solutions and even the dependence of @xmath150 on @xmath151 strongly depend on the equation of state and the gravity itself . the non - singular solutions is obvious , as @xmath30 is finite . the top - left panel shows the results at @xmath68 ( a ) , i.e. radiation - dominated ( perfect fluid approximation ) flrw universe . the hlg with non - detailed balance conditions , eq . ( [ eq : ndwp13 ] ) leads to a very rapid increase in @xmath151 , almost a parabolic shape . at a certain value of @xmath150 , the scale factor became @xmath150-independent , i.e. @xmath151 diverges . the results from hlg with detailed balance conditions , eq . ( [ eq : dtl2 ] ) , and gr , eq . ( [ eq : w13std ] ) , are also compared with . the bottom - left panel shows the results at @xmath62 ( b ) , i.e. matter - dominated ( dust approximation ) flrw universe . again , the hlg with non - detailed , eq . ( [ eq : add0a ] ) , and detailed balance conditions , eq . ( [ eq : bhlg1 ] ) , lead to a rapid increase in @xmath151 ; an almost a linear dependence . also here , the gr gravity , eq . ( [ eq : egw1 ] ) , has a non - singular solution . the bottom - right panel presents the results at @xmath77 ( d ) , i.e. vacuum or dark energy or finite cosmological constant . we find that the validity of hlg with non - detailed balance conditions is very limited . also , hlg with detailed balance conditions shows a very rapid change in @xmath151 against @xmath150 . the gr gravity results in an inflationary phase followed by a slower expansion . despite the non - singular solutions , we notice that the expansion depicted in the right - hand panel is faster than the one illustrated in the left - hand panel . again , the universe seems to start up its evolution from a very large scale factor . figure [ fig:2 ] presents the results from chaplygin ( a ) and qcd ( b ) equations of states . the results of the chaplygin gas strongly depends on the parameter @xmath81 . the hlg with detailed balance conditions results in an almost linear dependence of @xmath151 and @xmath150 . the non - singular solutions appear in both hlg approaches and equations of state . we conclude that the difference between the chaplygin and qcd results is due to non - baryonic and baryonic equations - of - state , respectively . while the present article was under review , we have completely conducted and published a research paper in which the proposed solutions are confronted to recent planck and bicepii observations @xcite . with single scalar field potentials describing power - law and minimal - supersymmetrically extended inflation , we have derived possible modifications in the friedmann equations . for various eos , the dependences of the tensorial and spectral density fluctuations ( and their ratio ) on the inflation field are characterized . the tensorial - to - spectral density fluctuations are calculated with varying spectral index . it was found that , they decrease when moving from hlg with non - detailed balance conditions , to friedmann gravity , to hlg without the projectibility conditions , and to hlg with detailed balance conditions . such a pattern remains valid for various eos and different inflation potential models . the calculations fit well with the recent planck observations @xcite . even , when studying the consequences of the quantum fluctuations on our understanding of landau - raychaudhuri equations , we found that cosmic eos play an essential role @xcite . a theory for quantum gravity with an anisotropic scaling in uv based on a scalar field theory was proposed by horava , who was inspired by lifshitz theory for the quantum critical phenomena in condensed matter physics . horava benefited from the success of qft in describing all forces ( except gravity ) and the standard model for the particle physics . we study the impacts of various equations - of - state on flrw cosmology based on horava - lifshitz gravity . we also implement the so - called hlg with detailed balance conditions , which was thought as a technical trick to reduce couplings in the theory . from the possible connection between detailed balance conditions and the entropic origin of the gravity , the constructed potential can be used instead of adopting an effective field theory . both are compared with the gr gravity . in this work , we have studied the dependence of scale factor ( @xmath151 ) and cosmic time ( @xmath150 ) in the context of horava - lifshitz gravity in early universe by using different equations - of - state . we have compared the results with the gr gravity . remarkable differences from what is predicted by gr are found . also , a noticeable dependence on the equations of state is observed in singular and non - singular big bang . these observations are precautionarily corresponding to the fixed parameters @xmath152 and @xmath153 , @xmath154 and @xmath155 . the results presented in present paper are strongly depending on these parameters , for instance , the non - singular solutions can be to a large extend understood due to finite cosmological constant , @xmath30 . the present work presents a systematic analysis for the equations of state characterizing the cosmic geometry and proposes a link between the conventional gravity by gr and its horava - lifshitz counterparts . the latter are power - counting renormalizable quantum theories which assure causal dynamical triangulations , renormalization group approaches based on asymptotic safety and symmetries of gr . the horava - lifshitz gravity introduces a new set of symmetries imposing invariance under foliation - preserving diffeomorphisms . various types of equations of state are implemented . in future works , we intend to extend this study to cover various topics , such as comparison with other theories such as the gravity s rainbow @xcite , and determining the possible impacts of the generalized uncertainty principle @xcite . also , we plan to study third quantization of this theory @xcite , such as in a simple cosmological model @xcite . we conclude that hlg can explain various epochs in the early universe . furthermore , hlg might be able to reproduce the entire cosmic history with and without singular big bang . it intends to study the stability of the given flrw solutions and then analyse the universe evolution scenarios , especially that hlg is conjectured to be a candidate for a quantum field theory of gravity , despite its yet - unresolved problems . for eos @xmath48 and friedmann equations with finite cosmological constant , the continuity equation reads @xmath156 accordingly , we get following solutions @xmath157}{3\ , \sqrt{b}\ , { \cal c}_1 } , \label{eq : egw1 } \\ \mathrm{at}\ ; \omega=\frac{1}{3},\;\;\ ; t & = & \frac{1}{\sqrt{b } } \ln \left[a^2 - \dfrac{2a}{b } \right ] , \label{eq : w13std } \\ \mathrm{at}\ ; \omega=-\frac{1}{3},\;\;\ ; a(t ) & = & c_1 \exp\left(\sqrt{b}\ , t\right ) + c_2 \exp\left(-\sqrt{b}\ , t\right ) , \\ \mathrm{at}\ ; \omega=-1:\;\;\ ; a(t ) & = & c_3 \exp\left(\sqrt{c}\ , t\right ) + c_4 \exp\left(-\sqrt{c}\ , t\right ) , \\ \mathrm{and}\ , \mathrm{for}\ , \mathrm{generalized}\ , \mathrm{chaplugin}\ , \mathrm{gas},\ ; a(t ) & = & c_5 \exp\left(\sqrt{d}\ , t\right ) + c_6 \exp\left(-\sqrt{d}\ , t\right ) , \ ] ] where @xmath158 , @xmath159 , @xmath160 , and @xmath161 , @xmath162 . @xmath163 are integration _ arbitrary _ constants . from the recent lattice qcd simulations @xcite , we illustrate in fig . [ fig : peqcd ] , the pressure in dependence on the energy density . as discussed in ref . @xcite , the ultimate goal of the high - energy experiments is first - principle determination of the underlying dynamics in the strongly interacting matter . determining the thermodynamical quantities , which are well - known tools to describe nature , degrees of freedom , decomposition , size and even the overall dynamics controlling evolution of the medium from which they are originating , is therefore very essential . for the thermodynamical pressure , an approximate attempt utilizing the higher - order moments of the particle multiplicity seems to be promising @xcite . on the other hand , determining the thermodynamical energy density and even relating the bjorken energy density to the lattice energy density depends on lattice qcd at finite baryon chemical potential and first - principle estimation of the formation time of the quark - gluon plasma ( qgp ) . the energy density can be deduced from the derivative of the free energy with respect to inverse temperature . this would explain the need to implement another variable different than the one responsible for the derivatives of the high - order moments . e. thomas and a. r. zhitnitsky , _ topological susceptibility and contact term in qcd . a toy model _ , phys . d * 85 * , 044039 ( 2012 ) . t. zhu , f .- w . shu , q. wu , and a. wang , _ general covariant horava - lifshitz gravity without projectability condition and its applications to cosmology _ , phys . d * 85 * , 044053 ( 2012 ) . d. vernieri and t p sotiriou , _ horava - lifshitz gravity : detailed balance revisited _ , phys . d * 85 * , 064003 ( 2012 ) . m. saravani , n. afshordi , and r. b. mann , _ dynamical emergence of universal horizons during the formation of black holes _ , phys . d * 89 * , 084029 ( 2014 ) . f. kheyri , m. khodadi , and h. r. sepangi , _ horava - lifshitz cosmology , entropic interpretation and quark - hadron phase transition _ , annals phys . * 332 * , 75 ( 2013 ) . r. loll and l. pires , _ role of the extra coupling in the kinetic term in hoava - lifshitz gravity _ , phys . rev . d * 90 * , 124050 ( 2014 ) . a. de felice and s. tsujikawa , _ inflationary gravitational waves in the effective field theory of modified gravity d * 91 * , 103506 ( 2015 ) . m. cruz , m. gonzalez - espinoza , j. saavedra , and d. vargas - arancibia , _ scalar perturbations of two - dimensional horava - lifshitz black holes _ , eur . j. c * 76 * , 75 ( 2016 ) . a. maciel , _ quasilocal approach to general universal horizons _ , d * 93 * , 104013 ( 2016 ) . m. chaichian , a. ghalee , and j. kluson , _ restricted f(r ) gravity and its cosmological implications _ , phys . d * 93 * , 104020 ( 2016 ) . j. ambjorn , j. jurkiewicz , and r. loll , _ quantum gravity as sum over spacetimes _ , _ lect . notes phys . _ * 807 * , 59 - 124 ( 2010 ) . d. f. litim , _ fixed points of quantum gravity _ , _ phys . rev . lett . _ * 92 * , 201301 ( 2004 ) . m. reuter and f. saueressig , _ renormalization group flow of quantum gravity in the einstein - hilbert truncation _ , _ phys . d _ * 65 * , 065016 ( 2002 ) . s. weinfurtner , t. p. sotiriou , and m. visser , _ projectable horava - lifshitz gravity in a nutshell _ , _ j. phys . ser . _ * 222 * , 012054 ( 2010 ) . t. p. sotiriou , _ horava - lifshitz gravity : a status report _ , _ j. phys . ser . _ * 283 * , 012034 ( 2011 ) . m. visser , _ status of horava gravity : a personal perspective _ , _ j. phys . ser . _ * 314 * , 012002 ( 2011 ) . luis piresa , master thesis on _ horava - lifshitz gravity : hamiltonian formulation and connections with cdt _ , institute of theoretical physics at utrecht university , ( 2012 ) . abdel nasser tawfik and abdel magied diab , _ `` generalized uncertainty principle and recent cosmic inflation observations '' _ , _ electron . j. theor . phys . _ * 12 * , 9 - 30 ( 2015 ) . f. kheyri , m. khodadi , and h. r. sepangi , _ horava - lifshitz early universe phase transition beyond detailed balance _ , _ eur . j. c _ * 73 * , 2286 ( 2013 ) . s. nojiri and s. d. odintsov , _ `` introduction to modified gravity and gravitational alternative for dark energy '' _ , _ int . phys . _ * 4 * , 115 ( 2007 ) . m. chaichian , sh . nojiri , s. d. odintsov , m. oksanen , and a. tureanu , _ class . _ * 27 * , 185021 ( 2010 ) ; erratum - ibid . * 29 * , 159501 ( 2012 ) . a. tawfik and a .diab , _ generalized uncertainty principle : approaches and applications * 23 * , 1430025 ( 2014 ) . a. kamenshchik , u. moschella , and v. pasquier , _ an alternative to quintessence _ , _ phys . b _ * 511 * , 265 ( 2001 ) . t. p. sotiriou , m. visser , and s. weinfurtner , _ phenomenologically viable lorentz - violating quantum gravity _ , _ phys . _ * 102 * , 251601 ( 2009 ) . t. p. sotiriou , m. visser , and s. weinfurtner , _ quantum gravity without lorentz invariance _ , _ jhep _ * 0910 * , 033 ( 2009 ) . p. horava . _ `` quantum gravity at a lifshitz point '' _ , _ phys . d _ * 79 * , 084008 ( 2008 ) . d. giulini and c. kiefer , _ wheeler - dewitt metric and the attractivity of gravity _ , _ phys . a _ * 193 * , 21 ( 1994 ) . shinji mukohyama , _ scale - invariant cosmological perturbations from horava - lifshitz gravity without inflation _ , _ jcap _ * 0906 * , 001 ( 2009 ) . t. zhu , y. huang , and a. wang , _ inflation in general covariant hoava - lifshitz gravity without projectability _ , _ jhep _ * 1301 * , 138 ( 2013 ) . y. huang , a. wang , and q. wu , _ inflation in general covariant theory of gravity _ , _ jcap _ * 1210 * , 010 ( 2012 ) . e. elizalde , s. nojiri , s. d. odintsov , and d. saez - gomez , _ unifying inflation with dark energy in modified f(r ) horava - lifshitz gravity _ , _ eur . j. c _ * 70 * , 351 ( 2010 ) . r. kase and s. tsujikawa , _ effective field theory approach to modified gravity including horndeski theory and hoava - lifshitz gravity _ , _ int . * 23 * , 1443008 ( 2015 ) . a. d. polyanin and v. f. zaitsev , _ exact solutions for ordinary differential equations _ , ( chapman & hall / crc , london , 2003 ) . s. borsanyi , z. fodor , ch . hoelbling , s. d. katz , s. krieg , and k. k. szabo , _ full result for the qcd equation of state with 2 + 1 flavors _ , _ phys . b _ * 730 * , 99 - 104 ( 2014 ) . abdel nasser tawfik , _ an estimate of the thermodynamic pressure in high - energy collisions _ , _ int * 30 * , 1550027 ( 2015 ) . e. p. verlinde , _ on the origin of gravity and the laws of newton _ , _ jhep _ * 1104 * , 029 ( 2011 ) . chien - i chiang , je - an gu , and pisin chen , _ constraining the detailed balance condition in horava gravity with cosmic accelerating expansion _ , _ jcap _ * 1010 * , 015 ( 2010 ) . h. lu , j. mei , and c. n. pope , _ solutions to horava gravity _ , _ phys . * 103 * , 091301 ( 2009 ) . a. tawfik and a. diab , _ `` emergence of cosmic space and minimal length in quantum gravity : a large class of spacetimes , equations of state , and minimal length approaches '' _ , _ indian j. phys . _ , * 90 * , 1095 ( 2016 ) . y .- w . kim , s. k. kim , and y .- j . park , _ thermodynamic stability of modified schwarzschild - ads black hole in rainbow gravity _ , eur . j. c * 76 * , 557 ( 2016 ) . s. gangopadhyay and a. dutta , _ constraints on rainbow gravity functions from black hole thermodynamics _ , europhys . 115 * , 50005 ( 2016 ) . r. garattini and e. n. saridakis , _ gravity s rainbow : a bridge towards hoava - lifshitz gravity j. c * 75 * , 343 ( 2015 ) . mir faizal and barun majumder , _ incorporation of generalized uncertainty principle into lifshitz field theories _ , annals phys . * 357 * , 49 ( 2015 ) . abdel nasser tawfik and eiman abou el dahab . _ corrections to entropy and thermodynamics of charged black hole using generalized uncertainty principle _ , int . a * 30 * , 1550030 ( 2015 ) . abdel nasser tawfik and abdel magied diab , _ black hole corrections due to minimal length and modified dispersion relation a * 30 * , 1550059 ( 2015 ) . s. robles - perez , y. hassouni , and p. f. gonzalez - diaz , _ coherent states in the quantum multiverse b * 683 * , 1 ( 2010 ) . s. gielen , d. oriti and l. sindoni , _ homogeneous cosmologies as group field theory condensates _ , jhep * 1406 * , 013 ( 2014 ) . mir faizal , _ multiverse in the third quantized horava - lifshits theory of gravity _ , mod . phys . lett . a * 27 * , 1250007 ( 2012 ) .

quantum information protocols have been demonstrated experimentally in both the discrete - variable ( dv ) and so - called continuous - variable ( cv ) settings . dv quantum information protocols employ qubits @xcite and qudits @xcite , and cv quantum information protocols regard continuously parameterized canonical position states as the logical elements analogous to qubits for the dv case @xcite . cv quantum information is experimentally appealing because sophisticated squeezed light experiments have led to claims of successful quantum information protocols such as teleportation @xcite , key distribution @xcite , and memory @xcite , but the theoretical status of cv quantum information is challenged by unresolved issues concerning quantum error correction @xcite , non - distillability @xcite , no - go theorems for quantum computation @xcite , and the absence of full security proofs for key distribution . cv information processing has also been studied for classical models , including the now named blum - shub - smale machine @xcite and continuous turing machines @xcite . these models are of background relevance to the research into cv quantum information and are referenced here for contextual purposes . in this paper , we establish a sound theoretical framework for studying quantum algorithms and apply this framework to study the cv analogue of the early dv quantum algorithm , known as the deutsch - jozsa ( dj ) algorithm @xcite . the problem solved by dj algorithm is the following . [ problem : balancedorconstant ] given a function @xmath0 that is promised to be either _ constant _ ( @xmath1 takes the same value everywhere ) or _ balanced _ ( @xmath1 takes the value 0 on exactly half the inputs ) , determine whether @xmath1 is constant or balanced . the best classical algorithm requires @xmath2 evaluations in the worst case . if error is tolerated , for any integer @xmath3 , to achieve an error of at most @xmath4 , any probabilistic algorithm requires a number of evaluations that is at least of order @xmath5 @xcite . if the function is accessible on a quantum computer as a quantum oracle , then the dj algorithm is exact and requires just one evaluation to solve the problem . our focus here is on the cv analogue of the dj algorithm , and we are inspired by the braunstein and pati formulation @xcite of the cv dj algorithm ; however , our work differs from theirs in that ours relies only on logical states that are elements in the hilbert space and thus provides a strict cv version of the dj problem . we introduce a particular model for the computation of the dj problem in a cv setting . within the constraints of this model , our analysis shows that the cv dj algorithm is necessarily probabilistic and its performance must therefore be compared to the classical case where bounded error is tolerated and not to the classical deterministic case . we choose the dj algorithm for the following reasons . two types of quantum algorithms dominate the field , those that implement a version of the hidden subgroup problem and those that use a version of grover s search algorithm @xcite . an early example of the former is the deutsch - jozsa ( dj ) algorithm @xcite , which is amongst the oracle class of problems @xcite that have been important in demonstrations of quantum speed - ups . finally the cv dj algorithm has a head start in the work of braunstein and pati so our analysis can build on their concepts @xcite . our paper is presented as follows . in sec . [ sec : background ] , we review the dj algorithm . although this algorithm is well known , our review serves as a foundation for careful construction of the cv version of this algorithm . furthermore we compare the dj algorithm s performance against both deterministic and probabilistic strategies , especially because the cv case can only be properly compared against probabilistic strategies because the cv dj algorithm can never be deterministic . our description of the dv dj algorithm comprises three steps so that these steps can be discussed separately during the construction of the cv analogue . our approach emphasizes a recasting of the dv dj algorithm in that we do not need the target qubit . this approach leads to an easier adaptation to the cv case . in sec . [ sec : background ] , we review the formalism of rigged hilbert spaces ( rhs ) @xcite since our cv algorithm , as well as any other cv quantum algorithms , must work in a rhs . this will have implications when we discuss the limitations of error inherent in our cv dj algorithm in sec . [ sec : cvrep ] and in sec . [ sec : bounding ] . in sec . [ sec : cvrep ] , we adapt the dj problem to the cv case and develop the cv dj algorithm through the same three fundamental steps of the algorithm . we pay particular attention to the challenge of encoding a finite @xmath6-bit string into functions over the real numbers . overcoming this challenge enables us to recognize that perfect encoding results in the inability to determine if the encoding is of a constant string or balanced in a single execution of the algorithm . we show that this probabilistic nature of the algorithm is the result of an uncertainty relation between the continuous representation and its fourier - transform dual representation . in sec . [ sec : bounding ] , we determine an upper bound on the query complexity of the cv dj algorithm . we note that because the cv dj algorithm is shown to be probabilistic , its performance can only logically be compared to the classical probabilistic algorithm and not to the classical deterministic algorithm . we conclude that the formalism presented herein is applicable to a wide range of oracle identification problems in a cv setting . we cast the dj problem into the class of ` oracle identification problems ' in subsec . [ subsec : oracle ] . we then review deterministic algorithms in subsec . [ subsec : deterministic ] and probabilistic algorithms in subsec . [ subsec : randomized ] . in subsec . [ subsec : quantumdj ] , we analyze an alternative representation of the quantum dj algorithm that uses @xmath7 qubits instead of the traditional @xmath8 qubits . in subsec . [ subsec : cv ] , we present a primer on the rigged hilbert space and close with a discussion of the concepts required to transition from discrete variables to continuous variables . the dj problem is an identification problem in which we are given a function from some candidate set @xmath9 of functions . the candidate set @xmath10 is the disjoint union of two collections of functions , and our task is to determine which of the two collections the function @xmath1 is drawn from . [ problem : oip ] consider the set of all functions from @xmath7 bits to one bit , @xmath11 . let @xmath12 and @xmath13 be disjoint subsets of @xmath14 . given some oracle @xmath15 with the promise that either @xmath16 or @xmath17 , determine the index @xmath18 such that @xmath19 . for @xmath20 , we impose lexicographic order on the @xmath6-bit strings of @xmath21 . we can then specify any function @xmath22 by writing all its @xmath6 function values in a list @xmath23 of length @xmath6 . the @xmath24 bit @xmath25 in the list is 1 if @xmath1 takes the value 1 on the @xmath24 bit - string of @xmath21 . there are @xmath26 functions from @xmath7 bits to one bit , and thus our candidate set has cardinality upper bounded by @xmath27 . in the following , we often write @xmath22 to denote the function that corresponds to the @xmath6-bit string @xmath28 . we are interested in finding an efficient strategy to identify the property of whether @xmath1 belongs to set @xmath12 or to set @xmath13 without necessarily determining @xmath1 itself . in the dj case , the property we are interested in is whether @xmath1 is balanced or constant @xcite . the cost of the algorithm is the number of queries made to the oracle . with the promise of balanced or constant functions , there are far fewer than @xmath26 functions . the number of balanced and the number of constant functions is readily ascertained from the binomial theorem applied to power sets . the strings @xmath28 of length @xmath6 that correspond to the constant functions are the string consisting only of 1s and the string consisting only of 0s . there are thus just two constant functions . the strings @xmath28 of length @xmath6 that correspond to balanced functions are the strings in which exactly half of the bits are 0 and half are 1 . there are thus precisely @xmath29 balanced functions . on a classical turing machine , problem [ problem : balancedorconstant ] can be solved deterministically . a deterministic algorithm corresponds to submitting queries in the form of @xmath7-bit inputs and obtain the one - bit output for each query . there are @xmath6 unique input strings , but the promise of balanced versus constant functions implies that only @xmath30 are required to determine whether the given function is balanced or constant , with certainty . the reason that fewer than @xmath30 queries is insufficient is that only @xmath31 queries may reveal all output bits being the same , suggesting a constant function , whereas the remaining @xmath31 outputs could all be the opposite of the first @xmath31 queries . in subsec . [ subsec : deterministic ] , we saw that fewer than @xmath30 queries is insufficient for a deterministic algorithm , but that case seems highly unlikely . more formally , fewer than @xmath30 queries will identify most of the balanced functions as non - constant in much fewer than @xmath30 queries . here we ask the question about how many queries are required if we are prepared to tolerate a small number of errors . in fact a probabilistic algorithm achieves an exponentially small error of @xmath4 with a number of queries that is only linear in @xmath5 @xcite . to understand how a probabilistic algorithm can help , consider that , although a single query with a random input provides no information , two queries with two random inputs can be highly informative . if the output from the second query differs from the first output , then the function is proved not to be constant and therefore must be balanced . if , on the other hand , the second output is the same as the first , then the outcome is not certain , but the more times the outputs are the same , the more confident one can be about the function being constant . -qubit `` control '' state , and the lower line represents the 1-qubit `` target '' state.,width=321 ] we calculate the probability of successfully determining whether the given function @xmath22 is balanced or constant . a lower bound on the success probability @xmath32 for @xmath5 queries can be achieved by examining a sampling - without - replacement strategy , which is expressed as @xmath33 here the equality is calculated assuming sampling without replacement and shows dependency on @xmath6 , whereas the inequality in eq . ( [ eq : classdjran ] ) is based on sampling with replacement and is independent of @xmath6 . the failure probability @xmath34 declines exponentially in @xmath5 , the number of queries . in subsec . [ subsec : quantumdj ] , we study the quantum dj algorithm next where we show that the problem can be solved with a single query independent of @xmath6 . although this exponential speed - up is impressive when compared to the classical deterministic approach , it is much less so when compared to the classical probabilistic approach . the quantum dj algorithm has been shown to solve problem [ problem : balancedorconstant ] in a single query @xcite . the quantum dj algorithm is usually studied via its corresponding quantum circuit . we present a standard circuit version @xcite in fig . [ fig : tradqdjcircuit ] . the state represented by the lower line in fig . [ fig : tradqdjcircuit ] is referred to as the _ target _ qubit . in order for easier adaption of this circuit to the cv setting , we choose an alternative , and equivalent , circuit formulation one without the target state . we take this approach to avoid some of the difficulties the target state introduces in @xcite . the unitary operator associated with the oracle function changes slightly in this alternative circuit . we discuss these differences before proceeding with analysis of the circuit . this simpler algorithm without the target qubit is given in fig . [ fig : targetlessqdjcircuit ] . defined in eq.[eq : qdjtargetlessuf],width=8 ] . [ fig : targetlessqdjcircuit ] oracle application is the critical part of the algorithm . the oracle construct originally proposed by dj is expressed , for @xmath35 and @xmath36 , as @xmath37 this construction yields a matrix representation for the @xmath38 as a permutation matrix , hence always unitary @xcite . with respect to the ordered basis @xmath39 the unitary matrix @xmath38 can be expressed in the following insightful form @xmath40 with @xmath41 the @xmath42 @xmath43 operator in this case . here @xmath38 is a @xmath44 matrix , which results from there being @xmath45 strings ( the arguments of @xmath1 ) and an additional target qubit . the operator @xmath38 can also be expressed in the alternative ordered basis @xmath46 as @xmath47 with @xmath48 the @xmath49 identity operator . furthermore the operator @xmath50 is expressed as the @xmath51 matrix @xmath52 and thus provides a reduced representation for @xmath38 . it is apparent that the operator @xmath50 acts on a @xmath49 subspace of @xmath38 since @xmath53 we make the assumption that if we have the oracle @xmath38 , we we also have the oracle @xmath50 . we thus conclude that the construction employing both control and target qubits is not strictly necessary . that is , one could construct this algorithm employing the @xmath7-qubit control state only . apparently the choice of representation simply depends on the nature of the actual physical implementation . we now present a step - by - step analysis of the alternative circuit presented in fig . [ fig : targetlessqdjcircuit ] . we shall analyze the cv circuit in the same steps for cross reference and comparison . we use the hat notation @xmath54 in order to emphasize that this analysis is of the algorithm presented in fig . [ fig : targetlessqdjcircuit ] , which employs @xmath7-qubit states and not of that presented in fig . [ fig : tradqdjcircuit ] , which employs @xmath55-qubit states . the @xmath7-qubit input state of the circuit in fig . [ fig : targetlessqdjcircuit ] is a string of qubits prepared in @xmath56 . the next step in state preparation is to place the state @xmath57 into an equal superposition of all computational basis states @xmath58 for @xmath59 the single qubit hadamard operator . given the definition of the reduced operator @xmath50 defined in eq . [ eq : qdjtargetlessuf ] , its effect on the equal superposition of basis states expressed in the state @xmath60 is to effectively encode the @xmath6-bit string @xmath28 unitarily into the state @xmath61 . we express this as @xmath62 which is a convenient representation . we shall show that this representation naturally extends to the cv setting . measurement proceeds by first undoing the superposition created during the state preparation step . this is achieved through the application of the operator @xmath63 , which modifies the state after oracle application @xmath64 the resultant state is @xmath65 we rewrite eq . ( [ eq : qdjstate3 ] ) with the operator @xmath66 expressed in terms of a recursive definition as follows @xmath67 given @xmath68 the combination of eq . ( [ eq : qdjtargetlessstate3recursive ] ) and eq . ( [ eq : hadamardsinglequbit ] ) allows us to see that all of the rows ( and columns ) of the operator @xmath66 have an equal number of positive and negative ones except for the first row , which consists entirely of plus ones . it is this feature that permits the constant and balanced functions to be distinguished in a single measurement . for the two constant cases , eq . ( [ eq : qdjtargetlessstate3recursive ] ) may be expressed as @xmath69 as only the first row does not result in amplitude cancellation of the @xmath45 constant amplitude components of the state @xmath61 . each of the balanced functions result in the amplitudes of the state @xmath61 having an equal number of positive and negative ones . this feature coupled with action of the operator @xmath66 results in the first component of the state @xmath70 having zero amplitude for all the balanced functions . we express this result as @xmath71 where we use the symbol @xmath72 to represent that the non - zero value(s ) will land on the other @xmath73 components depending on which of the @xmath29 balanced functions the oracle is set to . it is interesting to note that the number of rows in the state @xmath74 potentially having a non - zero value is @xmath73 whereas the number of balanced functions is exponential in @xmath6 . this means that many of the balanced states can be expressed as real - valued mixtures of the computational basis states with the condition that the amplitude of the first component is always zero . for the final measurement step , we employ the projection operator @xcite defined for @xmath75 as follows @xmath76 we are only concerned with the first component as discussed above , so for the constant cases we have @xmath77=\left\langle\hat{\psi}_{3\text{c}}\right|m_{(0\cdots 0 ) } \left|\hat{\psi}_{3\text{c}}\right\rangle=1,\ ] ] and for all balanced cases we have @xmath78=\left\langle\hat{\psi}_{3\text{b}}\right| m_{(0\cdots 0)}\left|\hat{\psi}_{3\text{b}}\right\rangle=0\ ] ] as required . we have completed the study of the quantum dj algorithm in a form that allows us to adapt readily to the cv setting . our strategy will be to construct a cv algorithm analogous to that shown in fig . [ fig : targetlessqdjcircuit ] and whose operator representation is given by @xmath79 this approach is simpler , and we can worry about whether or not an implementation will require target states when a particular implementation is considered . before delving into the cv algorithm , we present some background cv information . the transition from dv to cv quantum information requires an extension of hilbert spaces to rigged hilbert spaces @xcite , which allows the use of position states @xmath80 with @xmath81 but restricts dual states to so - called ` test functions ' . an inner product between position states and test functions is meaningful , but the inner product between two position states leads to the dirac relation @xmath82 , which must be treated carefully . for @xmath7 the size of problem [ problem : balancedorconstant ] , the target - less quantum dj algorithm requires @xmath7 qubits , which requires a hilbert space of size @xmath20 @xcite . the hilbert space for cv problems seems quite generous in this respect as it is infinite - dimensional . in fact the cv hilbert space is congruent to the space of square - integrable complex functions over the real field @xmath83 @xcite . a function @xmath84\rightarrow \mathbb{c}$ ] , for @xmath85\subset\mathbb{r}$ ] , is in @xmath83 if @xmath86 the inner product of two functions @xmath87 is @xmath88 with positive definite norm and distance metric defined by @xmath89 respectively . typically in cv quantum information discourse , the position states @xmath80 are introduced as a basis set of the hilbert space with each @xmath80 an eigenstate of a position operator @xmath90 , with @xmath81 . unfortunately the state @xmath80 does not exist in the hilbert space ; this problem is evident in the standard inner product @xmath91 as @xmath92 is not a proper function , position states are not proper states . fortunately the position states are correct as a representation ; for example @xmath93 is the position representation of test function @xmath1 within the context of the rigged hilbert space . also eq . ( [ eq : cv_deltafunction ] ) is meaningful in the context of distribution theory . a rigged hilbert space is a pair @xmath94 such that @xmath95 is a hilbert space and @xmath96 is a vector space that is included by a continuous mapping into @xmath95 : @xmath97 . elements of @xmath96 are referred to as ` test functions ' , and the dual to @xmath96 is @xmath98 , for @xmath99 dual to @xmath100 and @xmath101 comprising generalized functions , or ` distributions ' . the inner product @xmath102 is in @xmath103 $ ] for any @xmath104 and for any @xmath105 @xcite . note that the adaptation of the dv dj algorithm to the cv regime needs to be done in the context of a computational problem . here the relevant problem is still problem [ problem : balancedorconstant ] , and the notion of the oracle remains unchanged . thus , in the cv case , our task is still to determine whether the function @xmath22 belongs to the set of constant functions or to the set of balanced functions . we begin by giving a strategic overview in order to convey the key concepts of our approach to developing a cv computation model . we follow this with a subsection giving some preliminary definitions allowing us to set the stage for detailed analysis . we then proceed with a step - by - step analysis of our cv dj algorithm . although we are now working with cv , instead of dv , quantum information , the computational problem to be solved remains problem [ problem : balancedorconstant ] . in other words , we want to learn whether the function @xmath22 is constant or balanced with as few oracle queries as possible . another way to think of this is that we wish to determine the index @xmath106 such that @xmath107 . we now give a conceptual overview of our model for cv quantum computation of the dj problem , which we follow later with a rigourous treatment . in our model of cv quantum computation , we will use the continuous position and momentum variables of a particle . for @xmath108 , we use the particle s position wave function , @xmath109 , to describe where the particle is concentrated and the particle s momentum wave function , @xmath110 , to describe its momentum distribution . the position and momentum wave functions are fourier transform pairs , and the relationship between the particle s position and its momentum is governed by heisenberg s uncertainty principle . -bit string in a region of momentum extending from @xmath111 to @xmath112 using the @xmath113 , @xmath114 example . note that each of the bits @xmath115 are uniquely represented.,width=8 ] there are many position and momentum wave function pairs on which we could base our computational model . we select our particular pair as follows . first , we wish to encode the unknown @xmath6-bit string , @xmath28 , in the momentum domain . we do so because encoding in the momentum domain is the continuous analogue of the discrete case , where encoding is performed on an equal superposition of computational basis states . second in order to fix one of the degrees of freedom of the problem , we want each of the bits comprising the string @xmath28 to be unambiguously represented in the momentum space . by unambiguous we mean that each of the bits are represented by equal - sized , non - overlapping , contiguous regions in the momentum space . since we want each of the @xmath6-bits comprising the string to be represented unambiguously , we naturally think of each bit as being manifested by a finite - width square pulse whose position in momentum space represents the bit position in the string @xmath28 and whose magnitude represents the bit value . continuing along this line of reasoning to the representation of the entire string @xmath28 , we can imagine we have a region of momentum extending from @xmath111 to @xmath112 . all the contiguous momentum pulses within this region thus have width " @xmath116 , and for @xmath117 , the @xmath118 momentum pulse is centred at position @xmath119 and takes on value @xmath120 . we illustrate this concept in fig . [ fig : cvdjalgencodeconcept ] for a particular @xmath113 case . is that of a sinc function . ( b ) the fourier transform of the position state wave function is a pulse " function in the momentum domain , which acts as the encoding substrate " . ( c ) the @xmath6-bit string @xmath121 modulates this momentum substrate " . ( d ) the inverse fourier transform of the encoded square wave " produces a generalized " sinc function whose infinite position extent necessitates an optimal measurement window " parameterized by @xmath122.,width=8 ] the picture that thus emerges is that each of the @xmath26 possible strings may be represented by a uniquely shaped square wave " having extent @xmath123 comprising @xmath6 pulses each of width @xmath124 and having magnitude @xmath125 . with the encoding concept clear , we conceptually illustrate the four key stages of the algorithm in fig . [ fig : cvdjalgoverview ] . we begin with a position wave function centered at @xmath126 and illustrated in fig . [ fig : cvdjalgoverview ] ( a ) . note that this position wave function is a sinc function since sinc / pulse functions are fourier transform pairs . in fig . [ fig : cvdjalgoverview ] ( b ) , we present the momentum wave function , a pulse function , which acts as the substrate " into which the @xmath6-bit strings are encoded . in fig . [ fig : cvdjalgoverview ] ( c ) , the pulse function is encoded with the particular @xmath6-bit string @xmath121 . finally , the inverse fourier transform of this square wave " is presented in fig . [ fig : cvdjalgoverview ] ( d ) . since the inverse fourier transforms of finite pulses in the momentum domain have infinite extent in the position domain , we need to limit the extent of our measurement to @xmath122 . in summary , we see that our algorithm will need the parameters @xmath6 , @xmath127 and @xmath92 . we note that as @xmath6 gets large , the individual pulse width associated with a single bit becomes small appearing to pose a limit on the maximum value of @xmath6 . we will return to this issue once we have determined the relationship between @xmath127 and @xmath92 . there are many potential models for quantum computation in a cv setting . we have chosen to study one where we unambiguously encode an n - bit string into the continuous momentum variable of a particle . within the constraints of this model , we will show that the cv dj is necessarily probabilistic and prove an upper bound on the query complexity of the cv dj problem . we speculate that we ca nt do better than this . for example if the momentum / position pair are described by gaussian / gaussian functions , as would be the case for the physically meaningful states of quantum optics , imperfect encoding of the @xmath6-bit string in the momentum domain will result in increased position error . whether or not this will in turn impact the big oh " representation of the query complexity requires further research as does a general proof of a lower bound . the challenge will be to show that another strategy can do better than the model described herein . we now proceed to formalize some of the concepts presented in the previous subsection . here we describe a ` natural ' way of encoding a finite - dimension , @xmath6-bit string in a continuous domain . we define the following function , along with its fourier dual , to help us achieve this end . for @xmath128 , the ` top hat ' function @xmath129\\0,&\text{if}\quad p\notin[p_0-p , p_0+p]\end{array } \right.\end{aligned}\ ] ] will be especially useful in bridging the gap between dv and cv quantum information because @xmath130 so the state @xmath131 is , in some sense , a momentum eigenstate @xmath132 in the limit @xmath133 . the inverse fourier transform of the function @xmath134 is @xmath135 where @xmath136 defines the position of the sinc function . the limit of @xmath109 as @xmath127 goes to @xmath137 yields @xmath138 . the position eigenstate @xmath139 is likewise formed in the limit @xmath140 . now imagine we want to sum a contiguous string of pulses " described by the top hat function ( [ eq : tophat ] ) with all pulses having width @xmath141 and the @xmath142 pulse having complex amplitude @xmath143 . this results in the composite function @xmath144 this function can also be used as a basis @xcite of cv kets in dirac notation as @xmath145 thus allowing us to encode quantum information in the cv domain . note that @xmath146 is the complex amplitude for real - valued @xmath147 . this affords a consistent way of encoding a discrete wave function over a continuous domain . before proceeding with a formal analysis of the algorithm , we give an overview of our proof strategy . the oracle is either set to one of two constant strings or to one of @xmath148 balanced strings . a string and its complement have indistinguishable probability distributions , so there are a total of one constant probability distribution plus @xmath149 balanced probability distributions representing the possible oracle settings . in order to simplify the analysis , we wish to replace this exponential number of balanced probability distributions with a single worst - case " balanced probability distribution . thus we seek a particular balanced string ( and its complement ) whose probability distribution is most likely to fool " us into concluding it is a constant string . intuitively , the balanced strings that have the fewest number of changes between adjacent bits in the interval @xmath150 $ ] will be the most constant like " of the balanced strings . there are no balanced strings with zero changes - this is the key feature that separates the constant strings from the balanced strings . there is however , a single pair of balanced strings having only one change . these strings exhibit the feature that the first @xmath31 bits are constant and the second @xmath31 bits are the complement of the first . we call these strings the anti - symmetric balanced ( asb ) strings . one of these two strings is illustrated in fig . [ fig : cvdjalgoverview ] ( c ) . note that all other balanced strings have more than one change . our proof strategy begins by making the assumption that the asb case is the worst case " of all balanced cases . we use this assumption to determine the optimum value of @xmath92 , which is the extent of our measurement in the position domain and is illustrated conceptually in fig . [ fig : cvdjalgoverview ] ( d ) . given this optimum value of @xmath92 , we then prove by induction that the worse balanced case is indeed the asb case . our strategy is to create a cv analogue of the alternative formulation of the discrete dj algorithm presented in fig . [ fig : targetlessqdjcircuit ] . the cv extension of this is presented in fig . [ fig : targetlesscvdjcircuit ] . our construction of a cv dj algorithm employs some of braunstein and pati s techniques @xcite and avoids the pitfalls . in particular , we employ position states as a logical representation ( states in @xmath101 ) analogous to the discrete computational basis states . encoding is not , however , into the position states but rather into test functions @xmath151 with @xmath152 . furthermore we employ a fourier transform to operate as a cv version of the dv hadamard transform ( extending the hadamard transformation to the cv case is not unique @xcite ) . for @xmath72 the canonical position and @xmath147 the canonical momentum , the fourier transform maps a function @xmath109 to its dual @xmath110 according to @xcite @xmath153 such that @xmath154 and @xmath155 note that we make use of the momentum variable @xmath147 as the fourier dual of the position variable @xmath72 . the function @xmath156 can be a test function in @xmath96 and @xmath109 is the inner product of @xmath156 with the position in @xmath101 : @xmath157 . the momentum state @xmath158 is the fourier transform of @xmath80 and @xmath159 . with these concepts in order , we now proceed through the cv dj algorithm analogous to the three steps in the dv dj algorithm . the function notation @xmath109 and @xmath110 is more convenient here rather than the dirac notation in the previous section . we have argued previously that we need the fourier transform of the input state to be the top hat function defined in eq . ( [ eq : tophat ] ) . we add several conditions that do not take away from the generality of the solution . first , we want the top hat to have zero phase , which gives @xmath160 and to be centred at @xmath161 . second , we want the pulse to have extent @xmath123 . this gives us the simplest form of the sinc function for the initial state @xmath162 we note that the limit of @xmath163 as @xmath164 gives a @xmath165 . thus we can think of the quantity @xmath127 as playing the role of the standard deviation in a gaussian distribution . the final step in state preparation is to perform the fourier transform , which yields the top hat function with extent @xmath123 @xmath166\\0,&\text{if}\,\,p\notin[-p , p].\end{array } \right.\ ] ] this function forms the raw substrate , which will be ` modulated ' by the individual @xmath6-bit strings @xmath28 . we perform encoding by partitioning the real numbers representing momentum into non - overlapping , contiguous and equal - sized bins . in this digital - to - analogue strategy , the width of each @xmath147-bin is @xmath124 , and @xmath167\\0,&\text{otherwise . } \end{array } \right.\ ] ] the oracle encodes the index @xmath28 into the function @xmath22 as follows : @xmath168 where the factor @xmath169 serves to modulate the phase of the top hat function according to the bit value . consider the case @xmath170 ; hence @xmath171 . as one case , the function corresponding to the four - bit string @xmath172 is @xmath173 the only two four - bit strings yielding constant functions would be @xmath174 , for which the function is identically unity over the whole domain @xmath175 $ ] , and @xmath176 , for which the function is identically @xmath177 over @xmath175 $ ] . four cases are presented in fig . [ fig : cvdjsqwaveneq2 ] . we refer to the function @xmath178 as the lowest - order " antisymmetric balanced wave as it has just one zero crossing in @xmath175 $ ] . ( a ) @xmath179 , ( b ) @xmath180 , ( c ) @xmath114 , and ( d ) @xmath181.,width=8 ] in the limit that @xmath182 with @xmath127 fixed , @xmath183 for @xmath184 the midpoint of the @xmath185 bin . the limit @xmath182 thus gives a prescription for approaching a continuous variable representation where the @xmath28 index seems to approach a continuum ; however this limit yields a countable , rather than uncountable , set @xmath186 , and the finite domain @xmath175 $ ] has important ramifications on the nature of the functions corresponding to fourier transforms of @xmath187 . we express the state after encoding as @xmath188 where we observe the `` modulating '' effect of the encoded string @xmath189 on the momentum `` substrate '' @xmath110 . in the context of the digital - to - analogue strategy , the constant functions are analogous to direct current ( dc ) signals and the balanced functions to alternating current ( ac ) signals . the number of zero - crossings corresponds to frequency information , and the question of whether the output is balanced or constant is essentially a problem of querying whether there is a non - zero frequency component of the output signal . as noted previously , the asb function has the lowest frequency component . we now proceed to analyze the measurement stage . we have the strings @xmath152 encoded into the momentum state ( [ eq : encodedmomentum ] ) . the next step prior to the final measurement is to take the inverse fourier transform of this pulse train . for @xmath115 the @xmath142 bit of @xmath28 , this is expressed as @xmath190 the expression given in eq . ( [ eq : cvdjphaser1 ] ) can be simplified to yield @xmath191where we have defined @xmath192 we see that the magnitude of an individual generalized sinc function , @xmath193 , is determined by a vector sum of @xmath6 phasors , which is modulated by a particular @xmath6-bit string @xmath28 . note that the phasors , @xmath194 , are equiangular divisions of the angular interval @xmath195\nonumber,\ ] ] and they exhibit the pairwise complex conjugate property @xmath196 . defined in eq . ( [ eq : cvdjphaser2 ] ) for @xmath197 ( a ) @xmath198 phasors range between @xmath199 in steps of @xmath200 , and(b ) @xmath201 phasors range between @xmath202 in steps of @xmath203.,width=7 ] in fig . [ fig : cvdjsincphasers ] , we present the phasors for @xmath197 with @xmath198 and @xmath201 to illustrate these features . note that the phasors are added constructively or de - constructively depending on the phase of the angles , which results from the term @xmath120 . this effect defines the magnitude of the resulting sinc function . we note that the only functions with @xmath204 are the two constant sinc functions . this is clear given that @xmath205 for the two constant cases , and @xmath206 since for all balanced cases , the latter sum always resolves to @xmath207 . this feature of the set of @xmath208 sinc functions represented by @xmath209 defined by eq . ( [ eq : cvdjphaser2 ] ) implies the strategy for measurement that will distinguish between the constant and balanced cases . in order to refine this strategy , we focus on two cases . the first of these cases is for the two constant functions for which eq . ( [ eq : cvdjphaser2 ] ) gives the probability distribution @xmath210 where we have @xmath211 the second case deals with the two balanced functions having the lowest ` frequency ' content , which occurs when the first @xmath31 bits and the last @xmath31 bits have opposite values . we think of these two balanced functions as having the lowest frequency content since eq . ( [ eq : fz - encoding ] ) has a single zero crossing in the interval @xmath175 $ ] for these two balanced strings only . all other balanced strings have more than one zero crossing and thus higher frequency content . for this pair of balanced functions , which we call the antisymmetric balanced ( asb ) functions , eq . ( [ eq : cvdjphaser2 ] ) gives the probability distribution @xmath212 where we have @xmath213 note that of the @xmath214 balanced functions , there are many that are also antisymmetric about the midpoint . however , we reserve the term asb for these two _ lowest - order _ antisymmetric balanced functions . we will use these cases to bound the success probability of distinguishing between the constant and all balanced cases . we first illustrate the concept of frequency in the following example . again consider the case @xmath215 , @xmath170 ; hence @xmath171 . as one case , the function corresponding to the four - bit string @xmath172 is @xmath216 this function corresponds to the @xmath113 asb function . the probability distributions for the four distinct @xmath113 cases are presented in fig . [ fig : cvdjsincftnsneq2 ] . we clearly see that of the three balanced cases , the @xmath113 asb function has probability peaks closest to @xmath160 . for ( a ) @xmath179 , ( b ) @xmath180 , ( c ) @xmath114 , and ( d ) @xmath181 . we clearly see that of the three balanced cases ( b ) through ( d ) , the asb function ( b ) has probability peaks closest to @xmath160.,width=8 ] our measurement strategy is to measure the probability distribution in a small band around the position @xmath160 parameterized by @xmath122 . the cv analog of the projection operator given in eq.([eq : qdjmeasure])is defined as @xmath217 where @xmath218[eq : cvdjdoorway]due to the symmetry of the sinc functions about @xmath136 , we set @xmath219 and @xmath220 . we now need to determine the optimal value of @xmath92 that will maximize our ability to distinguish between the constant and balanced cases . we will determine the optimum value of @xmath92 by first assuming that the probability distribution @xmath221 given by eq . ( [ eq : cvdjstate3pdfabal ] ) dominates all other balanced probability distributions in the region @xmath222 $ ] . after using this assumption to determine a value for the optimal delta , we will state and prove a theorem justifying our assumption . as an illustration that our assumption is true for the @xmath113 case , we plot the four distinct cases in fig . [ fig : cvdjoptimalmeasurelimits ] . the ability to effectively distinguish between two random events is proportional to the separation of the individual probabilities of occurrence . thus we need to select @xmath92 such that we get as much separation between the constant distribution and the asb distribution as possible . given this concept we can think that when we make a measurement we are distinguishing between two events , the probabilities for which we define as follows @xmath223= e^{\delta}_{-\delta}\left(\mathcal{p}_{\text{\tiny{c}}}(x)\right ) \label{eq : cvdjprobconst},\ ] ] and @xmath224= e^{\delta}_{-\delta}\left(\mathcal{p}_{\text{\tiny{asb}}}(x)\right ) \label{eq : cvdjprobasb}.\ ] ] , the optimal value of @xmath225 . this graph shows that only the constant and the antisymmetric balanced functions significantly contribute to probability between @xmath226 . , width=6 ] we can determine the optimum value of @xmath92 by maximizing the expression @xmath227 . it suffices to find the value of @xmath92 for which @xmath228 , which may be expressed as @xmath229this occurs where @xmath230 for @xmath231 , which gives a global maximum at @xmath232 . it is interesting to think of this result as an uncertainty relationship @xmath233we shall return to this concept in our discussion in the conclusion . we have determined the optimum value for @xmath92 based on our assumption that for @xmath234 the balanced probability distribution @xmath235 dominates all other balanced probability distributions . we now proceed to prove this assumption . in order to proceed with the proof , we define a set @xmath96 of @xmath5 pairwise conjugate angles with @xmath236 . note that @xmath6 is not restricted to being equal to @xmath45 for the purpose of this proof . also for the purpose of this proof , we set @xmath215 and incorporate @xmath72 into the definition of @xmath237 for @xmath238 . we let @xmath239 with @xmath240 and note the pairwise conjugate property @xmath241 . now consider @xmath242 where @xmath243\mapsto\pm 1 $ ] subject to the balanced condition @xmath244 , then max @xmath245 occurs under the specific balanced conditions @xmath246 and @xmath247 which we refer to as the asymmetric balanced functions ( asb ) . proof is done by induction on @xmath5 . we begin with the base case @xmath248 , @xmath249 . this case is trivial since the only balanced cases are the two asb cases represented by the strings @xmath250 . we proceed with the base case for @xmath251 , @xmath113 . this case is a little more involved . we begin by labelling the angles and phasors as shown in fig . [ fig : cvdjoangledefintionsforproof ] . base base . note that the effect of varying @xmath72 over @xmath252 $ ] simply focuses or expands the double angles @xmath253 and @xmath254 proportionally . , width=6 ] there are @xmath255 balanced cases , and we have to consider the strings @xmath256 . since the latter three are complements of the first three , we have to consider only three vector sums . with reference to fig . [ fig : cvdjoangledefintionsforproof ] , we have @xmath257 . we simplify and express the resultant along with the three other cases as 1 . @xmath258 2 . @xmath259 3 . @xmath260 . clearly @xmath261 . we use the trigonometric identities,@xmath262 and @xmath263 . we note that @xmath264 and @xmath265 for all @xmath6 and the specified range of @xmath72 . since @xmath266 for @xmath267 , we conclude that @xmath268 and thus @xmath269 for @xmath270 . this proves that the theorem is true for the @xmath251 , @xmath113 base case . we are now ready to prove the inductive step . we consider two cases . case ( i ) assumes every pair is balanced . by this we mean that @xmath271 . by inspection , this gives the same result as @xmath272 and @xmath273 for the @xmath251 , @xmath113 case . case(ii ) assumes that case(i ) is not true and is proved by induction . since case(i ) is not true , there must exist two non - balanced pairs for which @xmath274 and @xmath275 . as an illustration in the @xmath276 , @xmath197 case , the balanced string @xmath277 has this property . the inductive step is @xmath278 where @xmath279 is maximized for the @xmath251 , @xmath113 base case . only when @xmath280 itself is maximized is equality achieved and the total sum maximized . this occurs for the asb strings . we have established that we can bound the probabilities of determining whether an unknown function is balanced or constant in a single query in a cv setting . in the next section will determine an upper bound for the query complexity of a cv algorithm in terms of success probability in terms of the number of queries . before we bound the query complexity , we make some important observations regarding the comparison between the discrete dj algorithm and the cv dj algorithm . first , we note that probability distributions @xmath281 and @xmath235 defined by eqs . ( [ eq : cvdjstate3pdfconst ] ) and ( [ eq : cvdjstate3pdfabal ] ) respectively , are in @xmath282 , the hilbert space of @xmath83 functions over the interval @xmath283 $ ] . this implies that since we are measuring over a finite interval , the cv dj algorithm is necessarily probabilistic . furthermore , we noted that @xmath127 and @xmath92 are related by the uncertainty relation given in eq . ( [ eq : cvdjuncertainty ] ) . this leads to the conclusion that even in the limit of the improper delta function @xmath138 , the cv dj algorithm remains probabilistic . this conclusion is contrary to that made in @xcite . second , we compare the operator descriptions of the dv dj and the cv dj , which we express as @xmath284 the first equation represents the quantum dj algorithm operator expression given in eq . ( [ eq : qdjtargetlessoperator ] ) . the second equation is the analogous cv operator expression determined by concatenating the steps of the previous section . there is a high degree of similarity between these two expressions , but there are mathematical subtleties . 1 . the cv position state @xmath163 is not a perfect analog to the computational basis state @xmath56 except in the limit . however , this limit creates a state that is not in the rhs we argued is necessary for consistency @xcite . the continuous fourier transform is not equal to the cv extension of the hadamard operator in a cv parameterized system with a finite hilbert space . it is however , a convenient extension when the hilbert space is infinite . 3 . finally , the diagonal operator @xmath50 given by eq . ( [ eq : qdjtargetlessuf ] ) has each entry taking on the value @xmath285 dependent on the value of @xmath189 . the cv analogue to this operator is the function @xmath286 , where each of the @xmath6 partitions of the real interval @xmath175 $ ] similarly take on the value @xmath125 . we now determine numerical values of the probabilities determined in eqs . ( [ eq : cvdjprobconst ] ) and ( [ eq : cvdjprobasb ] ) . we can readily calculate the probability of detecting if the function is constant @xmath287 where the sine integral is given by @xmath288 note this probability depends only on the product @xmath289 . if the function is the lowest - order antisymmetric balanced ( asb ) @xmath290[eq : cvdjmeasureabalprb]for @xmath291 , the numerical values of these two probabilities are @xmath292 and @xmath293 given this probabilistic nature of the cv dj algorithm , we need to develop a strategy to bound the error probability . we will employ the technique sometimes called _ probability amplification _ @xcite . our strategy will be to make @xmath5 repetitions of the cv dj algorithm where we assume that the oracle is set to the same function for each of the repetitions . each repetition ends with a measurement . from this sequence of measurements we want to determine whether the unknown function is balanced or constant with high probability . an error of @xmath294 can be achieved by making @xmath295 repetitions of the cv dj algorithm . we will adopt the convention that when we make a query to the cv dj algorithm we either detect something ( algorithm returns a 1 ) , or we do not ( algorithm returns a 0 ) . we can thus treat multiple queries as a sequence of bernoulli trials @xcite . we assume that we have set our measurement limits to the optimal @xmath226 . the two events we are trying to uncover are the constant cases where , for ease of calculation we set the probability of detecting something is @xmath296 , and the balanced cases where the probability detecting something is @xmath297 . note that we have set the probabilities to these rational numbers for illustrative purposes and to simplify the calculation . we can make this arbitrary setting , and we will get the same result as long as the probabilities are bound from @xmath298 by a constant . if each measurement is based on an independent preparation of the state @xmath163 , then each of the queries are independent . after a series of @xmath5 queries , we can use the chernoff bounds of the binomial distribution to amplify the success probability @xcite . the simplest ( but somewhat weak ) chernoff bound on the lower tail is given by @xcite as @xmath299<\text{e}^{-\frac{\mu \epsilon^2}{2}},\label{eq : cvdjchernoffweakl}\end{aligned}\ ] ] and on the upper tail as @xmath300<\text{e}^{-\frac{\mu \epsilon^2}{4}},\label{eq : cvdjchernoffweaku}\end{aligned}\ ] ] where @xmath301 is the expected mean of the resulting binomial distributions after @xmath5 queries , and @xmath302 is the relative distance from the respective means . first , we bound the lower tail corresponding to the distribution of the constant case for which we have @xmath303 . here we set @xmath304 , which expresses the probability for the value being less than half way between the two means as @xmath305<\text{e}^{-\frac{m}{24}}$ ] . similarly , we bound the upper tail for the balanced case for which we have @xmath306 . here we set @xmath307 , which expresses the probability for the value being greater than half way between the two means as @xmath308<\text{e}^{-\frac{m}{16}}$ ] . clearly the success is worse for the lower tail allowing us to bound the success probability of the cv dj algorithm after @xmath5 queries as @xmath309\geq1-\text{e}^{-\frac{m}{24}}.\label{eq : cvdjsuccessprob}\end{aligned}\ ] ] this gives an error probability that is @xmath310 as required . we note that this is of the same order as the exponentially good success probability we have for the classical probabilistic approach given by eq . ( [ eq : classdjran ] ) . also note that this query complexity is independent of the value of @xmath6 . we have made no attempt to obtain a tighter bound preferring to show only that the success probability of the cv dj algorithm is of the same order of that of the classical probabilistic approach to solving the dj problem . in this paper we have presented a rigorous framework for the analysis of the dj oracle identification problem in a cv setting . the rigged hilbert space ( rhs ) affords a consistent transition from the traditional discrete hilbert space to the cv setting . our framework allows us to define a consistent way of encoding @xmath6-bit strings into functions over the real numbers . we have used this framework , and the selection of the sinc / pulse fourier transform pair , to prove that a cv implementation of the dj algorithm can not provide the exponential speed - up of its discrete quantum counterpart . additionally , we have presented a bounded - error , upper bound on the query complexity of the dj problem within the constraints of our model . the lack of speed - up results from an uncertainty principle between the ability to encode perfectly in a continuous representation and the subsequent inability to measure perfectly in the fourier - dual representation . this uncertainty relationship is manifest in eq . ( [ eq : cvdjuncertainty ] ) , which relates @xmath127 , the encoding extent , to @xmath92 , the measurement extent . a natural extension of this work would be prove a lower bound perhaps exploring the techniques along the lines of @xcite from the perspective of different fourier transform pairs . this uncertainty relationship appears to be a natural feature of the cv setting , but it could also be used to advantage . there is likely to be oracle function symmetries that are particularly suited to different cv settings . for example in sec . [ sec : cvrep ] , we showed that balanced functions with a higher number of zero crossings create sinc functions with frequency components further away from @xmath136 . it appears that an oracle identification problem designed to separate balanced functions according to frequency separation could be implemented in a cv setting and possibly provide advantage over classical or discrete quantum settings . furthermore , it would be interesting to classify the balanced functions from the perspective of different coherent states @xcite in cv parameterized settings of both finite and infinite dimensions . the former would naturally involve the study of the coherent spin systems @xcite . furthermore the use of squeezed spin states should be studied @xcite . infinite dimension systems would naturally involve the study of implementations involving the coherent states of quantum optics @xcite . additionally , we have set up this framework in a manner that should allow any oracle identification problem to be analyzed in a similar manner in the cv setting . an implementation of a discrete quantum oracle , for example @xcite , requires a unitary operator representing the oracle . provided we can create a diagonal representation of this oracle along the lines of @xmath50 given in eq . ( [ eq : qdjtargetlessoperator ] ) , our framework will naturally extend to it . of course we need to be able to create an implementation of these oracles and that remains an important open question . other avenues of the extension of this framework include cv implementations of other hidden subgroup problems . the solution of simon s problem @xcite in this setting would be an obvious starting point as would the exploration of a cv implementation of shor s algorithm @xcite . additionally , the cv framework could be extended to include analysis of noisy oracles along the lines of @xcite . in closing we note that the transition from a discrete quantum information setting to a cv setting has many subtleties . in particular the improper delta functions must not be used . limiting behaviour can be explored , but only if the limits are taken from the perspective of functions defined in the rigged hilbert space . we appreciate financial support from the alberta ingenuity fund ( aif ) , alberta s informatics circle of research excellence ( _ _ i__core ) , canada s natural sciences and engineering research council ( nserc ) , the canadian network centres of excellence for mathematics of information technology and complex systems ( mitacs ) , and general dynamics canada . ph is a scholar and bcs is an associate of the canadian institute for advanced research ( cifar ) . d. akamatsu , y. yokoi , m. arikawa , s. nagatsuka , t. tanimura , a. furusawa , and m. kozuma , `` ultraslow propagation of squeezed vacuum pulses with electromagnetically induced transparency '' , * 99 * , 153602 ( 2007 ) . c. m. caves , `` physical resources , entanglement , and the power of quantum computation '' , lecture given in squint summer retreat university of southern california , 2005 july 7 . reference downloaded 03 dec . 2008 . m. adcock `` the classical and quantum complexity of the goldreich - levin problem with applications to bit commitment '' , _ master of science thesis , department of computer science , university of calgary ( 2004)_. r. cleve , w. van dam , m. nielsen , and a. tapp , `` quantum entanglement and the communication complexity of the inner product function '' , lecture notes in computer science * 1509 * ( springer - verlag ) , pp . 61 - 74 ( 1999 ) . m. adcock and r. cleve , `` a quantum goldreich - levin theorem with cryptographic applications '' _ proc . 19@xmath311 international symp aspects comp . ( stacs 2002 ) _ , h. alt and a. ferreira , eds . , lecture notes in computer science * 2285 * ( springer - verlag ) , pp . 323 - 334 ( 2002 ) .

the photometric variability of the eclipsing binary star lv her ( tyc 2076 - 1042 - 1 ; @xmath9 , @xmath10 , j2000.0 ; spt f9 , @xmath11 ) was announced by @xcite . its period was estimated by @xcite as 2.634 days , by @xcite as 5.2674 days , by @xcite as 9.218 days , by @xcite as 18.1312 days , and by @xcite as 18.4359350 days . all photometric and spectroscopic work performed since this last study has confirmed that it is basically correct . the prior erroneous period values resulted from the very long orbital period , the similar eclipse depths ( 0.68 mag and 0.66 mag in @xmath0 ; see below ) , the extremely high eccentricity of the orbit ( 0.613 ) , and the resultant shift of secondary eclipse to a phase of 0.86 . this sort of erroneous period estimation is not unusual in this kind of situation @xcite . largely because of this uncertainty , no determination of the absolute dimensions of the system has been made until now . in addition to an accurate determination of the eclipse ephemeris in [ sec : ephemeris ] , this work presents new high - quality photometric and spectroscopic observations ( [ sec : photometry ] and [ sec : spectroscopy ] ) that yield masses and radii on a par with the best determinations for eclipsing binaries to date ( [ sec : dimensions ] ) . in [ sec : evolution ] we compare these determinations with models of stellar structure and stellar evolution . the high eccentricity and long period of lv her make it an interesting case for comparison with tidal theory , which we present in [ sec : tidal ] . we discuss our results in [ sec : discussion ] . our photometric work on lv her began in 2001 and is based on observations obtained with two robotic instruments : the ursa telescope at kimpel observatory @xcite on the campus of the university of arkansas , and a robotic telescope at the nf observatory ( nfo ; * ? ? ? * ) near silver city , nm . kimpel observatory ( ursa.uark.edu ) consists of a meade 10-inch f/6.3 lx-200 telescope with a santa barbara instruments group st8 ccd camera ( binned @xmath12 to produce @xmath13 pixel images with 2.3 arcsec square pixels ) inside a technical innovations robo - dome , and controlled automatically by an apple macintosh g4 computer . the observatory is located on top of kimpel hall on the fayetteville campus of the university of arkansas , with the control room directly beneath the observatory inside the building . exposures of 60 or 120 seconds made through a bessell @xmath0 filter ( 2.0 mm of gg 495 and 3.0 mm of bg 39 ) were read out and downloaded by imagegrabber ( camera control software written by j. sabby ) to the control computer over a 30 second interval , then the next exposure was begun . the observing cadence was therefore about 90 to 150 sec per observation . the variable star would sometimes be monitored continuously for 46 hours . lv her was observed by ursa on 156 nights during parts of nine observing seasons from 2001 february 18 to 2009 april 8 . the total number of ursa observations is 6690 . the frames were analyzed by a virtual measuring engine application written by lacy that flat - fielded the images , automatically located the variable and comparison stars in the image , measured their brightnesses , subtracted the corresponding sky brightness , and corrected for the differences in airmass between the stars . extinction coefficients were determined nightly from the comparison star measurements . they averaged 0.25 mag / airmass . lv her is tyc 2076 - 1042 - 1 . the comparison stars were tyc 2076 - 0580 - 1 ( comp , @xmath14 , as listed in the tycho-2 catalogue ; * ? ? ? * ) , and tyc 2076 - 1387 - 1 ( ck , @xmath15 ) . both comparison stars are within 10 arcmin of the variable star ( var ) . the comparison star magnitude differences ( comp@xmath16ck ) were constant at the level of 0.013 mag ( standard deviation within a night ) , and 0.010 mag for the standard deviation of the nightly mean magnitude difference . the ursa differential magnitude ( var@xmath16comp ) of the variable star was referenced only to the magnitude of the comparison star , comp . the resulting 6690 @xmath0 magnitude differences ( var@xmath16comp ) are listed in table [ tab : ursa ] ( without any nightly corrections ) and plotted in figures [ fig : lcall][fig : lcsec ] ( after the nightly corrections discussed below have been applied ) . the eclipse depths are 0.68 mag for the primary and 0.66 mag for the secondary . the precision of the variable star differential magnitudes is about 0.016 mag . ccc 51958.87391 & @xmath160.195 & 0.15136 + 51958.87498 & @xmath160.147 & 0.15141 + 51958.87603 & @xmath160.174 & 0.15147 + 51958.87708 & @xmath160.179 & 0.15153 + 51958.87812 & @xmath160.183 & 0.15158 + [ -1.5ex ] ccc 53430.93043 & 1.003 & 0.99840 + 53430.93323 & 1.010 & 0.99855 + 53430.93602 & 1.035 & 0.99870 + 53430.93878 & 1.055 & 0.99885 + 53430.94158 & 1.068 & 0.99900 + [ -1.5ex ] the other telescope we used is the nfo webscope , a refurbished 24-inch group 128 cassegrain reflector with a 2k @xmath17 2k kodak ccd camera , located near silver city , nm . observations consisted of 120 second exposures also through a bessell @xmath0 filter . lv her was observed by nfo on 123 nights during parts of five observing seasons from 2005 march 1 to 2009 april 24 , yielding 2946 observations . extinction coefficients were determined nightly from the comparison star measurements . they averaged 0.18 mag / airmass . the same comparison stars were used as those of the ursa telescope . the comparison star magnitude differences ( comp@xmath16ck ) were constant at the level of 0.007 mag ( standard deviation within a night ) , and 0.020 mag for the standard deviation of the nightly mean magnitude difference . the differential magnitude ( var@xmath16comps ) of the variable star was referenced to the magnitude corresponding to the sum of the intensities of the comparison star , comp , and the check star , ck . the resulting 2946 @xmath0 magnitude differences ( var@xmath16comps ) are listed in table [ tab : nfo ] ( without nightly corrections ) and plotted in figure [ fig : lcall][fig : lcsec ] ( including nightly corrections ; see below ) . the precision of the variable star differential magnitudes is about 0.010 mag . we noticed early on during the observations that the nfo magnitudes showed a small but significant offset from night to night , on the order of a hundredth of a magnitude . the origin of the offset is a variation in responsivity across the field of view of the nfo combined with imprecise centering from night to night . these variations are a well - known effect of the optics when using wide - field imaging telescopes such as the nfo . we have removed most of this variation by using dithered exposures of open star clusters to measure this variation , fitting a 2-d polynomial ( see * ? ? ? * ) , and subtracting the variation during initial reductions ( photometric flat ) . the ursa observations , on the other hand , show this kind of effect to a very much smaller extent . we have removed these nightly offsets before further analysis by using a procedure discussed below in [ sec : lcfits ] . absolute photometry of lv her is available in the literature from several sources , and in several photometric systems including strmgren , 2mass , tycho-2 , and johnson - cousins . we collect these measurements in table [ tab : photometry ] . color indices formed from these magnitudes can be used to estimate a mean effective temperature for the binary , which we discuss below in [ sec : dimensions ] . the interstellar reddening can be estimated , for example , by comparison with the standard strmgren indices of @xcite , giving @xmath18 mag , or @xmath19 . another estimate is available from the reddening maps of the sky near the variable star from @xcite , which results in a larger value of @xmath20 mag at the estimated distance of the eclipsing binary . we adopt for further use the straight average of these two reddening estimates , @xmath21 mag , with a conservative error . lcc @xmath22 & 11.02 @xmath23 0.02 & 1 + @xmath24 & 1.552 @xmath23 0.030 & 1 + @xmath25 & 0.571 @xmath23 0.020 & 1 + @xmath26 & 0.367 @xmath23 0.015 & 1 + @xmath0 & 11.055 @xmath23 0.040 & 2 + @xmath27 & 10.367 @xmath23 0.058 & 2 + @xmath28 & 9.905 @xmath23 0.018 & 3 + @xmath29 & 9.665 @xmath23 0.025 & 3 + @xmath30 & 9.631 @xmath23 0.018 & 3 + @xmath31 & 11.832 @xmath23 0.066 & 4 + @xmath32 & 11.045 @xmath23 0.055 & 4 + [ -1.5ex ] since the work of @xcite , additional times of minimum for lv her have been reported in the literature , and some recent ones from our webscopes are available that have not yet been published . the previously published ones and these additional times of minimum light ( 37 for the primary , 30 for the secondary ) are listed in table [ tab : minima ] . all available minima have been used to estimate an eclipse ephemeris by a least squares technique . when available , the uncertainties have been adopted as published . for the others ( visual and photographic measurements ) , uncertainties have been estimated by iterations requiring the reduced @xmath33 to be near unity , separately for each class of measurement technique . for the visual meaurements we obtain @xmath34 days , and for the photographic timings @xmath35 days . separate solutions with the primary and secondary data give periods that are not significantly different . for the final ephemeris we enforced a common period , and obtained : @xmath36 where the reference epochs were chosen so as to minimize the correlations with the period and with each other . this ephemeris is the basis of the orbital phases we cite below . the phase of secondary eclipse is @xmath37 . a plot of the @xmath38 diagram from this fit is shown in figure [ fig : minima ] , and the residuals are included in table [ tab : minima ] . it is quite remarkable how much the accuracy has improved over the last century . lcccccc 15929.712 & 1902.4907 & 2 & pg & 0.053 & @xmath390.01915 & 1 + 16224.798 & 1903.2986 & 2 & pg & 0.053 & @xmath390.12989 & 1 + 16669.658 & 1904.5165 & 1 & pg & 0.053 & @xmath160.01072 & 1 + 17351.799 & 1906.3841 & 1 & pg & 0.053 & @xmath160.00000 & 1 + 18605.473 & 1909.8165 & 1 & pg & 0.053 & @xmath390.02916 & 1 + 23601.681 & 1923.4954 & 1 & pg & 0.053 & @xmath390.09376 & 1 + 24062.505 & 1924.7570 & 1 & pg & 0.053 & @xmath390.01892 & 1 + 26032.643 & 1930.1510 & 2 & pg & 0.053 & @xmath390.04762 & 1 + 26032.644 & 1930.1510 & 2 & pg & 0.053 & @xmath390.04862 & 1 + 26090.482 & 1930.3093 & 1 & pg & 0.053 & @xmath390.04103 & 1 + 26219.577 & 1930.6628 & 1 & pg & 0.053 & @xmath390.08436 & 1 + 26901.699 & 1932.5303 & 1 & pg & 0.053 & @xmath390.07608 & 1 + 27212.514 & 1933.3813 & 2 & pg & 0.053 & @xmath390.01760 & 1 + 27636.462 & 1934.5420 & 2 & pg & 0.053 & @xmath160.06134 & 1 + 27657.429 & 1934.5994 & 1 & pg & 0.053 & @xmath160.06802 & 1 + 28281.830 & 1936.3089 & 2 & pg & 0.053 & @xmath390.04829 & 1 + 28429.2740 & 1936.7126 & 2 & vis & 0.017 & @xmath390.00466 & 2 + 28487.1220 & 1936.8710 & 1 & vis & 0.017 & @xmath390.00708 & 2 + 30254.522 & 1941.7098 & 2 & pg & 0.053 & @xmath390.09326 & 1 + 31268.42 & 1944.4857 & 2 & vis & 0.017 & @xmath390.01382 & 3 + 31289.35 & 1944.5431 & 1 & vis & 0.04 & @xmath160.02986 & 3 + 31326.29 & 1944.6442 & 1 & vis & 0.017 & @xmath390.03823 & 3 + 31342.16 & 1944.6876 & 2 & vis & 0.017 & @xmath390.01001 & 3 + 34626.332 & 1953.6792 & 1 & pg & 0.053 & @xmath390.04456 & 1 + 37444.539 & 1961.3950 & 2 & pg & 0.053 & @xmath390.08839 & 1 + 38147.573 & 1963.3198 & 1 & pg & 0.053 & @xmath390.01843 & 1 + 38587.449 & 1964.5242 & 2 & pg & 0.053 & @xmath160.03072 & 1 + 40744.498 & 1970.4298 & 2 & pg & 0.053 & @xmath390.01171 & 1 + 41060.502 & 1971.2950 & 1 & pg & 0.053 & @xmath390.06678 & 1 + 41982.251 & 1973.8186 & 1 & pg & 0.053 & @xmath390.01810 & 1 + 42664.347 & 1975.6861 & 1 & pg & 0.053 & @xmath160.01618 & 1 + 46001.268 & 1984.8221 & 1 & pg & 0.053 & @xmath160.00276 & 1 + 46941.483 & 1987.3963 & 1 & pg & 0.053 & @xmath160.02139 & 1 + 47381.384 & 1988.6006 & 2 & pg & 0.053 & @xmath160.04555 & 1 + 48100.425 & 1990.5693 & 2 & pg & 0.053 & @xmath160.00674 & 1 + 48487.60 & 1991.6293 & 2 & vis & 0.017 & @xmath390.01328 & 1 + 48508.60 & 1991.6868 & 1 & vis & 0.04 & @xmath390.03956 & 1 + 48545.52 & 1991.7879 & 1 & vis & 0.04 & @xmath390.08765 & 1 + 48745.70 & 1992.3359 & 2 & vis & 0.017 & @xmath390.00989 & 1 + 48948.50 & 1992.8912 & 2 & vis & 0.017 & @xmath390.01440 & 1 + 49098.540 & 1993.3020 & 1 & pg & 0.053 & @xmath390.02904 & 1 + 49206.61 & 1993.5978 & 2 & vis & 0.017 & @xmath390.02105 & 1 + 49667.49 & 1994.8597 & 2 & vis & 0.017 & @xmath390.00221 & 1 + 49688.48 & 1994.9171 & 1 & vis & 0.017 & @xmath390.01853 & 1 + 49925.65 & 1995.5665 & 2 & vis & 0.02 & @xmath390.05886 & 1 + 52008.85 & 2001.2700 & 2 & ccd & 0.01 & @xmath160.00388 & 1 + 52045.74 & 2001.3710 & 2 & ccd & 0.01 & @xmath390.01421 & 1 + 52066.6993 & 2001.4283 & 1 & ccd & 0.0008 & @xmath160.00017 & 1 + 52432.88040 & 2002.4309 & 2 & ccd & 0.00050 & @xmath160.00042 & 4 * + 52490.72613 & 2002.5893 & 1 & ccd & 0.00020 & @xmath160.00027 & 4 + 52785.7012 & 2003.3969 & 1 & ccd & 0.0010 & @xmath160.00046 & 5 + 53154.4210 & 2004.4064 & 1 & ccd & 0.0017 & @xmath390.00027 & 6 + 53154.4212 & 2004.4064 & 1 & ccd & 0.0015 & @xmath390.00047 & 6 + 53209.7288 & 2004.5578 & 1 & ccd & 0.0004 & @xmath390.00021 & 7 + 53430.95992 & 2005.1635 & 1 & ccd & 0.00016 & @xmath160.00011 & 8 + 53870.8862 & 2006.3679 & 2 & ccd & 0.0005 & @xmath390.00101 & 9 * + 53907.7573 & 2006.4689 & 2 & ccd & 0.0002 & @xmath390.00020 & 9 + 53928.7308 & 2006.5263 & 1 & ccd & 0.0003 & @xmath390.00002 & 9 + 54297.44944 & 2007.5358 & 1 & ccd & 0.00040 & @xmath160.00041 & 10 + 54297.4498 & 2007.5358 & 1 & ccd & 0.0008 & @xmath160.00005 & 11 + 54297.4509 & 2007.5358 & 1 & ccd & 0.0004 & @xmath390.00105 & 12 + 54331.7839 & 2007.6298 & 2 & ccd & 0.0003 & @xmath160.00013 & 13 + 54368.6555 & 2007.7307 & 2 & ccd & 0.0003 & @xmath160.00043 & 13 + 54589.8873 & 2008.3364 & 2 & ccd & 0.0002 & @xmath160.00008 & 13 + 54647.7328 & 2008.4948 & 1 & ccd & 0.0006 & @xmath160.00016 & 13 + 54647.7332 & 2008.4948 & 1 & ccd & 0.0002 & @xmath390.00024 & 13 + 54868.9645 & 2009.1005 & 1 & ccd & 0.0007 & @xmath390.00010 & 13 + [ -1.5ex ] lv her was placed on the observing list at the harvard - smithsonian center for astrophysics ( cfa ) on 1991 may 31st , and was monitored spectroscopically for exactly 11 years with an echelle instrument on the 1.5 m tillinghast reflector at the f. l. whipple observatory on mount hopkins , arizona . we collected 42 spectra with a photon - counting intensified reticon detector @xcite in a single order 45 wide centered near 5187 . the main features in this spectral window are the lines of the b triplet . the resolving power provided by this setup is @xmath40 . two additional observations , for a total of 44 , were obtained with a nearly identical system on the 1.5 m wyeth reflector at the oak ridge observatory ( harvard , massachusetts ) . the signal - to - noise ratios of these spectra range from 13 to 28 per resolution element of 8.5 kms@xmath41 . radial velocities for the two components were derived using todcor , a two - dimensional cross - correlation technique @xcite . this method uses two templates , one for each component of the binary , which we selected from a large library of synthetic spectra based on model atmospheres by r. l. kurucz ( see * ? ? ? these templates have been calculated for a wide range of effective temperatures ( @xmath42 ) , surface gravities ( @xmath43 ) , rotational velocities ( @xmath44 when seen in projection ) , and metallicities ( [ m / h ] ) . following @xcite the optimum templates for each star were determined by means of extensive grids of cross - correlations with todcor , seeking to maximize the average correlation weighted by the strength of each exposure . because of the strong correlation between temperature , surface gravity , and metallicity in our narrow spectral window , we initially assumed solar metallicity , and surface gravities of @xmath45 for both components , close to the values determined from a preliminary analysis . the projected rotational velocities we obtained are @xmath46 kms@xmath41 for both stars , and the best - fit temperatures were near solar , with a difference of only 2030 k between the primary and secondary . for the radial velocity measurements we adopted @xmath47 k. in [ sec : dimensions ] we describe experiments in which we extended the grids of correlations to other compositions in order to fine - tune the temperatures and attempt to constrain the metallicity . these minor changes in the stellar parameters have no effect on the radial velocities . typical errors for our measurements are 0.8 kms@xmath41 for the primary and 1.1 kms@xmath41 for the secondary . the stability of the zero - point of the cfa velocity system was monitored by means of exposures of the dusk and dawn sky , and small systematic run - to - run corrections were applied in the manner described by @xcite . we determined also the light ratio directly from our spectra following @xcite , and obtained @xmath48 at the mean wavelength of our observations , which is essentially also the ratio in the visual band given that the stars are nearly identical in temperature . lc[b ] + @xmath49 ( days ) & 18.4359535 + @xmath50 ( hjd@xmath162,400,000 ) & 53,652.19147 + @xmath51 ( kms@xmath41 ) & 67.24 @xmath23 0.19 + @xmath52 ( kms@xmath41 ) & 68.59 @xmath23 0.27 + @xmath53 ( kms@xmath41 ) & @xmath54 @xmath23 0.094 + @xmath55 & 0.61273 @xmath23 0.00073 + @xmath56 ( deg ) & 352.20 @xmath23 0.24 + + @xmath57 ( @xmath58 ) & 1.193 @xmath23 0.010 + @xmath59 ( @xmath58 ) & 1.1697 @xmath23 0.0080 + @xmath60 & 0.9803 @xmath23 0.0047 + @xmath61 ( 10@xmath62 km ) & 13.471 @xmath23 0.037 + @xmath63 ( 10@xmath62 km ) & 13.743 @xmath23 0.054 + @xmath64 ( @xmath65 ) & 39.119 @xmath23 0.095 + + @xmath66 & 44 + time span ( days ) & 4018.0 + @xmath67 ( kms@xmath41 ) & 0.76 + @xmath68 ( kms@xmath41 ) & 1.10 + [ -1.5ex ] lccccccc[b ] 48407.8794 & @xmath1636.77 & @xmath3916.94 & 0.75 & 1.09 & @xmath160.38 & @xmath390.59 & 0.5389 + 48428.8481 & @xmath1630.28 & @xmath3910.75 & 0.77 & 1.12 & @xmath390.32 & @xmath390.30 & 0.6763 + 48435.8126 & @xmath392.41 & @xmath1624.64 & 0.81 & 1.17 & @xmath160.56 & @xmath160.85 & 0.0541 + 48459.7950 & @xmath1633.43 & @xmath3915.58 & 0.84 & 1.22 & @xmath390.97 & @xmath391.25 & 0.3549 + 48461.7375 & @xmath1635.70 & @xmath3918.08 & 0.85 & 1.23 & @xmath390.88 & @xmath391.53 & 0.4603 + 48462.6830 & @xmath1636.20 & @xmath3917.06 & 0.84 & 1.22 & @xmath390.44 & @xmath390.44 & 0.5116 + 51709.7370 & @xmath1633.05 & @xmath3912.51 & 0.82 & 1.19 & @xmath390.18 & @xmath160.62 & 0.6377 + 51710.7867 & @xmath1628.71 & @xmath398.12 & 0.69 & 1.00 & @xmath390.21 & @xmath160.61 & 0.6947 + 51711.7546 & @xmath1622.10 & @xmath390.39 & 0.72 & 1.04 & @xmath160.34 & @xmath161.05 & 0.7472 + 51712.7284 & @xmath166.38 & @xmath1613.66 & 0.81 & 1.18 & @xmath391.76 & @xmath161.20 & 0.8000 + 51715.7770 & @xmath3962.12 & @xmath1684.03 & 0.77 & 1.11 & @xmath160.20 & @xmath390.31 & 0.9653 + 51740.7610 & @xmath1634.04 & @xmath3912.50 & 0.64 & 0.93 & @xmath160.98 & @xmath160.46 & 0.3205 + 51741.7446 & @xmath1635.18 & @xmath3914.19 & 0.72 & 1.04 & @xmath160.18 & @xmath160.75 & 0.3739 + 51742.7170 & @xmath1636.06 & @xmath3915.62 & 0.74 & 1.07 & @xmath390.12 & @xmath160.52 & 0.4266 + 51744.7057 & @xmath1637.70 & @xmath3915.90 & 0.83 & 1.21 & @xmath161.26 & @xmath160.51 & 0.5345 + 51799.6219 & @xmath1636.83 & @xmath3917.30 & 0.73 & 1.05 & @xmath160.20 & @xmath390.69 & 0.5132 + 51800.6010 & @xmath1637.55 & @xmath3914.68 & 0.69 & 1.00 & @xmath161.65 & @xmath161.17 & 0.5663 + 51802.6229 & @xmath1630.48 & @xmath3910.83 & 0.66 & 0.95 & @xmath390.14 & @xmath390.36 & 0.6760 + 51803.6558 & @xmath1624.42 & @xmath393.48 & 0.64 & 0.92 & @xmath160.15 & @xmath160.51 & 0.7320 + 51831.5984 & @xmath1628.65 & @xmath398.50 & 0.75 & 1.08 & @xmath390.24 & @xmath160.21 & 0.2477 + 51833.6157 & @xmath1635.73 & @xmath3914.79 & 0.75 & 1.09 & @xmath161.25 & @xmath390.38 & 0.3571 + 51834.5731 & @xmath1635.52 & @xmath3917.04 & 0.72 & 1.04 & @xmath390.34 & @xmath391.22 & 0.4091 + 51858.5764 & @xmath1626.02 & @xmath395.88 & 0.73 & 1.06 & @xmath391.09 & @xmath161.01 & 0.7110 + 51861.5870 & @xmath3942.50 & @xmath1666.00 & 0.77 & 1.12 & @xmath161.39 & @xmath160.46 & 0.8743 + 51973.9772 & @xmath3956.88 & @xmath1678.61 & 0.79 & 1.14 & @xmath390.26 & @xmath160.09 & 0.9706 + 51977.9683 & @xmath1623.19 & @xmath394.08 & 0.68 & 0.98 & @xmath390.30 & @xmath390.88 & 0.1871 + 52006.9280 & @xmath1620.18 & @xmath160.38 & 0.83 & 1.21 & @xmath160.49 & @xmath390.30 & 0.7579 + 52009.8622 & @xmath3996.55 & @xmath16119.15 & 0.81 & 1.18 & @xmath390.79 & @xmath160.70 & 0.9171 + 52010.8700 & @xmath3954.74 & @xmath1677.11 & 0.80 & 1.16 & @xmath160.68 & @xmath390.19 & 0.9717 + 52011.9261 & @xmath3913.95 & @xmath1633.43 & 0.76 & 1.10 & @xmath390.58 & @xmath390.97 & 0.0290 + 52032.9120 & @xmath1621.81 & @xmath390.64 & 0.75 & 1.09 & @xmath160.63 & @xmath160.20 & 0.1673 + 52034.8450 & @xmath1628.93 & @xmath399.98 & 0.71 & 1.04 & @xmath391.59 & @xmath160.39 & 0.2722 + 52035.8449 & @xmath1633.46 & @xmath3914.00 & 0.70 & 1.02 & @xmath160.15 & @xmath390.78 & 0.3264 + 52039.9325 & @xmath1636.23 & @xmath3916.20 & 0.73 & 1.05 & @xmath390.02 & @xmath160.01 & 0.5481 + 52102.6290 & @xmath3980.99 & @xmath16102.64 & 0.87 & 1.26 & @xmath160.07 & @xmath390.81 & 0.9489 + 52138.6401 & @xmath3982.46 & @xmath16106.28 & 1.31 & 1.91 & @xmath391.04 & @xmath162.45 & 0.9022 + 52156.6725 & @xmath3951.60 & @xmath1673.90 & 0.71 & 1.04 & @xmath390.17 & @xmath160.67 & 0.8803 + 52157.6303 & @xmath3995.25 & @xmath16118.44 & 0.70 & 1.01 & @xmath160.44 & @xmath160.06 & 0.9323 + 52211.5883 & @xmath3927.39 & @xmath1642.70 & 0.89 & 1.30 & @xmath390.01 & @xmath395.99 & 0.8590 + 52360.9241 & @xmath3969.40 & @xmath1691.63 & 0.78 & 1.13 & @xmath390.24 & @xmath160.31 & 0.9593 + 52420.8295 & @xmath1625.10 & @xmath394.20 & 0.77 & 1.11 & @xmath390.58 & @xmath161.23 & 0.2087 + 52421.8579 & @xmath1630.47 & @xmath399.39 & 0.78 & 1.13 & @xmath160.43 & @xmath160.49 & 0.2645 + 52423.9196 & @xmath1635.64 & @xmath3915.54 & 0.73 & 1.06 & @xmath160.57 & @xmath390.53 & 0.3763 + 52425.8862 & @xmath1635.66 & @xmath3917.17 & 0.77 & 1.11 & @xmath391.03 & @xmath390.51 & 0.4830 + [ -1.5ex ] although todcor significantly reduces systematic errors in the radial velocities caused by line blending , residual effects can remain as a result of shifts of the spectral lines in and out of our narrow spectral window as a function of orbital phase . we investigated these effects by means of numerical simulations similar to those described by @xcite ( see also * ? ? ? * ; * ? ? ? we generated synthetic composite spectra matching our observations by combining copies of the primary and secondary templates used above , shifted to the appropriate velocities for each of the exposures as predicted by a preliminary orbital solution , and scaled to the observed light ratio . these synthetic observations were then processed with todcor in exactly the same way as the real spectra , and the resulting velocities were compared with the input shifts . the differences for lv her were typically well under 0.5 kms@xmath41 for both stars , but were nevertheless applied as corrections to the raw velocities . they affect the masses at the level of 1.3% for the primary and 0.8% for the secondary , which are similar to the formal errors . the corrected measurements are listed in table [ tab : rvs ] along with their uncertainties . similar adjustments based on the same simulations were made to the light ratio , and are already included in the value reported above . our orbital solution is shown in figure [ fig : sb2orbit ] , and was derived holding the ephemeris fixed according to [ sec : ephemeris ] . the residuals are shown as well , and are listed in table [ tab : rvs ] . the orbital elements and derived quantities ( minimum masses , semimajor axes , etc . ) are given in table [ tab : specorbit ] . lccc[b ] @xmath69 & 0.967 @xmath23 0.004 & 0.983 @xmath23 0.003 & 0.9772 @xmath23 0.0080 + @xmath70 & 0.06880 @xmath23 0.00009 & 0.06806 @xmath23 0.00006 & 0.06829 @xmath23 0.00037 + @xmath71 & 0.986 @xmath23 0.015 & 0.961 @xmath23 0.009 & 0.968 @xmath23 0.012 + @xmath72 & 0.03465 @xmath23 0.00026 & 0.03471 @xmath23 0.00015 & 0.03470 @xmath23 0.00028 + @xmath73 & 0.03415 @xmath23 0.00026 & 0.03335 @xmath23 0.00016 & 0.03357 @xmath23 0.00028 + @xmath74 ( deg ) & 89.489 @xmath23 0.010 & 89.509 @xmath23 0.009 & 89.500 @xmath23 0.010 + @xmath75 & 0.60662 @xmath23 0.00011 & 0.60739 @xmath23 0.00006 & 0.60721 @xmath23 0.00038 + @xmath76 & @xmath160.0938 @xmath23 0.0019 & @xmath160.0796 @xmath23 0.0012 & @xmath160.0836 @xmath23 0.0071 + @xmath55 & 0.61383 @xmath23 0.00018 & 0.61259 @xmath23 0.00010 & 0.61288 @xmath23 0.00062 + @xmath56 ( deg ) & 351.21 @xmath23 0.18 & 352.53 @xmath23 0.12 & 352.12 @xmath23 0.66 + @xmath77 & 0.516 @xmath23 0.007 & 0.524 @xmath23 0.004 & 0.5220 @xmath23 0.0040 + @xmath78 & 0.484 @xmath23 0.007 & 0.476 @xmath23 0.004 & 0.4780 @xmath23 0.0040 + @xmath79 & 0.938 @xmath23 0.024 & 0.908 @xmath23 0.014 & 0.916 @xmath23 0.015 + @xmath80 & 0.53 @xmath23 0.04 & 0.55 @xmath23 0.04 & 0.54 @xmath23 0.03 + @xmath66 & 6690 & 2946 & + time span ( days ) & 2970.9 & 1514.9 & + @xmath81 ( mmag ) & 12.49 & 6.42 & + in order to remove the small nightly offsets in the differential photometry , we performed preliminary fits of the ursa and nfo light curves by using the nelson - davis - etzel model @xcite as implemented in the ebop code . this model is well suited for well - detached systems such as lv her . the nightly residuals , which were typically less than 0.02 mag and were uncorrelated with phase , were removed from each light curve . this improved the residuals of the ursa data by about 27% , and those from nfo by a more significant 42% . the corrected light curves were then fitted by using the jktebop code of @xcite , based on the same model , which allows for more realistic estimates of the uncertainties . the adjustable parameters are the central surface brightness @xmath69 of the secondary relative to the primary , the inclination angle @xmath74 , the sum of the relative radii @xmath70 , the ratio of the radii @xmath82 , the eccentricity factors @xmath75 and @xmath83 , a phase shift , and the magnitude at quadrature . a linear limb - darkening law was adopted , consistent with our experience that with the amount and precision of our data , nonlinear laws do not improve the accuracy of the fits significantly ( e.g. , * ? ? ? * ; * ? ? ? the coefficient @xmath84 was left free , and constrained to be the same for the two stars . the gravity brightening exponent was set to 0.35 for both stars , based on the calculations by @xcite and the mean temperature and surface gravity of the components . the mass ratio @xmath85 was adopted from the spectroscopy . separate solutions for the ursa and nfo photometry are given in table [ tab : lcfit ] . the uncertainties for the fitted parameters were evaluated by monte carlo simulation with jktebop . tests indicated third light was not significant , and it was therefore set to zero . the last column of the table gives the adopted fit ( weighted average ) , with uncertainties that account for the difference between ursa and nfo . the uncertainties for the individual radii were derived following the prescription recommended by @xcite , based on the errors for @xmath70 and @xmath86 . the synthetic curve corresponding to the adopted solution is displayed in figures [ fig : lcall][fig : lcsec ] along with the observations . the eccentricity and @xmath56 from the light curves are in excellent agreement with the spectroscopic values in table [ tab : specorbit ] , and the photometric light ratio @xmath87 is consistent with the estimate of @xmath88 from the cfa spectra . the well - detached stars are essentially spherical , and both eclipses are partial , with 88.4% of the primary light blocked at the primary minimum , and 95.6% of the secondary light covered at the secondary minimum . we note that the linear limb - darkening coefficient we determined , @xmath89 , is consistent with theoretical value of 0.568 for the visual band according to @xcite , but is significantly smaller than the value 0.658 from the tables by @xcite , for a star of this temperature and surface gravity . similar discrepancies have been found in other systems ( see , e.g. , * ? ? ? * ) . as a test we repeated the light curve fits holding the limb - darkening coefficient fixed at this last theoretical value . the change in @xmath70 was only + 0.15% , the individual radii increased by less than 0.1% , and the inclination angle decreased by @xmath90 . these changes do not impact the absolute masses or radii at a significant level , compared to other sources of uncertainty . the photometric orbit and the spectroscopic orbit may be combined to yield the absolute masses and radii of the binary stars components . these are given in table [ tab : dimensions ] , along with other properties described below . the masses and radii are determined to 0.9% or better for both stars . the tests and checks for systematics described above suggest that they are also accurate at this level . the ratio of the effective temperatures is very well constrained from the light curves through the central surface brightness parameter @xmath69 , according to which the stars differ by only about 30 k ( * ? ? * table 1 ) . a similar difference was obtained from the cfa spectra . the _ absolute _ temperatures of the stars may in principle be estimated also from the spectroscopy , though in practice this is difficult with the material at hand because of the correlation with metallicity mentioned earlier . we first determined the magnitude of this dependence by repeating the grids of correlations described in [ sec : spectroscopy ] for metallicities between [ m / h ] @xmath91 and [ m / h ] @xmath92 in steps of 0.5 dex , and surface gravities of @xmath45 and 4.5 . at each composition we determined the primary and secondary temperatures interpolated to the surface gravities indicated in table [ tab : dimensions ] , as well as the luminosity - weighted mean temperature for the system ( where the weights depend only on the radius ratio and temperature ratio that come from the light curve fits ) . an additional constraint on the mean system temperature is available from the absolute photometry of lv her presented earlier , through color indices and color - temperature calibrations ( including metallicity terms ) . equating the two mean temperatures then allows to solve for the metallicity . with the photometry listed in table [ tab : photometry ] we formed 7 different color indices , we de - reddened them as indicated in [ sec : photometry ] , and we made use of three different calibrations by @xcite , @xcite , and @xcite . while the 7 indices are not independent , they are useful in providing an idea of the internal consistency of the results . the agreement between the different indices and different calibrations is excellent , generally within 100 k. the metallicity dependence of the photometric temperatures is small , but was nevertheless accounted for by iterations with the spectroscopic results . we obtained a mean system temperature of @xmath93 k and [ m / h ] @xmath94 , where the errors are dominated by the uncertain reddening . the primary and secondary temperatures are then @xmath6 k and @xmath7 k , respectively , though of course @xmath95 is known much more precisely . these temperatures correspond to a spectral type of approximately f9 ( e.g. , * ? ? ? lv her does not have an entry in the _ hipparcos _ catalog @xcite . the distance to the system is estimated here as @xmath96 pc , based on the bolometric luminosities , the apparent system magnitude of @xmath97 ( table [ tab : photometry ] ) , the adopted extinction of @xmath98 , and bolometric corrections from @xcite . separate distance estimates for each component agree nearly perfectly , showing the internal consistency of the results . also listed in table [ tab : dimensions ] are the rotational velocities ( projected along the line of sight ) expected if the stars were pseudo - synchronized ( see * ? ? ? * ) , as well as those that correspond to synchronization with the angular velocity at periastron , for comparison with the measured @xmath44 values . we discuss these below . lcc[b ] mass ( @xmath58 ) & 1.193 @xmath23 0.010 & 1.1698 @xmath23 0.0081 + radius ( @xmath65 ) & 1.358 @xmath23 0.012 & 1.313 @xmath23 0.011 + @xmath43 ( cgs ) & 4.2493 @xmath23 0.0082 & 4.2695 @xmath23 0.0081 + temperature ( k ) & 6060 @xmath23 150 & 6030 @xmath23 150 + @xmath99 ( @xmath100 ) & 0.349 @xmath23 0.044 & 0.311 @xmath23 0.044 + @xmath101 ( mag ) & 3.86 @xmath23 0.11 & 3.95 @xmath23 0.11 + @xmath102 & @xmath160.04 @xmath23 0.10 & @xmath160.04 @xmath23 0.10 + @xmath103 ( mag ) & 3.90 @xmath23 0.16 & 4.00 @xmath23 0.16 + @xmath104 ( @xmath65 ) & + distance ( pc ) & + @xmath105 $ ] & + @xmath44 ( kms@xmath41 ) & 13 @xmath23 1 & 13 @xmath23 1 + @xmath106 ( kms@xmath41 ) & 16.0 @xmath23 0.1 & 15.5 @xmath23 0.1 + @xmath107 ( kms@xmath41 ) & 19.6 @xmath23 0.2 & 19.0 @xmath23 0.2 + the high accuracy and precision of our measurements for lv her offers the opportunity to test various aspects of theoretical modeling . in this particular case , knowledge of the chemical composition , even though the precision is not as high , is especially interesting since it is rarely available for eclipsing binaries and it eliminates one of the free parameters in the comparison with theory . in figure [ fig : yalemrt]a we compare the two best determined parameters of lv her ( @xmath108 and @xmath109 ) against model isochrones from the yonsei - yale series by @xcite , computed for the metallicity of [ fe / h ] @xmath110 determined in [ sec : dimensions ] . these models include convective core overshooting ( see * ? ? ? * ) , which has an effect for stars of this mass and higher , and treats convection in the standard mixing length approximation , with a mixing length parameter @xmath111 , calibrated against the sun . an isochrone corresponding to 3.85 gyr ( solid line ) provides an excellent fit , and the high precision of the measurements suggests an age uncertainty of no more than 0.4 gyr for this fixed metallicity . however , the relatively large error in the measured [ fe / h ] adds significantly to the age uncertainty . this is shown by the shaded area around the 3.85 gyr isochrone corresponding to @xmath112 } = 0.21 $ ] dex , which contributes at least another 0.5 gyr . in the lower panel of the figure the temperature is shown as a function of mass , with the same models as in the top panel . once again the isochrone of 3.85 gyr matches the observations very well , but in this case the lower precision of @xmath42 provides a much weaker constraint on the age , and gives rise to a larger uncertainty than the error that comes from [ fe / h ] . the @xmath42@xmath43 diagram in figure [ fig : yalelogg ] implicitly compares all four measured quantities ( @xmath108 , @xmath109 , @xmath113 , and [ fe / h ] ) against evolutionary tracks from the yonsei - yale series . the models are computed for the observed metallicity of [ fe / h ] @xmath110 , and for the precise masses we measure . the shaded area indicates the uncertainty in the location of the tracks that comes from the mass errors . the same 3.85 gyr isochrone as before is shown with a dashed line . the stars are seen to be in the middle of the main - sequence band . a similarly good fit to the properties of lv her is provided by the stellar evolution models of @xcite , shown in figure [ fig : claretlogg ] , also for a metallicity constrained to the measured value . these models are similar to the ones used before in their treatment of convection ( @xmath114 ) , and also include overshooting ( @xmath115 ) , but differ in the details . in this case , we find that the best - fit age is 4.2 gyr , about 9% older than with the yonsei - yale models . the difference is of the same order as the age uncertainty discussed above . the high orbital eccentricity and our accurate measurement of the projected rotational velocities of both components constitute useful probes of tidal forces in the system and allow for an interesting comparison with the predictions from theory . for our initial comparisons we have used the radiative damping formalism of @xcite and @xcite , often invoked for this type of analysis , to calculate the critical times for lv her corresponding to synchronization of the axial rotations , alignment of the rotational and orbital axes , and circularization of the orbit . these are then compared with the evolutionary age , which is 4.2 gyr according to the models by @xcite used in this section . the procedure follows closely that described by @xcite , and consists of integrating the differential equations describing the changes in these quantities along the evolutionary track for each star . we find that rotational synchronization due to tidal forces is not expected to occur for the primary until the system reaches an age of @xmath116 gyr ( @xmath117 ) , and the secondary synchronization is predicted to happen shortly after ( @xmath118 gyr , @xmath119 ) . this appears to be in agreement with the fact that the measured @xmath44 values are slower than expected for pseudo - synchronization in the eccentric orbit ( table [ tab : dimensions ] ) , assuming the spin axes are parallel to the orbital axis ( but see below ) . the time of orbital circularization indicated by theory is 7.3 gyr ( @xmath120 ) , again consistent with the fact that the orbit is observed to be highly eccentric . these times are displayed graphically in figure [ fig : tidal1 ] , along with the evolution of the radius of each star according to the evolutionary tracks . a somewhat unexpected result is that the times of alignment predicted by theory are also significantly longer than the current age of the system : @xmath121 gyr and @xmath122 gyr ( or @xmath123 and @xmath124 , in reference to figure [ fig : tidal1 ] ) . at face value this indicates both spin axes may be inclined relative to the axis of the orbit , which is the opposite of what is almost universally assumed for binary systems . the long orbital period of lv her could provide a plausible explanation , in principle , but in any case the misalignment makes it somewhat problematic to interpret the apparent lack of ( pseudo-)synchronization inferred from the measured @xmath125 , since the projection factor can no longer be assumed to be known . the calculations above involve a number of approximations implicit in the equations we have used , as described , e.g. , by @xcite and @xcite . in particular , they are linearized around the equilibrium state , and are strictly valid only for relatively small eccentricities and near - synchronous rotation , and for small relative inclinations between the spin axes and the axis of the orbit . we have also ignored changes in the semimajor axis that occur concurrently with the evolution of other orbital elements . the condition on the eccentricity is most certainly not met for lv her ( @xmath1 ) , and it is unclear a priori to what extent this may affect the conclusions above . this , and the potentially interesting situation regarding the spin axes , have prompted us to explore the situation in greater detail . we have directly integrated the more general differential equations ( valid also for high eccentricities ) that describe the evolution of the semimajor axis ( @xmath126 ) , eccentricity ( @xmath127 ) , angular rotational rates ( @xmath128 , @xmath129 ) , and inclination of the spin axes relative to the orbital axis ( @xmath130 in our nomenclature ) as given by @xcite , using a fourth - order runge - kutta method . because they are coupled , these equations must be integrated simultaneously , and as before the stellar properties have been interpolated from the evolutionary tracks at each time step . the turbulent dissipation timescale for stellar phases with convective envelopes is taken to be @xmath131 , where @xmath108 , @xmath109 , and @xmath132 are the mass , radius , and luminosity of the star . for phases with radiative envelopes the timescales adopted follow closely those in eq.(17 ) and eq.(18 ) by @xcite . the initial conditions are of course not known , so we explored a range of values that give predictions matching the observed quantities at the current age of the system . this is shown in figure [ fig : tidal2]a , b for the eccentricity and orbital period . the initial values are @xmath133 and @xmath134 days at the starting age of @xmath135 . according to these calculations we expect the orbit to circularize at an age @xmath136 ( @xmath137 gyr ) , which is in fairly good agreement with our previous estimate despite the concerns expressed above . also shown in the figure is the evolution of the angular rotation rate @xmath138 for each star , which we normalize for convenience to the orbital rate @xmath139 ( figure [ fig : tidal2]c , d ; solid lines ) . for lv her we have no observational constraints on @xmath140 and @xmath141 , so the initial values are arbitrary and chosen to be the same . in principle they could be constrained by directly measuring the rotation periods of the stars due , e.g. , to rotational modulation by spots , although in practice this is difficult for a spatially unresolved binary . more importantly , these stars may be too hot to be affected significantly by spots . the ratio between the pseudo - synchronous velocity and the mean orbital motion is shown in figure [ fig : tidal2]e , and this curve is repeated in figure [ fig : tidal2]c , d ( dot - dashed line ) . as seen from the convergence of the dot - dashed and solid curves , the stars are predicted to reach pseudo - synchronization roughly at @xmath142 ( @xmath143 gyr ) , which is much younger than the evolutionary age . this is completely at odds with our previous estimate , showing the limitations of that approach for a system like lv her . we also have no direct constraint on the relative angle @xmath144 between the spin axes and the orbital axis . furthermore , the relation between @xmath144 and the orbital and rotational inclinations @xmath145 and @xmath146 , both measured relative to the plane of the sky , is given by @xmath147 which involves an unknown angle @xmath148 between the sky - projected angular momentum vectors of the orbit and the stellar spin . the spectroscopically measured projected rotational velocities of the stars , which we refer to more properly now as @xmath149 , do provide some indirect constraint on a combination of the theoretically predictable quantities , but this still involves the unknown angle @xmath148 . because @xmath150 is very nearly 90 for lv her ( see table [ tab : dimensions ] ) , we may make the approximation @xmath151 . we then have @xmath152 in this equation all quantities on the right - hand side are either known from stellar evolution calculations ( @xmath109 ) , or can be computed from the solution of the differential equations , except for @xmath153 , which depends on the observer s viewpoint . for the sake of illustration , we ignore this term in the following ( or consider @xmath148 to be small ) , so that @xmath154 . in figure [ fig : tidal2]f we show the evolution of @xmath144 for four different initial values ( 20 , 40 , 60 , 80 ) . curves for the primary and secondary are nearly indistinguishable , so we show only those for the primary . once again the conclusion is very different from the one we reached with the equations by zahn : alignment of the spin axes with the orbital axis is achieved much earlier than the age of the system , at @xmath155 ( or @xmath156 gyr ) . the corresponding @xmath149 curves for the primary and secondary , calculated using eq.([eq : tidal ] ) , are shown in the bottom panels and are compared with the measured rotations ( dashed line and shaded uncertainty region ) . the general trend is for the stars in lv her to be spun up by tidal forces , but the calculations disagree with the observations at the current age of the system , predicting the measured rotations should be 16 kms@xmath41 for the primary and 15.5 kms@xmath41 for the secondary ( see also table [ tab : dimensions ] ) . and @xmath141 have been chosen arbitrarily here , this does not affect the predicted rotational velocities at the current age because pseudo - synchronization occurs much earlier , independently of the starting values of @xmath140 and @xmath141 . ] these are nominally outside of the range allowed by the observational uncertainties . we consider it very unlikely that our measurements are off by as much as 3 kms@xmath41 for the primary and 2.5 kms@xmath41 for the secondary , since our formal uncertainty of @xmath231 kms@xmath41 is already conservative . the alternative would be a deficiency in the calculations , related to remaining approximations in the differential equations for this very challenging problem . a combination of both effects is also possible . an even more general treatment of the dynamical problem of tidal friction in binary systems with deformable components has been developed by @xcite , but the application is considerably more complex and is beyond the scope of the present work . we note , finally , that an empirical test of the prediction of alignment is available in eclipsing binaries by measuring the rossiter - mclaughlin effect ( e.g. , * ? ? ? * ; * ? ? ? this provides a direct measure of the angle @xmath148 between the sky projections of the spin axes and the axis of the orbit . as a result of our intensive spectroscopic and photometric monitoring , the absolute masses and radii of lv her are now determined to 0.9% or better , and are among the best available for any eclipsing binary . in addition we have established the effective temperatures , as well as the overall metallicity . the latter quantity has not been determined for many eclipsing binaries , and is important because it reduces the degrees of freedom in the comparison with theory . unfortunately the precision of [ m / h ] ( and @xmath42 to some extent ) is limited in this case by poor knowledge of interstellar extinction in the direction of the star . this could perhaps be remedied with additional absolute photometric observations . nevertheless , the combination of @xmath108 , @xmath109 , @xmath42 , and [ m / h ] provides unusually strong contraints on stellar evolution theory . comparison with current models yields an excellent fit to the observations , and indicates the stars are approximately half way through their main - sequence phase , at an age between 3.8 and 4.2 gyr , depending on the model . eclipsing binaries with periods as long as that of lv her are relatively rare and difficult to find , except perhaps in the course of automated variability surveys such as those designed to search for transiting planets , which often operate almost continuously for weeks or months at a time . the very high eccentricity of the system ( @xmath157 ) makes it the most extreme case among those with well - determined properties @xcite . although apsidal motion is expected , the apsidal period is predicted to be very long ( @xmath158 yr ) on account of the wide separation between the stars ; it may take decades for the effect to be detected and measured accurately . tidal theory is consistent with the observations in predicting that the orbit should not yet have been circularized by tidal forces . direct integration of the coupled differential equations for this problem according to @xcite indicates that the spin axes have already been forced into alignment with the orbital axis early on in the evolution of this system . the spins of both stars are also expected to be pseudo - synchronized with the orbital motion , although this is not quite in agreement with the measured rotations , and may indicate either large errors of measurement ( which we believe are unlikely ) or shortcomings in theory , or some combination of both . the case of lv her highlights the importance of bearing in mind the assumptions under which the equations for tidal evolution have been derived . because of their ease of use and simpler interpretation , the linearized equations by @xcite ( see also * ? ? ? * ; * ? ? ? * ) have occasionally been applied to binary systems that may violate those assumptions to some degree ( namely , the requirement of small eccentricities and spin - orbit inclinations , and near - synchronous rotations ) . exactly when and how the theory breaks down is difficult to predict . with its high eccentricity , lv her is one such case , as we have demonstrated here by comparing the critical times from the simpler formulation with the results from the more general equations given by @xcite . we are grateful to p. berlind , m. calkins , r. j. davis , e.horine , d. w. latham , j. peters , and j. zajac for their help in gathering the spectroscopic observations of lv her used in this work , and to r. j. davis for maintaining the echelle database at cfa . gt acknowledges partial support for this research through nsf grant ast-0708229 . chsl thanks bill neely who operates and maintains the nfo for our consortium , and who handles preliminary processing and storage of the images . he also wishes to thank summer 2009 arkansas reu student chris gong for her preliminary analysis of the ursa data . this research has made use of the simbad database and the vizier catalogue access tool , both operated at cds , strasbourg , france , of nasa s astrophysics data system abstract service , and of data products from the two micron all sky survey ( 2mass ) , which is a joint project of the university of massachusetts and the infrared processing and analysis center / california institute of technology , funded by nasa and the nsf .

by studying variable stars in the small magellanic cloud ( smc ) , henrietta leavitt discovered a linear relation between the pulsation period and magnitudes of @xmath2 cepheids , in the sense of that the brightest cepheids have longer periods @xcite . this correlation , commonly called period - luminosity ( pl ) relation , was lately renamed as the leavitt law ( ll ) in honor to its discoverer @xcite . this linear statistical correlation is a cornerstone in the measurement of the extragalactic distances using stellar standard candles . however , in order to obtain an accurate calibration of this relation , it is important to establish the effects of non - linearities , metallicity and companions on it , as it is mentioned by @xcite ; as well as other effects such as cepheids showing the hertzsprung progression and cepheids exhibiting changes in period . a brief literature review of these issues is presented below . * i ) the non - linearity effect : * the first works in this direction began twenty five years after the discovery by henrietta leavitt . in a study of the shape of the pl relation of nearby galaxies , kukarkin found that this relation has a break around @xmath3 d ( @xcite ; @xcite ) . three decades later , sandage & tammann ( 1968 , 1969 ) found evidence of curvature in the pl relation , and afterward , @xcite confirmed the break of the pl relation in @xmath3 d. @xcite confirmed again this result based on the statistical @xmath4-test . in the last years , the following studies have confirmed the non - linearity of the ll without proposing an explanation for it : the testimator method @xcite , the approaches of linear regression residuals and additive models @xcite , the multi - phase pl relations @xcite and the multiple least - squares regression @xcite . * ii ) the metallicity effect : * by using non - linear convective pulsating models , @xcite and @xcite studied synthetic multiband pl relations of populations of cepheids of different chemical compositions , uniformly distributed over the instability strip . the masses of these stars follow the law @xmath5 , and are distributed in the range of @xmath6 @xmath7 . for a wide range of log @xmath8 , those authors found clear evidence that the optical ll is better represented by a quadratic relationship than by a linear one , showing a dependence with the metallicity and the intrinsic width of the instability strip . for ir bands they showed that the ll is better fitted by a linear function showing a slightly dependence on the metallicity . @xcite and @xcite found strong evidence in favor of the universality of the optical pl relation , in the metallicity range from @xmath9 dex to @xmath10 dex . @xcite and @xcite found that the slopes of the pl relations in _ vi@xmath11 _ do not change significantly between the environments of the milky way and the lmc . * iii ) the companion effects : * physically bound companions to cepheids are difficult to detect at distances of tens of kpc . to produce a detectable effect on the light curve , the eclipses should be deep enough and a significant number of points associated with the eclipses should be observed . the first condition is reached with an adequate combination of radius , luminosity and effective temperature of the primary and secondary stars , jointly with the appropriate inclination and mass ratio of the system . these eclipsing systems exhibit light curves similar to those of cepheids , showing besides dispersed points below them , which are a signal of the eclipses ( @xcite , fig . 7 ) . when the cepheid light curve is subtracted , it emerges a typical eclipsing signal ( soszyski et al . 2008 , fig . @xcite claim to detect a few cepheids in eclipsing systems in the galactic bulge and the magellanic clouds , using a random forest supervised algorithm over the macho catalog of variable stars . however , due to these variables do not show a clear period of eclipses , it is difficult to affirm that they are binaries . it is very likely that these stars are blended with near neighbors . @xcite reported few magellanic clouds cepheids that could be blended , and also others that could be members of binary systems . detection of cepheid systems whose light curves exhibit eclipsing variations is too important : if the cepheid is a member of a physical binary system , it is possible to determine dynamical masses , radii and distance with a very high accuracy . until now , spectroscopic studies of double - lined eclipsing binary systems in the lmc have been made by the optical gravitational lensing experiment ( ogle ) in the systems : ogle - lmc - cep-@xmath12 @xcite , ogle - lmc - cep-@xmath13 @xcite , ogle - lmc - cep-@xmath14 @xcite , ogle - lmc - cep-@xmath15 @xcite and ogle - lmc@xmath16 ( @xcite , @xcite ) a system detected in the ogle gaia south ecliptic pole field . * iv ) the hertzsprung progression effect : * inspecting light curves of @xmath17 cepheids , @xcite found a relation between the position of a bump feature in the light curve and the pulsation period . later studies confirmed that this bump feature is present in the light and radial curves of fundamental - mode classical cepheids whose periods are in the range of @xmath18 d @xcite . this feature is present in the descending branch for pulsations in the range of @xmath19 d. for longer periods , this feature is observed in the ascending branch and disappears for periods longer than @xmath20 d ( @xcite , @xcite ) . the amplitude bump grows when the period increases , reaching its maximum value for periods near to the maximum light around @xmath21 d. for longer periods , the amplitude bump decreases until it vanishes @xcite . + there are two models that explain some of the properties of the observed bump feature . the first one , called the echo mechanism , proposes radial - pressure waves generated in the he ii ionization region @xcite . the traveling inward wave is reflected in the core and reaches the surface one period later , leading to the formation of the bump @xcite . however , this model has two points to solve : the contradiction with the acoustic - ray formalism ( @xcite ; aikawa & whitney 1984 , 1985 ) and the difficulty of predicting adequately the christy wave velocity near to the stellar surface ( @xcite , @xcite ) . the second model , known as the resonance mechanism , proposes that the feature bump arises from a resonance between the fundamental mode and the second overtone @xcite . by using numerical simulations @xcite studied the nonlinear saturation of the acoustic modes excited by the @xmath22mechanism . they found that this @xmath23:@xmath24 resonance causes that the bump appears in the ascending branch for @xmath25 . * v ) the changing period effect : * the long - term observations detect several thousand oscillation cycles for short period cepheids . this allows to determine , with very high accuracy , changing periods . for the long period cepheids , it is no possible to reach this accuracy , because the data have in the worse cases of sampling , tens of cycles . there are three scenarios explaining different characteristics of the period change rates . the first one proposes that the evolutionary changes on the stellar structure of cepheids crossing the instability strip are responsible of the period change rates ( @xcite ; @xcite ) . the second one proposes that the presence of magnetic fields is a possible explanation for the random changing periods exhibited by some cepheids . the last one suggests that the amplitude and phase variations of cepheids are caused by an analogous to the blazhko effect , exhibited by rr lyr stars @xcite . @xcite reported a few cepheids that could be exhibit this blazhko effect , being the number of these cepheids larger in the lmc than in the smc . + the period changes should produce a shift of the phase in the light curve , and as a consequence , a higher scatter in the light curve , detectable on years of observations . a study of @xmath26 lmc cepheids found period changes in @xmath27 of the fundamental mode , and in @xmath28 of the first overtone ( @xcite , table 3 ) . a visual inspection of the light curves of these ogle cepheids shows , for the most of them , a strong scatter of the points as a consequence of the shift phase . + since there are a few studies about magnetic fields on cool radial pulsating stars , it is not clear if the nature of these magnetic fields are fossil or are produced as a consequence of the stellar pulsation @xcite . based on a solar - like magneto - convective cycle , @xcite could explain the observed period change rates of two short period cepheids : polaris and v473 lyr . as the existence of local magnetic fields on the surface of late - type stars causes spots @xcite , @xcite proposed that convective hot spots can be a possible explanation to the random changes in the pulsation period detected in the single cepheid observed by the kepler mission , v1154 cyg . + the first detection of magnetic fields in variable stars was made on rr lyr @xcite . measurements of magnetic fields for cepheid stars began later on . in particular , for @xmath29 car and @xmath30 cyg there were reported values of @xmath31 g @xcite , and from @xmath32 g to @xmath33 g ( @xcite ; @xcite ) , respectively . the spectropolarimetric works over the bright bump cepheid @xmath34 aql ( @xmath35 , @xmath36 d ) reported controversial results . while @xcite and @xcite measured periodic variations of the longitudinal magnetic field with an amplitude of tens of gauss , @xcite found a non - significant detection of the longitudinal magnetic field at a level of @xmath3 g. + in order to establish whether or not the effects previously described are influential on the ll , we make a study using statistical techniques . we start - off , in the second section , with a brief description of the lmc ll and the ogle - ii and ogle - iv cepheid data . the third section is dedicated to review the statistical theory of the linear regression analysis , relevant for this work . the fourth section presents the results of applying statistical models in order to obtain optical ll of the lmc and smc . finally , our main conclusions are given in the fifth section . ogle - ii and ogle - iv observations of cepheid variables of the lmc and smc galaxies were collected with the @xmath37-m warsaw telescope , at las campanas observatory , chile ( @xcite , @xcite , @xcite ) . while ogle - ii fundamental mode cepheid catalogs contain 771 and 1319 stars for the lmc and smc , respectively , ogle - iv has a nearly complete collection ( 2429 and 2739 for the lmc and smc , respectively ) , covering practically the whole magellanic system area with the time baseline , a little more than five years . @xcite . the slope and zero point of the lmc ll in optical _ vi_-bands were computed by @xcite using ordinary least - squares ( ols ) regression , on ogle - ii fundamental - mode cepheids . points deviating by more than @xmath38 ( outliers ) were removed by applying the sigma - clipping algorithm @xcite . udalski realized that standard deviation of residuals of the lmc data was almost two times smaller than that of the smc . this greater dispersion is caused mainly by the spatial distribution of cepheids in the smc bar , whose thickness is placed along the line - of - sight with a typical depth of @xmath39 mag @xcite . these facts , and the manner in which the ll of the lmc is much better populated for periods longer than @xmath40 d , suggested udalski to adopt as the universal slope value the one obtained for the lmc , notwithstanding that the number of cepheids in the smc was around twice that of lmc . for ogle - iv lmc cepheids the slope and zero point of the ll were determined in analogous approach , but the smc slope was obtained independently , using only cepheids of this galaxy @xcite . + with the aim to establish a possible statistical cause of the non - linearity of the ll , we use the dereddened sample of ogle - ii . fundamental mode cepheids belonging to the lmc and smc galaxies reported by @xcite and @xcite , as well as the nearly complete collection of ogle - iv magellanic clouds cepheids reported by @xcite . in this section we summarize the topics of the statistical theory of linear regression analysis , most relevant to this paper . we start explaining what the outliers and influential points are . then , we present the theory of ordinary least square method and the conditions to apply it . next , we describe the statistics tools to identify influential points . then , we present the specification test , which verifies if all variables involved in a statistical problem are adequately included and represented by a proposed model . following , we present the structural break test , which looks for unstable parameters . finally , we present the robust @xmath0-regression that allows to do a linear regression of data with no - gaussian error distribution . the @xmath1-regression is applied in order to get a high breakdown point and fit the models in presence of influential points that are outliers at the same time . an extended statistical discussion of these topics can be found in @xcite . some applications in the astronomy context can be found in @xcite . the most common method to estimate parameters in a linear regression model is ols , based on the gauss - markov theorem . the model is ( in matrix notation ) : @xmath41 where @xmath42 is the dependent variable vector , @xmath43 is the independent variables matrix , @xmath44 is the parameters vector and @xmath45 is the errors vector . this theorem states that the ols estimators are the best linear unbiased estimators if the expected value of the errors vector is zero ( @xmath46 ) , the variance of the errors is constant ( @xmath47 ) and the errors are uncorrelated . it can be shown @xcite that ols estimators ( @xmath48 ) are obtained from : @xmath49 notice that : @xmath50 is the nule vector , @xmath51 the identity matrix and @xmath52 indicates @xmath53 matrix is transposed . once the model is fitted the adequacy must be checked . that is : i ) the error term @xmath45 has a constant variance @xmath54 . ii ) the errors are uncorrelated . these items are the gauss - markov theorem assumptions . iii ) the errors are normally distributed . this is very important for making hypothesis testing and estimating confidence intervals . iv ) all the observations have approximately the same weight . it refers to the fact that it is not desirable that model estimators depend more on few observations than on the majority of them . this could happen if some observations , called influential points , have a disproportional impact in the ols estimations . v ) there is no specification problems in the model and there is not a structural break , i.e. the parameters are stable . it refers to the correct functional form of the model . after fitting the model , a residual analysis is made to verify its adequacy . since the residuals are the difference between the original observed values and their fits , they also measure the variability of the dependent variable which is not explained by the model . therefore , any violations to the assumptions can be detected by analyzing model residuals . mainly the analysis on residuals is looking for evidence of : i ) errors come from a distribution with heavier tails than normal . large departures from normality in the error distribution mean that the _ f- _ and _ t- _ test are not longer valid as well as the confidence interval estimations . ii ) heteroskedasticity , meaning that the errors do not have constant variance . iii ) influential points . iv ) specification problems . v ) autocorrelated errors . in this study it is not necessary to check if errors are correlated because the data are not time dependent , i.e. they come from a cross - section in the statistical sense . + for checking the error distribution a normal probability quantil - quantil ( @xmath55 ) plot is used . if all the residuals lie along a line , it means that they come from a normal distribution . for more details see @xcite . this plot shows if the distribution has heavier or lighter tails or if it is skewed compared to the normal distribution . for checking the assumption that errors have constant variance ( homoskedasticity ) the white s test @xcite is used in the following equation : @xmath56 where @xmath57 is the residual from the linear regression defined as : @xmath58 this test does not make the assumption that errors come from a normal distribution . its null hypothesis is that @xmath59 for all @xmath60 . since the pl relation can be modeled as a simple linear regression , the test uses the auxiliary model given by @xmath61 where @xmath62 is the error term and @xmath63 is the number of stars . it can be shown that white s statistic @xmath64 is asymptotic distributed as @xmath65 with g-1 degrees of freedom , where @xmath63 is the number of observations and @xmath66 is the regression coefficient of determination of the equation ( [ eq5 ] ) . because the auxiliary model ( equation [ eq5 ] ) has three parameters ( @xmath67 ) the @xmath65 has @xmath23 degrees of freedom . + outliers are points that have an unusual behavior . let us imagine a scatter plot , outliers are data points that unusually are located out of the pattern or far from the data cloud . on the other hand , influential points are those data that have a disproportional impact on ols estimators ( slope and zero point ) . as a result , the model estimators depend more on them than on the majority of data . let us imagine the scatter plot again , a regression line is fitted with all points . then we delete a point and fit the regression line once more . if the regression line obtained after removing the point changes a lot ; the deleted observation is an influential point . what is a lot ? statistics helps us in this decision , as we shall describe later . now , let us imagine the scatter diagram with a far point from the cloud of data points , but located along the regression line . if we delete it , the regression line will not change . therefore , the point is an outlier but not an influential point , because it affects the @xmath68 coefficient and the ols estimator standard errors , but not the ols regression estimators . on the contrary , if the outlier affects the regression coefficient estimators it is also an influential point . finally , let us imagine that the point is not an outlier ; however , after deleting it , the new regression fitted - line is very different from the obtained with it . this means that it is an influential point . in general , influential points can or can not be outliers . + * detection of influential points : * in order to look for influential points cook s distance , _ dffits , dfbetas , covratio _ statistics are commonly used @xcite . the first of them is defined as follows : @xmath69 basically , it measures the square distance between the ols @xmath44 estimate , based on all observations and ols @xmath70 estimate obtained after deleting the @xmath71 observation , where @xmath72 in equation ( [ eq6 ] ) is the number of parameters in equation ( [ eq3 ] ) , i.e. the number of independent variables plus the zero point . moreover , @xmath73 is an estimate of the mean square error , defined as follows : @xmath74 observations with large cook s distance affect ols estimates of @xmath44 . to know what is a large cook s distance , @xmath75 is compared to the @xmath76 distribution @xcite . if the @xmath71 observation has a @xmath77 , it is considered an influential point . _ dffits _ measures how large is the influence of the @xmath71 observation on its own estimated @xmath78 . in other words , it shows how many standard deviations @xmath78 change due to delete the @xmath71 observation . it is defined as follows : @xmath79 @xmath80 and @xmath81 are the fitted values and the standard deviation respectively , obtained without the @xmath71 observation . the term @xmath82 is the @xmath71 diagonal element of the matrix @xmath83 that is defined as follows : @xmath84 the term @xmath85 is the amount of leverage that the @xmath71 observation has on the @xmath86 fitted value . a point is considerate influential when @xmath87 . for complementary explanations and references about _ dffits _ and the @xmath88 matrix see @xcite and @xcite . other statistics that detects influential points is _ dfbetas_. it indicates how much effect has the @xmath71 observation on each @xmath89 , measured in standard deviation units . in order to apply it , the coefficient estimators of @xmath89 obtained using all observations are compared with the coefficient estimators computed excluding the @xmath71 observation ( @xmath90 ) . as a result , a measure for each @xmath89 is obtained . the _ dfbetas _ statistics is defined as follows : @xmath91 where @xmath92 , so that @xmath93 is the @xmath86 diagonal element of the matrix @xmath94 and @xmath95 is the square root of the regression mean square error fitted without the @xmath71 observation . an observation is considered an influential point if @xmath96 . + the last statistics used to detect influential points is _ covratio _ , that measures how the covariance matrix is affected by the @xmath71 observation . it compares the generalized variance of the parameters estimators , obtained without the @xmath71 observation , with that obtained using all observations . + the variance - covariance matrix of parameters estimators with all observations is : @xmath97 the generalized variance is the determinant of the equation ( [ eq11 ] ) : @xmath98 given this , _ covratio _ is defined as follows : @xmath99 the @xmath71 observation is considered influential if its _ covratio _ lies out of the following interval @xmath100 , where @xmath72 is the number of parameters in the equation ( [ eq3 ] ) and @xmath63 is the total number of observations . the reported ll in this work are obtained by rejecting from the sample the influential points , and cepheids exhibiting the communalities , explained later . however , two questions arise that require detailed answers : why is it important to identify these influential points and give them an statistical treatment ? why is it necessary to apply robust techniques to make a linear regression instead of applying a well - known algorithm as sigma - clipping ? in the next paragraphs , we answer these questions , illustrating the associated statistical problem . outliers are observations with anomalous behavior . when they are caused by human errors or instrumental failures , they can be recognized and excluded from the analysis . moreover , they can be rejected from the analysis only if there are strong non - statistical reasons that support such decision . in this work it is possible to reject them based on astronomical reasons . however , the usual practice of deleting outliers without making a further analysis looking for reasons that could explain their behavior , could imply serious consequences in the estimators precision , because they are adjusted artificially @xcite . as a result , when there are not astronomical reasons that support excluding those points and/or there are influential points , robust estimation methods are needed . to validate that there is no evidence of specification errors in the model , we use tests to look if a non - linear combination of the fitted values are significant to explain the response variable . one of them is the ramsey s test given by the equations ( [ eq15 ] ) to ( [ eq17 ] ) @xcite . the other one , is a variant proposed by godfrey and orme that uses only the model given by equation ( [ eq15 ] ) @xcite . these auxiliary models and their respectively hypotheses used in this work are the following : @xmath101 the null and alternative hypothesis are @xmath102 and @xmath103 @xmath104 the null and alternative hypothesis are @xmath105 and @xmath106 @xmath107 the null and alternative hypothesis are @xmath108 and @xmath109 : at least one @xmath110 ; @xmath111 . + if any of these tests are significant because there is evidence to reject the null hypothesis @xmath112 , then there is enough evidence of misspecification . in fact , it means that a non - linear combination of the fitted values are significant to explain the variability in the dependent variable , i.e the functional form is incorrect . + to verify if there are structural breaks , dummy variables are used . before explaining what a dummy variable is , it is necessary to give a brief explanation about qualitative variables . qualitative or categorical variables are those that do not have a natural scale of measurement . in this study two categorical variables are used . the first variable indicates which data set the observation belongs to when data are divided in two sets , to make structural tests to the models . this variable has two categories : data set @xmath24 and data set @xmath23 . the second variable is the galaxy name where the cepheid is located , which also has two categories : lmc and smc . + dummy variables are used to indicate which category from a qualitative variable the observations belong to . each dummy variable takes two possible values : @xmath24 if the observation belongs to the category that the dummy variable is representing , and @xmath113 if not . if a qualitative variable has @xmath114 categories , it can be represented by @xmath114 dummy variables , one for each category . however , only @xmath115 dummy variables are used in the regression model . besides , it does not matter which @xmath115 dummy variables are used since @xmath116 matrix is the same , independently of which dummy variable is taken out to fit the model . in fact , dummy variables representing the categories of a qualitative variable are linked between each other because when one of them takes a value of @xmath24 for an observation , the others take @xmath113 for the same observation . therefore , when all the @xmath115 dummy variables in the model take a value of @xmath113 , they are representing the dummy variable that is no present in the model . for more details see @xcite . + the process of verifying the existence of structural breaks is made as follows : the data are divided in two data sets by dummy variables . then , a model that includes the dummy variables is estimated to verify if the estimated parameters associated with dummies are significant or not . if they are significant , the parameters are unstable , thus there are structural breaks . + to contrast hypothesis the @xmath117value is used . it means the lowest level of significance that leads to reject null hypothesis when it is true . therefore , if @xmath117value is greater than the significance level , there is not enough evidence to reject the null hypothesis . on the contrary , if the @xmath117value is less than or equal to the significance level there is enough evidence to reject the null hypothesis and accept the alternative one , which means the test is significant . sometimes ols assumptions are not accomplished ; for example , errors have a distribution with heavier tails than normal distribution and/or there are influential points . an alternative to fit a linear model in these kinds of scenarios is a robust regression , in which the residuals can be defined as follows : @xmath118 it is convenient to scale the residuals by using median absolute deviation : @xmath119 therefore the scaled residuals @xmath120 can be expressed as follows : @xmath121 if scaled residuals are not dependent between them , and all of them have the same distribution @xmath122 , the maximum likelihood @xmath44 estimators are those that maximize the likelihood function : @xmath123 in @xcite is shown that this condition is equivalent to minimize the equation ( [ eq22 ] ) . therefore , a class of robust estimator that minimizes this equation is called @xmath0-estimator and the regression based on it is called @xmath0-regression . @xmath0 comes from maximum - likelihood since function @xmath124 is related to equation ( [ eq21 ] ) for a proper choice of the error distribution . besides , @xmath124 weights the scaled residuals @xmath62 in the sum : @xmath125 there are more than one suggestion for @xmath124 in the literature , the function used in this work is the tukey s bi - square , with parameter equal to @xmath126 @xcite . for further information about this function and other choices see @xcite . to minimize the equation ( [ eq22 ] ) the first partial derivative of @xmath124 must be calculated : @xmath127 as a result , the following equation is obtained : @xmath128 @xmath72 is the number of parameters in the model . multiplying the equation ( [ eq24 ] ) by @xmath129 , it is obtained : @xmath130 where @xmath131 @xmath132 @xmath133 , otherwise is @xmath24 . + the determination of @xmath0-estimator is an iterative process , thus , the @xmath0-estimator for @xmath44 is : @xmath134 this iterative process stops when a convergence criteria is reached , i.e. when the estimators change less than a preselected amount in the @xmath135 iteration . notice that the equation ( [ eq27 ] ) is the weighted least square estimator where @xmath136 is a diagonal @xmath137 matrix of weights whose elements are @xmath138 . see @xcite and @xcite for a wider discussion . in the context of @xmath0-regression it is important to check that there are no bad influential points , i.e. observations that are outliers in the * x * space and influential points at the same time . in this scenario the @xmath1-estimator @xcite should be used because it has a high breakdown point ( bdp ) . in a finite - sample the bdp is the smallest fraction of contaminated data causing a divergence of the estimator from the value that it would take if data were not contaminated @xcite . + the @xmath1-estimator combines the asymptotic efficiency of @xmath0-estimator with the high bdp estimators such as @xmath139estimators and least treammed squares . @xmath1-regression is obtained in three steps . first , a high bdp parameter estimators model is calculated to compute the residuals of the model . second , based on residual computed in the previous step , a @xmath0-estimate of scale with high bdp is calculated . finally , the model parameters are estimated using @xmath0-estimators and the scale estimation computed in the previous step . + neither normal errors nor homoskedasticity are assumptions in @xmath0-regression and @xmath1-regression . however , these regressions assume that errors are uncorrelated , which is not an issue when data come from a cross - section , meaning that data are not time dependent . + robust regressions are based on asymptotic results . this means they could lead to wrong results when they are used with small data sets . @xcite say that small to moderate sample size could be less than @xmath140 points . in our case , the smallest data set has about @xmath141 observations . hence , we have enough data to apply these techniques without problems . in particular , ogle - iv has around two and three times of observations for the smc and the lmc than ogle - ii , respectively . the statistical theory explained previously are applied for each one of the lmc and smc galaxies . the following models are fitted by ols : @xmath142 @xmath143 @xmath144 @xmath145 is @xmath146 and @xmath8 is the pulsation period of a cepheid . in general , these models have violations to the ols assumptions , as it will be shown below . influential points are data that truly affect the linear regression results , although they can or can not be outliers . it is very important that the studies of the ll make an appropriate treatment of those points . deleting a subset of influential points does not prevent the appearance of new influential points , as will be shown below . + the sigma - clipping algorithm is an iterative procedure commonly used in astronomy in order to reject the outliers in ols regression . however , to apply the ols regression its assumptions must be accomplished . in order to investigate if the ogle - ii data do not fulfill these conditions , we make an experiment in v - band using the whole sample of @xmath147 ogle - ii lmc fundamental mode cepheids . we apply the sigma - clipping algorithm using five iterations and an optimum threshold of @xmath148 . these values are the same used in the studies of the ll made by @xcite and @xcite . simultaneously , we apply the white s test to check if the assumption of homoskedasticity is fulfilled . + before the first sigma - clipping iteration , the white s test finds enough evidence to reject the null hypothesis of homoskedasticity with a _ _ p-__value of @xmath149 at @xmath150 of significance . a number of @xmath151 influential points are detected using some of the cook s distance , _ dffits , dfbetas , _ and _ covratio_. on the other hand , by applying the criteria of the sigma - clipping algorithm , @xmath152 points are rejected in the first iteration , not all of them identified as influential points . during second iteration , the white s test finds again enough evidence to reject the null hypothesis of homoskedasticity with a _ _ p-__value of @xmath153 at @xmath150 of significance and @xmath154 influential points are identified . by using sigma - clipping algorithm @xmath155 points are rejected . in the following three iterations influential points are detected again . taking into account these facts , we notice that the main problems of applying the sigma - clipping algorithm in the context of the ll are : \i ) ols estimators obtained by using this procedure are not adequate since their precision are adjusted artificially @xcite . \ii ) in the two initial iterations , the models have no a constant error variance , i.e. , they show heteroskedasticity . in the remaining iterations , the models show homoskedasticity , as a result of deleting points using the sigma - clipping criterium . \iii ) the residuals in each iteration have distributions with heavier tails than the normal distribution . \iv ) influential points are detected on each iteration . _ summarizing , rejecting outliers using the sigma - clipping algorithm is not adequate in this case , since there are violations of ols assumptions . without strong reasons to support the rejection of points , this method is invalid in the ll linear regression context . _ as we mentioned before , models ( [ eq28 ] ) to ( [ eq30 ] ) have violations to the ols assumptions : for cepheids belonging to the lmc galaxy , white s test ( explained in 3.1 ) finds enough evidence of heteroskedasticity with @xmath117values of @xmath149 and @xmath156 for models ( [ eq28 ] ) and ( [ eq29 ] ) at @xmath157 of significance . on the contrary , for model ( [ eq30 ] ) , white s test does not find enough evidence to reject its null hypothesis because of its @xmath117value of @xmath158 , that is greater than the same fixed level of significance . for cepheids belonging to the smc galaxy , white s test does not find enough evidence to reject the null hypothesis of homoskedasticity with @xmath117values of @xmath159 , @xmath160 and @xmath161 for models ( [ eq28 ] ) to ( [ eq30 ] ) respectively , at @xmath150 of significance . + another problem is the influential points that are found in all models in both galaxies . as we have mentioned above , this fact makes necessary to search for communalities in the influential points to understand the causes of this behavior . to do that , the @xmath162-band light curves of all cepheids in the lmc and smc are visually inspected many times . as a result , we identify two main morphological characteristics in them : the dispersion and the shape . using these communalities we perform a bi - dimensional classification of the cepheids in some of the following groups : light curves with small ( 1 ) , moderate ( 2 ) , or large ( 3 ) dispersion . and light curves whose morphological shapes are sinusoidal - like ( a ) , sawtooth - like ( b ) or exhibit bumps , i.e. showing the hertzsprung progression ( c ) . dispersion of light curves of cepheids classified as 2 and 3 seem to be caused in several cases by blending , since a visual inspection of the finding charts of these cepheids , shows neighbors near to the stars . representative examples of this classification are shown in figure [ fig1 ] . the bi - dimensional classification of all ogle - ii lmc and smc fundamental mode cepheids is given in table [ tab1 ] . the last column reports our bi - dimensional classification : morphological shape is given by a , b or c , and the grade of dispersion is measured by the discrete scale : @xmath24 , @xmath23 or @xmath163 , the last one being the corresponding to light curves with larger dispersion . an appended i letter indicates that this star is detected as influential point by at least one of the dffits , dfbetas or covratio statistics , computed from @xmath164 data . in table [ tab0 ] , the number of cepheids belonging to each sub - class is reported . .**morphological classification of ogle - ii cepheids * * [ cols="^,^,^,^,^,^,^,^,^",options="header " , ] when the influential points are correlated with our bi - dimensional classification , an interesting fact arises : around one third of the lmc cepheids and one fifth of the smc cepheids , classified with a grade of dispersion moderate and large are influential points . taking this fact into account , and in order to avoid violations of ols assumptions , we propose the following hypothesis : if cepheids with light curves of moderate and large dispersion , exhibiting characteristics suggesting blending , bumps , eclipses and period changes , are excluded from the ols linear regression , and the remaining data do not present violations of ols assumptions , it is possible to affirm that a cause of non - linearities of the ll is due to the inclusion of a fraction of these stars , that despite to be cepheids are influential points . + for testing this hypothesis , we fit the models given by equations ( [ eq28 ] ) to ( [ eq30 ] ) by ols using only cepheids with small dispersion . the results confirm the hypothesis , as it will be described below . for this reason , we consider appropriate to summarize here the communalities of these influential cepheids : \i ) the scatter exhibited by light curves of some cepheids classified as a2 or a3 can be associated with phase shifts , suggesting changes in the pulsational period . this effect produces a dispersion of the mean magnitudes , which moves the locus cepheid in the pl relation . \ii ) the b2 light curves exhibit a behavior that could be explained by period changes or blending effects . the dispersion of these light curves do not allow to directly establish which of these effects could be present in these types of cepheids . in this work it is not clear which of them is more likely to be the best explanation . however , this point is out of the scope of this paper . this interesting fact could be investigated in more detail in future works . a small fraction of the light curves classified as b3 have characteristics suggesting eclipses or blending . in the first case , the main effect of the companion on the pl relation is to move the cepheid locus . due to that , the total mean brightness out - of - eclipse includes the companion contribution , adding scatter in some cases up to @xmath165 mag . , as it was reported by @xcite and @xcite . the second case is more probable to occur , since a visual inspection of the finding charts of the b3 cepheids , shows that most of them are blended . \iii ) light curves classified as c1 or c2 are bump cepheids , i.e. they exhibit the hertzsprung progression . the amplitude bump affects the cepheid mean magnitude . this effect is stronger for periods near to the maximum light , shifting the cepheid position in the pl relation . in order to use the ols regression , the white s test is applied . this test does not find enough evidence to reject its null hypothesis of homoskedasticity , with a @xmath150 of significance , since the @xmath117values associated with each model test are @xmath166 , @xmath167 and @xmath168 for the lmc , and @xmath169 , @xmath170 and @xmath171 for the smc , respectively . however , two issues are noted : the first one is again the presence of influential points . the second fact is that the residual distributions for @xmath164 lmc data ( and only for the @xmath11 smc data ) have heavier tails with respect to the normal distribution . + since there is neither an astronomical nor statistical reason to delete those influential points or a fraction of them , a new approach to estimate the parameters given in the models ( [ eq28 ] ) to ( [ eq30 ] ) is needed instead of ols . besides , if these influential points were deleted , new ones may appear . + we fit the models by using the @xmath0-regression . the @xmath172 function used is the bi - square . since no evidence of bad influential points is found , we check each model beginning with the slope test , which is a robust version equivalent to the @xmath173-test in ols . the null and alternative hypotheses are : @xmath174 and @xmath175 . + for each model fitted in both galaxies , the robust slope test has a @xmath8-value less than @xmath176 , so at @xmath150 of significance they are significant . the lmc @xmath177-band fitted model ( [ eq28 ] ) has a @xmath66 value ( the coefficient of determination ) of @xmath178 , meaning that the model explains about the @xmath179 of the variability of @xmath177 . for the lmc @xmath162-band model ( [ eq29 ] ) and the lmc @xmath11 model ( [ eq30 ] ) , they explain about @xmath180 and @xmath181 respectively of the observed variability . we also use the robust ramsey s test @xcite , given by the following equations : @xmath182 the null ( @xmath112 ) and alternative ( @xmath109 ) hypotheses are @xmath102 and @xmath103 . @xmath183 the null and alternative hypotheses are @xmath184 and @xmath185 at least one @xmath186 . @xmath187 the null and alternative hypotheses are @xmath108 and @xmath185 at least one @xmath188 . in equations ( [ eq31 ] ) to ( [ eq33 ] ) , @xmath189 is any of the magnitude vectors @xmath177 , @xmath162 or the @xmath11 index . moreover , values of @xmath190 and @xmath191 are different for each photometric band . this test shows that there is not enough evidence to reject @xmath112 due to its @xmath117values at @xmath150 of significance . this implies that there is no evidence of specification problems in the lmc models . on the contrary , for the smc models there is enough evidence to reject @xmath112 due to @xmath117values obtained by the test at the same level of significance . in particular , when equation ( [ eq31 ] ) is used , the lmc ( smc ) @xmath117values are @xmath192 ( @xmath193 ) for @xmath177-band , @xmath194 ( @xmath195 ) for @xmath162-band and @xmath196 ( @xmath197 ) for the @xmath11 index . by using equation ( [ eq32 ] ) , the lmc ( smc ) @xmath117values are @xmath198 ( @xmath199 ) for @xmath177-band , @xmath200 ( @xmath195 ) for @xmath162-band and @xmath201 ( @xmath202 ) for the @xmath11 index . finally , when equation ( [ eq33 ] ) is used , the lmc ( smc ) @xmath117values are @xmath203 ( @xmath197 ) for @xmath177-band , @xmath204 ( @xmath205 ) for @xmath162-band and @xmath206 ( @xmath207 ) for the @xmath11 index . in order to verify if there are structural breaks ( explained in 3.4 ) in model equations ( [ eq28 ] ) to ( [ eq30 ] ) , we fit the following model : @xmath208 each data set is split in two parts using a dummy variable represented by @xmath209 , where @xmath210 the lmc ( smc ) data are split around @xmath211 @xmath212 d , in order to secure that each data set has enough observations to avoid problems due to asymptotic results in robust regressions . as a result , each data set has @xmath213 ( @xmath214 ) and @xmath215 ( @xmath216 ) points . the hypotheses to contrast are : @xmath217 and @xmath185 some @xmath218 . the @xmath219 and @xmath220 parameters are associated with @xmath209 in order to detect structural breaks in the zero point @xmath221 and/or the slope @xmath222 of the ll , if these parameters are not significant there is not enough evidence of structural breaks . on the contrary , if both or some of them are significant there is enough evidence of structural breaks . + the obtained lmc @xmath117values are @xmath223 for @xmath224band , @xmath225 for @xmath226band , and @xmath227 for the @xmath11 index . whereby , no evidence of structural breaks is found , hence the parameters obtained using @xmath0-estimators are stable at @xmath150 of significance . we use the @xmath0-regression to estimate the lmc ll throughout the following equations : @xmath228 @xmath229 @xmath230 models explain about @xmath179 , @xmath180 and @xmath181 of the variability in @xmath177-band , @xmath162-band and @xmath11 , respectively . the confidence interval at @xmath231 for each @xmath0-estimator for the lmc zero point @xmath232 and slope @xmath221 are reported in table [ tab2 ] . it is worth to say that the estimators for zero point and slope obtained by ols are included between the confidence intervals obtained by @xmath0-regression . + _ to summarize , the robust tests have shown that there is no evidence to have doubts about lmc models adequacy ; as a result they are acceptable . _ @ccccc + & @xmath233 & @xmath234 & zp & @xmath34 + @xmath235 & 17.005 @xmath236 0.064 & -2.683 @xmath236 0.108 & 17.066 @xmath236 0.021 & -2.775 @xmath236 0.031 + @xmath237 & 16.549 @xmath236 0.044 & -2.930 @xmath236 0.074 & 16.593 @xmath236 0.014 & -2.977 @xmath236 0.021 + @xmath11 & 15.839 @xmath236 0.029 & -3.315 @xmath236 0.049 & 15.868 @xmath236 0.008 & -3.300 @xmath236 0.011 + @ccccc + & @xmath233 & @xmath234 & @xmath238 & zp + @xmath235 & 17.444 @xmath236 0.097 & 0.135 @xmath236 0.069 & -2.551 @xmath236 0.147 & 17.635 @xmath236 0.031 + @xmath237 & 17.019 @xmath236 0.074 & 0.107 @xmath236 0.053 & -2.869 @xmath236 0.110 & 17.149 @xmath236 0.025 + @xmath11 & 16.353 @xmath236 0.052 & 0.063 @xmath236 0.037 & -3.342 @xmath236 0.079 & 16.381 @xmath236 0.016 + @cccc + & @xmath233 & @xmath234 & @xmath238 + @xmath235 & 17.037 @xmath236 0.047 & 0.554 @xmath236 0.030 & -2.732 @xmath236 0.074 + @xmath237 & 16.590 @xmath236 0.034 & 0.540 @xmath236 0.022 & -2.995 @xmath236 0.054 + for the smc model the @xmath8-value obtained by the test of structural breaks is less than @xmath239 for all bands , implying enough evidence to reject the null hypothesis and accept @xmath109 at @xmath150 of significance ; i.e. there are structural breaks . because of the problems already detected in model equations ( [ eq28 ] ) to ( [ eq30 ] ) for smc , further models analysis is needed using equation ( [ eq34 ] ) . therefore the individual significance of the parameters is analyzed . to do that , the following hypotheses are contrasted : @xmath240 and @xmath241 . the individual robust test for @xmath219 shows it is significant at @xmath150 of significance , due to @xmath117values of @xmath242 in @xmath224band , @xmath199 in @xmath226band and @xmath243 in the @xmath11 index . on the contrary , the same test for @xmath220 is not significant at @xmath150 of significance since associated @xmath117values are @xmath244 in @xmath224band , @xmath245 in @xmath226band and @xmath246 in the @xmath11 index . there is not enough evidence to reject @xmath112 , i.e. @xmath247 . it means that the model has different zero point for each data set but the same slope . taking into account the specification problems and structural breaks detected in models ( [ eq28 ] ) to ( [ eq30 ] ) , we fit the following model for smc based on the previous individual significance analysis using ( [ eq34 ] ) : @xmath248 using ogle - iii cepheids , @xcite determined the inclination of the smc galaxy s disk in @xmath249 . the geometrical distribution of the cepheids across this disk could explain the behavior of the ll in the smc , observed in equation ( [ eq50a ] ) : linear regressions with the same slope and different zero points for each data set . + model ( [ eq50a ] ) is fitted by @xmath0-regression and no evidence of bad influential points is found . robust ramsey s test is applied again to check if there are specification problems in model ( [ eq50a ] ) . the equations and hypotheses of the test are : @xmath250 the null and alternative hypothesis are : @xmath102 and @xmath103 . @xmath251 the null and alternative hypothesis are : @xmath184 and @xmath252 . @xmath253 the null and alternative hypothesis are : @xmath108 and @xmath254 . the test does not find enough evidence of specification problems because the @xmath117values obtained at @xmath255 of significance . in particular , for model ( [ eq51a ] ) the smc @xmath117values are @xmath256 for @xmath177-band , @xmath257 for @xmath162-band and @xmath258 for the @xmath11 index . for model ( [ eq52a ] ) , smc @xmath117values are @xmath259 for @xmath224band , @xmath260 for @xmath226band and @xmath261 for the @xmath11 index . for equation ( [ eq53a ] ) , smc @xmath117values are @xmath262 for @xmath177-band , @xmath263 for @xmath162-band and @xmath264 for the @xmath11 index . @xmath0-regression estimations for the smc ll are shown in the following equations : @xmath265 @xmath266 @xmath267 these models explain about @xmath180 , @xmath268 and @xmath181 of the variability in @xmath177 , @xmath162 and the @xmath11 index , respectively . the confidence intervals at @xmath231 for each @xmath0-estimator in the equations ( [ eq54a ] ) to ( [ eq56a ] ) are reported in table [ tab3 ] . besides , it is interesting that the confidence intervals for @xmath0-estimators include the values obtained by ols , although there are slightly differences . + _ to summarize , robust tests show specification problems and structural break points in the smc models ( [ eq28 ] ) to ( [ eq30 ] ) ; therefore , analysis leads to model ( [ eq50a ] ) as a better choice . _ the universality hypothesis of the ll implies that the slope of the linear regression observed in the same filter in different galaxies has the same value , and shows a negligible dependence with the metallicity ( @xcite , and references therein . ) in order to test the universality of the ll , a linear model to fit simultaneously the data from the lmc and smc is proposed . this model is given by : @xmath269 that is fitted in @xmath177 , @xmath162 and the @xmath11 index after deleting the points previously discussed . the dummy variable @xmath209 indicates if the observation is from the lmc or the smc : @xmath270 if a cepheid is from the smc and @xmath271 otherwise . a number of @xmath272 cepheids from the smc and 487 from the lmc are used to fit this model . + the model ( [ eq73 ] ) is fitted by ols . white s test finds enough evidence to reject the null hypothesis of homoskedasticity , at @xmath150 of significance , since the @xmath8-value is @xmath273 in @xmath177 , less than @xmath239 in @xmath162 and less than @xmath176 in the @xmath11 index . as happened before , ols regression is not adequate due to heteroskedasticity problems and the presence of influential points . besides , the residuals distribution has a heavier tail than the normal distribution . therefore , model ( [ eq73 ] ) is fitted by @xmath0-regression using bi - square function . no evidence of bad influential points is found ; thus , the process to check the adequacy of the model continues . + the @xmath117values of the slope test are less than @xmath176 in @xmath177 , @xmath162 and for the @xmath11 index , so the models are significant at @xmath150 of significance . the hypotheses contrasted are : @xmath274 and @xmath275 . to test if both galaxies have the same zero point and slope the following hypotheses are contrasted : @xmath276 and @xmath185 some @xmath218 . + the @xmath117values obtained by the robust test are less than @xmath176 in @xmath177 , @xmath162 and for the @xmath11 index ; therefore , there is enough evidence to reject the null hypothesis at @xmath150 of significance . hence , the individual significance of each parameter must be analyzed . to do that , the following hypotheses are contrasted again : @xmath277 and @xmath241 . + the individual significance robust test for @xmath219 , shows it is significant at @xmath150 of significance , since its associated @xmath117values are less than @xmath176 in @xmath278 and the @xmath11 index . the same test for @xmath220 is not significant at @xmath150 of significance because its associated @xmath117values are @xmath279 and @xmath280 in @xmath177 and @xmath162 , respectively . this means that the model has different zero points , as we expect due to the galaxies are at different distances , but it has the same slope . as a result , the final equations fitted by the @xmath0-regression in @xmath177 and @xmath162 bands , after deleting @xmath220 are : @xmath281 @xmath282 confidence radius are reported in table [ tab4 ] . models ( [ eq76 ] ) and ( [ eq77a ] ) make an adjustment to the zero point when cepheids are from the smc but have the same slope for both galaxies at @xmath150 of significance . moreover , they explain about the @xmath283 and @xmath284 of the variability in @xmath177 and @xmath162 , respectively . the individual significance robust test for @xmath220 in the @xmath11 index is significant at @xmath150 of significance because its associated @xmath8-value is @xmath285 . it implies that each galaxy has its own regression line with different zero point and slope . it explains about @xmath181 of the @xmath11 variability . following equation gives the model fitted for the @xmath11 index : + @xmath286 the confidence intervals at @xmath231 for estimated parameters in equation ( [ eq78a ] ) are : @xmath287 for zero point , @xmath288 for zero point adjustment when cepheids are from smc , @xmath289 for slope , and @xmath290 for slope adjustment when cepheids are from smc . + _ as a general result of this work , robust tests have shown that the ll of the lmc and smc are universal in @xmath177 and @xmath162 bands . there are enough evidence of two parallel egression lines , with the same slope at @xmath150 of significance but different zero points , as it is shown by models ( [ eq76 ] ) and ( [ eq77a ] ) . however , for the @xmath11 index , it has been shown that there are not a universal ll : the lmc and smc have two completely different regression lines , as it is shown by equation ( [ eq78a ] ) . _ based on the experience obtained by fitting the models ( [ eq28 ] ) and ( [ eq29 ] ) , they are fitted for all @xmath278 mean magnitudes of fundamental mode cepheids observed by the ogle - ii project . we also fit these models using the ogle - iv fundamental mode cepheids , reported by @xcite . it is worthy to say that the published ogle - iv mean magnitudes are not corrected for extinction . all following models are fitted with the @xmath1-regression because of its high bdp of @xmath291 . the least treammed squares and tukey function are used . + as no evidence is found of skewness in the models for both galaxies , we proceed with the analysis . beginning with the lmc , model slope tests are significant , as it can be seen in the @xmath117values obtained : less than @xmath176 in @xmath224 and @xmath226band ( ogle - ii , ogle - iv ) . then , the test proposed by @xcite is performed obtaining @xmath117values of @xmath292 ( ogle - ii ) and @xmath293 ( ogle - iv ) in @xmath224band and @xmath294 ( ogle - ii ) and @xmath295 ( ogle - iv ) in @xmath226band ; therefore , there is evidence of specification problems in ogle - iv @xmath226band model at @xmath255 of significance . this could be due to blending problems and also to the lack of extinction correction in ogle - iv data . to perform stability tests , the data are split in two data sets around @xmath296 d ( ogle - ii ) and @xmath297 d ( ogle - iv ) , to have about half of observations in each set , by dummy variables . no evidence of structural breaks is found , because the @xmath117values obtained , by testing @xmath298 vs the hypothesis that at least one of them is not @xmath113 in equation ( [ eq34 ] ) , are @xmath299 ( ogle - ii ) and @xmath300 ( ogle - iv ) in @xmath224band , and @xmath301 ( ogle - ii ) and @xmath302 ( ogle - iv ) in @xmath226band . models explain about @xmath303 and @xmath304 of variability in @xmath224 and @xmath226bands respectively ( ogle - ii ) ; and @xmath305 and @xmath306 of variability in @xmath224 and @xmath226bands respectively ( ogle - iv ) . the estimated parameters of the models and their confidence intervals are reported in table [ tab6 ] . results for @xmath226band are shown only for information purposes because of their specification problems . the ogle - iv lmc data were also split in around @xmath307 d by equation ( [ eq34 ] ) . no evidence of structural break is found in the slope , because the @xmath8-value obtained is @xmath308 . however , the zero point adjustment is significative . as a result , there is evidence of two parallel regression lines . + the smc slope tests for models ( [ eq28 ] ) and ( [ eq29 ] ) are significant because the @xmath117values obtained are less than @xmath176 for @xmath224 and @xmath226bands ( ogle - ii , ogle - iv ) . however , these models show problems . godfrey and orme test finds enough evidence of specification problems because of @xmath309values obtained are less than @xmath176 for @xmath224 and @xmath226bands ( ogle - ii , ogle - iv ) . to perform stability tests , data are split around @xmath310 d ( ogle - ii ) and @xmath311 d ( ogle - iv ) to have about half of observations in each data set . enough evidence of structural breaks is found : @xmath117values are less than @xmath176 in both bands ( ogle - ii ) , for testing @xmath298 vs the hypothesis that at least one of them is not @xmath113 in equation ( [ eq34 ] ) . besides , when individual significance tests are performed for @xmath219 and @xmath220 , @xmath117values obtained are less than @xmath176 for @xmath219 in both bands ; @xmath312 and @xmath313 for @xmath220 in @xmath224band and @xmath226band respectively . as a result , in both bands there are two regression lines at @xmath150 of significance ( ogle - ii ) . + unlike smc ogle - ii models , there is not enough evidence to support that ogle - iv slope model parameters are unstable with @xmath117values @xmath314 and @xmath315 in @xmath224 and @xmath226bands respectively , while zero point adjustment , @xmath219 ( equation ( [ eq34 ] ) ) , is still significant with @xmath117value less than @xmath176 in both bands . as a result , stability tests show that there are two parallel regression lines in both bands . based on the individual significance analysis using model ( [ eq34 ] ) for @xmath224 and @xmath226bands , new models are fitted . however , godfrey and orme test shows that specification problems persist . nevertheless , only for information purposes , we report the slope estimated values at @xmath231 confidence intervals for the ogle - iv data:@xmath316 for @xmath224band and @xmath317 for @xmath226band . + _ summarizing , the lmc models fitted with ogle - ii and ogle - iv data are adequate except for the @xmath226band model ( ogle - iv ) . the ogle - ii models adequacy could be explained because only @xmath318 of cepheids in the lmc has intermediate and large dispersion . therefore , as cepheids with small dispersion are the largest fraction , excluding intermediate and large dispersion cepheids has no significant impact on the estimated slope value , because confidence interval for slope ( @xmath234 ) , showed in table [ tab6 ] , includes the slope estimation ( @xmath234 ) showed in table [ tab2 ] . moreover , lmc fitted models with all ogle - iv data have confidence intervals for slope that contain estimation obtained by using the models fitted with all ogle - ii data ( see @xmath234 values in table [ tab6 ] ) _ . + _ on the other hand , there is enough evidence showing that smc models are not adequate . unlike lmc , the largest fraction of ogle - ii smc data @xmath319 are intermediate and large dispersion cepheids ; as a result , when models are fitted with all data , those cepheids are majority in the sample . the presence of a large fraction of intermediate and large dispersion cepheids in the smc galaxy can be explained by the geometry of this galaxy , as it was discussed in the section @xmath320 . it is interesting that smc models improve when they are fitted with ogle - iv data because , despite of specification problems , there is not enough evidence of unstability of the slope parameter in both bands . _ our main conclusions can be summarized as follows : by applying the traditional ols and robust statistical linear regression models to optical ogle - ii and ogle - iv data , it is possible to affirm that the problems of break and non - linearity of the ll disappear for the lmc when @xmath0- or @xmath1-regressions are used . in that case , the ll is obtained without slope breaks , as it is shown in equations ( [ eq76 ] ) and ( [ eq77a ] ) , and it is not needed to exclude or reject points from the data sample . the models in this case are adequate and do not present specification problems or structural breaks , except in the ogle - iv @xmath162-band , probably due to the lack of extinction correction among others causes . these facts allow to conclude that the non - linearity of the lmc ll is a consequence of using a non - adequate statistical method . + for the smc , despite to use the robust regressions , specification , structural breaks and adequacy problems are found . this could be caused in part by blending problems due to the geometry of this galaxy , that could be generating a large fraction of intermediate and large dispersion cepheids , as it was discussed in section @xmath320 . this astronomical reason seems to imply that cepheid variables in galaxies with elongated structures like this of the smc are not appropriate to fit a reliable ll . in particular , for the smc , we found that excluding the intermediate and large dispersion cepheids , i.e. stars that exhibit the communalities of influential cepheids , leads to avoid the adequacy and specification problems . this could be indicating that this kind of cepheids really has influence in the fitting of the ll for galaxies with similar geometry to this of the smc . however , from a statistical point of view , there is no reason to exclude these stars from the analysis . + the ols method has violations to its assumptions for the lmc and smc galaxies , with ogle - ii and ogle - iv data , as it was discussed at the beginning of section @xmath321 . in particular , two important problems are worthy of mentioning here . the first one is the fact that rejecting outliers using the sigma - clipping algorithm is not adequate because the precision of the parameters is adjusted artificially , besides there are violations of ols assumptions . without strong non - statistical reasons to support the rejection of points , this method is invalid in the ll linear regression context . the second problem is the presence of influential points , that implies the necessity to find communalities in them . in this work we found that a fraction of cepheids with a grade of moderate and large dispersion are influential points . the communalities exhibited by them suggest in their light curves bumps , i.e. cepheids showing the hertzsprung progression , or eclipsing variations , or evidence of period changes . the exclusion of these influential cepheids from the sample using ols regression was made to solve adequacy problems ; however , new influential points appear that imply again violations to the ols assumptions . + although the obtained @xmath0-estimators ( slope and zero point ) and their respective confidence intervals include the ols estimators reported by @xcite , the relevance of using the @xmath0-estimators is that they allow to make a reliable statistical inference , since none of their assumptions are violated , unlike the ols estimations . + when cepheids from the lmc and smc are combined to fit a single model , in each one of the @xmath224 and @xmath226bands , the linear relations remain to be valid , after deleting the cepheids already discussed . the @xmath0-model shows that each galaxy has its own regression line differing only in the zero point . this implies that the ogle - ii leavitt law is universal in the range of metallicities of the magellanic clouds . on the contrary , the combined model for @xmath11 shows that each galaxy has its own linear regression with the expected difference in zero point , but with unexpected difference in slope . this result has already been reported in the literature , but its causes are not yet clear . lmc models fitted by using all ogle - ii and ogle - iv data by @xmath1-regression are adequate , but ogle - iv @xmath226band has specification problems . these problems could be due to blending and the lack of extinction correction in these data . there is no evidence of unstability in their zero point and slope parameters . on the other hand , despite of the specification problems in smc models for ogle - ii and ogle - iv data , there is not enough evidence of unstability in the slope parameters when they are fitted with ogle - iv data that have more observations compared to ogle - ii . however , when ogle - iv data are split in @xmath307 d , there is evidence of two parallel regression lines . + it is clear from our results that influential cepheids do not have an important effect on the ll for the lmc . however , they could have impact in the case of galaxies with similar geometry to this of the smc . the effect of these cepheids in the measurement of distances to these galaxies , will be studied in a forthcoming paper . + finally , based on the results of this work , we suggest to use the ll relations given by equations ( [ eq76 ] ) and ( [ eq77a ] ) as a universal law for the smc and lmc galaxies . we also suggest to use the slope and zero point for @xmath177-band , obtained for lmc using ogle - iv data , shown in table [ tab6 ] . for the smc using ogle - ii without the influential cepheids , we suggest to use the equations ( [ eq54a ] ) and ( [ eq55a ] ) . + authors thank the referee , dr . pawel pietrukowicz , for his important recommendations and suggestions that have improved this manuscript . agv , bes and jrm acknowledge support from facultad de ciencias , universidad de los andes , through proyecto semilla . svd acknowledges the physics department , universidad de los andes , for the fellowship supporting the postdoctoral position held within the astronomy group . aikawa , t. , & whitney , c. , 1984 , apj , 282 , 527 aikawa , t. , & whitney , c. , 1985 , apj , 296 , 165 alcock , c. , allsman , r. , alves , d. , et al . , 2002 , apj , 573 , 338 babcock , h. , 1958 , apjs , 3 , 141 bono , g. , marconi m. , & stellingwerf , r. , 2000 , a&a , 360 , 245 butkovskaya , v. , plachinda , s. , baklanova , d. , & butkovskyi , v. , 2014 , iaus , 301 , 393 caputo , f. , marconi , m. , & musella , m. , 2000 , a&a , 354 , 610 christy , r. , 1968 , quarterly journal of the royal astronomical society , 9 , 13 drapper , n. , & smith , h. , 1998 , applied regression analysis ( 3th ed . ; john wiley & sons publication ) fadeyev , y. , 2013 , astronomy letters , 39 , 746 fernie , j. d. , 1969 , pasp , 81 , 707 fouqu , p. , arriagada , p. , storm , j. , et al . , 2007 , a&a , 476 , 73f freedman , w. , & madore , b. , 2010 , ara&a , 48 , 673 feigelson , e. d. , & babu , g. j. , 2012 , modern statistical methods for astronomy with r applications ( 1st ed . ; cambridge ; cambridge university press ) garca - varela , a. , sabogal , b. , & ramrez - tannus , m. , 2013 , mnras , 431 , 2278 gastine , t. , & dintrans , b. , 2008 , a&a , 490 , 743 gieren , w. , storm , j. , barnes , t. , et al . , 2005 , apj , 627 , 224 g gieren , w. , pilecki , b. , pietrzyski , g. , et al . , 2014 , apj , 786 , 80 gieren , w. , pilecki , b. , pietrzyski , g. , et al . , 2015 , apj , 815 , 28 g godfrey , l. g. , & orme , c. d. , 1994 , international economic review , 48 , 489 harris , j. , & zaritsky , d. , 2006 , aj , 131 , 2514 hertzsprung , e. , 1926 , bulletin of the astronomical institutes of the netherlands , 3 , 115 kafadar , k. , 1983 , journal of reseach of the national bureau of standards , 88 , 105 kanbur , s. m. , & ngeow , c - c . , 2004 , mnras , 350 , 962k kanbur , s. m. , ngeow , c - c . , nanthakumar , a. , & stevens , r. , 2007 , pasp , 119 , 512 karp , a. , 1975 , apj 199 , 448 keller , s. , & wood , p. , 2006 , apj , 642 , 834 koen , c. , kanbur , s. m. , & ngeow c - c . , 2007 , mnras , 380 , 1440 kukarkin , b. , 1937 , astr . zhurnal u.s.s.r . , 14 , 125 leavitt , h. s. , & pickering , e. c. , 1912 , harvard college observatory circular , 173 , 1 madore , b. , & freedman , w. , 1991 , pasp , 103 , 933 marconi , m. , musella , m. , & fiorentino , g. , 2005 , apj , 632 , 590 molnr , l. , & szabados , l. , 2014 , mnras , 442 , 3222 montgomery , d. c. , peck , e. e. , & vining , g.g . , 2012 , introduction to linear regression analysis ( 5th ed . ; john wiley & sons publication ) neilson , h. , & ignace , r. , 2014 , a&a , 563 , l4 ngeow , c - c . , kanbur , s. m. , bellinger , e. p. , et al . , 2012 , ap&ss , 341 , 105 nun , i. , pichara , k. , protopapas , p. , et al . , 2014 , apj , 793 , 23 pietrzyski , g. , gieren , w. , udalski , a. , et al . , 2004 , aj , 128 , 2815p pietrzyski , g. , thompson , i. , gieren , w. , et al . , 2010 , nature , 468 , 542 pietrzyski , g. , thompson , i. , graczyk , d. , et al . , 2011 , apj , 742 , 20 pilecki , b. , graczyk , d. , gieren , w. , apj , 2015 , 809 , 29 plachinda , s. , 1990 , krymskaia astrofizicheskaia observatoriia , 81 , 112 plachinda , s. , 2000 , a&a , 360 , 642 poleski , r. , 2008 , acta astron . , 58 , 313 ramsey , j. b. , 1969 , journal of the royal statistical society series b , 31(2 ) , 350 ramsey , j. b. , 1974 . frontiers in econometrics . p. zarembka . nueva york ; academic press ) sandage , a. , & tammann , g. , 1968 , apj , 151 , 531 sandage , a. , & tammann , g. , 1969 , apj , 157 , 683 schwarzenberg - czerny , a. , 1989 , mnras , 241 , 153 severny , a. , kuvshinov , v. , & nikulin , n. , 1974 , krymskaia astrofizicheskaia observatoriia , izvestiia obs . 50 , 3 simon , n. , & schmidt , e. , 1976 , apj , 205 , 162 soszyski , i. , poleski , r. , udalski , a. , et al . , 2008 , acta astron . , 58 , 163 soszyski , i. , udalski , a. , poleski , r. , et al . , 2012 , acta astron . , 62 , 219 soszyski , i. , udalski , a. , szymaski m. , et al . , 2015 , acta astron . , 65 , 279 strassmeier , k. , 2009 , a&arv , 17 , 251 stothers , r. , 2009 , apj , 696 , l37 submramanian , s. , & subramanaiam , a. , 2015 , a&a , 573 , a135 tammann , g. a. , sandage , a. , & reindl , b. , 2003 , a&a , 404 , 423 turner , d. , abdel - sabour abdel - latif , m. , & berdnikov , l. , 2006 , pasp , 118 , 410 udalski , a. , soszyski , i. , szymaski , m. , et al . , 1999a , acta astron . , 49 , 223 udalski , a. , soszyski , i. , szymaski , m. , et al . , 1999b , acta astron . , 49 , 437 udalski , a. , szymaski , m. , kubiak , m. , et al . , 1999c , acta astron . , 49 , 201 udalski , a. , 2000 , acta astron . , 50 , 279 udalski , a. , wyrzykowski , . , pietrzyski , g. , et al . , 2001 , acta astron . , 51 , 221 udalski , a. , szymaski , m. , & szymaski , g. , 2015 , acta astron . , 65 , 1 wade , g. , chadid , m. , shorlin , s. , et al . , 2002 , a&a , 392 , l17 weiss , w. , 1986 , a&a , 160 , 243 white , h. , 1980 , econometrica , 48 , 817 whitney , c. , 1983 , apj , 274 , 830 yohai , v. , 1987 , the annals of statistics , 15 , 642

in this article we revisit the notion of _ state _ : when it is preceded by the adjective _ pure _ it is commonly understood within _ quantum _ formalism as a unit ray in an hilbert space . we revisit the way a state is transformed through the action of an operator named _ quantum observable _ , and the way it evolves with the introduction in the formalism of an evolution parameter , i.e. a _ time_. hilbert spaces in quantum physics are usually vector spaces on complex numbers , and are infinite - dimensional for many of them . here , our approach is utterly minimalist , not only for pedagogical purposes , but also to show that quantum formalism applies as well to unexpected situations , far beyond our modelling of the microscopical world . it is minimalist in the sense that our departure point is the two - dimensional hilbert space on the real numbers , i.e. the euclidean plane , denoted @xmath5 , so familiar to us since , say , the kindergarden - grade 12 . think about the following quote extracted from common core state standards initiative @xcite _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ although there are many types of geometry , school mathematics is devoted primarily to plane euclidean geometry , studied both synthetically ( without coordinates ) and analytically ( with coordinates ) . euclidean geometry is characterized most importantly by the parallel postulate , that through a point not on a given line there is exactly one parallel line . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ besides parallel lines , _ vector _ is also a central concept of euclidean geometry , and the terminology associated to it can be a source of confusions , as exemplified by the following question asked on march 23 2013 at 22:15 on the site _ math.stackexchange.com_ , _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ i m trying to understand the difference between the sense , orientation , and direction of a vector . according to this , sense is specified by two points on a line parallel to a vector . orientation is specified by the relationship between the vector and given reference lines ( which i m interpreting to be some basis ) . _ _ however , these two definitions seem to be synonymous with direction . how do these 3 terms differ ? _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ the most clear , detailed , correct in our sense , answer was given on january 1 2014 , at 9:44 , as _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ here is how i think of it , let s construct a vector from scratch using just two points in space @xmath6 , @xmath7 . draw a line segment between the two points @xmath6 , @xmath7 . the magnitude of the line segment is the `` length '' of the vector . the `` orientation '' of the line segment we can define as the angle that the line segment makes with the horizontal axis . to be clear this angle is measured counterclockwise from the positive @xmath8 axis and is an angle between @xmath9 and @xmath10 . ok so far we just have a line segment that is situated on the plane somewhere and we know how it is oriented , but there is no arrowhead yet . now the last thing you need is the `` sense '' , that basically tells you where to put the arrowhead and implies an order . we can define this vector as @xmath11 with an arrow over it , where you read it as the vector starting from @xmath6 and ending at @xmath7 . _ _ more generally you can define a vector by defining its magnitude ( length ) , its orientation , and then its sense . but keeping in mind , technically a vector is an equivalence class . so there are an infinite number of vectors which are parallel to each other ( have the same orientation ) and have the same sense or same choice of where to put the arrowhead ( there are only two possible senses , since the arrowhead can be placed on @xmath6 or on @xmath7 ) . but i do nt want to confuse you . usually we discuss vectors that are situated at the origin so we do nt concern ourselves over other equivalent vectors . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ hence , a unit ray in the euclidean plane is an orientation associated to an angle @xmath12 , equivalently to a point of the upper unit semi - circle deleted of its left boundary point . it is the unit circle @xmath13 where diametrically opposed points are made identical , i.e. @xmath14 as a physical realization of this set , we immediately think of the compass without considering the sense , i.e , by identifying north with south , east with west , .... , or even better , to the familiar protractor shown in figure [ protractor ] . is clearly apparent on the picture , with @xmath15 , @xmath16 , etc , sign minus being attributed to the internal graduation.,width=480 ] the true compass orientations @xmath17 senses are described by the unit circle itself . in order to avoid notational complications , we consider in this paper states as elements of the latter , keeping in mind the possibility to view as equivalent two diametrically opposed points . now , multiplying a unit vector or a ray by a real number has a deep signification both in euclidean geometry as in quantum formalism , since it lies at the heart of the superposition principle , which states that any linear combination of two vectors is a vector , and that a new unit vector or ray is built from this vector , if it is not zero , by dividing it by its length . an appreciable part of the material presented here is borrowed from references @xcite , @xcite and @xcite . the article is organized as follows . in section [ r2dirac ] we briefly recall the definition of hilbert space before rewriting the language of euclidean geometry by using _ bra _ and _ ket _ dirac notations . we then describe _ pure _ or _ mixed _ states , and symmetric @xmath18 matrices as observables in these real hilbert spaces . in section [ intquant ] we proceed with the integral quantization of the circle , after a rapid survey of the method . this method allows us to give in a straightforward way the quantum counterparts of classical models such as functions ( and even distributions ) defined on the circle . some probabilistic aspects of the obtained quantum objects and of their _ semi - classical _ portraits are examined in great details . section [ evolution ] is devoted to the description of a quantum dynamics adapted to our model . indeed , an adaptation is necessary since we are supposed to use real numbers , vectors and matrices in our quantum formalism , and the latter can not be derived from the quantization of some phase space . such a classical object does not exist in our formalism , and so our quantum dynamics can not be derived from an hamiltonian system . if we want to use quantum dynamics _ la _ dirac , we are forced to introduce at a certain stage pure imaginary numbers . hence , the most general hamiltonian compatible with the equation ruling the quantum evolution of symmetric @xmath18 real matrices must be a symmetric complex matrix . on the other hand , we can deal with a formalism using real numbers only and leading to a kind of majorana evolution equation . then unitary evolution operators are just @xmath18 rotation matrices , as expected in two - dimensional euclidean geometry . in section [ entanglement ] , we address the question of entanglement of two systems with states in the euclidean plane . states of such objects are elements of the tensor product @xmath19 . now , because of the elementary identity @xmath20 , this tensor product is isomorphic to @xmath21 ( and also to @xmath22 ) . since the plane @xmath5 can be also viewed as the complex plane @xmath23 , we get the possibility of understanding initial entangled objects either as elements of the more elementary complex hilbert spaces @xmath2 , e.g. , the space of quantum one - half spins , or as elements of the quaternion field @xmath3 . this set of possibilities sheds an interesting light on entangled objects viewed as cat states in @xmath2 . the purpose of section [ magnint ] is to give a physical exemple of the quantum hamiltonian appearing in euclidean plane quantum dynamics . this is done by considering for instance a charged spin one half particle , e.g. , an electron , interacting with a magnetic field with constant direction in the plane . the demonstration rests upon an integral quantization of the 2-sphere @xmath24 viewed as the phase space with canonical coordinates @xmath25 , where @xmath26 has the meaning of an angular momentum and @xmath27 is the azimutal angle . in section [ conclu ] we present some interesting perpectives of generalisations of the content of our work , centred on the continuation of the game @xmath28 ( as vector spaces ) with @xmath29 ( as vector spaces ) , the latter cartesian product being given an octonionic structure . in appendix [ pararho ] we describe different parametrizations of real @xmath30 density matrices together with interesting properties . quaternions in different languages and their basic properties are described in appendix [ allsu2 ] , by insisting on their @xmath31 representation . in appendix [ unitsphere ] we present the integral quantization of the @xmath0-sphere which is needed to establish our result in section [ magnint ] . we first recall that a finite - dimensional hilbert space is a finite - dimensional real or complex vector space @xmath32 equipped with a scalar ( or dot or inner ) product @xmath33 or @xmath23 , between any pair of vectors @xmath34 , @xmath35 , with the following properties @xmath36 where @xmath37 , @xmath38 , in are arbitrary complex or real numbers , and encodes the _ semi - definiteness _ of the scalar product . in the case of real vector space , the scalar product is fully symmetric and bilinear . hence , it is clear that the two - dimensional real vector space equipped with the usual dot product , i.e. the euclidean plane @xmath5 , is the simplest non - trivial hilbert space . we now adopt the dirac notations for vectors in @xmath5 after restoring the familiar arrow or hat superscripts and boldface symbols . given the orthonormal basis of @xmath5 defined by the two familiar unit vectors @xmath39 and @xmath40 , as is shown in figure [ r2fig ] , any vector @xmath41 in the plane decomposes as @xmath42 we also use in the sequel the column vector notation to designate @xmath41 with implicit mention of orthonormal basis , @xmath43 in such a way that the dirac _ ket _ and _ bra _ notations become transparent , @xmath44 hence , elementary matrix calculus allows to formulate scalar product and ( orthogonal ) projector in _ bra ket _ ( respectively in _ ket bra _ ) notations : @xmath45 since we view the euclidean plane as a space of quantum states , pure states are identified with unit vectors , discarding , as announced in the introduction , the imposed equivalence between opposed senses along the orientation support . hence , the unit circle @xmath13 in the plane is the natural set of parameters which distinguish these states . to each @xmath46 , polar angle of the unit vector @xmath47 , corresponds the pure state @xmath48 . in particular , with these notations and associated quantum terminology , the horizontal orientation " or _ spin up _ state is @xmath49 , @xmath50 is the vertical orientation " or _ spin down _ state . their quality to form an orthonormal basis is encoded in the relations @xmath51 together with the fact that the sum of their corresponding orthogonal projectors _ resolves the unity _ @xmath52 ( 60,60 ) ( 10,20)(1,0)30 ( 10,20)(0,1)30 ( 43 , 24)(0,0)@xmath53 ( 29 , 52)(0,0)@xmath54 ( 45 , 41)(0,0)(quantum ) state in `` spin representation '' ( 10 , 16.5)(0,0)@xmath55 ( 30 , 10)(0,0)@xmath56 ( 30 , 1)(0,0)@xmath57 ( 8 , 52)(0,0)@xmath58 ( 16 , 38)(0,0)@xmath59 ( 20 , 26)(0,0)@xmath27 ( 14,20)(10,15)[tr ] ( 10,20)(1,2)14 pursuing with our quantum terminology , the spin representation of a pure state lies in the form of the expansion or decomposition @xmath60 there exists another representation which could be abusively named `` phase space '' , as will become clearer in the next section . it is the `` unit circle '' representation given by the trigonometric function @xmath61 however , we should be aware that in our model there is no quantum observable underlying localisation on the whole unit circle , at the difference of ordinary quantum mechanics for which there exists a position operator , @xmath62 , underlying configuration representation , or a momentum operator , @xmath63 , underlying momentum representation . on the other hand , we know that in ordinary quantum mechanics there is no sharp localisation in the corresponding phase space @xmath64 due to the non commutativity of these two operators ( heisenberg ! ) . a similar situation is encountered here with the fuzzy localisation exemplified by the probability distribution families on the unit circle derived from , @xmath65 to the pure state @xmath66 corresponds the orthogonal projector @xmath67 given by : @xmath68 where @xmath69 is the matrix of rotation in the plane by the angle @xmath27 , @xmath70 in quantum formalism , the projector @xmath71 is also called pure state . this designation is consistent with the fact that the projector ignores orientation , @xmath72 . then , _ mixed states _ or _ density matrices _ are unit trace and non - negative matrices of the form @xmath73 with @xmath74 the conditions express the non - negativeness of @xmath75 , i.e. @xmath76 for all @xmath77 . pure states are rank - one such matrices corresponding to @xmath78 , i.e. @xmath79 . more details are given in subsection [ ssecqcirc ] and in appendix [ pararho ] . a more enlightening parametrisation is given in terms of the highest eigenvalue @xmath80 , @xmath81 given in and polar angle @xmath27 , @xmath82 of the corresponding pure eigenstate @xmath77 pointing toward the upper half - plane . introducing the parameter @xmath83 , @xmath84 , the most general form of a real density matrix can be given , as a @xmath85-periodic matrix , in terms of the polar coordinates @xmath86 of a point in the upper half unit disk : @xmath87 this expression epitomizes the map from the closed unit disk to the set of density matrices . more precisely , each @xmath88 is univocally ( but not biunivocally ) determined by a point in the unit disk @xmath89 , with polar coordinates @xmath90 , @xmath91 , @xmath92 . we check that for @xmath93 the density matrix is just the orthogonal projector on the unit vector @xmath77 with polar angle @xmath27 , i.e. @xmath94 . the parameter @xmath95 can be viewed as a measure of the distance of @xmath75 to the pure state @xmath67 while @xmath96 measures the degree of mixing " . a statistical interpretation is made possible through the von neumann entropy@xcite defined as @xmath97 as a function of @xmath98 $ ] @xmath99 is nonnegative , concave and symmetric with respect to its maximum value @xmath100 at @xmath101 , which corresponds to @xmath102 . the graph of @xmath99 as a function of @xmath95 is presented in figure [ srho1 ] . in fact , this function is the basic boltzmann - gibbs or shannon entropy in the case of two possibilities with probabilities @xmath80 and @xmath103 . as a function of parameter @xmath104$].,width=288 ] according to the canons of quantum formalism , a quantum observable is a self - adjoint operator in the hilbert space of quantum states , because self - adjointness is essential in the copenhagen - dirac - von neumann interpretation of quantum physics @xcite , @xcite , @xcite , @xcite . indeed , the spectral theorem for bounded or unbounded self - adjoint operators is the key for ( sharp ) quantum measurement , as it was mentioned above . in our case , a self - adjoint operator @xmath6 is represented by a @xmath30 real symmetric matrix and its spectral decomposition reads in terms of the two orthogonal projectors corresponding to its real eigenvalues : @xmath105 an immediate example is provided by the spin representation , which is also understood through the superposition of the two basic projectors @xmath106 in terms of quantum measurement , this decomposition means that the observable `` orientation '' assumes two values only : @xmath107 , associated to the projector on the horizontal or spin up state @xmath108 and the other one , @xmath109 , associated to the projector on the vertical or spin down @xmath110 . those measurements of orientation with only two issues are sharp in the sense that the corresponding projectors are orthogonal . at this point , we understand why no angular sharp localisation can be established on the unit circle . by sharp localisation we mean the existence of an angle operator @xmath6 acting as a multiplication operator in `` angle '' representation @xmath111 . such an operator does not existe because no @xmath30 real symmetric matrix @xmath6 can satisfy the equation @xmath112 , for all @xmath113 . three basic matrices generate the jordan algebra @xcite of all real symmetric @xmath18 matrices . they are the identity matrix and the two real pauli matrices , @xmath114 any element @xmath6 of the algebra decomposes as : @xmath115 the product in this algebra is defined by @xmath116 which entails on the level of components @xmath117 , the relations : @xmath118 in this context , a pure state @xmath77 in the plane can be viewed as the eigenstate of a system , e.g. a spin @xmath119 , for the eigenvalue @xmath120 of the following operator @xmath121 whereas it is eigenstate of @xmath122 for the eigenvalue @xmath123 . it is interesting to notice that the density operator can be expressed in terms of @xmath124 as @xmath125 any quantization of functions defined on a set @xmath126 ( e.g. a phase space or something else ) should reasonably meet four basic requirements @xcite : it is a linear map @xmath127 : @xmath128 where * @xmath129 is a vector space of complex or real - valued functions @xmath130 on a set @xmath126 , i.e. a `` classical '' mathematical model , * @xmath131 is a vector space of linear operators in some real or complex hilbert space @xmath132 , i.e , a `` quantum '' mathematical model . the map is such that the function @xmath133 is mapped to the identity operator @xmath134 on @xmath132 , a real @xmath135 is mapped to a self - adjoint operator @xmath136 in @xmath132 or , at least , a symmetric operator ( in the infinite dimensional case ) if @xmath126 is acted on by a symmetry group @xmath137 , i.e. for @xmath138 , @xmath139 , then there exists a unitary representation @xmath140 of the group such that @xmath141 , with @xmath142 . in our case , @xmath132 is finite - dimensional and the operators @xmath143 can be viewed as matrices . of course , further requirements are necessarily added , depending on the mathematical structures equipping @xmath126 and @xmath129 and their respective physical relevance . an easy way to achieve the above minimal programme is to use resources offered by some adapted integral calculus . suppose that the set @xmath126 is equipped with a measure @xmath144 allowing to deal with integrals like @xmath145 , in which the functions @xmath146 are matrix valued , the integral being performed on each matrix element . now , suppose that there exists an @xmath126-labeled family of matrices @xmath147 resolving the identity @xmath134 in the following sense : @xmath148 the case we are specially interested in occurs when the @xmath149 s are unit trace and non - negative , i.e. are density matrices , @xmath150 because of their possible probabilistic content ( see below ) . with this material at hand , the integral quantization of functions @xmath130 on @xmath126 is the linear map @xmath151 which fulfills at least our quantization requirements ( i ) and ( ii ) and it takes the classical function @xmath130 to the operator @xmath143 . if the matrices @xmath149 are real symmetric , then ( iii ) is also satisfied . to get ( iv ) , i.e. covariance , we need to add more structure on the measure space @xmath152 . such structures will appear in a natural way in the examples we consider in this paper , namely the circle , @xmath153 , and the sphere , @xmath154 . any quantization map should be complemented by a dequantization " map , associating to @xmath143 a function @xmath155 , more or less similar to the original @xmath130 , and giving a semi - classical portrait " of @xmath143 . in quantum mechanics , examples are provided by the husimi function or the wigner function@xcite , more generally by lower ( lieb ) @xcite or contravariant ( berezin ) @xcite symbols . the terminology _ lower _ , which will appear below together with _ upper _ , is justified by the two berezin - lieb inequalities which follow from the symbol formalism . as a matter of fact , some properties of the operator @xmath136 , may be derived or at least well grasped from functional properties of its lower symbol , which is defined in the present context by : @xmath156 when @xmath157 ( density matrix ) for all @xmath8 , this new function is the local average of the original @xmath135 with respect to the probability distribution @xmath158 with respect the measure @xmath159 : @xmath160 in many cases , this generalisation of convolution has a regularising effect on the original @xmath135 . for instance , if @xmath161 is a dirac peaked at @xmath162 , @xmath163 for any test function @xmath164 , then @xmath165 , a function which is expected to be more regular around @xmath166 . here , the measure space @xmath167 is the unit circle equipped with its uniform ( lebesgue ) measure : @xmath168 the hilbert space is the euclidean plane @xmath169 . note that @xmath13 is also a group , isomorphic to the abelian group so@xmath4 of rotations in the plane . thus , we can expect that rotational covariance mentioned in point ( iv ) plays a central role here . it is indeed the case since the covariance property described in appendix [ pararho ] allows one to define the family of density operators obtained from rotational transport of the initial density operator introduced in , @xmath170 where the rotation matrix @xmath171 is defined by . it is straightforward to prove that this family resolves the identity @xmath172 the proof can be made by hand , integrating each matrix element . actually it is the direct consequence of a famous result in group representation theory , named schur s lemma @xcite . it follows the @xmath13-labelled family of probability distributions on @xmath173 @xmath174 such an expression reminds us of the cardioid distribution ( see @xcite , page 51 ) . at @xmath175 we get the uniform probability on the circle whereas at @xmath93 we get the simplified resolution of the identity @xmath176 and the pure state probability distribution @xmath177 previously introduced in . at given @xmath95 and @xmath178 a distance @xmath179 on the circle is defined as @xmath180 where the subscript @xmath181 refers to the hilbertian norm @xmath182 defined on matrices @xmath6 @xcite . the quantization of a function ( or distribution ) @xmath183 on the circle based on leads to the 2@xmath1842 matrix operator @xmath185 where @xmath186 is the average of @xmath135 on the unit circle and @xmath187 . the symbols @xmath188 and @xmath189 are for the cosine and sine doubled angle fourier coefficients of @xmath135 , @xmath190 the simplest function to be quantized is the angle function @xmath191 , i.e. the @xmath192-periodic extension of @xmath193 for @xmath46 , @xmath194 its eigenvalues are @xmath195 with corresponding eigenvectors @xmath196 . its lower symbol is given by the smooth function @xmath197 the existence of the set @xmath198 as the particular case @xmath93 and @xmath199 of the density matrices permits to proceed with the simplified coherent state " integral quantization @xmath200 the quantized version of the angle function @xmath201 introduced above becomes : @xmath202 with eigenvalues @xmath203 . thus , a localisation measurement on the circle would yield those two values only , along the spectral decomposition @xmath204 with @xmath205 the family of pure states @xmath206 allows one to carry out the symbol calculus _ la _ berezin - lieb mentioned above . to any _ self - adjoint _ operator @xmath6 one associates the two types of _ symbol _ , functions @xmath207 et @xmath208 respectively defined on the unit circle by @xmath209 and @xmath210 where @xmath208 is named upper or contravariant symbol of @xmath6 , besides the fact that it is also the classical counterpart of @xmath6 in the context of integral quantization . the simplest upper symbols and the lower symbols of non - trivial basic elements are respectively given by : @xmath211 for the symmetric matrix ( [ msym13 ] ) there follows the two symbols : @xmath212 for instance , the lower symbol of the quantum angle ( [ qangle13 ] ) is equal to @xmath213 a @xmath85-periodic function that smoothly varies between the two eigenvalues of @xmath214 , as is shown in figure [ quantangle13 ] . -periodic function @xmath215 that varies between the two eigenvalues of @xmath214.,width=384 ] one should notice that all the symbols defined above belong to the subspace @xmath216 of real fourier series that is the linear span of the three functions @xmath59 , @xmath217 , @xmath218 . hence @xmath208 is highly non unique since it is defined up to the addition of any function @xmath219 lying in the orthogonal complement of @xmath216 , i.e. such that @xmath220 . let us make explicit the berezin - lieb inequalities in the present context . let @xmath221 be a convex function . denoting by @xmath222 the eigenvalues of the symmetric matrix @xmath6 , we have @xmath223 this double inequality is not trivial . independently of the euclidean context it reads as : @xmath224 where @xmath225 and @xmath226 denotes the mean value on a period . if one applies ( [ berlieb13 ] ) to the exponential function @xmath227 , we get an intertwining of inequalities involving bessel functions of the second kind and the hyperbolic cosine : @xmath228 behind the resolution of the identity lies an interesting interpretation in terms of geometrical probability . let us consider a borel@xcite subset @xmath229 of the interval @xmath230 and the restriction to @xmath229 of the integral : @xmath231 one easily verifies the following properties : @xmath232 the application @xmath233 defines a normalized measure on the @xmath234-algebra @xcite of the borel sets in the interval @xmath230 , assuming its values in the set of positive or null @xmath30 matrices ( povm ) . let us now put into evidence the probabilistic nature of the measure @xmath235 . let @xmath236 be a unit vector . the application @xmath237 is clearly a probability measure , since it involves precisely the distribution @xmath238 introduced in . for instance , for @xmath239 $ ] and @xmath240 , we have @xmath241\right)= \frac{\theta_2-\theta_1}{2\pi } - \frac{1}{\pi}\sin\left(\theta_2-\theta_1\right ) \sin\left(\theta_2+\theta_1\right)\ , .\ ] ] now , the quantity @xmath242 means that direction @xmath243 is examined from the point of view of the family of vectors @xmath244 . as a matter of fact , it has a _ geometrical probability _ interpretation in the plane @xcite . with no loss of generality let us choose @xmath240 . recall here the canonical equation describing a straight line @xmath245 in the plane : @xmath246 where @xmath247 is the direction normal to @xmath245 and the parameter @xmath248 is equal to the distance of @xmath245 to the origin . there follows that @xmath249 is the ( non - normalized ) probability measure element on the set @xmath250 of the lines randomly chosen in the plane . picking a certain @xmath27 , consider the set @xmath251 of the lines normal to @xmath252 that intersect the segment with origin @xmath55 and length @xmath253 equal to the projection of @xmath252 onto @xmath254 as shown in figure [ probge13 ] ( 8,6 ) ( 5.1,4.9)(0,-1)3.9 ( 2.8,.6)o ( 7.5,1.3)@xmath254 ( 5.2,5)@xmath255 ( 3.1,1)(1,2)2 ( 3.1,1)(1,0)4.67 ( 2.1,1.5)(.1,.2)5(2,-1)2.7 ( 3.1,.1)@xmath256 0.5 cm the measure of this set is equal to : @xmath257 integrating ( [ measline13 ] ) over all directions @xmath252 gives the area of the unit circle . hence we can construct @xmath258 as the probability for a straight line in the plane to belong to the set of secants of segments that are projections @xmath259 of the unit vectors @xmath252 , @xmath260 , onto the unit vector @xmath243 . one could think in terms of _ polarizer _ @xmath261 and _ analyzer _ `` sandwiching '' the directional signal @xmath236 . as was asserted in the introduction , from a strictly quantum point of view , we should have made equivalent any vector @xmath263 of the euclidean plane with its symmetric @xmath264 , which amounts to deal with the half - plane viewed as the coset @xmath265 . following the same procedure as above , we start from the hilbert space @xmath266 and we choose the subset @xmath267 that is orthonormal . note that @xmath268 and @xmath269 are now exactly wigner semicircle distributions @xcite . we then consider the continuous family of pure states or coherent states @xmath270 . they are normalized and they solve the identity ( just change the factor @xmath271 into @xmath272 ) . the previous material can be repeated _ in extenso _ but with the restriction that @xmath12 . we now deal with the question of the existence of a quantum dynamics compatible with the simplicity of our model . let us introduce into the game a time ( or evolution ) parameter @xmath273 . can we associate with it a self - adjoint hamiltonian or energy operator @xmath274 , i.e. a @xmath18 real symmetric matrix @xmath274 ? a first problem concerns the construction of @xmath274 without the existence of a classical phase space as a guidance . let us circumvent this difficulty by simply postulating its existence . then , another issue makes our effort totally pointless . indeed , if we rely on the copenhagen - dirac - von neumann canons @xcite , the evolution of a time - dependent quantum observable or state is ruled by the equation @xmath275 + \ii \hbar\frac{\partial a}{\partial t}\ , .\ ] ] the presence of the imaginary unit @xmath276 in is unavoidable since the latter has to be a relation between symmetric operators acting in complex hilbert spaces . on the other hand , we want to keep the reality of our euclidean plane and of the matrices acting on it . thus , this number @xmath276 should be canceled because of its appearance somewhere on the right - hand side of . the only place where it can stand lies in the expression of @xmath274 as a @xmath18 complex symmetric matrix , and we easily check that the most general allowed matrix form of @xmath274 is @xmath277 actually the diagonal hamiltonian parameter @xmath278 does not play any role . moreover @xmath279 can be time - dependent . hence , no real symmetric matrix , except the trivial multiple of the identity , can represent an hamiltonian in the context of our approach . if we wish to keep real our matrices , we have to abandon the requirement of being symmetric . instead , let us introduce the real `` pseudo - hamiltonian '' @xmath280 then , by dividing by @xmath276 the evolution equation , one obtains an equation involving only real quantities , our _ baby majorana equation _ @xcite , @xmath281 + \hbar\frac{\partial a}{\partial t } \ , .\ ] ] the price to pay with our manipulation is that the eigenvalues @xmath282 of @xmath283 are real only in the trivial case @xmath284 : no physical energy comes out from our formalism . on the other hand , if @xmath279 is constant , the pseudo - hamiltonian @xmath283 may generate unitary time evolution . more precisely , in the case of time explicit independence of @xmath6 , the integration of is straightforward : @xmath285 i.e. the appearance of the @xmath30 proper rotation matrix in the plane as the solely possible continuous unitary operator in the plane is unescapable . in the case of the time evolution of a pure state @xmath286 , this solution corresponds to a uniform rotation , with frequency @xmath287 , of the unit vector @xmath288 . on the algebraic side , we note that our real quantum dynamics involves the antisymmetric matrix @xmath289 . together with @xmath290 and @xmath291 these 3 matrices generate the lie algebra @xmath292 of the group sl@xmath293 @xmath294= 2\sigma_3\ , , \quad [ \tau_2,\sigma_3]= 2\sigma_1\ , , \quad [ \sigma_3,\sigma_1]= -2\tau_2= 2\tau_2\,.\ ] ] if we had sticked to the complex formalism imposed by , we would have to introduce the third pauli matrix @xmath295 , which was missing in . then the self - adjointness of @xmath274 is preserved , and quantum evolution restores the full real lie algebra of su@xmath4 generated by the @xmath296 s , @xmath297 . actually , we will use in section [ magnint ] the equation for presenting a physical example of . the possible entanglement property of quantum states appears with the construction of tensor products @xcite of hilbert spaces for describing quantum states of composite systems . in the present case , we are in presence of a remarkable sequence of vector space isomorphisms due to the fact that @xmath298 : @xmath299 where @xmath3 is the field of quaternions ( see appendix [ allsu2 ] for definition and properties of these objects ) . let us first write the ( canonical ) orthonormal basis of the tensor product @xmath300 , the first factor being for system `` @xmath6 '' and the other for system `` @xmath7 '' , as latexmath:[\[\label{tensbas } these factors @xmath108 , @xmath302 pertaining to @xmath6 or to @xmath7 , are termed `` @xmath303-bit '' or qubit " in standard language of quantum information . they can be associated to a pointer measuring the horizontal ( resp . vertical ) direction or polarisation described by the state @xmath108 ( resp . @xmath302 ) . we now introduce the celebrated bell pure states @xcite in @xmath300 as special linear superpositions of the above basis elements @xmath304 or in matrix form , @xmath305 they form an orthonormal basis of @xmath300 . they are four specific _ maximally entangled _ quantum states of two qubits . `` maximally '' refers to the degree to which a state in a quantum system consisting of two entities is entangled . this degree is measured by the von neumann entropy @xcite of either of the two reduced density operators of the state , i.e. , obtained by tracing out one of the partners in the composite systems . they exhibit perfect correlation . consider for instance the state @xmath306 . if the pointer associated to @xmath6 measures its qubit in the standard basis the outcome would be perfectly random , either possibility having probability 1/2 . but if the pointer associated to @xmath7 then measures its qubit , the outcome , although random for it alone , is the same as the one @xmath6 gets . there is quantum correlation . let us now turn our attention to the complex two - dimensional hilbert space @xmath2 , one of the most familiar mathematical objects in our initiation to quantum formalism , after having been so nicely popularised by feynman @xcite with examples like two - level atom or spin one - half objects . now , as a complex vector space , @xmath2 , with canonical basis @xmath307 , @xmath308 , has a real structure . the latter is isomorphic to the real vector space @xmath309 , itself isomorphic to @xmath19 , from the above . a standard real structure is obtained by considering the vector expansion @xmath310 i.e. , by writing @xmath311 , @xmath312 , and considering the set of vectors @xmath313 as forming a basis of @xmath309 . now , we need to define a map from this @xmath309 to @xmath19 , forgetting about the superfluous subscripts @xmath6 and @xmath7 . for this , we use the map euclidean plane @xmath5 @xmath314 complex `` plane '' @xmath23 determined by @xmath315 or , in matrix form , @xmath316 interpreting @xmath2 as the space of quantum states of a two - level system , e.g. spin one - half with notations @xmath317 the correspondances and might be interpreted as giving a real composite nature to `` up '' and down " complex objects . note that the @xmath318-matrix on the right of is its own inverse . finally , we get the matrix map from the bell basis and the basis @xmath319 in terms of components of vectors in their respective spaces , @xmath320 equivalently , in complex notations , with @xmath321 , @xmath322 where we have introduced the conjugation operator @xmath323 , i.e. the mirror symmetry with respect to the real axis , @xmath324 being the mirror symmetry with respect to the imaginary axis . note that the operator @xmath325 can be expressed in the quaternionic language for su@xmath4 , as introduced in , by @xmath326 therefore , with the above choice of isomorphisms , bell entanglement in @xmath327 is represented by a linear superposition in @xmath2 . it involves also the two mirror symmetries @xmath328 . interpreting @xmath2 as in , we see that the operator @xmath329 is a kind of `` flip '' whereas `` cat '' operator @xmath325 builds from the _ up _ and _ down _ basic states the two elementary schrdinger cats @xmath330 we can now understand better the real bell state nature of these elementary cat superpositions if we remember the identification between composite real states and complex _ up _ and _ down _ basic states . interesting too is the appearance of the flip @xmath329 in the construction of the spin one - half coherent state defined in terms of spherical coordinates @xmath331 by @xmath332 indeed , they result from the action on the up state by the operator @xmath333 given in . this operator represents the element @xmath334 in su@xmath4 , quaternionic conjugate or inverse of @xmath335 . the latter corresponds through the homomorphism so@xmath336 @xmath314 su@xmath4 to the specific rotation @xmath337 that maps the unit vector pointing to the north pole , @xmath338 , to @xmath339 , as is shown in figure [ rotkton ] and explained at length in appendix [ allsu2 ] : @xmath340 now we notice that the second column of @xmath333 is precisely the flip of the first one , @xmath341 actually , this feature is the key for grasping the isomorphisms @xmath342 . with the notations of appendix [ allsu2 ] and using quaternionic algebra , e.g. , @xmath343 @xmath17 even permutations , @xmath344 after identifying @xmath345 , as both are roots of @xmath123 . then the flip appears naturally in the final identification @xmath346 as @xmath347 the fact that the sole acceptable form the hamiltonian may assume in our model is might suggest a kind of magnetic field interacting with some one - half spin magnetic moment . how can we integrate this physics in our toy model ? the ( potential ) energy of a magnetic moment @xmath348 ( whatever the origin of the latter ) submitted to a magnetic field @xmath349 is given by @xmath350 in the case of a moving charge , the magnetic moment is proportional to the angular momentum of the latter . here we consider a magnetic moment as a physical system on its own , without any consideration of space location or something else , and proportional to an angular momentum or classical spin denoted by @xmath351 : @xmath352 the factor @xmath353 is usually named gyromagnetic factor . now , the energy of a magnetic moment submitted to a constant magnetic field alone is conserved . choosing the direction of the unit vector @xmath354 , part of the orthonormal basis in space @xmath355 , as the same as @xmath349 , we have for the hamiltonian @xmath356 and so the polar angle @xmath357 is conserved : the magnetic moment experiences a precession about the @xmath354-axis at constant azimuthal velocity @xmath358 in such a way that @xmath359 is also conserved . actually , this precession motion which takes place on the surface of a sphere @xmath360 of radius @xmath361 has to be viewed as a phase trajectory in @xmath360 , the latter being as the classical phase space for the magnetic moment . the canonical conjugate coordinates are in the present case action - angle variables , the angle variable being the azimuthal angle @xmath27 whereas the action variable @xmath362 is precisely @xmath363 . indeed , from the expression @xmath364 of the rotational invariant measure on @xmath360 , one infers that @xmath363 is canonically conjugate to @xmath27 , and the action variable defined as @xmath365 is identical to @xmath363 in the case of a constant external magnetic field . note that the pair @xmath366 are a part of the cylindrical coordinates of the sphere @xmath360 with radius @xmath361 . let us turn our attention to the quantum model of a magnetic moment interacting with a constant magnetic field . following the integral quantization procedure of the two - sphere @xmath24 and of functions ( or distributions on it , as is explained in details in appendix [ unitsphere ] , and yielding the expression ) , the quantum version of the classical hamiltonian with @xmath367 reads as @xmath368 where the parameter @xmath369 plays the same rle as in . now , contrarily to our previous choice , we take the direction of the magnetic field along the unit vector @xmath370 , , @xmath371 . this new choice gives a quantum hamiltonian of the form , @xmath372 this formula leaves to us the freedom to choose @xmath95 and @xmath361 in order to comply with observations , for instance for the electron . we just remind that , in its quantum description , the electron _ total _ spin magnetic moment is given by the formula @xcite : @xmath373 where @xmath374 is the charge of the electron , @xmath221 is the land @xmath221-factor , and by the quantization equation along the direction determined by @xmath339 : @xmath375 where @xmath376 is the bohr magneton . basic aspects of quantum formalism have been presented in this paper , taking the euclidean plane as the simplest mathematical model to be considered in a pedagogical approach . the corresponding quantum dynamics has been given a physical flavour at the price of increasing the dimension from 2 to three in order to implement what we call integral quantization , and a simple illustration is presented in terms of magnetic moment . we have shown that entanglement , a central notion of modern developments in quantum physics , allows to follow an interesting path from the euclidean plane to the quantum one - half spin formalism and to the quaternionic language as well . now , this path can be extended in order to include entanglement of two @xmath119 spins and to give an interesting description of the latter in an octonionic language . indeed , like @xmath28 ( as vector spaces ) , we can continue with @xmath29 ( as vector spaces ) , the cartesian product of two quaternion field being given an octonionic structure _ la _ cayley dickson . this is the content of a work in progress . there are various expressions for a density matrix acting on the euclidean plane , i.e. a @xmath18 real positive matrix with trace equal to @xmath59 . the most immediate one is the following with parameters @xmath377 and @xmath378 : @xmath379 the above inequalities imply the following ones @xmath380 let @xmath381 be the highest eigenvalue of @xmath75 ( the lowest one is @xmath382 ) . the spectral decomposition of @xmath75 reads as @xmath383 where @xmath384 is the corresponding unit eigenvector , chosen as pointing toward the right half - plane . we could have as well chosen the opposite @xmath385 pointing in the left half - plane since @xmath386 . our choice corresponds to the most immediate one in terms of orthonormal basis of the plane issued from the canonical one @xmath387 through the rotation by @xmath27 . let us make explicit the decomposition , @xmath388 we derive from this expression the polar parametrization of the @xmath389 parameters of @xmath75 : @xmath390 in return , we have the angle @xmath391 $ ] in function of @xmath377 and @xmath378 @xmath392 in this way , each @xmath75 is univocally ( but not biunivocally ) determined by a point in the unit disk , with polar coordinates @xmath393 , @xmath91 , @xmath394 . also note the alternative expression issued from : @xmath395 where @xmath396 is the rotation matrix in the plane @xmath397 and @xmath291 is the diagonal pauli matrix @xmath398 in order to get the second equality in we used the important property @xmath399 therefore the expression of a matrix density to which we refer mostly often through the paper reads @xmath400 from and we derive the interesting multiplication formula @xmath401 and the resulting ( non - closed ! ) `` algebra '' of real density matrices , @xmath402= -irr'\sin(\phi -\phi^{\prime } ) \sigma_2\ , , \quad \{\rho,\rho'\ } = \rho + \rho ' + ( \cos(\phi - \phi^{\prime})- 1/2)i\ , .\ ] ] the expression is convenient to examine the way a density matrix transforms under a rotation @xmath403 in the plane . we have , @xmath404 the computation of the three following integrals is straightforward : @xmath405 @xmath406 @xmath407 @xmath408 where @xmath89 is the unit disk and @xmath409 . a convenient representation of rotations in space is possible thanks to quaternion calculus . we recall that the quaternion field as a multiplicative group is @xmath411su@xmath4 . the correspondence between the canonical basis of @xmath412 , @xmath413 , and the pauli matrices is @xmath414 , with @xmath415 . hence , the @xmath18 matrix representation of these basis elements is the following : @xmath416 any quaternion decomposes as @xmath417 ( resp . @xmath418 ) in scalar - vector notation ( resp . in euclidean metric notation ) . we also recall that the multiplication law explicitly reads in scalar - vector notation : @xmath419 . the ( quaternionic ) conjugate of @xmath420 is @xmath421 , the squared norm is @xmath422 , and the inverse of a nonzero quaternion is @xmath423 . unit quaternions , i.e. , quaternions with norm @xmath59 , which form the multiplicative subgroup isomorphic to su@xmath4 , constitute the three - sphere @xmath424 . on the other hand , any proper rotation in space is determined by a unit vector @xmath339 defining the rotation axis and a rotation angle @xmath425 about the axis , as illustrated in figure [ rotation ] . ( 40,40 ) ( 20,10)(2,1)30 ( 18,6)@xmath55 ( 35,12)@xmath426 ( 20,10)(-4,1)30 ( 16,20)@xmath427 ( 18,18)@xmath428 ( 17,26)@xmath339 ( 20,10)(-1,4)5 ( 0,10)@xmath429 the action of such a rotation , @xmath430 , on a vector @xmath429 is given by : @xmath431 the latter is expressed in scalar - vector quaternionic form as @xmath432 where @xmath433 or , in matrix form resulting from the identification , @xmath434 in which case quaternionic conjugation corresponds to the transposed conjugate of the corresponding matrix , @xmath435 in particular , for a given unit vector @xmath436 one considers the specific rotation @xmath437 that maps the unit vector pointing to the north pole , @xmath338 , to @xmath339 , as is shown in figure [ rotkton ] , @xmath438 with , in quaternionic matrix notations , @xmath439 @xmath440 ( 50,30 ) ( 20,10)(4,3)20 ( 20,10)(4,1)15 ( 18,6)@xmath55 ( 40,21)@xmath339 ( 33,9)@xmath441 ( 20,12.5)@xmath442 ( 27,11.5)@xmath442 ( 22,16)@xmath357 ( 17,28)@xmath354 ( 20,10)(0,4)23 the unitary irreducible representations @xmath443 of su@xmath4 are labelled by half - integers @xmath444 . for a given @xmath444 , the corresponding complex hilbert space is finite - dimensional , with dimension @xmath445 . the matrix elements of the linear operator @xmath446 in the basis @xmath447 are given by @xmath448 in agreement with talman @xcite . in the lowest non trivial dimensional case we are concerned with in this paper , @xmath449 , these matrix elements are given by @xmath450 hence , we have to be careful when we work with matrix representations of quaternions , due to the choice of basis . consequently , besides the `` canonical '' one , we use preferably ( and safely ! ) the notation @xmath451 for the @xmath30 matrix given by in the `` up '' and `` down '' spin basis , @xmath452 precisely , we have @xmath453 and for the particular case @xmath454 , essential in the construction of spin @xmath119 coherent states , @xmath455 we consider the unit sphere equipped with its rotationally invariant measure : @xmath456\ , , \quad \phi\in [ 0,2\pi)\ , .\ ] ] the hilbert space is @xmath457 . the unit ball @xmath458 in @xmath410 parametrizes the set of @xmath18 complex density matrices @xmath75 . indeed , given a 3-vector @xmath459 , viewed as a pure quaternion @xmath460 , and such that @xmath461 , a general density matrix @xmath75 can be written as @xmath462 where @xmath463 ( resp . @xmath464 ) are the cartesian ( resp . spherical ) coordinates of vector @xmath429 . we have used the representation @xmath465 defined by in order to avoid any ambiguity about the basis in which is expressed the density matrix @xmath75 . if @xmath93 , with spherical coordinates @xmath466 , then @xmath75 is the pure state @xmath467 where the column vector @xmath468 is the spin @xmath449 coherent state introduced in . let us now transport the density matrix @xmath75 by using the two - dimensional complex representation of rotations in space , namely the matrix su@xmath4 representation . for @xmath469 , one defines the family of density matrices labelled by @xmath470 : @xmath471 in order to get a one - to - one correspondence with the points of the 2-sphere , we restrict the elements of su@xmath4 to those corresponding to the rotation @xmath472 bringing the unit vector @xmath354 pointing to the north pole to the vector @xmath473 with spherical coordinates @xmath474 , as described in and but with an opposed sign for @xmath27 in order to get for @xmath93 . @xmath475 the value of the integral @xmath476 shows that the resolution of the unity is achieved with @xmath477 only . then , it is clear that @xmath478 it is with this strong restriction and the simplified notation @xmath479 that we go forward to the next calculations with the resolution of the unity @xmath480 note that the resolution of the identity with the su@xmath4 transport of a generic density operator is possible only if we integrate on the whole group , as it was done in @xcite . the @xmath24-labelled family of probability distributions on @xmath481 @xmath482 at @xmath175 we get the uniform probability on the sphere whereas at @xmath93 we get the probability distribution corresponding to the spin 1/2 cs , @xmath483 the quantization of a function ( or distribution ) @xmath484 on the sphere based on leads to the 2@xmath1842 matrix operator @xmath485 where @xmath486 is the average of @xmath135 on the unit sphere and @xmath487 and @xmath488 are fourier coefficients of @xmath135 on the sphere defined as @xmath489 since the sphere is a phase space with canonical coordinates @xmath490 , @xmath491 , and @xmath492 , the latter may be thought as the simplest functions to be quantized . we find for the quantization of @xmath303 @xmath493 its eigenvalues are @xmath494 with corresponding eigenvectors @xmath495 . its lower symbol is given by the smooth function @xmath496 the quantization of @xmath248 yields the diagonal matrix @xmath497 with immediate eigenvalues @xmath498 and lower symbol @xmath499 we note the commutation rule @xmath500 now , applying the quantization formula of the three components of the generic unit vector provides , as expected , the three pauli matrices , up to a common factor in agreement with general results given in @xcite , @xmath501 j.p . gazeau thanks cbpf and cnpq for financial support and cbpf for hospitality . curado acknowledges cnpq and faperj for financial support . the authors are grateful to pierre martin - dussaud ( ens lyon ) for helpful comments . for a quantum - mechanical system described by a density matrix @xmath503 , the von neumann entropy is @xmath504 , the last expression being actually its shannon form . it is zero if and only if @xmath75 is a pure state , i.e. , a projector . j. von neumann , mathematische grundlagen der quantenmechanik , springer , berlin ( 1932 ) . english translation by r. t. beyer , mathematical foundations of quantum mechanics , princeton university press ( 1955 ) . a borel set is any set in a topological space that can be formed from open sets ( or , equivalently , from closed sets ) through the operations of countable union , countable intersection , and relative complement . a very pedagogical explanation of the semi - circle law is given in the junior research seminar on probability , y. sinai , s. miller , l. silberman , @xmath505-functions , random matrix theory and ramanujan graphs , online at https://web.williams.edu/mathematics/sjmiller/public_html/mathlab/public_html/jr02spring/plfrmt.pdf ( 2003 ) . see the answer to the question _ what is the difference between cartesian and tensor product of two vector spaces ? _ at http://math.stackexchange.com/questions/710481/what-is-the-difference-between-cartesian-and-tensor-product-of-two-vector-spaces

string duality is a recently discovered symmetry of string theory . string theory itself is over twenty - five years old , and has been under intensive development since 1984 as the leading candidate for a unified theory of particle physics and gravity . the reason that some of its symmetry has been overlooked until now is simple , and important : string duality is not manifest in the weak - coupling perturbation expansion by which the theory is usually studied , but it is a property of the exact theory . as a result , string duality gives information about the behavior of string theory at strong coupling . in a period of a little over a year , we have gone from near - complete ignorance of the behavior of strongly - coupled strings to a rather detailed understanding of the intricate dynamics which occurs , at least in vacua having enough supersymmetry , and the subject continues to develop at a rapid pace . the central idea of string duality is that the strongly coupled limit of any string theory is equivalent to the weakly coupled limit of some other theory . all string theories are connected in this way , as well as something new and surprising : an eleven - dimensional theory known provisionally as ` m - theory . ' besides the ordinary vibrating strings which are the basic quanta of string theory , the multiplets of string duality include smooth classical objects ( solitons ) , singular classical objects ( black holes ) , and a new type of topological defect which is unique to string theory ( d - branes ) . with the improved understanding of string dynamics it has become possible to address one of the long - standing problems of quantum gravity to count the number of states of certain black holes in a controlled way , giving for the first time a statistical mechanical interpretation to the bekenstein - hawking entropy . beyond these specific results , string duality has greatly changed the way string theorists think about the fundamental principles of the theory . many ideas which once seemed to be central are now seen as technicalities , while other ideas which were neglected are now central . in this colloquium i would like to try to explain these developments and to put them in the context of the search for the unified theory of particle physics and gravity . i will use units in which @xmath0 . the gravitational coupling @xmath1 is then a length - squared , @xmath2 where @xmath3 is the planck length , @xmath4 cm . this is the natural length scale for the effects of quantum gravity to become important and so for the unification of gravity with the other interactions . it is far shorter than the length scales which can be probed directly ; the corresponding energy scale is @xmath5 far beyond the reach of accelerators . thus , to construct a unified theory one must rely heavily on theoretical reasoning such as the internal consistency of the theory , and any experimental tests will be indirect . fortunately , consistency is a very restrictive guide . general relativity is a nonrenormalizable field theory , meaning that its quantum mechanical perturbation theory has uncontrollable divergences . to see the significance of this , let us recall the four - fermi theory of the weak interaction . the weak interaction was originally described as an interaction of four fermionic fields at a spacetime point as depicted in figure 1a . the weak coupling constant @xmath6 also has units of length - squared , or inverse energy - squared . in a process with a characteristic energy @xmath7 the effective dimensionless coupling is then @xmath8 . it follows that at sufficiently high energy the coupling becomes arbitrarily strong , and this also implies divergences in the perturbation theory . a second order weak amplitude is dimensionally of the form @xmath9 where @xmath10 is the energy of the virtual state in the second order process , and this diverges at large energy . in position space the divergence comes when the two weak interactions occur at the same spacetime point ( high energy = short distance ) . the divergences become worse at each higher order of perturbation theory and so can not be controlled even with renormalization . the natural interpretation of such divergences is that the theory one is working with is only valid up to some energy scale , beyond which new physics appears . the new physics should have the effect of smearing out the interaction in spacetime and so soften the high energy behavior . one might imagine that this could be done in many ways , but in fact the combined constraints of lorentz invariance and causality are very restrictive . this is because lorentz invariance requires that if the interaction is spread out in space it is also spread out in time . in fact , for the weak interaction there is only one known way to solve the short - distance problem . this is depicted in figure 1b , where the four - fermi interaction is resolved into the exchange of a vector boson . moreover , this vector boson must be of a very specific kind , coming from a spontaneously broken gauge invariance . this is the only known solution , and in fact it is the one that nature chooses . for gravity the discussion is much the same . the gravitational interaction is depicted in figure 2a . as we have noted already , the gravitational coupling @xmath1 has units of length - squared and so the dimensionless coupling is @xmath11 . this grows large at high energy and signifies a nonrenormalizable perturbation theory . again the natural suspicion is that short - distance physics smears out the interaction , and again there is only one known way to do this , shown in figure 2b . it involves a bigger step than in the case of the weak interaction : it requires that at the planck length the graviton and other particles turn out to be not points but one - dimensional objects , loops of ` string . ' it is certainly not obvious that this should be the right thing to do , but if one tries to make a consistent lorentz - invariant quantum theory of one - dimensional objects , one finds that it is possible but that the theory is highly constrained . in particular , one of the states of the string is massless and has spin two . consistency requires that such a particle be a graviton , that its long - wavelength interactions be described by general relativity , and indeed this is what is found from the sum - over - histories depicted in figure 2b ( yoneya , 1974 ; scherk and schwarz , 1974 , 1975 ) . in a sense string theory predicts gravity , in that every consistent string theory includes general relativity , even though the theory initially seems to be formulated in flat space . further , calculation of the quantum corrections to amplitudes shows that they are finite . it is also not obvious why _ only _ this idea should work . many others have been tried without success . one natural question is , why one - dimensional objects and not two - dimensional or higher ? for this at least there is a simple answer . extended objects have an infinite number of internal degrees of freedom ( the fourier modes describing their shape ) . spreading out point particles into extended objects softens the spacetime divergences but introduces potential new divergences from the internal degrees of freedom . the latter grow worse as the dimension of the object increases , and the one - dimensional case is the only one for which both the spacetime and internal behavior is under control . it could be that we suffer from a failure of imagination , and that there are other solutions . however , we see from the example of the weak interaction that if we can find even one solution to a short - distance problem we should take it very seriously and see where it leads . the result here is quite striking . besides the graviton , other states of the string with different internal oscillators excited behave as gauge bosons ( neveu and scherk , 1972 ) , and yet others are fermions ( ramond , 1971 ) . in fact , string theory automatically incorporates ( and in some ways generalizes ) three earlier ideas for explaining the patterns in the standard model , namely grand unification , supersymmetry ( gliozzi , scherk and olive , 1977 ) , and kaluza - klein theory . some of the simplest string theories , the calabi - yau models , closely resemble unified versions of the supersymmetric standard model . as a particle physicist i usually emphasize these particle physics motivations for string theory , but many of those educated in general relativity and in mathematics find string theory compelling for reasons that are not entirely the same . from newton s gravity to einstein s , from quantum mechanics to non - abelian gauge theory , new physical theories have often required new mathematics , or least mathematics that had not previously been used in physics . if one searches for higher symmetries or other more mathematical structures that might be useful in physics , one finds many connections to string theory . one reason i emphasize this now is that in the recent work on string duality , all of these different points of view have had a role to play . it is worthwhile to note that these three kinds of motivation solving the divergence problem , the connection with geometry , and explaining the broad patterns in the standard model , were also present in the weak interaction . weinberg ( 1980 ) emphasized the divergence problem as i have done . salam ( 1980 ) was more guided by the idea that non - abelian gauge theory was a beautiful structure that should be incorporated in physics . experiment gave no direct indication that the weak interaction was anything but the pointlike interaction of figure 1a , and no direct clue as to the new physics that smears it out , just as today it gives no direct indication of what lies beyond the standard model . but it did show certain broad patterns universality and the @xmath12 structure , which were telltale signs of a gauge interaction . it appears that nature is kind to us , in providing many trails to a correct theory . the key question is , how do we go from explaining broad patterns to making precise predictions ? to understand the situation , it is useful to look again to history , to the state of quantum field theory in the early 60 s . at that time there was a good technical control of the weak - coupling perturbation theory , the feynman graph expansion , but little else . important dynamical ideas were missing , such as the higgs mechanism , dynamical symmetry breaking , and confinement . these have largely to do with the fact that the vacuum , the ground state of quantum field theory , is generally a more interesting and less symmetric object than one might naively expect from the hamiltonian . without understanding these ideas one can not make sense of the standard model : it would seem to predict fractionally charged particles and a long - ranged force coupled to isospin . beyond these dynamical ideas , it remained to discover the defining principle of the theory that one should organize the physics not _ graph by graph _ but rather _ length scale by length scale _ with the shortest distances first . this realization not only made it possible to define the theory beyond perturbation theory , but also provided a framework for understanding the dynamics , both analytically and numerically . the situation is similar in string theory . there has been a good technical control of the perturbation theory , but little else , at least up until the recent developments . the need to understand the dynamics is even more acute than in the case of the standard model for two reasons . the first is that because the planck energy is so large , there is no hope to explore a ` partonic ' regime where the stringy behavior is directly visible rather , we see only the extreme low energy limit of string theory , filtered through all the dynamics at intervening scales . the second reason has to do with one of the very attractive features of string theory , that it has no free dimensionless parameters at all . instead it has , in perturbation theory , many degenerate ground states parameterized by the values of various scalar fields ( moduli ) . the parameters of the standard model come ultimately from the values of these scalars , so it is necessary to understand the dynamics which selects one of the many ground states . string theory contains quantum field theory as its low energy limit , so all of the familiar dynamics of the latter is still present , but because string theory has many more degrees of freedom it is likely to have interesting new dynamics of its own . beyond this , the central defining principle of string theory is not known . we are trying to answer the question ` what is string theory ? ' , just as wilson and others addressed the question ` what is field theory ? ' . the various properties that make string theory an attractive unifying idea also imply that the theory exists as something more than an asymptotic weak - coupling expansion . so there is good reason to expect that we will find a central principle as powerful as that in field theory , which will again enable us to better understand the dynamics . with this background i can summarize what has been learned from string duality . we have learned a number of new dynamical ideas though certainly not yet all we need . we have not yet found the central principle but we have many new clues , some of which are surprising and have taken us in unexpected directions . string duality involves an interplay of many different ideas from quantum field theory and string theory . in this part i would like to explain some of the central principles . here are two ways to think about the meaning of the coupling constant @xmath13 of a quantum field theory . the first is in terms of the weak - coupling perturbation expansion . the amplitude @xmath14 for any process can be expanded @xmath15 where @xmath13 is the amplitude for a single interaction to occur . , or only odd , contribute to a given amplitude . ] the coefficient @xmath16 is given by the sum over feynman graphs with @xmath17 interaction vertices . interpreted with sufficient care this is an asymptotic series , meaning that it gives an approximation of any desired accuracy by taking @xmath13 to be sufficiently small . thus it is also a good qualitative guide to the small-@xmath13 physics . but for @xmath13 of order @xmath18 or larger it is of limited usefulness and can miss important aspects of the physics . another way to think about the meaning of @xmath13 is in terms of the quantum fluctuation of the fields . for the purpose of this discussion it is useful to keep explicit factors of @xmath19 ; we will return to @xmath20 units after section 3.2 . when @xmath13 is small the fluctuations are small and their equations of motion can be approximated by linear equations in other words , small @xmath13 is approximately free field theory . for larger @xmath13 the fluctuations of the fields , and the nonlinearities , become larger . and the size of the quantum fluctuations is in terms of the path integral , the sum over field histories weighted by @xmath21 with @xmath22 the classical action . the fields can be rescaled in such a way that @xmath13 appears in the action only as an overall factor @xmath23 , so @xmath13 and @xmath19 appear only in the combination @xmath24 . it is then clear that small @xmath13 and small @xmath19 are equivalent with this scaling of the fields . to relate this to the familiar example of qed , one must note that the electric charges of individual quanta are related to the charge in the classical field action by @xmath25 , so the familiar expansion parameter @xmath26 is indeed @xmath27 . ] in qcd , for example , the coupling is very strong at long distance , and correspondingly the color magnetic fields undergo large fluctuations in the vacuum this is the source of confinement . weak / strong duality ( in field theory or string theory ) means that as the string coupling @xmath13 becomes large , one can find new ` dual ' fields whose fluctuations become _ small_they are characterized by a new coupling @xmath28 which is something like @xmath29 . this is similar to a fourier transform , where a function which becomes spread out in position space can become very narrow in momentum space . here though , the fourier transform is in a complicated nonlinear field space . string theory includes gravity , and so one of the fields is the spacetime metric . one might therefore have expected strongly coupled string theory to have new and very exotic physics , including large fluctuations of the spacetime geometry and all other fields . this might correspond to some sort of confined phase of gravity . one of the surprises of string duality is that this is not so . as @xmath30 , @xmath31 and so the metric of the dual theory behaves more and more classically . it is quite likely that in nature the string coupling is close to 1 rather than to 0 or @xmath32 , so that string duality does not allow one to relate the theory directly to a weakly coupled theory in which one can calculate accurately . but having an understanding of both limits , large and small coupling , constrains the kinds of qualitative physics that can occur at intermediate coupling . even more important , duality gives a great deal of information about the exact theory and its symmetries , information that is not contained in the perturbation expansion , and so new clues to the answer to ` what is string theory ? ' . another perspective on the meaning of duality starts with maxwell s equations , @xmath33 the symmetry between @xmath34 and @xmath35 in these equations is striking . if one ignores the sources , or adds magnetic sources , the equations are invariant under @xmath36 , @xmath37 . this curious fact was made more interesting by dirac s study ( 1931 ) of the quantum mechanics of a charge moving in a magnetic monopole field . he found that the wavefunction could be consistently defined only if the electric charge @xmath38 and magnetic charge @xmath39 satisfy a quantization condition @xmath40 note that if a monopole of some charge @xmath39 exists , then all electric charges must be multiples of the unit @xmath41 . this would ` explain ' why the magnitudes of the electron and proton charges should be _ exactly _ equal , a fact known to hold to one part in @xmath42 . the subject took another step forward when t hooft ( 1974 ) and polyakov ( 1974 ) showed that in any grand unified theory , magnetic monopoles actually do exist . they are classical solutions , with a nonsingular core whose size is set by the scale of spontaneous symmetry breaking . at weak coupling , the electrically and magnetically charged objects look very different . the electrically charged objects are weakly coupled and have pointlike interactions . the magnetically charged objects are strongly coupled by the dirac condition ( [ dirac ] ) , the magnetic fine structure constant @xmath43 is roughly the reciprocal of the usual one and as noted above they have cores of finite size . montonen and olive ( 1977 ) were led by various evidence to conjecture that at strong coupling the situation would be reversed : the electrically charged objects would be strongly coupled and have nonsingular cores , while the magnetically charged objects would become weakly coupled and pointlike . the strongly coupled theory would be equivalent to weakly coupled theory in which the basic quanta carried magnetic rather than electric charges . in subsequent work this conjecture was refined ( witten and olive , 1978 ; osborn , 1979 ) . it was argued to hold specifically in supersymmetric gauge theories , in particular @xmath44 theories ( @xmath45 is the number of conserved supersymmetries ) . relating this discussion to the previous section , the weakly coupled dual fields described there would be the fields corresponding to the magnetic quanta . these conjectures were greeted with wide skepticism because the evidence for them was rather circumstantial , while attempts to construct the dual fields directly in terms of the original ones did not succeed . this skepticism is now largely gone , not because the dual fields have been found ( they have not ) , but because of a substantial strengthening of the circumstantial evidence . this evidence , based on supersymmetry , will be described in the next section . the reader should notice that this whole section has been concerned with duality in field theory , not string theory . the extension to string theory will be discussed in section 3.5 . to test the duality conjectures we need to be able to say something about the physics of the strongly coupled theories . the methods available for this are very limited , but in supersymmetric theories it is possible . supersymmetry is of interest for a number of reasons . it is likely that it is associated with the breaking of the electroweak symmetry . if so , it should be discovered by the lhc if not before , and it is the one piece of new physics associated with string theory that might be accessible at accelerators . beyond this , it is an appealing mathematical structure that extends general coordinate invariance and unifies fermions and bosons . any consistent string theory must have supersymmetry , at least at the planck scale . finally , supersymmetric field theories , and string theories with unbroken supersymmetry below the planck scale , have various nice properties that make them easier to study . it is difficult to give an intuitive picture of supersymmetry . one way to think about it is that in addition to the usual spacetime dimensions , whose coordinates are real numbers @xmath46 ( and so their multiplication commutes ) , there are additional ` fermionic ' coordinates @xmath47 which anticommute , @xmath48 . these extra dimensions have no size ; the anticommuting property means that they are essentially infinitesimal . one can make sensible field theories on this space . in fact , they can be thought of as quantum field theories on ordinary spacetime but with fields that have both fermionic and bosonic components , with a symmetry relating the masses and charges of the particles with different statistics . to see why supersymmetry is valuable in studying strong coupling we need the algebra of the quantum mechanical generators of the symmetry . for an ordinary internal symmetry , such as baryon number , the generator @xmath49 commutes with the hamiltonian @xmath50 , @xmath51 = 0.\ ] ] this is the definition of a symmetry . for supersymmetry , with generator @xmath52 , there is an additional relation ( see , for example , haag , lopuszansky , and sohnius , 1975 ) @xmath53 = 0 \qquad \{q , q\ } = h + g. \label{susy}\ ] ] in the second , anticommutation , relation , both the hamiltonian and various internal generators appear on the right - hand side . for clarity we have omitted the indices which would distinguish the various supersymmetry and internal symmetry generators and various associated constants , so this anticommutation relation is schematic . the anticommutation relation has the hamiltonian on the right - hand side , and as a result supersymmetry gives much more information about the dynamics than ordinary internal symmetries . to see one example of this ( witten and olive , 1978 ) , consider a one - particle state @xmath54 which has the special property that is invariant under part of the supersymmetry algebra ; in other words , @xmath55 for some of the @xmath52 s . this is known as a _ bogomolnyi - prasad - sommerfeld ( bps ) state_. take the expectation value of the anticommutator in eq . ( [ susy ] ) , @xmath56 for those @xmath52 s which annihilate @xmath57 , the left hand side is zero . the two terms on the right are just the mass of the particle and its @xmath49-charge . thus , the mass of any bps particle is determined entirely by its charge . this is a consequence of symmetry and does not depend on dynamics at all ; in particular it remains true even if the coupling is large . further analysis puts strong constraints on the interactions and on the phase diagram . for example , any bps state with zero charge has zero energy . the anticommutation relations imply that no other state can have lower energy , so any such state will be the ground state . a typical supersymmetric theory has many such states , which are characterized by the expectation values of some scalar fields . all these states must be degenerate . this is similar to spontaneous symmetry breaking , but with spontaneous symmetry breaking the degenerate vacua all have the same physics while here they are physically inequivalent they are not related to one another directly by any symmetry , rather the degeneracy follows indirectly from the bps argument . the scalar fields which label these vacua , known as moduli , must be massless for the same reason that goldstone bosons are massless . the low energy physics , the phase structure , is determined by the physics of the moduli , which is strongly constrained by supersymmetry . applying these methods to test the @xmath44 duality conjecture , one first finds that the bps mass formula is duality symmetric . that is , electrically charged bps states at coupling @xmath13 have the same masses as magnetically charged states at coupling @xmath29 . secondly , one can compare the degeneracies of bps states with different charges and spins . the degeneracy of magnetic monopoles can be determined by semiclassical methods when @xmath13 is small . in @xmath44 theories there can not be a phase transition as the coupling is varied and the supersymmetry algebra prevents the number of bps states from changing , so this also determines the degeneracy at strong coupling . for @xmath44 theories it is the same as that of the electrically charged states in the dual theory . as another test , the effective low energy physics of the moduli is duality invariant . this evidence was widely regarded as unconvincing , an accidental consequence of supersymmetry , until seiberg ( 1994 ) and others began to apply these methods systematically to determine the phase structures of theories with less supersymmetry , @xmath58 and @xmath59 . the arguments in these cases are more intricate , the physics is richer , and there are more consistency checks . many more examples of duality were found ( reviewed in seiberg , 1995 , and intriligator and seiberg , 1995 ) , as well as convincing evidence for various associated phenomena such as composite gauge bosons . , the duality holds at all energies . in many others it is holds only in the low energy theory . ] duality in supersymmetric gauge theory is now well - established . it is important to note , though , that the dual fields still have not been constructed . it is believed that to make duality manifest one may need a new formulation of these quantum field theories , something more intrinsically quantum - mechanical than the path integral over classical histories . it is also suspected that an understanding of the dualities in field theory will ultimately come from string theory . we will need one more idea in order to fully understand string duality . this is that spacetime may have more than four dimensions . the three spatial dimensions we see are expanding , and once were highly curved . it is a logical possibility that there are other dimensions which did not expand but remain small and highly curved . moreover , this is an attractive idea for a number of reasons . a good model for physics in such a spacetime is a waveguide , a cavity which has some finite cross section in the @xmath60-@xmath61 plane and is very long in the @xmath62 dimension . seen from far away , or with low resolution , this looks one - dimensional . an infinite number of different fields ( functions of @xmath62 and @xmath63 ) with different dispersion relations ( masses ) move along the waveguide . seen up close , the three - dimensional structure is evident and one sees that there is only one field , the electromagnetic field , and that the different dispersion relations come from modes with different @xmath60 , @xmath61 dependence . physics is much the same in a spacetime with three large spatial dimensions and additional ones which are small and compact . in nature , the additional dimensions would be quite small , close to the planck scale , so we would have seen only the very lowest modes . one reason this is attractive is that it unifies gravitational and gauge interactions . depending on whether its polarization is aligned along the long or compact directions , a higher - dimensional graviton can look like a graviton , photon , or scalar from the lower - dimensional point of view . this is the kaluza - klein mechanism . in addition , the dirac equation on such a space typically gives rise to multiple copies of the same set of quantum numbers ( that is , generations ) in the lower - dimensional spectrum . consistent weakly - coupled string theories necessarily live in ten spacetime dimensions . the origin of this condition is difficult to explain in simple terms , but it can be understood in various ways . calculation of quantum effects shows that they spoil essential symmetries unless the dimension is ten . also , the properties of fermions depend in an essential way on the number of dimensions and ten is special here for a number of reasons related to supersymmetry . for example , ten is the highest dimension in which the number of states of a massless vector ( eight : the spacetime dimension minus the timelike and longitudinal polarizations ) can be equal to the number of states of a massless fermion in higher dimensions the spinor representations are too large . the field equations of ten - dimensional string theory are consistent with four - dimensional physics that looks very much like what we have . some of the simplest solutions give rise to the same gauge groups and matter representations found in grand unification ( candelas , horowitz , strominger , and witten , 1985 ) . it is notable that the four - dimensional gauge couplings can be chiral , meaning that the right- and left - handed fermions have different gauge interactions . this was impossible for a number of previously considered unifying ideas including standard kaluza - klein theory ( witten , 1981 ) . for future reference it is useful to look at a simple example . this is a massless scalar field in five spacetime dimensions , with one dimension periodic with period @xmath64 . let @xmath46 with @xmath65 be the coordinates for the large dimensions , and @xmath66 be the periodic coordinate ; the index @xmath67 runs over all five values . the wave equation is @xmath68 as in the waveguide , separate variables and expand the @xmath66 dependence in a complete set of modes , @xmath69 then the @xmath17th mode satisfies @xmath70 which is the klein - gordon equation for a scalar field of mass @xmath71 . at low energies only the @xmath72 mode is detectable , but at energies above @xmath73 one sees a characteristic infinite tower of states . for this simple geometry the states are spaced evenly in mass . more complicated compact spaces would of course give a more complicated spectrum , but the average density of states depends only on the number of additional dimensions . from another point of view , consider how the spectrum behaves as @xmath74 : the infinite tower comes down in mass and forms the continuum characteristic of a noncompact dimension . string theory includes gauge theory . weak / strong duality in gauge theory is thus necessary if the same is to be true in string theory , but it is not sufficient . after the work of seiberg , it was still possible that duality was a property only of the low energy limit of string theory , not of the full massive spectrum . there were a number of early duality conjectures in string theory . a partial list includes the self - duality of @xmath58 heterotic string theories in four dimensions ( font , ibanez , lust , and quevedo , 1990 ) , self - duality of @xmath44 heterotic strings in four dimensions ( font , ibanez , lust , and quevedo , 1990 ; sen , 1994 ) , self - duality of the heterotic string in six dimensions ( duff,1995 ) , duality of certain string theories with theories of five - dimensional objects ( five - branes ) in ten dimensions ( duff , 1988 ; strominger , 1990 ) , and a relation not specifically duality of other string theories in ten dimensions to theories of membranes ( two - branes ) in eleven dimensions ( duff , howe , inami , and stelle , 1987 ) . these conjectures did not attract broad attention , due to their rather scattered nature and the limited evidence for any of them . the subject took a major step forward with the papers of hull and townsend ( 1995 ) , townsend ( 1995 ) , and witten ( 1995a ) . these authors proposed a nearly complete set of duals for all known string theories with at least @xmath44 supersymmetry . these new proposals elevated duality from a set of isolated conjectures to a general principle applying to any string theory in any dimension . the structure is quite rigid for each string theory the spectrum and symmetries determine a unique candidate for its dual , and the entire pattern fits together in an intricate way as dimensions are compactified and decompactified . some of the earlier conjectures , such as the @xmath44 conjecture , were incorporated whole , while others were incorporated in a modified or limited form ; we will return to the membrane idea in section 4.2 . these systematic conjectures made possible many new tests , and evidence for weak / strong duality in string theory accumulated rapidly . this led also to the discovery of various new dynamical ideas , some of which will be described in part 4 . there is now an integrated picture of the strongly coupled dynamics , at least for string theories with enough supersymmetry , and the methods are being extended to theories with less supersymmetry . figure 3 is a schematic picture of the space of string vacua . the parameters are the string coupling and the sizes and shapes of the compact dimensions ; in string theory these are not fixed but are determined by the expectation values of the moduli . over most of the space , the string coupling @xmath13 is of order 1 . in various limits , there is an effective description in terms of one or another weakly coupled string theory . the different string theories are characterized by the number and kind of supersymmetries and by the world - sheet topologies allowed ( oriented vs. unoriented and closed vs. open ) . in some limits new theories are encountered ; the most interesting of these , m - theory , will be described in section 4.1 . figure 3 is actually an oversimplification , in that the different limits are characterized not only by different string theories but by different topologies for the compact dimensions . it has been known for some time that even in weakly coupled string theory spacetime topology is not invariant ( aspinwall , greene , and morrison , 1993 ) , and string duality has provided many more examples ( greene , morrison , and strominger , 1995 ) . also , a more accurate picture would have many pieces resembling figure 3 , touching one another at points or along curves . in field theory the duality multiplets included the elementary quanta plus smooth classical configurations ( magnetic monopoles ) . the string duality multiplets include these ( though the elementary quanta are now loops rather than points ) , plus singular classical configurations ( black holes ) and also a new type of object unique to string theory , the d - brane . d - branes are topological defects on which the ends of a string can be trapped , as shown in figure 4 . they can be pointlike , one - dimensional , two - dimensional , and so on . these were discovered in the study of perturbative dualities of string theories , dualities that relate different weakly coupled theories ( dai , leigh , and polchinski , 1989 ; polchinski , chaudhuri , and johnson , 1996 ) . their relevance to weak / strong duality was noticed more recently ( polchinski , 1995 ) . let me illustrate one of the methods by which the strongly coupled limits are determined . in addition to the fundamental strings , various string theories have in their spectra one - dimensional objects which are either smooth solitons or d - branes . at weak coupling these are much heavier than the fundamental strings , but at strong coupling they are much lighter ( again this is guaranteed by the bps formula ) . in this limit it is natural to reinterpret the theory with the soliton or d - brane being the fundamental string . which string theory one gets depends on the degrees of freedom of the object , which can be determined by a weak - coupling calculation . this has been applied to determine , or confirm , the strongly coupled limits of various string theories ( sen , 1995a ; harvey and strominger , 1995 ; dabholkar , 1995 ; hull , 1995 ; schwarz , 1995 ; witten , 1996a ; polchinski and witten , 1996 ) . i have described the principal methods and general results of string duality . in this final part i would like to focus on some specific highlights , some of the discoveries that have most greatly changed our understanding . as described in section 3.4 , weakly coupled string theory requires specifically ten dimensions . one of the striking consequences of string duality is the existence of an eleventh dimension , not visible in weakly coupled string theory . how does one discover a new dimension ? the experimental signature was discussed in section 3.4 , a tower of new states above the energy @xmath73 . the theoretical signature is the same . a certain string theory , the iia theory in ten dimensions , has zero - dimensional d - branes , d - particles . these are heavy , with a mass @xmath75 at weak coupling ( @xmath76 is the string mass scale ) . for a pair of d - particles , the supersymmetric bound state problem can be solved and there is a single bound state with mass @xmath77 ( witten , 1996a ; sen , 1995b ) . similarly , @xmath17 d - particles have a single bound state of mass @xmath78 . this bound state spectrum was actually inferred from lower - dimensional duality symmetries before the explicit construction of the bound states . at strong coupling all of these become light , and in the @xmath79 limit form a continuum . we have seen just such a spectrum before , in section 3.4 . it is the signature of a new , eleventh , dimension with @xmath80 ( townsend , 1995 , witten , 1995a ) . this makes it clear why this dimension is invisible in string perturbation theory : the latter is an expansion in @xmath81 , so an expansion around the zero - radius limit . eleven dimensions is an interesting number . this is the maximum in which supersymmetry is possible beyond eleven the massless multiplets would contain spins higher than two , something which seems to be impossible in a consistent theory . eleven - dimensional supergravity is therefore the most symmetric theory based on supersymmetry ( cremmer , julia , and scherk , 1978 ) , and was considered as a possible unifying idea before string theory . the principal problems were that it is nonrenormalizable and that it is not possible to obtain a chiral spectrum . string theorists generally regarded eleven - dimensional supergravity as an irrelevant curiosity because of its difficulties and because strings seemed to live in ten , though some supergravity experts retained an interest in the theory because of its high degree of symmetry . now it develops that strings have an eleven - dimensional limit whose low energy physics is determined by symmetry considerations to be eleven - dimensional supergravity and whose short distance physics is not understood . this has been given the provisional name ` m - theory . ' the following exercise gives an interesting insight into m - theory . consider a long iia string in ten dimensions , shown in figure 5a . follow this state as the coupling is increased . we now know that a new orthogonal dimension will appear , but there are two possibilities for the form of the resulting state : in eleven dimensions it could look like a string as in figure 5b , or it could turn out to be a membrane wrapped around the eleventh dimension as shown in figure 5c . the answer can be determined in various ways from the conserved charges carries by the various objects , from the scaling of the string tension with coupling , from symmetry and it is figure 5c , the wrapped membrane . this is a puzzle . it seems to say that string theory is really the quantum mechanics of membranes , but in section 2.1 we have said that theories of fundamental membranes do not seem to exist . it is possible that the problems of quantizing membranes will be overcome and string theory will become membrane theory this was one of the origins of the name m - theory . it is likely , though , that the resolution lies in a different direction , that strings and membranes will turn out to be composites of something else . this has long been suspected in the case of string theory . some of the reasons for this are reviewed in polchinski ( 1994 ) . they include the general lack of success of string field theory and the fact that string perturbation theory diverges more rapidly than in field theory ( shenker , 1991 ) . string duality seems to give further support for this . in figure 3 , asymptotic _ behaviors in various limits are generated by different string theories , but there is no sign that the string descriptions do more than this . sums over string or membrane world - sheets are directly analogous to the feynman graph expansion and so are intrinsically perturbative . as for duality in field theory , we need a more quantum - mechanical description , one in which @xmath19 is not a free parameter ( witten , 1996c ) . incidentally , the membrane can also be oriented perpendicular to the eleventh dimension , so that in ten dimensions it still looks like a membrane . in string theory it appears as a d - brane . from the eleven - dimensional point of view one sees a symmetry in which a @xmath82 rotation takes the eleventh dimension into one of the others . from the string theory point of view this is a nonperturbative symmetry , mixing radii and couplings and relating strings to d - branes . in this section i would like to explain how all this business of extra dimensions might enter into an important piece of physics , the unification of the couplings . figure 6a shows the three gauge couplings and the dimensionless gravitational coupling @xmath83 as a function of energy . the logarithmic running of the gauge couplings is familiar ( amaldi , de boer , and furstenau , 1991 ) , as is the fact that they meet within experimental errors in the minimal supersymmetric model . it is also striking that the gravitational coupling nearly meets the other three , missing by a little over one order of magnitude out of fourteen.s and other factors reduces this to a factor of 20 ( kaplunovsky , 1988 ) . ] this is a near miss , less than ten percent in scale , but it is larger than the experimental error and so must be understood . there are many ways that the four couplings might turn out to unify . additional particles at an intermediate scale would change the running of the gauge couplings and can cause them to meet at a higher energy . enough extra states near the unification scale can have the same effect . or , the three gauge couplings may actually unify at the lower energy and a grand unified field theory with a single gauge coupling describe the physics from there until the string scale . all of these ideas modify the behavior of the gauge couplings . since the gauge couplings already meet , it would seem more economical to find instead a mechanism to change the running of the gravitational coupling . but this seems impossible because the running of the gauge coupling is just dimensional , as @xmath84 ; gravity is still classical at these scales . it is interesting to consider the effect of a new dimension below the unification scale . this changes the dimensional analysis . both the gauge and gravitational couplings grow more rapidly as shown in figure 6b , but they do not meet any sooner . in the @xmath85 heterotic string , the strong coupling limit leads to a new dimension which is slightly different from that considered before . instead of being periodic , a circle , it is a segment ( horava and witten , 1996 ) . the gauge fields and matter live at the endpoints only , while gravity propagates in the bulk . suppose that a fifth dimension of this type exists below the unification scale . that is , spacetime is a narrow five - dimensional layer bounded by four - dimensional walls . since the gauge fields and matter live in the walls , the evolution of the gauge couplings is the same as in four dimensions , but since gravity propagates in the full five dimensions the gravitational coupling behaves as in figure 6b . the result is shown in figure 6c : for a fifth dimension of appropriate size , the kink in the gravitational coupling makes it meet the others at the unification scale ( witten , 1996b ) . this is no more predictive than the idea of a unified theory : it adds one new parameter , the size of the fifth dimension . nevertheless it is an economical solution to the problem of the unification of all couplings , as likely to be true as any other . with only the data of the four couplings , there is not enough to distinguish among the possibilities . if supersymmetry is found and the masses of the superpartners measured with precision , many new renormalization group analyses become possible and it may be possible to determine the physics at the unification scale . local symmetries gauge invariance and coordinate invariance are a key ingredient in physics , appearing in the strong , weak , electromagnetic , and gravitational interactions . yet as has often been noted , these are not really symmetries in the usual sense . rather , they are _ redundancies _ : different vector potentials or coordinate systems describe the same physical state . why redundancy should be so important is puzzling , but it seems to be , and most attempts to unify are based on embedding the local symmetries at low energy into larger groups such as @xmath86 . in weakly coupled four dimensional theories , the size of the gauge group can only get larger at higher energies . there have been various arguments made that at strong coupling the reverse could be true , but convincing examples have only come with the understanding of strongly coupled dynamics in supersymmetric gauge and string theories . there are now many examples of composite gauge fields in the low energy theory , gauge fields not among the original short - distance fields ( intriligator and seiberg , 1995 ) . the interpretation of gauge symmetry is reversed : gauge symmetry is ubiquitous not because it is present in the underlying ultraviolet theory , but because it is infrared stable . that is , it is one of the few kinds of interesting long - distance dynamics which is _ natural _ , which does nt disappear when small changes are made in the short - distance parameters . in the early 1970 s it was found that classical black holes obey laws directly analogous to the laws of thermodynamics . this analogy was made sharper by hawking s discovery ( 1975 ) that black holes radiate as black bodies at the corresponding temperature . under this analogy , the entropy of a black hole ( the bekenstein - hawking entropy ) is the area of its horizon divided by @xmath87 . it has long been a goal to find a statistical mechanical theory associated with this thermodynamics , and in particular to associate the entropy with the density of states of the black hole . many arguments ( too many to review here ) have been put forward in this direction . while it may develop that some of the principles behind these are correct , until recently there was no example where the states of a black hole could be counted in a controlled way . the idea is to count supersymmetric ( bps ) black hole states . such black holes always carry gauge charges and have the maximum allowed charge to mass ratio they are extremal . as discussed in section 2.2 , the number of bps states can not change as we vary the string coupling constant . as the coupling is reduced and so the gravitational interaction weakened , certain black holes will at sufficiently weak coupling no longer look like black holes . rather , they look like a collection of d - particles as depicted in figure 7 ; this has the same charge - to - mass ratio as the black hole . the d - particles are a weakly coupled quantum system whose spectrum is explicitly known from d - brane methods . the density of bps states is just that given by the bekenstein - hawking entropy ( strominger and vafa , 1996 , and much subsequent work ; for a review see horowitz , 1996 ) . closely associated with all this is the black hole information paradox . a black hole of given mass and charge can be formed in a very large number of ways . it will then evaporate , and the final state is black body radiation that does not depend on how the black hole formed . thus , many initial states evolve to a single final state . this violates the usual laws of quantum mechanics . there are various schools of thought here ( reviewed in page , 1993 ) . the proposal of hawking ( 1976 ) is that this is just the way things are : the laws of quantum mechanics need to be changed . there is also strong opposition to this view . the problem is not that theorists believe that the laws of quantum mechanics are in their final form they may be , though most of us would like to see a less mysterious structure . rather , it is that the specific modification required here , the replacement of wavefunctions with density matrices , seems ugly and very possibly inconsistent . the principal alternative , that the initial state is encoded in subtle correlations in the hawking radiation , sounds plausible but in fact is even more radical . the problem is that hawking radiation emerges from the region of the horizon , where the geometry is smooth and so ordinary low energy field theory should be valid . one can follow the hawking radiation and see correlations develop between the fields inside and outside the black hole ; the superposition principle then forbids the necessary correlations to exist strictly among the fields outside . to evade this requires that the locality principle in quantum field theory break down in some long - ranged but subtle way . there is a proposal for how this might occur in string theory ( susskind , 1993 , 1995 ; lowe , polchinski , susskind , thorlacius , and uglum , 1995 ) , but it is controversial . the recent progress in string duality suggests that black holes do obey the ordinary rules of quantum mechanics . the multiplets include black holes along with various nonsingular states , and in figure 7 we have continuously deformed a black hole into a system which obeys ordinary quantum mechanics , at least to high accuracy . this is certainly not decisive , however . the problem is that dynamical properties are not like the counting of bps states ; they are not invariant under changes in parameters . it could be that as the coupling constant is increased , a critical coupling is reached where the d - particles collapse into a black hole . at this coupling there could be a discontinuous change ( or a smooth crossover ) from ordinary quantum behavior to information loss . it is an open question whether d - branes will give insight into the dynamical behavior of black holes ; there are hints that it may be so . so when will string theory make sharp predictions ? probably not until the vacuum is understood , as we discussed in section 2.3 . an important problem remains in our understanding of the vacuum , when we try to calculate the vacuum energy density , the cosmological constant . the vacuum is a complicated place , and there are many different contributions to the energy density the zero point energies of the quantum fields , the potential energy of the higgs field , the energy of the strongly fluctuating qcd fields . the total of the separate densities is at least @xmath88 gev@xmath89 , and there is no reason known that they should cancel to any degree of accuracy . yet the experimental bound on the cosmological constant is 55 orders of magnitude smaller than this ( see weinberg , 1989 , for a review ) . this is just as much problem in quantum field theory as in string theory but field theory can still make precise predictions because it is less ambitious . one can ignore gravity , or add a free parameter and adjust it to cancel the other contributions to the cosmological constant . in string theory , neither of these is possible . does string duality help here ? possibly . it is an interesting fact that in supersymmetric theories the energy density can cancel naturally . for example the zero point energies of bosons and their fermionic partners are equal and opposite . one can see this in the discussion of bps states : a supersymmetric vacuum has zero energy ( though the inclusion of supergravity makes things a bit more subtle ) . in nature , though , supersymmetry must be spontaneously broken there is no charged boson degenerate with the electron and spontaneous breaking spoils the cancellation of the cosmological constant . we need some new phase of supersymmetric theories , in which the boson / fermion degeneracy is removed while the vacuum energy remains zero . witten ( 1995b ) has suggested how such a phase might appear in strongly coupled string theory . from another point of view , the cosmological constant and black hole information problems both seem to require subtle nonlocalities in spacetime . for the information problem we have discussed this ; for the cosmological constant , the point is that the low energy value of this constant must somehow feed back into the short - distance physics that determines it . for example , it was argued that spacetime wormholes could accomplish this ( coleman , 1988 ) . this proposal was the subject of intense and skeptical scrutiny , but the subject ground to a halt because a nonperturbative formulation of quantum gravity was lacking . it may then be that it will return , possibly in some transmuted form , when we know what string theory is . in closing , let me say that there is a sense among those working in string theory that we are dealing with a unique and remarkable structure , one which has many points of contact with the physics we know and with earlier attempts to unify it quantum mechanics , gravity , gauge symmetry , chirality , grand unification , supersymmetry , and kaluza - klein theory . there is also a sense that we are discovering this structure , not inventing it . the distinction is illustrated , for example , by the question of the nature of spacetime . there have been many previous attempts to modify spacetime in a way that would look the same at long distance but would solve the problems of quantum gravity . this approach has often been sterile it is not a problem that seems to yield to direct effort . string theory , on the other hand , starts with a flat spacetime . one discovers first that the spacetime is dynamical , that the theory contains gravity . later one finds , as mentioned in section 3.5 , that spacetime topology is not a physical invariant but can change in specific and controlled ways . there is still more to be learned . in perturbation theory it seems that the shortest sensible distance is the string length scale , the typical radius of zero point vibrations of the string ( gross and mende , 1987 ; amati , ciafaloni , and veneziano , 1989 ; reviewed in witten , 1996c ) . this works out to be the planck length divided by a power of the string coupling . there is evidence for a somewhat shorter scale in the exact theory ( shenker , 1995 ) , and that d - branes may probe it ( bachas , 1995 ; danielsson , ferretti and sundborg , 1995 ; kabat and pouliot , 1995 ) . moreover , there is a sense in which the spacetime coordinates for d - branes are elevated from numbers to matrices ( witten , 1996a ) ; only at low energy the matrices are diagonal and an ordinary spacetime picture holds . it may turn that this is a curiosity , or it may signal a new uncertainty principle relating to a minimum distance . exploring the connection between d - branes and black holes is a likely way to learn which . in effect , string theory is smarter than we are . it knows what spacetime is , and we do nt , and we have to figure out how to ask it . we are very fortunate that this remarkable structure exists and that it seems to be within our power to understand it . i would like to thank gary horowitz , clifford johnson , and andy strominger for their suggestions on the manuscript . this work is supported by nsf grants phy91 - 16964 and phy94 - 07194 . u. amaldi , w. de boer , and h. furstenau , 1991 , `` comparison of grand unified theories with electroweak and strong coupling constants measured at lep , '' phys . lett . * b260 , * 447 . + d. amati , m. ciafaloni , and g. veneziano , 1989 , `` can spacetime be probed below the string scale ? '' phys . lett . * b216 , * 41 . + p. s. aspinwall , b. r. greene , and d. r. morrison , 1993 , `` multiple mirror manifolds and topology change in string theory , '' phys . lett . * b303 , * 249 , e - print hep - th/9301043 . + c. bachas , 1996 , `` d - brane dynamics , '' phys . * b374 , * 37 , e - print hep - th/9511043 . + e. bergshoeff , e. sezgin , and p.k . townsend , 1987 , `` supermembranes and eleven dimensional supergravity , '' phys . lett . * b189 , * 75 . + p. candelas , g. horowitz , a. strominger , and e. witten , 1985 , `` vacuum configurations for superstrings , '' nucl . phys . * b258 , * 46 . + b. carter , 1979 , `` the general theory of the mechanical , electromagnetic , and thermodynamic properties of black holes , '' in _ general relativity . an einstein centennial survey , _ ed . s. w. hawking and w. israel ( cambridge , cambridge ) . + s. coleman , 1988 , `` why there is nothing rather than something : a theory of the cosmological constant , '' nucl . phys . * b310 , * 643 . + e. d. commins and p. h. bucksbaum , 1983 , _ weak interactions of leptons and quarks _ ( cambridge , new york ) . + e. cremmer , b. julia , and j. scherk , 1978 , `` supergravity theory in eleven dimensions , '' phys . lett . * 76b , * 409 . + a. dabholkar , 1995 , `` ten - dimensional heterotic string as a soliton , '' phys . * b357 , * 307 , e - print hep - th/9506160 . + j. dai , r. g. leigh , and j. polchinski , 1989 , `` new connections between string theories , '' mod . . lett . * a4 , * 2073 . + u. h. danielsson , g. ferretti , and b. sundborg , 1996 , `` d - particle dynamics and bound states , '' preprint usitp-96 - 03 , e - print hep - th/9603081 . + p. a. m. dirac , 1931 , `` quantized singularities in the electromagnetic field , '' proc . * a133 , * 60 . + m. j. duff , p.s . howe , t. inami , and k.s . stelle , 1987 , `` superstrings in @xmath90 from supermembranes in @xmath91 , '' phys * 191b , * 70 . + m. j. duff , 1988 , `` supermembranes : the first fifteen weeks , '' class . * 5 , * 189 . + m. j. duff , 1995 , `` strong / weak duality from the dual string , '' nucl . * b442 , * 47 , e - print hep - th/9506057 . + a. font , l. ibanez , d. lust , and f. quevedo , 1990 , `` strong - weak coupling duality and nonperturbative effects in string theory , '' phys . * b249 , * 35 . + f. gliozzi , j. scherk , and d. olive , 1977 , `` supersymmetry , supergravity theories , and the dual spinor model , '' nucl . phys . * b122 , * 253 ; reprinted in schwarz , 1985 , vol . 1 , 300 . + m. b. green and j. h. schwarz , 1984 , `` anomaly cancellations in supersymmetric @xmath90 gauge theory and superstring theory , '' phys . lett . * 136b , * 367 ; reprinted in schwarz , 1985 , vol . 2 , 1007 . + m. b. green , j. h. schwarz , and e. witten , 1987 , _ superstring theory , vols . 1 and 2 _ ( cambridge , cambridge ) . + b. r. greene , d. r. morrison , and a. strominger , 1995 , `` black hole condensation and the unification of string vacua , '' nucl . * b451 , * 109 , e - print hep - th/9504145 . + d. j. gross , j. a. harvey , e. martinec , and r. rohm , 1985 , `` heterotic string , '' phys . * 54 , * 502 . + r. haag , j. t. lopuszanski , and m sohnius , 1975 , `` all possible generators of supersymmetries of the s matrix , '' nucl . phys . * b88 , * 257 . + j. a. harvey and a. strominger , 1995 , `` the heterotic string is a soliton , '' nucl . * b449 , * 535 , ( e ) * b458 , * 456 , e - print hep - th/9504047 . + j. a. harvey , 1996 , `` magnetic monopoles , duality , and supersymmetry , '' chicago preprint efi-96 - 06 , e - print hep - th/9603086 . + s. w. hawking , 1975 , `` particle creation by black holes , '' comm . * 43 , * 199 . + s. w. hawking , 1976 , `` breakdown of predictability in gravitational collapse , '' phys . rev . * d14 , * 2460 . + p. horava and e. witten , 1996 , `` heterotic and type i string dynamics from eleven dimensions , '' nucl . * b460 , * 506 , e - print hep - th/9510209 . + g. t. horowitz , 1996 , `` the origin of black hole entropy in string theory , '' ucsb preprint ucsbth-96 - 07 , e - print gr - qc/9604051 . + c. m. hull and p. k. townsend , 1995 , `` unity of superstring dualities , '' nucl . * b438 , * 109 , e - print hep - th/9410167 . + c. m. hull , 1995 , `` string - string duality in ten - dimensions , '' phys . * b357 , * 545 , e - print hep - th/9506194 . + k. intriligator and n. seiberg , 1995 , `` lectures on supersymmetric gauge theories and electric - magnetic duality , '' presented at tasi 95 , boulder , colo . , trieste summer school , trieste , italy , and cargese summer school , cargese , france , rutgers preprint ru-95 - 48 , e - print hep - th/9509066 . + d. kabat and p. pouliot , `` a comment on zero - brane quantum mechanics , '' rutgers preprint ru-96 - 17 , e - print hep - th/9603127 . + v. s. kaplunovsky , 1988 , `` one loop effects in string unification , '' nucl . phys . * b307 , * 145 , ( e ) * b382 , * 436 ( 1992 ) , e - print hep - th/9205070 . + d. a. lowe , j. polchinski , l. susskind , l. thorlacius , and j. uglum , 1995 , `` black hole complementarity versus locality , '' phys . rev . * d52 , * 6997 , e - print hep - th/9506138 . + c. montonen and d. olive , 1977 , `` magnetic monopoles as gauge particles ? '' phys . * 72b , * 117 . + a. neveu and j. scherk , 1972 , `` connection between yang - mills fields and dual models , '' nucl . phys . * b36 , * 155 . + h. osborn , 1979 , `` topological charges for @xmath44 supersymmetric gauge theories and monopoles of spin 1 , '' phys . * 83b , * 321 . + d. n. page , 1993 , `` black hole information , '' canadian gen . rel . * 1993 , * 1 , e - print hep - th/9305040 . + j. polchinski , 1994 , `` what is string theory ? '' les houches summer school , session 62 : fluctuating geometries in statistical mechanics and field theory , 1994 , e - print hep - th/9411028 + j. polchinski , 1995 , `` dirichlet - branes and ramond - ramond charges , '' phys . * 75 , * 4724 , e - print hep - th/9510017 . + j. polchinski and e. witten , 1996 , `` evidence for heterotic - type i string duality , '' nucl . phys . * b460 , * 525 , e - print hep - th/9510169 . + j. polchinski , s. chaudhuri , and c. v. johnson , 1996 , `` notes on d - branes , '' itp preprint nsf - itp-96 - 003 , e - print hep - th/9602052 . + p. ramond , 1971 , `` dual theory for free fermions , '' phys . * d3 , * 2415 ; reprinted in schwarz , 1985 , vol . 1 , 53 . + a. salam , 1980 , `` grand unification of fundamental forces , '' rev . * 52 , * 525 . + j. scherk and j. h. schwarz , 1974 , `` dual models for non - hadrons , '' nucl . * b81 , * 118 ; reprinted in schwarz , 1985 , vol . 1 , 191 . + j. scherk and j. h. schwarz , 1975 , `` dual model approach to a renormalizable theory of gravitation , '' caltech preprint calt-58 - 488 ; reprinted in schwarz , 1985 , vol . 1 , 218 . + j. h. schwarz , 1985 , _ superstrings . the first 15 years of superstring theory , vols . 1 and 2 _ ( world scientific , singapore ) . + j. h. schwarz , 1995 , `` an @xmath92 multiplet of type iib superstrings , '' phys . * b360 , * 13 , ( e ) * b364 , * 252 , e - print hep - th/9508143 . + n. seiberg , 1994 , `` exact results on the space of vacua of four - dimensional susy gauge theories , '' phys * d49 , * 6857 , e - print hep - th/9402044 . + n. seiberg , 1995 , `` the power of duality : exact results in 4-d susy field theory , '' rutgers preprint ru-95 - 37 , e - print hep - th/9506077 . + a. sen , 1994 , `` strong - weak coupling duality in four - dimensional string theory , '' int . * a9 , * 3707 , e - print hep - th/9402002 . + a. sen , 1995a , `` string string duality conjecture in six - dimensions and charged solitonic strings , '' nucl . * b450 , * 103 , e - print hep - th/9504027 . + a. sen , 1995b , `` a note on marginally stable bound states in type ii string theory , '' mehta preprint mri - phy-23 - 95 , e - print hep - th/9510229 . + s. h. shenker , 1991 , `` the strength of nonperturbative effects in string theory , '' in _ random surfaces and quantum gravity _ , eds . o. alvarez , e. marinari , and p. windey ( plenum , 1991 ) . + s. h. shenker , 1995 , `` another length scale in string theory ? '' rutgers preprint ru-95 - 53 , e - print hep - th/9509132 . + a. strominger , 1990 , `` heterotic solitons , '' nucl . * b343 , * 167 , ( e ) * b353 , * 565 . + a. strominger and c. vafa , 1996 , `` microscopic origin of the bekenstein - hawking entropy , '' preprint hutp-96-a002 , e - print hep - th/9601029 . + l. susskind , 1993 , `` string theory and the principles of black hole complementarity , '' phys . * 71 , * 2367 , e - print hep - th/9307168 + l. susskind , 1995 , `` the world as a hologram , '' j. math . phys . * 36 , * 6377 , e - print hep - th/9409089 . + p. k. townsend , 1995 , `` the eleven - dimensional supermembrane revisited , '' phys . * b350 , * 184 , e - print hep - th/9501068 . + s. weinberg , 1980 , `` conceptual foundations of the unified theory of weak and electromagnetic interactions , '' rev . * 52 , * 515 . + s. weinberg , 1989 , `` the cosmological constant problem , '' rev . * 61 , * 1 . + k. wilson , 1983 , `` the renormalization group and critical phenomena , '' rev . mod . * 55 , * 515 . + e. witten and d. olive , 1978 , `` supersymmetry algebras that include topological charges , '' phys . lett . * 78b , * 97 . + e. witten , 1981 , `` search for a realistic kaluza - klein theory , '' nucl . * b186 , * 412 . + e. witten , 1995a , `` string theory dynamics in various dimensions , '' nucl . * b443 , * 85 , e - print hep - th/9503124 . + e. witten , 1995b , `` strong coupling and the cosmological constant , '' mod . * a10 , * 2153 , e - print hep - th/9506101 . + e. witten , 1996a , `` bound states of strings and @xmath93-branes , '' nucl . * b460 , * 335 , e - print hep - th/9510135 . + e. witten , 1996b , `` strong - coupling expansion of calabi - yau compactification , '' ias preprint iassns - hep-96 - 08 , e - print hep - th/9602070 . + e. witten , 1996c , `` reflections on the fate of spacetime , '' phys . today * 49 , * 4 , 5 .

the formation of a quark - gluon plasma ( qgp ) is achieved in ultra - relativistic heavy - ion collision experiments at relativistic heavy - ion collider ( rhic ) in bnl and at the large hadron collider ( lhc ) in cern . more and more information on properties of strongly interacting quark - gluon matter in quantum chromodynamics ( qcd ) is being accumulated . however , despite a number of interesting discoveries and new notions established in experiments , the thermalization mechanism that leads to the qgp still lacks a deep understanding based on qcd first principles . there have been many attempts to estimate the thermalization time scale , that is phenomenologically known to be @xmath0 from analysis in hydrodynamic models @xcite . it was argued @xcite that perturbative qcd processes lead to a thermalization time of order @xmath1 where the saturation momentum @xmath2 is a characteristic hard scale in the high - energy reactions ; @xmath3-@xmath4 in the typical rhic kinematics . if the prefactor is of order unity , this estimate would yield a thermalization time around several @xmath5 , which is too slow to account for the rhic observations . it has also been suggested that qcd perturbation theory , no matter at which order one truncates it , may be unable to describe thermalization at all @xcite . plasma instabilities , particularly the weibel instability @xcite , could be promising candidates for the early thermalization mechanism @xcite and have been well investigated in the hard - loop formalism @xcite . there are also various other approaches @xcite including a framework based on the ads / cft correspondence @xcite , in which thermalization appears as the formation of a black hole horizon in the dual theory . parallel to the analysis on plasma instabilities , an important observation was made in a related theoretical framework . the occupation number of gluons at high energy is so huge that color field amplitudes @xmath6 can be realized , and then non - linearities must be taken into account to all orders . instead of a plain perturbative expansion , then , an approximation in terms of classical field equations becomes more suitable because it effectively makes a resummation of large - amplitude fields @xmath7 . this non - linear regime , whose natural time scale is of order @xmath8 at rhic , is expected to be prevalent during the early stages in heavy - ion collisions . the qcd framework known as the color glass condensate ( cgc ) @xcite rearranges the perturbative expansion in terms of classical fields ( that correspond to the resummation of infinite sets of graphs ) , and perturbations on top of these classical fields are of higher order in @xmath9 . the cgc was first developed in order to deal with the saturation effect in the gluon distribution function @xcite . because we are interested in the early time ( @xmath10 ) dynamics before thermalization , the cgc description seems to be the most suited for our purpose . the initial condition just above the forward light - cone ( i.e. at @xmath11 ) for the classical field that appears at lowest order in the cgc description can be nicely formulated in the bjorken ( expanding ) coordinates @xcite . because in the regime of large occupation numbers , classical field theory and kinetic theory essentially describe the same dynamics @xcite , it is conceivable that the cgc can describe the isotropization and thermalization processes qualitatively . however , for this program to have a chance to work , it is necessary to go beyond the lowest order description , since at zeroth order it is well known that the pressure tensor never isotropizes . it is unfortunately impossible to obtain an analytical solution of the classical yang - mills equations of motion for the heavy - ion collision ( two - source ) problem , and therefore one has to resort to numerical simulations . in refs . @xcite , heavy - ion collisions were simulated in the mclerran - venugopalan ( mv ) model ( i.e. the cgc at lowest order , with a local gaussian distribution of color sources ) , formulated gauge - invariantly in terms of link variables . these numerical calculations were later confirmed by an independent simulation in ref . the peculiar feature in the cgc description at leading order is that the initial condition for the classical fields is independent of the space - time rapidity variable @xmath12 , and the equations of motion maintain @xmath12-independence during the time evolution . therefore , the space - time evolution is entirely independent of @xmath12 , that is , invariant under boosts in the direction of the collision axis . intuitively , one can expect a longitudinally extended structure of the initial fields , which is indeed a characteristic property of the so - called `` glasma '' state @xcite , which could be a source for the so - called ridge structure seen in the two - particle correlations at rhic and lhc @xcite . within the cgc framework , an instability has been found when the boost invariant glasma fields are disturbed @xcite . this instability has some similarities to the weibel instability in anisotropic plasmas , but the origin of the instability is not yet fully clarified , though several interpretations have been proposed @xcite . also , it is still an issue how to properly formulate disturbances to the boost invariant cgc background . however , it is clear that such perturbations arise when one considers higher order corrections to the leading cgc picture @xcite , and therefore they should not be ignored since they can alter phenomena such as particle production @xcite , etc . moreover , without including these perturbations , one can not quantify the thermalization processes . one of the reasons why the origin of the glasma instability lacks a clear understanding is that the available numerical data from earlier works @xcite is rather limited . this motivates us to carry out a more systematic survey of the glasma instability , exploiting a wider range of simulation parameters . in this work we will specifically investigate the dependence of the glasma instability on the following parameters : * [ i ] * _ transverse grid size @xmath13 : _ it is known that the initial energy density at @xmath11 is sensitive to the ultraviolet and infrared regulators in the transverse direction , and this sensitivity becomes milder at finite @xmath14 @xcite . it is therefore important to check whether continuum - limit results can be obtained reliably by looking systematically at the dependence on the transverse lattice spacing @xmath15 , where @xmath16 is the length of the box and @xmath13 the number of lattice points in the transverse direction . * [ ii ] * _ longitudinal grid size @xmath17 : _ physical results should also not be contaminated by ultraviolet singularities in the longitudinal direction . to avoid this , we will use initial fluctuations with sufficiently small longitudinal wave - number @xmath18 ( @xmath18 is the fourier conjugate to the rapidity @xmath12 ) . we will confirm that there is no dependence on the longitudinal grid size @xmath17 or the longitudinal lattice spacing @xmath19 , at least until the longitudinal spectrum extends into the ultraviolet regions at very late time . * [ iii ] * _ magnitude @xmath20 of the seeds : _ in general the strength of the instability increases with the magnitude @xmath20 of the initial @xmath12-dependent perturbations . this dependence is found to be very simple until the unstable modes become large enough to be self - interacting : at moderate times , the finite-@xmath18 modes obey the linearized dynamics and their amplitudes grow linearly as @xmath20 increases . * [ iv ] * _ initial wave - number @xmath21 : _ the time evolution depends on the initial condition in the longitudinal direction , and in particular on the wave - number @xmath21 of the seeds . we will show that the instability produced by the smallest wave - number is defined to be an integer between @xmath22 and @xmath23 . the lowest non - zero wave - number is thus @xmath24 . ] ( @xmath25 ) is the fastest and strongest in a wide range of times . we will also confirm the dominance of linearity and the appreciable presence of residual non - linear effects by verifying that the result from an initial condition with multiple @xmath21 s can be approximated as the superposition of the results from initial conditions with single wave - numbers @xmath21 s . in this paper , we report on numerical results about these dependences for the growing behavior at the late stage of the glasma evolution . these numerical results provide a pool of observations that we hope will help to understand the microscopic mechanisms that drive the glasma instabilities . in the section [ sec : formulas ] , we recall the most important formulas for a lattice study of this problem . in the section [ sec : bi ] , we briefly expose some results in the purely boost - invariant case . these are well - known results already , and merely serve as a check of our implementation . in the section [ sec : notbi ] , we describe what happens when the glasma fields are perturbed by a rapidity - dependent seed , and we investigate thoroughly how the result depends on the parameters listed above . we explore the time evolution of the @xmath18-spectrum in the section [ sec : lspectrum ] . finally , the section [ sec : summary ] is devoted to a summary and outlook . in this section , we summarize our equations , with emphasis on the lattice specificities . the main goal of this section is to avoid confusions due to the existence of different conventions in the literature for some quantities . the link variables are defined as @xmath26 with the gauge potential @xmath27 that connects the neighboring sites from @xmath28 and @xmath29 . this expression for the link variable is the most important one in order to establish the correspondence between the continuum and the lattice ( link - variable ) formulations . under a gauge transformation by @xmath30 the link variables transform as @xmath31 and the chromo - electric fields as @xmath32 the plaquette variable is then defined as @xmath33 \ ; , \label{eq : approxf}\ ] ] where the last form is an approximation which gives rise to the field strength @xmath34 $ ] of the continuum formulation . this is a very useful expression to study the continuum limit . the lattice spacing should be the longitudinal one , i.e. @xmath35 , which is distinct from the transverse one @xmath36 , if the longitudinal direction is involved in the displacements @xmath37 or @xmath38 in the above formula . the canonical momenta in terms of link variables are expressed as @xmath39 here we note that in the above we already use variables made dimensionless by the transverse lattice spacing @xmath36 , and thus @xmath36 will never appear explicitly in our formula . because @xmath12 is a dimensionless number , @xmath35 does not bring any mass dimension unlike @xmath36 . we can also discretize the time in the above equations in order to solve the time evolution numerically for @xmath40 and @xmath41 ; @xmath42 u_i(\tau ) \;,\\ u_\eta(\tau '' ) & = \exp\bigl [ -2{\delta\tau}\cdot { \mathrm{i}}g a_\eta \tau ' e^\eta(\tau ' ) \bigr ] u_\eta(\tau ) \;,\end{aligned}\ ] ] where @xmath43 and @xmath44 . note that we have not written explicitly the position arguments in the quantities that appear in these equations . here , the exponentiation of @xmath45 is important ; otherwise the updated @xmath46 would not be a unitary matrix . the equations of motion are discretized in the same manner as @xmath47_\tau \\ & \qquad + 2{\delta\tau}\frac{{\mathrm{i}}\tau}{2g}\ , \sum_{j\neq i } \bigl [ u_{ji}(x)+u_{-ji}(x ) - \text{(h.c . ) } \bigr]_\tau \ ; , \end{split}\ ] ] for the transverse components and @xmath48_\tau \;,\ ] ] for the longitudinal component . we recall again that all variables including @xmath14 and @xmath49 above are dimensionless , and expressed in unit of the transverse spacing @xmath36 . one can make sure that these equations are equivalent to the ordinary equations of motion in the continuum limit , by using the approximate form . the initial conditions are derived from the requirement of avoiding the singularity at the collision point . it is written in terms of @xmath50 , the classical solution of the mv model for a single color source ( the label @xmath51 indicates which of the two nuclei produces the corresponding field ) . this single - source classical solution is a transverse pure - gauge , @xmath52 where the gauge rotation matrix @xmath53 can be written as @xmath54 \;. \label{eq : wilson}\ ] ] here , @xmath55 is the solution of the poisson equation , @xmath56 . in the mv model , the random color source @xmath57 is gaussian distributed , with a two - point correlation @xmath58 higher - point correlation functions are obtained as products of two - point contractions , via wick s theorem . in the mv model , @xmath59 is the only dimensionful parameter and it is related to the saturation momentum @xmath2 . then , the initial conditions are @xcite ; @xmath60 \ ; , \label{eq : initial_e}\end{aligned}\ ] ] written here in the slightly simpler case of the su(2 ) color group . these discretized equations of motion and initial conditions completely define the numerical mv model . our conventions are mostly consistent with those of previous works , except for the fact that the field @xmath61 is not treated as an adjoint scalar - field variable but instead we also describe it in terms of a link variable @xmath62 . we note that our formulation is reduced to the conventional one with the adjoint scalar - field variable if we take @xmath35 to be sufficiently small . in addition , since our interest lies primarily in the qualitative behavior of the glasma instability , we disregard the color neutralization that has been imposed onto the mv model color sources in other works @xcite . it is now straightforward to implement the numerical calculation for gauge - invariant observables such as the components of the energy - momentum tensor . they are in the continuum convention written down as @xmath63\bigr\rangle \ ; , \label{eq : e } \\ & p_{_t } = \frac{1}{2}\bigl\langle t^{xx } + t^{yy } \bigr\rangle = \bigl\langle { \mathrm{tr}}\bigl [ e_{_l}^2 + b_{_l}^2 \bigr ] \bigr\rangle \ ; , \label{eq : p_{_t } } \\ & p_{_l } = \bigl\langle \tau^2 t^{\eta\eta } \bigr\rangle = \bigl\langle{\mathrm{tr}}\bigl [ e_{_t}^2 + b_{_t}^2 - e_{_l}^2 - b_{_l}^2 \bigr]\bigr\rangle \ ; , \label{eq : p_{_l}}\end{aligned}\ ] ] where we define @xmath64 these formulas are implicitly summed over the color index @xmath36 , for @xmath65 and @xmath66 to be gauge invariant . here , the chromo - magnetic field squared should be expressed in a way consistent with eq . , that is , @xmath67 these definitions are , however , not very convenient when we decompose the energy into its fourier components , as we will see later . we confirm that the energy - momentum tensor is traceless ; @xmath68 . it should also be mentioned that our definition of @xmath69 is different from that in refs . @xcite by a factor @xmath70 so that @xmath71 for an isotropic system . we first check the consistency with the known results in the boost - invariant case , i.e. in @xmath72 dimensions . this is useful also for later discussions on the spectral decomposition when we consider @xmath12-dependent fluctuations . in this section , we focus on @xmath12-independent classical solutions , with initial conditions specified in eqs . and that are themselves independent of @xmath12 . eventually , however , our goal is to superimpose @xmath12-dependent fluctuations to these boost invariant initial conditions in order to incorporate quantum effects @xcite . we note here that our numerical calculations are limited to the su(2 ) color group , so that we can get more samples and thus better statistics for the same computational cost . as discussed in ref . @xcite , the su(2 ) calculations capture the essential properties of the glasma instability , and we expect that the results we report here for the su(2 ) case would be qualitatively the same for su(3 ) . let us first take a look at a typical fixed initial configuration , without taking the ensemble average over the color sources , in order to develop an intuitive picture for the glasma state . in the mv model , the initial color distribution in the transverse plane is akin to white noise , having no correlation between different transverse sites . ] . since the solution of the poisson equation @xmath73 involves a convolution with a massless ( i.e. long ranged ) transverse propagator , @xmath74 has a smooth structure as shown in the left panel of fig . [ fig : lam ] . . the mv - model parameters are chosen as @xmath75 ( number of transverse sites ) and @xmath76 . left : the solution @xmath74 of the poisson equation . only the first color component @xmath77 is plotted . right : corresponding gauge field @xmath78 calculated from @xmath79 $ ] . only the first color component @xmath80 is plotted.,title="fig:",scaledwidth=49.0% ] . the mv - model parameters are chosen as @xmath75 ( number of transverse sites ) and @xmath76 . left : the solution @xmath74 of the poisson equation . only the first color component @xmath77 is plotted . right : corresponding gauge field @xmath78 calculated from @xmath79 $ ] . only the first color component @xmath80 is plotted.,title="fig:",scaledwidth=49.0% ] although @xmath74 is fairly smooth due to the regularizing effect of this convolution , the corresponding gauge field @xmath81 , which is a phase of @xmath82 defined in eq . , is far from smooth . this is actually a consequence of the pure - gauge form ; contributions from globally smooth parts in @xmath74 cancel each other and the resulting @xmath81 fluctuates strongly . in other words , @xmath81 has a rough structure because a derivative of @xmath74 undoes the regularization provided by the convolution . the @xmath81 shown in the figure [ fig : lam ] is the gauge field from a single source only . in a collision , the initial fields such as @xmath83 and @xmath84 involve a product of two @xmath81 s from two independent sources , and thus have an even rougher structure . it is worth mentioning that the `` color - flux tube '' structures , whose transverse size is expected to be of order @xmath85 , do not emerge manifestly from the mv model since the source distribution has no correlation length in the transverse direction . this means that the nielsen - olesen instability @xcite is unlikely to occur at least within the mv model . nevertheless , we expect that color - flux tube structures would appear after the jimwlk evolution from the mv - model distribution . in the mean - field approximation , in fact , the transverse correlation function of color source has been already studied in ref . @xcite , which should be taken into account in the future works . also , to make a clear comparison with the weibel instability and the hard - loop estimates ( especially early works @xcite ) , it would be useful to simplify the mv - model setup in ( 1 + 1 ) dimensions without transverse dependences in such a way not to lose the qualitative features of the glasma instability that we will see in the section [ sec : notbi ] . in this sense the results reported in the present paper should serve as a foundation for future investigations in these directions . . an ensemble average is taken over @xmath86 random configurations of @xmath87 for @xmath88 and @xmath89 , and over @xmath90 configurations for @xmath91 , @xmath92 , @xmath93 and @xmath94 . the mv model parameter is chosen as @xmath76 . right : chromo - electric and chromo - magnetic fields and for the same mv model parameters and @xmath88.,title="fig:",scaledwidth=49.0% ] . an ensemble average is taken over @xmath86 random configurations of @xmath87 for @xmath88 and @xmath89 , and over @xmath90 configurations for @xmath91 , @xmath92 , @xmath93 and @xmath94 . the mv model parameter is chosen as @xmath76 . right : chromo - electric and chromo - magnetic fields and for the same mv model parameters and @xmath88.,title="fig:",scaledwidth=49.0% ] defined in eq . and transverse pressure @xmath69 defined in eq . , multiplied by @xmath95.,scaledwidth=49.0% ] figure [ fig : energy ] shows the time evolution of the energy density and . we set @xmath76 which means that @xmath96 with @xmath97 according to the convention for the rhic physics in the literature @xcite . this choice corresponds to @xmath98 [ @xmath99 ( i.e. the radius of the au atom ) ] and @xmath100 ( where @xmath101 as usually chosen , meaning that @xmath102 ) . in most of the numerical results presented in this work , we keep using these parameters . we note that the strength of the glasma instability as seen in the next section is set by the instability seed @xmath20 relative to @xmath103 which characterizes the cgc background . in this sense , for our purpose in later discussions , the precise value of @xmath104 is of no qualitative importance . in fig . [ fig : energy ] , we investigate the dependence on the transverse grid size @xmath13 : the left plot shows how the energy density increases with increasing @xmath13 , varying from @xmath105 to @xmath88 from the bottom curve to the topmost one . it is already known in the literature @xcite that the @xmath13-dependence ( or the uv - cutoff dependence ) is significant at @xmath106 , but becomes almost irrelevant for @xmath107 ( except for small lattice sizes @xmath108 ) . as is well - known from earlier works @xcite , the energy density decreases as @xmath109 , which is the scaling behavior of a free - streaming expansion . in fact the energy density multiplied by the time , @xmath110 , saturates at a time @xmath111 ( which is sometimes called the formation time , i.e. the time at which partons are freed from the wave - function of the nuclei ) where the free - streaming expansion starts . this is natural on physical grounds , because the classical fields keep interacting strongly until @xmath111 , and become quasi - free at @xmath112 . we see that the longitudinal chromo - electric and chromo - magnetic fields behave in the same way , which is the characteristic feature of the glasma initial state @xcite . in our convention @xmath65 , @xmath113 , @xmath66 , and @xmath114 are all approaching a common value for @xmath115 , and the system still remains anisotropic ( isotropy requires that @xmath116 and @xmath117 ) . this becomes clearer if we consider the transverse pressure @xmath69 and the longitudinal one @xmath118 defined in eqs . and , respectively , as shown in fig . [ fig : p ] . then , we can find that @xmath119 and @xmath120 both approach an asymptotic value and @xmath120 is much smaller than @xmath119 . if @xmath118 is exactly zero , this means that the system reaches the limit of the free - streaming expansion leading to @xmath121 . if the system is completely thermalized , on the other hand , the pressure must be isotropic and thus @xmath71 . because an expansion with positive @xmath118 produces work against the expansion of the matter , the energy density decreases faster than in the free - streaming case , e.g. @xmath122 if @xmath123 and if the expansion is purely longitudinal ( i.e. bjorken expansion ) . therefore , the ( leading order ) glasma state alone can not reach isotropization ; a candidate for causing the longitudinal pressure to grow is the instability due to the @xmath12-dependent fluctuations that occur in higher order corrections . let us now consider the time evolution of the spectral composition of the energy content . that is , the energy is written in terms of fourier components as @xmath124 in terms of the continuum variables . if written in terms of the lattice variables , the transverse momenta take discrete values @xmath125 and the integration is replaced by a summation over @xmath126 . the energy density in the mode @xmath127 is given by @xmath128\bigr\rangle , \label{eq : ee_k}\\ \varepsilon_{_b}(k_\perp ) & \equiv \bigl\langle { \mathrm{tr}}\bigl [ b^{\eta a}(-{\boldsymbol{k}_\perp})b^{\eta a}({\boldsymbol{k}_\perp } ) + \tau^{-2}\bigl ( b^{ia}(-{\boldsymbol{k}_\perp})b^{ia}({\boldsymbol{k}_\perp } ) \bigr ) \bigr]\bigr\rangle\ ; , \label{eq : eb_k}\end{aligned}\ ] ] where @xmath129 , @xmath130 , @xmath131 , and @xmath132 are the fourier transforms of the chromo - electric and chromo - magnetic fields . the above expressions are given in terms of the continuum variables , but it is non - trivial how to re - express them in terms of the link variables : equation for the chromo - electric component is not changed , but we can not use eq . as it is and @xmath114 in eq . are perfectly fine to calculate the chromo - magnetic energy the problem is that they are not simply a sum of squares . ] . it is of course possible to introduce the chromo - magnetic field using the relation in such a way that @xmath133 \notag\\ b_{x}^a & = f_{\eta y}^a \approx \frac{2}{{\mathrm{i}}g a_\eta}{\mathrm{tr}}\bigl [ t^a \bigl ( 1-u_{\eta y } \bigr ) \bigr]\ ; , \quad b_{y}^a = f_{\eta x}^a \approx \frac{2}{{\mathrm{i}}g a_\eta}{\mathrm{tr}}\bigl [ t^a \bigl ( 1-u_{\eta x } \bigr ) \bigr]\ ; . \label{eq : approxb}\end{aligned}\ ] ] this approximation works for sufficiently large number of the transverse sites , i.e. @xmath134 . then , we in principle have two options here : we abandon the information on @xmath135 and just evaluate @xmath136 to see the spectral pattern , or , we adopt a sufficiently large value of @xmath13 so that we can use the approximation . in this work we take the former option , because simulations with @xmath134 would be too time - consuming in the forthcoming studies that include @xmath12-dependent fluctuations ( and where one thus needs to solve ( @xmath137 ) dimensional equations ) . at @xmath138 , the initial spectrum is calculable analytically in the mv model , leading to the expression @xcite , @xmath139 , \label{eq : ee_pert}\ ] ] with an ir cutoff @xmath140 . this formula means that @xmath136 behaves as @xmath141 at large @xmath127 , which is the expected perturbative tail . at @xmath142 , @xmath143 , @xmath144 , and @xmath145 . the transverse size is chosen as @xmath146 and the mv model parameter as @xmath147 . the spectrum at @xmath142 shows the perturbative scaling @xmath148 . an ensemble average is taken over @xmath86 configurations.,scaledwidth=49.0% ] at non - zero @xmath14 , we have evaluated @xmath136 numerically , with results shown in fig . [ fig : time - trans ] . to draw the figure , we have calculated @xmath149 first as a function of @xmath150 and @xmath151 and then have taken an average over the results that lie in a bin , @xmath152 , to obtain @xmath136 with @xmath153 . we use logarithmic horizontal and vertical axes in the plot of fig . [ fig : time - trans ] , to facilitate the identification of the initial - time power - law scaling in the spectrum . we can notice that the energy spectrum at the initial time @xmath142 ( where we started the simulation ) obeys the perturbative scaling @xmath154 up to the uv - cutoff , as indicated by a line for eye guide in fig . [ fig : time - trans ] in agreement with the perturbative expression in eq . . the spectral shape quickly approaches the asymptotic one within @xmath155 and it hardly changes after that time . in fact the shape at @xmath156 is already close to that at @xmath157 ( where we stopped the simulation ) . thanks to this , we do not show the results at any intermediate time between @xmath156 and @xmath157 in order to keep the plot legible . in later discussions in the section [ sec : lspectrum ] we will perform a similar analysis in the longitudinal direction to find non - trivial asymptotic behavior of the spectral shapes . after this summary of the boost - invariant results , including the time evolution of the energy spectrum , let us proceed with the analysis of the instability with respect to @xmath12-dependent fluctuations . we will first confirm the existence of the instability in a similar setup as the one used in previous studies . for this purpose , we choose the initial fluctuations according to the prescription used in refs . @xcite and we fix the mv model parameter @xmath103 in the same way as in the boost - invariant situation . we take the extent in the @xmath12 direction to be @xmath158 units of rapidity and we use periodic boundary conditions for @xmath12 as well as for transverse coordinates . we note that this extent covers roughly the mid - rapidity region where the boost - invariant plateau is expected , and therefore it is legitimate to have a boost - invariant background . that is , our parameter choice reads ; @xmath159 and we perform calculations at various values of the transverse grid size @xmath13 and the longitudinal grid size @xmath17 later . introducing an arbitrary function @xmath160 , we can write the @xmath12-dependent chromo - electric fields in such a way that gauss s law is satisfied by construction , @xmath161\ , \xi^i({\boldsymbol{x}_\perp } ) \ ; , \label{eq : deltaei}\\ \delta e^\eta(x ) & = -f(\eta)\sum_{i = x , y } \bigl [ u_i^\dagger(x-\hat{i})\xi^i({\boldsymbol{x}_\perp}-\hat{i})u_i(x-\hat{i } ) -\xi^i({\boldsymbol{x}_\perp } ) \bigr]\;. \label{eq : deltaeeta}\end{aligned}\ ] ] before applying eq . we can also add @xmath12-dependent fluctuations in @xmath162 . for the moment we will discuss the case with @xmath163 and @xmath164 only . here the choice of @xmath165 is arbitrary and we choose it as a random variable in the transverse plane , i.e.@xmath166 it is important to note that @xmath167 has the dimension of a momentum , which is obvious from the above equation . if all the dimensional quantities are scaled with the transverse spacing @xmath36 as is the case in the numerical implementation , @xmath167 scales as @xmath168 , which is artificial . the strength of the longitudinal disturbance should be given independently . to cancel this artificial @xmath36 dependence , we use a dimensionless @xmath169 satisfying @xmath170 ( on the lattice ) and make @xmath160 scale as @xmath171 , so that the fluctuations are proportional overall to @xmath172 which is a constant . in the literature @xcite @xmath160 is chosen as a random variable too containing all infrared and ultraviolet modes in the spectrum . here , in order to study the instability behavior in a well - controlled situation , let us perturb the system with a single @xmath12-mode , and consider superposition of multiple modes only later . hence , we will adopt the following simple form for the moment , the frequency @xmath21 is an integer between @xmath22 and @xmath23 . ] , @xmath173 where @xmath174 is chosen to be @xmath70 as we explained before and @xmath20 should scale as @xmath175 to make the fluctuations insensitive to the way we discretize the transverse plane . we will later fix @xmath176 and then vary @xmath13 to study the ( unphysical ) dependence on the transverse grid size . here we fix @xmath177 in eq . because , as we will confirm later , this lowest non - zero mode leads to the fastest and strongest growth in the longitudinal pressure . figure [ fig : size_t ] shows how the instability occurs in the longitudinal pressure @xmath118 in the long - time run . the left - panel is the pressure multiplied by time ( @xmath178 ) in the unit of @xmath179 . we note that the plotted quantity is for @xmath180 which maximizes the fourier transform of @xmath120 ( except for the zero mode @xmath181 ) . that is , we define , @xmath182 which is complex valued in general in our definition ranges from @xmath22 to @xmath183 and the corresponding wave - number is @xmath184 . ] . we thus plot its modulus , @xmath185 , in the left panel of fig . [ fig : size_t ] corresponding to an initial disturbance at @xmath177 and @xmath186 in eqs . and ( i.e. we choose @xmath187 as a reference point and then @xmath188 for @xmath187 ) . the horizontal axis represents @xmath189 , which is a natural variable in a longitudinally expanding geometry . note that the fourier transform @xmath190 at finite @xmath18 is not a measurable quantity in experiments . its real and imaginary parts both fluctuate from negative to positive values depending on the initial conditions and they would be vanishing if we compute an average without taking the modulus . only the @xmath181 mode has a physical meaning as the longitudinal pressure . for our purpose of studying the instability , we can nevertheless consider @xmath191 ( i.e. we compute the modulus first , and we perform the average over initial conditions next ) in order to confirm some known results reported previously @xcite . in later discussions , we will also decompose the energy density into @xmath18 components , which is a more well - defined quantity to see the instability . , where @xmath177 is chosen for the initial condition . the results are for @xmath105 , @xmath93 , @xmath92 , @xmath192 , @xmath193 in the transverse direction ( from the top to the bottom ) and @xmath194 fixed in the longitudinal direction ( except for @xmath195 for @xmath75 and @xmath196 for @xmath197 ) . the seed magnitude is @xmath188 for @xmath187 , i.e. @xmath186 in eq . . an ensemble average is taken over @xmath198 configurations . right : spectrum of the longitudinal `` pressure '' as a function of @xmath18 at @xmath199 ( @xmath200).,title="fig:",scaledwidth=49.0% ] , where @xmath177 is chosen for the initial condition . the results are for @xmath105 , @xmath93 , @xmath92 , @xmath192 , @xmath193 in the transverse direction ( from the top to the bottom ) and @xmath194 fixed in the longitudinal direction ( except for @xmath195 for @xmath75 and @xmath196 for @xmath197 ) . the seed magnitude is @xmath188 for @xmath187 , i.e. @xmath186 in eq . . an ensemble average is taken over @xmath198 configurations . right : spectrum of the longitudinal `` pressure '' as a function of @xmath18 at @xmath199 ( @xmath200).,title="fig:",scaledwidth=49.0% ] from the left panel of fig . [ fig : size_t ] , the exponential growth with the square root of proper time is obvious . the onset of the instability depends on the transverse size @xmath13 but the instability slope is rather insensitive to @xmath13 . with increasing @xmath13 , the instability occurs earlier , and eventually the shift in this onset position saturates for @xmath201 . this means that the instability does not start earlier by increasing @xmath13 further . so , we stop increasing @xmath13 and conclude that simulations at @xmath187 are reasonably close to the continuum limit . in the right panel , we see how the fields , initially localized in the modes @xmath202 ( background field itself ) and @xmath203 ( perturbation ) , populate the higher @xmath18 modes due to the non - linearities in the yang - mills equations . for a small perturbation , the amplitude in the higher modes decreases very fast ( exponentially with @xmath18 ) , because in order to produce a field at a given @xmath18 , one needs to multiply @xmath18 seeds at @xmath177 . . the results are for @xmath188 and @xmath187 fixed in the transverse direction and @xmath194 and @xmath204 in the longitudinal direction . an ensemble average is taken over @xmath198 configurations . right : spectrum of the longitudinal `` pressure '' as a function of @xmath18 at @xmath199.,title="fig:",scaledwidth=49.0% ] . the results are for @xmath188 and @xmath187 fixed in the transverse direction and @xmath194 and @xmath204 in the longitudinal direction . an ensemble average is taken over @xmath198 configurations . right : spectrum of the longitudinal `` pressure '' as a function of @xmath18 at @xmath199.,title="fig:",scaledwidth=49.0% ] physical results must also not depend on how we discretize the longitudinal coordinate . in order to see the dependence on the longitudinal grid size , we make a plot in fig . [ fig : size_l ] in the same way as in fig . [ fig : size_t ] , where we compare two longitudinal grid sizes that differ by a factor @xmath70 . as is clear from fig . [ fig : size_l ] we can not find any sizable dependence on @xmath17 , which means that our numerical calculations are safely free from lattice artifacts in the longitudinal direction . the influence of @xmath17 can be seen more clearly in the spectrum of @xmath205 shown in the right panel of fig . [ fig : size_l ] . a different choice of @xmath17 with fixed @xmath174 changes the uv cutoff ( the spectrum ranges between @xmath22 and @xmath23 ) : in the right panel of fig . [ fig : size_l ] , @xmath18 varies from @xmath206 to @xmath207 for the @xmath194 calculation and from @xmath208 to @xmath209 for the @xmath204 calculation . since the seeds are at @xmath180 in our initial condition , we naturally anticipate that the uv cutoff at @xmath210 does not affect physical properties around @xmath211 . one may think that the results for a larger seed amplitude @xmath212 in fig . [ fig : size_l ] could be sensitive to the choice of @xmath17 because all the amplitudes in the spectrum are substantially larger and thus a change in @xmath17 leading to a change in the uv cutoff may have an influence on the spectrum . we have also checked this and found that it is not the case . as is clear from fig . [ fig : size_l ] , the difference between the results for @xmath213 with @xmath212 and that for @xmath214 with @xmath212 is almost invisible . from all these results we can conclude that a value of @xmath194 is sufficiently large as long as @xmath21 is set to be of order unity . if @xmath21 takes a larger value , as we will see soon below , the spectral shape has a wider distribution and the uv cutoff at @xmath210 could influence the results . . the results are for @xmath215 , @xmath216 , @xmath217 and a fixed lattice size @xmath213 . right : spectrum of the longitudinal `` pressure '' as a function of @xmath18 at @xmath199 , which has a uv cutoff at @xmath218 for the @xmath194 calculation.,title="fig:",scaledwidth=49.0% ] . the results are for @xmath215 , @xmath216 , @xmath217 and a fixed lattice size @xmath213 . right : spectrum of the longitudinal `` pressure '' as a function of @xmath18 at @xmath199 , which has a uv cutoff at @xmath218 for the @xmath194 calculation.,title="fig:",scaledwidth=49.0% ] in this subsection , we study the dependence on the seed magnitude @xmath20 of the longitudinal fluctuations in eq . . in fig . [ fig : magnitude ] , we show some of numerical results for different values of @xmath20 . it is interesting that the results obviously scale as @xmath219 up to the point where the instability growth eventually saturates . this @xmath20 dependence is almost trivial , which suggests that the non - linear effects in terms of finite-@xmath18 modes are small unless the unstable modes substantially develop at large @xmath95 . one might have thought that any physical observables should be expanded in even powers of @xmath20 and thus @xmath220 should be proportional @xmath221 rather than @xmath20 . this is not the case because we took the modulus of complex @xmath222 before performing the average over initial conditions , thus preventing the linear term in @xmath20 from vanishing . we can also notice from the right panel of fig . [ fig : magnitude ] that , once the instability at @xmath223 gets saturated , the mode at @xmath223 stops growing but the spectrum spreads quickly to higher @xmath18-modes . once this happens , the scaling property with @xmath20 is lost . actually the amplitude of the @xmath223 mode reaches one third of the zero - mode amplitude ( for @xmath212 results at @xmath199 ) before this saturation occurs . then the non - linearities or self - interaction effects can not be neglected any longer . now we investigate the @xmath21-dependence of the instability behavior . first , we have chosen three different @xmath21 s as @xmath177 , @xmath224 , and @xmath225 to uncover the general trend . then we study a more general case , that consists in a superposition of these initial wave - numbers . , @xmath224 , and @xmath225 with the initial condition given as a superposition . right : spectrum of the longitudinal `` pressure '' as a function of @xmath18 at @xmath199 . the solid line represents the results from the non - linear evolution with the initial condition given as a superposition , while the dotted one is a simple linear superposition of the results shown in fig . [ fig : nu].,title="fig:",scaledwidth=49.0% ] , @xmath224 , and @xmath225 with the initial condition given as a superposition . right : spectrum of the longitudinal `` pressure '' as a function of @xmath18 at @xmath199 . the solid line represents the results from the non - linear evolution with the initial condition given as a superposition , while the dotted one is a simple linear superposition of the results shown in fig . [ fig : nu].,title="fig:",scaledwidth=49.0% ] , @xmath224 , and @xmath225 with the initial condition given as a superposition . right : spectrum of the longitudinal `` pressure '' as a function of @xmath18 at @xmath199 . the solid line represents the results from the non - linear evolution with the initial condition given as a superposition , while the dotted one is a simple linear superposition of the results shown in fig . [ fig : nu].,title="fig:",scaledwidth=49.0% ] , @xmath224 , and @xmath225 with the initial condition given as a superposition . right : spectrum of the longitudinal `` pressure '' as a function of @xmath18 at @xmath199 . the solid line represents the results from the non - linear evolution with the initial condition given as a superposition , while the dotted one is a simple linear superposition of the results shown in fig . [ fig : nu].,title="fig:",scaledwidth=49.0% ] the figure [ fig : nu ] shows the numerical results . it is clear from the figure that the onset of the instability is delayed for larger @xmath21 . this is naturally understandable from the kinetic term for finite-@xmath18 modes . the ( positive ) kinetic term generally tends to stabilize the system , and so the ( negative ) term responsible for the instability is overwhelmed if @xmath18 is large enough . for dimensional reasons , however , this kinetic term for the finite-@xmath18 modes is of the form @xmath226 , and therefore it decreases with time . in contrast , the negative term that generates the instability has a constant coefficient of order unity therefore , it dominates over the kinetic term at sufficiently large time . in other words , the kinetic term in @xmath226 only delays the start of the instability by a time proportional to @xmath18 . we can confirm qualitatively this interpretation from fig . [ fig : nu ] . the growth appears to start around @xmath227 for @xmath228 , respectively , and the starting times are indeed ordered proportionally to @xmath21 . the spectral shape shown in the right panel of fig . [ fig : nu ] looks discontinuous for @xmath229 and @xmath225 , but this is a trivial artifact of the fact that our initial seed contains a single @xmath21 mode . indeed , the non - linear couplings in the yang - mills equations can only produce linear combinations with integer coefficients of the frequencies present in the initial condition . for instance , with @xmath229 , only modes with @xmath18 a multiple of @xmath224 can be produced by the non - linear evolution . the fact that the amplitude if these higher harmonics is suppressed compared to the amplitude of the base frequency @xmath21 is a good indication that we are still in a regime where the non - linearities are rather small . we can also see that the non - linear effects are certainly present but not very large by repeating the above calculation with an initial condition given as a superposition of three modes , namely , @xmath230 with @xmath231 , @xmath232 , @xmath233 , instead of the single - mode fluctuation of eq . with @xmath177 , @xmath229 , @xmath234 individually . if the non - linearity is significant , the results should not be a simple superposition of the individual calculation . in view of fig . [ fig : nu_mix ] , however , we can see only small deviations from the individual results shown in fig . [ fig : nu ] at @xmath199 . this is an important observation because the initial fluctuations in reality should have some continuous spectral distribution ( see ref . @xcite for instance ) and we simplified our analysis by considering the special case of a single - mode fluctuation . as long as the non - linearity is a minor effect , we can just add up the single - mode outputs in order to get approximate results for general initial spectra . although they are several orders of magnitude smaller at @xmath199 , we see that higher harmonics mixing the three @xmath21 s are enhanced by non - linear effects in the right panel of fig . [ fig : nu_mix ] . for instance , the amplitude of the mode @xmath235 is much larger than the sum of the amplitudes at @xmath235 for the three seeds considered in isolation . indeed , when the seed @xmath177 is alone for example , the mode @xmath235 is produced only after 20 interactions , which explains why it is totally negligible . furthermore , the seeds @xmath229 or @xmath225 can never lead to @xmath235 by themselves . when the three seeds are superposed in the initial condition , 21 can be reached more efficiently as @xmath236 , i.e. with 2 interactions only . apart from small non - linear effects , it is certainly true that the smallness of the field fluctuations we have considered gives a justification for our simplified , single mode , analysis . however , at the same time , thermalization is hardly expected to occur in these circumstances , precisely because in this regime the interaction effects are small . in realistic conditions , the parameter @xmath20 that controls the fluctuations is related to the strong coupling @xmath9 , and therefore it could be larger than @xmath237 that we have considered here . as a consequence , we expect that the non - linearities would start much sooner and be much more important in applications of this framework to actual heavy - ion collisions . the study with such large @xmath20 is beyond our current scope because a large @xmath20 requires full information on quantum fluctuations and needs appropriate renormalization procedures , which is technically involved @xcite . in spite of the instability the zero - mode is still dominant among all the modes in @xmath190 . with an initial condition that contains a single mode as in eq . , the @xmath180 mode is the second largest , which is obvious from the spectrum shown in the right panel of fig . [ fig : size_t ] . moreover , the spectrum is spreading into higher wave - numbers as the time elapses . in this section , we shall discuss the scaling property of the energy spectrum with respect to the frequency @xmath18 , and show that it possibly displays also a kolmogorov s turbulent spectrum . a turbulent spectrum has already been reported in the context of the qcd plasma instability @xcite , but these results with the power @xmath238 instead of @xmath239 are in conflict with kolmogorov s wave turbulence . on the other hand , the study performed in ref . @xcite has shown that a kolmogorov spectrum appears due to instabilities in yang - mills theory , for a fixed - volume system . therefore , it is still worth addressing the turbulent spectrum in the present glasma simulation , that has longitudinal expansion . this is because the glasma system is expanding along the beam axis , which tends to tame the turbulences and to delay the isotropization or thermalization in general . it is thus non - trivial whether the universal scaling still holds or not in the expanding system . besides , our description of the glasma evolution is purely classical and the precise relationship between the glasma instability and the qcd plasma instability is not fully understood yet . . the lattice size is @xmath213 and the seed is @xmath188 . left : results at early times up to @xmath240 . note that the horizontal axis is not logarithmic and also that the vertical axis is for not @xmath241 but @xmath242 ( see the text for details ) . right : results at later times up to @xmath243.,title="fig:",scaledwidth=49.0% ] . the lattice size is @xmath213 and the seed is @xmath188 . left : results at early times up to @xmath240 . note that the horizontal axis is not logarithmic and also that the vertical axis is for not @xmath241 but @xmath242 ( see the text for details ) . right : results at later times up to @xmath243.,title="fig:",scaledwidth=49.0% ] two plots in fig . [ fig : time - scale - long ] show the longitudinal energy spectrum ( for the chromo - electric part only ) , that is , the energy density decomposed in terms of the longitudinal wave - number @xmath18 as @xmath244 with @xmath245\bigr\rangle .\ ] ] ( the transverse momentum @xmath246 of the fields is integrated over . ) the question is then what kind of scaling spectrum we can expect for @xmath247 at later times . for this purpose , we consider three characteristic quantities ; the wave - number , the fourier decomposed energy , and the rate of energy flow in @xmath18-space ( denoted by @xmath248 below ) . the wave - number @xmath18 itself is , however , dimensionless and it seems difficult to cope with the dimensional matching argument by kolmogorov . we propose that one should use a scaled variable @xmath249 instead of @xmath18 , in the case of an expanding geometry . this is because @xmath250 is approximately equal to the longitudinal coordinate @xmath251 if @xmath12 is small enough , and @xmath249 is approximately equal to the momentum @xmath252 . then , the dimension of the scaled wave - number is @xmath253=l^{-1}$ ] . we note that the integration variable in the decomposition should also be @xmath254 for this scaling analysis , which means that the relevant energy density per mode should be @xmath255 . this in addition should be multiplied by the expanding length @xmath256 . therefore , the dimensions of the three relevant quantities are @xmath257 = l^{-1 } , \qquad [ v_\perp(c\tau)^2\varepsilon_{_e}(\nu ) ] = l^3 t^{-2 } , \qquad [ \psi ] = l^2 t^{-3 } , \ ] ] which are identical to the standard analysis of kolmogorov s spectrum . provided that @xmath258 is expressed in a form of @xmath259 , we can fix @xmath260 and @xmath261 uniquely from the dimensions and thus expect the kolmogorov scaling , that is , @xmath262 . in fact we can confirm this scaling property asymptotically from the right panel of fig . [ fig : time - scale - long ] . one might be wondering what the fate of the power - law spectrum should be at even later times . we understand from fig . [ fig : stage ] that the glasma time evolution in the present mv - model simulation has three stages ; the linear regime , the non - linear regime , and the uv - cutoff regime . in the linear regime the unstable modes grow up and the non - linear effects are still minor . when the amplitudes of @xmath263 modes become so large that the non - linearities can affect the time evolution , the instability stops , as shown in the right panel of fig . [ fig : stage ] , and the non - abelianization occurs @xcite . because non - linearities are no longer negligible in this regime , the energy cascade due to non - linearities is naturally expected , which would lead to the kolmogorov - type spectrum . eventually , as the turbulence decays into higher uv modes , the uv - cutoff effect flattens the whole spectrum as plotted in fig . [ fig : stage ] up to @xmath264 . up to @xmath264 . left : results including later time spectra . the horizontal axis is logarithmic to make it evident how the power - law spectrum spreads . right : longitudinal `` pressure '' with three regimes identified schematically.,scaledwidth=49.0% ] to the best of our knowledge the present work is the first to exhibit a kolmogorov spectrum from the glasma time evolution . our results imply that the framework of the glasma , in terms of classical variables together with @xmath12-dependent fluctuations , may encompass a rich physics contents including the qcd plasma instability . here , as a final remark in this section , we draw attention to related works in analogous systems . in the context of an ultra - cold gas , the kolmogorov spectrum has been obtained as a solution of the gross - pitaevskii equation ( gpe ) , which is a classical equation of motion like the glasma equations @xcite . the calculational procedure by means of the gpe is similar to the glasma simulation , but in that case the microscopic mechanism of the energy flow is understood in terms of the vortex dynamics . in future studies on the glasma dynamics , it will be important to clarify the structure of the energy flow , and it would be very interesting to see whether the so - called non - thermal fixed point @xcite appears in the strongly correlated yang - mills theory and glasma . indeed , a possible deviation in the power could be expected in yang - mills theory by non - perturbative resummations @xcite . in this paper we investigated the time evolution of the classical equations of motion in the su(2 ) pure yang - mills theory , especially in the framework of the glasma or the mclerran - venugopalan model in ( 3 + 1 ) dimensions . our emphasis has been put on the systematic study of the instability behavior that occurs in fluctuations with respect to the space - time rapidity @xmath12 when the initial conditions break boost invariance . computing the longitudinal pressure component at non - zero wave - number @xmath18 , that is a conjugate of @xmath12 , we have numerically confirmed the glasma instability that had been found in preceding works @xcite . we carefully checked the system size dependence of the instability to conclude that the glasma instability is a robust consequence from @xmath12-dependent disturbances and is not sensitive to how the transverse and longitudinal coordinates are discretized . hence , the glasma instability is certainly a physically well - defined phenomenon . at the same time , however , this means that taking the continuum limit does not help with an earlier onset of the instability . the typical time scales we found for the glasma instability is of order of hundreds in terms of @xmath95 , which is arguably too slow for being useful in heavy ion collisions . one should however keep in mind that the seeds we have used in this numerical study are very small compared to the glasma fields , while in a realistic situation they are suppressed at most by a power of the strong coupling constant . there may also be a chance that the non - linear effects could more or less accelerate the instability . we compared two results for the spectral distribution , one from the superposition of individual calculations with the initial condition that is given by a different wave - number @xmath21 , and the other from a single calculation with the initial condition that contains all @xmath21 s . we could observe the non - linear effects in our numerical results , which would not affect the instability behavior substantially , however , as long as the instability seed @xmath20 is reasonably small . we have decomposed the energy density into fourier modes to see how the low wave - number modes cascade toward higher wave - number modes due to non - linearity in the equations of motion . our motivation to do so is to examine whether the so - called kolmogorov s scaling can appear in the turbulent system described by the yang - mills theory put in an expanding geometry . in contrast to the boost - invariant case , a kolmogorov - type scaling behavior in the energy spectrum a function of the longitudinal wave - number @xmath18 may appear when the boost invariance is broken by the initial fluctuations . in this case , the violation of boost invariance increases with time due to instability , but the @xmath181 mode remains very large compared to the other @xmath263 modes , which enables us to regard the @xmath181 mode as the source for the energy injection . then , we could clearly observe that the energy spectra approach an asymptotic form that shows the kolmogorov s scaling with the power @xmath265 . as far as we know , our work is the first example that exhibits the kolmogorov - type scaling law in the expanding systems . although it is difficult to find any indication that the system indeed becomes isotropic and thermalized in our simulation , the confirmation of the kolmogorov - type spectra is a promising signal for the general tendency heading for thermalization . there remain very interesting questions to be investigated in the future . in order to unveil the microscopic mechanism for the instability , it would be very useful to make approximations on the classical equations of motion in such a way that the analytical treatment could be feasible . for instance , as we have confirmed , the instability mostly lies in the linear regime as long as its magnitude specified by @xmath20 is small enough . therefore the linearization of the equations of motion in terms of fluctuating fields on top of the boost - invariant cgc background should correctly describe the instability in early times . in this work we did not exploit such a comparison between the numerical output and the analytical result from the linearized equations . this is because it is not quite efficient to linearize the equations of motion given in terms of the lattice variables . the linearization is of course possible , but it would eventually be reduced to similar calculations in the continuum formulation . then , it would make much more sense to make a comparison between the numerical and analytical results that are both formulated in the continuum variables . we have already executed test programs to make sure if the numerical results from the glasma simulations are consistent with each other , given they are written in terms of lattice and continuum variables . we will report on this analysis elsewhere . it is also an intriguing question to look for a link between kolmogorov s turbulent spectrum and the chaotic behavior generally seen in the solution of the classical yang - mills equations . this problem may have some relevance for a deeper understanding of the microscopic description of the entropy generating process and how the thermalization is achieved at all . the time evolution of the glasma and its physics contents deserve much more investigations in the future . we would like to thank kevin dusling , tuomas lappi , and raju venugopalan for discussions . this work has been supported in part by jsps japan - france integrated action program ( sakura ) . thanks the yukawa institute for theoretical physics , where part of this work has been done , for its hospitality and support . p. f. kolb and r. rapp , `` transverse flow and hadrochemistry in au+au collisions at @xmath266 , '' http://dx.doi.org/10.1103/physrevc.67.044903[_phys . rev . _ * c67 * ( 2003 ) 044903 ] , http://arxiv.org/abs/hep-ph/0210222 [ arxiv : hep - ph/0210222 [ hep - ph ] ] . r. baier , a. h. mueller , d. schiff , and d. t. son , `` bottom - up thermalization in heavy ion collisions , '' http://dx.doi.org/10.1016/s0370-2693(01)00191-5[_phys . lett . _ * b502 * ( 2001 ) 5158 ] , http://arxiv.org/abs/hep-ph/0009237[arxiv:hep-ph/0009237 ] . y. v. kovchegov , `` can thermalization in heavy ion collisions be described by qcd diagrams ? , '' http://dx.doi.org/10.1016/j.nuclphysa.2005.08.009[_nucl . _ * a762 * ( 2005 ) 298325 ] , http://arxiv.org/abs/hep-ph/0503038[arxiv:hep-ph/0503038 ] . b. arnold , j. lenaghan , and g. d. moore , `` qcd plasma instabilities and bottom up thermalization , '' _ jhep _ * 0308 * ( 2003 ) 002 , http://arxiv.org/abs/hep-ph/0307325[arxiv:hep-ph/0307325 [ hep - ph ] ] . erratum added online , sep/29/2004 . b. arnold , j. lenaghan , g. d. moore , and l. g. yaffe , `` apparent thermalization due to plasma instabilities in quark - gluon plasma , '' http://dx.doi.org/10.1103/physrevlett.94.072302[_phys . * 94 * ( 2005 ) 072302 ] , http://arxiv.org/abs/nucl-th/0409068 [ arxiv : nucl - th/0409068 [ nucl - th ] ] . p. b. arnold , g. d. moore , and l. g. yaffe , `` the fate of non - abelian plasma instabilities in 3 + 1 dimensions , '' http://dx.doi.org/10.1103/physrevd.72.054003[_phys . * d72 * ( 2005 ) 054003 ] , http://arxiv.org/abs/hep-ph/0505212 [ arxiv : hep - ph/0505212 [ hep - ph ] ] . p. b. arnold and g. d. moore , `` qcd plasma instabilities : the nonabelian cascade , '' http://dx.doi.org/10.1103/physrevd.73.025006[_phys . rev . _ * d73 * ( 2006 ) 025006 ] , http://arxiv.org/abs/hep-ph/0509206[arxiv:hep-ph/0509206 [ hep - ph ] ] . j. randrup and s. mrowczynski , `` chromodynamic weibel instabilities in relativistic nuclear collisions , '' http://dx.doi.org/10.1103/physrevc.68.034909[_phys . rev . _ * c68 * ( 2003 ) 034909 ] , http://arxiv.org/abs/nucl-th/0303021[arxiv:nucl-th/0303021 ] . s. mrowczynski , `` instabilities driven equilibration of the quark - gluon plasma , '' _ acta phys . * b37 * ( 2006 ) 427454 , http://arxiv.org/abs/hep-ph/0511052[arxiv:hep-ph/0511052 ] . a. dumitru , y. nara , and m. strickland , `` ultraviolet avalanche in anisotropic non - abelian plasmas , '' http://dx.doi.org/10.1103/physrevd.75.025016[_phys . rev . _ * d75 * ( 2007 ) 025016 ] , http://arxiv.org/abs/hep-ph/0604149 [ arxiv : hep - ph/0604149 [ hep - ph ] ] . j. berges , d. gelfand , s. scheffler , and d. sexty , `` simulating plasma instabilities in su(3 ) gauge theory , '' http://dx.doi.org/10.1016/j.physletb.2009.05.008[_phys . * b677 * ( 2009 ) 210213 ] , http://arxiv.org/abs/0812.3859[arxiv:0812.3859 [ hep - ph ] ] . d. bodeker , `` the impact of qcd plasma instabilities on bottom - up thermalization , '' http://dx.doi.org/10.1088/1126-6708/2005/10/092[_jhep_ * 0510 * ( 2005 ) 092 ] , http://arxiv.org/abs/hep-ph/0508223 [ arxiv : hep - ph/0508223 [ hep - ph ] ] . a. rebhan , p. romatschke , and m. strickland , `` hard - loop dynamics of non - abelian plasma instabilities , '' http://dx.doi.org/10.1103/physrevlett.94.102303[_phys . lett . _ * 94 * ( 2005 ) 102303 ] , http://arxiv.org/abs/hep-ph/0412016[arxiv:hep-ph/0412016 ] . a. rebhan , p. romatschke , and m. strickland , `` dynamics of quark - gluon plasma instabilities in discretized hard - loop approximation , '' http://dx.doi.org/10.1088/1126-6708/2005/09/041[_jhep_ * 09 * ( 2005 ) 041 ] , http://arxiv.org/abs/hep-ph/0505261[arxiv:hep-ph/0505261 ] . p. romatschke and a. rebhan , `` plasma instabilities in an anisotropically expanding geometry , '' http://dx.doi.org/10.1103/physrevlett.97.252301[_phys . _ * 97 * ( 2006 ) 252301 ] , http://arxiv.org/abs/hep-ph/0605064[arxiv:hep-ph/0605064 ] . a. rebhan , m. strickland , and m. attems , `` instabilities of an anisotropically expanding non - abelian plasma : 1d+3v discretized hard - loop simulations , '' http://dx.doi.org/10.1103/physrevd.78.045023[_phys . rev . _ * d78 * ( 2008 ) 045023 ] , http://arxiv.org/abs/0802.1714 [ arxiv:0802.1714 [ hep - ph ] ] . a. ipp , a. rebhan , and m. strickland , `` non - abelian plasma instabilities : su(3 ) vs. su(2 ) , '' http://arxiv.org/abs/1012.0298[arxiv:1012.0298 [ hep - ph ] ] . b. muller and a. trayanov , `` deterministic chaos in nonabelian lattice gauge theory , '' http://dx.doi.org/10.1103/physrevlett.68.3387[_phys * 68 * ( 1992 ) 33873390 ] . t. s. biro , c. gong , and b. muller , `` lyapunov exponent and plasmon damping rate in nonabelian gauge theories , '' http://dx.doi.org/10.1103/physrevd.52.1260[_phys . * d52 * ( 1995 ) 12601266 ] , http://arxiv.org/abs/hep-ph/9409392[arxiv:hep-ph/9409392 ] . t. kunihiro _ et al . _ , `` chaotic behavior in classical yang - mills dynamics , '' http://dx.doi.org/10.1103/physrevd.82.114015[_phys . rev . _ * d82 * ( 2010 ) 114015 ] , http://arxiv.org/abs/1008.1156[arxiv:1008.1156 [ hep - ph ] ] . a. nishiyama and a. ohnishi , `` entropy production in gluodynamics in temporal axial gauge in 2 + 1 dimensions , '' _ prog . * 125 * ( 2011 ) 775795 , http://arxiv.org/abs/1011.4750[arxiv:1011.4750 [ nucl - th ] ] . p. m. chesler and l. g. yaffe , `` horizon formation and far - from - equilibrium isotropization in supersymmetric yang - mills plasma , '' http://dx.doi.org/10.1103/physrevlett.102.211601[_phys . * 102 * ( 2009 ) 211601 ] , http://arxiv.org/abs/0812.2053 [ arxiv:0812.2053 [ hep - th ] ] . p. m. chesler and l. g. yaffe , `` boost invariant flow , black hole formation , and far - from - equilibrium dynamics in n = 4 supersymmetric yang - mills theory , '' http://dx.doi.org/10.1103/physrevd.82.026006[_phys . rev . _ * d82 * ( 2010 ) 026006 ] , http://arxiv.org/abs/0906.4426 [ arxiv:0906.4426 [ hep - th ] ] . y. v. kovchegov and s. lin , `` toward thermalization in heavy ion collisions at strong coupling , '' http://dx.doi.org/10.1007/jhep03(2010)057 [ _ jhep _ * 03 * ( 2010 ) 057 ] , http://arxiv.org/abs/0911.4707[arxiv:0911.4707 [ hep - th ] ] . v. balasubramanian _ et al . _ , `` thermalization of strongly coupled field theories , '' http://arxiv.org/abs/1012.4753[arxiv:1012.4753 [ hep - th ] ] . f. gelis , t. lappi , and r. venugopalan , `` high energy scattering in quantum chromodynamics , '' http://dx.doi.org/10.1142/s0218301307008331 [ _ int.j.mod.phys . _ * e16 * ( 2007 ) 25952637 ] , http://arxiv.org/abs/0708.0047[arxiv:0708.0047 [ hep - ph ] ] . l. d. mclerran and r. venugopalan , `` computing quark and gluon distribution functions for very large nuclei , '' http://dx.doi.org/10.1103/physrevd.49.2233[_phys . rev . _ * d49 * ( 1994 ) 22332241 ] , http://arxiv.org/abs/hep-ph/9309289[arxiv:hep-ph/9309289 ] . l. d. mclerran and r. venugopalan , `` gluon distribution functions for very large nuclei at small transverse momentum , '' http://dx.doi.org/10.1103/physrevd.49.3352[_phys * d49 * ( 1994 ) 33523355 ] , http://arxiv.org/abs/hep-ph/9311205[arxiv:hep-ph/9311205 ] . l. d. mclerran and r. venugopalan , `` fock space distributions , structure functions , higher twists and small x , '' http://dx.doi.org/10.1103/physrevd.59.094002[_phys . rev . _ * d59 * ( 1999 ) 094002 ] , http://arxiv.org/abs/hep-ph/9809427[arxiv:hep-ph/9809427 ] . a. kovner , l. d. mclerran , and h. weigert , `` gluon production from nonabelian weizsacker - williams fields in nucleus - nucleus collisions , '' http://dx.doi.org/10.1103/physrevd.52.6231[_phys . rev . _ * d52 * ( 1995 ) 62316237 ] , http://arxiv.org/abs/hep-ph/9502289[arxiv:hep-ph/9502289 ] . a. kovner , l. d. mclerran , and h. weigert , `` gluon production at high transverse momentum in the mclerran - venugopalan model of nuclear structure functions , '' http://dx.doi.org/10.1103/physrevd.52.3809[_phys . _ * d52 * ( 1995 ) 38093814 ] , http://arxiv.org/abs/hep-ph/9505320[arxiv:hep-ph/9505320 ] . d. son , `` semiclassical approach for multiparticle production in scalar theories , '' http://dx.doi.org/10.1016/0550-3213(96)00386-0[_nucl . phys . _ * b477 * ( 1996 ) 378406 ] , http://arxiv.org/abs/hep-ph/9505338[arxiv:hep-ph/9505338 [ hep - ph ] ] . a. mueller and d. son , `` on the equivalence between the boltzmann equation and classical field theory at large occupation numbers , '' http://dx.doi.org/10.1016/j.physletb.2003.12.047[_phys . _ * b582 * ( 2004 ) 279287 ] , http://arxiv.org/abs/hep-ph/0212198 [ arxiv : hep - ph/0212198 [ hep - ph ] ] . a. krasnitz and r. venugopalan , `` non - perturbative computation of gluon mini - jet production in nuclear collisions at very high energies , '' http://dx.doi.org/10.1016/s0550-3213(99)00366-1[_nucl . _ * b557 * ( 1999 ) 237 ] , http://arxiv.org/abs/hep-ph/9809433[arxiv:hep-ph/9809433 ] . a. krasnitz and r. venugopalan , `` the initial energy density of gluons produced in very high energy nuclear collisions , '' http://dx.doi.org/10.1103/physrevlett.84.4309[_phys . * 84 * ( 2000 ) 43094312 ] , http://arxiv.org/abs/hep-ph/9909203[arxiv:hep-ph/9909203 ] . a. krasnitz and r. venugopalan , `` the initial gluon multiplicity in heavy ion collisions , '' http://dx.doi.org/10.1103/physrevlett.86.1717 [ _ phys . * 86 * ( 2001 ) 17171720 ] , http://arxiv.org/abs/hep-ph/0007108[arxiv:hep-ph/0007108 ] . t. lappi , `` production of gluons in the classical field model for heavy ion collisions , '' http://dx.doi.org/10.1103/physrevc.67.054903[_phys . * c67 * ( 2003 ) 054903 ] , http://arxiv.org/abs/hep-ph/0303076[arxiv:hep-ph/0303076 ] . t. lappi and l. mclerran , `` some features of the glasma , '' http://dx.doi.org/10.1016/j.nuclphysa.2006.04.001[_nucl . phys . _ * a772 * ( 2006 ) 200212 ] , http://arxiv.org/abs/hep-ph/0602189[arxiv:hep-ph/0602189 ] . a. dumitru , f. gelis , l. mclerran , and r. venugopalan , `` glasma flux tubes and the near side ridge phenomenon at rhic , '' http://dx.doi.org/10.1016/j.nuclphysa.2008.06.012[_nucl . phys . _ * a810 * ( 2008 ) 91108 ] , http://arxiv.org/abs/0804.3858 [ arxiv:0804.3858 [ hep - ph ] ] . k. dusling , f. gelis , t. lappi , and r. venugopalan , `` long range two - particle rapidity correlations in a+a collisions from high energy qcd evolution , '' http://dx.doi.org/10.1016/j.nuclphysa.2009.12.044 , 10.1016/j.nuclphysa.2009.12.044[_nucl . phys . _ * a836 * ( 2010 ) 159182 ] , http://arxiv.org/abs/0911.2720[arxiv:0911.2720 [ hep - ph ] ] . p. romatschke and r. venugopalan , `` collective non - abelian instabilities in a melting color glass condensate , '' http://dx.doi.org/10.1103/physrevlett.96.062302[_phys . _ * 96 * ( 2006 ) 062302 ] , http://arxiv.org/abs/hep-ph/0510121[arxiv:hep-ph/0510121 ] . p. romatschke and r. venugopalan , `` the unstable glasma , '' http://dx.doi.org/10.1103/physrevd.74.045011[_phys . rev . _ * d74 * ( 2006 ) 045011 ] , http://arxiv.org/abs/hep-ph/0605045[arxiv:hep-ph/0605045 ] . k. fukushima , `` initial fields and instability in the classical model of the heavy - ion collision , '' http://dx.doi.org/10.1103/physrevc.76.021902[_phys . _ * c76 * ( 2007 ) 021902 ] , http://arxiv.org/abs/0704.3625[arxiv:0704.3625 [ hep - ph ] ] . k. fukushima , `` erratum : initial fields and instability in the classical model of the heavy - ion collision , '' http://arxiv.org/abs/0711.2634[arxiv:0711.2634 [ hep - ph ] ] . h. fujii and k. itakura , `` expanding color flux tubes and instabilities , '' http://dx.doi.org/10.1016/j.nuclphysa.2008.05.016[_nucl . phys . _ * a809 * ( 2008 ) 88109 ] , http://arxiv.org/abs/0803.0410[arxiv:0803.0410 [ hep - ph ] ] . h. fujii , k. itakura , and a. iwazaki , `` instabilities in non - expanding glasma , '' http://dx.doi.org/10.1016/j.nuclphysa.2009.06.020 [ _ nucl . phys . _ * a828 * ( 2009 ) 178190 ] , http://arxiv.org/abs/0903.2930[arxiv:0903.2930 [ hep - ph ] ] . k. dusling , t. epelbaum , f. gelis , and r. venugopalan , `` role of quantum fluctuations in a system with strong fields : onset of hydrodynamical flow , '' http://dx.doi.org/10.1016/j.nuclphysa.2010.11.009[_nucl . _ * a850 * ( 2011 ) 69109 ] , http://arxiv.org/abs/1009.4363[arxiv:1009.4363 [ hep - ph ] ] . k. fukushima , f. gelis , and l. mclerran , `` initial singularity of the little bang , '' http://dx.doi.org/10.1016/j.nuclphysa.2007.01.086[_nucl . _ * a786 * ( 2007 ) 107130 ] , http://arxiv.org/abs/hep-ph/0610416[arxiv:hep-ph/0610416 [ hep - ph ] ] . f. gelis and r. venugopalan , `` particle production in field theories coupled to strong external sources , '' http://dx.doi.org/10.1016/j.nuclphysa.2006.07.020[_nucl . * a776 * ( 2006 ) 135171 ] , http://arxiv.org/abs/hep-ph/0601209 [ arxiv : hep - ph/0601209 [ hep - ph ] ] . f. gelis and r. venugopalan , `` particle production in field theories coupled to strong external sources . ii . generating functions , '' http://dx.doi.org/10.1016/j.nuclphysa.2006.08.015[_nucl . phys . _ * a779 * ( 2006 ) 177196 ] , http://arxiv.org/abs/hep-ph/0605246 [ arxiv : hep - ph/0605246 [ hep - ph ] ] . f. gelis , s. jeon , and r. venugopalan , `` how particles emerge from decaying classical fields in heavy ion collisions : towards a kinetic description of the glasma , '' http://dx.doi.org/10.1016/j.nuclphysa.2008.11.011 [ _ nucl . phys . _ * a817 * ( 2009 ) 6189 ] , http://arxiv.org/abs/0706.3775[arxiv:0706.3775 [ hep - ph ] ] . t. lappi , `` energy density of the glasma , '' http://dx.doi.org/10.1016/j.physletb.2006.10.017[_phys . lett . _ * b643 * ( 2006 ) 1116 ] , http://arxiv.org/abs/hep-ph/0606207[arxiv:hep-ph/0606207 ] . e. iancu , k. itakura , and l. mclerran , `` a gaussian effective theory for gluon saturation , '' http://dx.doi.org/10.1016/s0375-9474(03)01477-5 [ _ nucl . * a724 * ( 2003 ) 181222 ] , http://arxiv.org/abs/hep-ph/0212123[arxiv:hep-ph/0212123 ] . a. krasnitz , y. nara , and r. venugopalan , `` gluon production in the color glass condensate model of collisions of ultrarelativistic finite nuclei , '' http://dx.doi.org/10.1016/s0375-9474(03)00636-5[_nucl.phys._ * a717 * ( 2003 ) 268290 ] , http://arxiv.org/abs/hep-ph/0209269 [ arxiv : hep - ph/0209269 [ hep - ph ] ] . a. krasnitz , y. nara , and r. venugopalan , `` classical gluodynamics of high - energy nuclear collisions : an erratumn and an update , '' http://dx.doi.org/10.1016/j.nuclphysa.2003.08.004[_nucl.phys._ * a727 * ( 2003 ) 427436 ] , http://arxiv.org/abs/hep-ph/0305112 [ arxiv : hep - ph/0305112 [ hep - ph ] ] . h. fujii , k. fukushima , and y. hidaka , `` initial energy density and gluon distribution from the glasma in heavy - ion collisions , '' http://dx.doi.org/10.1103/physrevc.79.024909[_phys . rev . _ * c79 * ( 2009 ) 024909 ] , http://arxiv.org/abs/0811.0437 [ arxiv:0811.0437 [ hep - ph ] ] . j. berges , s. scheffler , and d. sexty , `` turbulence in nonabelian gauge theory , '' http://dx.doi.org/10.1016/j.physletb.2009.10.032[_phys . _ * b681 * ( 2009 ) 362366 ] , http://arxiv.org/abs/0811.4293[arxiv:0811.4293 [ hep - ph ] ] . j. berges , a. rothkopf , and j. schmidt , `` non - thermal fixed points : effective weak - coupling for strongly correlated systems far from equilibrium , '' http://dx.doi.org/10.1103/physrevlett.101.041603[_phys . lett . _ * 101 * ( 2008 ) 041603 ] , http://arxiv.org/abs/0803.0131 [ arxiv:0803.0131 [ hep - ph ] ] .

the discovery of persistence has recently generated considerable interest in understanding the statistics of first passage problems in spatially extended nonequilibrium systems , both theoretically and experimentally . the definition of persistence is as follows . let @xmath11 be a stochastic variable fluctuating in space and time according to some dynamics . the persistence probability is simply the probability @xmath12 that at a fixed point in space , the quantity @xmath13 $ ] does not change up to time @xmath7 . in many systems of physical interest a power law decay , @xmath14 , is observed , where @xmath15 is the persistence exponent and is , in general , nontrivial . the nontriviality of @xmath15 emerges as a consequence of the coupling of the field @xmath11 to its neighbours , since such coupling implies that the stochastic process at a fixed point in space and time is non - markovian . persistence phenomena have been widely studied in recent years @xcite . theoretical and computational studies include spin systems in one @xcite and higher @xcite dimensions , diffusion fields @xcite , fluctuating interfaces @xcite , phase - ordering dynamics @xcite , and reaction - diffusion systems @xcite . experimental studies include the coarsening dynamics of breath figures @xcite , soap froths @xcite , and twisted nematic liquid crystals @xcite . persistence in nonequilibrium critical phenomena has also been studied in the context of the global order parameter , @xmath16 , ( e.g. the total magnetization of a ferromagnet ) , regarded as a stochastic process @xcite . in the present work we consider spatially extended systems with a nonequilibrium field @xmath11 , which takes discrete values , at each lattice site @xmath17 . the field then evolves in time @xmath7 through interactions with its neighbours . the persistence probability at time @xmath7 is defined as the fraction of sites in which the stochastic field @xmath11 did not change its value in the time interval @xmath18 $ ] . the physical interpretation of @xmath11 could be for , for example , the coarsening spin field in the ising model after being quenched to low temperature from an initial high temperature , the sign of a diffusion field starting from a random initial configuration or the sign of the height , relative to the mean height , of a fluctuating interface . as the stochastic field evolves in time , such systems develop regions of persistent and nonpersistent sites . since the number of persistent sites decays with time according to @xmath6 , the persistent clusters shrink in size and number with a corresponding growth in the size of the nonpersistent regions . recently , manoj and ray ( mr ) @xcite have studied such spatially extended systems in one dimension ( 1d ) within the context of the @xmath19 reaction diffusion model , which is equivalent to the 1d ising model . they found that the length scale which characterises the interval sizes between persistent clusters _ apparently _ has a different time - dependence from the length scale characterising the walker separations ( or spin correlations , in the ising representation ) , and futhermore seems to depend on the initial walker density . if true , this result would be surprising . naively , one would expect the initial walker density to be irrelevant to the asymptotic dynamics , and the coarsening length scale , set by the mean distance between walkers , to be the relevant length scale for all spatial correlations . we shall show that the former expectation is correct , but the latter , in general , is not . in this paper we generalise , and reinterpret , this study in the context of the @xmath2-state potts model , which has @xmath2 distinct but equivalent ordered phases . the @xmath19 model corresponds to the @xmath20 potts model , i.e. the ising model , with the walkers identified as the domain walls between up and down ising spins . in the @xmath2-state potts model the random walkers represent domain walls between potts states . at each time step , every walker hops randomly left or right . persistent potts sites are those at which the potts state has never changed , i.e. sites which have never been crossed by a walker . if two walkers occupy the same site they either _ coalesce _ , with probability @xmath21 , or _ , with probability @xmath22 , these numbers being the probabilities that the states on the farther sides of the walkers are the same ( annihilation ) or different ( coalescence ) . the persistence probability ( the fraction of sites that have never been jumped over by a walker ) decays as a power of time , with a @xmath2-dependent exponent , @xmath23 . among the questions which emerge naturally in such a study are ( i ) what is the dominant length scale in the problem , as far as the persistent structures are concerned ? ( ii ) what is the nature of the spatial distribution of the persistent sites ? ( iii ) what is the average size of a persistent cluster ? an important point to make at the outset concerns the different length scales associated with the walkers and with the persistent sites . the walker density decays as @xmath24 for any @xmath2 , so the mean distance between walkers grows as @xmath25 . the density of persistent sites decays as @xmath26 , so the mean distance between these sites grows as @xmath27 . this simple observation immediately suggests that the spatial structure of the persistent sites for @xmath8 , where @xmath28 is the larger length scale , will be very different from when @xmath29 , where @xmath30 is the larger length . this is precisely what we find : the characteristic length scale , @xmath31 , controlling the distribution of the _ intervals _ between clusters of persistent sites , is given by @xmath32 . for the ising case studied earlier @xcite , one has @xcite @xmath33 , which is ` close ' to @xmath34 . we believe the proximity of these two exponents is responsible for the apparently non - universal behavior in the ising system reported in mr . derrida et al . @xcite have obtained an exact expression for @xmath15 for 1d potts models with arbitrary @xmath2 : @xmath35 ^ 2\ . \label{theta(q)}\ ] ] the value of @xmath2 corresponding to @xmath36 is @xmath37 = 2.70528\ldots$ ] , so @xmath38 for all integer @xmath39 . note that the probabilistic algorithm for implementing the potts model through the annihilation or coalescence of random walkers ( domain walls ) allows @xmath2 to be a real number , @xmath40 , while an equivalent ising spin representation ( see section iv ) of the potts persistence problem even allows @xmath41 ! by this means means we can explore a range of @xmath15 above and below 1/2 . our main result is that the scale length controlling the distribution of interval sizes between persistent clusters is given by @xmath10 when @xmath29 , but by @xmath9 when @xmath8 . we find no evidence for any dependence of the asymptotic scaling distribution on the initial walker density ( other than through nonuniversal amplitudes ) . this paper is organised in the following manner . in section ii we outline a scaling phenomenology within which the above questions can be addressed . in section iii we give exact results for the @xmath42 potts model . finally , in section iv , we present extensive numerical simulations for the @xmath2-state potts model which confirm our predictions based on general scaling arguments and the exact large-@xmath2 results . section v concludes with a discussion and summary of the results . the basic scaling phenomenology was introduced by manoj and ray , although we will adopt a slightly different notation . we are interested in the distribution of the sizes of the _ intervals _ between persistent sites . we define an interval size , @xmath1 , as the number of _ non - persistent _ sites between two consecutive persistent sites . if the interval size is zero , the sites belong to the same _ cluster _ of persistent sites , and we henceforth consider only intervals of non - zero size . we define @xmath0 to be the number of intervals ( per site ) of size @xmath1 at time @xmath7 a natural dynamical scaling assumption is @xmath43 where @xmath44 is the ` characteristic ' length scale at time @xmath7 . ( a dynamical exponent @xmath45 is conventionally defined via @xmath46 , rather than @xmath47 . here we are following the notation of mr ) . the rationale for the scaling form ( [ scaling ] ) is as follows . the number of non - persistent sites ( per site ) is given by @xmath48 but since the number of persistent sites ( per site ) , @xmath12 , decays to zero as @xmath6 , and @xmath49 , it follows that @xmath50 for @xmath51 . converting the sum ( [ np ] ) to an integral , with lower limit zero ( valid as @xmath52 provided the integral converges ) , we see that the prefactor @xmath53 in ( [ scaling ] ) is precisely what is needed to satisfy the required condition @xmath54 for @xmath51 . consider next the number , @xmath55 , of persistent clusters per site . since the number of clusters is equal to the number of intervals , @xmath56 is given by @xmath57 converting the sum to an integral ( with lower limit zero ) , using ( [ scaling ] ) and assuming the integral converges , gives @xmath58 . thus the mean distance between persistent clusters , i.e. the mean interval size , increases as @xmath47 . is this reasonable ? let us recall that there are two length scales in the system , @xmath59 , the mean distance between walkers , and @xmath60 , the mean distance between persistent sites . to make further progress we make the following two assumptions , both of which are confirmed by our numerical studies , by exact results for @xmath61 , and by heuristic arguments to be expounded below : ( i ) the mean cluster size tends to a constant for @xmath51 , and ( ii ) the length scale @xmath62 that controls the interval size distribution is the _ larger _ of @xmath30 and @xmath28 , i.e. @xmath63 . from assumption ( i ) we deduce that @xmath64 , i.e. @xmath65 , which is consistent with assumption ( ii ) provided @xmath8 . what if @xmath29 ? then we still require @xmath64 , but now @xmath66 , so the result @xmath58 , derived from ( [ cl ] ) breaks down . going back to ( [ cl ] ) , we infer that this breakdown requires that the conversion of the sum to an integral be invalid the sum must have its dominant contribution from @xmath1 of order unity , rather than @xmath1 of order @xmath47 . this in turn requires that the scaling function @xmath67 in ( [ scaling ] ) have a singular small-@xmath17 limit of the form @xcite @xmath68 with @xmath69 . using this form in ( [ scaling ] ) , with @xmath66 , and inserting the result into ( [ cl ] ) , gives @xmath70 . comparing this with @xmath64 fixes the value of @xmath71 : @xmath72 we first present a heuristic argument for the result @xmath73 . consider first the case @xmath8 . in this regime , the distance between clusters is much greater that the distance between walkers , i.e. there are , on average , many ( @xmath74 ) random walkers between each consecutive pair of persistent clusters , as illustrated in figure [ fig0 ] . however , walkers separated by distances large compared to @xmath10 are essentially uncorrelated , because any correlations are mediated by the random walkers , and correlations between walkers decay on the length scale @xmath10 . therefore one expects the intervals between clusters , measured on the scale @xmath9 , to be independent random variables , and the interval size distribution to scale with this length . this argument also suggests that the locations of the clusters as a function of position on the lattice are described by a poisson process , i.e. the interval size distribution is an exponential , @xmath75 , with @xmath76 , for @xmath8 . these expectations are borne out by our numerical studies . for @xmath29 , the opposite is true : there are many clusters between each pair of walkers ( see the @xmath77 snapshot in figure [ fig0 ] ) . the largest length scale is set by the walker spacing , @xmath59 , and the interval size distribution will scale with this length . on smaller scales , there will be residual structure in the cluster distribution left from an earlier epoch when the number of walkers was larger . indeed , on these smaller scales the persistent sites form a fractal set @xcite . consider two sites separated by a distance @xmath78 . the probability , @xmath79 , that _ both _ are persistent has the scaling form , @xmath80 where @xmath81 for @xmath82 , since for @xmath83 the sites are uncorrelated . the residual structure remaining on scales @xmath84 , however , means that if the first site is persistent , which occurs with probability of order @xmath26 , the probability that the second one is also persistent depends only on @xmath78 . this implies @xmath85 for @xmath86 . the number of persistent sites within a distance @xmath87 of a given persistent site can therefore be estimated as @xmath88 , where @xmath89 is the fractal dimension of the persistent set . clearly this result only makes sense for @xmath29 , since @xmath90 can not be negative . this suggests @xmath91 for @xmath92 , corresponding to point - like objects , i.e. isolated finite clusters . for @xmath61 , for example , where @xmath93 , we find that there is typically one cluster of persistent sites , with a fixed mean size , per scale length @xmath94 . we now present a heuristic argument in support of our claim that the mean cluster size , @xmath95 , approaches a constant as @xmath51 for all values of @xmath15 . the initial steps of the argument follow the approach of mr . the total number of clusters per site can be written as @xmath96 where @xmath97 is the fraction of walker sites which have never been visited by a walker , i.e. the persistence probability for walker sites . such sites form ` spacers ' between each pair of persistent potts states in a cluster , there being exactly one more persistent potts site than persistent walker site in each cluster . the mean cluster size is @xmath98^{-1}\ . \label{lav}\ ] ] we expect , on universality grounds , that the exponents describing the decay of @xmath12 and @xmath97 should be the same ( i.e. @xmath15 ) , but the ( nonuniversal ) amplitudes will be different , i.e. @xmath99 , while @xmath100 , with @xmath101 . inserting these forms in ( [ lav ] ) gives the limiting value of the mean cluster size as @xmath102 . this number is nonuniversal and is determined by the initial distribution of cluster sizes . many of the general features of the dynamics for @xmath104 are exemplified by the @xmath105 limit , for which the walker dynamics reduces to @xmath106 , i.e. the walkers always aggregate on contact . at @xmath107 all sites are persistent . the random walkers initially present divide the sites into clusters of persistent sites . clearly no cluster can increase in size . consider a cluster of initial size @xmath108 . we first calculate the probability density , @xmath109 , that the cluster survives to time @xmath7 and has size @xmath110 . the key point is that , for @xmath106 dynamics , we need only consider the two walkers on either side of the initial cluster , since subsequent coalescence processes do not affect the random walk dynamics of these two walkers . the cluster and the two walkers can therefore be treated in isolation . for simplicity we begin by treating the continuum limit in which the walkers are regarded as continuous - time random walkers on a continuous space . this should be correct in the limit @xmath111 , an assertion we can subsequently test . let the ends of the initial cluster ( and the two walkers ) be located at @xmath112 and @xmath113 . each walker obeys an equation of the form @xmath114 , where @xmath115 is a gaussian white noise with mean zero and correlator @xmath116 . we first write down the probability distribution @xmath117 of the rightmost excursion , @xmath118 , up to time @xmath7 , of the left walker . an elementary calculation gives @xmath119 for the cluster to survive , we require @xmath120 , so in the limit @xmath121 the exponential factor in ( [ p_r ] ) can be dropped . the probability distribution , @xmath122 , of the leftmost excursion of the right walker ( from its initial position ) is given by a similar expression . clearly @xmath123 is the residual size of the cluster at time @xmath7 , where @xmath124 means the cluster has disappeared . the probability density @xmath109 that the cluster has survived and has size @xmath110 is therefore given , for @xmath121 , by @xmath125 the probability , @xmath126 , that the cluster survives till time @xmath7 is given by @xmath127 immediately we see that the mean interval between surviving clusters grows as @xmath7 ( @xmath128 , since @xmath129 ) , which is much larger than the mean interval between walkers , which grows as @xmath10 . the mean length of surviving clusters , of given initial size @xmath108 , in the @xmath51 limit is @xmath130 if the initial clusters have a distribution of sizes , a straightforward calculation gives @xmath131 , where @xmath132 indicates an average over this distribution , while the fraction of clusters which survive is @xmath133 . the above calculations demonstrate that the mean cluster size approaches a constant at late times , and that the scale length for the sizes of the intervals between clusters grows as @xmath7 , and not as the naive scaling length @xmath10 associated with the underlying domain wall coarsening . in fact we can calculate the interval size distribution , @xmath0 , explicitly for this model . first recall that the surviving clusters are separated , at late times , by many ( @xmath134 ) walkers . the fate of a given cluster depends only on the nearest walker on either side . the motion of these walkers is uncorrelated with that of the walkers bordering other clusters , because the correlation length for walkers grows only as @xmath10 . therefore we can assume , in the scaling limit @xmath135 , @xmath51 , with @xmath136 fixed , that clusters survive _ independently _ with probability @xmath137 ( where we have specialized to initial clusters of fixed size @xmath108 ) . the probability distribution of the interval size @xmath1 between neighboring clusters is therefore exponential , with mean @xmath138 , while @xmath139 , which has the scaling form ( [ scaling ] ) with @xmath140 . this exponential form is in excellent agreement with simulation data presented in section iv . since the initial clusters used in the @xmath61 simulations are quite small ( 2 or 4 ) , the continuum limit is not expected to be quantitatively correct . to conclude this section , therefore , we quote the asymptotic mean cluster size for clusters of arbitrary initial size @xmath108 . the details are given in the appendix : the result is @xmath141 , which generalizes the continuum result @xmath142 . let us recall the zero - temperature coarsening dynamics of the 1d @xmath2-state potts model , starting from a random initial configuration . to maximize the speed of the program , we adopt two - sublattice parallel updating . the zero - temperature dynamics proceeds as follows : in each time step every spin on one of the sublattices changes its colour to that of one of its two neighbours with equal probability , the two sublattices being updated alternately . the dynamics of such systems can equivalently be formulated in terms of the motions of the domain walls as a reaction - diffusion model . the domain walls can be considered as random walkers performing independent random walks . whenever two walkers meet , they annihilate ( @xmath19 ) with probability @xmath22 or aggregate ( @xmath143 ) , to become a single walker , with probability @xmath21 . the persistence is the probability that a fixed point in space has not been traversed by any random walker . for @xmath20 , the particles only annihilate and hence this is equivalent to the glauber model , whereas in the @xmath61 limit the walkers only aggregate . this algorithm is , however , restricted to modelling @xmath2-state potts models with @xmath40 , since @xmath144 generates negative probabilities , while the general result ( [ theta(q ) ] ) allows any real @xmath145 . in order to simulate values of @xmath2 in the interval @xmath146 we can map the potts model ( as far as persistence properties are concerned ) on to an ising representation with a fraction @xmath147 of the sites initially pointing up , and initially persistent , while the down spins are nonpersistent . studying the persistence of the minority ( majority ) spins then corresponds to the case @xmath148 ( @xmath144 ) @xcite . the domain walls between the up and down spins behave as random walkers , annihilating each other on contact with probability one . as far as the persistence is concerned , the dynamics of the potts and ising models are completely equivalent @xcite . for @xmath148 , results obtained from both types of simulation are entirely consistent . the numerical simulations are performed on a 1d lattice of size @xmath149 with periodic boundary conditions . the ising spins , or potts variables , occupy the odd sites of the lattice , while the random walkers are restricted to even - numbered sites . therefore the effective size of the lattice , in terms of ising / potts spins , is @xmath150 . each walker jumps to an adjacent even - numbered site with equal probability , turning any persistent site it hops over into a nonpersistent site . the positions of all walkers are updated simultaneously at each monte - carlo step . the initial configurations are chosen to eliminate the possibility of any ` crossover ' of random walkers when they jump . this can be done by placing them only on sites whose positions on the lattice are of the form @xmath151 , where @xmath1 is an integer . the walkers then occupy subsets of the sites @xmath151 and @xmath152 alternately . for the potts simulations ( @xmath148 ) , the lattice is initially completely persistent . the random walkers are periodically laid down , in clusters of uniform initial size @xmath108 . the walkers then execute random walks , aggregating or annihilating according to the prescribed probabilities , @xmath21 and @xmath22 respectively . the simulations are performed for initial cluster sizes of 2 and 4 . the values of @xmath2 chosen are @xmath153 and 4 , for which the corresponding values of @xmath15 are @xmath154 , @xmath155 , and @xmath156 . for @xmath157 the ising representation is employed , in which persistent sites ( up spins ) are periodically laid down on the lattice , with a fraction @xmath147 of all sites persistent , and the domain walls execute random walks annihilating according to @xmath19 . the value @xmath77 was chosen , for which @xmath158 . the simulations were performed for initial cluster sizes 4 and 8 . the ising simulations were also run for @xmath159 and @xmath160 . the results for scaling functions are consistent with the potts data , but the statistics are poorer due to the smaller number of clusters generated . for this reason , the data presented in figs . 3 - 9 were generated from the potts runs . all the simulations ( potts and ising ) were run for @xmath161 monte - carlo steps , and the results averaged over 30 independent runs . the simulations investigate the limiting values of the average sizes of the persistent clusters , and the distribution , @xmath0 , of intervals between persistent clusters with emphasis on its dynamic scaling form . while the first of these lends itself readily to direct measurement , the dynamic scaling form for @xmath0 manifests itself most clearly in the data through a study of the ( complement of the ) cumulative distribution . we define @xmath162 inserting the scaling form ( [ scaling ] ) and converting the sum to an integral in the scaling limit @xmath135 , @xmath51 with @xmath163 fixed but arbitrary , gives @xmath164 we now discuss the numerical results for the @xmath61 model . in figure [ fig1 ] we plot the mean cluster size against time for initial cluster sizes of 2 and 4 . the numerical results are summarized in table 1 . the numerics clearly support the exact results given in section iii [ and derived in the appendix as equation([l_av ] ) ] , namely @xmath165 where @xmath108 is the initial cluster size . in order to investigate the dynamic scaling form ( [ newscaling ] ) for @xmath61 , we plot , in figure [ fig2 ] , @xmath166 against @xmath136 for @xmath167 500 , 1000 , 2000 , 4000 , 8000 , for initial cluster size 2 , where we have used @xmath168 for @xmath61 . excellent data collapse occurs in agreement with the dynamic scaling form ( [ newscaling ] ) . [ cols="^,^,^",options="header " , ] . average cluster size at late times , @xmath169 , for various values of @xmath2 and various initial cluster sizes @xmath108 , where p or i indicate whether potts or ising representations were employed . in all cases the data were averaged over times between 1000 and 9000 monte carlo steps . in section iii we argued that the scaling function @xmath67 in ( [ scaling ] ) should be a simple exponential , which implies that @xmath170 in ( [ newscaling ] ) is also a simple exponential . to test this prediction , we plot @xmath171 against @xmath136 in figure [ fig3 ] . the data lie on the expected straight line except at the latest times , where the deviations from the line are scatter due to statistical noise . very similar results are obtained for initial cluster size 4 , confirming that the scaling function @xmath67 is independent of the initial walker density . in figure [ fig4 ] the mean cluster size is plotted against time for various finite values of @xmath2 and initial cluster size @xmath108 . the numerical results indicate that the average cluster size tends to a constant ( see table i ) in all cases , as expected on the basis of the heuristic argument in section ii . for @xmath172 , @xmath173 and the dominant length scale is still determined by the mean distance between persistent clusters . we therefore anticipate a scaling form of the kind given by ( [ newscaling ] ) with @xmath174 . the scaling plot presented in figure [ fig5 ] shows good data collapse with this value of @xmath45 , while figure [ fig6 ] indicates that @xmath170 , and therefore @xmath67 , is again a simple exponential . the data collapse is not as good as for @xmath61 , which we attribute to a poorer separation of the length scales @xmath28 and @xmath30 at the time scales available . this problem becomes more acute for @xmath175 ( see the discussion below ) . for @xmath175 , @xmath176 , so we plot ( figure [ fig7 ] ) @xmath177 against @xmath163 with @xmath178 . this time the scaling collapse is definitely not as good as for @xmath61 and @xmath172 , reflecting an apparent deviation from the dynamic scaling form ( [ newscaling ] ) . we believe that this deviation reflects the proximity of the length scales @xmath179 and @xmath59 , which are not sufficiently well separated on the timescales achievable in the simulation , and would disappear for asymptotically large times . for example , @xmath180 at @xmath181 , while the corresponding ratio is about 14 for @xmath172 and 90 for @xmath61 . a similar effect may explain the apparent deviations from universality ( e.g. an apparent dependence of @xmath45 on the initial walker density ) in the @xmath20 simulations of mr . in this case @xmath33 , and @xmath182 at @xmath183 , the largest time reached by the mr simulations . so in this case also there is not a strong separation of length scales at the times achieved in the simulations ( a point recently emphasized by manoj and ray @xcite ) . figure [ fig8 ] gives the log - linear plot for @xmath175 . again , the data are consistent with an exponential scaling function , except at small scaling variable where there is a perceptible deviation from a straight line . in this region , however , the data also do not scale at all well either ( see inset in figure [ fig8 ] ) , which may be another manifestation of the poor separation of length scales : the scaling limit requires @xmath135 , @xmath51 with @xmath163 fixed . in particular the condition @xmath184 , which should be satisfied for good scaling , is violated at the small values of @xmath163 where the upturn in the data occurs . we will return to this point in the discussion . to investigate the spatial distribution of the sites in the regime @xmath29 we study @xmath77 , for which @xmath185 . in figure [ fig4 ] we show that the mean cluster size is a constant for initial cluster sizes of 4 and 8 . in the regime @xmath29 the density of the persistent sites decays more rapidly than that of the walkers . the dominant length scale is therefore no longer given by the persistence length , @xmath60 , but is given instead by the mean separation , @xmath59 , of the the walkers . hence we expect asymptotic dynamic scaling of the form ( [ scaling ] ) and ( [ newscaling ] ) with @xmath66 . in figure [ fig9 ] we plot @xmath186 against @xmath187 for an initial cluster size @xmath188 . the numerical results give excellent data collapse , supporting the proposed dynamic scaling form . apart from a change of scale , very similar results are obtained for @xmath189 , supporting the universality of the scaling function , i.e. the independence of @xmath170 from the initial walker density . we choose @xmath77 rather than @xmath20 in order that the length scales @xmath28 and @xmath30 be reasonably well separated at late times : @xmath190 at @xmath181 . figure [ fig9 ] reveals a pronounced upturn at small @xmath1 , which can be seen more clearly in the log - linear plot of figure [ fig10 ] and its inset , suggesting a divergence for @xmath191 . this is , in fact , expected from the analysis of section ii , which predicts , for @xmath29 , @xmath192 for @xmath86 , with @xmath193 , i.e. @xmath71 is in the range @xmath69 . the scaling function @xmath170 is given by @xmath194 for @xmath86 , i.e. @xmath195 , where @xmath196 is the fractal dimension of the persistent set on scales smaller than @xmath30 [ equation ( [ df ] ) ] . according to this prediction , the product @xmath197 should approach a constant at small @xmath17 . this product is shown in figure [ fig11 ] . the small-@xmath1 divergence has clearly been removed , the function approaching a constant at small @xmath17 as predicted ( the erratic behaviour at very small @xmath17 is due to statistical noise ) . for @xmath77 all of the numerical simulations have been performed using the ising representation , with periodic initial conditions in which clusters of @xmath108 ` up spins ' are placed at uniform intervals in a ` down spin ' background , and occupy a fraction @xmath147 of the sites . for @xmath148 , the potts representation has mainly been used , but we expect all universal properties , such as exponents and scaling functions , to be the same for the two representations . this expectation is confirmed by the results for @xmath5 and the scaling function @xmath170 . in order to obtain a _ precise _ correspondence between the ising and the potts representations , however , it is necessary to run the ising simulations with random initial conditions , rather than periodic initial conditions which we have mostly employed , and to scale the axes appropriately with @xmath2 to account for the different interval sizes in the two cases . this is because the reaction - diffusion representation of the potts model , in which walkers annihilate or coalesce with probabilities which depend on @xmath2 , is only an exact representation if the potts states occur in a completely random sequence . this is not true for the periodic initial conditions employed in the ising representation , whose correlations spoil the exact correspondence between ising and potts simulations . our expectation , then , is that the ising representation with random and periodic initial conditions should give the same _ scaling functions _ as the reaction - diffusion implementation of the potts model , but non - universal amplitudes will be different for the periodic initial conditions . in order to test the exact equivalence between random ising and potts simulations we plot , in figure [ fig12 ] , the results for the @xmath172 model using the potts ( reaction - diffusion ) method . the lattice is initially all persistent and the initial cluster size is @xmath198 . this corresponds to a sequence of pairs of potts states in which the members of a pair are in same state , but there are no correlations between pairs ( except that neighboring pairs must be in different states ) . the random initial conditions for the ising simulation are generated from the potts initial state by setting all pairs in potts state ` 1 ' ( say ) to be up spins and all other pairs to be down spins . in the potts simulations we keep track of all persistent sites , while in the ising simulation we track the persistence of the spins initially ` up ' . the interval sizes ( between persistent clusters ) are then naturally larger by a factor @xmath2 in the ising simulations relative to the potts simulations . figure [ fig12 ] shows both scaling functions in log - linear form , for @xmath199 , with this factor of @xmath2 scaled out ( i.e. @xmath200 is plotted against @xmath201 for these data ) . the data overlap almost perfectly . the ising data for a periodic initial condition are consistent with the same scaling function ( exponential ) , but with a different amplitude . in this paper we have investigated the nature of persistent structures in the coarsening dynamics of 1d potts models . a central concept has been the existence of two characteristic length scales , the mean separation , @xmath59 , of domain walls ( or ` walkers ' ) and the ` persistence length ' , @xmath60 , which measures the mean separation of persistent sites or clusters . the focus of our attention has been the distribution function , @xmath0 , for the number of intervals ( between clusters of persistent sites ) of length @xmath1 . this distribution has the scaling form ( [ scaling ] ) , with characteristic length scale , @xmath47 , which is the _ larger _ of @xmath30 and @xmath28 , i.e. @xmath4 . within the 1d potts model both regimes , @xmath29 and @xmath8 , can be accessed by varying the number of potts states , @xmath2 . from the general result , ( [ theta(q ) ] ) , we see that @xmath5 is a monotonically increasing function with , @xmath202 marking the boundary between the two regimes . the regime @xmath8 is conceptually simpler . the @xmath61 limit can been solved exactly ( section iii ) , and results for the long - time limit of the mean cluster size obtained . the fact that the locations of the surviving clusters are statistically independent in the scaling regime leads to the result that the scaling function @xmath67 in ( [ scaling ] ) is a simple exponential . we have argued that the same result should hold for all @xmath8 , since in this regime there are , on average , many walkers between each pair of persistent clusters . the _ relevant _ walkers for this argument ( those which are turning persistent sites in neighboring clusters nonpersistent ) are uncorrelated since their separations , of order @xmath28 , are much larger than the typical separation , @xmath30 , of neighboring walkers . on this basis we expect the asymptotic scaling function @xmath67 to be exponential for all @xmath8 ( i.e. @xmath203 ) . the data for @xmath172 ( figure [ fig6 ] ) and @xmath175 ( figure [ fig8 ] ) are consistent with an exponential form for @xmath204 , although the scaling is not perfect and there is a small upturn in the scaling function at small scaling variable for @xmath175 . we attribute these features to an imperfect separation of length scales on the time scales achieved in the simulations , and conjecture that the true scaling function is exponential for all @xmath104 . the case @xmath29 is more tricky . in this case the persistent clusters outnumber the walkers . the scaling function @xmath67 is clearly _ not _ a simple exponential , though it seems ( figure [ fig10 ] ) to have an exponential tail . there is a small - argument singularity of the form @xmath192 , with @xmath193 . this is related to the fractal dimension , @xmath196 , of the persistent sites : @xmath205 . note that the borderline , @xmath206 , between the two regimes occurs at @xmath207 . the existence of a small-@xmath17 singularity for @xmath29 raises the possibility of an alternative scenario for @xmath104 , in which the @xmath208 singularity , with @xmath209 , persists for @xmath8 , where it becomes an integrable singularity . the small-@xmath17 singularity in @xmath170 would then take the form of a cusp : @xmath210 . we have not been able to rule out this scenario from the data , but think it unlikely for the reasons given elsewhere in this paper . further insight could be obtained if it were possible to perform an expansion around the soluble @xmath61 limit to first order in @xmath147 , but we leave this as a challenge for the future . we conclude by discussing briefly the possibility of the existence of these two qualitatively different regimes in spatial dimension @xmath211 . first we generalize the result ( [ df ] ) , relating @xmath90 and @xmath15 , to any dimension . starting from ( [ both ] ) , the result @xmath212 follows generally , and the number of persistent sites within a distance @xmath87 of a given site is estimated as @xmath213 , where @xmath214 . generalizing still further , if the coarsening exponent is @xmath215 , rather than @xmath34 , the result is @xmath216 ( the @xmath217 version of this result is given in @xcite , based on the same reasoning ) . clearly this result requires @xmath218 , since @xmath90 can not be negative . if this inequality is violated , as in the 1d potts model with @xmath104 , the persistent sites no longer have a fractal structure but become point - like objects , with mean density @xmath219 , where @xmath220 is the coarsening length scale . cases where @xmath221 are easy to find , e.g. in the coarsening of the 2d ising model @xcite , the 2d diffusion equation @xcite , or the time - dependent ginzburg - landau equation in 2d . these all exhibit fractal persistent structures with the expected fractal dimension @xcite . it would be interesting to look for examples , in addition to the 1d potts model , where one can have @xmath222 . this work was supported by epsrc ( uk ) . for a recent review see s. n. majumdar , curr . india * 77 * , 370 ( 1999 ) . b. derrida , a. j. bray , and c. godrche , j. phys . a * 27 * , l357 ( 1994 ) . b. derrida , v. hakim , and v. pasquier , phys . lett . * 75 * , 751 ( 1995 ) ; j. stat . 85 , 763 ( 1996 ) . d. stauffer , j. phys . a * 27 * , 5029 ( 1994 ) . s. n. majumdar , c. sire , a. j. bray , and s. j. cornell , phys . lett . * 77 * , 2867 ( 1996 ) ; b. derrida , v. hakim , and r. zeitak , ibid . * 77 * , 2871 ( 1996 ) . j. krug , h. kallabis , s. n. majumdar , s. j. cornell , a. j. bray , and c. sire , phys . e * 56 * , 2702 ( 1997 ) . a. j. bray , b. derrida , and c. godrche , europhys . * 27 * , 175 ( 1994 ) ; p. l. krapivsky , e. ben - naim , and s. redner , phys . e * 50 * , 2474 ( 1994 ) ; s. n. majumdar and c. sire , phys . * 77 * , 1420 ( 1996 ) . j. cardy , j. phys . a * 28 * , l19 ( 1995 ) ; e. ben - naim , phys . e * 53 * , 1566 ( 1996 ) ; m. howard , j. phys . a * 29 * , 3437 ( 1996 ) ; s. j. cornell and a. j. bray , phys . e * 54 * , 1153 ( 1996 ) . m. marcos - martin , d. beysens , j - p . bouchaud , c. godrche , and i. yekutieli , physica ( amsterdam ) 214d , 396 ( 1995 ) . w. y. tam , r. zeitak , k. y. szeto , and j. stavans , phys . lett . * 78 * , 1588 ( 1997 ) . b. yurke , a. n. pargellis , s. n. majumdar , and c. sire , phys . rev . e * 56 * , r40 ( 1997 ) . s. n. majumdar , a. j. bray , s. j. cornell , and c. sire , phys . lett . * 77 * , 3704 ( 1996 ) ; k. oerding , s. j. cornell , and a. j. bray , phys . e * 56 * , r25 ( 1997 ) . g. manoj and p. ray , preprint , cond - mat/9901130 ( 1999 ) ; preprint , cond - mat/0003203 ( 2000 ) . d. h. zanette , phys . e * 55 * , 2462 ( 1997 ) . g. manoj and p. ray , j. phys . a * 33 * , l109 ( 2000 ) . s. jain and h. flynn , preprint , cond - mat/0004148 ( 2000 ) . e. ben - naim , l. frachebourg , and p. l. krapivsky , phys . e * 53 * , 3078 ( 1996 ) . s. j. cornell and a. j. bray , unpublished . see http://theory.ph.man.ac.uk/@xmath223bray/pictures.html for some illustrations of the persistent structures which arise . in this appendix we compute the probability that a cluster of initial size @xmath108 survives to time @xmath7 , and the mean size of surviving clusters , for the process @xmath106 . first note that initially there is a random walker at each end of the domain . at each time step the walkers move independently left or right with probability @xmath34 , so we can treat each walker independently . consider , therefore , a single walker moving at discrete time steps on a discrete 1d lattice , starting , at time @xmath107 , at position @xmath78 . let the ` origin ' be the point @xmath224 , and let @xmath225 be the probability that the walker has not yet reached the origin at time @xmath7 . clearly , @xmath226,\ \ \ r \ge 2\ . \label{recurrence}\end{aligned}\ ] ] we are interested only in the limit @xmath51 . in this limit , we know from standard random walk theory that every @xmath225 decays like @xmath24 , with an @xmath78-dependent amplitude . to leading order in @xmath24 , therefore , the @xmath7-dependence drops out of equations ( [ recurrence ] ) , which then become equations for the amplitudes . by inspection , the solution in this regime is @xmath227 now consider a walker starting immediately to the right of a cluster of @xmath108 persistent sites . the probability that after @xmath7 steps the walker has jumped over exactly @xmath78 of these ( making them nonpersistent ) is @xmath228 , where the final result follows from ( [ leading ] ) , to leading order for large @xmath7 . the same result holds for a walker starting immediately to the left of the cluster . the probability that @xmath110 sites remain persistent after time @xmath7 is @xmath229 times the number of ways of partitioning the cluster of length @xmath108 into 3 sections , with the central section of length @xmath110 ( and zero lengths are allowed for the outer sections ) . this number is @xmath230 . so the probability of the cluster surviving and having length @xmath110 is @xmath231 a generalization of ( [ pl0 ] ) to the discrete system .

there have been a number of attempts to study hypernuclei containing not only nucleons but also hyperons with strangeness @xcite . it has been of high interest to investigate how the presence of strangeness broadens and/or modifies the knowledge achieved in the field of ordinary nuclear physics . it is a fundamental problem to understand properties of a quantum many - body system in terms of basic interactions between constituent particles . in hypernuclear physics a unified treatment of all the baryon - baryon interactions , especially nucleon - nucleon ( @xmath4 ) and hyperon - nucleon ( @xmath5 ) interactions , will be necessary if one wish to understand general properties of hypernuclei . however , our knowledge on the @xmath5 interaction is quite inadequate in contrast to that of the @xmath6 interaction due to experimental difficulties . in such a situation many - body theoretical studies of hypernuclei could provide alternative information on the @xmath5 interaction . it has been discussed that physics of hypernuclei is different from conventional nuclear physics in some aspects . first , the channel couplings such as the @xmath7-@xmath8 coupling play a significant role in the structure of hypernuclei in contrast to the very limited role of the @xmath9-@xmath10 coupling appeared in nuclear many - body systems @xcite . second , the anti - symmetric spin - orbit force , which is forbidden in ordinary nuclei , emerges in hypernuclei in addition to the symmetric spin - orbit force . the spin - orbit splitting of @xmath11 single - particle levels in @xmath11 hypernuclei was considered experimentally to be much smaller than their nucleonic counterpart @xcite . however , some recent studies as in ref . @xcite suggest that the splitting would not be so small as discussed earlier . it will be highly desirable for the spin - orbit splitting to be established in experiment . the magnitude of the spin - orbit splitting will put further constraints on the @xmath12 interaction . there have been several possibilities of solving the structure of hypernuclei . one possibility is to treat few - body or light hypernuclei by solving the faddeev equation @xcite or by applying the cluster model @xcite . another possibility is microscopic studies of hypernuclei with a larger mass number . while a few lower partial waves play a decisive role in few - body or light hypernuclei , higher partial waves could give some contributions in heavier hypernuclei . these microscopic studies of hypernuclei could make us understand interesting physics of their structure itself , and also help us in determining the @xmath5 interaction . microscopic structure calculation of heavier hypernuclei is usually performed by introducing a @xmath12 effective interaction . for this purpose the @xmath13-matrix theory has been applied in two ways : in some works an effective @xmath5 interaction has been derived by simulating a nuclear matter @xmath13-matrix . in this approach the non - locality and energy dependence of the @xmath13-matrix are neglected , and the effective @xmath5 interaction is represented in the two- or three - range gaussian form with the adjustable fermi momentum @xmath14 . this type of the effective interaction has been referred to as the yng interaction @xcite . the yng interaction has been successfully used in many of hypernuclear structure calculations @xcite . it seems to us that the yng interaction has made an important contribution to the considerable progress in overall description of hypernuclei . the other approach , in the framework of the @xmath13-matrix theory , is to calculate the @xmath13-matrix in finite hypernuclei by treating rigorously the pauli exclusion operator and making a self - consistent calculation of single - particle potential of hyperons @xcite . in contrast to studies of using the yng interaction this approach possibly gives a better description of the state dependence of the effective @xmath5 interaction . it has been established that the @xmath13-matrix can be used as a basic ingredient in constructing a microscopic many - body theory starting with bare interactions . in spite of a marked success in the applications of the @xmath13-matrix theory , we may say from a formal point of view that the @xmath13-matrix is energy - dependent and non - hermitian , and does not have the property of decoupling between a model space of low - lying two - particle states and its complement . the @xmath13-matrix is not considered to be an effective interaction in a formal sense of the effective interaction theory . in order to derive an effective interaction we should add some higher - order corrections such as folded diagrams @xcite . it would be desirable if we could have a theory describing many - body systems in terms of an energy - independent and hermitian effective interaction with the property of decoupling . two of the present authors , k. s. and r. o. , proposed a many - body theory , the unitary - model - operator approach ( umoa ) that was formulated on the basis of such an effective interaction . the umoa was applied to finite nuclei , @xmath15o and @xmath16ca @xcite , and some improvements have been attained . the method may be viewed as an alternative way of deriving an effective interaction in finite many - body systems . a unified study of hypernuclei and ordinary nuclei would be helpful in understanding many - body systems of baryons in a microscopic way . therefore , we extended the formulation of the umoa , in the previous work @xcite , to the description of hypernuclei and made a calculation of the properties in @xmath17o with a realistic @xmath12 interaction . the purposes of this work are to present a general formulation of the extended umoa and apply it to @xmath17o by employing some realistic @xmath12 interactions . this paper is organized as follows : in section 2 we present a general formulation of the umoa for deriving the @xmath18 effective interaction . in section 3 the approximation procedure is given for making actual calculations . in section 4 we apply the umoa to @xmath17o by employing the nijmegen soft - core ( nsc ) @xcite , jlich model - ( j ) and model - b ( jb ) @xcite @xmath12 potentials . in section 5 we make some concluding remarks . we present a formulation of the umoa for applying to a calculation of the properties in @xmath11 hypernuclei , especially , closed - shell nucleus plus one @xmath11 systems . a main purpose is to give a method of calculating an effective interaction between @xmath11 and @xmath19 . in the derivation of the @xmath8 effective interaction we should note that , because of the small mass difference of about 77mev between @xmath11 and @xmath20 , the coupling of the @xmath8 and @xmath7 channels plays an important role . we first consider a two - body hamiltonian of a @xmath7-@xmath8 coupled system given by @xmath21 where @xmath22 @xmath23 and @xmath24 the terms @xmath25 and @xmath26 for @xmath27 and @xmath19 are the kinetic and single - particle potential energies of @xmath28 and @xmath19 , respectively , and @xmath29 is the bare two - body @xmath12 interaction including the @xmath8 and @xmath7 channels . the terms @xmath30 and @xmath31 are the rest masses of @xmath32 and @xmath33 , respectively . the hamiltonian @xmath34 describes a subsystem of interacting two particles @xmath8 or @xmath7 in the one - body potentials @xmath35 , @xmath36 and @xmath37 . in order to derive a @xmath8 effective interaction we introduce a unitarily transformed @xmath5 interaction given by @xmath38 where @xmath39 is the correlation operator defined in a space of two - particle states of @xmath8 and @xmath7 . the @xmath39 is an anti - hermitian operator satisfying @xmath40 in the formulation of the umoa the correlation operator @xmath39 is determined by the equation of decoupling for @xmath41 between a certain model space consisting of low - momentum two - particle states and its complement . this determination of the correlation operator @xmath39 is one of the characteristics of the present approach . for the determination of a @xmath8 effective interaction , we introduce a model space consisting of low - momentum @xmath8 states . let @xmath42 be the projection operator onto the model space . the complement of the @xmath43 space includes a space of high - momentum @xmath8 states referred to as the @xmath44 space and a space of all the @xmath7 states referred to as the @xmath45 space . the condition of decoupling for the transformed interaction @xmath41 is given by @xmath46 hereafter , we refer to the above equation as the decoupling equation . we assume that the one - body part @xmath47 in eq . ( [ eq : tvyn ] ) is decoupled as @xmath48 on the above assumption the decoupling equation ( [ eq : dec ] ) becomes @xmath49 in general , the decoupling equation ( [ eq : decr ] ) does not determine @xmath39 uniquely . the usual restrictive conditions are @xmath50 these conditions are called the minimal effect requirements @xcite , which mean that the transformation exp@xmath51 does not induce unnecessary transformation within each of the @xmath52 and @xmath53 spaces . in the general theory of constructing a hermitian effective interaction , the solution for @xmath54 in eq . ( [ eq : decr ] ) has been known @xcite and is given by @xmath55 with latexmath:[\ ] ] the @xmath59 in eq . ( [ eq : oyn ] ) is the biorthogonal state of @xmath60 satisfying latexmath:[\[\begin{aligned } \label{eq : bio } \langle \overline{\psi_{{\sst { \it \lambda n}}k } } where @xmath60 is the @xmath52-space component of @xmath57 defined by latexmath:[\[\begin{aligned } \label{eq : psc } that the solution @xmath39 in eq . ( [ eq : syn ] ) is determined dependently on the choice of a set of @xmath63 eigenstates @xmath64 as given in eq . ( [ eq : oyn ] ) . we choose a set of @xmath65 so that they have the largest @xmath52-space overlaps among all the eigenstates in eq . ( [ eq : eige ] ) . this is a usual choice in the derivation of the effective interaction . with the solution @xmath39 in eq . ( [ eq : syn ] ) , the @xmath8 effective interaction in the @xmath52 space is given by @xmath66 it has been known that this @xmath8 effective interaction can be written explicitly @xcite as @xmath67 where @xmath68 ( @xmath69 ) and @xmath70 ( @xmath71 ) are defined through the eigenvalue equation @xmath72 and @xmath73 is given by @xmath74 the @xmath75 is a @xmath42-space operator that agrees with the effective interaction of non - hermitian type . equation ( [ eq : evln ] ) is a formula for converting the non - hermitian effective interaction to the hermitian one . the term @xmath76 in eq . ( [ eq : shyn ] ) is introduced as a self - consistent potential of @xmath32 calculated with the @xmath8 effective interaction @xmath77 as @xmath78 ( @xmath79 ) and @xmath80 are the single - particle states of @xmath32 and @xmath19 , respectively . from eqs . ( [ eq : evln])-([eq : nhevln ] ) we see that the @xmath8 effective interaction @xmath81 can be calculated once the operator @xmath82 is given . the operator @xmath82 is determined completely , as given in eq . ( [ eq : oyn ] ) , by solving the eigenvalue equation ( [ eq : eige ] ) and selecting @xmath63 eigenstates @xmath65 with the largest @xmath52-space overlaps . it is noted here that eqs . ( [ eq : syn])-([eq : ul ] ) give a set of equations that determines the @xmath8 effective interaction @xmath77 self - consistently with the single - particle potential @xmath76 . here we discuss a calculation procedure for obtaining the @xmath8 effective interaction acting in a small model space of low - lying @xmath8 states . we use the harmonic oscillator ( h.o . ) wave functions as basis states . it is desirable that the single - particle potentials of @xmath83 , @xmath33 and @xmath19 in intermediate states are taken into account in a self - consistent way . however , as our main concern is to derive a @xmath8 effective interaction , the self - consistent treatment of @xmath35 would have predominant importance in the present calculation . for this reason we make a self - consistent calculation only for @xmath35 , and as for @xmath37 we employ a fixed potential calculated previously for closed shell nuclei in the umoa @xcite . the effect of the single - particle potential @xmath36 on the @xmath8 effective interaction has not necessarily been clarified , although some calculations with @xmath36 have been made for the @xmath13-matrices in hyperonic nuclear matter @xcite . in the actual calculations made so far the single - particle potential @xmath36 has been considered to be rather shallow @xcite . although we have not known clearly how much the effect of @xmath36 on the @xmath8 effective interaction , we here assume @xmath84 . in solving the self - consistent equation for @xmath35 with @xmath81 we should be careful in the following two respects : ( a ) in the calculation of energies of intermediate states in propagators we should use @xmath35 that is given for @xmath83 states including high - momentum states . in order to do this we need a @xmath8 effective interaction acting in a large model space , because in general the @xmath44 space contains high - momentum @xmath8 states and the @xmath85-space effective interaction @xmath86 is not always well behaved . we thus make a calculation by introducing two model spaces , namely , large and small model spaces . the @xmath8 effective interaction in each of the model spaces is calculated by following a two - step procedure . in the first - step procedure , we define a large model space and calculate the @xmath8 effective interaction and single - particle potential of @xmath83 in high - lying states . the @xmath8 effective interaction acting in this large model space will not be suitable for the effective interaction in a usual sense . by introducing a small model space and using the @xmath8 effective interaction determined in the first - step calculation , we proceed to the second - step calculation where we calculate a @xmath8 effective interaction acting in this small model space of low - lying @xmath87 states . ( b ) it is important in actual calculations how to determine the @xmath88 and @xmath85 spaces because an effective interaction is determined dependently on the choice of the @xmath88 and @xmath85 spaces . if a state in the @xmath88 space mixes strongly with @xmath85-space states in the eigenstate in eq . ( [ eq : eige ] ) , the matrix element of @xmath89 or @xmath54 defined in sec . 2 becomes large and as a result higher - order many - body correlations @xcite give rise to non - negligible contributions . in general , the strong mixing takes place when some of the @xmath88- and @xmath85-space states are quasi - degenerate in energy . for this reason , we should choose the @xmath88 and @xmath85 spaces which are well separated in energy . with due regard to points ( a ) and ( b ) we perform a calculation of the effective interaction through the two - step procedure . in the following subsections we present the two - step method in detail . let us write a two - particle state of @xmath83 and @xmath19 as the product of the h.o . single - particle wave functions in the usual notation as @xmath90 we define a model space denoted by @xmath91 and its complement @xmath92 with a boundary number @xmath93 as @xmath94 we also define a space of all the @xmath7 states denoted by @xmath95 as @xmath96 note that the model space @xmath91 does not contain the @xmath7 states . if we make a graph of two - particle states , the @xmath97 , @xmath92 and @xmath98 spaces are represented in fig . 1 . the @xmath93 means a number such that @xmath99 gives the maximum kinetic energy of the @xmath8 two - particle states in the @xmath97 space . the @xmath100 denotes the unit of the h.o . energy which we use commonly for @xmath32 , @xmath33 and @xmath19 . as we require to introduce a sufficiently large model space in the first - step calculation , we should choose @xmath93 as large as possible . another requirement is that the states in the @xmath91 space should be separated in energy from the states in the @xmath101 space . for this reason , we adopt a model space @xmath97 of a triangular shape . this choice of the @xmath97 space guarantees the separation in energy . [ fig:1 ] = 13 cm in fig . 1 the spaces @xmath102 for @xmath103 and @xmath33 , specified by the numbers @xmath104 and @xmath105 , stand for the excluded spaces due to the pauli principle for nucleons . the number @xmath106 is defined as @xmath107 with the quantum numbers @xmath108 of the highest occupied orbit in the closed - shell core nucleus . the @xmath109 means a number such that @xmath110 gives the maximum kinetic energy of the @xmath12 two - particle states in the @xmath111 and @xmath112 spaces . we then write the @xmath111 and @xmath112 spaces as @xmath113 and @xmath114 we also define the sum of the excluded spaces as @xmath115 furthermore , by using the projection operators , @xmath101 , @xmath116 , @xmath95 and @xmath117 , we introduce projection operators onto the spaces of allowed @xmath12 states as @xmath118 @xmath119 and we also define @xmath120 considering the pauli - exclusion principle for nucleons the terms @xmath121 and @xmath122 should be removed in treating the problem of two - body correlations . however , if we remove all the matrix elements of these terms , this leads to overmuch counting of the pauli principle . the matrix elements which are diagonal in nucleon states should be restored even though nucleons are in occupied states . therefore , instead of @xmath123 , we should use a @xmath12 interaction @xmath124 defined as @xmath125 where @xmath126 denotes the set of the h.o . quantum numbers @xmath127 of an occupied state of a nucleon , and the operator @xmath128 is defined as @xmath129 the two - body @xmath12 equation to be solved in the first - step calculation is now written as @xmath130 where we define @xmath131 as @xmath132 the @xmath133 is the single - particle potential of @xmath32 to be calculated with the @xmath8 effective interaction self - consistently . the number @xmath134 stands for a set of quantum numbers to specify a two - body @xmath12 eigenstate . the unitarily transformed @xmath12 interaction in the first - step calculation is given by @xmath135 the decoupling equation for determining the correlation operator @xmath136 becomes @xmath137 the @xmath8 effective interaction @xmath138 in the @xmath97 space is then given by @xmath139 in the first - step procedure we solve the decoupling equation ( [ eq : dece1 ] ) by introducing some assumptions . as we take a sufficiently large number as @xmath140 , we may assume that the one - body potential of @xmath83 for a state in the @xmath101 space is negligible in comparison with the sum of the energies , @xmath141 . therefore , @xmath142 takes nonzero values only for states in the @xmath91 space , and we have @xmath143 the definitions of @xmath97 and @xmath92 in eq . ( [ eq : dtps ] ) allow us to divide a space of @xmath8 states written in relative and center - of - mass ( c.m . ) coordinates into the same subspaces @xmath97 and @xmath92 as @xmath144 where @xmath145 and @xmath146 are the h.o . quantum numbers of relative and c.m . motions , respectively . the @xmath147 and @xmath148 denote , respectively , spin and angular momentum defined through @xmath149 . the isospin @xmath150 is omitted because @xmath150 takes a constant @xmath151 in the @xmath8 system . a @xmath7 state in the @xmath152 space in fig . 1 is also written by the product of the h.o . relative and c.m . states . the @xmath5 interaction @xmath153 in eq . ( [ eq : vbyn ] ) is not diagonal in the c.m . quantum numbers @xmath154 because the projection operators @xmath97 , @xmath155 and @xmath156 are not diagonal in @xmath154 . thus , the @xmath8 effective interaction @xmath157 is not diagonal in @xmath154 . it has been shown , however , in the calculation of ordinary nuclei that @xmath154 non - diagonal elements of the @xmath13-matrix are generally small @xcite . therefore , we make an assumption that the @xmath154 non - diagonal matrix elements of @xmath157 can be neglected . the two - body eigenvalue equation ( [ eq : eige1 ] ) is solved for the relative motion with a fixed set of @xmath158 written by @xmath159 where we rewrite @xmath160 as @xmath161 the @xmath162 and @xmath163 are the kinetic energies of the @xmath5 relative and c.m . motions , respectively . the state @xmath164 expresses a superposition of the @xmath8 and @xmath7 h.o . states and gives a general form of the eigenstate when we consider the @xmath7-@xmath8 coupling and the @xmath12 interaction with the tensor and antisymmetric spin - orbit couplings . the @xmath165 is an additional quantum number specifying a two - body @xmath5 eigenstate . in general , the one - body part @xmath166 does not satisfies the assumption of decoupling in eq . ( [ eq : assume ] ) . however , it has been shown in the calculation in nuclei that if we take a larger number as @xmath167 , the effect of violating the assumption of decoupling becomes considerably smaller @xcite . we have confirmed that this non - decoupling effect becomes negligible in calculating nuclear effective interactions . we here assume this to be the case also in the determination of the @xmath8 effective interaction . furthermore , we introduce another approximation that the sum of @xmath168 and @xmath169 be diagonal in the c.m . quantum numbers @xmath170 and @xmath171 . this assumption has also been made sure to be acceptable in the previous works @xcite . in order to solve the eigenvalue equation ( [ eq : eige1r ] ) we need to calculate the matrix elements of @xmath172 in the relative and c.m . coordinates representation . however , the calculation of these matrix elements is difficult because the term @xmath172 can not be separated into a sum of operators written in relative and c.m . coordinates . therefore , as have made in the previous work @xcite , we employ the angle - average approximation and have @xmath173 where , as was assumed previously , @xmath174 for @xmath175 . the @xmath176 is the h.o . transformation bracket @xcite for the @xmath12 two - body state which depends on the rest masses for @xmath177 and @xmath33 . here the angle - average approximation means that , when a function @xmath178 is given with an angular momentum @xmath179 and its @xmath180-component @xmath181 , we approximate @xmath178 as @xmath182 . in the present case the matrix element on the left - hand side in eq . ( [ eq : uyn ] ) is dependent on the @xmath180-components @xmath183 and @xmath184 of @xmath148 and @xmath171 , respectively . the average over @xmath183 and @xmath184 leads to the expression given in eq . ( [ eq : uyn ] ) . this approximation is essentially the same as has been made to derive an angle - averaged pauli operator by wong and sauer @xcite . the use of the basis states @xmath185 makes it difficult to represent the operator @xmath155 , but it has been confirmed that the angle - average technique works well . we thus make an approximation to the operator @xmath155 as @xmath186 where the summation should be made on the conditions @xmath187 here the coefficient @xmath188 is defined as @xmath189 where in the summation the following conditions should be satisfied @xmath190 furthermore , we need an expression of @xmath191 given in eq . ( [ eq : qynd ] ) in the angle - average approximation @xmath192 where the summation should be made on the same conditions as in eq . ( [ eq : sumc1 ] ) . the coefficient @xmath193 is given by @xmath194 where @xmath195 and @xmath196 should satisfy the conditions given in eq . ( [ eq : sumc2 ] ) . once the set of the eigenstates @xmath197 in eq . ( [ eq : eige1r ] ) is solved , the @xmath8 effective interaction @xmath198 acting in the @xmath97 space is given according to eqs . ( [ eq : evln])-([eq : nhevln ] ) by substituting @xmath199 , where @xmath200 is defined as latexmath:[\[\begin{aligned } \label{eq : oyn1 } \hspace{-5mm}\omega _ { \sst { \it yn}}^{\sst ( 1 ) } = \sum_{k_{1},ll'ss'j_{r},n_{c}l_{c } } \overline{q}_{\sst { \it yn}}^{\sst ( 1 ) } \langle \overline{{\it yn};k_{1},ll'ss'j_{r},n_{c}l_{c } } the @xmath8 effective interaction @xmath157 thus determined , we calculate a new single - particle potential @xmath202 as an averaged matrix element of @xmath203 . the @xmath202 can be given explicitly in terms of the effective interaction @xmath157 as @xmath204 in the above treatment of the single - particle potential of @xmath32 in the first - step calculation , we do not consider the one - body spin - orbit splitting . the @xmath8 effective interaction @xmath205 in eq . ( [ eq : tvln1 ] ) is dependent on the single - particle potential @xmath133 , and @xmath133 is determined with @xmath205 as in eq . ( [ eq : ul1 ] ) . therefore , the equation for determining @xmath205 and @xmath133 should be solved simultaneously . the model space @xmath91 in the first - step procedure is chosen as a _ low - momentum _ space , but the @xmath91 space is still large when we wish to use @xmath205 in the calculation of bound or low excited states of @xmath32 hypernuclei . with @xmath205 , we further proceed to the calculation of the @xmath8 effective interaction acting in a smaller model space and correspondingly the single - particle potential of @xmath32 in low - lying states . we separate the @xmath91 space into the @xmath206 , @xmath207 , @xmath208 and @xmath209 spaces specified by the numbers @xmath140 , @xmath106 , and @xmath210 as shown in fig . 2 . the number @xmath211 is introduced to specify the uppermost bound state of @xmath32 . here these spaces are defined as @xmath212 the @xmath213 space is the excluded space due to the pauli principle for nucleons . we are principally interested in the @xmath8 effective interaction that gives a description of the bound states of @xmath83 . it would be desirable that such a @xmath8 effective interaction satisfies the condition of decoupling between two @xmath8 states consisting of the bound and unbound states of @xmath83 . this requirement is equivalent to the situation that the @xmath8 effective interactions should be decoupled between two @xmath8 states in the @xmath214 and @xmath209 spaces . therefore , we solve the decoupling equation for @xmath215 between the @xmath8 states in the @xmath214 and @xmath209 spaces . the space @xmath208 is not a space to be excluded due to the pauli principle . however , some states in the @xmath208 and @xmath214 spaces overlap each other in energy . therefore , we treat separately the effect of the coupling between the @xmath208 and @xmath214 spaces . actually the interaction between states in the @xmath208 and @xmath214 spaces induces the core polarization in the core nucleus , as discussed in refs . @xcite and @xcite . we take into account the effect of the core polarization in the usual perturbative calculation . = 6 cm in the second - step calculation the unitarily transformed interaction is given by @xmath216 where @xmath217 the term @xmath218 is the self - consistent potential of @xmath32 to be determined in the second - step procedure . the decoupling equation for determining the correlation operator @xmath219 in eq . ( [ eq : tvyn2 ] ) becomes @xmath220 we note here that the assumption of decoupling @xmath221 is satisfied exactly because the two two - body states in the @xmath214 and @xmath209 spaces do not contain a common single - particle state of @xmath83 or @xmath19 . the two - body eigenvalue equation to be solved in the second - step calculation is written as @xmath222 where @xmath223 denotes the total angular momentum , and @xmath224 a set of quantum numbers of a @xmath8 two - body eigenstate . it should be noted that this eigenvalue equation can be solved exactly in the @xmath225 space by the diagonalization of the hamiltonian written with the basis states @xmath226 if the @xmath227 space is chosen suitably as a small space . once a set of the eigenstates @xmath228 in eq . ( [ eq : eige2 ] ) is solved , the @xmath8 effective interaction @xmath229 in the @xmath214 space is given from eqs . ( [ eq : evln])-([eq : nhevln ] ) with @xmath230 defined by @xmath231 is calculated with the @xmath8 effective interactions , @xmath229 and @xmath157 , as @xmath232 we solve the set of equations for determining @xmath229 and @xmath233 iteratively until the result converges . we remark here that we employ @xmath157 as the effective interaction in the @xmath213 space in the calculation of @xmath233 . since the effective interaction in the @xmath213 space is unchanged in the transformation made in the second - step procedure , we may write the @xmath8 effective interaction in the @xmath213 space as @xmath234 . we , therefore , have used partly the @xmath87 effective interaction @xmath157 in the calculation of @xmath233 as in eq . ( [ eq : ul2 ] ) . first we note that the one - body hamiltonian @xmath235 in eq . ( [ eq : hln2 ] ) has non - diagonal terms because we did not make the hartree - fock calculation . these terms could bring about sizable corrections to the single - particle energies of @xmath236 . we adopt the same h.o . energy both for @xmath83 and @xmath19 , though it has been considered that the wave function of @xmath83 is basically different from that of @xmath237 . therefore , we must evaluate the corrections arising from the non - diagonal terms of @xmath235 . by taking the diagonal part of @xmath235 as the unperturbed hamiltonian we calculate these corrections perturbatively up to second order as shown in diagram ( a ) of fig . 3 . next , the interaction between the @xmath8 states in the @xmath214 and @xmath208 spaces induces the core polarization in the core nucleus , and is important in the calculation of the corrections . we here assume that these corrections can be calculated approximately , using the interaction @xmath238 it will be desirable to take into account the effect of intermediate nucleon states with high momentum . in this perturbative calculation we include only the nucleon states belonging to the @xmath208 space , that is , the nucleon h.o . state should satisfy @xmath239 , where @xmath240 are the h.o . quantum numbers of a @xmath83 state . the validity of this truncation of nucleon states will be checked in the numerical calculation in the next section by increasing the number @xmath241 . the diagrams corresponding these corrections are given in diagram ( b ) and ( c ) of fig . we performed calculations employing the jlich model - ( j ) , model - b ( jb ) @xcite potentials in the momentum - space representation made by b. holzenkamp and a. reuber . ] and nijmegen soft - core ( nsc ) @xcite potentials for the _ yn _ interaction . we used the h.o . basis states with a common @xmath242mev for @xmath32 , @xmath33 and @xmath19 . the dependence of the effective interaction and single - particle energies on the value @xmath100 has been already discussed in @xmath15o @xcite and we confirmed that the dependence was quite small around @xmath243mev if perturbation corrections were included . as for the nucleon single - particle potential @xmath244 , we used the fixed data @xcite calculated using the paris potential @xcite , and we have assumed that @xmath245 . the boundary numbers @xmath246 , @xmath247 and @xmath109 are taken as @xmath248 , @xmath249 and @xmath250 for @xmath0o . = 13.8 cm we present , in fig . 4 , the @xmath93 dependence of the single - particle energies of @xmath32 including the perturbation corrections shown in fig . 3 . the parameter @xmath93 specifies the @xmath91 space in which the @xmath8 effective interaction @xmath229 is determined rigorously without any approximation as discussed in sec . 3 . therefore , as @xmath140 becomes larger , the calculated result does more accurate . if the result converges at a certain number @xmath251 , we may say that the method of the two - step calculation of the @xmath8 effective interaction would be trustworthy . as shown in fig . 4 , the results are convergent and stable for @xmath252 for each @xmath12 potential . if we take @xmath93 larger than 6 , our model space is considered to be sufficiently large . we note that the model space with @xmath253 corresponds to that used in the @xmath13-matrix calculations made by halderson @xcite and by hao and kuo @xcite . .dependence of typical diagonal matrix elements @xmath254 on @xmath255 in units of mev , where @xmath256 and @xmath257 denote the @xmath32 and @xmath19 single - particle orbits , respectively , labeled as @xmath258 , @xmath259 and @xmath260 , and @xmath223 is the total angular momentum . [ cols="^,^,^,^,^,^,^,^ " , ] next , the partial - wave contributions to the first - order potential energy of @xmath32 in the 0@xmath1 state are given for the different treatments of the pauli - blocking effect . we see from table 4 that the contributions of the @xmath261 and @xmath262-@xmath263 channels are very different dependently on the @xmath5 interactions employed . in the @xmath261 channel the largest pauli - blocking effect is brought about in the result for the j potential . on the other hand , in the @xmath262-@xmath263 channel we have an extremely large pauli - blocking effect for the nsc potential . this result shows that the nsc potential has the strong @xmath7-@xmath8 coupling in the @xmath262-@xmath263 channel as discussed in ref . the absolute values contributing to the first - order potential energy of @xmath32 in the channels with angular momenta @xmath264 are much smaller than those in the @xmath261 and @xmath262-@xmath263 channels . in table 5 we display the single - particle energies of @xmath32 with the contributions of the perturbation corrections given in fig . 3 , where the pauli - blocking effect is treated accordingly to case ( 3 ) . from this table we see that the j and nsc potentials yield almost the same values of the first - order potential energy pe and the perturbation corrections in spite of having different partial - wave contributions as shown in table 4 . the single - particle energies obtained for the jb potential are most attractive among the @xmath265 interactions used . it should be noted here that the values of the ke plus pe in the 0@xmath266 states for three kinds of @xmath5 interactions are positive ( unbound ) . therefore , the perturbation corrections play a decisive role when we argue that the @xmath32 in the 0@xmath266 states is bound or not . correction ( a ) in fig . 3 is induced by the presence of non - diagonal terms in the one - body hamiltonian @xmath267 . this effect should be treated in the hartree - fock ( hf ) calculation . however , in the present calculation , we treat this effect perturbatively . these non - diagonal terms work to change the single - particle wave function of @xmath32 . therefore , this procedure is indispensable in this approach since we have taken a fixed value of @xmath100 commonly for @xmath32 and @xmath19 . this correction ( a ) contributes most attractively to the single - particle energies among all the corrections in fig . the magnitude of every correction is considerably small in comparison with the first - order potential energies for all the @xmath5 interactions used . all the perturbation corrections contribute , however , attractively to the single - particle energies of @xmath32 , and do not change the mutual spacings of the single - particle states significantly . this trend is very different from the situation observed in ordinary nuclei such as @xmath15o and @xmath16ca , in which the correction terms contribute to about 40@xmath268 of the calculated spin - orbit splitting of the 0@xmath266 or 0@xmath269 states @xcite . the role of many - body correlations , such as the core polarization , does not seem to be important in the spin - orbit splitting of @xmath32 . we have presented a formulation of the umoa to apply it to a structure calculation of hypernuclei . we have introduced a unitary transformation exp(@xmath270 ) , with the correlation operator @xmath270 , that describes two - body correlation of hyperon and nucleon . an equation has been given for determining the correlation operator @xmath270 as the equation of decoupling so that the unitarily transformed interaction does not have non - zero matrix elements between the model space @xmath88 of low - lying @xmath8 states and its complement @xmath85 . the effective interaction has been given by the projected interaction onto the @xmath88 space . the effective interaction thus determined has properties of being hermitian , energy - independent and decoupled between the @xmath88 and @xmath85 spaces . in comparison with the usual @xmath13-matrix approach , the use of such an effective interaction would have some advantages in describing many - body systems . the umoa has been applied to the calculation of the structure of @xmath0o . the @xmath8 effective interaction has been calculated by employing three @xmath12 potentials , namely , the j , jb and nsc potentials . with the @xmath8 effective interaction , we have calculated the single - particle energies of @xmath32 in the 0@xmath1 , 0@xmath2 and 0@xmath3 states . we have seen that how to treat the pauli - blocking effect is very important , especially when we solve the problem of the @xmath7-@xmath8 coupling . in the usual treatment one excludes all the interactions acting in a space of @xmath271 and @xmath272 states in which a nucleon is in occupied states . however , this treatment of the pauli - blocking effect would lead to overmuch counting . we have introduced a new treatment in which , while we exclude the interactions acting among forbidden @xmath5 states as in the usual treatment , we restore the interactions that are diagonal in occupied nucleon states . we have observed that this treatment of the pauli - blocking effect has given rise to a significant effect as compared with the usual treatment . the present results have shown that @xmath32 in the 0@xmath3 state is unbound for the j and nsc potentials and , on the other hand , bound for the jb potential . the spin - orbit splitting of the single - particle levels of @xmath83 in the 0@xmath266 states has been given by 1.41 , 1.57 and 1.66 for the j , jb and nsc potentials , respectively . these values are very large compared with the results in ref . it will be of high interest for the single - particle nature of @xmath32 in @xmath0o to be established experimentally . in the application of the umoa some problems have still remained . the unitarily transformed hamiltonian @xcite contains originally three - or - more - body interactions . the effects of these many - body interactions have been evaluated in the previous works of calculating nuclear properties , and we have concluded that the contributions of these many - body terms are much smaller than those of two - body interactions but sizable in some cases . we therefore should estimate the effects of these many - body interactions also in hypernuclei . another refinement to be made is that , in the present calculation , the single - particle energies of @xmath33 in intermediate states are neglected . however , the single - particle spectrum of @xmath33 has a possibility of giving rise to some contributions to the @xmath8 effective interaction . although the properties of the single - particle energies of @xmath33 have not always been made clear , we should investigate how much the spectrum of @xmath33 affects in the determination of the @xmath8 effective interaction . this problem is also a remaining task in the umoa . the authors would like to thank y. yamamoto and k. miyagawa for useful discussions and their help for using the nijmegen and jlich @xmath265 potential codes . one of the authors ( s. f. ) would like to express his thank to k. takada , m. kamimura and y. r. shimizu for their continuous encouragement and instructive discussions . a. gal , advances in nuclear physics , vol . 8 , eds . m. baranger and e. vogt ( plenum , new york , 1975 ) p. 1 . b. povh , ann . nuc . part . 28 ( 1978 ) 1 . h. band et al . , prog . 81 ( 1985 ) . h. band , t. motoba and j. @xmath273ofka , int . j. mod . a 5 ( 1990 ) 4021 , and references therein . t. motoba , y. akaishi and k. ikeda , prog . 117 ( 1994 ) . b. f. gibson and e. v. hungerford iii , phys . 257 ( 1995 ) 349 . b. f. gibson , i. r. afnan , j. a. carlson and d. r. lehman , prog . 117 ( 1994 ) 339 . w. brckner et al . , phys . . 79 b ( 1976 ) 157 . r. h. dalitz , d. h. davis , t. motoba and d. n. tovee , nucl . phys . a 625 ( 1997 ) 71 . k. miyagawa , and w. glckle , phys . rev . c 48 ( 1993 ) 2576 ; + k. miyagawa , h. kamada , w. glckle and v. stoks , phys . rev . c 51 ( 1995 ) 2905 . t. yamada , h. ohkura , t. motoba and k. ikeda , prog . 117 ( 1994 ) 65 . e. hiyama , m. kamimura , t. motoba , t. yamada and y. yamamoto , prog . 97 ( 1997 ) 881 . h. band and y. yamamoto , prog . theor . phys . 81 ( 1985 ) 9 . y. yamamoto , t. motoba , h. himeno , k. ikeda and s. nagata , prog . 117 ( 1994 ) 361 . h. band , prog . 66 ( 1981 ) 1349 . d. halderson , phys . rev . c 48 ( 1993 ) 581 . j. hao , t. t. s. kuo , a. reuber , k. holinde , j. speth and d. j. millener , phys . 71 ( 1993 ) 1498 ; + t. t. s. kuo and j. hao , prog . 117 ( 1994 ) 351 . m. hjorth - jensen , a. polls , a. ramos and h. mther , nucl . phys . a 605 ( 1996 ) 458 . i. vidaa , a. polls , a. ramos and m. hjorth - jensen , nucl . phys . a 644 ( 1998 ) 201 . t. t. s. kuo and e. osnes , in : folded - diagram theory of the effective interaction in nuclei , atoms and molecules , lecture notes in physics , vol . 264 , ( springer , berlin , 1990 ) . j. hao and t. t. s. kuo , phys . 264 ( 1996 ) 233 . k. suzuki , r. okamoto and h. kumagai , phys . c 36 ( 1987 ) 804 ; + k. suzuki and r. okamoto , prog . theor . ( 1994 ) 1045 . h. kumagai , k. suzuki and r. okamoto , prog . 97 ( 1997 ) 1023 . s. fujii , r. okamoto and k. suzuki , prog . 99 ( 1998 ) 151 . m. m. nagels , t. a. rijken and j. j. de swart , phys . rev . d 15 ( 1977 ) 2547 ; + p. m. m. maessen , t. a. rijken and j. j. de swart , phys . c 40 ( 1989 ) 2226 . b. holzenkamp , k. holinde and j. speth , nucl . phys . a 500 ( 1989 ) 485 ; + a. reuber , k. holinde and j. speth , nucl . phys . a 570 ( 1994 ) 543 . m. lacombe , b. loiseau , j. m. richard , r. vinh mau , j. ct , p. pirs and r. de tourreil , phys . c 21 ( 1980 ) 861 . b. h. brandow , in : effective interactions and operators in nuclei , lecture notes in physics , vol . 40 , ed . b. r. barrett ( springer , berlin , 1975 ) p. 1 . i. shavitt and l. t. redmon , j. chem . 73 ( 1980 ) 5711 . p. westhaus , int . j. quantum chem . 20 ( 1981 ) 1243 . k. suzuki , prog . 68 ( 1982 ) 246 . t. t. s. kuo , p. j. ellis , j. hao , z. li , k. suzuki , r. okamoto and h. kumagai , nucl . phys . a 560 ( 1993 ) 621 . y. yamamoto and h. band , prog . 83 ( 1990 ) 254 . schulze , a. lejeune , j. cugnon , m. baldo and u. lombardo , phys . b 355 ( 1995 ) 21 ; phys . c 57 ( 1998 ) 704 . k. suzuki and r. okamoto , prog . 76 ( 1986 ) 127 . e. m. krenciglowa , c. l. kung , t. t. s. kuo and e. osnes , ann . of phys . 101 ( 1976 ) 154 . b. buck and a. c. merchant , nucl . a 600 ( 1996 ) 387 . c. w. wong , nucl . phys . a 91 ( 1967 ) 399 . p. u. sauer , nucl . phys . a 150 ( 1970 ) 467 .

the unconventional temperature scaling of the in plane resistivity in high-@xmath3 cuprates has been of considerable interest for many years . it is now widely accepted that overdoped cuprates exhibit a @xmath4-linear resistivity at high temperature which crosses over to a @xmath5 dependence at lower temperature [ ] . the onset of the linear temperature dependence appears to coincide with the onset of superconductivity and the crossover temperature to a conventional @xmath5 dependence drops monotonically with the doping level to zero , roughly around optimal doping [ ] . hussey and collaborators measured the resistivity on a number of cuprates particularly the single layer tl2201 in the overdoped region [ ] and lsco into the underdoped region [ ] . very recently the barii et al . [ ] reported results for the resistivity for more cuprates , e.g. hg1201 and ybco over a wider density range . they found that when the resistivity was normalized to a give a resistance per cu@xmath2o@xmath2 plaquette , @xmath6 , universal results for all cuprates followed with identical coefficients . further they argued for a universal trend in both the linear @xmath7 and quadratic @xmath8 regimes , with the coefficients @xmath9 and @xmath10 proportional to @xmath11 , the inverse hole density . as barii et al . [ ] remarked , if one uses a drude formula for resistance , this behavior is indicative of a carrier density that is proportional to the density of holes rather than electrons . such behavior is consistent with a doped mott insulator scenario . anderson [ ] has argued that the cuprates as doped mott insulators should be in the strong coupling regime of the hubbard model at all hole densities . he has put forward the hidden fermi liquid [ hfl ] ansatz to describe the connection between the fermi surface before gutzwiller projection and the electronic state that results from the removal of doubly occupied states by gutzwiller projection . this hfl ansatz gives a resistivity with the observed temperature dependence , crossing over from a linear to a quadratic form as the temperature is lowered . an earlier viewpoint interprets the cuprate phase diagram in terms of an underlying quantum critical point separating the overdoped and underdoped regions . in these theories the effective interactions between electrons are mediated and controlled by the order parameter fluctuations associated with the quantum critical point [ ] . another approach recently put forward by kokalj and coworkers [ ] is based on an anisotropic marginal fermi liquid self energy . their phenomenological theory gives good fits to the resistivity and also to the angular dependent magnetoresistance [ admr ] results [ ] and a number of other properties . in an earlier study , which employed a functional renormalization group [ frg ] treatment of the two - dimensional hubbard model to describe the overdoped region , ossadnik et al . [ ] found good qualitative agreement with the experimental results on overdoped tl2201 for the temperature and angular dependence of the relaxation rate determined from the magnetoresistance [ ] . since overdoped tl2201 shows a conventional full fermi surface in angle - resolved photoemission spectroscopy ( arpes ) at zero magnetic field [ ] and in quantum oscillation experiments at finite magnetic fields [ ] a perturbative treatment is of interest . note that the frg calculations are evaluated only to one - loop order and therefore are quantitatively reliable only for weak to moderate values of the hubbard onsite interaction , @xmath12 . as a result , a quantitative comparison to the experimental data shows a substantial underestimate of the magnitude of the calculated temperature dependent resistance , e.g. see the comparison in kokalj and mckenzie [ ] . in their calculation of unusual transport behavior , ossadnik and collaborators analyzed the quasiparticle scattering vertex for overdoped cuprates within a functional renormalization group ( frg ) approach [ ] . they found an anisotropic term in the quasiparticle scattering rate with an angular form similar to the experiment and with an unconventional @xmath4-linear dependence which increased when approaching optimal doping starting from the overdoped side . they argued that this unusual term appearing in the imaginary part in the quasiparticle self - energy could lead also to an equivalent temperature dependence for the in - plane resistivity . the goal of this study is to expand this idea and move beyond a self energy analysis of the transport properties to a full calculation of transport properties using solutions of the boltzmann equation . this is possible within a frg approach since the method computes the strongly renormalized quasiparticle interactions . we determine the matrix elements for the scattering events within the frg framework presented in [ ] and use numerical techniques to solve the linearized boltzmann equation . a detailed numerical treatment is required in view of the strong anisotropy in the transport life times [ ] due to both the anisotropic quasiparticle interactions and the strongly angle dependent fermi velocities introduced by the anisotropy of the band structure . the importance of full transport calculations was stressed by hlubina and rice in an early paper [ ] . as will be discussed later , our calculations agree qualitatively with the unconventional temperature dependence but the magnitude of the resistance is substantially lower . this disagreement in the absolute values is in line with the restrictions on @xmath12 required to justify the one - loop approximation in the frg . another discrepancy between our numerical results and the experiment is the lack of scaling of the resistance coefficients , @xmath13 and @xmath14 , with @xmath11 the inverse hole density as reported by barii et al . note however , our results are for the overdoped samples . the data in fig . 5 of ref . [ ] deviate from the inverse doping scaling in the overdoping density range . in the case of the linear coefficient , @xmath13 the data show a linear decrease and vanish at the onset of superconductivity in line with the earlier results of hussey and coworkers [ ] . a good example of such an onset of a non - fermi - liquid - like @xmath4-linear contribution to the resistivity can be found in the resistivity measurements by hussey and coworkers [ ] upon reducing the hole concentration from the overdoped regime toward optimal doping in lsco ( cf . [ fig : experimental resistivity ] ) . they observed a strong growth of the temperature dependent contribution to the resistivity with a decrease in the hole doping together with a remarkable trend toward linear resistivity that goes down to lowest temperatures . they used strong magnetic fields to explore the normal state transport properties within the superconducting dome . the analysis of the temperature dependence of the resistivity was based on a second - order polynomial fit to the experimental data over a wide temperature range : @xmath15 . this allows to distinguish doping regimes with dominant linear versus dominant quadratic temperature dependence . we also solve the boltzmann equation to obtain the seebeck coefficient which describes the charge transport in response to a thermal gradient . standard fermi liquid theory predicts a metallic seebeck coefficient with a linear @xmath4-dependence . the experiments by lalibert and coworkers [ ] showed substantial deviations from the standard form with a complex density dependence . again our calculations reproduce the qualitative trends in the temperature and density dependence very well . in sec . [ sec : numerical results ] we show our numerical results and compare with the experimental data of refs . we also examine the validity of matthiessen s rule which relies on a decoupling of impurity and two - particle scattering in sec . [ sec : momentum relaxation mechanism in the cuprates and matthiessens rule ] . finally in sec . [ sec : the seebeck effect in overdoped lsco ] , we study the thermoelectric effect of lsco , use the seebeck coefficient as a sign of non fermi liquid transport and examine the evidence for quantum critical behavior . the numerical method that we have developed to study the transport properties of correlated materials such as lsco is explained in detail in the appendix , app . [ sec : model for the quasiparticle dispersion of lsco]-[sec : effective model for the normal state charge transport of the high - temperature superconductor lsco ] . we solve the semiclassical boltzmann equation taking the full angular and energy dependence of the distribution function into account . the two important inputs of the model are the explicit form of the band structure and the temperature dependent calculation of the quasiparticle scattering rates . we begin by reviewing our numerical results . we use a semiclassical model to describe the normal state charge transport of overdoped lsco . for the quasiparticle dispersion we used a two - dimensional tight binding model and a doping dependent hopping parameter renormalization scheme based on the values determined in ref . [ ] from the fermi surface in apres data on lsco . the use of phenomenological input for the quasiparticle dispersion is important because the renormalization of the hopping integrals is of two - loop order in the frg analysis that we use to determine the scattering vertex . the calculation of two - loop diagrams for the rg flow is , however , beyond the scope of this study . for details about the dispersion , cf . [ sec : model for the quasiparticle dispersion of lsco ] . the frg technique that we use to compute the renormalized quasiparticle scattering rates is outlined in app . [ sec : strongly renormalized quasiparticle interactions ] together with brief overview of the doping and temperature dependence of the scattering vertex and a discussion of the parameters that enter the rg calculation . taking the quasiparticle dispersion and the scattering vertex as input , the conductivity is computed by solving the linearized boltzmann equation with its full angular and energy dependence . due to the strong anisotropy in the scattering rates and the quasiparticle velocities , a full solution of the boltzmann equation rather than a single relaxation time approximation is required . the collision integral is computed by introducing an efficient discretization of the brillouin zone and the boltzmann equation is then solved numerically . this method of solving the boltzmann equation and discretizing the brillouin zone is explained in app . [ sec : effective model for the normal state charge transport of the high - temperature superconductor lsco ] . the in - plane resistivity in our model of the normal state of overdoped lsco is shown in fig . [ fig : rhovstemperaturelsco ] ( data for @xmath17 ) . we also plot the resistivity for different initial values for the unrenormalized onsite interaction parameter @xmath12 . the data corresponding to the different values of @xmath18 are derived from renormalized scattering vertices while the data for @xmath19 are computed for particles interacting via the bare onsite repulsion . for a comparison of these calculations , cf . [ subsec : discussion of the parameters ] . the values for the hole - doping @xmath20 are chosen to match the experimental measurements of fig . [ fig : experimental resistivity ] . our results compare well to many qualitative aspects of the experimental data of ref . first , there is the strong growth of the resistivity when reducing the hole doping from strongly overdoped values of @xmath21 to optimal doping around @xmath22 . second , there is clear evidence for a growing linear term in the temperature dependence of @xmath23 around optimal doping . for large values of the hole doping , @xmath16 displays a rather quadratic temperature dependence . on the other hand , for doping concentrations below @xmath24 a linear @xmath25-dependence becomes obvious . we analyze our data in the same scheme as in the experimental paper [ ] using fits with the form @xmath26 ( sec . [ subsubsec : linear plus quadratic fit ] ) or also by introducing a `` local scaling exponent '' @xmath27 , e.g. @xmath28 ( sec . [ subsubsec : local scaling exponent ] ) . for a quantitative comparison to the experiment we convert our two - dimensional data to the experimental units of a three - dimensional sample ( cf . app . [ sec : conversion of computation to experimental units ] ) . from fig . [ fig : rhovstemperaturelsco ] we find that our calculated values for @xmath16 at @xmath29 are roughly by a factor four or five smaller than in experiment ( cf . [ fig : experimental resistivity ] ) . an underestimation of the resistivity is consistent with the fact that our unrenormalized interaction strength @xmath30 is considerably lower than expected for lsco . in spite of the weak coupling approximation in our calculation of the scattering rates , the resistivity is in good qualitative agreement with the observed temperature dependence . next we analyze the numerical resistivity data using the polynomial fit up to second order @xmath26 , in analogy to cooper et al , cf . they fit their resistivity data over the entire temperature range and obtain the fit coefficients @xmath32 and @xmath33 as a function of the hole doping , cf . fig . [ fig : coefficientslsco ] ( left panel ) . they found the linear coefficient @xmath32 grows strongly with a decrease in doping while there is a constant or decreasing value of @xmath33 . the linear @xmath32 and quadratic @xmath33 coefficients from fitting our numerical results are shown in fig . [ fig : coefficientslsco ] ( right panel ) for the entire range of hole doping from @xmath34 to @xmath21 . the fit was performed for temperatures ranging from @xmath35 up to @xmath36 . for the comparison of the two coefficients we introduce an average temperature scale @xmath37 in the conversion to resistivity units . the coefficient representing the residual resistivity , @xmath38 , is discussed later . the direct comparison of our fit coefficients to the experiment yields very good qualitative agreement . the linear term shows a strong increase with reduced hole concentration toward optimal doping while above @xmath39 it becomes indistinguishable from zero . on the other hand , the quadratic contribution does not grow and even decreases almost down to zero toward optimal doping for larger @xmath12 . a vanishing @xmath33 indicates a within our model a purely linear temperature dependence . whether or not the quadratic coefficient of the fit to the experimental data vanishes is not really clear from ref . [ ] , although it appears that the @xmath33 determined with the conventional @xmath31 fit indeed decreases and finally vanishes around @xmath40 , while the quadratic coefficient determined from a parallel resistor fit remains constant . the coefficients in the overdoped regime disagree with the reported @xmath11 dependence by barii et al . as already mentioned in the introduction , in the overdoped regime , their data suggest a deviation from the remarkable @xmath11 dependence of the linear term in the underdoped regime . the experimental data of both ref . [ ] and [ ] indicate that the linear term eventually goes to zero for sufficiently large hole density , @xmath41 . an alternative analysis of the experiments was recently put forward by hussey et al . taking the derivative @xmath42 they distinguished two regimes which are not immediately obvious in the resistivity data , but in which @xmath16 scales differently with temperature . hussey and coworkers identified a high - temperature regime where @xmath43 is constant , i.e. the resistivity depends linearly on @xmath25 . toward lower temperatures @xmath44 deviates downward from this constant value . with this rather abrupt change one can associate a temperature @xmath45 , called coherence temperature in ref . [ ] which decreases with lowering the hole concentration and disappears probably around the same value of @xmath20 where the characteristic temperature for the pseudogap vanishes extrapolated from the underdoped side . in fig . [ fig : phase diagram and coherence temperature ] the doping dependence of @xmath45 measured in ref . [ ] is shown . it was experimentally shown that @xmath45 coincides with the loss of the quasiparticle coherence peak [ ] . plotting @xmath46 for different doping levels in our calculation , we can analogously identify distinct regimes in the temperature scaling of the resistivity , cf . [ fig : coherence_temperature ] ( left panel ) . there is a low - temperature regime with finite slope of @xmath46 , an intermediate temperature regime with finite , but smaller slope , and a high - temperature regime where @xmath46 is constant . the temperatures that separate these three regimes will be referred to as @xmath47 and @xmath48 . the temperature scales @xmath47 and @xmath48 are displayed in fig . [ fig : exponentphasediagram ] . the doping dependence of @xmath48 shows a very similar trend as @xmath45 . this is interesting because there is no real loss of quasiparticle coherence in our semiclassical model based on the boltzmann transport equation . note that we have not considered the values for @xmath47 and @xmath48 for @xmath49 as it becomes difficult to distinguish the regimes for @xmath40 it is interesting to relate the onset of the unconventional scaling , @xmath47 , to the proximity of the fermi energy to the van hove level , @xmath50 . we introduce the temperature scale @xmath51 , which is shown as the red squares in fig . [ fig : exponentphasediagram ] . the factor of 4 is introduced for convenience , as we consider only those low - energy states relevant for the transport , that live within an energy range of @xmath52 from the fermi level . the doping dependence of @xmath53 and @xmath47 is very similar and we conclude that proximity to a van hove singularity plays an important role in the development of non - fermi liquid behavior . to compare to the analysis in ref . [ ] we fit the high - temperature linear part of @xmath43 with a constant @xmath54 and the low - temperature regime by a linear function @xmath55 . in the high - temperature regime , @xmath54 represents the coefficient of the dominating linear term in the resistivity , while @xmath56 corresponds to the coefficient of the dominating linear term in the limit @xmath57 . the results for these coefficients are shown in fig . [ fig : coherence_temperature ] ( right panel ) . ( @xmath58 ) as a function of the hole doping of lsco . the coherence temperature is where the temperature derivative of the conductivity deviates from its constant value at high temperature ( data adopted from ref . [ ] ) . , scaledwidth=33.0% ] a different way to illustrate the regime change on the resistivity data is to define an exponent of @xmath16 for each temperature formulated as @xmath59 the exponent @xmath60 is obtained by taking the logarithmic derivative of the @xmath16 , @xmath61 where @xmath38 denotes the residual resistivity . the results for @xmath60 over a wide range in doping are shown in fig . [ fig : exponentphasediagram ] . defined in eqs . and for our model of lsco in color - code . the temperature scales @xmath47 and @xmath48 , cf . [ subsubsec : scaling regimes ] , are shown as black squares . the temperature scale @xmath53 corresponds to the energy difference between the chemical potential and the energy level of the van hove singularity , @xmath62 ( we consider low - energy states up to an energy of @xmath52 around the fermi energy ) . note that the unconventional scaling @xmath63 goes down to the lowest temperatures at optimal doping . the lifshitz transition is indicated as the horizontal dashed line . ] even though such a local scaling law has large uncertainties , it is intriguing to find a very good qualitative match between the function of @xmath27 and the generic scaling behavior of the cuprates , cf . and ref . in the overdoped low - temperature region of the phase diagram , the local exponent rises to the value of @xmath64 corresponding to a standard fermi liquid . around optimal doping , @xmath27 is always well below @xmath65 down to the lowest temperature of our study . in the high - temperature regime , the onset of unconventional scaling ( @xmath66 ) is shifted to higher temperature with increasing doping . using the non - equilibrium distribution function , we can analyze the angular dependence of this unconventional scaling and compare to the difference in the temperature dependence between isotropic and anisotropic scattering rates in the frg study [ ] . the quasiparticle scattering rates are strongly anisotropic and especially the quasiparticles close to the saddle points ( anti - nodal - direction ) in the brillouin zone have very short transport life times . the scattering rates that cause these short life times are strongly enhanced as the doping level decreases toward optimal doping . to examine the role of strong scattering of the anti - nodal quasiparticles for the unconventional temperature dependence of the overall resistivity , we interpret the angular patches of our brillouin zone discretization ( cf . [ fig : discretization ] in the appendix ) as individual resistors and parametrize the low - energy quasiparticle states as a parallel resistor network . each resistor is characterized by the conductivity @xmath67 and resistivity @xmath68 where the index @xmath69 labels the the angular patch . the total conductivity and resistivity are determined by @xmath70 in fig . [ fig : angularresolutioncoefficients ] we depict the the angle - resolved scaling by plotting the linear and quadratic contributions to the patch resistivity as functions of the angular patch index @xmath69 and the hole concentration . the temperature dependence of each resistor was analyzed by a linear plus quadratic fit , @xmath71 . in order to compare linear and quadratic term , we give them a weight factor that represents their contribution to the total resistivity and normalize their sum , the total resistivity @xmath23 , to unity , @xmath72 thus @xmath73 represents the normalized contribution of the angular patch @xmath69 to the term in the total resistivity that scales with temperature exponent @xmath74 . ) . on top , we have plotted the weighted linear term @xmath75 , and the lower plot shows the weighted quadratic term @xmath76 , defined in eq . . these @xmath73 s are weighted according to the contribution of the angular patch @xmath69 to the total charge transport . the patches range in angular direction from @xmath77 ( patch 1 ) to @xmath78 ( patch 10 ) . the @xmath73 s are evaluated at the temperature @xmath37 , the average temperature of the fit region . ] from fig . [ fig : angularresolutioncoefficients ] one can clearly see that in the strongly overdoped regime only the quadratic term is present and is essentially independent of the angle . on the other hand , below @xmath79 , the linear term grows showing strong anisotropy . the linear term appears first close to the nodal direction ( patches 4 to 7 ) , and constantly spreads out . below @xmath80 the linear term starts to dominate the transport . the quadratic term , however , remains finite in the nodal direction down to the lowest doping levels . it is interesting to see that around optimal doping , where the fermi surface is very close to the van hove singularity , the charge transport from the angular patches 1 and 10 , that contain the saddle points , essentially vanishes . this is consistent with the strong growth of the scattering rates from the frg analysis . these angular patches are short - circuited as a consequence of strong scattering , cf . our discussion here is fully compatible with the behavior of the angle - dependent scattering rates found by the frg study by ossadnik and collaborators [ ] . we can also analyze the current distribution as a function of the temperature . in fig . [ fig : angular distribution of the current ] we display the angular distribution of the charge current for two different doping levels . at the lowest temperatures the distribution function changes its angular dependence strongly and the weight of the current is shifted from small angles to larger angles , while above a certain temperature the angular distribution remains invariant . these regimes can be associated with two different kinds of scattering mechanisms . while at low temperature impurity scattering dominates , for sufficiently high temperature the transport is mainly determined by particle - particle scattering . since only umklapp scattering contributes noticeably to the resistivity , the two - particle scattering contribution is anisotropic , while the @xmath81-potential impurities give isotropic scattering . this difference accounts for a qualitative change in the angular dependence of the distribution function . we will return to the interplay of the two scattering mechanisms in the context of a discussion of matthiessen s rule , sec . [ sec : momentum relaxation mechanism in the cuprates and matthiessens rule ] . we turn now to the role of the renormalization of the scattering vertex in the unconventional nature of the transport properties . as explained in app . [ sec : choice of parameters ] we used a value of @xmath82 for lsco . when we extend our calculation to a series of lower initial values of @xmath83 the temperature dependence of the resistivity and the coefficients of the corresponding linear plus quadratic fit are shown fig . [ fig : rhovstemperaturelsco ] and in fig . [ fig : coefficientslsco ] . in addition we show the resistivity for the unrenormalized interaction @xmath84 which corresponds to an frg treatment with infinite cutoff . in this case the scattering rates are determined only by the scattering phase space which is specified by the fermi function and the geometry of the fermi surface . comparing the resistivity curves for @xmath17 and @xmath85 , we observe very similar overall trends as function of doping . the high - temperature resistivity grows upon reducing the hole - concentration and the @xmath4-linear part increases at the same time . it is obvious that the scattering phase space plays an important role in the change of the behavior of resistivity for different hole - concentrations . overall the behavior is , however , less pronounced for the unrenormalized than renormalized vertex as is clear in fig . [ fig : coefficientslsco ] . the renormalization of the vertex causes a sizable increase of the resistivity as one can see in the comparison of the curves for @xmath86 and @xmath85 , and it also strongly enhances the trend toward linear resistivity . this feature is also easily identified from the @xmath12-dependence of the quadratic term in fig . [ fig : coefficientslsco ] . for stronger initial interaction , and therefore stronger renormalization effects , the quadratic term is substantially reduced , even as the total resistivity , the sum of linear and quadratic term is increased . thus , it is evident that correlation effects included in the vertex renormalization by the frg scheme , make a substantial contribution of the unconventional transport properties , in particular , in the regime close to optimal doping . the vertex renormalization leads to a non - trivial momentum space structure of the effective scattering vertex which can be translated to a longer ranged real space interaction compared to the local hubbard-@xmath12 . as a second effect of the frg analysis , the momentum space structure but also the absolute values of the scattering vertex become temperature dependent . in order to discriminate between these two effects we modified our calculations in a way as to include only the momentum space renormalization of the vertex , but neglect the temperature dependence this was implemented by stopping the rg flow with a large cutoff @xmath87 . in this case _ large _ refers to the cutoff exceeding the largest temperature of our study . while this completely suppresses the temperature dependence of the scattering vertex , it still introduces momentum space anisotropy . the result resembles the behavior of the simple unrenormalized onsite repulsion @xmath88 of fig . [ fig : rhovstemperaturelsco ] . the introduction of momentum anisotropy through the rg flow has almost no qualitative effect on the temperature dependence of the resistivity . from this comparison we can conclude that the pronounced linear resistivity in the limit of @xmath57 is a consequence of the strongly temperature dependent quasiparticle scattering rates . this supports the idea of ref . [ ] where the unconventional temperature dependence of the scattering rates through the rg flow were described . at high temperature , however , the growth of the linear - temperature term when the doping is decreased toward optimal doping , is mainly due to the fermi surface geometry and the temperature dependence of the scattering volume , while the momentum anisotropy of the scattering interaction is responsible for the enhanced value of the resistivity . in the cuprates the resistivity is caused by electron - electron interactions which raises the question , how the total momentum of the entire distribution function is relaxed . in the following we will discuss which contributions are combined in the resistivity calculations of the previous section . in impurity scattering processes , the momentum difference between initial and final state is transferred to the lattice and determines the residual resistivity as the zero - temperature limit . a deeper analysis shows that the resistivity from impurity scattering can have a weak temperature dependence due to thermal broadening of the distribution function and a possible bias toward higher or lower quasiparticle velocity . in our calculation the impurities are modeled as @xmath81-potentials . a second and more complex contribution to the electrical resistivity originates from electron - electron scattering with finite momentum transfer to the lattice , so - called umklapp scattering , where a momentum corresponding to a reciprocal lattice vector is transferred to the lattice . note that the fermi surface has to be sufficiently large in order to allow for umklapp assisted scattering between different momenta on the fermi surface . for a square lattice we can distinguish between two kinds of umklapp scatterings : with momentum transfers of the type @xmath89 and @xmath90 to the lattice . the first kind of umklapp scattering has a large phase space for a sufficiently large fermi surface with regions where the @xmath91-component of the fermi vector satisfies the condition @xmath92 and other regions with @xmath93 . the second type of umklapp process is only possible , if the fermi surface intersects with the so - called umklapp surface , which is the rotated square ( diamond ) that connects the four saddle points at @xmath94 and @xmath95 . this umklapp scattering has relatively small phase space due to the stringent momentum constraints . within our model for lsco the @xmath89 umklapp process largely dominates over the @xmath90 umklapp process for all doping levels as far as the charge transport resistance is concerned . on the strong overdoped side , @xmath90 umklapp scattering is even completely absent in the low - temperature limit . due to the geometrical constraints for umklapp scattering the momentum space structure of the scattering vertex can play an important role . the temperature dependence of the resistivity is controlled by umklapp scattering . if the scattering rates become especially high on certain parts of the fermi surface , their contributions to transport will be suppressed yielding the possibility for unconventional temperature scaling . often different contributions to the resistivity are assumed to be decoupled and one adds them as serial resistors . this decoupling leads to matthiessen s rule , which is only justified , if each individual contribution can be well described by a single relaxation time . in this case the collision integral in the boltzmann equation directly splits into the scattering channels . in the general case , when such a single relaxation time approximation is insufficient , the collision integral does not decouple . in fig . [ fig : angular distribution of the current ] we have seen that in the low - temperature regime the angular distribution of the current changes its form . this was attributed to a crossover from an impurity scattering dominated regime to a two - particle scattering regime . these different scattering mechanisms lead to different current distributions due to their difference in the scattering geometry . while impurity scattering is isotropic , the umklapp processes responsible for the momentum relaxation in the two - particle scattering channel are highly anisotropic leading to a non - trivial interplay of impurity scattering with two - particle interactions and a potential breakdown of matthiessen s rule . to examine the deviations from matthiessen s rule we compare our calculation with the sum of the individual contributions to the resistivity . first we remove the impurities to obtain a bare electron - electron resistivity @xmath96 , second we estimate @xmath97 by considering free ( non - interacting ) electrons scattered by impurities . in a first step we compare the linear and the quadratic coefficients of a @xmath98-fit for the clean limit , @xmath99 , to the coefficients in the original calculation including impurities of fig . [ fig : coefficientslsco ] . the quadratic coefficient grows weakly down to @xmath24 and remains constant at @xmath100 , while for the full calculation the quadratic coefficient decreased below @xmath79 and even went to zero for sufficiently strong coupling . furthermore , the linear term is somewhat smaller without impurities . this difference in the behavior of the coefficients is a violation of matthiessen s rule . in order to make this more evident , we plot the deviation from matthiessen s rule , @xmath101 , quantitatively defined as @xmath102/\rho_{\text{full } } , \label{eq : deviation from matthiessens rule}\end{aligned}\ ] ] as a function of the temperature for different doping levels . the result is shown in a color plot in fig . [ fig : matthiessenruledeviation ] . we immediately observe that the deviation from matthiessen s rule is strongest at the optimally doped regime at low temperature . the difference in the scattering geometry for impurity scattering and two - particle scattering is most pronounced in this part of the diagram . here the deviation @xmath101 reaches values as large as 16% . it should be noted that the one - loop frg approximation is less reliable at low temperature and low doping , as pointed out earlier . hence the magnitude of @xmath101 in our calculation carries considerable uncertainties in this part of the diagram . nevertheless , the violation of matthiessen s rule is an generic property for systems featuring a crossover from isotropic impurity scattering to strongly anisotropic inter - particle scattering dominated transport . [ t ] and normalize it to @xmath103 . it is clearly visible that the deviation @xmath101 ( cf . is most pronounced in the critical region of the phase diagram . ] the seebeck effect is a further example which allows us to test fermi liquid properties in transport . this effect as the electric response to a temperature gradient is based on thermo - diffusion of charge carriers and can be analyzed with boltzmann transport theory too . lalibert et al . measured the seebeck coefficient @xmath104 of @xmath105 [ eu - lsco ] , cf . ref . while their study aims at probing the fermi surface reconstruction scenarios in the underdoped regime of the phase diagram , they also present data up to a hole - doping level of @xmath106 . data of the seebeck coefficient are displayed in fig . [ fig : experimental seebeck eu - lsco and numerical seebeck ] ( left panel ) for the doping regime @xmath107 . europium doping of lsco is believed to have a stronger influence on the properties of underdoped samples , but little effect is expected in the overdoped regime . for underdoped samples @xmath108 shows a sign change , starting negative at low temperature it turns positive at higher temperature . on the other hand , for overdoped samples @xmath108 remains positive for all temperatures measured and shows a rather strong increase toward low temperature near optimal doping . the inset of fig . [ fig : experimental seebeck eu - lsco and numerical seebeck ] ( left panel ) demonstrates the @xmath109 dependence of @xmath110 at low temperature , which has been interpreted as a signature of quantum criticality . we have numerically calculated the seebeck coefficient for lsco within our semiclassical approach , as explained in app . [ subsec : calculation of the seebeck coefficient ] . the results are shown in figs . [ fig : experimental seebeck eu - lsco and numerical seebeck ] ( right panel ) and [ fig : seebeck phase diagram ] . a good overview on the behavior of the seebeck coefficient as a function of doping and temperature is obtained from fig . [ fig : seebeck phase diagram ] ( left panel ) . at the doping @xmath111 we observe a sign change as we cross the lifshitz transition at low temperature . interestingly , for @xmath112 increasing the temperature leads to a sign change from negative at low to positive at higher temperature . turning to the other limit of strong overdoping the seebeck coefficient turns negative again as expected for electrons is a band with small filling . we also observe a doping dependent temperature scale where @xmath104 displays a rapid increase . interestingly , this temperature compares well with @xmath47 . in order to compare our numerical with the experimental results , we have plotted @xmath110 in fig . [ fig : experimental seebeck eu - lsco and numerical seebeck ] for some selected values of hole doping . the inset shows the low - temperature regime for the doping level @xmath113 on a logarithmic temperature scale . the general structure of the calculated seebeck coefficient is in good qualitative agreement with the experimental data measured for eu - lsco fig . [ fig : experimental seebeck eu - lsco and numerical seebeck ] . in both cases , there is a critical doping level below which the seebeck coefficient changes sign with increasing temperature , starting from negative in the @xmath57 limit . above this critical doping level , @xmath108 becomes very large for low temperature and strongly decreases with temperature . the low - temperature rise of @xmath108 becomes less pronounced with increasing hole doping . in the low - temperature regime , @xmath108 acquires a @xmath109 dependence which is clearly visible in both the experimental data and the numerical data below @xmath114 . a small discrepancy between our model and the experiment is the value of the critical doping level . we have designed the band structure that we use to describe lsco such that the lifschitz transition is located at @xmath115 , cf . [ sec : model for the quasiparticle dispersion of lsco ] . the van hove singularity at the lifschitz transition is the origin for the critical behavior in our model and thus constrains the critical doping level to @xmath116 . in the experimental study , however , a critical doping level of @xmath117 is reported . in the last section , we have discussed the low - temperature @xmath109 dependence of @xmath108 and the sign changes with doping or temperature . these properties of the seebeck coefficient are often taken as indication for the presence of a quantum critical point [ ] . these features are found in both the experimental data and the numerical simulation . we now want to locate the areas of critical behavior in the phase diagram . for conventional metals , the seebeck coefficient is a linear function of the temperature manifesting fermi liquid behavior . deviations from this linear dependence may be used as a signature for non - fermi liquid physics . thus , we analyze the the temperature derivative of @xmath110 which should vanish for fermi liquids . fig . [ fig : seebeck phase diagram ] ( right panel ) displays this function . this value is largest at low temperature around optimal doping and a sign change appears at the lifshitz transition . this is the region where the strongest differences with the fermi liquid picture emerge . from this finding we may identify the vicinity of the van hove singularity and the presence of a lifshitz transition as the origin of non - fermi liquid behavior and the signatures of quantum criticality . interestingly , fig . [ fig : seebeck phase diagram ] ( right panel ) allows us also to see the temperature of the sharp increase of @xmath108 , cf . [ fig : seebeck phase diagram ] ( left panel ) , appearing as a maximum of @xmath118 . this characteristic temperature @xmath119 separates the high - doping low - temperature regime coming closest to fermi liquid behavior ( @xmath120-dependence of resistivity ) from the lower - doping high - temperature regime which is characterized by the non - fermi liquid @xmath4-linear resistivity , as discussed in sec . [ subsubsec : scaling regimes ] . we investigate the temperature and doping dependence of charge transport coefficients using a combination of an experimentally determined band structure and a renormalized scattering matrix calculated in a one - loop renormalization group approximation . the band dispersion matches the quasiparticle spectrum in arpes for lsco [ ] , and as such includes the renormalization of the dispersion with hole density due to interactions . the electron - electron scattering vertex develops momentum - space and temperature dependence as the hole density is reduced . we have chosen a moderate to weak value for the bare onsite coulomb repulsion consistent with the one - loop approximation . we also refrain from including the pseudogap region which would require a solution of the strong coupling regime that results from divergences in the frg flow . the boltzmann transport equation is solved numerically with collision integrals for electron - electron and impurity scattering determined by fermi s golden rule . this semiclassical approach determines the quasiparticle distribution functions taking into account the full angular and energy dependence of the collision integral . we find good qualitative agreement with the main trends in the experiments on the in - plane resistivity of lsco [ ] . this includes the increase of the temperature dependent term in the resistivity upon decreasing the hole - concentration and the trend toward a linear temperature dependence of the resistivity @xmath121 around optimal doping . a parameterization of our results using a polynomial fit in @xmath4 up to second order , inspired by the experimental analyses [ ] , gives consistency between the trends in experiment and our model . while the @xmath4-linear behavior of @xmath121 is often attributed to the existence of a quantum critical point around optimal doping [ ] , our discussion shows that this feature may also be understood as a consequence of the unusual temperature dependence of the renormalized scattering vertices and the increased scattering phase space around this doping regime , where a lifshitz transition of the band structure leads the chemical potential very close to a van hove singularity . the magnitude of the calculated resistivity is substantially lower than in the experiment , consistent with the moderate to weak interaction values and one - loop rg approximation motivated by the experimental data analysis [ ] we also identified crossover regimes in @xmath122 and find rather sharp changes in the temperature dependence . there are two characteristic temperatures @xmath123 and @xmath48 where the latter denotes the lower bound of the @xmath124-linear regime of @xmath121 and @xmath125 the upper bound for a purely @xmath126-dependence ( fermi liquid regime ) . the intermediate temperature range may be considered as combination of both . the resulting phase diagram gives a very good account of the trends of @xmath121 observed in the experiment . in particular , we can show that there is a clear correlation of the unconventional temperature scaling with the van hove singularity , quantified through the effective distance of the chemical potential from the van hove level , @xmath127 , cf . [ fig : exponentphasediagram ] . by comparing the temperature dependence of @xmath16 for renormalized and bare scattering rates , we could demonstrate the importance of the scattering vertex renormalization . while in the strongly overdoped regime , the difference in the numerical data determined from renormalized and unrenormalized scattering rates is not very pronounced , around optimal doping the phase space driven growth of a linear term is strongly enhanced due to renormalization of the quasiparticle interactions . it also turns out that the impurity scattering is important in this respect as well , contributing to the unconventional temperature dependence . this is surprising as impurities contribute usually only a weakly temperature dependent offset to resistivity . we found deviations from matthiessen s rule in large parts of the considered phase diagram . this deviation is especially pronounced as we approach optimal doping at intermediate to low temperature due to the different momentum dependence of impurity and electron - electron scattering . matthiessen s rule , however , is recovered in the more conventional , strongly overdoped regime at low temperature . we have also studied the seebeck coefficient within our model . we find qualitative agreement between our numerical data and the main trends in the measurements performed on a slightly different compound , eu - lsco . we believe that the eu - doping does not strongly influence the seebeck coefficient in the optimal to overdoped doping regime which we are interested in . the calculated seebeck coefficient , also displays a critical region in the phase diagram with pronounced deviations from fermi liquid behavior . we have related the critical behavior to proximity of the van hove singularity to the fermi energy which is linked to the lifschitz transition . we are grateful to carsten honerkamp , nigel hussey , lszl forr , and stevan arsenijevi for helpful discussions . this study was supported by the swiss nationalfonds , the nccr manep , the hittec project of the competence center energy & mobility and the synergia teo . we use a band structure parametrized within a tight - binding model on a square lattice , including nearest , next - nearest , and next - to - next - nearest neighbor hopping @xmath128 the hopping parameters @xmath129 , @xmath130 , @xmath131 , and the chemical potential @xmath132 define the geometry of the fermi surface and the filling of the band . based on their arpes study , yoshida et al . [ ] have fitted the low - energy quasiparticle spectrum of lsco and the fermi surface in the doping range from @xmath133 to @xmath134 using the model . fixing he magnitude of @xmath129 to @xmath135 and the ratio @xmath136 for all dopings they used @xmath132 and @xmath130 as fitting parameters . their results are shown in fig . [ fig : discretization ] . the relation between @xmath132 and @xmath130 as a function of the doping is linear to a very good approximation , @xmath137 yoshida et al . identified a lifshitz transition at @xmath138 . our hopping parameter renormalization scheme , eq . , locates the lifshitz transition at @xmath139 . changing @xmath140 to @xmath141 shifts the lifshitz transition within our model to the experimental doping level . note , that a second arpes study by ino et al . [ ] also confirms the presence of a lifschitz transition at a doping level of about @xmath142 . the calculation of the renormalized vertex in ref . [ ] is based on the frg equation for the one - particle irreducible generating functional derived in refs . [ ] which gives hierarchically coupled flow equations for the one - particle irreducible vertices . the rg - flow follows from a wilsonian flow scheme with a sharp momentum cutoff . the rg equations are solved using the standard truncation of all vertices with more than four legs , so that the self energy and the four - point vertex are the only remaining quantities in the calculation . for the vertex , the momentum dependence is discretized and constrained to momenta lying on the fermi surface . the frequency dependence of the vertex is suppressed . for the rg - flow , the flow parameter @xmath87 is taken from @xmath143 to zero . in general , the truncated flow tends to diverge for low temperature at finite energy scales @xmath87 , indicating the appearance of an instability . the energy scale at which the divergence occurs is related to the critical temperature @xmath3 of the corresponding instability , in our case superconductivity . note , however , that due to the truncation of the rg equations the @xmath3 is largely overestimated . for a study of normal state transport , the divergence of the vertex should be avoided . while in the experiment superconductivity is suppressed by a magnetic field which is difficult to incorporate in the rg analysis , we keep thge cutoff energy scale @xmath144 finite and so suppress the leading @xmath145-wave pairing instability in the rg flow . the choice for @xmath146 is discussed in app . [ sec : choice of parameters ] . the quasiparticle scattering vertex @xmath147 is peaked for momenta @xmath148 lying close to the saddle - points @xmath149 and @xmath150 and has especially strong contributions for momentum transfer of @xmath151 , cf . [ fig : vertex ] . further discussions of the rg flow and interpretations of vertex diagrams as in fig . [ fig : vertex ] are given in refs . the frg method is a weak coupling analysis . on the underdoped side , below a temperature scale @xmath152 a pseudogap opens as an indication that the system is driven into a strong coupling phase . the strong - correlation physics are not included in our analysis . hence , we restrict our study to dopings from optimal to strongly overdoped . the onsite repulsion @xmath12 for the cuprates is a large energy scale of the order of the band width as at half - filling a mott insulating phase is realized . using such large values of @xmath12 , however , leads to a divergence of the rg flow within the normal state temperature range , since the instability temperatures are overestimated . this urges us to start with moderate value of the onsite repulsion of @xmath153 which then is renormalized to much larger values for the low - energy quasiparticles after integrating out the high - energy states . moreover , we suppress the @xmath145-wave pairing instability by keeping the cutoff @xmath144 finite in the rg flow , as mentioned above . for this purpose we choose @xmath154 corresponding to the lowest temperature in our study . the boltzmann equation relates the substantial derivative of the quasiparticle distribution function to the collision integral , taking all scattering events into account which yield a change of the distribution function , @xmath155\cdot{\boldsymbol{\nabla}}_{p}f({\boldsymbol{r}},{\boldsymbol{p}},t)\phantom{\int}\nonumber\\ & = \left.\frac{\partial f({\boldsymbol{r}},{\boldsymbol{p}},t)}{\partial t}\right|_{\text{coll}}. \label{eq : the boltzmann equation}\end{aligned}\ ] ] we consider two contributions in the collision integral , two - particle collisions labeled by `` ee '' and impurity scattering labeled by `` imp '' : @xmath156 the first contribution is expressed as , @xmath157with@xmath158 where the transition rates @xmath159 are generated through fermi s golden rule from the matrix elements of the quasiparticle interaction vertex @xmath160 , cf . [ sec : strongly renormalized quasiparticle interactions ] . the second contribution to the collision integral is simpler due to its single particle form , @xmath161with@xmath162 we ignore electron - phonon contributions here because , in the cuprates , the momentum relaxation of the quasiparticles is dominated by electron - electron interactions as was concluded e.g. from the temperature dependence of the hall coefficient [ ] and from the observation of a quadratic resistivity for overdoped lsco [ ] . it is important to notice that two - particle scattering can only yield a finite resistivity , if momentum is transferred to the lattice , which only can be realized through umklapp scattering ( see sec . [ sec : momentum relaxation mechanism in the cuprates and matthiessens rule ] ) . umklapp scattering has special geometrical constraints and , therefore , introduces strong anisotropy in the brillouin zone and requires a detailed analysis of the collision integral beyond a single - relaxation - time approach . we linearize the boltzmann equation in terms of the deviation from equilibrium @xmath163 . the collision integral is interpreted as a linear integral operator acting on @xmath163 . we evaluate the integral kernel on discrete patches of a finite mesh ( see fig.[fig : discretization ] ( a ) ) . note in this context that that the energy conservation appearing as delta function in the scattering rates has to be treated carefully . in discretized momentum space , the linearized boltzmann equation is reduced to a set of linear equations which are solved numerically . in order to match the fermi surface , we use polar angle and the energy to represent each momentum vector . for our curved - patch discretization we can adjust the grid spacing in radial ( energy ) direction to the temperature scale and always cover the relevant scattering phase space . a sketch of this discretization is given in fig . [ fig : discretization ] ( a ) . for the calculations that we present here we have chosen a discretization of 24 patches in radial direction and 120 patches in angular direction . the latter are not distributed equally but are much finer in anti - nodal than in nodal direction to take care of the flat dispersion in the proximity of the saddle points . for a study of the seebeck effect , the boltzmann equation is extended by a new drift term that accounts for the response of the distribution function to the external thermal gradient , @xmath164 with the electro - chemical potential defined as @xmath165 . the collision integral remains unchanged . the distribution function that solves this equation carries a heat current and an electrical current . these two currents are defend as the linear response to the external fields as @xmath166 considering an open circuit geometry with @xmath167 , the typical setup for thermopower measurement , the seebeck coefficient is defined as the proportionality factor between the applied thermal gradient and the induced electrochemical potential , @xmath168 . for simplicity we assume all forces aligned along one direction and , thus , we can use scalar quantities . it is important to note that we do not make use of analytic simplifications as mott s formula or sommerfeld expansion . according to the arpes study [ ] used above to find the effective tight - binding parameters of lsco , a value of @xmath135 for the nearest neighbor hopping parameter @xmath129 fits the arpes data best . this defines an energy scale which translates the energy scales of temperatures in units of @xmath129 to @xmath169 . if we want to study the temperature range up to room temperature of @xmath36 the corresponding energy scale is @xmath170 . in order to cover at least one order of magnitude in temperatures we choose @xmath171 and @xmath172 . our model is based on a two - dimensional interacting electron gas on a lattice . the experimental setup is based on the measurement of the resistivity of a three - dimensional but layered sample . for a quantitative comparison of the experimental and numerical data , a conversion of the two - dimensional resistivity to a three - dimensional resistivity is required . for a layered system this is very easily achieved by multiplication of the two - dimensional resistivity with the interlayer distance . the c - axis lattice constant of lsco is given by @xmath173 [ ] and the interlayer distance by half the lattice constant , because the unit cell of the 2 - 1 - 4-compounds contains 2 layers . f. lalibert , j. chang , n. doiron - leyraud , e. hassinger , r. daou , m. rondeau , b.j . ramshaw , r. liang , d.a . bonn , w.n . hardy , s. pyon , t. takayama , h. takagi , i. sheikin , l. malone , c. proust , k. behnia , and l. taillefer : nature comm . * 2 * ( 2011 ) 432 t. yoshida , x.j . zhou , k. tanaka , w.l . yang , z. hussain , z .- x . shen , a. fujimori , s. sahrakorpi , m. lindroos , r.s . markiewicz , a. bansil , seiki komiya , yoichi ando , h. eisaki , t. kakeshita , and s. uchida : phys . b * 74 * ( 2006 ) 224510

there is a general lack of information about the nature of uv sources fainter than the completeness limit of the td-1 photometric catalog ( thompson 1978 ) . this is , primarily , the result of a lack of surveys of the uv sky . only partial surveys , in limited regions of the sky , have been conducted since the td-1 mission . one of the more extended , in terms of general sky coverage , was by the faust experiment . faust is a @xmath1 instantaneous field - of - view ( fov ) telescope , covering the spectral range between 1400 and 1800 with an angular resolution of 3.5 arcminutes . it has a 20 cm entrance aperture and it utilizes a microchannel plate detector with wedge and strip anode , which records the position of each detected photon . faust operated on board the space shuttle ( sts ) atlantis in march 1992 as part of the atlas1 mission . a description of faust and its operation aboard atlas1 was given by lampton ( 1993 ) . during this flight 19 pointed exposures were obtained , of which three were short exposures for pointing checks . the other fixed - pointing images were exposed for 12 to 18 minutes . in addition , two @xmath030@xmath2 long scans of the sky were obtained by rolling the sts during the exposure . details of the image reconstruction from the time - tagged photon stream and subsequent reductions are given in bowyer ( 1993 ) . a catalog of 4660 faust sources , originating from the fixed - pointings as well as from the sky scans , was produced by bowyer ( 1995 ; fsc ) . this used a uniform thresholding algorithm for source detection , and the identifications were obtained from correlations against catalogs ( mainly from the simbad data base ) . the faust images are being studied systematically at tel aviv in order to identify sources and perform , as much as possible , ground - based follow - up studies . in this context , we already presented results from five images at high galactic latitude sky regions visible from the northern hemisphere : the north galactic pole ( brosch 1995 ) , the virgo cluster region ( brosch 1997 ) , and in the direction of the coma cluster ( brosch 1998 ) . this paper presents our analysis of four faust fields in the southern sky . it uses , for the first time , data from an objective - prism survey ( the hamburg - eso survey = hes , wisotzki 1996 ; reimers 1997 ) to provide additional ground - based spectroscopic information about the possible optical counterparts . the uv sources in this part of the milky way ( mw ) may help the understanding of the morphology of the disk and halo of the mw . this is because stellar uv sources represent special stellar populations ; the majority are rather massive main - sequence objects with spectral types a & f , as our previous studies have shown . their study allows us to test the mw structure in four spatial windows . in this region there are indications of bends , corrugations , valleys , and other morphological peculiarities of the galactic disk ( alfaro & efremov 1996 ) . the paper is arranged as follows . we first describe the faust observations and the source detection on the uv images . we then describe the process of source identification based on correlations with published data , mainly from electronic data banks . we describe our use of the he s objective - prism plates to confirm or modify our proposed identification . the discussion concentrates on ( a ) a comparison of the fields which have he s data with those fields lacking this information , and ( b ) the general properties of the uv sources we detected , in the context of the fourth galactic quadrant in which they are located . one field is located at the border between the third and the fourth quadrant ; it is analyzed here as though it is one of the 4th quadrant fields . the regions observed by faust , and in which the source identification process has been performed , are listed in table 1 . the faust images are shown in figures 1 to 4 , for the dorado , centaurus , m83 , and telescopium ( hereafter dor , cen , m83 , and tel ) regions . all four regions are located in , or very close to , the fourth galactic quadrant , the part of the mw galaxy which ( as seen from the solar neighbourhood ) lies between the galactic center ( gc ) and l@xmath3270@xmath2 . they are all located at intermediate galactic latitudes , two north of the mw plane and two south of it . the lines of sight sample , therefore , the disk and halo of the mw at similar `` elevation '' angles from the galactic equator and at approximately the same aspect angles with respect to the galactic bulge . therefore , apart from the general interest in source detection and identification , the present observations allow , in principle , an inter - comparison of regions which sample approximately the same mw environments . each faust image has location - specific exposure levels within the image . the reason is that the space shuttle attitude control system permits some platform motion , and the information was collected as a time - tagged photon stream . the images were reconstructed on the ground by tracking relatively bright stars . this implies that during a pointed exposure a larger area of the sky was imaged than the instantaneous fov of faust . table 1 lists , for each region , the actual sky area covered by each image and the lowest and highest exposure levels , generally collected near the edge and in the central part of the image , respectively . the sky area covered by each image was calculated from the total number of non - zero exposure pixels , and by assuming that each pixel is a square of 1.1 arcmin sides . this is an approximation , because electrostatic distortions of the image caused by the detector causes pixels to have slightly different areas . an impartial source detector algorithm , described in brosch ( 1995 ) , operated on the faust images and produced lists of uv source candidates with their faust uv fluxes and equatorial coordinates . because of the physical nature of the faust frames , the exposure times , and therefore the image depths , differ for different regions in each image . variations range from a few hundred seconds in the central regions to a few tens of seconds at the frame periphery . this results in regions with reasonably high signal , whereas other regions ( with low exposure ) have low signal levels . the detector used in faust has as only sources of noise the poisson statistics of photon arrival and detection , and the sky background . as a result , the detection method must be signal - to - noise limited rather than flux - limited . the source detector algorithm used here required an acceptance threshold ( at ) level of 5.5 times the standard deviation of the local background ( @xmath4 ) to decide that a source was probably real . this was determined by examining the behavior of the number of source candidates with threshold level ; the number of possible sources increased exponentially with at for thresholds lower than 5 , but only linearly for thresholds higher than this value . we infer that , if the at is too low , large numbers of `` false '' sources are introduced by the detection method . by adopting a high threshold level one is in danger of losing valuable information . we selected the specific 5.5@xmath5 level used here at the inflexion point in the plot of the number of detected sources ( n@xmath6 ) _ vs. _ at , where the linear regime and the exponential regime cross over . the total number of sources detected by this method in each field is also listed in table 1 . the astrometry of the sources positions is based on a best - fit to the @xmath040 brightest uv sources in each field . these were identified in predicted uv images , which were created with data from the sao catalog and the hipparcos input catalog ( hic ) , together with the algorithm for predicting the uv magnitude using the visual magnitude and spectral type of stars ( brosch 1991 , and further modifications ) . the fits were usually good to @xmath030 " and were checked using the derived coordinates of the remaining @xmath020 visually identified sources . the photon flux from each detected source was measured by integrating all count rates within a simulated round aperture centered on the location of the object , and subtracting the local background . a round aperture of diameter 12 pixels , equivalent to 13.2 arcmin , was used for the majority of sources except for the very strong sources where a diameter of 24 pixels ( 26.4 arcmin ) was used . this aperture is much larger than the resolution of faust and it allows the collections of virtually all the uv photons from a source . the photometric error is composed of the aperture photometry error , that originating from the background estimation , and a systematic error in flux resulting from the laboratory calibration of faust and the intrinsic errors of the iue photometric scale ( 15.8% of the observed flux ) . monochromatic faust magnitudes ( [ uv]=2.5 log(flux)21.175 ) are used here to represent the detected brightness , where the flux is in erg/@xmath7/sec / and was derived from the count rate of faust , as explained in brosch ( 1995 ) . the photometric error translates to a minimum error of 0.16 mag in the uv magnitude used here . the lists of uv sources , separate for each faust field , are presented in table 2 . each source is identified by a running number and two leading letters , the second that identifies the faust field in which this source was identified . for instance , source fd1 is the first detected uv source in the dorado ( dor ) field , and fm83 in the 83rd source in the m83 field . we give the j2000 coordinates of the sources in columns 2 and 3 , the faust magnitude and error in column 4 , and the corresponding uv source number from the faust source catalog ( fsc : bowyer 1995 ) , for the cases where the fsc lists a source in column 5 . it is important to emphasize that the source detection algorithm used here is different from the one used in creating the fsc ( bowyer 1995 ) . because of this , a comparison of the number of detected sources listed here and in the fsc is in order . we searched the fsc for all entries within 3 of each of the sources detected here . the corresponding fsc number is listed in table 2 for all entries where a match was found . in a few cases more than one match was found . this was usually due to the proximity of another source and the `` ownership '' was resolved by visually inspecting the uv image . for three sources , all in the m83 field , bowyer ( 1995 ) detected two components , while we detected only one . the vast majority of the fsc detections have matches in our lists . in the three other fields ( dor , cen , and tel ) most fsc sources undetected by us are all very faint and remained unidentified in the fsc . they could be artefacts , but there is one exception : fsc4623 = sao193381 in the tel field , apparently a bright uv source , whose presence we could not confirm with our detection technique . a comparison for the m83 field is problematic due to the overlap of one of the faust scans with the pointed m83 exposure . nevertheless , the number of fsc entries for the m83 region not detected here is small ; ten such entries were found within 3@xmath2.4 of the m83 image center . this number does not include the three cases where the fsc showed two adjacent entries for only one source detected here . in these cases we included both fsc numbers in the lists , but caution that this may be due to a single object having slightly different fsc astrometric solutions , from the scan and from the fixed pointing . since almost all of the ten other `` undetected here '' m83 fsc entries have either nearby detected sources or are very close the edge of the image , we conclude that they originate from the faust scan , not from the pointed observation . we estimate the number of fsc sources we did not detect and which originate from the m83 fixed exposure to be at most five . table 3 summarizes this comparison of results and emphasizes the significant addition of uv sources to those in the fsc by using the present detection algorithm . on average , fsc sources make up only 2/3 of the uv sources detected by us . therefore , with our detection scheme we increased by @xmath050% the number of faust uv sources in comparison with the fsc . figure 5 shows histograms of the uv magnitude distribution in each of the four fields . in general , all fields show a steady increase in the number of detected sources with increasing apparent magnitude , but there are some subtle differences . the cen field , which is at the lowest galactic latitude , has many more sources than the other fields . on the other hand , the source distribution with uv magnitude in the cen field starts to become incomplete already at m@xmath811 , approximately 1 - 2 magnitudes brighter than in the three other fields . one important feature is noticeable by even a cursory inspection of table 1 , which gives the total number of detections along with the location of each image and its depth . the m83 image has almost the same number of detections as has the tel field , yet the two images differ by a factor of @xmath9 in exposure depth , with the tel field being shallower . the discrepancy can not be due to a higher stellar concentration in tel , because there are no ( listed ) star clusters in this direction and the galactic latitudes of the two fields are almost identical . this will be discussed further , below . the primary identification used on - line data bases , such as the sao catalog ( sao 1966 ) , the hipparcos input catalogue ( turon 1993 ) , the tycho catalog ( esa 1997 ) , and the ned . the identification criterion was the angular distance : the nearer an object is to the calculated uv source position , the more likely it is to be the actual counterpart . additional constraints were imposed on a physical basis . the bluer an optical candidate , the more likely it is to be the real counterpart . the spectral type of the object , if it was listed in the catalog source , was an important input to the determination of the accepted counterpart . sources for which no physical properties could be found were correlated with guide star catalog ( lasker , jenkner , & russell 1988 ) objects , if such could be found in the neighbourhood of a source . the remaining identifications were extracted mainly from the first edition of the us naval observatory two - colour catalog ( usno - a1.0 , hereafter usno : monet 1996 ) . this catalog is based on scans of the palomar and eso / serc sky survey blue and red schmidt telescope plates . the absolute scale of the blue and red magnitudes is not accurate , but a `` local '' comparison of colours between a number of candidates for the same uv source is useful for identifying the bluest and brightest object , which is most likely to be the counterpart of the uv source . the second version of this catalog ( usno - a2.0 : monet 1998 ) has improved astrometric positions and photometric calibration and was used to check a few problematic sources . the photometric improvement is based on the b and v magnitudes of the tycho catalog for the brighter objects and on ccd observations of standard stars for some sky regions . however , the final reported photometric accuracy ( 0.15 mag internal error and up to 0.5 mag due to systematic effects ) is not much better that that of the a1.0 version of the catalog . the usno catalog , whose candidates are designated by a `` usno '' prefix in the lists , contains both stellar and non - stellar sources . on completing the identification process and analyzing the results , a few ( @xmath1010 for all four fields combined ) fairly strong sources , all very near the edge of the images , remained unidentified . as these are relatively strong uv sources they are very probably not artefacts , and a special effort was exerted to identify them . an additional search in the tycho catalog , using a search radius of 5 arcmin , produced good identifications for most sources with optical counterparts located within 3 - 4 arcmin of the uv source . analysis of other identified sources in the same vicinity showed that a systematic positional error appears sometimes at the edge of the images . this is presumably due to the electrostatic distortions being the greatest at the edges of the detector , and not having enough pre - identified stars in these locations as input for the astrometric solution used to compute the celestial coordinates . these additional identifications have been included in the lists shown in tables 2 and 4 . the entries of table 4 , which reports the identifications , repeat the source identifier used here in table 2 column 1 , give a leading catalog identifier for the proposed counterpart ( column 2 ) , its v - band magnitude ( or b - band , in some cases ) in column 3 , a note on its spectral type if it is a star or on its morphological classification if it is a galaxy ( column 4 ) , its listed b - v colour ( or b - r , in some cases ) in column 5 , and the candidate s celestial position in columns 6 and 7 . all positions reported here are for j2000 . the he s ( wisotzki 1996 ) is an objective - prism survey aimed at detecting bright qsos . it was performed with the eso schmidt telescope using iiia - j plates and a 4@xmath2 objective prism . the plates have a resolution of 15 at h@xmath11 ( seeing - dependent ) and cover the range 3200 to 5400 . the celestial area covered by the he s is mainly in the extragalactic sky , roughly above @xmath12b@xmath12=30@xmath2 . because of this sky coverage limitation , two of the faust regions analysed here could not be compared with the he s data sets . in particular , we limit the discussion of the objective - prism spectra to the fields of m83 and of dorado , as the other two regions have a galactic latitude too low to be included in the he s . the he s plates were scanned on the hamburg pds with a 30@xmath1330@xmath14m@xmath15 aperture , and spectra were automatically extracted using object detection on the digitized sky survey . each spectrally observed candidate has , therefore , an astrometric solution good to @xmath02 arcsec and a linearized spectrum ( using the plate sensitometric relation ) covering the range of the he s objective - prism plates , which is @xmath0300 pixels long . on average , @xmath075% of the spectra do not show overlap with other neighbouring spectra and can be used to verify or revise the classification and assignation of an optical candidate to a faust uv source . the search for optical counterparts on the he s data set was done by lw in early - november 1998 . a search in a region 3 arcmin wide around each faust source was performed and the candidates were extracted and examined by nb . in a number of instances , the candidate optical object was so bright as to cause partial or even full saturation of the objective - prism plate . for several cases , the spectral information allowing a rough classification could be recovered using additional scan lines from the two - dimensional scan data . we could thus use the balmer absorption lines and the 4000 ca ii break as classifier criteria in objects as bright as b@xmath169.5 . an optical candidate was accepted as the probable counterpart of a uv source if it : 1 . was the brightest source , 2 . was the nearest source , 3 . showed strong balmer absorption lines , 4 . showed a significantly strong blue - uv `` tail '' of the spectrum not all these conditions had to be fulfilled simultaneously for a source to be accepted as a counterpart . we are aware that these may seem to be subjective criteria , but adopt them here in absence of a systematic spectral classification of all he s spectra . an example of the output produced from the he s plates for this search of optical counterparts , one page of the many produced for the various candidates and inspected by us , is shown in figure 6 . three objects are displayed in this figure ; the bottom one is source fm 31 identified as a new sub - dwarf star . the various entries for each candidate include the number of the eso - schmidt plate on which the object appears , various classification parameters , the image of the field from the digitized sky survey with the candidate object centered , and two representations of the objective - prism spectrum , one at full resolution ( centered ) and another binned to low resolution ( to the right ) . the strong blue tail , indicative of a very hot object , is evident for candidate he1342 - 2728=fm 31 . the he s plates yield also blue ( b@xmath17 ) magnitudes , from the blue part of the spectrum and a calibration against standard stars ( vanelle 1996 ) . the calibration is good to @xmath00.2 mag provided that the spectrum is not overexposed ( b@xmath1811.5 ) . in principle , it is also possible to determine the u b and b v colour indices of the candidates from the spectra , using a calibration of spectral slope against colour indices ; however , here one has to guard even more carefully against chance partial spectrum overlap . we emphasize that the assignation of optical counterparts in the dor and m83 fields was done in a completely independent way from the assignments based on catalog correlations . in most cases , the he s spectra confirmed the original catalog selection of a counterpart . the counterparts selected from the he s spectra are listed in table 4 if they were `` better '' than those originally selected using catalog correlations , or if they offered the choice of a single counterpart , when the catalog searches yielded a number of optical sources in the same error ellipse . the he s sources can be recognized by the he prefix identifier and by the approximate spectral type , derived from the visual inspection of the spectrum tracing . note that in many cases the location of the uv source on the he s contains no object with a spectrum which could be associated with the faust source . in the dor field 37 uv sources had no such he s counterpart . in a number of cases , we could not solve the question of the proper counterpart , even using the he s data . in such cases we reverted to the usno catalog ( version a2.0 ) and selected the best counterpart to be the nearest and bluest listed object , and this is the entry adopted for table 4 . in this section we summarize results from the detection and identification process , including the confirmatory or enhancing results with the he s spectra using the data from table 4 . the summary , which splits the source identification into three broad classes ( extragalactic , stars , and unknown ) is given in table 5 . one item stands out immediately , that is the low number of `` unknown '' sources in the dorado and m83 fields . in these two faust fields the percentage of such sources is 19% and 5% , respectively , while it is @xmath030% in the other two fields . the only difference between the first pair of fields and the second is that for dor and m83 we used the additional information from the he s survey . this indicates that using the he s resource we could classify ( in broad spectral types ) an additional @xmath015% of the faust uv source , some of which belong to the interesting `` hot , evolved '' stellar variety . in this context , an important statistic is the number of such sources ( sub - dwarfs and white dwarfs ) counted in each field . whereas the images without he s data have extremely few such sources ( one in cen and none in tel ) , the other two fields contain significantly more sources ( nine in dor and five in m83 ) . this emphasizes the importance of additional spectral information in untangling the nature of the ( optically faint ) hot uv sources . a comparison of the detected and identified stars with the predicted number of stars , using the updated model of the uv galaxy of brosch ( 1991 ) . the original model uses the bahcall & soneira ( 1980 ) predictor routines , where the milky way is approximated as an exponential disk and a bulge , each with its own luminosity function . the extinction by dust is treated by assuming a slab - like distribution of dust along the galactic equator with an exponential scale height above the disk . the bahcall & soneira model yields predictions for optical bands , and it has also been extended to the near - ir . the uv application of this model uses an optical - to - uv transformation for stars based on their colours . this was derived from iue observations and was extended to late - type stars using kurucz ( 1991 ) model atmospheres . in addition , the uv predictor model includes the gould belt , and a thick disk of white dwarfs with the scale height determined by boyle ( 1989 ) . the comparison of the observations with the model predictions is shown in figures 7 and 8 . we show there the distribution of the differential star counts ( number of stars per uv magnitude ) vs. the model predictions in the left panels , where the actual star counts are represented by squares with error bars derived from poisson statistics . the right - hand plot for each region corresponds to the distribution of colours . to produce this , we deleted from the star list those objects classified as `` hot evolved stars '' , such as hot wds and sd s , based on the catalog information or on data extracted from the he s spectra , because they are not predicted well by our model . we also deleted all the extragalactic objects , which are not included in the predictor model . our model predicts the star numbers and magnitude distributions per square degree , and the predicted colour distribution depends on the faint magnitude cutoff . the fainter this cutoff , the redder is the peak and the median colour of the stellar distribution . allowing fainter stars in the calculation of the colour distribution brings in more red objects than blue ones , driving the total distribution to the red . the actual cutoff in a specific faust exposure is location - dependent . this is because the exposure depth is shallower the closer an object is to the edge of the faust image , the part of the area with the low exposure time . for this reason , we adopted the following strategy when scaling the predictions to the `` effective area '' surveyed : 1 . we selected a magnitude cutoff for the colour prediction at 70% of the peak in the magnitude distribution . that is , we set the cutoff to a fainter magnitude than the peak of the magnitude distribution of the specific field , at the ( interpolated ) magnitude level where the stellar magnitude distribution reached 70% of the peak value . we decided on a scaling constant from the `` per square degree '' prediction of the model and the real star counts in * both * the magnitude and colour distributions using the total number of stars detected to be brighter than or equal to m@xmath19=11 [ n(m@xmath20 ) ] and the similar number of stars , brighter than 11 mag , predicted by the model [ n(m@xmath21 ) ] . the ratio of these values was used to scale both the magnitude distribution prediction and the colour distribution prediction . the actual detected distributions , and those predicted by the model and scaled as explained above , are shown in figures 7 and 8 . in deriving the uv - v colour of an object we used the magitude of the counterpart as listed in table 4 . in many cases , this is the usno catalog `` blue '' magnitude , which may be substantially different from the actual v magnitude of the objects . however , we estimate that this is a random error , which will only increase the scatter of the results but will not introduce a systematic bias , as most of the stars that have spectral types tend to be a and f , which have small b v values . the results , shown in figures 7 and 8 for the four fields , indicate that the procedure is valid and that the predicted stellar populations , as well as the general colour distributions , correspond to those actually observed . we emphasize that , while the choice of the scaling factor seems arbitrary by using the total number of stars to 11 mag . , the use of the same constant for scaling the colour distribution is a constraint , and the subsequent reasonable fit represents an independent confirmation of this procedure . the possible sources of discrepancy between the model predictions and the actual measurements are presumably stars that are too blue in the uv than the model is able to predict . such stars are , for example , hot evolved objects that are not included in the model . the discrepancy is emphasized by the trend of the plots in the right panels of figs . 7 and 8 to under - predict the observed colour distribution at the blue end . this indicates that probably there are more hot evolved stars among those observed by faust than revealed by the spectroscopic classification . the assumption is supported by the size of the under - prediction : the discrepancy is higher for the two fields where we do not have he s spectroscopic data than in the two fields with such data . this finding indicates that collections of uv observations have the potential of enlarging the samples of such objects , with obvious influences on our understanding of the structure and evolution of the milky way . in this section we discuss those stellar sources classified as either hot evolved stars ( sd s , wds , etc . ) , or as red stars ( k0 and later ) , which represent the hot and the cold extremes of the stellar distribution . it is noticeable that of 24 such objects in the four faust fields treated here , only two are in a image for which we have no he s information . this demonstrates , again , the value of ancilliary spectral information from which we extract the special - interest objects . we identified 15 stars as hot evolved objects , mostly in the dor and m83 fields . these are relatively faint uv sources , with a median magnitude of 11.44 , and with very blue colours ( median uv - v , or uv - b=3.65 ) that imply rather faint optical magnitudes . the optical faintness explains the lack of spectral information ; only wide - field surveys using multi - band colours or an objective prism could have picked these objects out . most are sdo , sdb , or sda , but one ( fc 89 ) is classified sdf2 and is indeed very red , with uv - v=3.94 . this is a high proper motion star , as expected of a subdwarf . the source with the most negative colour index is fd 46 , with uv - v=5.03 . we searched the lists of objects for those with extremely blue uv - v colours ( @xmath224 ) and found very few of these . specifically , we detected three such stars in the dor field , three in cen , seven in m83 , and two in tel . six of these are classified in simbad or by us as sub - dwarfs , one is an emission - line galaxy , and for the rest we do not have information beyond the identification with a star - like object and a rough optical magnitude . it is possible that most are also hot sub - dwarfs or binary systems containing hot , evolved stars , and spectroscopic follow - ups are necessary to clarify this point . there are 10 objects classified k0 or later in the four faust fields . as expected of red stars , their uv - v colours are in general large and positive . the colour indices are , though , much bluer than could be expected from the effective temperature which corresponds to their spectral type . the variations of the colour index within the same allocated spectral type are considerable ; among four k2 stars the spread of uv - v is more than 5.5 mag ! normally , k stars are expected to have uv - v@xmath163.2 - 7.4 , but some stars in this class are more uv - bright than this , presumably because of some form of coronal activity ( see jordan 1987 ) , or because of the presence of a hot secondary . one object , fm 81 , is classified as a m0 star with an expected uv - v@xmath169.45 colour , yet it shows an almost nil colour index . checking with simbad , we find this star to be rw hya , a well - known symbiotic system consisting of an m2 giant and a @xmath080,000k wd ( kenyon & mikolajewska 1995 ) . this demonstrates the advantage of a wide wavelength base , from the uv to the optical , with which to identify potentially peculiar objects . another one , fc 168 , is an m3iii star with uv - v=+5.4 , while we expect for such an object uv - v=+9.67 . although there are no indications in simbad of any peculiarity concerning this star , it is likely that this is yet another binary system , with a hot component responsible for the uv excess . we remarked in earlier papers that metallic - line a stars ( am ) appear fainter in the uv than their regular counterparts . the present study is particularly useful to check this behavior because it yielded a relatively large stellar sample . table 4 lists six am stars ( earlier than a5 ) and one late - bm star . these are plotted as filled squares in the colour - colour diagram shown in figure 9 . the comparison is done with regular a stars from the same faust images . the a0 stars are plotted as circles and the late - a stars ( a7 , a8 , and a9 ) are plotted as filled triangles . we added the reddening vector assuming the standard savage & mathis ( 1979 ) relation as a vector for e(b - v)=0.2 to demonstrate that the discrepancy could not be due to reddening by a standard milky way extinction law . it is clear that the am stars are uv - deficient in comparison to most of the a - type regular stars , including the latest a stars ( a9 ) . this deficiency has been noted before ( by van dijk 1978 ) and is presumably due to line blanketing of their spectral energy distributions . note though that the study of van dijk , and others dealing with the issue of uv deficiency of a stars , concentrated mostly on ap objects rather than on am stars . in principle , this points a way to use the property of these a stars to be less uv - bright than expected from their rough optical spectral classification , to identify possible candidate am ( or ap ) stars from among those objects in our four fields for which we found spectral classifications . the requirement is for good uv and optical photometry of the candidate objects , which should have been previously classified as a stars . fifty - four objects in table 4 have been identified as extragalactic objects . in two instances two optical counterparts could correspond to specific uv sources ; the extragalactic ones are listed in table 6 . as seen in previous cases , faust detected more uv flux than iue in the few instances where such observations are available . in the case of ngc 5253 , the iue spectra yield monochromatic uv magnitudes of [ 1482]=11.03 and [ 1913]=11.39 , whereas faust measured 9.89@xmath230.19 mag . the iue measurements refer to `` standard '' bands in the sw spectral range , centered at the listed wavelengths . this discrepancy between faust and iue is undoubtedly the result of measuring a large object with a small aperture . for ngc 5236=fm 83 the faust magnitude is 7.92@xmath230.17 whereas iue measured [ 1482]=10.48 and [ 1913]=10.82 . ngc 5135 has [ faust]=13.84@xmath231.01 and [ 1482]=13.94 and [ 1913]=13.87 , presumably the result of most of the uv emission being produced in the `` composite '' seyfert 2+starburst compact nucleus ( phillips 1983 ) . the amount of uv flux external to the iue aperture may be quite significant in some cases . among the more interesting extragalactic objects detected by faust in these four fields we note m83 ( source fm 70 ) , a starburst galaxy in which no less than six supernovae were discovered in a period of 60 years . a similar case is tol 34 ( source fm 181 ) , which contains a starburst and a sy 2 nucleus . some interacting galaxies appear as uv sources : the compact group of ngc 6845 with four peculiar spirals is such an example . unfortunately , the angular resolution of faust is not sufficient to show which of the objects is the uv source . this group would make an interesting target for a future uv imaging observation . we detected nine uv - emitting galactic nuclei , of which six were identified from the he s spectroscopic data . one of the more interesting such sources is the optically - violent - variable bl lac object euve j2009 - 48.8 ( source ft 34 ) . ours is the first listed uv measurement of this object , which has been detected in @xmath11-rays and xrays , as well as in the euv range . note that the uv - v colour index we measure is based on the optical magnitude listed in the ned . because of the intrinsic variability of a bl lac object , the real colour index may be different from the `` instantaneous '' one . note also the detection of the cd galaxy eso 444-g046 , the largest in the cluster abell 3558 , as a faust uv source . this object has a neutral colour index , uv - v@xmath16 + 0.06 . one other object detected by faust could belong to the same cluster : this is source fm 185 , a star - like object listed only with its gsc number but which showed emission lines in he s spectra and appears almost star - like on the blue plates of the he s . the object was classified by us as an `` emission - line galaxy '' ( elg ) . the colour index exhibited by this galaxy is one magnitude bluer than that of the cd galaxy . the elg is apparently part of a chain of larger galaxies in the cluster , extending from north - west to south - east . lacking a redshift , the assumption that this galaxy is a member of the a3558 cluster remains to be proven . we remarked above on the puzzling discrepancy in the source count between the m83 and the tel areas . briefly , the discrepancy is that these two sky regions , at almost identical @xmath24b@xmath24 , have approximately the same number of uv sources but one ( m83 ) has three times more exposure than the other . the puzzle only deepens when we consider the normalized source count , that is , the projected source density . we first define a representative projected source density for a field @xmath25 as @xmath26 , where @xmath27 is the number of sources brighter than m@xmath19=11 detected in the specific field , and @xmath28 is the angular coverage obtained by faust and listed in table 1 . this magnitude cutoff makes it likely that we are not missing sources that bright even in the image with the shortest exposure time ( 280 sec for the central region of the tel image ) , as indeed the source detection histograms in fig . 5 show . comparing the four values of @xmath29 , where @xmath30 is the exposure time for the image of area @xmath25 , we find that the tel regions has a @xmath31 higher than the three other regions , which have almost identical @xmath32 values . the tel region appears to have many more uv sources that it is `` entitled '' to have , given its projected sky area and depth of exposure . the discrepancy can * not * be explained by the galactic model . the modified bahcall - soneira predictions appear to fit reasonably well both the colour and the magnitude distributions of uv sources . one possibility is that the tel region contains even more of the hot , evolved stellar types than do the other three regions . this could , in a way , be understood when considering the pointing direction : the line - of - sight to the tel region grazes the bulge of the galaxy whereas the three other images have wider separations from the galactic center . if the excess of uv sources in the direction of telescopium would be due to hot , horizontal - branch stars in the bulge , these would appear as the sources near the faint completeness limit of the faust exposure . we estimate this from the colours and magnitudes of zero - age horizontal - branch models listed in table 2 of mould ( 1996 ) and from a galactocentric distance of 8 kpc . another possibility could be for the m83 region to have , for some reason , more extinction than the tel area . we checked this using the stars identified as a0 in table 4 . there are 15 such stars in the m83 frame and 11 in the tel image . the mean b v values are 0.09@xmath230.10 and 0.06@xmath230.07 , respectively . the mean uv v values , on the other hand , are 0.68@xmath231.06 and 0.16@xmath2377 , respectively . the difference between the mean b v values of the two regions could imply an expected difference in the mean uv v values of @xmath00.15 , assuming the wavelength dependence of the extinction as in the `` typical '' galactic law ( savage & mathis 1979 ) , whereas we measure @xmath33(uv v)@xmath160.84 . this could be the case if the slope of the extinction law would be steeper in the faust - observed regions than the savage & mathis relation , and is supported also by the slope of the uv v _ vs. _ b v relation for all the a0 stars in table 4 . the 43 objects have a regression slope of 6.36@xmath231.29 , whereas the savage & mathis relation indicates a slope of 4.9 . a steeper dependence of the uv extinction may be an indicator for the presence of smaller dust grains than encountered on average in the milky way . we note this as a possibility , but do not explore it further in this paper . we presented faust uv observations in four sky area , mostly in the fourth galactic quadrant . our detection algorithm identified 777 uv sources in these regions . we identified the sources with optical counterparts mainly through correlations with catalogued object , and succeeded in increasing the number and quality of identifications by including objective - prism spectra . the stellar objects detected and identified in this program fit reasonably well our model of the uv galaxy . we discussed the more interesting stellar and extragalactic sources , and pointed out that the tel area may contain horizontal branch stars on the outskirts of the galactic bulge . this paper made use of the simbad data bank , established and maintained by the centre du donnes stellaire in strasbourg . the paper is dedicated to the memory of dr . barry lasker , one of the `` founding fathers '' of the guide star catalog and of other catalogs derived from scanning of photographic plates . uv research at tel aviv university is supported by grants from the ministry of science and arts through the israel space agency , from the austrian friends of tel aviv university , and from a center of excellence award from the israel science foundation . nb acknowledges support from a us - israel binational award to study uv sources measured by the faust experiment . we are grateful to mr . benny bilenko for updating the model of uv stellar distributions in the milky way . constuctive remarks by an anonymous referee are gratefully acknowledged . alfaro , e. & efremov , yu . n. 1996 , rev.mex.a.a . , 4 , 1 bahcall , j.n . & soneira , r.m . 1980 astrophys . j. suppl 44 , 73 . boyle , b.j . 1989 mon . not . 240 , 533 . bowyer , s. , sasseen , t.p . , lampton , m. & wu , x. 1993 , apj , 415 , 875 bowyer , s. , sasseen , p.t . , xiauyi , w. & lampton , m. 1995 , apjs , 96 , 461 ( fsc ) brosch , n. 1991 , mnras , 250 , 780 brosch , n. , almoznino , e. , leibowitz , e.m . , netzer , h. , sasseen , t. , bowyer , s. , lampton , m. & wu , x. 1995 , apj , 450 , 137 brosch , n. , formiggini , l. , almoznino , e. , sasseen , t. , lampton , m. & bowyer , s. 1997 , apjs , 111 , 143 brosch , n. , ofek , e. , almoznino , e. , sasseen , t. , lampton , m. , & bowyer , s. 1998 , mnras , 295 , 959 esa `` the hipparcos and tycho catalogues '' 1997 , esa sp-1200 hagen , h .- j . , groote , d. , engels , d. & reimers , d. 1995 , a&as , 111 , 195 jordan , c. , ayres , t.r . , brown , a. , linsky , j.l . & simon , t. 1987 , mnras , 225 , 903 kenyon , s.j . & mikolajewska , j. 1995 , aj , 110 , 391 kurucz , r.l . 1991 nato asi ser . c. , math . , 341 , 441 . lampton , m. , sasseen , t.p . , wu , x. & bowyer , s. 1993 , geophys . res . lett . * 20 * , 539 lasker , b. , jenkner , h. & russell , j.l . 1988 in `` mapping the sky : past heritage and future directions '' ( s. debarat , ed . ) , dordrecht : kluwer , p. 229 . monet , d. 1996 , `` usno - a1.0 catralog '' , washigton : u.s . naval observatory monet , d. 1998 , `` usno - a2.0 catalog '' , washigton : u.s . naval observatory mould , j.r . , watson , a.m. , gallagher , j.s . iii , ballester , g.e . , burrows , c.j . , casertano , s. , clarke , j.t . , crisp , d. , griffiths , r.e . , hester , j.j . , hoessel , j.g . , holtzman , j.a . , scowen , p.a . , stapelfeldt , k.r . , trauger , j.t . & westphal , j.a . 1996 , apj , 461 , 762 paturel , g. , vauglin , i. , garnier , r. , marthinet , m.c . , petit , c. , di nella , h. , bottinelli , l. , gouguenheim , l. & durand , n. 1992 _ leda : the lyon - meudon extragalactic database _ , cd - rom observatorie de lyon phillips , m.m . , charles , p.a . & baldwin , j.a . 1983 , apj , 266 , 485 reimers , d. & wisotzki , l. 1997 , the messenger , 88 , 14 sao 1966 , `` smithsonian astrophysical observatory star catalog '' , washington : smithsonian institution , publication no . 4652 savage , b.d . & mathis , j.s . 1979 , araa , 17 , 73 thompson , g.i . , nandy , k. , jamar , c. , monfils , a. , houziaux , l. , carnochan , d.j . & wilson , r. 1978 , _ catalogue of stellar ultraviolet fluxes _ , the science research council ( td-1 ) turon , c. 1993 bull . cds 43 . van dijk , w. , kerssiers , a. , hammerschlag - hensberg , g. & wesselius , p.r . 1978 , a&a , 66 , 187 vanelle , c. 1996 , msc thesis , hamburger sternwarte wisotzki , l. , kohler , t. , groote , d. & reimers , d. 1996 , a&as , 115 , 227 cccccccccc dor & 04 15 36 & -56 04 12 & 266.0 & -43.7 & 200 - 610 & 60.5 & 142 & 58 & 12.7 cen & 12 47 24 & -40 46 12 & 302.1 & 22.1 & 300 - 1100 & 62.0 & 249 & 129 & 12.5 m83 & 13 34 48 & -28 45 00 & 314.3 & 33.2 & 250 - 850 & 60.1 & 197 & 81 & 13.0 tel & 20 11 38 & -47 06 00 & 352.5 & -32.8 & 100 - 280 & 52.6 & 189 & 105 & 12.35 fc 1 & 12 47 36 & -37 23 22 & 11.85 & 0.40 & 3466 fc 2 & 12 55 45 & -37 17 35 & 11.26 & 0.32 & 3574 fc 3 & 12 40 22 & -37 51 52 & 7.99 & 0.17 & 3350 fc 4 & 12 56 32 & -37 39 28 & 11.75 & 0.29 & 3580 fc 5 & 12 41 54 & -37 53 32 & 10.07 & 0.21 & 3378 fc 6 & 12 42 37 & -37 54 09 & 11.30 & 0.23 & 3392 fc 7 & 13 03 21 & -37 44 39 & 10.73 & 0.29 & 3654 fc 8 & 12 56 50 & -37 57 31 & 12.19 & 0.36 & 3584 fc 9 & 12 52 21 & -38 02 20 & 13.23 & 0.76 & 3538 fc 10 & 12 53 32 & -38 03 43 & 12.56 & 0.45 & 3548 fc 11 & 12 38 03 & -38 15 09 & 11.50 & 0.33 & 3304 fc 12 & 12 49 58 & -38 18 24 & 12.36 & 0.39 & 3510 fc 13 & 12 43 48 & -38 20 52 & 11.85 & 0.34 & 3410 fc 14 & 12 48 21 & -38 22 30 & 11.40 & 0.25 & 3481 fc 15 & 12 39 42 & -38 24 46 & 11.98 & 0.37 & 3333 fc 16 & 13 04 38 & -38 23 54 & 9.39 & 0.19 & 3665 fc 17 & 12 39 09 & -38 44 46 & 10.08 & 0.20 & 3317 fc 18 & 12 34 51 & -38 49 43 & 10.39 & 0.23 & 3253 fc 19 & 12 59 41 & -38 42 32 & 8.81 & 0.17 & 3614 fc 20 & 12 47 45 & -38 51 16 & 10.49 & 0.20 & 3467 fc 21 & 12 45 27 & -38 51 25 & 11.19 & 0.24 & 3434 fc 22 & 13 00 35 & -38 44 38 & 12.24 & 0.28 & 3631 fc 23 & 12 58 01 & -38 55 04 & 7.48 & 0.16 & 3599 fc 24 & 12 40 19 & -39 02 42 & 13.07 & 0.83 & 3346 fc 25 & 12 51 54 & -38 58 24 & 13.58 & 0.92 & 3526 fc 26 & 12 31 48 & -39 13 14 & 10.70 & 0.30 & 3208 fc 27 & 12 52 43 & -39 03 40 & 12.16 & 0.40 & 3542 fc 28 & 12 58 53 & -39 06 28 & 6.73 & 0.16 & 3608 fc 29 & 13 03 37 & -38 58 24 & 10.58 & 0.20 & 3657 fc 30 & 12 57 11 & -39 06 51 & 10.80 & 0.22 & 3587 fc 31 & 13 01 04 & -39 03 05 & 11.79 & 0.30 & 3633 fc 32 & 12 36 10 & -39 22 40 & 12.51 & 0.58 & 3276 fc 33 & 12 58 01 & -39 16 06 & 9.85 & 0.18 & 3600 fc 34 & 12 50 26 & -39 26 33 & 10.94 & 0.25 & ? fc 35 & 13 00 40 & -39 17 38 & 11.67 & 0.27 & 3629 fc 36 & 12 58 12 & -39 22 23 & 11.24 & 0.20 & 3601 fc 37 & 12 53 46 & -39 29 10 & 10.85 & 0.22 & 3553 fc 38 & 12 39 18 & -39 39 35 & 9.89 & 0.22 & 3320 fc 39 & 12 52 01 & -39 41 41 & 5.23 & 0.16 & 3529 fc 40 & 12 49 09 & -39 43 25 & 5.82 & 0.16 & 3498 fc 41 & 12 35 58 & -39 52 00 & 5.22 & 0.16 & 3271 fc 42 & 12 44 12 & -39 46 12 & 9.60 & 0.18 & 3420 fc 43 & 12 39 53 & -39 58 08 & 5.42 & 0.16 & 3338 fc 44 & 12 42 08 & -40 04 42 & 10.66 & 0.26 & 3386 fc 45 & 12 43 26 & -40 10 42 & 8.47 & 0.17 & 3407 fc 46 & 12 53 25 & -40 11 10 & 6.04 & 0.16 & 3545 fc 47 & 12 31 29 & -40 16 34 & 10.63 & 0.24 & 3198 fc 48 & 12 49 41 & -40 13 26 & 12.30 & 0.46 & 3507 fc 49 & 13 03 33 & -40 14 59 & 12.64 & 0.40 & 3655 fc 50 & 12 40 21 & -40 37 31 & 10.91 & 0.24 & 3348 fc 51 & 12 49 12 & -40 37 58 & 10.12 & 0.19 & 3501 fc 52 & 12 34 13 & -40 49 54 & 11.17 & 0.24 & 3245 fc 53 & 12 48 40 & -40 42 36 & 12.77 & 0.32 & 3486 fc 54 & 12 59 07 & -40 37 53 & 11.85 & 0.33 & 3610 fc 55 & 12 51 35 & -40 44 18 & 13.81 & 1.04 & 3524 fc 56 & 12 35 46 & -41 01 15 & 6.94 & 0.16 & 3270 fc 57 & 12 46 54 & -40 53 27 & 11.60 & 0.27 & 3458 fc 58 & 13 04 47 & -40 42 34 & 11.85 & 0.30 & 3666 fc 59 & 12 52 31 & -40 56 22 & 10.71 & 0.21 & 3539 fc 60 & 12 59 55 & -40 52 40 & 9.57 & 0.18 & 3619 fc 61 & 12 45 15 & -41 02 35 & 13.06 & 0.57 & 3431 fc 62 & 13 07 39 & -40 45 59 & 12.50 & 0.49 & 3680 fc 63 & 12 38 19 & -41 12 22 & 9.76 & 0.19 & 3308 fc 64 & 12 44 40 & -41 13 17 & 12.85 & 0.52 & 3429 fc 65 & 12 51 01 & -41 15 06 & 10.22 & 0.18 & 3519 fc 66 & 12 33 35 & -41 25 24 & 8.87 & 0.17 & 3232 fc 67 & 12 52 16 & -41 19 40 & 8.01 & 0.16 & 3536 fc 68 & 13 10 37 & -40 56 50 & 11.09 & 0.28 & 3702 fc 69 & 12 31 32 & -41 31 47 & 10.15 & 0.21 & 3199 fc 70 & 13 02 40 & -41 13 42 & 11.58 & 0.26 & 3648 fc 71 & 12 47 50 & -41 28 01 & 12.58 & 0.43 & 3470 fc 72 & 12 59 55 & -41 23 11 & 11.04 & 0.22 & 3620 fc 73 & 12 43 14 & -41 36 15 & 11.97 & 0.35 & 3403 fc 74 & 13 08 15 & -41 23 05 & 9.00 & 0.17 & 3684 fc 75 & 13 04 01 & -41 24 09 & 12.65 & 0.44 & 3662 fc 76 & 12 50 25 & -41 42 59 & 11.24 & 0.26 & 3516 fc 77 & 12 59 14 & -41 36 27 & 8.94 & 0.17 & 3611 fc 78 & 12 39 26 & -41 50 48 & 9.57 & 0.18 & 3328 fc 79 & 13 05 00 & -41 30 41 & 14.48 & 2.02 & 3667 fc 80 & 12 33 50 & -41 57 49 & 11.24 & 0.28 & 3239 fc 81 & 12 30 29 & -41 57 11 & 11.05 & 0.27 & 3185 fc 82 & 12 50 25 & -41 54 09 & 10.20 & 0.20 & 3515 fc 83 & 12 37 24 & -42 01 19 & 12.22 & 0.36 & 3292 fc 84 & 13 03 01 & -41 46 57 & 10.48 & 0.20 & 3651 fc 85 & 12 56 46 & -41 55 04 & 12.09 & 0.38 & 3582 fc 86 & 12 44 19 & -42 05 35 & 12.70 & 0.50 & 3422 fc 87 & 13 01 06 & -41 58 55 & 11.48 & 0.26 & 3634 fc 88 & 12 56 09 & -42 09 10 & 7.87 & 0.17 & 3576 fc 89 & 12 36 14 & -42 18 47 & 11.92 & 0.35 & 3275 fc 90 & 12 58 23 & -42 05 43 & 13.31 & 0.80 & 3603 fc 91 & 13 07 33 & -42 01 59 & 9.37 & 0.18 & 3678 fc 92 & 12 55 10 & -42 18 26 & 4.98 & 0.16 & 3569 fc 93 & 12 39 45 & -42 21 41 & 12.09 & 0.39 & 3335 fc 94 & 13 10 06 & -42 12 23 & 9.21 & 0.18 & 3697 fc 95 & 12 37 30 & -42 35 45 & 11.20 & 0.25 & 3293 fc 96 & 13 11 08 & -42 12 60 & 11.43 & 0.25 & 3703 fc 97 & 13 00 16 & -42 29 48 & 11.74 & 0.29 & 3624 fc 98 & 13 01 28 & -42 29 59 & 12.20 & 0.34 & 3636 fc 99 & 13 02 12 & -42 28 60 & 13.25 & 0.40 & 3646 fc 100 & 12 39 33 & -42 58 09 & 11.60 & 0.33 & 3329 fc 101 & 12 52 05 & -42 55 28 & 10.53 & 0.21 & 3530 fc 102 & 12 38 01 & -43 01 17 & 12.49 & 0.33 & 3305 fc 103 & 12 45 59 & -43 05 06 & 7.91 & 0.16 & 3444 fc 104 & 12 38 03 & -43 03 35 & 12.51 & 0.33 & ? fc 105 & 12 53 29 & -43 04 17 & 9.41 & 0.18 & 3546 fc 106 & 12 51 40 & -43 10 34 & 11.46 & 0.30 & 3525 fc 107 & 12 49 28 & -43 12 44 & 14.67 & 2.49 & 3503 fc 108 & 12 48 22 & -43 19 37 & 11.89 & 0.35 & 3482 fc 109 & 12 41 31 & -43 37 55 & 9.46 & 0.18 & 3372 fc 110 & 13 01 30 & -43 22 37 & 10.42 & 0.20 & 3639 fc 111 & 13 02 39 & -43 20 59 & 10.36 & 0.21 & 3647 fc 112 & 12 48 45 & -43 42 36 & 7.55 & 0.16 & 3489 fc 113 & 13 08 57 & -43 27 50 & 8.38 & 0.17 & 3689 fc 114 & 12 35 49 & -43 49 27 & 11.35 & 0.31 & 3269 fc 115 & 12 42 02 & -43 53 10 & 11.96 & 0.35 & 3379 fc 116 & 12 56 14 & -43 46 40 & 11.70 & 0.32 & 3577 fc 117 & 12 49 02 & -43 56 26 & 9.51 & 0.18 & 3494 fc 118 & 13 00 55 & -43 47 50 & 10.83 & 0.24 & 3632 fc 119 & 12 43 22 & -44 00 13 & 12.09 & 0.28 & 3404 fc 120 & 12 42 49 & -44 05 49 & 11.72 & 0.26 & 3395 fc 121 & 12 59 20 & -43 59 42 & 11.01 & 0.24 & 3613 fc 122 & 13 04 58 & -43 54 50 & 10.49 & 0.23 & 3668 fc 123 & 12 55 10 & -44 10 27 & 12.27 & 0.35 & 3567 fc 124 & 12 41 25 & -44 17 34 & 13.25 & 1.03 & 3368 fc 125 & 13 06 11 & -44 06 08 & 11.80 & 0.45 & 3674 fc 126 & 12 54 15 & -44 23 15 & 12.16 & 0.35 & 3559 fc 127 & 12 43 06 & -44 34 02 & 10.56 & 0.23 & 3401 fc 128 & 12 48 18 & -44 43 22 & 5.92 & 0.16 & 3479 fc 129 & 12 41 14 & -44 41 51 & 11.06 & 0.30 & 3363 fc 130 & 13 06 15 & -44 24 36 & 9.92 & 0.22 & 3673 fc 131 & 12 45 30 & -44 43 49 & 11.01 & 0.25 & 3437 fc 132 & 13 03 25 & -44 34 04 & 8.75 & 0.18 & 3653 fc 133 & 13 01 20 & -44 45 55 & 5.69 & 0.16 & 3635 fc 134 & 12 56 24 & -44 51 50 & 11.93 & 0.40 & 3578 fc 135 & 13 04 04 & -44 55 37 & 10.34 & 0.27 & 3663 fc 136 & 12 53 52 & -45 05 47 & 11.64 & 0.36 & 3555 fc 137 & 13 01 29 & -45 11 02 & 9.40 & 0.21 & 3637 fc 138 & 12 48 39 & -45 25 42 & 10.85 & 0.27 & 3488 fc 139 & 12 46 05 & -45 29 59 & 9.98 & 0.23 & 3448 fc 140 & 12 52 24 & -45 30 01 & 11.16 & 0.37 & 3537 fc 141 & 12 44 33 & -37 18 20 & 11.95 & 0.46 & ? fc 142 & 12 53 14 & -37 34 32 & 11.62 & 0.16 & ? fc 143 & 12 45 01 & -37 43 04 & 12.41 & 0.51 & ? fc 144 & 12 50 55 & -37 50 52 & 12.81 & 0.64 & ? fc 145 & 13 00 32 & -37 49 28 & 12.16 & 0.44 & ? fc 146 & 12 32 32 & -38 31 40 & 11.72 & 0.37 & ? fc 147 & 12 32 38 & -38 34 39 & 11.77 & 0.37 & ? fc 148 & 12 50 20 & -38 29 37 & 13.22 & 0.80 & 3514 fc 149 & 12 36 00 & -38 38 02 & 11.62 & 0.16 & ? fc 150 & 12 56 00 & -38 39 57 & 15.18 & 3.90 & ? fc 151 & 12 53 32 & -38 47 20 & 13.23 & 0.76 & 3544 fc 152 & 12 44 09 & -38 58 16 & 13.96 & 1.38 & ? fc 153 & 12 49 39 & -39 01 08 & 11.62 & 0.16 & ? fc 154 & 12 44 42 & -39 06 17 & 13.55 & 0.86 & ? fc 155 & 12 40 10 & -39 22 23 & 11.77 & 0.33 & 3344 fc 156 & 12 41 23 & -39 25 15 & 12.36 & 0.45 & ? fc 157 & 13 00 02 & -39 14 40 & 12.64 & 0.28 & 3621 fc 158 & 12 54 19 & -39 19 14 & 12.19 & 0.36 & 3560 fc 159 & 12 43 12 & -39 31 42 & 12.25 & 0.40 & 3402 fc 160 & 12 47 27 & -39 30 35 & 13.94 & 0.68 & ? fc 161 & 12 47 27 & -39 32 58 & 13.32 & 0.45 & ? fc 162 & 12 30 26 & -39 40 36 & 12.49 & 0.71 & ? fc 163 & 12 45 13 & -39 37 46 & 14.01 & 1.42 & ? fc 164 & 12 33 23 & -39 48 36 & 11.62 & 0.16 & ? fc 165 & 12 47 33 & -39 47 24 & 11.51 & 0.27 & 3465 fc 166 & 12 49 37 & -39 54 43 & 12.85 & 0.48 & ? fc 167 & 12 34 38 & -40 01 57 & 11.21 & 0.32 & ? fc 168 & 12 47 21 & -40 06 59 & 13.81 & 1.32 & ? fc 169 & 13 09 18 & -39 51 04 & 16.14 & 0.16 & ? fc 170 & 12 30 34 & -40 16 59 & 12.41 & 0.37 & ? fc 171 & 12 45 27 & -40 18 07 & 13.81 & 1.09 & ? fc 172 & 12 47 53 & -40 22 24 & 11.62 & 0.73 & ? fc 173 & 12 38 12 & -40 29 26 & 11.30 & 0.30 & ? fc 174 & 12 56 36 & -40 21 02 & 12.97 & 0.58 & ? fc 175 & 12 59 20 & -40 25 47 & 10.04 & 0.19 & 3612 fc 176 & 12 58 18 & -40 36 17 & 14.97 & 1.65 & ? fc 177 & 12 36 57 & -40 50 55 & 13.29 & 0.82 & ? fc 178 & 13 01 46 & -40 36 22 & 12.02 & 0.32 & 3642 fc 179 & 13 04 42 & -40 32 08 & 13.64 & 0.54 & ? fc 180 & 13 07 47 & -40 30 09 & 11.62 & 0.16 & ? fc 181 & 13 11 23 & -40 28 46 & 12.16 & 0.55 & ? fc 182 & 12 38 51 & -41 01 41 & 13.65 & 1.08 & ? fc 183 & 12 33 56 & -41 02 06 & 11.62 & 0.16 & ? fc 184 & 12 32 04 & -41 04 27 & 15.87 & 9.24 & 3213 fc 185 & 12 50 00 & -41 05 35 & 11.62 & 0.16 & ? fc 186 & 12 59 17 & -40 59 13 & 13.68 & 0.61 & ? fc 187 & 12 49 05 & -41 11 57 & 12.92 & 0.57 & ? fc 188 & 13 05 57 & -40 56 55 & 13.97 & 1.38 & ? fc 189 & 12 41 45 & -41 17 18 & 12.78 & 0.59 & ? fc 190 & 12 38 17 & -41 22 22 & 14.59 & 1.11 & ? fc 191 & 12 30 06 & -41 24 33 & 12.02 & 0.44 & ? fc 192 & 12 58 50 & -41 10 32 & 13.02 & 0.59 & ? fc 193 & 12 53 41 & -41 18 59 & 15.09 & 3.51 & 3550 fc 194 & 12 33 02 & -41 38 48 & 17.55 & 0.16 & ? fc 195 & 13 10 44 & -41 19 13 & 12.03 & 0.38 & ? fc 196 & 12 45 48 & -41 49 08 & 14.74 & 2.19 & ? fc 197 & 12 52 02 & -41 52 05 & 13.06 & 0.39 & ? fc 198 & 12 52 36 & -41 56 33 & 12.85 & 0.38 & ? fc 199 & 12 52 46 & -41 58 17 & 13.02 & 0.42 & ? fc 200 & 12 53 18 & -42 00 55 & 12.61 & 0.32 & ? fc 201 & 12 42 43 & -42 10 01 & 13.14 & 0.70 & ? fc 202 & 13 08 17 & -41 57 20 & 13.00 & 0.49 & ? fc 203 & 12 42 01 & -42 17 44 & 14.13 & 1.69 & 3383 fc 204 & 13 01 25 & -42 07 19 & 13.80 & 0.62 & ? fc 205 & 13 04 28 & -42 06 35 & 11.62 & 0.16 & ? fc 206 & 12 49 14 & -42 27 49 & 16.28 & 9.22 & ? fc 207 & 12 46 01 & -42 31 38 & 15.17 & 3.01 & ? fc 208 & 12 39 02 & -42 34 49 & 11.62 & 0.16 & ? fc 209 & 12 52 11 & -42 31 26 & 12.81 & 0.64 & ? fc 210 & 12 35 17 & -42 38 58 & 12.48 & 0.53 & 3264 fc 211 & 12 49 07 & -42 38 11 & 13.50 & 0.83 & 3495 fc 212 & 12 51 22 & -42 43 03 & 12.66 & 0.48 & ? fc 213 & 12 59 42 & -42 35 02 & 13.16 & 0.44 & ? fc 214 & 12 44 16 & -42 46 34 & 13.98 & 1.30 & 3418 fc 215 & 13 08 51 & -42 27 57 & 13.66 & 1.17 & 3688 fc 216 & 12 47 52 & -42 48 46 & 14.45 & 2.25 & ? fc 217 & 12 44 22 & -43 02 09 & 12.64 & 0.50 & ? fc 218 & 13 10 02 & -42 41 48 & 12.14 & 0.46 & ? fc 219 & 12 41 21 & -43 15 57 & 13.56 & 1.15 & ? fc 220 & 13 04 17 & -43 02 40 & 14.07 & 0.91 & ? fc 221 & 13 01 51 & -43 05 11 & 14.34 & 1.69 & ? fc 222 & 12 39 00 & -43 25 44 & 12.54 & 0.49 & 3312 fc 223 & 13 04 29 & -43 08 36 & 13.27 & 0.54 & ? fc 224 & 12 44 43 & -43 25 14 & 11.62 & 0.16 & ? fc 225 & 12 57 47 & -43 17 47 & 12.78 & 0.35 & 3591 fc 226 & 12 57 04 & -43 19 39 & 12.98 & 0.43 & ? fc 227 & 12 58 53 & -43 27 14 & 11.62 & 1.01 & ? fc 228 & 13 00 32 & -43 25 24 & 13.44 & 0.49 & ? fc 229 & 13 10 05 & -43 16 00 & 11.92 & 0.46 & ? fc 230 & 13 04 32 & -43 23 45 & 12.77 & 0.59 & ? fc 231 & 13 02 55 & -43 31 37 & 13.48 & 0.63 & ? fc 232 & 13 07 20 & -43 35 40 & 13.25 & 1.03 & ? fc 233 & 13 02 19 & -43 40 23 & 14.42 & 2.37 & ? fc 234 & 13 08 22 & -43 40 12 & 14.24 & 3.01 & ? fc 235 & 12 59 06 & -43 51 13 & 12.74 & 0.34 & 3609 fc 236 & 12 51 07 & -44 12 28 & 13.62 & 1.21 & ? fc 237 & 12 58 52 & -44 18 45 & 15.73 & 7.63 & ? fc 238 & 13 00 44 & -44 15 45 & 12.58 & 0.61 & ? fc 239 & 12 48 57 & -44 28 34 & 12.98 & 0.46 & ? fc 240 & 13 03 53 & -44 20 17 & 12.57 & 0.42 & ? fc 241 & 13 03 58 & -44 22 34 & 12.86 & 0.52 & ? fc 242 & 12 59 55 & -44 27 30 & 11.71 & 0.40 & ? fc 243 & 12 51 09 & -44 51 50 & 11.84 & 0.38 & ? fc 244 & 12 57 28 & -45 07 36 & 12.00 & 0.48 & ? fc 245 & 12 57 23 & -45 26 17 & 12.69 & 1.26 & ? fc 246 & 12 58 50 & -40 27 18 & 11.23 & 0.21 & ? fc 247 & 12 51 41 & -41 18 49 & 9.52 & 0.17 & 3522 fc 248 & 12 52 43 & -41 11 23 & 12.18 & 0.27 & 3541 fc 249 & 12 52 10 & -43 02 07 & 12.10 & 0.27 & 3531 fd 1 & 3 48 31 & -54 16 44 & 9.22 & 0.20 & 497 fd 2 & 3 52 58 & -54 05 47 & 9.36 & 0.19 & 499 fd 3 & 3 53 27 & -54 50 39 & 6.13 & 0.16 & 500 fd 4 & 4 03 38 & -52 55 38 & 10.87 & 0.24 & 516 fd 5 & 3 48 06 & -56 41 44 & 12.74 & 0.93 & 496 fd 6 & 3 59 59 & -54 10 23 & 12.25 & 0.43 & 509 fd 7 & 3 57 09 & -55 05 33 & 12.32 & 0.49 & 505 fd 8 & 4 00 59 & -54 23 16 & 8.82 & 0.17 & 510 fd 9 & 3 51 37 & -56 39 25 & 9.56 & 0.19 & 498 fd 10 & 4 03 16 & -54 39 51 & 11.66 & 0.30 & 514 fd 11 & 3 56 30 & -56 19 37 & 11.72 & 0.33 & 504 fd 12 & 3 57 42 & -56 15 16 & 11.17 & 0.26 & 506 fd 13 & 4 03 49 & -55 24 23 & 9.14 & 0.18 & 518 fd 14 & 4 15 23 & -52 09 32 & 9.33 & 0.19 & 534 fd 15 & 3 59 35 & -56 28 21 & 9.46 & 0.18 & 508 fd 16 & 4 10 25 & -53 34 56 & 12.27 & 0.36 & 525 fd 17 & 4 01 12 & -56 19 48 & 8.91 & 0.17 & 512 fd 18 & 3 58 37 & -57 06 20 & 10.56 & 0.22 & 507 fd 19 & 4 11 07 & -54 03 12 & 12.73 & 0.56 & 528 fd 20 & 4 11 43 & -53 51 17 & 12.93 & 0.68 & ? fd 21 & 4 13 41 & -53 26 35 & 11.26 & 0.25 & 531 fd 22 & 3 55 52 & -58 19 19 & 11.99 & 0.46 & 503 fd 23 & 4 18 36 & -52 52 09 & 11.27 & 0.26 & 542 fd 24 & 4 09 24 & -55 44 52 & 12.53 & 0.53 & 524 fd 25 & 4 02 39 & -57 27 48 & 11.38 & 0.30 & 513 fd 26 & 4 06 34 & -56 34 50 & 10.94 & 0.25 & 522 fd 27 & 4 16 05 & -54 01 25 & 11.93 & 0.36 & 537 fd 28 & 4 04 25 & -57 09 57 & 11.89 & 0.33 & 519 fd 29 & 4 15 26 & -54 22 43 & 10.60 & 0.21 & 536 fd 30 & 4 19 22 & -53 19 25 & 12.26 & 0.39 & 545 fd 31 & 4 00 53 & -58 39 30 & 7.76 & 0.17 & 511 fd 32 & 4 18 46 & -54 08 43 & 7.49 & 0.16 & 543 fd 33 & 4 19 05 & -53 51 01 & 12.45 & 0.45 & 544 fd 34 & 4 19 59 & -54 04 45 & 10.53 & 0.21 & 548 fd 35 & 4 17 09 & -54 58 14 & 11.30 & 0.24 & 539 fd 36 & 4 04 58 & -58 09 44 & 12.59 & 0.55 & 520 fd 37 & 4 22 54 & -53 32 18 & 12.57 & 0.43 & 552 fd 38 & 4 05 35 & -58 28 31 & 9.36 & 0.18 & 521 fd 39 & 4 14 07 & -56 11 16 & 12.14 & 0.42 & 533 fd 40 & 4 25 31 & -53 06 01 & 10.40 & 0.23 & 555 fd 41 & 4 19 57 & -54 56 46 & 9.68 & 0.19 & 547 fd 42 & 4 22 34 & -54 09 18 & 10.09 & 0.19 & 551 fd 43 & 4 08 49 & -58 29 04 & 9.56 & 0.18 & 523 fd 44 & 4 23 32 & -54 17 51 & 13.01 & 0.70 & 553 fd 45 & 4 27 24 & -53 23 44 & 11.06 & 0.28 & 562 fd 46 & 4 28 30 & -53 27 15 & 10.10 & 0.20 & 565 fd 47 & 4 17 30 & -56 44 23 & 12.74 & 0.52 & 540 fd 48 & 4 13 40 & -58 01 10 & 9.14 & 0.18 & 532 fd 49 & 4 29 39 & -53 13 03 & 11.62 & 0.16 & ? fd 50 & 4 11 39 & -58 41 36 & 12.30 & 0.44 & 529 fd 51 & 4 25 48 & -55 07 15 & 11.90 & 0.35 & 556 fd 52 & 4 31 12 & -53 35 18 & 11.14 & 0.31 & 567 fd 53 & 4 27 55 & -55 04 09 & 12.29 & 0.48 & 564 fd 54 & 4 22 06 & -56 59 34 & 11.42 & 0.27 & 550 fd 55 & 4 21 19 & -57 32 32 & 12.69 & 0.59 & 549 fd 56 & 4 27 03 & -56 05 19 & 10.12 & 0.20 & 561 fd 57 & 4 24 06 & -57 15 42 & 7.71 & 0.16 & 554 fd 58 & 4 16 39 & -59 17 34 & 9.74 & 0.19 & 538 fd 59 & 4 26 35 & -56 27 53 & 12.15 & 0.37 & 559 fd 60 & 4 19 13 & -58 44 49 & 8.75 & 0.17 & 546 fd 61 & 4 36 18 & -53 42 58 & 8.15 & 0.17 & 572 fd 62 & 4 25 56 & -57 12 13 & 12.03 & 0.36 & 557 fd 63 & 4 34 20 & -55 00 41 & 5.03 & 0.16 & 570 fd 64 & 4 31 08 & -55 55 31 & 11.35 & 0.32 & 566 fd 65 & 4 32 43 & -55 35 23 & 10.28 & 0.21 & 569 fd 66 & 4 36 33 & -54 37 25 & 9.92 & 0.24 & 576 fd 67 & 4 26 33 & -59 06 07 & 10.51 & 0.23 & 560 fd 68 & 4 34 48 & -56 36 07 & 11.51 & 0.29 & 571 fd 69 & 4 36 06 & -56 17 32 & 12.22 & 0.49 & 573 fd 70 & 4 40 50 & -56 02 32 & 10.58 & 0.24 & 584 fd 71 & 4 40 20 & -56 11 39 & 12.29 & 0.32 & 583 fd 72 & 4 37 03 & -57 28 29 & 12.27 & 0.50 & 578 fd 73 & 4 39 21 & -57 19 34 & 11.03 & 0.25 & 582 fd 74 & 4 36 10 & -58 36 40 & 12.21 & 0.56 & 575 fd 75 & 4 42 39 & -56 41 59 & 11.77 & 0.33 & 587 fd 76 & 4 38 51 & -58 58 05 & 8.83 & 0.18 & 580 fd 77 & 4 40 14 & -58 44 35 & 11.64 & 0.39 & ? fd 78 & 4 03 33 & -52 28 53 & 11.07 & 0.29 & 515 fd 79 & 3 55 31 & -53 29 43 & 13.68 & 1.56 & ? fd 80 & 3 57 34 & -54 20 33 & 13.39 & 0.94 & ? fd 81 & 3 48 33 & -56 32 11 & 14.59 & 3.76 & ? fd 82 & 4 06 00 & -52 39 49 & 12.62 & 0.57 & ? fd 83 & 3 58 19 & -54 51 39 & 12.38 & 0.39 & ? fd 84 & 3 51 34 & -56 23 23 & 12.24 & 0.56 & ? fd 85 & 4 09 29 & -52 28 06 & 12.69 & 0.63 & ? fd 86 & 4 04 01 & -54 04 22 & 12.90 & 0.67 & ? fd 87 & 4 13 30 & -52 24 19 & 13.56 & 1.12 & ? fd 88 & 4 00 14 & -56 05 14 & 11.62 & 0.16 & ? fd 89 & 4 06 09 & -54 35 51 & 11.62 & 0.16 & ? fd 90 & 4 13 23 & -52 53 27 & 13.71 & 1.18 & 530 fd 91 & 3 55 11 & -57 45 16 & 13.22 & 1.01 & ? fd 92 & 4 03 47 & -56 01 00 & 13.17 & 0.89 & ? fd 93 & 4 06 59 & -55 19 29 & 12.89 & 0.68 & ? fd 94 & 4 12 55 & -54 01 46 & 13.64 & 0.97 & ? fd 95 & 4 08 50 & -55 37 09 & 11.62 & 0.16 & ? fd 96 & 4 03 03 & -57 41 48 & 13.50 & 0.98 & ? fd 97 & 4 10 56 & -56 28 58 & 12.92 & 0.58 & 527 fd 98 & 4 00 56 & -58 55 36 & 11.44 & 0.32 & ? fd 99 & 4 03 41 & -58 57 02 & 12.94 & 0.83 & ? fd 100 & 4 26 31 & -53 09 01 & 11.73 & 0.37 & 558 fd 101 & 4 10 21 & -57 47 14 & 12.89 & 0.69 & 526 fd 102 & 4 14 18 & -57 02 59 & 14.31 & 1.89 & ? fd 103 & 4 29 47 & -52 58 52 & 13.13 & 0.95 & ? fd 104 & 4 17 12 & -57 01 21 & 12.93 & 0.64 & ? fd 105 & 4 26 43 & -54 16 53 & 13.17 & 0.81 & ? fd 106 & 4 23 34 & -55 53 51 & 13.36 & 0.90 & ? fd 107 & 4 11 47 & -59 08 38 & 12.61 & 0.52 & ? fd 108 & 4 31 54 & -53 29 58 & 13.93 & 1.01 & ? fd 109 & 4 17 15 & -58 11 30 & 12.45 & 0.52 & 541 fd 110 & 4 33 51 & -53 09 36 & 14.89 & 4.19 & ? fd 111 & 4 32 00 & -53 46 44 & 14.16 & 2.76 & ? fd 112 & 4 26 03 & -56 20 39 & 13.21 & 0.41 & ? fd 113 & 4 29 38 & -57 10 21 & 13.23 & 0.81 & ? fd 114 & 4 28 51 & -57 36 39 & 13.02 & 0.70 & ? fd 115 & 4 40 48 & -54 04 44 & 13.03 & 1.01 & ? fd 116 & 4 37 31 & -55 41 40 & 11.28 & 0.34 & ? fd 117 & 4 30 36 & -58 10 39 & 12.07 & 0.42 & ? fd 118 & 4 37 41 & -55 56 25 & 12.19 & 0.60 & ? fd 119 & 4 39 14 & -55 38 19 & 14.35 & 3.66 & ? fd 120 & 4 33 05 & -57 32 35 & 13.22 & 0.70 & ? fd 121 & 4 37 34 & -56 20 24 & 11.62 & 0.16 & ? fd 122 & 4 36 53 & -56 38 57 & 11.84 & 0.39 & 577 fd 123 & 4 40 20 & -55 34 29 & 11.62 & 0.16 & ? fd 124 & 4 43 14 & -54 41 49 & 12.18 & 0.58 & ? fd 125 & 4 30 60 & -58 52 52 & 13.05 & 1.05 & ? fd 126 & 4 43 58 & -54 47 09 & 12.73 & 0.86 & ? fd 127 & 4 36 11 & -57 52 50 & 15.23 & 5.16 & ? fd 128 & 4 43 22 & -55 46 25 & 13.01 & 0.75 & ? fd 129 & 4 33 51 & -58 51 57 & 14.51 & 3.81 & ? fd 130 & 4 42 19 & -56 07 32 & 12.83 & 0.73 & 586 fd 131 & 4 45 20 & -55 06 26 & 13.33 & 1.40 & ? fd 132 & 4 35 31 & -58 28 25 & 14.37 & 2.75 & ? fd 133 & 4 40 38 & -56 54 25 & 14.25 & 1.51 & ? fd 134 & 4 43 47 & -56 46 10 & 12.69 & 0.59 & ? fd 135 & 4 44 51 & -56 42 02 & 12.23 & 0.45 & ? fd 136 & 4 40 46 & -58 29 47 & 12.19 & 0.53 & ? fd 137 & 4 46 02 & -56 54 28 & 14.87 & 4.03 & ? fd 138 & 4 03 41 & -52 11 25 & 9.11 & 0.18 & 517 fd 139 & 3 56 01 & -52 51 56 & 11.07 & 0.35 & 502 fd 140 & 3 47 06 & -58 06 32 & 7.79 & 0.17 & 495 fd 141 & 4 31 20 & -59 45 28 & 8.92 & 0.19 & 568 fd 142 & 3 55 17 & -54 52 36 & 11.55 & 0.25 & 501 fm 1 & 13 55 36 & -28 59 15 & 12.40 & 0.50 & 3942/3 fm 2 & 13 53 60 & -28 09 49 & 7.70 & 0.17 & 3938 fm 3 & 13 53 60 & -27 56 29 & 10.24 & 0.21 & 3937 fm 4 & 13 53 42 & -27 07 16 & 11.35 & 0.33 & 3935 fm 5 & 13 53 14 & -27 20 60 & 11.67 & 0.35 & 3934 fm 6 & 13 52 56 & -27 41 09 & 10.87 & 0.24 & 3930 fm 7 & 13 53 54 & -30 01 20 & 9.29 & 0.18 & 3936 fm 8 & 13 52 37 & -27 31 43 & 9.37 & 0.18 & 3929 fm 9 & 13 52 18 & -26 48 08 & 11.35 & 0.30 & 3928 fm 10 & 13 52 17 & -30 47 45 & 9.17 & 0.18 & 3927 fm 11 & 13 51 23 & -29 07 27 & 11.35 & 0.25 & 3923 fm 12 & 13 51 52 & -30 48 28 & 9.16 & 0.18 & 3924 fm 13 & 13 50 03 & -29 05 48 & 5.92 & 0.16 & 3915 fm 14 & 13 51 22 & -30 48 10 & 12.36 & 0.34 & ? fm 15 & 13 49 53 & -28 11 48 & 10.30 & 0.20 & 3914 fm 16 & 13 49 11 & -26 55 26 & 11.13 & 0.25 & 3911 fm 17 & 13 50 39 & -30 15 24 & 12.21 & 0.38 & 3917 fm 18 & 13 49 46 & -28 38 55 & 11.77 & 0.31 & 3913 fm 19 & 13 49 12 & -27 25 49 & 11.61 & 0.29 & 3912 fm 20 & 13 50 40 & -31 12 27 & 9.02 & 0.18 & 3918 fm 21 & 13 50 21 & -31 03 57 & 9.58 & 0.18 & 3916 fm 22 & 13 49 11 & -28 35 27 & 12.67 & 0.32 & 3910 fm 23 & 13 48 27 & -27 56 22 & 13.13 & 0.70 & 3907 fm 24 & 13 48 06 & -29 19 50 & 12.59 & 0.46 & 3905 fm 25 & 13 45 38 & -26 06 26 & 5.51 & 0.16 & 3895 fm 26 & 13 47 51 & -31 32 40 & 11.40 & 0.30 & 3903 fm 27 & 13 46 27 & -28 02 03 & 12.25 & 0.35 & 3898 fm 28 & 13 47 23 & -30 27 02 & 12.57 & 0.47 & 3902 fm 29 & 13 46 57 & -29 41 34 & 13.35 & 0.70 & 3900 fm 30 & 13 45 33 & -27 51 33 & 12.17 & 0.32 & 3894 fm 31 & 13 45 11 & -27 43 07 & 11.44 & 0.25 & 3892 fm 32 & 13 46 27 & -30 26 01 & 13.52 & 0.85 & ? fm 33 & 13 46 06 & -30 12 11 & 11.89 & 0.30 & 3896 fm 34 & 13 44 38 & -26 46 33 & 10.63 & 0.20 & 3891 fm 35 & 13 46 24 & -31 45 34 & 12.60 & 0.54 & 3897 fm 36 & 13 45 16 & -29 01 15 & 12.80 & 0.52 & 3893 fm 37 & 13 43 52 & -26 47 37 & 12.22 & 0.37 & 3880 fm 38 & 13 44 10 & -28 16 50 & 12.86 & 0.53 & 3886 fm 39 & 13 43 53 & -27 46 10 & 11.85 & 0.29 & 3882 fm 40 & 13 43 54 & -28 42 03 & 12.75 & 0.46 & 3881/3 fm 41 & 13 44 25 & -30 07 07 & 10.94 & 0.21 & 3888 fm 42 & 13 44 10 & -31 49 07 & 11.81 & 0.32 & 3885 fm 43 & 13 43 00 & -30 10 02 & 9.71 & 0.18 & 3877 fm 44 & 13 43 28 & -31 26 27 & 12.11 & 0.35 & 3879 fm 45 & 13 41 31 & -27 16 16 & 9.93 & 0.19 & 3871 fm 46 & 13 42 15 & -29 45 20 & 10.74 & 0.21 & 3874 fm 47 & 13 41 42 & -28 45 09 & 12.31 & 0.26 & 3872 fm 48 & 13 42 32 & -31 02 37 & 12.32 & 0.35 & 3875 fm 49 & 13 41 42 & -29 53 43 & 11.14 & 0.23 & 3873 fm 50 & 13 40 47 & -28 14 29 & 12.01 & 0.34 & 3866 fm 51 & 13 40 53 & -28 57 18 & 12.40 & 0.42 & 3867 fm 52 & 13 39 54 & -27 08 47 & 10.00 & 0.19 & 3863 fm 53 & 13 41 13 & -31 34 28 & 12.43 & 0.27 & 3868 fm 54 & 13 39 42 & -28 08 13 & 10.71 & 0.21 & 3860 fm 55 & 13 40 14 & -29 33 34 & 12.02 & 0.32 & 3865 fm 56 & 13 38 44 & -27 15 36 & 9.64 & 0.18 & 3853 fm 57 & 13 38 31 & -26 54 19 & 9.63 & 0.18 & 3850 fm 58 & 13 39 57 & -31 38 05 & 9.89 & 0.19 & 3862 fm 59 & 13 39 10 & -30 14 09 & 12.44 & 0.37 & 3856 fm 60 & 13 38 42 & -29 32 48 & 9.95 & 0.18 & 3851 fm 61 & 13 39 05 & -30 44 04 & 12.34 & 0.34 & 3855 fm 62 & 13 39 44 & -32 29 31 & 12.38 & 0.46 & 3861 fm 63 & 13 37 35 & -27 03 44 & 10.86 & 0.22 & 3848 fm 64 & 13 38 23 & -29 49 40 & 11.02 & 0.23 & 3849 fm 65 & 13 36 48 & -26 30 13 & 7.37 & 0.16 & 3838 fm 66 & 13 37 14 & -26 42 39 & 12.18 & 0.35 & 3845 fm 67 & 13 38 46 & -31 50 34 & 11.74 & 0.31 & 3852 fm 68 & 13 37 04 & -27 39 32 & 12.87 & 0.57 & 3843 fm 69 & 13 37 03 & -27 54 14 & 12.59 & 0.46 & 3842 fm 70 & 13 37 00 & -29 50 41 & 7.92 & 0.17 & 3840 fm 71 & 13 35 53 & -25 56 39 & 12.21 & 0.38 & 3833 fm 72 & 13 37 36 & -32 08 53 & 11.28 & 0.24 & 3847 fm 73 & 13 36 29 & -29 07 03 & 12.25 & 0.35 & 3835 fm 74 & 13 36 06 & -28 41 47 & 11.86 & 0.29 & 3834 fm 75 & 13 36 41 & -30 17 51 & 13.03 & 0.56 & 3836 fm 76 & 13 36 59 & -32 18 02 & 12.28 & 0.28 & 3839 fm 77 & 13 36 49 & -32 24 45 & 10.66 & 0.22 & 3837 fm 78 & 13 35 01 & -26 07 13 & 12.67 & 0.51 & 3824 fm 79 & 13 35 31 & -29 08 32 & 8.42 & 0.17 & 3827 fm 80 & 13 35 42 & -29 21 49 & 11.48 & 0.25 & 3830 fm 81 & 13 34 12 & -25 25 52 & 8.86 & 0.17 & 3818 fm 82 & 13 35 39 & -31 25 04 & 12.52 & 0.26 & 3828 fm 83 & 13 35 40 & -32 01 15 & 10.73 & 0.21 & 3829 fm 84 & 13 34 13 & -27 55 12 & 11.99 & 0.34 & 3819 fm 85 & 13 35 15 & -32 13 16 & 9.82 & 0.19 & 3826 fm 86 & 13 34 47 & -31 11 39 & 12.80 & 0.48 & 3821 fm 87 & 13 32 54 & -26 36 00 & 9.77 & 0.18 & 3811 fm 88 & 13 33 10 & -29 17 04 & 10.54 & 0.18 & 3815 fm 89 & 13 32 34 & -28 41 37 & 5.47 & 0.16 & 3806 fm 90 & 13 33 21 & -31 39 08 & 10.08 & 0.18 & 3816 fm 91 & 13 32 36 & -29 17 04 & 9.91 & 0.17 & 3805 fm 92 & 13 31 49 & -26 09 20 & 12.10 & 0.37 & 3803 fm 93 & 13 32 37 & -29 33 07 & 11.41 & 0.25 & 3808 fm 94 & 13 33 09 & -31 47 29 & 10.83 & 0.21 & 3812 fm 95 & 13 31 33 & -28 06 44 & 6.69 & 0.16 & 3801 fm 96 & 13 31 01 & -25 42 36 & 9.06 & 0.18 & 3796 fm 97 & 13 31 22 & -27 32 14 & 11.69 & 0.29 & 3798 fm 98 & 13 32 45 & -32 46 54 & 12.04 & 0.39 & 3810 fm 99 & 13 31 48 & -29 42 25 & 12.26 & 0.39 & 3802 fm 100 & 13 32 12 & -32 06 22 & 12.16 & 0.25 & 3804 fm 101 & 13 31 36 & -32 08 57 & 10.86 & 0.22 & 3799 fm 102 & 13 30 02 & -27 02 52 & 10.71 & 0.22 & 3792 fm 103 & 13 29 48 & -26 35 11 & 8.56 & 0.17 & 3790 fm 104 & 13 30 57 & -31 44 42 & 11.77 & 0.29 & 3794 fm 105 & 13 29 33 & -27 23 39 & 10.23 & 0.19 & 3789 fm 106 & 13 29 54 & -28 56 11 & 10.00 & 0.18 & 3791 fm 107 & 13 29 34 & -29 04 58 & 11.08 & 0.19 & 3788 fm 108 & 13 28 17 & -26 34 47 & 11.17 & 0.24 & 3780 fm 109 & 13 28 42 & -31 59 43 & 11.69 & 0.31 & 3783 fm 110 & 13 28 01 & -29 46 48 & 12.83 & 0.49 & 3779 fm 111 & 13 27 12 & -29 03 28 & 11.15 & 0.22 & 3777 fm 112 & 13 26 51 & -29 51 53 & 12.87 & 0.50 & 3775 fm 113 & 13 26 03 & -27 28 15 & 11.62 & 0.27 & 3771 fm 114 & 13 25 60 & -30 23 11 & 11.71 & 0.20 & 3769 fm 115 & 13 25 14 & -30 22 12 & 8.68 & 0.17 & 3765 fm 116 & 13 24 58 & -31 20 15 & 11.17 & 0.25 & 3764 fm 117 & 13 23 51 & -26 53 60 & 11.05 & 0.24 & 3761 fm 118 & 13 22 43 & -28 47 03 & 11.08 & 0.23 & 3758 fm 119 & 13 22 45 & -30 11 08 & 11.43 & 0.24 & 3757 fm 120 & 13 21 58 & -28 26 52 & 12.02 & 0.34 & 3754 fm 121 & 13 20 46 & -29 29 29 & 11.11 & 0.24 & 3745 fm 122 & 13 19 39 & -30 20 40 & 6.21 & 0.16 & 3737 fm 123 & 13 18 02 & -27 36 19 & 11.28 & 0.33 & 3730 fm 124 & 13 55 02 & -29 07 28 & 12.83 & 0.69 & ? fm 125 & 13 51 24 & -27 24 05 & 14.75 & 2.96 & 3922 fm 126 & 13 51 18 & -28 06 27 & 12.22 & 0.43 & 3921/0 fm 127 & 13 51 10 & -29 20 03 & 11.67 & 0.28 & 3919 fm 128 & 13 49 02 & -28 17 46 & 11.73 & 0.29 & 3909 fm 129 & 13 49 35 & -30 02 06 & 14.11 & 1.11 & ? fm 130 & 13 48 05 & -27 16 56 & 12.45 & 0.46 & 3904 fm 131 & 13 48 25 & -29 06 48 & 12.98 & 0.55 & ? fm 132 & 13 48 49 & -30 12 30 & 13.62 & 0.85 & ? fm 133 & 13 47 38 & -28 55 02 & 14.30 & 1.64 & ? fm 134 & 13 47 11 & -30 46 12 & 12.87 & 0.53 & 3901 fm 135 & 13 45 53 & -28 34 06 & 14.57 & 1.65 & ? fm 136 & 13 44 34 & -26 18 22 & 11.46 & 0.29 & 3889 fm 137 & 13 43 39 & -26 29 55 & 13.11 & 0.71 & ? fm 138 & 13 44 40 & -29 39 34 & 13.69 & 0.88 & ? fm 139 & 13 44 14 & -28 48 53 & 13.39 & 0.42 & 3887 fm 140 & 13 45 17 & -31 22 33 & 13.48 & 0.83 & ? fm 141 & 13 43 21 & -28 55 36 & 13.50 & 0.46 & 3878 fm 142 & 13 43 18 & -29 00 16 & 14.07 & 0.71 & ? fm 143 & 13 42 59 & -29 07 15 & 13.80 & 0.56 & ? fm 144 & 13 42 11 & -28 28 59 & 16.06 & 3.81 & ? fm 145 & 13 41 13 & -27 29 26 & 14.53 & 1.73 & 3869 fm 146 & 13 41 56 & -29 28 49 & 14.33 & 0.80 & ? fm 147 & 13 41 25 & -29 06 22 & 14.45 & 0.89 & ? fm 148 & 13 40 27 & -27 43 12 & 13.69 & 1.00 & ? fm 149 & 13 41 59 & -32 09 29 & 12.59 & 0.51 & ? fm 150 & 13 41 23 & -30 52 59 & 12.78 & 0.44 & ? fm 151 & 13 41 43 & -31 37 13 & 13.47 & 0.45 & ? fm 152 & 13 39 30 & -26 40 11 & 13.04 & 0.60 & 3859 fm 153 & 13 40 59 & -31 14 31 & 13.44 & 0.76 & ? fm 154 & 13 40 15 & -29 53 12 & 14.01 & 1.10 & 3864 fm 155 & 13 40 34 & -32 17 11 & 13.79 & 1.22 & ? fm 156 & 13 39 13 & -29 05 45 & 14.07 & 1.18 & ? fm 157 & 13 39 28 & -29 54 45 & 11.62 & 0.16 & ? fm 158 & 13 36 57 & -25 54 28 & 12.26 & 0.38 & 3841 fm 159 & 13 38 00 & -31 06 14 & 13.22 & 0.43 & ? fm 160 & 13 38 01 & -31 10 40 & 13.15 & 0.36 & ? fm 161 & 13 37 40 & -32 21 50 & 12.50 & 0.47 & 3846 fm 162 & 13 37 11 & -31 29 23 & 13.78 & 0.87 & ? fm 163 & 13 35 50 & -30 51 28 & 13.74 & 0.58 & 3831 fm 164 & 13 33 21 & -27 42 04 & 13.02 & 0.51 & 3817 fm 165 & 13 32 10 & -30 44 25 & 13.86 & 0.84 & ? fm 166 & 13 30 34 & -25 29 15 & 14.32 & 1.11 & ? fm 167 & 13 30 56 & -30 18 29 & 13.11 & 0.50 & 3795 fm 168 & 13 30 41 & -30 47 08 & 12.93 & 0.53 & ? fm 169 & 13 29 00 & -25 36 03 & 12.49 & 0.65 & 3785 fm 170 & 13 29 20 & -28 27 46 & 12.74 & 0.50 & 3787 fm 171 & 13 29 40 & -29 36 24 & 16.34 & 7.80 & ? fm 172 & 13 29 06 & -31 02 01 & 13.73 & 0.90 & ? fm 173 & 13 28 46 & -29 16 29 & 13.41 & 0.73 & 3784 fm 174 & 13 27 48 & -30 46 28 & 13.40 & 0.66 & 3778 fm 175 & 13 27 41 & -31 31 45 & 13.91 & 0.56 & ? fm 176 & 13 27 07 & -30 04 33 & 13.63 & 0.79 & 3776 fm 177 & 13 26 20 & -26 51 42 & 13.01 & 0.56 & ? fm 178 & 13 27 06 & -31 30 15 & 14.23 & 0.77 & ? fm 179 & 13 26 48 & -32 28 51 & 13.27 & 0.59 & ? fm 180 & 13 25 36 & -27 39 21 & 13.28 & 0.66 & ? fm 181 & 13 25 47 & -29 49 49 & 13.84 & 1.01 & 3767 fm 182 & 13 24 57 & -28 15 51 & 14.52 & 1.88 & ? fm 183 & 13 24 02 & -28 24 56 & 13.94 & 1.19 & 3762 fm 184 & 13 23 17 & -27 05 24 & 12.12 & 0.38 & ? fm 185 & 13 20 29 & -29 52 49 & 13.64 & 1.23 & ? fm 186 & 13 19 53 & -27 48 56 & 13.03 & 0.81 & 3738 fm 187 & 13 17 42 & -28 48 03 & 12.37 & 0.54 & 3728 fm 188 & 13 31 21 & -29 33 53 & 13.31 & 0.40 & ? fm 189 & 13 31 36 & -29 30 21 & 12.77 & 0.29 & 3800 fm 190 & 13 26 07 & -32 33 26 & 10.36 & 0.23 & 3770 fm 191 & 13 48 09 & -32 13 29 & 11.17 & 0.32 & 3906 fm 192 & 13 32 05 & -32 11 40 & 12.50 & 0.30 & ? fm 193 & 13 34 47 & -25 18 49 & 11.91 & 0.28 & 3822 fm 194 & 13 41 19 & -28 45 37 & 12.89 & 0.35 & ? fm 195 & 13 35 20 & -31 24 27 & 13.07 & 0.32 & ? fm 196 & 13 40 10 & -31 31 06 & 13.06 & 0.36 & ? fm 197 & 13 36 21 & -29 41 57 & 13.86 & 0.64 & ? ft 1 & 19 55 30 & -49 27 45 & 10.27 & 0.24 & 4536 ft 2 & 19 53 16 & -48 27 36 & 11.25 & 0.37 & 4532 ft 3 & 20 00 09 & -50 09 49 & 10.09 & 0.24 & 4544 ft 4 & 19 54 51 & -48 31 15 & 11.27 & 0.34 & 4534 ft 5 & 19 55 18 & -48 16 52 & 10.60 & 0.27 & 4535 ft 6 & 20 01 44 & -49 58 38 & 10.16 & 0.23 & 4551 ft 7 & 20 04 50 & -49 56 01 & 11.11 & 0.30 & 4560 ft 8 & 20 00 10 & -48 43 38 & 12.18 & 0.54 & 4546 ft 9 & 19 56 08 & -47 23 08 & 9.34 & 0.19 & 4538 ft 10 & 20 04 46 & -49 44 03 & 11.19 & 0.30 & 4559 ft 11 & 20 03 31 & -49 22 26 & 11.28 & 0.30 & 4555 ft 12 & 19 53 15 & -46 22 23 & 9.35 & 0.20 & 4533 ft 13 & 20 00 01 & -48 11 02 & 11.80 & 0.39 & 4545 ft 14 & 20 05 07 & -49 29 09 & 10.78 & 0.27 & 4562 ft 15 & 20 09 35 & -50 25 49 & 10.43 & 0.25 & 4584 ft 16 & 20 05 59 & -49 30 24 & 11.84 & 0.29 & 4567 ft 17 & 20 07 50 & -49 58 27 & 12.45 & 0.58 & 4572 ft 18 & 19 59 41 & -47 33 41 & 12.26 & 0.54 & ? ft 19 & 20 03 54 & -48 37 00 & 13.09 & 0.91 & ? ft 20 & 19 57 02 & -46 35 36 & 10.35 & 0.22 & 4540 ft 21 & 20 06 16 & -48 59 27 & 9.60 & 0.20 & 4568 ft 22 & 20 04 59 & -48 36 25 & 11.36 & 0.32 & 4561 ft 23 & 20 02 55 & -47 58 16 & 12.14 & 0.44 & 4554 ft 24 & 20 01 26 & -47 23 38 & 10.15 & 0.18 & 4550 ft 25 & 20 12 44 & -50 18 27 & 10.99 & 0.28 & 4597 ft 26 & 20 02 07 & -47 18 50 & 8.69 & 0.17 & 4552 ft 27 & 20 13 54 & -50 19 21 & 8.65 & 0.18 & 4602 ft 28 & 20 05 56 & -48 16 17 & 8.92 & 0.18 & 4566 ft 29 & 20 08 57 & -49 3 54 & 12.30 & 0.53 & 4581 ft 30 & 20 12 53 & -49 58 40 & 11.18 & 0.29 & 4599 ft 31 & 20 07 39 & -48 39 25 & 12.06 & 0.46 & 4570 ft 32 & 20 05 35 & -47 56 03 & 10.08 & 0.20 & 4564 ft 33 & 19 58 60 & -46 06 29 & 11.91 & 0.38 & 4542 ft 34 & 20 09 26 & -48 49 48 & 11.56 & 0.34 & 4583 ft 35 & 20 05 13 & -47 29 42 & 9.47 & 0.19 & 4563 ft 36 & 19 58 41 & -45 36 50 & 9.24 & 0.18 & 4541 ft 37 & 20 18 21 & -50 39 31 & 10.59 & 0.25 & 4609 ft 38 & 20 00 53 & -45 42 34 & 8.04 & 0.17 & 4548 ft 39 & 20 11 25 & -48 22 45 & 10.75 & 0.23 & 4593 ft 40 & 20 08 32 & -47 33 54 & 10.96 & 0.25 & 4578 ft 41 & 20 19 11 & -50 09 55 & 12.04 & 0.50 & 4615 ft 42 & 20 08 09 & -47 11 30 & 9.28 & 0.18 & 4575 ft 43 & 20 00 46 & -45 06 44 & 8.33 & 0.17 & 4547 ft 44 & 20 15 37 & -48 58 12 & 11.91 & 0.38 & 4606 ft 45 & 20 07 35 & -46 51 45 & 12.59 & 0.59 & 4571 ft 46 & 20 21 01 & -50 00 27 & 7.51 & 0.17 & 4619 ft 47 & 20 15 11 & -48 44 17 & 12.28 & 0.48 & ? ft 48 & 20 06 25 & -46 05 07 & 9.26 & 0.18 & 4569 ft 49 & 20 18 09 & -48 57 25 & 12.53 & 0.61 & 4608 ft 50 & 20 02 04 & -44 26 49 & 7.11 & 0.16 & 4553 ft 51 & 20 08 37 & -46 19 58 & 11.59 & 0.31 & 4579 ft 52 & 20 05 43 & -45 34 38 & 11.78 & 0.37 & 4565 ft 53 & 20 22 22 & -49 29 13 & 9.17 & 0.18 & 4625 ft 54 & 20 04 29 & -44 57 08 & 10.44 & 0.22 & 4558 ft 55 & 20 23 35 & -49 46 36 & 12.57 & 0.76 & 4631 ft 56 & 20 03 25 & -44 35 11 & 9.59 & 0.19 & 4556 ft 57 & 20 18 34 & -48 32 33 & 11.51 & 0.33 & 4613 ft 58 & 20 10 30 & -46 25 39 & 12.32 & 0.42 & 4590 ft 59 & 20 23 01 & -49 19 55 & 11.28 & 0.24 & 4628 ft 60 & 20 20 06 & -48 28 29 & 9.02 & 0.18 & 4617 ft 61 & 20 03 46 & -43 49 14 & 9.01 & 0.18 & 4557 ft 62 & 20 13 56 & -46 33 51 & 11.56 & 0.30 & 4603 ft 63 & 20 12 25 & -46 10 46 & 10.62 & 0.21 & 4598 ft 64 & 20 07 47 & -44 55 47 & 12.58 & 0.62 & 4574 ft 65 & 20 16 42 & -47 05 24 & 12.12 & 0.40 & ? ft 66 & 20 19 18 & -47 35 05 & 11.32 & 0.31 & 4616 ft 67 & 20 12 23 & -45 51 35 & 11.14 & 0.26 & 4596 ft 68 & 20 09 40 & -44 46 27 & 12.70 & 0.70 & 4586 ft 69 & 20 08 55 & -44 31 16 & 11.83 & 0.43 & 4582 ft 70 & 20 13 37 & -45 36 15 & 8.02 & 0.17 & 4601 ft 71 & 20 21 57 & -47 43 25 & 10.33 & 0.21 & 4622 ft 72 & 20 23 58 & -48 13 09 & 11.38 & 0.31 & 4632 ft 73 & 20 30 13 & -49 21 49 & 8.83 & 0.18 & 4650 ft 74 & 20 09 38 & -44 19 20 & 12.87 & 0.82 & 4587 ft 75 & 20 18 19 & -46 33 12 & 12.04 & 0.38 & 4611 ft 76 & 20 27 12 & -48 34 42 & 11.38 & 0.34 & 4641 ft 77 & 20 10 27 & -44 08 21 & 11.39 & 0.34 & 4589 ft 78 & 20 26 36 & -47 56 53 & 10.29 & 0.23 & 4639 ft 79 & 20 12 03 & -44 16 19 & 9.84 & 0.20 & 4595 ft 80 & 20 10 46 & -43 36 02 & 10.34 & 0.20 & 4591 ft 81 & 20 26 22 & -47 30 36 & 11.94 & 0.43 & 4638 ft 82 & 20 10 60 & -43 28 51 & 9.81 & 0.19 & 4592 ft 83 & 20 22 29 & -46 27 09 & 12.16 & 0.45 & 4626 ft 84 & 20 12 40 & -43 50 20 & 10.53 & 0.23 & 4600 ft 85 & 20 14 11 & -44 10 37 & 12.69 & 0.74 & 4604 ft 86 & 20 27 00 & -47 15 39 & 10.41 & 0.22 & 4640 ft 87 & 20 29 33 & -47 47 07 & 9.56 & 0.19 & 4645 ft 88 & 20 15 34 & -44 28 42 & 11.57 & 0.33 & ? ft 89 & 20 11 59 & -43 24 30 & 11.60 & 0.43 & 4594 ft 90 & 20 20 45 & -45 41 25 & 11.80 & 0.32 & 4620 ft 91 & 20 26 06 & -46 39 46 & 9.10 & 0.18 & 4636 ft 92 & 20 22 00 & -45 39 05 & 8.86 & 0.17 & 4624 ft 93 & 20 14 58 & -43 39 54 & 9.85 & 0.21 & 4605 ft 94 & 20 28 24 & -46 56 53 & 10.68 & 0.24 & 4643 ft 95 & 20 33 45 & -47 56 28 & 11.65 & 0.39 & 4653 ft 96 & 20 26 05 & -46 10 09 & 11.74 & 0.40 & 4637 ft 97 & 20 29 39 & -46 09 44 & 7.00 & 0.16 & 4647 ft 98 & 20 22 40 & -44 35 38 & 11.14 & 0.31 & 4627 ft 99 & 20 33 58 & -46 49 56 & 10.07 & 0.21 & 4654 ft 100 & 20 29 51 & -45 55 25 & 11.55 & 0.40 & 4649 ft 101 & 20 29 33 & -45 23 16 & 10.17 & 0.21 & 4646 ft 102 & 20 27 16 & -44 45 25 & 11.20 & 0.29 & 4642 ft 103 & 20 23 00 & -43 39 26 & 10.14 & 0.20 & 4630 ft 104 & 20 22 46 & -43 29 04 & 9.18 & 0.20 & 4629 ft 105 & 20 35 48 & -46 37 18 & 10.91 & 0.32 & 4655 ft 106 & 20 23 53 & -43 36 56 & 11.09 & 0.27 & 4633 ft 107 & 20 29 25 & -44 57 48 & 12.42 & 0.42 & ? ft 108 & 20 24 20 & -43 40 09 & 11.75 & 0.33 & ? ft 109 & 20 36 52 & -46 26 44 & 9.32 & 0.20 & 4657 ft 110 & 20 32 19 & -45 21 27 & 10.72 & 0.28 & 4652 ft 111 & 20 36 33 & -45 32 42 & 9.22 & 0.20 & 4656 ft 112 & 20 30 33 & -43 58 04 & 10.59 & 0.34 & 4651 ft 113 & 20 01 32 & -50 15 50 & 11.88 & 0.53 & ? ft 114 & 20 02 55 & -50 17 38 & 12.32 & 0.72 & ? ft 115 & 19 56 53 & -48 19 13 & 12.58 & 0.42 & ? ft 116 & 20 01 22 & -49 33 41 & 12.20 & 0.56 & ? ft 117 & 19 54 21 & -47 36 48 & 12.62 & 0.83 & ? ft 118 & 19 55 12 & -47 47 48 & 12.39 & 0.73 & ? ft 119 & 19 57 16 & -48 20 54 & 12.59 & 0.46 & ? ft 120 & 20 03 60 & -49 47 48 & 12.68 & 0.46 & ? ft 121 & 19 56 56 & -47 25 48 & 10.75 & 0.21 & 4539 ft 122 & 19 57 07 & -47 02 54 & 12.84 & 0.90 & ? ft 123 & 19 55 18 & -46 48 39 & 12.23 & 0.61 & ? ft 124 & 20 01 20 & -48 27 16 & 12.70 & 0.76 & ? ft 125 & 19 54 34 & -46 27 26 & 12.33 & 0.58 & ? ft 126 & 20 00 13 & -47 55 55 & 12.78 & 0.81 & ? ft 127 & 19 54 45 & -46 13 16 & 12.49 & 0.69 & ? ft 128 & 20 07 01 & -49 31 30 & 12.57 & 0.44 & ? ft 129 & 20 12 03 & -50 39 49 & 12.48 & 0.81 & ? ft 130 & 19 59 28 & -47 05 44 & 12.37 & 0.55 & 4543 ft 131 & 20 01 07 & -47 04 17 & 12.50 & 0.54 & 4549 ft 132 & 19 58 28 & -46 08 04 & 13.46 & 0.73 & ? ft 133 & 19 57 45 & -45 50 31 & 12.82 & 0.71 & ? ft 134 & 20 13 13 & -49 38 29 & 12.71 & 0.71 & ? ft 135 & 20 09 55 & -48 35 38 & 12.79 & 0.66 & ? ft 136 & 19 58 27 & -45 06 34 & 12.77 & 0.88 & ? ft 137 & 20 02 44 & -45 59 33 & 12.46 & 0.57 & ? ft 138 & 20 16 08 & -48 49 27 & 12.86 & 0.43 & 4607 ft 139 & 20 08 01 & -46 44 14 & 12.99 & 0.48 & ? ft 140 & 20 20 44 & -49 48 58 & 12.47 & 0.43 & ? ft 141 & 20 18 32 & -49 15 10 & 12.36 & 0.50 & ? ft 142 & 20 14 38 & -48 12 26 & 12.78 & 0.68 & ? ft 143 & 20 19 56 & -49 21 59 & 12.71 & 0.72 & ? ft 144 & 20 24 37 & -50 23 17 & 11.27 & 0.34 & ? ft 145 & 20 06 18 & -45 51 36 & 12.34 & 0.50 & ? ft 146 & 20 20 30 & -49 06 44 & 12.77 & 0.66 & 4618 ft 147 & 20 09 01 & -46 07 16 & 14.47 & 1.20 & ? ft 148 & 20 09 38 & -46 09 04 & 12.33 & 0.45 & 4585 ft 149 & 20 09 50 & -46 02 37 & 13.59 & 0.63 & ? ft 150 & 20 05 11 & -44 40 02 & 12.81 & 0.78 & ? ft 151 & 20 04 53 & -44 30 40 & 12.48 & 0.59 & ? ft 152 & 20 10 57 & -45 54 08 & 13.54 & 1.11 & ? ft 153 & 20 10 01 & -45 40 43 & 13.70 & 1.22 & 4588 ft 154 & 20 08 36 & -45 06 48 & 12.34 & 0.53 & ? ft 155 & 20 23 31 & -48 49 47 & 13.26 & 0.90 & ? ft 156 & 20 07 49 & -44 37 54 & 14.62 & 3.42 & ? ft 157 & 20 06 22 & -44 02 51 & 12.27 & 0.56 & ? ft 158 & 20 15 33 & -46 21 51 & 14.23 & 1.98 & ? ft 159 & 20 10 12 & -44 53 29 & 12.29 & 0.53 & ? ft 160 & 20 07 59 & -44 12 04 & 14.41 & 2.52 & 4577 ft 161 & 20 16 55 & -46 31 54 & 13.05 & 0.74 & ? ft 162 & 20 15 15 & -46 04 49 & 12.87 & 0.68 & ? ft 163 & 20 07 59 & -43 33 44 & 12.10 & 0.61 & 4576 ft 164 & 20 27 07 & -48 21 14 & 13.07 & 0.92 & ? ft 165 & 20 25 28 & -47 50 18 & 12.34 & 0.57 & 4635 ft 166 & 20 14 04 & -44 50 33 & 11.69 & 0.34 & ? ft 167 & 20 34 15 & -49 00 19 & 11.84 & 0.49 & ? ft 168 & 20 18 43 & -45 20 15 & 12.63 & 0.56 & 4614 ft 169 & 20 16 43 & -44 42 22 & 12.44 & 0.52 & ? ft 170 & 20 26 44 & -46 27 28 & 13.36 & 1.21 & ? ft 171 & 20 35 41 & -48 22 03 & 11.26 & 0.35 & ? ft 172 & 20 24 51 & -45 43 39 & 12.90 & 0.70 & ? ft 173 & 20 29 19 & -46 28 36 & 12.04 & 0.45 & ? ft 174 & 20 30 17 & -46 28 32 & 11.80 & 0.44 & ? ft 175 & 20 27 48 & -45 41 12 & 11.62 & 0.16 & ? ft 176 & 20 19 11 & -43 32 24 & 12.46 & 0.72 & ? ft 177 & 20 24 44 & -44 38 13 & 14.14 & 2.42 & ? ft 178 & 20 21 37 & -43 47 53 & 12.43 & 0.62 & 4621 ft 179 & 20 35 22 & -46 56 34 & 12.40 & 0.61 & ? ft 180 & 20 25 13 & -44 26 49 & 11.91 & 0.45 & ? ft 181 & 20 27 59 & -44 59 28 & 12.93 & 0.77 & ? ft 182 & 20 29 19 & -45 06 47 & 11.92 & 0.29 & 4644 ft 183 & 20 31 20 & -44 47 35 & 11.89 & 0.53 & ? ft 184 & 20 29 26 & -49 43 37 & 8.74 & 0.17 & ? ft 185 & 20 29 52 & -49 47 23 & 8.52 & 0.17 & 4648 ft 186 & 19 52 07 & -46 50 40 & 10.33 & 0.25 & 4531 ft 187 & 20 15 08 & -43 34 26 & 10.15 & 0.22 & ? ft 188 & 20 05 12 & -47 20 23 & 11.50 & 0.29 & ? ft 189 & 20 24 09 & -43 40 39 & 11.98 & 0.40 & ? fc 1 & usno0525 - 12907188 & b11.1 & u & ? & ? & 12 47 28.6 & -37 22 34 fc 2 & usno0525 - 13025279 & b12.6 & u & ? & ? & 12 55 26.4 & -37 19 18 fc 3 & sao203688 & 7.7 & s & a1v & 0.09 & 12 40 16.0 & -37 51 25 fc 4 & tyc7772 - 851 - 1 & 10.93 & s & ? & 0.39 & 12 56 30.5 & -37 40 10 fc 5 & sao203718 & 9.3 & s & a0 & 0.17 & 12 41 47.0 & -37 53 12 fc 6 & sao203731 & 8.1 & s & f0v & 0.38 & 12 42 32.2 & -37 54 10 fc 7 & tyc777 - 835 - 1 & 9.93 & s & ? & 0.47 & 13 03 39.7 & -37 46 52 fc 8 & usno0450 - 15027418 & b13.7 & u & ? & ? & 12 56 43.5 & -37 58 06 c 9 & usno - a2 - 0450 - 14831165 & b14.7 & u & ? & * -0.7 & 12 52 18.2 & -38 02 35 fc 10 & sao203908 & 9.7 & s & a & 0.31 & 12 53 25.7 & -38 03 32 c 11 & usno - a2 - 0450 - 14517802 & b12.0 & u & ? & * 0.4 & 12 38 01.9 & -38 15 11 fc 12 & usno0450 - 14861545 & b11.2 & u & ? & * 0.5 & 12 49 54.7 & -38 18 16 fc 13 & usno0450 - 14722885 & b14.0 & u & ? & ? & 12 43 53.0 & -38 21 25 fc 14 & sao203822 & 9.2 & s & a3 & 0.23 & 12 48 18.4 & -38 22 04 fc 15 & tyc7758 - 812 - 1 & 9.68 & s & ? & 0.32 & 12 39 35.9 & -38 24 39 fc 16 & sao204099 & 9.0 & s & a0 & 0.06 & 13 04 38.5 & -38 23 40 fc 17 & usno0450 - 14623199 & b11.9 & u & ? & * 0.0 & 12 39 09.8 & -38 44 51 fc 18 & sao203609 & 9.0 & s & a3ii & 0.23 & 12 34 55.5 & -38 49 55 fc 19 & usno0450 - 15096908 & b12.9 & u & ? & * -1.0 & 12 59 39.5 & -38 42 35 fc 20 & sao203816 & 8.9 & s & a3 & 0.24 & 12 47 48.3 & -38 51 45 fc 21 & usno0450 - 14756756 & b12.6 & u & ? & * -0.1 & 12 45 27.0 & -38 51 15 fc 22 & tyc7773 - 1670 - 1 & 10.09 & s & ? & 0.29 & 13 00 33.8 & -38 44 58 fc 23 & sao203979 & 6.8 & s & a3v & 0.13 & 12 58 01.0 & -38 55 02 fc 24 & sao203690 & 9.8 & s & f2 & 0.39 & 12 40 21.9 & -39 02 10 fc 25 & usno - a2 - 0450 - 14821964 & b11.8 & u & ? & * 0.2 & 12 51 54.7 & -38 58 10 fc 26 & sao203562 & 9.6 & s & a0 & 0.17 & 12 31 52.82 & -39 13 57 fc 27 & eso323-g025 & 13.39 & ig & sa(rs)b & ? & 12 52 38.7 & -39 01 42 fc 28 & sao204001 & 6.94 & s & a0v & 0.03 & 12 58 52.8 & -39 06 20 fc 29 & sao204080 & 7.88 & s & f0v & 0.36 & 13 03 36.7 & -38 58 07 fc 30 & usno0450 - 15038765 & b13.7 & u & ? & * -0.3 & 12 57 12.7 & -39 06 54 fc 31 & usno - a2 - 0450 - 15040174 & b14.9 & u & ? & * -0.8 & 13 01 04.0 & -39 03 19 fc 32 & ngc4553 & 13.5 & gp & sa0 & 0.93 & 12 36 08.0 & -39 26 19 fc 33 & sao203981 & 7.53 & s & a9iv & 0.34 & 12 58 03.1 & -39 16 02 fc 34 & usno0450 - 14874256 & b11.9 & u & ? & * 1.1 & 12 50 24.5 & -39 25 37 fc 35 & sao204034 & 8.2 & s & f2 & 0.41 & 13 00 44.4 & -39 17 43 fc 36 & sao20398 & 10.0 & s & a0 & 0.26 & 12 58 15.4 & -39 22 59 fc 37 & tyc7776 - 1116 - 1 & 10.95 & s & ? & 0.32 & 12 53 53.2 & -39 28 38 fc 38 & usno0450 - 14627245 & b13.6 & u & ? & * 0.5 & 12 39 21.4 & -39 41 04 fc 39 & sao203881 & 5.97 & s & b8v & -0.1 & 12 51 57.0 & -39 40 50 fc 40 & sao203835 & 8.1 & s & b5iv & -0.09 & 12 49 10.1 & -39 43 02 fc 41 & sao203621 & 5.78 & s & a0v & 0.00 & 12 36 01.1 & -39 52 11 fc 42 & tyc7775 - 798 - 1 & 10.73 & s & ? & 0.03 & 12 44 13.6 & -39 46 17 fc 43 & sao203681 & 4.63 & s & b8ii & -0.08 & 12 39 52.6 & -39 59 15 fc 44 & usno - a2 - 0450 - 14607200 & 9.9 & g & sabcd & * -1.3 & 12 42 16.2 & -40 06 25 fc 45 & sao203746 & 6.44 & s & a7iii & 0.24 & 12 43 26.3 & -40 10 41 fc 46 & sao203907 & 4.28 & s & a4iv & 0.22 & 12 53 26.2 & -40 10 44 fc 47 & sao203554 & 8.6 & s & a6iii & 0.29 & 12 31 29.1 & -40 16 28 fc 48 & usno - a2 - 0450 - 14768318 & b14.8 & u & ? & * -0.1 & 12 49 36.4 & -40 13 56 fc 49 & tyc7777 - 1416 - 1 & 10.37 & s & ? & 0.35 & 13 03 36.3 & -40 15 14 fc 50 & sao223587 & 8.9 & s & a5 & 0.31 & 12 40 22.0 & -40 37 52 fc 51 & sao223687 & 9.8 & s & a2 & 0.07 & 12 49 15.1 & -40 37 59 fc 52 & sao223519 & 8.0 & s & a5 & 0.17 & 12 34 13.8 & -40 49 44 fc 53 & usno0450 - 14828057 & b12.2 & u & ? & ? & 12 48 35.9 & -40 44 36 fc 54 & tyc7776 - 1480 - 1 & 10.09 & s & ? & 0.25 & 12 59 13.0 & -40 38 03 fc 55 & usno0450 - 14907321 & b13.6 & u & ? & * -0.4 & 12 51 42.7 & -40 44 09 fc 56 & sao223542 & 5.12 & s & a7iii & 0.22 & 12 35 45.5 & -41 01 19 fc 57 & tyc7775 - 757 - 1 & 10.34 & s & ? & 0.15 & 12 46 55.7 & -40 53 24 fc 58 & sao223879 & 7.93 & s & f6v & 0.45 & 13 04 45.2 & -40 40 58 fc 59 & sao223727 & 8.3 & s & a9iv & 0.28 & 12 52 34.3 & -40 56 28 fc 60 & sao223824 & 8.0 & s & a0 & 0.19 & 13 00 01.1 & -40 53 01 fc 61 & pgc042966 & 15.17 & ig & e & ? & 12 45 03.8 & -41 00 47 fc 62 & sao223916 & 10.26 & s & a3 & 0.28 & 13 07 38.1 & -40 46 30 fc 63 & tyc7762 - 215 - 1 & 11.11 & s & ? & 0.21 & 12 38 19.8 & -41 12 41 fc 64 & tyc7775 - 1763 - 1 & 10.63 & s & ? & 0.16 & 12 44 45.2 & -41 13 38 fc 65 & sao223706 & 8.47 & s & a2 & 0.21 & 12 51 03.9 & -41 15 05 fc 66 & sao223511 & 6.7 & s & a7v & 0.24 & 12 33 28.8 & -41 24 51 fc 67 & tyc7780 - 107 - 1 & 9.95 & s & ? & -0.07 & 12 52 19.4 & -41 19 38 fc 68 & tyc7778 - 1705 - 1 & 10.98 & s & ? & 0.18 & 13 10 27.0 & -40 57 24 fc 69 & sao223484 & 7.7 & s & a5iv & 0.32 & 12 31 18.8 & -41 30 17 fc 70 & sao223859 & 9.0 & s & a3 & 0.29 & 13 02 43.2 & -41 14 01 fc 71 & ngc4683 & 13.8 & ig & sb(s)0 & 0.93 & 12 47 42.6 & -41 31 44 fc 72 & sao223823 & 7.4 & s & f2 & 0.37 & 12 59 59.0 & -41 23 27 fc 73 & usno0450 - 14710416 & b14.1 & u & ? & * 0.5 & 12 43 18.5 & -41 37 04 fc 74 & sao223924 & 8.1 & s & a3v & 0.09 & 13 08 12.4 & -41 23 41 fc 75 & ngc4930 & 12.0 & g & sb(rs ) & 0.76 & 13 04 04.7 & -41 24 44 fc 76 & sao223701 & 8.8 & s & f3v & 0.43 & 12 50 31.7 & -41 41 47 fc 77 & sao223808 & 9.5 & s & a0 & 0.01 & 12 59 16.9 & -41 36 34 fc 78 & sao223577 & 9.11 & s & b9 & 0.11 & 12 39 25.2 & -41 51 03 fc 79 & tyc7781 - 616 - 1 & 10.04 & s & ? & 0.27 & 13 04 59.3 & -41 31 54 fc 80 & sao223513 & 7.95 & s & f2 & 0.37 & 12 33 40.8 & -41 57 04 fc 81 & sao223475 & 8.53 & s & ? & 0.29 & 12 30 11.6 & -41 55 28 fc 82 & sao223700 & 8.93 & s & a3 & 0.22 & 12 50 30.4 & -41 54 24 fc 83 & sao223558 & 8.72 & s & a5 & 0.38 & 12 37 21.7 & -42 01 42 fc 84 & sao223864 & 9.2 & s & a2 & 0.19 & 13 03 03.1 & -41 47 21 fc 85 & sao223778 & 8.2 & s & f3v & 0.44 & 12 56 50.2 & -41 54 50 fc 86 & sao223635 & 8.04 & s & f5v & 0.49 & 12 44 20.1 & -42 04 48 fc 87 & usno - a2 - 0450 - 15041646 & b11.3 & u & ? & * -0.1 & 13 01 07.7 & -41 58 55 fc 88 & tyc7780 - 1433 - 1 & 10.38 & s & ? & -0.03 & 12 56 14.2 & -42 09 11 fc 89 & sao223545 & 7.98 & s & sdf2 & 0.45 & 12 36 05.3 & -42 18 35 fc 90 & usno0450 - 15063833 & b11.6 & u & ? & * 0.7 & 12 58 16.0 & -42 03 15 fc 91 & tyc7781 - 83 - 1 & 9.85 & s & ? & 0.10 & 13 07 30.9 & -42 02 40 fc 92 & sao223756 & 6.82 & s & b5vn & -0.13 & 12 55 07.0 & -42 17 28 fc 93 & sao223580 & 9.51 & s & a3 m & 0.32 & 12 39 40.6 & -42 22 29 fc 94 & sao223946 & 8.95 & s & a0 & 0.10 & 13 10 03.7 & -42 13 16 fc 95 & tyc7766 - 1334 - 1 & 10.03 & s & ? & 0.14 & 12 37 25.3 & -42 36 07 fc 96 & sao223960 & 5.79 & s & f7iv & 0.51 & 13 11 08.9 & -42 13 59 fc 97 & sao223828 & 7.83 & s & f3iv & 0.43 & 13 00 17.8 & -42 30 04 fc 98 & tyc7781 - 907 - 1 & 10.52 & s & ? & 0.37 & 13 01 29.9 & -42 29 58 fc 99 & tyc7781 - 802 - 1 & 10.71 & s & ? & 0.35 & 13 02 11.6 & -42 29 13 fc 100 & sao223578 & 9.0 & s & f0v & 0.4 & 12 39 31.9 & -42 59 04 fc 101 & tyc7780 - 609 - 1 & 10.42 & s & ? & 0.15 & 12 52 08.7 & -42 55 26 fc 102 & sao223561 & 9.19 & s & f2 & 0.41 & 12 38 00.2 & -43 02 02 fc 103 & tyc7779 - 1941 - 1 & 10.56 & s & ? & -0.13 & 12 45 59.1 & -43 05 22 fc 104 & tyc7766 - 345 - 1 & 10.75 & s & ? & 0.57 & 12 37 52.3 & -43 05 24 fc 105 & sao223737 & 7.29 & s & a5v & 0.25 & 12 53 33.5 & -43 04 30 fc 106 & usno0450 - 14908873 & b12.8 & u & ? & * -1.2 & 12 51 46.9 & -43 10 44 fc 107 & sao223691 & 9.27 & s & f2 & 0.40 & 12 49 31.4 & -43 12 44 fc 108 & sao223678 & 8.2 & s & f0iii & 0.37 & 12 48 22.4 & -43 19 59 fc 109 & sao223605 & 9.38 & s & a2 & 0.08 & 12 41 30.5 & -43 38 43 fc 110 & tyc7785 - 239 - 1 & 10.89 & s & ? & 0.05 & 13 01 30.2 & -43 22 53 fc 111 & sao223857 & 7.9 & s & a7v & 0.26 & 13 02 38.0 & -43 21 18 fc 112 & sao223681 & 8.82 & s & b9iii & -0.03 & 12 48 47.2 & -43 42 56 fc 113 & tyc7785 - 2089 - 1 & 10.5 & s & ? & -0.10 & 13 08 46.2 & -43 27 52 fc 114 & usno0450 - 14550141 & b13.0 & u & ? & * -0.3 & 12 35 42.8 & -43 49 17 fc 115 & sao223609 & 10.01 & s & a2 & 0.32 & 12 41 58.9 & -43 53 52 fc 116 & tyc7784 - 810 - 1 & 10.93 & s & ? & 0.29 & 12 56 15.1 & -43 46 45 fc 117 & sao223683 & 7.8 & s & a3iii & 0.17 & 12 49 04.0 & -43 56 58 fc 118 & sao223836 & 9.67 & s & a0 & 0.16 & 13 00 53.3 & -43 47 45 fc 119 & usno0450 - 14711408 & b12.4 & u & ? & * 0.3 & 12 43 21.2 & -43 58 34 fc 120 & usno0450 - 14699795 & b12.0 & u & ? & * 0.2 & 12 42 47.8 & -44 06 09 fc 121 & sao223809 & 9.61 & s & a0 & 0.23 & 12 59 19.5 & -43 59 32 fc 122 & sao223880 & 8.79 & s & a2 & 0.25 & 13 04 48.3 & -43 55 29 fc 123 & sao223753 & 5.88 & s & g2iv & 0.63 & 12 54 58.6 & -44 09 07 fc 124 & tyc7783 - 1498 - 1 & 10.49 & s & ? & 0.40 & 12 41 17.9 & -44 18 10 fc 125 & sao223896 & 9.6 & s & a2 & 0.07 & 13 05 58.2 & -44 05 44 fc 126 & sao223742 & 9.1 & s & f0v & 0.38 & 12 54 15.6 & -44 24 06 fc 127 & tyc7783 - 1223 - 1 & 11.36 & s & ? & -0.07 & 12 43 02.4 & -44 34 25 fc 128 & sao223677 & 6.88 & s & b9v & -0.02 & 12 48 16.5 & -44 43 08 fc 129 & sao223598 & 10.28 & s & a2 & 0.21 & 12 41 11.8 & -44 42 26 fc 130 & sao223899 & 9.38 & s & a2iv & 0.18 & 13 06 08.8 & -44 25 14 fc 131 & sao223646 & 9.42 & s & a2 & 0.26 & 12 45 26.3 & -44 44 06 fc 132 & sao223866 & 7.8 & s & a2iv & 0.13 & 13 03 17.9 & -44 33 46 fc 133 & sao223842 & 7.66 & s & b6v & -0.08 & 13 01 15.2 & -44 45 07 fc 134 & sao223769 & 9.44 & s & a0 & 0.31 & 12 56 24.0 & -44 52 07 fc 135 & tyc7785 - 1568 - 1 & 10.88 & s & ? & 0.02 & 13 04 20.5 & -44 59 37 fc 136 & sao223740 & 9.0 & s & a9v & 0.34 & 12 53 54.3 & -45 06 33 fc 137 & sao223846 & 9.74 & s & b9 & 0.11 & 13 01 30.1 & -45 11 23 fc 138 & sao223679 & 8.4 & s & f0 & 0.33 & 12 48 33.8 & -45 25 11 fc 139 & sao223655 & 8.31 & s & f0 & 0.27 & 12 46 03.3 & -45 30 05 fc 140 & sao223722 & 9.33 & s & a2 & 0.03 & 12 52 18.7 & -45 29 29 fc 141 & tyc7259 - 115 - 1 & 9.94 & s & ? & 0.311 & 12 44 52.9 & -37 18 48 fc 142 & usno0450 - 14941695 & b12.3 & u & ? & * 0.4 & 12 53 07.4 & -37 31 45 fc 143 & usno0450 - 14747828 & b12.3 & u & ? & * 1.4 & 12 45 02.2 & -37 43 30 fc 144 & usno0450 - 14885379 & b11.3 & u & ? & * 0.7 & 12 50 49.6 & -37 49 39 fc 145 & usno0450 - 15121279 & b12.4 & u & ? & * 0.3 & 13 00 40.9 & -37 49 05 fc 146 & usno0450 - 14481710 & b13.7 & u & ? & * 0.4 & 12 32 26.5 & -38 32 45 fc 147 & usno0450 - 14485189 & b10.7 & u & ? & * -0.1 & 12 32 36.5 & -38 39 11 fc 148 & usno0450 - 14870154 & b13.2 & u & ? & * 0.0 & 12 50 14.8 & -38 29 34 fc 149 & usno0450 - 14551298 & b12.6 & u & ? & * 1.0 & 12 35 46.2 & -38 39 54 fc 150 & usno0450 - 15008287 & b12.8 & u & ? & * -0.5 & 12 55 55.5 & -38 39 59 fc 151 & usno0450 - 14951320 & b11.6 & u & ? & * 0.9 & 12 53 31.5 & -38 47 16 fc 152 & usno - a2 - 0450 - 14652712 & b15.9 & u & ? & * -0.2 & 12 44 24.5 & -38 56 15 fc 153 & usno0450 - 14854637 & b15.7 & u & ? & * -0.1 & 12 49 38.7 & -39 02 20 fc 154 & tyc7771 - 925 - 1 & 10.43 & s & ? & 0.42 & 12 44 37.7 & -39 05 51 fc 155 & usno0450 - 14643933 & b11.6 & u & ? & * 1.4 & 12 40 08.3 & -39 23 40 fc 156 & tyc7775 - 674 - 1 & 9.99 & s & ? & 0.46 & 12 41 33.8 & -39 27 38 fc 157 & usno0450 - 15104855 & b12.6 & u & ? & * 0.5 & 12 59 59.8 & -39 14 56 fc 158 & tyc7772 - 1576 - 1 & 9.81 & s & ? & 0.36 & 12 54 20.1 & -39 19 32 fc 159 & usno0450 - 14709560 & b11.5 & u & ? & * 1.0 & 12 43 16.0 & -39 32 22 fc 160 & sao203807 & 8.81 & s & g3v & 0.63 & 12 47 23.2 & -39 28 54 fc 161 & ngc4679 & 13.15 & g & sa(s)bc & ? & 12 47 30.2 & -39 34 21 fc 162 & tyc7762 - 1018 - 1 & 11.20 & s & ? & 0.30 & 12 30 31.9 & -39 39 53 fc 163 & tyc7775 - 54 - 1 & 10.99 & s & ? & 0.53 & 12 45 11.9 & -39 38 34 fc 164 & tyc7762 - 206 - 1 & 10.03 & s & ? & 1.03 & 12 33 31.5 & -39 49 01 fc 165 & usno - a2 - 0450 - 14720843 & b16.9 & u & ? & * 1.7 & 12 47 33.0 & -39 46 46 fc 166 & usno0450 - 14850792 & b13.1 & u & ? & * 0.4 & 12 49 30.1 & -39 53 31 fc 167 & usno0450 - 14528258 & b12.2 & u & ? & * 1.6 & 12 34 40.1 & -40 03 17 fc 168 & sao203806 & 8.4 & s & m3iii & 1.58 & 12 47 19.4 & -40 07 377 fc 169 & sao204171 & 8.06 & s & f5v & 0.48 & 13 09 10.5 & -39 52 24 fc 170 & sao203541 & 8.8 & s & a2 & 0.41 & 12 30 27.9 & -40 16 06 fc 171 & sao223652 & 9.1 & s & f8 & 0.56 & 12 45 42.5 & -40 16 41 fc 172 & tyc7775 - 1831 - 1 & 10.14 & s & ? & 0.44 & 12 47 49.7 & -40 22 53 fc 173 & tyc7762 - 1543 - 1 & 10.56 & s & ? & 0.35 & 12 38 21.2 & -40 30 58 fc 174 & usno0450 - 15031167 & b14.0 & u & ? & * 0.0 & 12 56 53.2 & -40 21 00 fc 175 & tyc7776 - 987 - 1 & 10.96 & s & ? & 0.14 & 12 59 25.0 & -40 25 53 fc 176 & usno0450 - 15063491 & b13.3 & u & ? & * 0.4 & 12 58 15.1 & -40 36 25 fc 177 & usno0450 - 14575914 & b13.7 & u & ? & * 0.5 & 12 36 55.9 & -40 52 01 fc 178 & usno - a2 - 0450 - 15059539 & b12.2 & u & ? & * 0.0 & 13 01 52.8 & -40 37 45 fc 179 & usno0450 - 15215568 & b13.5 & u & ? & * -0.7 & 13 04 37.6 & -40 32 15 fc 180 & tyc7777 - 392 - 1 & 10.5 & s & ? & 0.81 & 13 07 40.9 & -40 30 02 fc 181 & tyc7778 - 1639 - 1 & 9.36 & s & ? & 0.33 & 13 11 12.2 & -40 28 26 fc 182 & tyc7762 - 1945 - 1 & 10.18 & s & ? & 0.56 & 12 38 52.6 & -41 00 55 fc 183 & tyc7762 - 1926 - 1 & 10.57 & s & ? & 0.24 & 12 33 49.9 & -40 59 52 fc 184 & sao223494 & 7.3 & s & k2iii & 1.24 & 12 31 59.2 & -41 03 19 fc 185 & usno0450 - 14860017 & b11.4 & u & ? & * 0.8 & 12 49 50.9 & -41 04 23 fc 186 & eso323-g054 & 14.86 & ig & s & ? & 12 59 18.2 & -40 57 41 fc 187 & usno - a2 - 0450 - 14757081 & b12.4 & u & ? & * 0.0 & 12 49 08.7 & -41 12 26 fc 188 & usno0450 - 15248516 & b13.2 & u & ? & * -0.4 & 13 05 59.3 & -40 57 58 fc 189 & usno0450 - 14676419 & b14.7 & u & ? & * 1.4 & 12 41 41.4 & -41 16 59 fc 190 & tyc7766 - 717 - 1 & 10.64 & s & ? & 0.37 & 12 38 18.1 & -41 24 34 fc 191 & sao223465 & 8.03 & s & f3iv & 0.39 & 12 29 48.4 & -41 24 09 fc 192 & tyc7776 - 1098 - 1 & 10.55 & s & ? & 0.39 & 12 58 49.4 & -41 09 07 fc 193 & usno - a2 - 0450 - 14865062 & 16.4 & u & ? & * -0.4 & 12 53 42.8 & -41 18 29 fc 194 & sao223501 & 9.38 & s & f0 & 0.26 & 12 32 49.5 & -41 37 58 fc 195 & usno - a2 - 0450 - 15278266 & b11.8 & u & ? & * -0.2 & 13 10 52.4 & -41 18 04 fc 196 & tyc7779 - 1219 - 1 & 11.05 & s & ? & 0.43 & 12 45 45.7 & -41 49 32 fc 197 & usno - a2 - 0450 - 14825935 & b14.5 & u & ? & * 1.0 & 12 52 04.9 & -41 52 03 fc 198 & usno0450 - 14926401 & b14.6 & u & ? & * 0.7 & 12 52 30.3 & -41 55 28 fc 199 & usno0450 - 14931221 & b14.7 & u & ? & * 0.4 & 12 52 41.9 & -41 57 41 fc 200 & usno0450 - 14945366 & b15.0 & u & ? & * 1.0 & 12 53 16.5 & -42 00 40 fc 201 & tyc7779 - 733 - 1 & 10.93 & s & ? & 0.32 & 12 42 37.3 & -42 10 13 fc 202 & tyc7781 - 114 - 1 & 11.19 & s & ? & 0.08 & 13 08 12.6 & -41 57 12 fc 203 & usno0450 - 14682772 & b14.3 & u & ? & ? & 12 41 59.7 & -42 18 16 fc 204 & usno0450 - 15142256 & b14.3 & u & ? & * 0.5 & 13 01 34.0 & -42 07 48 fc 205 & usno0450 - 15213947 & b12.1 & u & ? & ? & 13 04 33.6 & -42 08 21 fc 206 & usno0450 - 14842856 & b13.2 & u & ? & * 0.6 & 12 49 11.6 & -42 26 10 fc 207 & tyc7779 - 1570 - 1 & 10.28 & s & ? & 0.52 & 12 45 58.5 & -42 31 52 fc 208 & tyc7766 - 1132 - 1 & 10.51 & s & ? & 0.37 & 12 38 58.0 & -42 34 53 fc 209 & usno - a2 - 0450 - 14919623 & b13.1 & u & ? & * 0.5 & 12 52 13.9 & -42 31 01 fc 210 & sao223535 & 8.95 & s & f2 & 0.36 & 12 35 13.1 & -42 40 21 fc 211 & sao223684 & 7.85 & s & f5iv & 0.54 & 12 49 07.0 & -42 38 03 fc 212 & tyc7780 - 747 - 1 & 10.7 & s & ? & 0.45 & 12 51 24.6 & -42 44 48 fc 213 & sao223819 & 8.8 & s & g5v & 0.71 & 12 59 52.0 & -42 33 41 fc 214 & usno - a2 - 0450 - 14647502 & 9.5 & u & ? & * 0.4 & 12 44 18.5 & -42 47 25 fc 215 & usno0450 - 15318910 & b11.7 & u & ? & ? & 13 08 52.4 & -42 27 49 fc 216 & tyc7779 - 2014 - 1 & 9.82 & s & ? & 0.37 & 12 47 55.1 & -42 50 35 fc 217 & tyc7779 - 2027 - 1 & 9.89 & s & ? & 0.31 & 12 44 10.1 & -43 04 30 fc 218 & sao223942 & 9.56 & s & ? & 0.37 & 13 09 44.9 & -42 43 14 fc 219 & sao223596 & 9.32 & s & f5 & 0.43 & 12 41 10.5 & -43 16 59 fc 220 & usno - a2 - 0450 - 15116369 & b14.5 & u & ? & * 1.5 & 13 04 15.1 & -43 02 04 fc 221 & usno0450 - 15150211 & b12.4 & u & ? & * 0.1 & 13 01 54.1 & -43 03 39 fc 222 & sao223571 & 9.35 & s & a0 m & 0.37 & 12 39 01.6 & -43 25 26 fc 223 & usno0450 - 15209781 & b11.6 & u & ? & ? & 13 04 23.1 & -43 08 30 fc 224 & tyc7783 - 60 - 1 & 9.81 & s & ? & 0.73 & 12 44 25.1 & -43 22 45 fc 225 & tyc7784 - 1348 - 1 & 10.77 & s & ? & 0.24 & 12 57 53.9 & -43 18 20 fc 226 & usno0450 - 15036393 & b12.2 & u & ? & * 0.4 & 12 57 06.7 & -43 20 59 fc 227 & usno0450 - 15075910 & b11.6 & u & ? & * 0.8 & 12 58 46.7 & -43 27 41 fc 228 & tyc7785 - 188 - 1 & 10.26 & s & ? & 0.27 & 13 00 36.8 & -43 27 17 fc 229 & eso269-ig056 & 14.01 & ig & sb(s)mp & ? & 13 10 01.2 & -43 12 46 fc 230 & sao223878 & 9.89 & s & a0 & 0.33 & 13 04 28.6 & -43 23 56 fc 231 & usno0450 - 15171172 & b12.6 & u & ? & * 1.3 & 13 02 46.7 & -43 32 10 fc 232 & usno0450 - 15282695 & b12.6 & u & ? & * 0.5 & 13 07 23.3 & -43 36 50 fc 233 & usno0450 - 15158550 & b12.9 & u & ? & * -1.2 & 13 02 14.6 & -43 39 19 fc 234 & tyc7785 - 367 - 1 & 10.54 & s & ? & 0.40 & 13 08 18.1 & -43 35 18 fc 235 & eso269-ig022/3 & 14.7 & ip & s0 & * -1.3 & 12 58 50.6 & -43 52 35 fc 236 & tyc7784 - 485 - 1 & 9.71 & s & ? & 0.54 & 12 51 01.0 & -44 11 32 fc 237 & usno0450 - 15075030 & b14.7 & u & ? & * 0.1 & 12 58 44.4 & -44 19 32 fc 238 & pgc044774 & 16.0 & gp & sab(s)p & ? & 13 00 31.1 & -44 14 39 fc 239 & usno0450 - 14840957 & b12.0 & u & ? & * 1.4 & 12 48 46.7 & -44 27 22 fc 240 & usno0450 - 15201324 & b13.5 & u & ? & * 0.1 & 13 04 01.7 & -44 19 42 fc 241 & usno0450 - 15199021 & b13.3 & u & ? & * -0.3 & 13 03 56.1 & -44 22 31 fc 242 & usno0450 - 15100880 & b12.7 & u & ? & * 1.1 & 12 59 49.7 & -44 27 49 fc 243 & sao223704 & 9.61 & s & f2 & 0.36 & 12 51 02.6 & -44 53 04 fc 244 & usno0375 - 17118731 & b12.9 & u & ? & * 0.8 & 12 57 28.5 & -45 05 36 fc 245 & tyc8246 - 51 - 1 & 10.82 & s & ? & 0.34 & 12 57 29.5 & -45 25 06 fc 246 & tyc7776 - 1164 - 1 & 9.57 & s & ? & 0.28 & 12 58 58.9 & -40 27 40 fc 247 & sao223716 & 9.52 & s & a0v & 0.11 & 12 51 45.8 & -41 18 27 fc 248 & sao223728 & 8.5 & s & f3v & 0.43 & 12 52 44.7 & -41 11 04 fc 249 & sao223721 & 9.45 & s & a5 & 0.32 & 12 52 16.0 & -43 02 14 fd 1 & sao233280 & 6.92 & s & a4v & 0.59 & 3 48 24.9 & -54 17 37 fd 2 & sao233313 & 8.69 & s & a0 & 0.08 & 3 53 01.3 & -54 06 231 fd 3 & sao233314 & 7.75 & s & b6v&-0.13 & 3 53 22.8 & -54 51 08 fd 4 & sao233384 & 7.52 & s & f0iv & 0.34 & 4 03 40.6 & -52 55 44 fd 5 & tyc8503 - 860 - 1 & 11.15 & s & a & 0.17 & 3 48 14.0 & -56 41 13 fd 6 & sao233351 & 7.81 & s & f5v & 0.42 & 3 59 54.3 & -54 09 41 fd 7 & tyc8504 - 1517 - 1 & 10.42 & s & ? & 0.11 & 3 57 13.7 & -55 05 25 fd 8 & sao233363 & 7.68 & s & a3v & 0.14 & 4 00 59.1 & -54 23 31 fd 9 & sao233306 & 9.25 & s & a0 & 0.07 & 3 51 44.2 & -56 39 06 fd 10 & usno0300 - 01218129 & 13.19 & s & b a & 0.45 & 4 03 21.8 & -54 39 54 fd 11 & sao233334 & 8.66 & s & f0iii & 0.33 & 3 56 34.5 & -56 19 30 fd 12 & sao233341 & 8.36 & s & a9iii & 0.28 & 3 57 45.5 & -56 15 01 fd 13 & hic18969 & 7.66 & s & a3v & 0.17 & 4 03 53.0 & -55 24 00 fd 14 & sao233451 & 9.51 & s & b9&-0.03 & 4 15 18.3 & -52 09 48 fd 15 & hic18656 & 6.84 & s & a2 m & 0.24 & 3 59 37.5 & -56 28 22 fd 16 & sao233423 & 9.49 & s & f8 & 0.53 & 4 10 31.0 & -53 35 06 fd 17 & sao233368 & 9.20 & s & a0 & 0.06 & 4 01 15.4 & -56 19 34 fd 18 & hic18592 & 6.05 & s & f2iv & 0.45 & 3 58 42.9 & -57 06 09 fd 19 & hic19543 & 8.55 & s & f3v & 0.42 & 4 11 08.9 & -54 02 22 fd 20 & gsc0850201085 & 14.78 & s & f & ? & 4 11 31.6 & -53 49 57 fd 21 & usno0300 - 01267605 & b13.7 & s & a & * 0.6 & 4 13 44.9 & -53 26 19 fd 22 & he0365 - 5827 & b13.54 ? & s & sdo & * 1.8 & 3 55 53.6 & -58 18 23 fd 23 & hic20109 & 6.09 & s & a8v & 0.48 & 4 18 40.0 & -52 51 36 fd 24 & usno0300 - 01253020 & 11.87 & s & a & * 0.9 & 4 09 25.4 & -55 44 27 fd 25 & hic18878 & 8.45 & s & f0iv & 0.28 & 4 02 49.7 & -57 27 34 fd 26 & sao233400 & 8.78 & s & a5 & 0.31 & 4 06 37.0 & -56 34 32 fd 27 & sao233460 & 9.63 & s & f2 & 0.28 & 4 16 10.1 & -54 00 53 fd 28 & sao233392 & 9.64 & s & a2 & 0.29 & 4 04 32.8 & -57 09 35 fd 29 & gsc0850200572 & 14.80 & s & sdob & ? & 4 15 30.3 & -54 21 58 fd 30 & he0418 - 5326 & 15.76 & s&wfd a & ? & 4 19 24.8 & -53 19 18 fd 31 & hic18731 & 7.04 & s & a2v & 0.11 & 4 00 44.9 & -58 39 42 fd 32 & hic20117 & 7.17 & s & a1v & 0.11 & 4 18 48.2 & -54 08 08 fd 33 & iue - bpfm 1 7731 & 15.54 & s & wfd a & * -1.2 & 4 19 10.1 & -53 50 43 fd 34 & usno0300 - 01307552 & 14.28 & s & sdo+cool & * 0.3 & 4 20 02.7&-54 04 14 fd 35 & iue - wd0416 - 55 & 15.49 & s & wd & * -1.4 & 4 17 11.4 & -54 57 47 fd 36 & tyc8507 - 189 - 1 + 2 & 9.40 & s & f5iv & 0.47 & 4 04 50.3 & -58 10 40 fd 37 & sao233500 & 9.85 & s & a2 & 0.24 & 4 22 59.4 & -53 32 06 fd 38 & hic19086 & 7.97 & s & a4v & 0.16 & 4 05 31.3 & -58 28 42 fd 39 & hic19759 & 7.55 & s & f5v & 0.45 & 4 14 11.3 & -56 10 36 fd 40 & hic20650 & 6.95 & s & f0v & 0.35 & 4 25 29.8 & -53 06 43 fd 41 & ngc1566 & 10.33 & g & sab(rs ) & 0.56 & 4 20 00.6 & -54 56 17 fd 42 & usno0300 - 01320497 & 13.87 & s & sdo & * -0.8 & 4 22 37.3 & -54 08 49 fd 43 & hic19353 & 9.40 & s & a1v & 0.07 & 4 08 45.9 & -58 29 14 fd 44 & usno0300 - 01325652 & 15.06 & s & a & * -0.4 & 4 23 36.7 & -54 17 17 fd 45 & hic20799 & 8.06 & s & b ? & 0.36 & 4 27 24.2 & -53 24 16 fd 46 & usno0300 - 01351412 & 15.13 & s & sdb & * -0.9 & 4 28 31.0 & -53 27 53 fd 47 & usno0300 - 01295274 & 12.51 & s & a & * -0.4 & 4 17 35.3 & -56 43 36 fd 48 & hic19726 & 7.32 & s & a5iv & 0.22 & 4 13 39.9 & -58 01 19 fd 49 & usno0300 - 01357213 & 13.04 & s & a & * 0.0 & 4 29 35.4 & -53 13 13 fd 50 & usno0300 - 01265933 & b15.7 & u & ? & * -0.8 & 4 11 45.7 & -58 41 06 fd 51 & tyc8505 - 1427 - 1 & 10.62 & s & b - a & 0.19 & 4 25 53.9 & -55 06 11 fd 52 & usno0300 - 01365856 & b12.8 & s & b(l)-a & * -0.5 & 4 31 11.0 & -53 35 27 fd 54 & sao233498 & 9.71 & s & a2 & 0.25 & 4 22 06.0 & -56 58 59 fd 55 & sao233493 & 9.40 & s & f2 & 0.37 & 4 21 23.9 & -57 32 30 fd 56 & sao233526 & 9.80 & s & a0 & 0.01 & 4 27 04.8 & -56 04 24 fd 57 & hic20543 & 7.53 & s & a2v & 0.01 & 4 24 06.9 & -57 15 11 fd 58 & hic19921 & 4.44 & s & k2iv & 1.08 & 4 16 28.9 & -59 18 07 fd 59 & sao233524 & 9.10 & s & f0 & 0.31 & 4 26 33.0 & -56 26 41 fd 63 & hic21281 & 3.30 & s & a0v & -0.08 & 4 33 59.8 & -55 02 42 fd 64 & sao233547 & 9.70 & s & a0 & 0.01 & 4 31 10.2 & -55 54 43 fd 65 & hic21200 & 8.17 & s & f2v & 0.38 & 4 32 43.2 & -55 35 49 fd 66 & hic21460 & 8.50 & s & ap & 0.28 & 4 36 30.8 & -54 37 16 fd 67 & usno0300 - 01340927 & b13.6 & u & ? & * -0.4 & 4 26 29.3 & -59 06 10 fd 68 & tyc8512 - 2014 - 1 & 9.55 & s & f & 0.29 & 4 34 48.1 & -56 35 30 fd 69 & usno0300 - 01391935 & b14.4 & u & ? & * 0.4 & 4 35 59.8 & -56 16 50 fd 70 & hic21787 & 7.75 & s & a8iv & 0.34 & 4 40 48.9 & -56 01 52 fd 71 & sao233619 & 9.35 & s & a & 0.28 & 4 40 18.5 & -56 11 34 fd 75 & tyc8512 - 2052 - 1 & 10.17 & s & ? & 0.16 & 4 42 36.9 & -56 41 35 fd 76 & hic21645 & 7.20 & s & a4v & 0.19 & 4 38 54.1 & -58 58 50 fd 77 & eso118-g34 & 13.52 & g & s0 pec & * 0.4 & 4 40 17.4&-58 44 43 fd 78 & hic18929 & 8.39 & s & f0iii & 0.34 & 4 03 28.1 & -52 30 21 fd 127 & usno - a2 - 0300 - 01385078 & b14.7 & u & ? & * 1.3 & 4 36 17.0 & -57 53 32 fd 128 & tyc8512 - 1176 - 1 & 11.33 & s & ? & 0.84 & 4 43 34.8 & -55 49 34 fd 129 & usno - a2 - 0300 - 01371953 & b12.2 & u & ? & * 0.9 & 4 33 53.7 & -58 50 21 fd 130 & hic21874 & 8.04 & s & f5v & 0.46 & 4 42 10.5 & -56 07 08 fm 2 & hfd 121081 & 8.9 & s & b9v & 0.08&13 53 58.4 & -28 8 34 fm 3 & usno0600 - 09386883 & 14.01 & s & b ? & * -1.1&13 54 00.2 & -27 55 16 fm 4 & sao182054 & 9.34 & s & a2 & 0.28&13 53 44.8 & -27 06 02 fm 5 & hfd 120970 & 10.3 & s & a2 & 0.22&13 53 15.0 & -27 18 37 fm 6 & tyc6724 - 799 - 1 & 11.68 & s & a - f ? & 0.42&13 52 54.8 & -27 40 08 fm 7 & sao182056 & 8.1 & s & a6iv & 0.24&13 53 53.1 & -30 00 39 fm 8 & hfd 120867 & 10.2 & s & a0 & 0.09&13 52 36.0 & -27 30 00 fm 9 & sao182034 & 10.24 & s & a0 & 0.11&13 52 18.7 & -26 47 16 fm 10 & sao204927 & 7.9 & s & a3 & 0.10&13 52 17.6 & -30 47 06 fm 11 & hfd 120657 & 10.5 & s & a2 & 0.24&13 51 23.1 & -29 08 44 fm 12 & hfd 120733 & 9.69 & s & a0 & 0.10 & 13 51 48.7 & -30 48 12 fm 13 & sao181999 & 6.19 & s & a0v & -0.03&13 50 06.4 & -29 04 51 fm 14 & tyc7279 - 1224 - 1 & 12.99 & s & ? & -0.04 & 13 51 28.9 & -30 48 27 fm 15 & sao181996 & 7.70 & s & a6 & 0.45&13 49 54.2 & -28 11 11 fm 16 & usno0600 - 09347896 & 15.37 & u & ? & * -0.1&13 49 10.2 & -26 55 30 fm 17 & sao204899 & 8.6 & s & f6v & 0.41&13 50 38.2 & -30 15 41 fm 18 & sao181993 & 8.7 & s & a2 & 0.36&13 49 47.3 & -28 38 24 fm 19 & usno0600 - 09348084 & b15.2 & s & f ? & * 0.4&13 49 11.6&-27 25 30 fm 20 & sao204900 & 7.1 & s & a5 & 0.17&13 50 39.6 & -31 12 20 fm 21 & sao204892 & 7.2 & s & f0 & 0.29 & 13 50 20.9 & -31 04 03 fm 22 & tyc6728 - 518 - 1 & 11.16 & s & a ? & 0.27&13 49 10.9 & -28 34 31 fm 23 & usno0600 - 09343173 & 13.52 & s & a ? & * -0.2&13 48 35.8 & -27 56 33 fm 24 & tyc6727 - 804 - 1 & 11.41 & s & b - a ? & 0.19&13 48 05.8 & -29 20 22 fm 25 & sao181931 & 5.8 & s & a0v & 0.02&13 45 36.3 & -26 06 57 fm 26 & tyc7266 - 692 - 1 & 10.69 & s & a0 & 0.10 & 13 47 50.9 & -31 32 49 fm 27 & usno0600 - 09325326 & 11.83 & u & ? & * 0.5&13 46 27.3 & -28 02 45 fm 28 & n5291/b & 14.00 & g & s pec & ? & 13 47 24.1 & -30 24 27 fm 29 & sao181949 & 8.6 & s & f8 & 0.51&13 46 57.5 & -29 40 18 fm 30 & hfd 119753 & 8.2 & s & f3v & 0.60&13 45 35.0 & -27 51 15 fm 32 & sao204825 & 8.3 & s & g0 & 0.59&13 46 17.0 & -30 28 28 fm 33 & tyc7266 - 1276 - 1 & 11.17 & s & a - f ? & 0.13 & 13 46 07.6 & -30 12 29 fm 34 & sao181919 & 8.6 & s & a3 & 0.30 & 13 44 36.7 & -26 46 48 fm 35 & tyc7266 - 1014 - 1 & 11.47 & s & a - f ? & 0.17 & 13 46 25.2 & -31 45 48 fm 36 & sao181928 & 8.8 & s & f0 & 0.31&13 45 17.8 & -29 01 00 fm 37 & hfd 119463 & 10.5 & s & a3 & 0.24 & 13 43 52.4 & -26 47 38 fm 38 & usno0600 - 09305363 & 15.35 & s & a - f(e ) ? & * -0.2&13 44 10.0 & -28 17 21 fm 39 & sao181910 & 8.8 & s & a2 & 0.26 & 13 43 53.9 & -27 45 59 fm 41 & sao181914 & 9.3 & s & a2 & 0.07&13 44 25.4 & -30 07 50 fm 42 & sao204787 & 8.5 & s & f0 & 0.37&13 44 09.9 & -31 48 59 fm 43 & sao181891 & 7.0 & s & a2v & 0.90&13 43 00.9 & -30 10 58 fm 44 & sao204771 & 9.0 & s & f0iv & 0.38&13 43 27.1 & -31 26 39 fm 45 & sao181862 & 8.7 & s & a2v & 0.19&13 41 31.2 & -27 16 21 fm 46 & sao181876 & 8.5 & s & f0 & 0.31&13 42 15.5 & -29 46 12 fm 47 & sao181868 & 8.0 & s & f2 & 0.37&13 41 43.6 & -28 45 44 fm 48 & usno0525 - 13768846 & 12.17 & s & a ? & * -0.1&13 42 29.8 & -31 3 21 fm 49 & n5264 & 12.6 & g & ib(s)m & 0.55&13 41 31.9 & -29 55 08 fm 50 & hic66742 & 8.01 & s & f3 & 0.44 & 13 40 47.0 & -28 14 32 fm 51 & usno - a2 - 0600 - 15938674 & b13.7 & u & ? & * -2.0 & 13 40 46.1 & -28 56 54 fm 52 & sao181838 & 8.4 & s & a3 & 0.10&13 39 52.9 & -27 09 05 fm 53 & usno0525 - 13747141 & b13.8 & s & a ? & * 0.1&13 41 13.3 & -31 34 54 fm 54 & gsc06726001897 & 14.58 & s & ? & ? & 13 39 42.5 & -28 8 37 fm 56 & sao181827 & 8.1 & s & a3iv & 0.16&13 38 44.5 & -27 15 35 fm 57 & sao181823 & 8.4 & s & a1v & 0.09&13 38 31.2 & -26 54 21 fm 58 & ngc5253 & 10.87 & g&im pec & 0.38&13 39 56.8 & -31 38 11 fm 59 & sao204696 & 9.59 & s & a3 & 0.35&13 39 07.3 & -30 15 30 fm 60 & sao181825 & 5.8 & s & f3v & 0.42 & 13 38 42.1 & -29 33 39 fm 61 & sao204695 & 9.37 & s & f5 & 0.38&13 39 03.6 & -30 44 53 fm 62 & sao204711 & 9.8 & s & a0 & 0.18 & 13 39 41.8 & -32 29 21 fm 63 & sao181807 & 8.6 & s & a5 & 0.17&13 37 34.8 & -27 03 31 fm 64 & sao181821 & 7.0 & s & f2iii & 0.40&13 38 23.3 & -29 50 20 fm 69 & he1334 - 2739 & 16.41 & ag & qso cand . & & 13 37 04.7 & -27 54 20 fm 70 & n5236 & 8.2 & g & sab(s)c & 0.61&13 36 58.8 & -29 51 52 fm 71 & i4293 & 13.50 & g & sa0 & * 1.5&13 36 01.7 & -25 52 53 fm 72 & sao204668 & 9.08 & s & a3 & 0.35&13 37 33.4 & -32 08 24 fm 73 & he1333 - 2853 & 15.92 & s & sda or qso cand . & & 13 36 29.5 & -29 08 53 fm 74 & hfd 118250 & 10.7 & s & a2 & 0.25&13 36 06.8 & -28 42 20 fm 75 & usno0525 - 13671585 & b13.4 & s & a ? & * 0.5&13 36 40.5 & -30 18 36 fm 76 & hfd 118350 & 9.8 & s & a0 & 0.31 & 13 36 58.8 & -32 17 38 fm 77 & sao204651 & 7.6 & s & a9v & 0.42&13 36 46.7 & -32 24 13 fm 78 & usno0600 - 09235666 & 13.7 & us & ? & * 1.1&13 35 00.6 & -26 05 32 fm 79 & sao181775 & 8.4 & s & a0 & 0.01&13 35 30.8 & -29 09 18 fm 80 & sao181779 & 8.1 & s & f2v & 0.41&13 35 41.1 & -29 22 40 fm 81 & sao181760 & 8.9 & s & m0 & 1.27&13 34 18.0 & -25 22 49 fm 82 & hfd 118147 & 10.03 & s & a3 & 0.27 & 13 35 34.3 & -31 25 26 fm 83 & hfd 118162 & 9.7 & s & a0 & 0.17 & 13 35 37.8 & -32 01 04 fm 85 & sao204623 & 6.8 & s & f0 & 0.37&13 35 12.2 & -32 12 45 fm 86 & sao204613 & 9.7 & s & f2 & 0.37&13 34 42.4 & -31 12 13 fm 87 & sao181742 & 6.9 & s & a9iv & 0.33&13 32 54.0 & -26 35 15 fm 88 & sao181749 & 8.7 & s & a2 & 0.07&13 33 09.2 & -29 17 58 fm 89 & sao181737 & 5.7 & s & a1v & 0.04&13 32 35.9 & -28 41 34 fm 90 & sao204584 & 10.2 & s & a0 & 0.01&13 33 19.7 & -31 39 13 fm 91 & sao181736 & 8.2 & s & a4v & 0.18&13 32 35.1 & -29 17 45 fm 93 & sao181735 & 6.4 & s & f5iv & 0.44&13 32 34.5&-29 33 55 fm 94 & ppm291518 & 10.49 & s & a ? & 0.20 & 13 33 06.6 & -31 47 36 fm 95 & sao181723 & 6.4 & s & a1v & 0.09&13 31 33.2 & -28 06 46 fm 96 & sao181716 & 7.5 & s & a4iv & 0.18&13 31 07.4 & -25 39 25 fm 97 & hfd 117531 & 9.70 & s & a2 & 0.27&13 31 20.8 & -27 31 46 fm 98 & tyc7269 - 1285 - 1 & 11.21 & s & ? & 0.04 & 13 32 42.5 & -32 47 00 fm 105 & sao181690 & 7.6 & s & a5 & 0.17&13 29 30.4 & -27 23 22 fm 106 & sao181698 & 9.2 & s & a3v & 0.16&13 29 53.0 & -28 56 41 fm 107 & sao181691 & 9.3 & s & a0 & 0.01&13 29 32.5 & -29 05 28 fm 108 & gsc06721 - 01689 & 13.4 & s & hot star ? & * -0.8&13 28 18.2 & -26 33 51 fm 119 & sao181601 & 8.8 & s & a2 & 0.07&13 22 42.7 & -30 11 51 fm 120 & gsc06712 - 01800 & 13.4 & s & a ? & * -0.3&13 21 56.2 & -28 26 31 fm 121 & usno - a2 - 0600 - 15616353 & b12.1 & s & a ? & * 0.3 & 13 20 45.5 & -29 29 33 fm 122 & sao204359 & 8.48 & s & b5iii & -0.11&13 19 43.0 & -30 19 56 fm 136 & sao181916 & 9.2 & s & k0 & 0.86&13 44 26.8 & -26 19 27 fm 137 & usno0600 - 09301026 & b14.6 & u & ? & * 1.3&13 43 36.1 & -26 29 45 fm 138 & gsc06727 - 01025 & 13.0 & s & f ? & * 1.2&13 44 43.6 & -29 37 49 fm 139 & hfd 119523 & 10.54 & s & f5 & 0.31 & 13 44 14.6 & -28 48 56 fm 174 & sao204493 & 8.3 & s & f3v & 0.48&13 27 45.0 & -30 47 03 fm 175 & eso444 - 46 & 15 & g & e & ? & 13 27 56.8 & -31 29 43 fm 176 & eso444 - 37 & 14.82 & g & sb(s)dm & * 0.6&13 26 59.8 & -30 04 27 fm 177 & ppm261656 & 9.7 & s & f8 & 0.38&13 26 21.8 & -26 50 54 fm 190 & sao204469 & 6.84 & s & f0v & 0.40&13 26 08.1 & -32 32 51 fm 191 & sao204853 & 9.1 & s & a2 & 0.07&13 48 10.1 & -32 14 18 fm 192 & sao204562 & 9.1 & s & f0 & 0.31&13 32 03.0 & -32 12 44 fm 193 & sao181770 & 9.00 & s & f8 & 0.54&13 35 05.1 & -25 18 15 fm 194 & hfd 119027 & 10.5 & s & a3 & 0.48&13 41 19.8 & -28 47 00 fm 195 & sao20462 & 8.0 & s & f3 & 0.46&13 35 15.9 & -31 24 31 fm 196 & gsc07266 - 00123 & 12.97 & s & ? & * 1.8&13 40 06.6 & -31 31 11 fm 197 & sao181785 & 9.6 & s & g5 & 0.64 & 13 36 13.4 & -29 39 40 ft 1 & ppm325630 & 10.57 & s & ? & 0.03 & 19 55 29.2 & -49 31 21 ft 2 & sao229911 & 8.97 & s & ? & 0.36 & 19 53 14.7 & -48 28 10 ft 3 & hic98444 & 6.67 & s & f2iv & 0.40 & 20 00 08.4 & -50 13 00 ft 4 & usno0375 - 38494969 & b12.8 & u & ? & * -0.5 & 19 54 53.6 & -48 32 33 ft 5 & hic98040 & 8.14 & s & f0v & 0.31 & 19 55 20.4 & -48 16 56 ft 6 & tyc8399 - 146 - 1 & 9.05 & s & ? & 0.49 & 20 02 07.3 & -49 58 54 ft 7 & usno0375 - 38670988 & b13.3 & u & ? & * -0.3 & 20 04 52.6 & -49 57 14 ft 8 & hic98449 & 7.67 & s & f6v & 0.47 & 20 00 11.8 & -48 44 54 ft 9 & sao229930 & 9.87 & s & a0 & 0.07 & 19 56 10.6 & -47 23 15 ft 10 & hic98880 & 8.55 & s & f0v & 0.37 & 20 04 47.8 & -49 45 18 ft 11 & usno0375 - 38647724 & b14.1 & u & ? & * 0.3 & 20 03 31.2 & -49 21 06 ft 12 & sao229910 & 8.19 & s & a2 & 0.14 & 19 53 09.0 & -46 22 12 ft 13 & sao229967 & 8.92 & s & a5 & 0.29 & 19 59 57.9 & -48 11 25 ft 14 & sao230028 & 9.6 & s & a2 & 0.03 & 20 05 07.1 & -49 30 17 ft 15 & sao246465 & 9.2 & s & a3 & 0.28 & 20 09 33.0 & -50 26 25 ft 16 & tyc8400 - 1697 - 1 & 9.56 & s & ? & 0.22 & 20 05 59.3 & -49 31 10 ft 17 & tyc8400 - 1353 - 1 & 11.1 & s & ? & 0.68 & 20 07 46.2 & -49 56 35 ft 18 & tyc8395 - 622 - 1 & 9.08 & s & f2 & 0.36 & 19 59 45.4 & -47 32 59 ft 19 & tyc8396 - 100 - 1 & 10.18 & s & ? & 0.36 & 20 03 51.2 & -48 37 04 ft 20 & sao229945 & 9.23 & s & a2 & 0.17 & 19 57 03.4 & -46 35 26 ft 21 & sao230036 & 10.42 & s & a0 & -0.05 & 20 06 17.2 & -48 59 39 ft 22 & tyc8396 - 580 - 1 & 9.66 & s & ? & 0.30 & 20 04 59.6 & -48 36 49 ft 23 & sao230006 & 10.12 & s & a3 & 0.16 & 20 02 52.5 & -47 58 52 ft 24 & hic98577 & 6.99 & s & f2iv & 0.35 & 20 01 25.3 & -47 23 51 ft 25 & hic99595 & 7.92 & s & f3v & 0.39 & 20 12 43.2 & -50 18 41 ft 26 & tyc8395 - 1024 - 1 & 10.5 & s & ? & 0.16 & 20 02 06.0 & -47 18 51 ft 27 & hic99703 & 8.42 & s & a2v & 0.12 & 20 13 54.9 & -50 19 16 ft 28 & usno0375 - 38688655 & b13.3 & u & ? & * -0.5 & 20 05 55.8 & -48 16 10 ft 29 & hic99260 & 8.83 & s & f2v & 0.38 & 20 08 58 & -49 04 42 ft 30 & usno0375 - 38800440 & b12.3 & u & ? & * -0.7 & 20 12 52.0 & -49 58 28 ft 31 & sao230049 & 7.95 & s & f2 & 0.43 & 20 07 39.5 & -48 39 08 ft 32 & ngc6851a / b & 14.81 & gp & sb & 0.84 & 20 05 39.7 & -47 58 40 ft 33 & hic98355 & 7.49 & s & f6v & 0.47 & 19 58 58.6 & -46 05 17 ft 34 & euvej2009 - 48.8 & 15.3 & qs & bl lac & * 0.6 & 20 09 25.4 & -48 49 54 ft 35 & sao230030 & 9.06 & s & a0 & 0.10 & 20 05 10.8 & -47 29 17 ft 36 & usno0375 - 38564499 & b11.6 & u & ? & * 1.0 & 19 58 43.0 & -45 36 31 ft 37 & tyc8417 - 667 - 1 & 10.02 & s & ? & 1.10 & 20 18 17.8 & -50 39 33 ft 38 & sao229984 & 8.78 & s & a0 & 0.00 & 20 00 53.4 & -45 41 58 ft 39 & sao230088 & 9.02 & s & a3 & 0.24 & 20 11 23.6 & -48 22 46 ft 40 & sao230060 & 8.09 & s & f0 & 0.29 & 20 08 30.1 & -47 33 34 ft 41 & tyc8413 - 1194 - 1 & 10.96 & s & ? & 0.58 & 20 19 24.6 & -50 12 41 ft 42 & sao230054 & 8.77 & s & a2 & 0.02 & 20 08 07.1 & -47 10 56 ft 43 & hic98512 & 5.8 & s & a8iii & 0.00 & 20 00 48.3 & -45 06 47 ft 44 & sao230121 & 9.75 & s & ? & 0.22 & 20 15 36.5 & -48 58 21 ft 45 & hic99130 & 8.72 & s & f0v & 0.39 & 20 07 31.0 & -46 50 50 ft 46 & hic100349 & 7.31 & s & a3v & 0.13 & 20 21 02.2 & -50 01 02 ft 47 & tyc8409 - 5 - 1 & 10.71 & s & ? & 1.2 & 20 15 15.6 & -48 43 43 ft 48 & tyc8392 - 1957 - 1 & 11.21 & s & ? & -0.05 & 20 06 24.1 & -46 04 05 ft 49 & usno0375 - 38880708 & b14.1 & u & ? & * -2.2 & 20 18 07.4 & -48 57 32 ft 50 & sao230000 & 7.91 & s & a0 & -0.03 & 20 02 10.2 & -44 27 58 ft 51 & sao230062 & 8.28 & s & f5v & 0.43 & 20 08 37.9 & -46 18 05 ft 52 & usno0375 - 38684700 & b15.7 & u & ? & * -0.3 & 20 05 41.5 & -45 33 25 ft 53 & tyc8413 - 818 - 1 & 8.03 & s & ? & 1.08 & 20 22 17.1 & -49 26 51 ft 54 & hic98859 & 7.83 & s & a9iv & 0.30 & 20 04 29.8 & -44 56 49 ft 55 & sao230181 & 9.54 & s & f5 & 0.35 & 20 23 36.6 & -49 47 44 ft 56 & sao230012 & 8.34 & s & a2 & 0.17 & 20 03 29.4 & -44 35 36 ft 57 & tyc08409 - 887 - 1 & b11.9 & u & ? & 0.58 & 20 18 29.1 & -48 32 11 ft 58 & usno0375 - 38762110 & b11.2 & u & ? & * 1.2 & 20 10 25.9 & -46 25 34 ft 59 & tyc8413 - 1203 - 1 & 10.68 & s & ? & 0.90 & 20 23 18.8 & -49 20 47 ft 60 & tyc8409 - 365 - 1 & 9.09 & s & ? & 0.05 & 20 20 07.0 & -48 28 41 ft 61 & sao230017 & 9.25 & s & a0 & 0.06 & 20 03 51.4 & -43 50 01 ft 62 & sao230111 & 9.68 & s & ? & 0.21 & 20 13 54.5 & -46 33 23 ft 63 & sao230099 & 9.07 & s & a0 & 0.19 & 20 12 22.3 & -46 10 00 ft 64 & usno - a2 - 0450 - 38582716 & b15.8 & u & ? & * 1.5 & 20 07 43.4 & -44 56 45 ft 65 & usno0375 - 38856050 & b12.8 & u & ? & * 1.1 & 20 16 32.2 & -47 06 59 ft 66 & hic100184 & 6.13 & s & f5v & 0.48 & 20 19 17.7 & -47 34 47 ft 67 & tyc8392 - 1398 - 1 & 10.42 & s & ? & 0.46 & 20 12 25.6 & -45 49 29 ft 68 & tyc7959 - 412 - 1 & 10.74 & s & ? & 0.28 & 20 09 38.3 & -44 44 52 ft 69 & tyc7959 - 1706 - 1 & 11 & s & ? & 0.19 & 20 08 59.3 & -44 30 19 ft 70 & hic99671 & 7.79 & s & ap & 0.00 & 20 13 34.8 & -45 35 26 ft 71 & tyc8409 - 1802 - 1 & 9.78 & s & ? & 0.05 & 20 21 56.8 & -47 43 25 ft 72 & sao230183 & 8.92 & s & f0 & 0.30 & 20 24 00.7 & -48 13 43 ft 73 & tyc8414 - 154 - 1 & 8.93 & s & ? & 0.15 & 20 30 20.6 & -49 26 29 ft 74 & tyc7959 - 1973 - 1 & 10.57 & s & ? & 0.34 & 20 09 39.5 & -44 18 56 ft 75 & tyc8405 - 1915 - 1 & 10.85 & s & ? & 0.10 & 20 18 15.0 & -46 31 36 ft 76 & hic100872 & 7.84 & s & f3iv & 0.43 & 20 27 11.1 & -48 35 35 ft 77 & sao230082 & 9.58 & s & a5 & 0.18 & 20 10 30.7 & -44 08 04 ft 78 & tyc8410 - 1739 - 1 & 8.41 & s & a7iv & 0.26 & 20 26 38.3 & -47 57 42 ft 79 & sao230096 & 9.26 & s & a0 & 0.10 & 20 12 03.3 & -44 15 42 ft 80 & hic99426 & 7.84 & s & a3 m & 0.25 & 20 10 50.2 & -43 36 11 ft 81 & usno - a2 - 0375 - 38789037 & b13.6 & u & ? & * -3.0 & 20 26 22.0 & -47 30 44 ft 82 & hic99448 & 7.71 & s & a4v & 0.22 & 20 11 04.7 & -43 28 28 ft 83 & hic100467 & 9.39 & s & f0v & 0.32 & 20 22 26.3 & -46 26 58 ft 84 & usno0450 - 38842338 & b14.4 & u & ? & * -0.4 & 20 12 41.5 & -43 50 53 ft 85 & hic99722 & 8.42 & s & f8iv & 0.44 & 20 14 06.4 & -44 10 24 ft 86 & sao230212 & 7.65 & s & f0 & 0.33 & 20 27 00.9 & -47 15 34 ft 87 & usno0375 - 39045904 & b13.3 & u & ? & * -0.8 & 20 29 34.1 & -47 47 26 ft 88 & usno0450 - 38892109 & b12.5 & u & ? & * 0.7 & 20 15 38.7 & -44 29 59 ft 89 & tyc7960 - 215 - 1 & 9.7 & s & a3 & 0.17 & 20 12 04.3 & -43 23 53 ft 90 & tyc8405 - 1234 - 1 & 9.78 & s & ? & 0.33 & 20 20 44.3 & -45 39 49 ft 91 & hic100787 & 6.73 & s & a9iii & 0.29 & 20 26 04.8 & -46 39 36 ft 92 & sao230170 & 8.65 & s & a0 & 0.09 & 20 21 57.2 & -45 38 20 ft 93 & hic99806 & 8.19 & s & a7v & 0.21 & 20 14 59.8 & -43 40 42 ft 94 & sao230230 & 9.36 & s & a2 & 0.20 & 20 28 23.9 & -46 57 21 ft 95 & tyc8410 - 743 - 1 & 11.44 & s & ? & 0.22 & 20 33 45.99 & -47 56 38 ft 96 & sao230200 & 8.95 & s & ? & 0.33 & 20 26 04.0 & -46 09 29 ft 97 & sao230238 & 7.42 & s & a0 & 0.05 & 20 29 38.6 & -46 09 05 ft 98 & hic100487 & 7.94 & s & a5 m & 0.34 & 20 22 38.1 & -44 35 14 ft 99 & tyc8406 - 29 - 1 & 10.71 & s & ? & 1.06 & 20 34 00.4 & -46 51 16 ft 100 & tyc8406 - 795 - 1 & 9.57 & s & ? & 0.33 & 20 29 53.0 & -45 54 48 ft 101 & sao230237 & 8.35 & s & ? & 0.23 & 20 29 30.5 & -45 22 58 ft 102 & sao230215 & 8.16 & s & f0 & 0.33 & 20 27 13.6 & -44 45 31 ft 103 & hic100528 & 8.39 & s & a6iv & 0.23 & 20 23 02.1 & -43 40 14 ft 104 & usno - a2 - 0450 - 38843012 & b13.6 & u & ? & * 1.1 & 20 22 45.4 & -43 29 43 ft 106 & tyc7961 - 435 - 1 & 9.89 & s & ? & 0.14 & 20 23 54.1 & -43 39 11 ft 107 & tyc7961 - 1763 - 1 & 10.5 & s & ? & 0.17 & 20 29 19.0 & -44 57 18 ft 108 & ngc6902 & 11.64 & ig & sa(r)b & 0.62 & 20 24 27.7 & -43 39 09 ft 109 & sao230295 & 10.23 & s & a0 & 0.03 & 20 36 54.6 & -46 27 22 ft 110 & tyc8406 - 1679 - 1 & 8.00 & s & f0 & 0.3 & 20 32 09.7 & -45 21 14 ft 111 & tyc8406 - 457 - 1/2 & 7.52 & s & a7iii & 0.28 & 20 36 30.5 & -45 33 28 ft 112 & usno0450 - 39114689 & b12.2 & u & ? & * 1.0 & 20 30 14.1 & -44 00 30 ft 113 & tyc8399 - 463 - 1 & 10.52 & s & ? & 0.17 & 20 01 32.4 & -50 17 46 ft 114 & usno0375 - 38638499 & b11.9 & u & ? & * 0.7 & 20 02 59.4 & -50 17 45 ft 115 & usno0375 - 38534372 & b13.7 & u & ? & * 1.4 & 19 57 01.8 & -48 18 58 ft 116 & usno - a2 - 0375 - 38402237 & b12.3 & u & ? & * 0.7 & 20 01 19.7 & -49 33 04 ft 117 & usno0375 - 38485210 & b13.9 & u & ? & ? & 19 54 22.9 & -47 37 44 ft 118 & usno0375 - 38502981 & b11.9 & u & ? & ? & 19 55 19.0 & -47 48 48 ft 119 & tyc8395 - 953 - 1 & 11.49 & s & ? & 0.04 & 19 57 16.1 & -48 21 29 ft 120 & tyc8400 - 336 - 1 & 10.97 & s & ? & 0.34 & 20 03 56.4 & -49 48 25 ft 121 & sao22994 & 9.61 & s & ? & 0.21 & 19 56 57.0 & -47 25 22 ft 122 & usno - a2 - 0375 - 38326701 & b12.4 & u & ? & * 1.0 & 19 57 01.1 & -47 03 45 ft 123 & tyc8391 - 843 - 1 & 10.47 & s & ? & 0.24 & 19 55 20.5 & -46 47 57 ft 124 & usno0375 - 38606749 & b13.8 & u & ? & * 0.5 & 20 01 08.3 & -48 28 31 ft 125 & tyc8391 - 2312 - 1 & 10.04 & s & ? & 0.40 & 19 54 35.9 & -46 30 37 ft 126 & usno0375 - 38593648 & b15.4 & u & ? & ? & 20 00 22.2 & -47 55 22 ft 127 & tyc8391 - 1762 - 1 & 10.69 & s & ? & 0.28 & 19 54 37.7 & -46 13 35 ft 128 & usno0375 - 38709646 & b12.3 & u & ? & * -0.4 & 20 07 11.4 & -49 30 04 ft 129 & tyc8404 - 1258 - 1 & 8.53 & s & ? & 0.56 & 20 12 02.9 & -50 43 23 ft 130 & tyc8395 - 1998 - 1 & 10.5 & s & ? & 0.83 & 19 59 33.0 & -47 04 30 ft 131 & ngc6845/a / b / c / fd & 14.28 & ig & sbb?pec & ? & 20 01 05.8 & -47 03 34 ft 132 & pgc063901 & 14.27 & gp & sab & * 0.2 & 19 58 17.6 & -46 06 46 ft 133 & sao98233 & 8.64 & s & f5iv & 0.01 & 19 57 45.2 & -45 49 39 ft 134 & usno0375 - 38806618 & b15.3 & u & ? & * 0.1 & 20 13 16.6 & -49 36 20 ft 135 & tyc8396 - 154 - 1 & 10.66 & s & ? & 0.24 & 20 09 52.3 & -48 34 30 ft 136 & tyc8391 - 677 - 1 & 10.85 & s & ? & 0.59 & 19 58 53.2 & -45 05 24 ft 137 & usno - a2 - 0375 - 38425789 & b11.1 & u & ? & * 0.5 & 20 02 43.4 & -45 57 22 ft 138 & usno - a2 - 0375 - 38637967 & b13.6 & u & ? & * 0.2 & 20 15 59.7 & -48 51 28 ft 139 & tyc8392 - 28 - 1 & 10.59 & s & ? & 0.15 & 20 07 57.6 & -46 43 45 ft 140 & tyc8413 - 614 - 1 & 9.98 & s & ? & 0.35 & 20 20 38.8 & -49 49 00 ft 141 & usno0375 - 38888302 & b12.7 & u & ? & * -0.8 & 20 18 36.2 & -49 13 07 ft 142 & tyc8396 - 418 - 1 & 10.63 & s & ? & 0.91 & 20 14 43.0 & -48 12 50 ft 143 & usno0375 - 38910011 & b13.4 & u & ? & * 0.8 & 20 20 01.1 & -49 19 57 ft 144 & tyc8413 - 1738 - 1 & 10 & s & ? & 0.34 & 20 24 36.6 & -50 25 20 ft 145 & tyc8392 - 1262 - 1 & 10.06 & s & ? & 0.65 & 20 06 16.8 & -45 54 16 ft 146 & eso234-g004 & 15.27 & gp & sbpec & ? & 20 20 43.4 & -49 07 29 ft 147 & usno - a2 - 0375 - 38528298 & b15.8 & u & ? & * -0.2 & 20 08 55.8 & -46 05 47 ft 148 & usno0375 - 38748771 & b11.3 & u & ? & * 0.3 & 20 09 35.9 & -46 07 28 ft 149 & usno0375 - 38753431 & b12.6 & u & ? & * 1.0 & 20 09 53.4 & -46 03 28 ft 150 & tyc7959 - 742 - 1 & 11.12 & s & ? & 0.15 & 20 05 11.9 & -44 39 32 ft 151 & usno0450 - 38690235 & b13.5 & u & ? & * 1.2 & 20 04 58.4 & -44 29 13 ft 152 & usno - a2 - 0375 - 38561392 & b15.7 & u & ? & * 0.0 & 20 10 59.6 & -45 53 50 ft 153 & usno0375 - 38758273 & 13.3 & u & ? & * 0.8 & 20 10 11.8 & -45 41 40 ft 154 & usno0375 - 38732664 & b15.9 & u & ? & * -0.5 & 20 08 36.3 & -45 05 16 ft 155 & usno0375 - 38963244 & b12.7 & u & ? & * 1.4 & 20 23 35.0 & -48 49 06 ft 156 & usno - a2 - 0450 - 38583089 & b17.7 & u & ? & * 0.1 & 20 07 44.6 & -44 37 38 ft 157 & usno0450 - 38718073 & b14.7 & u & ? & * -0.2 & 20 06 21.4 & -44 03 01 ft 158 & tyc8405 - 885 - 1 & 10.36 & s & ? & 0.37 & 20 15 24.7 & -46 19 50 ft 159 & usno0450 - 38794373 & b14.1 & u & ? & * 1.0 & 20 10 16.8 & -44 52 15 ft 160 & sao230055 & 8.49 & s & f5 & 0.441 & 20 08 01.3 & -44 11 33 ft 161 & eso284-ig045 & 14 & ig & sabbc & ? & 20 16 48.6 & -46 31 55 ft 162 & sao230119 & 9.75 & s & f5 & 0.38 & 20 15 12.5 & -46 03 12 ft 163 & tyc7959 - 962 - 1 & 10.18 & s & ? & 0.44 & 20 07 49.5 & -43 37 54 ft 164 & usno0375 - 39013302 & b14.4 & u & ? & * 0.9 & 20 27 11.4 & -48 19 47 ft 165 & usno - a2 - 0375 - 38776135 & b12.0 & u & ? & * 0.0 & 20 25 27.5 & -47 51 23 ft 166 & tyc7960 - 1677 - 1 & 9.77 & s & ? & 0.31 & 20 14 09.1 & -44 51 02 ft 167 & tyc8414 - 685 - 1 & 10.21 & s & ? & 0.36 & 20 34 24.2 & -49 05 05 ft 168 & usno0375 - 38887188 & b15.1 & u & ? & ? & 20 18 32.0 & -45 20 59 ft 169 & pgc064404 & 14.31 & pm & sabbcp & 0.95 & 20 16 45.5 & -44 41 43 ft 169 & pgc064399 & 14.51 & pm & sbdm & 1.17 & 20 16 31.0 & -44 42 32 ft 170 & sao230206 & 9.23 & s & f2 & 0.37 & 20 26 48.6 & -46 26 34 ft 171 & tyc8410 - 1555 - 1 & 11.42 & s & ? & 0.37 & 20 35 23.8 & -48 19 20 ft 172 & usno0375 - 38983401 & b15.2 & u & ? & * 1.1 & 20 25 03.5 & -45 42 55 ft 173 & tyc8406 - 1146 - 1 & 11.11 & s & ? & 0.07 & 20 29 17.2 & -46 27 58 ft 174 & hic101168 & 9.17 & s & f5v & 0.41 & 20 30 22.4 & -46 27 20 ft 175 & usno0375 - 39019592 & b14.2 & u & ? & * 1.3 & 20 27 39.3 & -45 40 49 ft 176 & usno0450 - 38948511 & b12.6 & u & ? & * 0.5 & 20 19 12.5 & -43 34 02 ft 177 & usno0450 - 39036632 & b13.6 & u & ? & * 1.4 & 20 24 53.5 & -44 37 31 ft 178 & sao230167 & 9.75 & s & f0 & 0.47 & 20 21 34.2 & -43 49 17 ft 179 & usno0375 - 39120597 & b12.8 & u & ? & * 0.8 & 20 35 14.8 & -46 58 10 ft 180 & usno0450 - 39039977 & b15.9 & u & ? & * 0.8 & 20 25 06.8 & -44 27 39 ft 181 & usno - a2 - 0375 - 38811091 & b13.9 & u & ? & * 1.2 & 20 27 58.9 & -45 00 52 ft 182 & tyc8406 - 452 - 1 & 10.59 & s & ? & 0.36 & 20 29 12.5 & -45 07 14 ft 183 & tyc7962 - 710 - 1 & 9.85 & s & ? & 0.52 & 20 31 12.8 & -44 45 57 ft 184 & tyc8414 - 669 - 1 & 8.46 & s & a2vn & 0.10 & 20 29 26.78 & -49 46 59 ft 185 & tyc8414 - 387 - 1 & 10.66 & s & ? & -0.12 & 20 29 48.9 & -49 52 24 ft 186 & hic97756 & 8.31 & s & a7iiim & 0.28 & 19 51 58.9 & -46 51 42 ft 187 & tyc7960 - 817 - 1 & 9.72 & s & ? & 0.38 & 20 15 08.2 & -43 34 14 ft 188 & sao230029 & 9.67 & s & f0 & 0.43 & 20 05 08.9 & -47 21 42 ft 189 & usno0450 - 39025207 & b13.2 & u & ? & * 2.0 & 20 24 07.9 & -43 41 10 lllccccc fc 27 & eso 323-g025 & sa(rs)b & 12.16@xmath230.20 & 13.39 & 1.17 & fc 32 & ngc 4553 & sa(r)0 & 12.51@xmath230.58 & 12.17 & 1.02 & fc 44 & usno & sab(r)cd & 10.66@xmath230.26 & 9.9 & * -1.3 & fc 61 & pgc 042966 & e ? & 13.06@xmath230.57 & 12.72 & 1.08 & fc 71 & ngc 4683 & sb(s)0 & 12.58@xmath230.43 & 13.8 & 0.93 & fc 75 & ngc 4930 & sb(rs)bc & 12.65@xmath230.44 & 12.0 & 0.76 & fc 161 & ngc 4679 & sa(s)bc & 13.32@xmath230.45 & 11.94 & 1.38 & fc 186 & eso 323-g054 & s & 13.68@xmath230.61 & 14.86 & & fc 229 & eso 269-ig056 & sb(s)m pec & 11.92@xmath230.46 & 14.01 & & fc 235 & eso 269-ig022/3 & s0+e & 12.74@xmath230.34 & 14.75 + 16.17 & & int.pair fc 238 & pgc044774 & sab(s ) pec & 12.58@xmath230.61 & 16.0 & & fd 41 & ngc 1566 & sab(rs)bc sy1 & 9.68@xmath230.19 & 10.33 & 0.56 & xrays fd 53 & ngc 1602 & ib(s)m pec & 12.29@xmath230.48 & 12.98 & 0.35 & fd 62 & euvej0425 - 57.1 & sy1 & 12.03@xmath230.36 & 14.1 & * -0.6 & xrays fd 77 & eso 118-g034 & s0 pec & 11.64@xmath230.39 & 13.04 & 0.44 & hii gal . fd 82 & ngc 1522 & s0 pec & 12.62@xmath230.57 & 13.56 & 0.27 & fd 86 & ngc 1515 & sab(s)bc & 12.90@xmath230.67 & 11.20 & 0.85 & fd 93 & ic 2032 & iab(s)m pec & 12.89@xmath230.68 & 14.73 & * -3.6 & fd 97 & ngc 1536 & sb(s)c pec & 12.92@xmath230.58 & 12.52 & 0.63 & fd 100 & ic 2073 & sb(s)cd ? pec & 11.73@xmath230.37 & 13.78 & 0.77 & fd 111 & pgc 015445 & e & 14.16@xmath232.76 & 15.0 & * 0.7 & fd 118 & eso 157-g044 & ib(s)m & 12.19@xmath230.60 & 14.87 & & fd 134 & he0442 - 5652 & qso z=0.34 & 12.69@xmath230.59 & 16.3 & * 0.2 & fd 142b & usno & qso z=0.29 & 11.55@xmath230.25 & 15.6 & * -0.2 & iras fsc 25 + 60 fm 28 & ngc 5291/b & s pec & 12.57@xmath230.47 & 14.0 & & int.pair fm 49 & ngc 5264 & ib(s)m & 11.14@xmath230.23 & 12.00 & 0.60 & ddo 242 fm 55 & he1337 - 2919 & qso & 12.02@xmath230.32 & 16.06 & & candidate fm 58 & ngc 5253 & i m pec & 9.89@xmath230.19 & 10.87 & 0.38 & * burst , iue fm 69 & he1334 - 2739 & qso & 12.59@xmath230.46 & 16.41 & & candidate fm 70 & ngc 5236=m83 & sab(s)c & 7.92@xmath230.17 & 7.54 & 0.66 & * burst ; xray ; iue fm 71 & ic 4293 & sa0 & 12.21@xmath230.38 & 13.50 & * 1.5 & fm 104 & gsc07265 - 02231 & em.ln.gal . & 11.77@xmath230.29 & 12.7 & * 0.0 & abell 3558 z@xmath160.04 ? fm 112 & ic 4248 & s ? & 12.87@xmath230.50 & 13.85 & * 0.7 & fm 124 & eso 445-g076 & s & 12.83@xmath230.69 & 14.66 & * 0.6 & fm 126 & he1348 - 2751 & qso & 12.22@xmath230.43 & 16.62 & & candidate & fm 130 & he1348 - 2751 & qso & 12.45@xmath230.46 & 16.62 & & candidate & fm 132 & ic 4327 & sb(s)c ? & 13.62@xmath230.85 & 13.89 & 0.61 & fm 141 & ic 4318 & s ? & 13.50@xmath230.46 & 14.54 & & fm 142 & eso - lv4450171 & s & 14.07@xmath230.71 & 17.03 & & fm 152b & he1336 - 2624 & qso & 13.04@xmath230.60 & 16.53 & & candidate fm 163 & eso 444-g075 & i ? & 13.74@xmath230.58 & 14.27 & * 0.7 & fm 175 & eso 444-g046 & cd & 13.91@xmath230.56 & 13.85 & & abell 3558 z@xmath160.04 ? fm 176 & eso 444-g037 & sb(s)dm : & 13.63@xmath230.79 & 14.82 & * 0.6 & lsb ; ugca 358 fm 181 & ngc 5135 & sb(l)ab & 13.84@xmath231.01 & 12.11 & 0.77 & tol 34 ; * burst+sy2 ; iue fm 185 & he1317 - 2937 & em.ln.gal . & 13.64@xmath231.23 & 17.79 & & ft 32 & ngc 6851 a+b & sb ( rs)d+sb(s)m & 10.08@xmath230.20 & 14.81 + 16.67 & 0.84 & ft 34 & euvej2009 - 48.8 & qso & 11.56@xmath230.34 & 15.3 & * 0.6 & bl lac z=0.071 ; xray ft 108 & ngc 6920 & sa0 & 11.75@xmath230.33 & 12.35 & 1.22 & ft 131 & ngc 6845 abcd & group & 12.50@xmath230.54 & 14.28 & & interacting ft 132 & pgc 063901 & sab(rs)bc & 13.46@xmath230.73 & 14.27 & * 0.2 & ft 146 & eso 234-g004 & sb(s)dm pec & 12.77@xmath230.66 & 15.27 & & ft 161 & eso 284-ig045 & sab(rs)bc : & 13.05@xmath230.74 & 13.32 & 0.68 & ft 169a & pgc 064404 & sab(s)bc pec & 12.44@xmath230.52 & 14.31 & 0.95 & nelg ft 169b & pgc 064399 & sb(s)dm & 12.44@xmath230.52 & 14.51 & 1.17 &