sentences
stringlengths 11
4.62k
| source
stringclasses 3
values |
|---|---|
In conclusion, our results show that combining firstprinciples transport calculations and high resolution low temperature IETS it is possible to characterize the configuration of single-molecule junctions. The agreement between experiment and theory is excellent and, in addition to identifying the presence of pentanedithiol molecules, allows us to extract structural information of the molecular junction, e.g., its state of strain. Indeed, the IETS signal indicates that the electrons are passing through the backbone of the molecule and that the it is bound to the electrodes by the thiol groups. TF, DSP and AA acknowledge stimulating discussions with Magnus Paulsson. This work has been supported by the Spanish MICINN (MAT2008-01735, FIS2007-6671-C02-00 and MAT2007-62732, and Consolider-Ingenio 2010 CSD2007-0010), the CAM ("CITECNOMIK" P-ESP-000337-0505), the UPV/EHU (IT-366-07), and the Basque Depto. de Industria and the Diputación Foral de Guipuzcoa ("ETORTEK"), the EC (STREP "SURF-MOF" NMP4-CT-2006-032109, and FP7 ITN "FUN-MOLS" 212942). | cond-mat.mes-hall |
( 4) and a finite difference approximation, one can express the Hamiltonian of Eq. (1) as, Ĥ = lml ′ m ′ b † lm H lm:l ′ m ′ +v lm:l ′ m ′ + χ l | U imp ( ρ, z m ) | χ l ′ ]δ mm ′ bl ′ m ′ . (5) Here, the first term describes motion along the longitudinal direction for each transverse mode and it's elements are given by, H lm:l ′ m ′ = [δ m,m ′ (ǫ lm + 2t H )-t H (δ m,m ′ +1 +δ m,m ′ -1 )]δ ll ′ with ǫ ml = ǫ l + χ l | U ( ρ, z m )-U L ( ρ) | χ l and the hopping energy of t H = 2 /2m z a 2 (hereafter, we use bold characters to denote a matrix displayed on the basis {χ l ψ m }). The v matrix in Eq. ( 5) accounts for the deviated potential distribution from that of deep source and drain regions. As a result, it gives rise to the hybridization among transverse modes as, v lm:l ′ m ′ = [ χ l | U ( ρ, z m )-U L ( ρ) | χ l ′ (1 -δ l,l ′ )δ mm ′ . | cond-mat.mes-hall |
With use of the lesser Green's function in (9), HMF becomes, HMF = kασ ε kασ n F (ω -∆ k ). (10) In the same way, the self-consistent equations and expectation value of free energy and current are obtained in terms of G < ασ (k, ω). Figure 2 shows the obtained ground state energies of horizontal and 3-fold CO in the θ d type crystal for the case with V p /V c = 0.9. If the electric field is applied more than the critical field, E c , the 3-fold state is stabilized in comparison to the horizontal state. We also calculated the gournd state energies for 0.8 ≤ V p /V c ≤ 1.0. When V p /V c is larger, the 3-fold CO state is more stable than the horizontal CO state. | cond-mat.str-el |
From here on we will only show the dependence of I 1 on ω and suppress the rest such that I 1 (n, n ′ , x ′ 0 ; ω) → I 1 (ω). SdH and Weiss oscillations are found to occur in the magnetoconductivity of both electric and magnetically modulated 2DEG. These transport measurements can be explained without taking into account electron-electron interactions. In order to investigate collective excitations of the system such as magnetoplasmons it is essential to consider electron-electron interactions. Magnetoplasmons arise due to the coherent motion of electrons as a result of electron-electron interactions. Two types of magnetoplasmons can be identified: Those arising from electronic transitions involving different Landau bands (inter Landau band plasmons) and those within a single Landau band (intra Landau band plasmons). | cond-mat.mes-hall |
Finally, aSe films were measured as prepared, and then remeasured immediately after annealing at 60 • C, 80 • C, and finally 100 • C for three hours each. The X-ray measurements were performed at the Hard X-ray Microprobe Analysis (HXMA) beamline of the Canadian Light Source (CLS) [14]. The measurements were performed with a Si(220) crystal monochromator and an Rh-coated harmonic rejection mirror. The measurements were performed in transmission mode. A liquid helium cryostat was used for the low temperature measurements, the temperature was stabilized between 30 K and 45 K. Since no difference was observed between the room temperature and low temperature XANES measurements, the measurements of the progressively annealed pure selenium were conducted at room temperature. A standard pure selenium film was measured concurrently with all spectra as a reference for calibration. | cond-mat.mtrl-sci |
In the Fi gure,we see the i ncreasi ng l oop centershi ftsofthe pi nned l ayeras the tem perature decreases. T hi s i s because the pi nni ng strength of the anti ferrom agneti c FeM n gets l arger at l ower tem perature (i . e. ,the exchange bi as ofthe pi nned l ayer i ncreases as the tem perature decreases). H owever, the l oop centershi ftsofthe free l ayerrem ai n al m ostconstantaccordi ng to the tem perature change.Ifthere were pi nhol es, the free l ayer woul d be di rectl y coupl ed w i th the pi nned l ayer. T hen,as the pi nned l ayer l oop shi fts w i th tem perature, the di rect m agneti c coupl i ng woul d force the free l ayerl oop to shi ft as wel l . T hi s resul t suggests that, accordi ng to Pong et al . | cond-mat.mtrl-sci |
Several theoretical works were done to investigate the effects of ionized-impurity scattering on one-dimensional electron * Electronic address: dahn@uos.ac.kr gas, and revealed their effects on the electronic structure. Most of these studies were for uniformly doped or remote-impurity systems [7,8] and adopted empirical models based on the so-called Büttiker probes for simulating the device. [9,10] The empirical methods are appealing due to relatively simple implementation but the methods often require parameters that need to be adjusted using more rigorous calculations or values from experiments. In this work, we take into account the ionized impurity scattering in simulating the gate-all-around nanowire using non-equilibrium Green's function approach. By averaging the Green's function over impurity configurations and expanding the arising term perturbatively, we treat the impurity scattering within a self-consistent Born approximation and apply the formula to the Si nanowire as realized in Ref. [6]. | cond-mat.mes-hall |
With the application of a magnetic field along c-axis, the antiferromagnetic signal is suppressed and the peak at 3 K diminishes in magnitude. Above 30 kOe, the peak vanishes corroborating a gradual field-driven transformation of the magnetic structure in this field range. When a magnetic field is applied along the ab plane (figure 7(b)), one observes a shift in the peak to higher temperatures, in addition to a decrease in the magnitude of χ . Specific heat The specific heat C p of h-DyMnO 3 measured at zero applied field is presented in figure 8 (inset magnifies the two apparent transitions). The antiferromagnetic transition of the Mn sublattice is evident as a sharp peak at about 57 K. Thus, in h-DyMnO 3 the magnetic ordering transition in the Mn-sublattice appears only in specific heat, but is not detectable in the magnetic susceptibility data. This is in contrast to the properties of hexagonal HoMnO 3 , where T Mn N is detected by peaks in C p and in χ as well [11]. | cond-mat.str-el |
3 we show a comparison of the closing and reopening of the band gap as obtained from calculations with and without self-consistency, in the relevant weak pairing regime (U 0 < 0). The self-consistency does not change the qualitative behavior. In particular, the gap only closes at k x = π/a for the parameters chosen (c.f. inset in Fig. 2) and the self-consistent determination of ∆ only shifts the critical potential δU slightly. In Fig. | cond-mat.mes-hall |
We also showed that a simple model of the tunnel barrier can explain the response behavior well, and can be used to predict that a higher barrier would increase the responsivity. For typical AlO x tunnel barriers, φ 0 ∼ 1 eV [24], which is well below the bulk value closer to 10 eV. Thus if one could find a way to make barriers closer to the bulk values, the responsivity would increase by a factor three. Low temperature measurements would also help in reducing the noise generated by the junction, and would provide more information on the microscopic behavior of AlO x barriers under strain, because information of the charging energy (capacitance) variation could then be measured [24]. We thank Thomas Kühn for useful discussions. This work has been supported by the Academy of Finland under projects 128532 and 118231. | cond-mat.mes-hall |
A promising solution is to apply the periodic boundary condition (PBC) here, so that one can rule out the possibility of dimerization acquired from open boundary condition. Therefore, we have also done some calculation by DMRG for system with PBC. For the system size N = 100 -200 sites, we keep up to m = 3000 states with truncation error smaller than 10 -8 . Finally, we find that both OBC and PBC systems give us consistent results. Observed the possible presence of dimerization, one can readily work out the phase diagram, see Fig. 5. | cond-mat.str-el |
To monitor the corresponding temporal evolution of the QD resonance, we measure the I(V ) curve in a fast acquisition mode (∼ 10 ms) and at regular intervals of time. Following illumination, measurements of the resonant peak in I(V ) in the fast mode are free of telegraph noise and, with time, the peak shifts from 5 mV to higher biases, back towards its voltage position in the dark, see Figure 4b. The I(V ) curve can be restored to its initial "dark" state by applying a short negative bias reset pulse. This causes a short discharging current pulse after which the system is ready to detect the next optical pulse, see Figure 4c. III. DISCUSSION To explain these data, we consider the effect of illumination on an electron tunneling through a single QD. | cond-mat.mes-hall |
We can take a high symmetry grain boundary and add or subtract flaws a distance d apart as shown in figure 8. By the same reasoning as used for the low angle grain boundaries, the energy near the high symmetry grain boundary will have the form E GB = E 0 + µb 4π(1 -ν) |θ -θ 0 | log eα ′ 2π|θ -θ 0 | (6) where E 0 is the energy of the high symmetry grain boundary which occurs at the angle θ 0 and α ′ incorporates the core energy of the flaw within the pattern of flaws. The grain boundary energies for all of the symmetric grain boundaries that we have measured are shown in figure 9. Note that cusps occur at the angles listed in table I. We are able to fit the data for symmetric grain boundaries to a function of the form FIG. 8: Adding a Flaw to a High Symmetry Grain Boundary. | cond-mat.mtrl-sci |
For a symmetric Anderson impurity model, this naturally leads to the manifestly spinrotationally symmetric decoupling (4.3) at strong coupling. With our decoupling scheme, this symmetry is also respected by the truncated flow equations. Our approach still has two major shortcomings: (i) We have not been able to reproduce the exponential dependence Z ∝ exp[-π 2 u 0 /8] of the wave function renormalization factor for U → ∞. Although we have tried several modifications of our approach, the strong fluctuations responsible for an exponential suppression of Z for U → ∞ are apparently not correctly described within our truncation of the exact FRG flow equations. (ii) In order to make progress without using complicated numerical methods, we have made approximations which are only accurate at low energies |ω| ∆. However, a direct numerical solution of our FRG flow equations should give rise to a much better agreement with NRG data at small couplings U and might also correctly reproduce the high-energy behavior of the spectral function. | cond-mat.str-el |
Two additional curves from another dataset are plotted in Fig. 2 (d) for comparison (hollow triangles). We now move to case ii where the magnetic field was perpendicular to the wire. Theory for confined one-dimensional hole systems predicts a strong anisotropy of the g-factor and low values for g 14,18 . Our measurements are consistent with this expectation. This is particularly visible in Fig. | cond-mat.mes-hall |
We start with the dots in the ground state |0h0e at δε > ∆. Adiabatically changing δε to δε < -∆, we drive the levels through the anticrossing and end up with the state |2h2e which relaxes to |0h0e via the emission of an entangled pair of photons. Changing δε adiabatically back to the original situation, an additional pair of photons is produced leading to a total of two photon pairs per cycle in the ε h,e parameter space. Ψ 1 after the first emission is a linear superposition of the photon being in state |+ 1 or |-1 with the same amplitude for both. After the first photon emission, the state of the system (dot and photon) reads {|0 h 0 e , |0 h 2 e , |2 h 2 e }. |↓ h |- |+ |- |0 h 0 e |- |+ |2 h 2 e |↓ h ↓ e |↑ h ↑ e |- |+ |2 h 2 e |↓ h ↓ e |↑ h ↑ e kick-out electron (a) (b) |↑ h |+V h V e V L V R e h Ψ 1 ∝ |↓ h ↓ e |+ 1 + |↑ h ↑ e |-1 . | cond-mat.mes-hall |
Obviously, f = ln for Ω = Ω log , but it remains an open question for general Ω whether a strain measure function exists such that Ω = Ω f . 5. Eulerian conjugate stress-strain pairs 5A. Arbitrary strain and corotational rate. It was noted in the Introduction that the concept of work-conjugate stress-strain pairs is more complicated for Eulerian quantities owing to the fact that the connection between the strain rate and the stretching tensor is not evident a priori. This issue was resolved by Lehmann and Liang [8] who introduced the notion that the Eulerian pair τ and e are defined to be conjugate ifẇ = tr Qτ Q t QeQ t ,(5-1) for some rotation Q. | cond-mat.mtrl-sci |
1, if there are sufficient number of electrons [9] (n el > 0.1 • 10 15 m -2 ). Consider a case where the q component approaches to zero, then the external (damped) potential is well screened, hence the long range part of the disorder potential. Whereas, the short range part remain unaffected, i.e. high q Fourier components. Now we turn our attention to the second type of impurities considered, the Gaussian ones. As well known, the Fourier transform of a Gaussian is also of the form of a Gaussian, therefore, similar arguments also hold for this kind of impurity. | cond-mat.mes-hall |
We argue that it is likely the former. First for all CMR samples the DE coupled FM sites are easily aligned at 0.4T and the same should apply for the main FM domains for slightly lower Ca concentrations. However as the Ca concentration decreases there will be an increasing number of tiny regions with no Ca, and hence no local holes. These nanoscale regions, of order a few unit cells, will possess mostly AFM coupling between a few Mn sites (but no long range AFM order) since that is the dominant magnetic coupling when no holes are present. In addition there will be Mn spins on the boundary between these AFM-coupled nanoclusters and the large FM clusters. Such spins may be frustrated -having AFM coupling to the nano AFM cluster but FM coupling to the FM clusters as a result of occasional hole hopping onto these sites. | cond-mat.str-el |
This coherent control substantially shapes the nonclassical light emission and is also a key capability in successful approaches to deterministic single photon generation with atoms 18,19 and ions 20 . The indirect excitation via continuum or quasi-resonant states in QD-based sources 7 can efficiently pump the emitters, but not coherently. True resonant excitation and collection of the light has not been demonstrated yet, although it is well-known that quantum interference 21 and Rabi oscillations [12][13][14]22,23 can be achieved in QDs. Extracting the single photons generated by an individual QD that is resonantly driven by strong laser fields has been challenging because of sizable laser scattering in the host crystal that blinds the single photon detection. Thus, so far, QD states have been either probed coherently using non-linear optical techniques without collecting the single photon response 12,13,24 , or they have been pumped incoherently via an excited state [8][9][10][11] . Based on a new approach to this problem that utilizes a micro-cavity, we show here that a single QD can in fact be coherently driven in resonance fluorescence. | cond-mat.mes-hall |
The general expression for the net current 28 can be rewritten by using pseudofermion and slave boson Green's functions 16 . The final expression for the current is I(t) = 2 ΓIm(i t -∞ dt 1 (G R pseu (t, t 1 )B < (t 1 , t) + G < pseu (t, t 1 )B R (t 1 , t)) (f L (t -t 1 ) -f R (t -t 1 ))),(2) where f L (t-t 1 ) and f R (t-t 1 ) are the convolution of the density of states function with the Fermi-Dirac distribution 16 . The conductance G is given by the current divided by the bias voltage. A quantum coherent many-body state called the Kondo effect emerges when the dot level is situated below the Fermi energy at sufficiently low temperatures. A spin singlet is formed from the free spin localized in the dot and the Fermi sea of electrons in the contacts. Its manifestation is a sharp resonance pinned to the Fermi levels of the contacts in the dot density of states. | cond-mat.mes-hall |
DISTORTED WINDMILL LATTICE Mean field calculations 4 on the distorted windmill lattice also revealed a partially ordered ground state, with a line of degenerate wavevectors along the (qqq) direction. In this paper we show that Monte Carlo simulations of the AF Heisenberg model on this lattice find an ordered ground state as on the trillium lattice. However, the physical origin of the two phase transitions differs. While the ordering in the trillium lattice can be understood by minimization, being the result of energetically different states, the ordering in the distorted windmill lattice can- not. Rather, it must proceed via an order by disorder mechanism. As on the trillium lattice it is interesting to ask whether the partial ordering features obtained by large-N (or mean field) theory can be found at finite T, where one might expect the spins to fluctuate strongly. | cond-mat.str-el |
. , a κ ), (B1) where χ is the degenerate part of the wave function (further discussed in Sec. IV B), and where we used that a † D a = a • † Ka • . This result is nothing else than Eq. (3.13) with the inverses of D and K replaced by the pseudoinverses. Substituting the density fluctuations δρ α for the gauge fields a • α using the constraint (3.3) yields exactly (3.16) by virtue of the property of K that K K K = K. This result is exactly equal to the steps we followed before, but only with K -1 replaced by K. Therefore, the ground state (3.16) found for the case of strictly positive eigenvalues is also valid if there are zero eigenvalues. | cond-mat.mes-hall |
2. Even though all spinon momenta of given spin orientation are distinct from one another, their exclusion statistics is semionic, not fermionic. This is demonstrated by applying the defining relation [16] ∆d σ = - σ ′ =± g σσ ′ ∆N σ ′ (3.11) for the statistical interaction coefficients g σσ ′ to the situation described by (3.10), taking into account that the number of available momentum states ∆d σ with N s = N + + N -spinons already present is affected both by the next particle added and by the shifting bounds. The result is g σσ ′ = 1/2 for all combinations of spinon spin orientations. With all these spinon properties in the XX model established we are ready to analyze their thermodynamics via TBA from the bottom up. This is the theme of Sec. | cond-mat.str-el |
Two kind of anomalies appear often in strongly correlated systems. The first situation is determined by a negative compressibility region. A notorious example is the uniform electron gas (EG) at low density [6] but this feature appears in several other models [7,8,9,10,11,12,13,14,15] including neutron star matter [5]. The other possibility is that the inverse electronic compressibility has a point with a Dirac-delta-like negative divergence at some density n c . This happens when the free energies of two states which are separated by a barrier, cross each other leading to a cusp singularity. An example is also provided by the EG. | cond-mat.str-el |
5), the singular behavior of χ S ( x) is regulated purely by finite temperature. Notice that f (T ) depends on irrelevant operators of the QCP; we conclude that the infrared divergence is regulated by irrelevant operators, including the gauge-like coupling between smectic and the nematic Goldstone mode, which implies that this gauge-like coupling is a dangerous irrelevant term near the QCP and become relevant at finite T . This conclusion agrees with the theory of the classical nematic-smectic transition at finite T , where the gauge coupling is known to be relevant. The boundary between the nematic and the quantum critical regime is determined by the condition f (T ) ∼ ∆ S , which is T ∼ ∆ S . On the smectic side, the equal-time correlation function of the smectic Goldstone mode fluctuations, i.e. the transverse smectic susceptibility χ S ⊥ ( q) (where q is the momentum measured from the ordering wave vector Q S of the smectic phase) is χ S ⊥ ( q) ≡ | Φ| 2 φ S ( q, τ )φ S (-q, τ ) ret = 1 βg 2 S N S (0) ωn 1 B ω 2 n k F g S | Φ| + A |ω n | κ 1 κ -1 2 q 4 x + q 2 y k F g S | Φ| + κ 1 q 4 x + κ 2 q 2 y (8.3) The most divergent term in χ S ⊥ ( q) has the asymptotic behavior T /(κ 1 q 4 x + κ 2 q 2 y ), which implies that the Fourier transform of χ S ⊥ ( q), the transverse susceptibility of the smectic Goldstone mode, is infrared divergent at finite T . | cond-mat.str-el |
Thus we come to the conclusion that the state closest to the center of excited subband is the best candidate for the transport state. If the QW number N is even, it will be one of states {|N/2 + 1 , |N/2 + 2 } of the central doublet, whereas for odd N this is the unpaired state |1 + (N + 1)/2 . The plots of the maximum values of p 1 , see Fig. 7, demonstrate quasiperiodic behavior of the transfer probabilities on the field energy for each state of the pair {|4 , |5 } for N=6. We observe that the state |4 displays better transport characteristics in comparison with the state |5 . The maximum of the transfer probability max(p 1 ) ≈ 0.999 for ω = ω 0, 4 is attained at quite large field energy ε f ield = 0.083, thus allowing for rapid electron transfer (about 45 ps for the GaAs parameters). | cond-mat.mes-hall |
In samples with x = 0.20 and 0.30, m * reaches values typical of small polarons, namely of charges which are assumed to be self-trapped within a single cell. In Table I, the effective mass in the Ga-free La-Sr manganite is somewhat smaller than those reported (at room temperature only) for a single crystal of La 0.7 Sr 0.3 MnO 3 in Ref. 31 and for a thin film of the Mid-and near-infrared bands The Drude-Lorentz fit to σ(ω) allows one to identify in all spectra two bands. The first one, the MIR band, shows at all doping a characteristic behavior with temperature: its peak frequency softens as the insulator-tometal (IM) transition is approached, to remain stable at about 2000 cm -1 in the whole metallic phase. The other one, the NIR band, is instead centered at about 10000 cm -1 in all samples. The band intensities S 2 MIR and S 2 N IR in Eq. | cond-mat.str-el |
One clearly sees, over a wide range of magnetic fields (B = 18 to 34 T), two distinct magnetopolaron branches separated by as much as 11 meV (∼ 0.4ω LO ) at resonance (Fig. 12). The theoretical curves show the results of calculations for coupling to the LO phonons in bulk (3D), sheet (2D) and after correction for the quasi-2D character of the system using α = 0.29. The agreement between theory and experiment is reasonable for the 3D case, but better for the quasi-2D system, if the finite spatial extent of the 2D electron gas in the symmetric planar layer is taken into account. Cyclotron-resonance measurements performed on semiconductor quantum wells with high electron density [93,94] reveal anticrossing near the TO-phonon frequency rather than near the LO-phonon frequency. In Ref. | cond-mat.mes-hall |
In the former case, heat diffuses predominantly along the ladder direction, while little dynamics is seen in the latter case. These results support the notion of anisotropic heat transport in this material, 34 due to the contribution of magnetic excitations. Besides thermal transport measurements, questions of diffusive versus ballistic transport have been experimentally probed in nuclear magnetic resonance 36,37 (see Ref. 38 for related theoretical work) as well as muon spin resonance experiments. 39 In a more recent development, transport properties of low-dimensional ultra-cold atom gases have gained attention as well, with experiments focusing on the detection of Anderson localization. 40,41 Interacting two-component Bose gases in optical lattices have been suggested to potentially realize spin-1/2 Hamiltonians. | cond-mat.str-el |
2). A simple physical picture for the situation described above is: In a basis where the Hamiltonian for the isolated dot is diagonal, the bare dot level energy is split by the magnetic field, and for non-interacting electrons the density of states has peaks at the two single-particle energies at ∓B, see Fig. 3. The widths of the two peaks depend on the angle φ, and for fully polarized leads they are proportional to cos 2 (φ/2) or sin 2 (φ/2), respectively. For φ = 0 and φ = π one of the peaks is infinitely narrow and 3: Schematic energy spectrum in the linear conductance regime for the non-interacting case. The bare resonant level is split due to magnetic field, and the angle between the magnetizations of the leads and the applied magnetic field is denoted by φ. | cond-mat.mes-hall |
In a MFEG, due to the negative interacting energy E int = -A 3D n 1/4 favoring high electron density, the density is expected to further reduce at the edges of the attractive potential and increase at the center of the well. In a repulsive potential the opposite should occur. C. Density limits In this section we will derive approximate expressions for the upper and lower density limits over which the many-flavor limit applies, these will be used to check the theory against numerical results [35] and to predict a lower bound on the number of flavors required for the theory to apply. To find the upper density limit one notes that Eqn. (13) implies that an acceptable upper limit to the momentum integral would scale as q = α( a * -1/4 0 )n 1/4 , the constant α ≈ 4 was determined numerically and was chosen to give the q upper limit on the integral that recovered 95 % of the interacting energy. Additionally, the two regions of integration defined by the Fermi Dirac distributions in Eqn. | cond-mat.str-el |
Traditionally 6 one chooses the orthonormal basis {|ψ ⊥ } = { Ô-1/2 |ψ } = { Ô1/2 |ψ } and diagonalizes the matrix (H H ) ϕ ⊥ ,ψ ⊥ = ϕ | Ô1/2 ĤH Ô-1/2 |ψ = (H eff ) ϕ,ψ ,(13)where Ĥeff = Ô1/2 ĤH Ô-1/2 = Ô-1/2 Ĥ Ô-1/2 . (14) Note that on the one hand-side Ĥeff is expressed in terms of the matrices O ϕ,ψ and H ϕ,ψ , which are both symmetric and easily calculated in terms of the loop structure formed by the NNVB states. On the other hand-side Ĥeff arises from the Heisenberg Hamiltonian ĤH by a similarity transformation and therefore both operators are equivalent. However, working with Ĥeff rather than with ĤH allows for using the nonorthogonal NNVB basis without any explicit knowledge of its dual basis, which is the aim of this transformation. In the original scheme 6,17 the dual basis was omitted, but the definition of the effective Hamiltonian is the same in terms of Ô and Ĥ. In order to achieve this, one had to introduce some generalized eigenvalue problem, however missing the fact that Ĥ and ĤH are indeed distinct operators (i.e. | cond-mat.str-el |
In close-packed metal particle arrays individual electrons are unlikely to move further than to their nearest neighbor, at least at temperatures above 1K, in contrast to hopping between sparse doping sites. Comparison of transport measurements taken in-plane and perpendicular to the arrays (Figs. 3 and4) indicate that typical cotunnel events in Au nanoparticle arrays involve up to 4 electrons (and thus reach net distances corresponding to 4 tunnel junctions in a row) at 10K (Fig. 6). The absence of clean evidence for magnetoresistive effects up to the highest applied magnetic fields (10T) further supports the picture of inelastic cotunneling as the mechanism for charge transport inside the Coulomb blockade regime. VI. | cond-mat.mes-hall |
The boundary conditions for (B1) are obtained from the boundary conditions for ( 14): f j,+ (0, t) = v j , f j,-(L, t) = 0. We denote f = (f 1,+ , . . . , f N,+ , f 1,-, . . | cond-mat.mtrl-sci |
Dzyaloshinskii reminded us that the Luttinger theorem, in fact, does not relate the particle density to the FS volume [14] (see also [15]). The latter relation exists only in weakly correlated systems where the electron self energy Σ(ω, k) does not have singularities at zero frequency. In strongly correlated systems where Σ(ω = 0, k) goes to infinity at certain surface in momentum space, the volume inside of this surface contributes to the particle density. A Mott insulator is an obvious example of an insulating state with a half filled Brillouin zone which seemingly violates the Luttinger theorem (in its traditional, or rather, according to Dzyaloshinskii, confused understanding). This idea was illustrated by Essler and Tsvelik [16] who considered a strongly correlated model which allowed a controlled approximation. This was a model of Hubbard chains coupled by a long range tunneling (the inverse tunneling radius κ served as a small parameter of the theory). | cond-mat.str-el |
serge.florens@grenoble.cnrs.fr Introduction. -The Mott metal-insulator transition, wherein electronic waves are localized by short-range electron-electron interactions (see [1] for a review), is one of the most complex phenomenon observed in strongly correlated electronic systems. Even though the appearance of a Mott gap is purely driven by the charge degrees of freedom, it is expected that magnetic fluctuations play a very crucial role in determining the true nature of this phase transition. In the paramagnetic Mott insulator, local moments are indeed well defined objects after their creation at high temperature (at a scale set by the local Coulomb interaction) and before their ultimate antiferromagnetic ordering at the Néel temperature, offering a window in which complex behavior of the spin excitations is yet to be clearly understood. Experimentally, the simplest situation in this respect occurs when the low-temperature magnetic ordering is first order, as in the case of Cr-doped V 2 O 3 . Since the magnetic correlations are expected to be weak in this case, many predictions can be made from a single-site approach like the Dynamical Mean Field Theory (DMFT) [2], where local moments are described as freely fluctu-ating from each other in the insulating state. | cond-mat.str-el |
The purpose of this paper is to point out that this minimum conductivity in ballistic samples arises from contact induced states and depends strongly on the structure and configuration (two-vs. four-terminal) used. While actual samples are probably not ballistic and involve scattering processes, our results show that these contact induced effects need to be taken into account in interpreting experiments. Contact induced states are similar to the well-known metal induced gap states (MIGS) in metal-semiconductor Schottky junctions which typically penetrate only a few atomic lengths into the semiconductor, while the depth of penetration decreases with increasing bandgap. However, in graphene we find that these states penetrate a much longer distance of the order of the contacts' width, which seems reasonable since the graphene acts like a semiconductor with a small gap that decreases with increasing W. In this paper, we will present model calculations showing how these contact induced states can help understand many experimental measurements of minimum conductivity in different multi-probe configurations. Model: The basic theoretical model presented here is based on the general non-equilibrium Green's function (NEGF) approach, which has been described elsewhere in detail 19 . The structure is partitioned into channel and contact regions with the channel properties described by a tightbinding Hamiltonian (H) appropriate for graphene with a single π-orbital for each carbon atom: all elements of [H] are equal to zero except for nearest neighbors for which ; 2.71mn H t t = -= eV. | cond-mat.mes-hall |
V y is in fact zero, as it breaks time-reversal symmetry. However, it should be noted that V y = 0 does not hold for the renormalized vertex, since higher order terms are not necessarily local. We will also assume V z ≫ V x as proposed previously 23 . In fact, this system is very similar to the Kondo model, with the two orbitals d xz and d yz representing the up and down-spin states on the magnetic impurity. We are going to perform a similar scaling analysis following Ref. 23. | cond-mat.str-el |
Regular close packing behavior must occur at f c = π/4. Distributions Field distributions are studied as in Sec. III B 4. Quite similar features are observed, summarized hereafter, along with some differences. The density P εPS is non-symmetric, and supported by the interval [g B (a), g B (0)]. A rescaling similar to (61), using µ instead of λ, provides P σPS , plotted in Fig. | cond-mat.mtrl-sci |
25 As heterostructure, with gate insulation obtained with a 60 nm-thick layer of hydrogen silsesquioxane (HSQ) 17 . We report a crossover from a singlet to a triplet spin state that occurs with an inplane magnetic field of 0.7 T. The observation that singlet and triplet spin states can be realized and manipulated in these lateral QDs at moderate fields offers promising venues for their exploitation in studies of spin physics and for quantum information processing. We employ two dimensional electron gases (2DEGs) confined in metamorphic In .75 Ga .25 As/ In .75 Al . 25 As heterostructures grown on undoped (001) GaAs substrates by solid-source molecular beam epitaxy 18,19 . A ≈1 µmthick In x Al 1-x As "virtual crystal" with stepwise increasing indium concentration (x = 0.15 to 0.75) is grown between the GaAs substrate and the active region in order to match the GaAs lattice constant to that of In .75 Ga .25 As and In .75 Al .25 As. Our heterostructure is designed as follows: a Si δ-doped layer is followed by a 11 nm In .75 Al . | cond-mat.mes-hall |
In contrast the curve of the approximant C Ag 9 shows a completely different behavior, because here the wave packet is located in a different local environment in the beginning. Further, by repeating similar numerical experiments for various values of the coupling strength v, we found that the peak structure becomes less distinct with in- creasing v. For the silver mean model we observed that it persists up to v ≈ 0.4. For this value of v the widths of the energy bands become smaller than the gaps between them, which in turn means that the impurityrelated eigenstates coincide very often with the energy bands and thus peaks merge so that almost no valleys occur. These results show that in quasiperiodic quantum wires one can strongly influence the long-range electronic transport properties by inducing local perturbations at different positions and of various strengths. The characteristics are related to the nature of the eigenstates, which spread only across one type of cluster in the limit v → 0. Knowing the structure of the energy bands and of the eigenstates allows one to design quasiperiodic chains with impurities that can act as sort of control gates. | cond-mat.mes-hall |
The reason for the difference in ν c between these two isostructural compounds is unclear, but may reflect the extreme sensitivity of the electronic structure to the outof-plane atoms. Clearly, both the magnetic order parameter, given by H int (T ), and a measure of the structural distortion, given by ∆ν c (T ) = |ν c (T ) -ν c (T N )|, are discontinuous at T N , indicating the first-order nature of the transition in CaFe 2 As 2 . Upon warming the sample from the ordered state, the paramagnetic signal is recovered at 168 K, revealing a thermal hysteresis of 1 K in excellent agreement with results from neutron diffraction [25]. We emphasize that there is no temperature range in which we observe either the magnetic or structural order parameter finite and the other one zero, indicating that both are intimately related. The temperature dependence of H int observed in Fig. 3 is remarkably close to the temperature dependence of the ordered moment that develops below a first order magnetic transition in isostructural SrFe 2 As 2 [24]. | cond-mat.str-el |
As usual, 26,27,28 we parametrize this infinitesimal neighborhood with the help of infinitesimal matrices δU X , with symmetry properties dictated by those of M X , such that the measure is simply the product of the independent matrix elements of δU X . For class D, the manifold M D is isomorphic to SO(N ), with U = νOν † , O ∈ SO(N ), and ν = 1 2 1 + i 1 -i 1 -i 1 + i , ν 2 = Σ 1 . (2) It might be worthwhile to note here, that solely from the unitarity of U and the symmetry U = Σ 1 U * Σ 1 only det U = ±1 follows. In Ref. 10, the manifold M D was identified through the exponentiation of the Bogoliubovde Gennes Hamiltonian, which leads to det U = 1 due to the mirror symmetry of the energy levels around zero. (The energies are measured relative to the Fermi level.) | cond-mat.mes-hall |
In one configuration, we can change the magnitude of transferred momentum whilst keeping its direction fixed and, in the other configuration, we can change the direction of transferred momentum, whilst keeping its magnitude fixed. The first configuration is shown in Fig. 2(a). The magnitude of transferred momentum ∆q can be changed, as ∆q ≃ 2|q 1 | cos θ, by means of changing the value of θ without changing its direction (1 -1 -1). The incident photon tuned at the Si K edge excites the 1s electron to the X-point ( ξ 1 = 0.0 eV). The corresponding hole in the final state is created according to the law of crystal momentum conservation (Eq. | cond-mat.mtrl-sci |
[15] have shown using a Hartree-Fock approximation with U as a free parameter that calculated and experimental values of the local magnetic moments and the direction of the easy axes agree well for U =0.76 eV. In this work we fit an equilibrium volume of US by varying U -parameter. The result is shown on the Fig. 1. The calculated equilibrium volume is close to the experimental one at the value of U =1.25 eV and it will be used for the uranium compounds under investigation. Another way to calculate the value of screened Coulomb interaction from first principles is the supercell procedure [14] that takes into account s-, p-and dscreening channels. | cond-mat.str-el |
As a toy model, we may assume for instance that the probability of adsorption varies linearly with z through the following criterion: we allow the incorporation ifz > h 0 + h i ,(4) where h i is a random number varying from 0 to z max -h 0 , and h 0 ∈ [-z max , z max ] is the lower cut-off of the distribution, with z max as the maximum height. The choice of another distribution function such as, for instance, a step function, θ(z -h 0 ), or some other distribution with a finite tail for z < h 0 , should lead to similar conclusions regarding the physical properties, as far as the probability of adsorption on the top of the hills is large compared to the probability adsorption in the valleys. If the number of atoms which are available for adsorption is fixed, by varying h 0 we obtain different incorporation fractions, as shown in Figures 1 (b),(c). Notice that for a given realization of ripples, as one increases the adatom concentration, one obtains a percolative structure. At low coverage densities the adatoms form clusters. Hence, ripples naturally lead to clustering. | cond-mat.str-el |
For the band structures of the metallic boron nanotubes, there are several bands in the vicinity of the Fermi level, which ensures a large carrier density. Take the (0, 8) tube for example, the bands are degenerate at the Γ-point and these bands are highly dispersive along the Γ-X axis, implying that the effective mass of the charge carriers should be very small, leading to high mobility and high conductivity. Hence the BNTs could have potential as the metallic interconnects in electronic devices. Conclusion In summary, the structural and electronic properties of a novel boron sheet and the related boron nanotubes have been investigated at the DFT-GGA level. The new class of boron sheet is sparser than the α-B sheet and after relaxation it remains flat and metallic. Within the scope of our research, except for the (8, 0) thin tube, all the nanotubes rolled from this sheet are metallic independent of their chirality. | cond-mat.mtrl-sci |
4. The DOS corresponding to the up and down spin channels are shown along regular and reversed y-axis, respectively. No hard gap is observed for U as large as 6 eV. The calculated results reveal many interesting features. Firstly, the near negligible Ir 5d contributions observed below -2 eV energy range in LSDA results (see Fig. 1) becomes more and more significant with the increase in U. | cond-mat.str-el |
schemes we consider the nearest-neighbor magnetic exchange constant J 1 H Heisenberg = J 1 i,j S i S j . (12) In leading perturbation order one obtains J The results are shown for various truncation schemes, where 'min' denotes the minimal model in which only the Heisenberg exchange term is kept in addition to the terms present in the initial Hamiltonian. The NN truncation represents a nearest-neighbor calculation defined by the maximum extension 1. This calculation reproduces the second order perturbative result for J 1 . The plaquette calculation contains all terms which fit on the clusters shown in Fig. 5. | cond-mat.str-el |
Although less precise (though, as will be shown below, many results are in very good agreeement with the results obtained with the ab-initio based computations), this model is less expensive and was therefore used to study CNTs of larger diameters. We compare in Fig. 6 phonon dispersions for (5,5) armchair tubes obtained with the empirical model and with ab-initio computed IFC. The agreement is fair enough, though the number of acoustic modes is not correct, and the lowest modes do not have a quadratic dispersion relation as is the case for real nanotubes. Despite that, the thermal transport results for both models should agree quantitatively as shown by a comparison of ballistic conductances computed with the empirical model used here and ab-initio reference results. Actually, a more intrinsic property is the ballistic thermal conductance divided by the cross sectional surface of the material g/A (this is the so-called ballistic conductivity per unit length, compare with Eq. | cond-mat.mtrl-sci |
22,23 Although our findings relied on the Yb 3+ ESR results to witness the Yb 3+ rattling mode, they suggest that the R ions in other skutterudites and clathrate compounds may be also rattling in an analogous form as long as they are inside an oversized cage. However, it may not be always observable in an ESR experiment. We believe that the evidence for predominant off-center rattling Yb 3+ ions in these skuterudites is a result that could justify the existence of Einstein oscillators and help to understand the low thermal conductivity and the strongly correlated phenomena exhibited by these type of materials. VI. ACKNOWLEDGMENTS We thank FAPESP-SP and CNPq for financial support. PS is supported by DOE grant No. | cond-mat.str-el |
1 and2. Surfaces that satisfy the ECM are generally semiconducting, while those that do not may be metallic. The degree to which a given surface satisfies the ECM can be measured by the excess electron count, ∆ν, which we define here as the difference between the number of available electrons and the number required to satisfy the ECM, per (1×1) surface unit cell. Excess electron counts for the structural models in Figs. 1 and 2 are tabulated in Table I. To compare the surface energies of reconstruction models with different periodicities and stoichiometries, we consider the surface energy per unit area, γ = E surf /A = (E tot -n III µ ′ III -n V µ ′ V )/A,(1) where E tot is the total energy of a reconstructed surface, of area A, containing n III group-III and n V group-V adatoms in excess with respect to the bulk-truncated, Sb-terminated AlSb(001) or GaSb(001). | cond-mat.mtrl-sci |
In Fig. 2, this optimized φ CuAg (r) exhibits a wider well than the force-matched pair potential. Fig. 3 shows that the optimized φ CuCu (r) gives shorter and weaker bonding between Cu atoms on the Ag surface than in the Mishin EAM bulk Cu. We scale the original Mishin φ CuCu (r) in 1% steps from 80% to 120%, and translate in 0.01Å steps from -0.15Å to 0.15Å; potentials with Cu lattice parameter outside of ±5% of the bulk value are removed. A 82% scaling and a -0.13Å translation reproduces all relative energy differences with a final 0.5meV range optimization of the φ CuAg (r). | cond-mat.mtrl-sci |
The probability distribution P ph can be obtained by noticing that the corresponding phase operator iλ(b-b + ) is analogous to that of an electromagnetic LC circuit with resonant frequency 1 √ LC = ω [9] (this result can also be obtained directly by tracing out the phonon degrees of freedom in eq.4): P ph (E) = e -λ2 coth( β ω 2 ) k δ(E -k ω 0 )e kβ ω/2 I k ( λ2 sinh(β ω/2) ) (6) where β is the inverse of the phonon temperature T ph . Using Eqs. (5 and 6) and standard expressions for current as a function of P (E) [9] it is possible to compute the current through the sample. For example in the case of symmetric contacts (α = 1/2) and constant density of states, dI/dV = 1 RT dE (P (eV /2 -E) + P (-eV /2 -E)) where R T is the tunnel resistance of each contact. Note the similarity with the DCB in a tunnel junction in series with a LC resonator [10]. The case of an energy dependent density of states , and of asymmetric contacts can readily be included by straightforward generalization of this expression according to eq.1. | cond-mat.mes-hall |
The two curves H c vs. t/w seem to converge with two different slopes for t/w → 0 (increasing w). At t/w > 0.01 the slope of the curve flattens and a plateau or inflection point is visible at t/w ≈ 0.017. For greater values, t/w > 0.02, the curve shows an increasing slope for t/w → ∞ (decreasing w). Considering the w-t-phase diagram for Py nanowires [9] one expects a TDW-VDW transition at w ≈ 450 nm for a base wire thickness, as in our case, of t = 10 nm and hence t/w ≈ 0.022. This value matches relatively well with the inflection point of the curve. Obviously it marks a crossover between two different regimes of domain wall propagation through the gradient. | cond-mat.mtrl-sci |
. . , H (2) = ij 1 4 (θ i -θ j ) 2 + σ i σ j + i σ 2 i ; (2a) H (3) dom = α η α H (3,α) dom ,(2b) H (4) dom = ij 1 16 (σ 2 i -σ 2 j ) 2 . (2c) In Eq. (2b), α indexes the center of each triangle, and H (3,α) dom ≡ √ 3 4 3 m=1 σ 2 αm (θ α,m+1 -θ α,m-1 ) . (3) From here on, I use "αm" (m = 1, 2, 3) to denote the site on triangle α in sublattice m, as an alias for the site index "i"; the index m is taken modulo 3 (in expressions like "m + 1") and runs counterclockwise around the triangles whose centers are even sites on the honeycomb lattice of triangle centers. | cond-mat.str-el |
Near T C , one expects an interplay between the large thermal fluctuations and the nonequilibrium spin transfer torque. Generally speaking, theories of critical phenomena in out-of-equilibrium systems have only recently been developed [5,6], and there remain many open questions on this topic. Even far from the Curie temperature, temperature plays an important role in quantitatively analyzing the dependence of the magnetic orientation on the applied field and applied current. The effect of finite temperature on spin dynamics in the presence of spin transfer torque has been modeled the macrospin approximation (fixed magnetization length) by adding a Slonczewski torque to the Langevin equation describing the stochastic spin dynamics [7,8], and by solving the Fokker-Planck equation with the spin transfer torque term added to the deterministic dynamics [9]. The Keldysh formalism provides a formal derivation of the stochastic equation of motion [10] for the non-equilibrium (i.e., current-carrying) system for a single spin of fixed magnitude. These treatments successfully describe the thermal characteristics of nanomagnets under the action of spin torques, such as dwell times and other details of thermally activated switching. | cond-mat.mes-hall |
splitting ∆E h s,p when both contributions are introduced in the TB approach. Since the hole p shell is no longer degenerate, we average over the single-particle energies of the states ψ h 1 and ψ h 2 . The energy splitting ∆E h s,p with SO coupling and CF splitting is also shown in Fig. 4. From the comparison with ∆E h s,p in the absence of these contributions, we find that the SO coupling and the CF splitting have only a negligible effect on the energy splitting ∆E h s,p . Furthermore, the ordering of the first three bound hole states is unaffected by SO coupling and CF splitting. | cond-mat.mtrl-sci |
(4) in the following way. The total shot noise P 12 comprises contributions from particles arriving at contacts at any time during the driving period T . Since we are interested in the contribution of electrons, we restrict the integral over t (t ′ ) to the interval τ m around t L (t R ). = -R C T C T l L T r L T l R T r R and δN B 12 = -R C T C R l L R r L R l R R r R are due to the capacitors A and B alone, and the correlation contribution, δN AB 12 = 2R C T C γ L γ R L(∆t e -∆τ ) cos (Φ L -Φ R ) ,(8) depends on the wave-packet overlap L(∆t e -∆τ ). To test a BI [8,23] we use the four joint probabilities to detect one electron at the contacts 1 (or 3) at the time t L and another electron at the contacts 2 (or 4) at time t R . For the normalized correlation function, E = (N 12 + N 34 -N 14 -N 32 ) / (N 12 + N 34 + N 14 + N 32 ), we find, E = L(∆t e -∆τ ) cos (Φ L -Φ R ), if the transmissions at all the QPCs are 1/2. | cond-mat.mes-hall |
The phenomenon of quantized pumping can be sustained in high fields in contrast to the conceptually related charge pumps using surface acoustic waves [16], allowing spin polarized quantized charge pumping. So far we have tested and observed quantized pumping in such devices up to fields of 16 T. The magnetic field as an additional control parameter makes single-parameter pumps in AlGaAs/GaAs heterostructures promising candidates for an accurately quantized, large-current source as needed for fundamental experiments in metrology and quantum electronics. Acknowledgments The research conducted within this EURAMET joint research project has received funding from the European Community's Seventh Framework Programme, ER-ANET Plus, under Grant Agreement No. 217257. Assistance with device fabrication from Th. Weimann, P. Hinze and H. Marx and helpful discussions with L. Schweitzer are greatly acknowledged. | cond-mat.mes-hall |
4). The lower inset of Fig. 4 shows clearly that the decrease we observe in K(T ) of K 2 CuP 2 O 7 is much more pronounced . The susceptibility of a 1D S = 1 2 HAF at T = 0 is exactly known 26 , χ(T = 0) = g 2 µ 2 B kB (J1)π 2 . Then, K(T ) at zero temperature can be written as K theo (T = 0) = K 0 + A hf g 2 µB kB (J1) × 1 π 2 . Using the parameters (K 0 , A hf , g, and J 1 ) determined from our K(T ) analysis, K theo (T = 0) was calculated to be 260 ppm. | cond-mat.str-el |
Given Eq. ( 11), one may attempt to solve Eqs. ( 1), ( 5), ( 8) and ( 9). Due to the rotational symmetry in the plane of the 2DEG, the azimuthal integration in Eq. ( 9) yields U m,l m ′ ,n ′ (0) ∝ δ m+m ′ ,l+l ′ , whereupon we obtain U(0) = U ↑↑ ↑↑ 0 0 0 0 U ↑↑ ↓↓ U ↑↓ ↓↑ 0 0 U ↓↑ ↑↓ U ↓↓ ↑↑ 0 0 0 0 U ↓↓ ↓↓ (12) in the {|m 1 , m 2 } = {| ↑↑ , | ↑↓ , | ↓↑ , | ↓↓ } basis. Assuming that the band splitting is small (λk F , h z << ǫ F ), the matrix elements of Eq. | cond-mat.mtrl-sci |
The GB begins to deform and settles in a configuration where the applied forces equate the restoring elastic tensions deriving form the term [1] above. Given the particular geometry of our problem, the GB profile is given by the displacement field η which satisfiesδE δη n = sbτ (r),(2) where the restoring force on the right-hand side is calculated as the (variational in the continuum case) derivative of the elastic energy functional in (1). Carrying on the above calculation, Equation ( 2) can be rewritten as a simple linear non-homogeneous problem, in the formm V nm η m = sbτ (r),(3)by defining V nm = -K 1 (n-m) 2 , n = m K k =n 1 (n-k) 2 , n = m. (4) A simple case -uniform stress Let us first review the simple case in which the applied stress is uniform, i.e. τ (r) = τ [23]. In order to solve Equation (3), which can be rewritten in operatorial terms asV H = sbτ ⇔ H = V -1 sbτ,(5) one needs to diagonalize the matrix of interactions V nm . This corresponds to solving the eigenvalue problemm V nm γ l m = λ l γ l n ,(6) where we have called λ l each of the N eigenvalues and γ l n each component of the eigenvector corresponding to λ l . | cond-mat.mtrl-sci |
Pt produces strong spin-flipping in Cu [18] with only modest elastic scattering [17]. For those samples, the inverse critical switching currents, (1/I + ) or (1/I -), were found to be directly proportional to the change in resistance ΔR = R(AP) -R(P) between the states with moments of the two F-layers anti-parallel (AP) or parallel (P) to each other [20]. As linear fits to the data of ref. [20] were compatible with zero intercept for both positive and negative (1/I), we recast the data into the form ΔRΔI = constant, where ΔI = I + -I -was defined above (because of uncertainties in the data, we do not claim this recasting as 'definitive'-there might be small non-zero intercepts). We used this relationship in two ways: (a) to check if our new values of ΔRΔI with Cu agree with earlier ones [20][21][22]-they do; and (b) to check if the values of ΔRΔI for CuGe agree with those for Cu-Table II shows that they do. These agreements at both 295 K and 4.2 K give us confidence that any systematic errors in our data should not be large. | cond-mat.mtrl-sci |
( 15) is shown with solid (blue) lines. Here η = 0.035 meV and other parameters are as in Fig. 2. We identify four characteristic frequencies, two defining the optical absorption edges, ω-and ω+, while the other two correspond to the peaks of the spin conductivity occuring at ωa and ω b . The latter two arises due to the symmetry of the spin-split conduction bands in k-space at the Fermi level. F,+ (θ) ≤ k ≤ k F,-(θ), for which ε + (k, θ) ≤ ε F ≤ ε -(k, θ). | cond-mat.mes-hall |
The films were subsequently washed several times in pure C 2 H 4 Cl 2 to remove excess Alq 3 . There are cracks of size 1-2 nm in the anodic alumina film produced in sulfuric acid [16,19,20]. Ref. [16] claims that when the anodic alumina film is soaked in Alq 3 solution, Alq 3 molecules of 0.8 nm size diffuse into the cracks and come to rest in nanovoids nestled within the cracks. Since the cracks are 1-2 nm wide, only 1-2 molecules of Alq 3 can reside in the nanovoids. Surplus molecules, not in the nanovoids, will be removed by repeated rinsing in C 2 H 4 Cl 2 [16]. | cond-mat.mes-hall |
Experimental Details Single crystals of TbMnO 3 with dimensions 2 × 2 × 1 mm 3 were grown at the University of Oxford using the flux growth method. They were cut with either the [0 1 0] or [0 0.28 1] directions as the surface normal and polished with 0.1µm diamond followed by 0.02µ Al 2 O 3 pastes, to a flat, shiny surface. Experiments were carried out on both beamline 5U1 at the SRS, Daresbury Laboratory and ID08 at the European Synchrotron Radiation Facility. The former beamline was used to collect the temperature dependence of the scattering, while the latter was used for its superior flux and high incident photon energy resolution to determine the energy dependence of the scattering. These measurements were conducted in a similar fashion to that of Wilkins et al. [20,21,22] on both beamlines. | cond-mat.str-el |
[18] in spite of their different protocol simulations. They used an impulse dynamics that can be mapped into our NVE simulations varying the initial conditions with respect to the temperature of simulations. In Table 2 we present the equivalent results for the case of a Ne atom inside a (5,5)BNNT. Again, for the cases investigated, the temperatures where the Ne atom is not encapsulated were 300K for the tube of 25 Å, and 240 and 300 K for the tube of Å. We believe that this occurs by the same reasons (large end atom tube movements creating a dynamic barrier) as in CNT case. That this occurs at a lower temperature in the BNNT case can be attributed in part to different symmetries (buckling) and the slightly diameter differences between the CNT (6.8 Å) and BNNT (6.9 Å). | cond-mat.mtrl-sci |
These series starts from g ω , whereas the average of any higher term vanishes. This profs T1. Next, it is possible also to show that the DMFT Green's function is optimal with respect to the variations of the Gaussian trial action (call this statement T2). With a variation S = -G -1 dual f * f → S = -G -1 dual f * f + 0f * 1 f 2 , formula (B2) becomes < (S+G -1 dual f * f )f * 1 f 2 > S =< S+G -1 dual f * f > S < f * 1 f 2 > S . (B7) The essential point is again that since all local momenta of f, f * are vanished because of (B6), and the dual potential V is local in space, all the nonlinearity drops out from the (B7). It means that both left-and right-hand sides of (B7) equal the same value, if -G -1 dual f * f equals the Gaussian part of the dual action. | cond-mat.str-el |
Second, b is then calculated from a based on equation ( 21), with χ depending on i * as given in (11). With respect to the following discussion we assume here that desorption caused deviations of χ from equation ( 11) can be neglected, which in the framework of the standard nucleation theory is valid in the long-time limit. In any case, more refined estimates of χ can easily be implemented. As has been shown in equation (19) a does not depend directly on p, but only on β and the boundary conditions (ŝ). Thus the parameters for the numerical computations can be chosen arbitrarily. Using u and t as the respective arbitrary units of length and time, the chosen parameters a all set to unity: s = 1 u, D = 1 u 2 /t and F = 1 pt/(u 2 t). | cond-mat.mtrl-sci |
Film thicknesses were determined using low-angle XRD, while film quality was verified with Laue oscillation measurements. X-ray photoemission spectroscopy (XPS) measurements were made using a Mg anode (1253.6eV) and hemispherical analyzer in a UHV measurement chamber with base pressure on the order of 10 -10 mbar. Samples were either baked overnight in a load-lock at approximately 100 • C to desorb surface contaminants, or brought into the measurement chamber rapidly to retain any contamination present on the sample surface. Some films were plasma etched using both O 2 and Ar at a calibrated etch rate of 1 nm/minute. This combination has been shown to etch STO without depleting its oxygen and rendering it metallic. 21 B. STM STM measurements were made under various conditions, ultra high vacuum (UHV), flowing helium gas, and ambient. | cond-mat.str-el |
Solving the one-dimensional Poisson equation, we obtain the electric field E in the spacer to be E = ∆E c -(E 0 + E F ) -∆ ed = en sat 2ǫ 0 ǫ r . (1) The conduction band mismatch between AlAs and Al 0.45 Ga 0.55 As is ∆E C = 140 meV 21 . From Hartree simulations, we calculate the binding energy of the ground state in the well to be E 0 = 12 meV (dark) and 14 meV (PIA) relative to the X-band minimum at the AlAs/AlGaAs interface (using m t as the mass in the confinement direction). The Fermi energy is E F = 0.53 meV for the dark saturation density of 2.0 × 10 11 cm -2 and 1.06 meV for the PIA saturation density of 4.0 × 10 11 cm -2 . ǫ 0 denotes the permittivity of free space and the relative permittivity of Al 0.45 Ga 0.55 As is ǫ r = 11.6 22 . The density-of-states mass used for Fermi energy is m * = (m l m t ) 1/2 = 0.45m e . | cond-mat.mes-hall |
Simple models such as the Rashba coupled 2D electron system with an out-of-plane magnetization should be under the reach of modern numerical algorithms. Acknowledgments I thank M. P. Anatska for the encouragement and the critical reading of this manuscript and I thank J. Sinova, Q. Niu and A. H. MacDonald for the useful collaboration. This work was funded in part by DOE under Contract No. DE-AC52-06NA25396. The total distribution function f l in the steady state (∂f l /∂t = 0) can be written as where g s l , g a1 l , g a2 l and g adist are non-equilibrium corrections to the distribution function of linear order in the electric field (the label adist stands for the anomalous distribution). They solve self-consistent time-independent equations: [53] and One can deduce the dependence of the distribution corrections on the impurity concentration by noticing that ω ll ′ ∼ n, then from (91) follows that g s l ∼ n -1 . | cond-mat.mes-hall |
Moreover, there seems to be no universal way to extract the Markov limit in various physical situations. By using the general Eq. ( 4), we obtain an all-purpose equation of motion at a modest computational cost. In the evolution generated by the H DQD , the states |0 and |3 are invariant and nontrivial evolution takes place only in the subspace spanned by |1 , |2 . The corresponding evolution operator can therefore be found easily. Transforming Eq. | cond-mat.mes-hall |
Note that it could be a difficult or even unfeasible task. Here, it was fortunate that classical potential calculations spuriously allow to stabilize the mobile core. Finally, it is interesting to discuss if this result is a rarity, specific to silicon, or whether it could occur in other systems. Certainly, an important condition is that several core configurations should exist for a given dislocation. This is certainly true in materials where reconstruction of the dislocation core could occur, leading to several more or less complex and stable configurations. All materials showing a covalent bonding character could therefore be concerned: usual semiconductors (Group IV, III-V, or II-VI) or more complex systems such as minerals present in the Earth mantle, or ceramics. | cond-mat.mtrl-sci |
By the choice of these model potentials and model structures we qualitatively probe the influence of structural aspects of the investigated compounds without being specific to a special kind of material. Model potentials are often used in fracture simulations and have led to useful insight into fundamental mechanisms (see e.g. Abraham (2003)). The minimum of the Lennard-Jones potential for the interactions between atoms of different kind is set to twice the value of that for atoms of the same type. However, the conclusions drawn from our simulations remain essentially unaffected by setting all binding energies equal, which again indicates that we are mainly probing structural effects. A fundamental building unit of the simulated quasicrystal -the prolate rhombohedronin a slightly deformed way forms the cubic C15 Laves phase A 2 B by periodic arrangement. | cond-mat.mtrl-sci |
IV, the results are only valid when the coupling between the grain and the lead is dominant and much smaller in time scale than the intrinsic relaxation. On the other hand, the Born and Markoff approximations enforce that Γ is much smaller than k B T (bath temperature) and the energy difference between manybody states in the grain. These two requirements confine the validity range of Γ in this work. Besides, the particle-hole excitation terms in Eq. (10) of Ref. 61 are also absent in this work, because the energy scale of these excitations are of the order δ a , δ i > 1meV, which already one order larger than the spin bias (V s ∼ 0.1meV) that can be generated according to Fig. | cond-mat.mes-hall |
Figure 4 also shows three unstable fixed points. In contrast to the situation at other points on the flow diagram, each of the fixed points exhibits a many-body spectrum that can be interpreted as the direct product of a set of bosonic excitations and a set of fermionic excitations. The Kondo fixed point corresponds in the renormalization-group language of Fig. 4 to effective couplings λ = 0 and ∆ = ∞. The many-body0 0 1 1 0 L ∆ λ K C LM FO λ c0 FIG. 4: Schematic renormalization-group flows on the λ-∆ plane for the symmetric charge-coupled BFA model with a bath exponent 0 < s < 1. | cond-mat.str-el |
Before introducing the winding number of the spin edge states on the complex energy RS, let us try to interpret the above two numerical characteristics. The first one is that there are always two energy-degenerate edge states appearing localized at opposite edges. To explain this issue, we suppose the spin edge states to be exponentially localized on the boundary with the following ansatz [33,34] ψ n = λ n ψ, (17) where λ is a complex number. Inserting Eq. ( 17) into the Harper equation ( 8), one can easily get the complex number λ satisfying the following equation λ + λ -1 2 -b λ + λ -1 -(d + 2) = 0. (18)That meansλ + λ -1 = ±t ± . | cond-mat.mes-hall |
Then, the Hamiltonian is reduced to a quantum rotor model H = U 2 j n j (n j -1) -µ j n j -2t <ij> cos(θ i -θ j ),(2) where t = n 0 (n 0 + 1)w. Here we take the number of bosons n j ≥ 0. Through a path integral mapping, we can construct the corresponding classical action [14] S = r ǫU 2 J τ r (J τ r -1) -ǫµJ τ r -ln I J x r (2ǫt) -ln I J y r (2ǫt)(3)with the partition functionZ = ∇• J=0 { Jr} e -S[ J] ,(4) where ǫ = β/L τ is a lattice constant in the imaginary time axis for an inverse temperature β, J r is an integer current at site r = (j, τ ) with a spatial index j and a temporal index τ , which is conserved at each site as denoted by ∇ • J = 0, and I m (x) is the modified Bessel function given by the relation e K cos θ = ∞ m=-∞ I m (K)e imθ . In this work, we investigate the properties of the model in Eq. 3 via Monte Carlo simulations using a recently proposed worm algorithm [13]. In order to reduce the systematic errors in discretizing the imaginary time axis, we need to take ǫ √ tU ≪ 1. We take U ǫ = 0.5 -2 for t ≪ U and set the energy unit U = 1. | cond-mat.str-el |
We add here that the IPT results for the field-dependence of the spectral functions are not in agreement with the NRG+DMFT results of Bauer et. al 27 . In the latter, the spectral function seems to be pinned at the Fermi level until the metamagnetic transition, whence there is a sudden drop in the density of states. We do not understand this difference with NRG completely, but speculate that this disagreement could be due to the non-conserving nature of IPT. The evolution of the spectral function for U = 1.5 and U = 2.0 (figure 8) with field is quite dramatic, reflecting the metamagnetic transition. The spin bands shift rapidly with field, and for a critical field at which the transition happens, the shift from a metallic to a quasiferromagnetic insulator takes place. | cond-mat.str-el |
We argue here that this is merely an artifact of the extrapolation procedure. Indeed, as it is discussed in section II E, negative ImG in our theory could only result from a negative residual. However, the graph of Σ at Matsubara frequencies for all k-points is qualitatively similar to whose shown in the upper panel of Figure 15. It is obvious these graphs have a negative derivative at Fermi energy, so that the residual Z = (i -∂Σ ∂ω ) -1 must be positive. IV. CONCLUSIONS To summarize, the transformation to dual fermion variables completely reconstructs perturbation theory, starting with the zeroth-order approximation which is accurate in the limits of both very weak and very strong interactions. | cond-mat.str-el |
In V ef f {η} = V ij η i (1 -η j )(2) with V > 0 being a measure of the ordering tendency. The ground state in this model would correspond to η = 1, 0, 1, 0, .. along each axis, i.e, B, B', B, B'.. Notice that this approach tries to incorporate the effect of complex interactions between the A, B, B' and O ions (as also the electrons) into a single parameter (See Appendix). This binary model is equivalent to a nearest neighbour Ising model, so the equilibrium physics is very well understood 6 . We, however, want to explore the consequences of different annealing protocols on this model, to examine the consequences of imperfect annealing. The qualitative ordering effects in the lattice gas model are similar in 2D and 3D, so we work with a 2D structure since it allows ease of visualisation. IV. | cond-mat.mtrl-sci |
The choice of Ising order parameter dictates the value of the scaling dimension of ∆ as d = 1/8. However in the absence of full microscopic derivation of Eq. ( 2) the scaling dimension d should be really considered as a fitting parameter with fluctuations enhanced for larger vales of d. In fact, we have found that d = 1/4 more accurately describes our data. We additionally assume that fluctuations remain critical in a wide temperature range as is the case for the XY model below the BKT transition. This assumption probably introduces a bias towards order thus making the spectral lines narrower than they must be. We remind the reader that the scenario described in [23], where quasiparticles have ∼ ω 2 + T 2 decay rate, is likely to lead to even sharper peaks. | cond-mat.str-el |
Applying the divergence operator to the equation ( 21) and differentiating (20) with respect to time, one obtains the equations of continuity in the form:T p X i j ij ∂ ∂ = ∂ ∂σ . (22) The equations (22) allow only one interpretation of i p in Eq. (20), namely as the momentum, as its left-hand part in ( 22) is a force (17). It follows herefrom that the equations of continuity (22) represent the second Newton's law for the locally inhomogeneous dynamical model which describes the dynamics of defects in the lattice. I.e., in the local Landau theory, the fundamental law of dynamics is determined by invariance of the nonequilibrium potential with respect to the spatial translations onto the periods of the unit cell. It is worth noting that if we define ij ij A Σ ≡ , the Eq. | cond-mat.mtrl-sci |
Such energy products are not competitive with high performances NdFeB permanent magnets [9]. However, preliminary high temperature measurements show a good stability up to 250°C (Fig. 2), thus we think that these materials could be serious contenders for permanent magnet applications at high temperatures. As shown in Figure 5, they could fill a gap between RE magnets like SmCo [9] and AlNiCo magnets [10]. RE magnets have higher coercivities but they show stronger softening upon warming while AlNiCo magnets exhibit much lower coercivities in the whole temperature range of interest. In addition, the fabrication method is a low temperature chemical process which does not need any advanced metallurgical skills. | cond-mat.mtrl-sci |
( 9) and schematically shown in the inset for positive values of α and β. At |V M | < V T the slope of f (V M ) is determined by α and at |V M | > V T it is determined by β. where θ(•) is the step function. The simulation results [20] depicted in Fig. 2 clearly show a pinched hysteresis loop and a hysteresis collapse with increasing frequency ω of the alternating (ac) voltage source V (t) = V 0 sin (2πωt). We can now extend the above definitions to capacitances and inductances. The memory devices that result share many characteristics of memristive systems, but with a fundamental difference: they store energy. | cond-mat.mes-hall |
Achieving resistance quantization, with the precision of few parts per billion required by metrology is, however, a much more complex and material-dependent issue. Suitability of material for QHE metrology depends on its homogeneity, stability, low intrinsic noise, possibility to attach lowresistance contacts to it 5 . Despite nearly 30 years of history, metrologically sound precision of the quantum Hall quantization at the level of parts per billion (ppb) was obtained only in two types of devices 6 -Si field effect transistors and GaAs heterostructures -two well-established materials of semiconductor microelectronics. New materials are sought to expand the horizon of quantum metrology and to advance the understanding of QHE itself. Graphene -a single layer of graphite -is a truly two-dimensional gapless semiconductor with electrons mimicking the behaviour of relativistic (Dirac) electrons 7 . This last feature of charge carriers in graphene is manifested most spectacularly through an unusual QHE 8 In reality, an impressive range of unconventional transport properties of electrons in graphene 13 , including QHE, have been seen almost exclusively in flakes mechanically exfoliated from bulk graphite. | cond-mat.mes-hall |
This can be decoupled in hopping and pairing channels, using the order parameters defined in Eq. ( 38). This modifies the coefficients of the quadratic Hamiltonian of Eq. ( 32) as follows In addition, the Hamiltonian acquires a constant contribution given by δE [4] The Hamiltonian is solved by a Bogoliubov transformation. For fixed s, the d and ∆ order parameters are determined self-consistently, while µ is tuned to make sure the constraint on bond operators is satisfied. s is chosen to minimize the ground state energy for every J 2 /J 1 . | cond-mat.str-el |
This paper is organized as follows. We begin by reviewing the experimental studies of low-temperature magnetoresistance phenomena in (Ga,Mn)As (Section II). We follow in Section III with an outline of the general formalism used here to evaluate the Cooperon in multiband ferromagnets. In Section IV we apply the formalism to the two dimensional electron gas ferromagnet (M2DEG) model studied earlier by Dugaev et al.. 15 The reults in this section are useful in discussing the competition between WL and WAL in ferromagnets generally. Section V is devoted to the more complicated 4-band Kohn-Luttinger model with a kinetic-exchange meanfield, which captures the essentials 19 of ferromagnetism in (Ga,Mn)As. We find that in this model, which employs the disordered valence-band picture of states near the Fermi level in metallic (Ga,Mn)As and typically overestimates the effect of SO interactions, very small exchange fields are sufficient to convert the positve MR (WAL) of the paramagnetic state to negative MR (WL) in the ferromagnetic state. | cond-mat.mtrl-sci |
In particular, it is shown that the induced spin polarization in these systems remains zero even when the frequency of the applied field goes to zero. The calculations also show that the static spin current in the electron systems described by the Hamiltonian (1) is zero even if a non-parabolicity of electron band spectrum is taken into account. The presented theory neglects the spin-orbit corrections to the scattering potential, which means that the effects of extrinsic spin-orbit coupling, such as the extrinsic spin currents and Elliot-Yafet spin relaxation, are not included in the calculations. The paper is organized as follows. In Sec. II we consider the quantum kinetic equation and present its analytical solutions. | cond-mat.mes-hall |
Richter were able to include the leading off-diagonal terms into the summation. 38,39 With off-diagonal contributions, the trajectory-based approach could successfully capture τ E dependences, not only of the weak localization correction and Fano factor of a quantum dot, 18,23,24,25,28,29 but also of quantum signatures whose τ E dependence was not known from the field-theoretic approach, such as the conductance fluctuations or the current pumped through a quantum dot with timedependent shape (a "quantum pump"). 25,26,27 Unlike weak localization and the Fano factor, the variance of the conductance var G and the mean square pumped current were found not to disappear in the limit of large Ehrenfest times. In fact, in the absence of dephasing, var G is independent of τ E in a quantum dot, as was first observed by Tworzydlo et al. and Jacquod and Sukhorukov on the basis of numerical simulations. 19 In the semiclassical theory the remarkable τ E -insensivity of var G in quantum dots has its origin in a large contribution to var G from trajectories that spend a long time in the vicinity of periodic orbits. | cond-mat.mes-hall |
1, labelled H = 0. This Boltzmann model has the advantages of interpolating between the ballistic and diffusive limits, and generally including other approaches [8][9][10][11][12][13][14][15] as limiting cases. We chose CuGe because Ge in Cu: (a) adds a large resistivity per atomic percent Ge (≈ 3.7 μΩcm/at.% [17]), thereby greatly shortening the mean-free-path, λ; (b) has a small spin-orbit cross-section in Cu (σ so ≈ 5.2 x 10 -19 cm 2 [18]), giving weak spin-flipping and thus a long spin-diffusion length, l N sf ; and (c) is soluble in Cu to ≈ 10 % [19]. As shown in Table I, with Cu, λ is at least 3 times the layer thickness of 10 nm (ballistic transport). With CuGe, λ is less than 40% of 10 nm (diffusive transport). We can thus determine, at both 295K and 4.2K whether: (1) just changing the scattering in the N-layer from ballistic to diffusive significantly affects the switching current, ΔI; and (2) how well the Boltzmann equation model of current-induced switching describes the ratio of switching currents. | cond-mat.mtrl-sci |
If the method is applicable, the regularization parameter ε can be varied over several orders of magnitude without effect, but should of course be chosen as small as possible (10 -7 -10 -6 in our calculations). An efficient procedure to fix ε is to examine plots of the eigenvalues α i (k) of ( 9) as functions of the quasimomentum k for several ε. For too small ε, noisy scatter and many abrupt jumps appear, resulting in completely wrong lineshapes for some k-values; ε is sufficiently big when α i (k) shows smooth "band structures" with only a few abrupt jumps. In cases where this is only possible with large ε, the prediction method is not applicable. B. Isotropic Heisenberg spin-1 2 antiferromagnet For the Heisenberg antiferromagnet (XXX model)ĤHAFM = i Ŝi • Ŝi+1(12) we have calculated the longitudinal spin structure factor (spectral function) S zz (ℓ, t) = Ŝz ℓ (t) Ŝz 0 (0) β . This case is interesting and challenging as already at T = 0 a whole continuum of excitations contributes to the spin structure factor as opposed to the simple dispersion of the XX model with a single peak in S(k, ω) each momentum. | cond-mat.str-el |
Summations over all modes as well as the size distribution of NPs in order to get a quantity closer to the experimentally INS spectrum will be considered later. Let σ H and m H respectively denote the neutron scattering cross section and mass of a hydrogen atom. Assuming that only the scattering by H surface atoms matters and neglecting the scattering by other atoms (Ti and O), we can rewrite this summation as being over all the H atoms σ H m H ν H e 2 ν H . ρ is the density of the NP and V is the volume. N H is the number of hydrogen atoms fixed to the surface of the NP. Introducing the mean square displacement at the surface U2S defined as the ratio of the surface and volume average of the square displacement as in Ref. | cond-mat.mtrl-sci |
The logarithmic corrections generated by the higher-order terms in classical smectics [132] will not be present. Notice that these scaling dimensions are the same as what we got from the phenomenological theory in Sec. IV, before it was coupled to the fermions. However, the physics is now very different. Most important of all, in the presence of fermions, the smectic Goldstone mode is damped, which is not the case in the phenomenological theory. A one-loop calculation of the fermion self-energy shows that the fermionic quasiparticles in the electronic smectic phase is a FL (Appendix E 2). | cond-mat.str-el |
Part of this flake (II) exhibits a higher optical density. We estimate this region to have a thickness of 5 graphene layers. The fabrication and identification of single-and multi-layer graphene samples by Raman scattering spectroscopy are described below. Figure 1(b) shows an optical image of the sample after electrode deposition around the flake. Most of the single-layer region (I) and a portion of the multi-layer region (II) are accessible for STM studies; the rest of the flake is buried beneath the gold film. Since the thin gold film is partially transparent, the contour of the entire flake of Figure 1(a) can still be seen. | cond-mat.mes-hall |
11): E(t) = Ee ⊥ sin ωt, e ⊥ is the unit vector perpendicular to the NT axis. An interaction between the electric field and an electron in a NT, which leads to electric-dipole transitions, is given by the following operator: H E = |e|E m 0 ω cos ωtP ⊥ = -i|e| E m 0 Rω cos ωt sin ϕ ∂ ∂ϕ ,(A2) where m 0 is the bare electron mass and P ⊥ = -i sin ϕ∂ ϕ /R is the electron momentum along e ⊥ . Here we assume that the influence of the lattice potential can be neglected for the estimation of the electric-dipole transitions. Therefore, using Eq. ( 37), the matrix element of the electric-dipole transitions can be expressed as where τ 1 is the Pauli matrix operating on valley-index space. The eigenvalues of the operator H 0 + V (ζ) + H curv SO + H K-K (for m = 0 subband) are given by E 0,n,S ζ = ± v ν 2 /9R 2 + k 2 n + k 2 K-K + (ϕ AB /R + 2S ζ k SO ) 2 +2β (ϕ AB /R + 2S ζ k SO ) 2 ν 2 /9R 2 + k 2 K-K (ν 2 /9R 2 + k 2 n ) 1/2 + S ζ ω Z ,(B4) where k K-K = ∆ K-K / v, k SO = ∆ curv / v. The energy spectrum of the lowest electron energy levels and highest hole levels described by Eq. | cond-mat.mes-hall |
In section 3 we provide a summary of our main results. In section 4 we derive analytic formulas pertaining to the kinetic steady state: In section 4.2 we use the mass fluxes as inputs and derive the ES effect [11,32]; and in section 4.3 we use the mass densities as inputs to derive asymmetric, θ-dependent step-edge permeability rates. In section 5 we apply perturbation theory to find β(θ) by using primarily the mass fluxes as inputs: In section 5.1 we carry out the perturbation analysis to first order for the edge-atom and kink densities as κ → 0; in section 5.3 we derive the step stiffness as a function of θ; and in section 5. 4 we discuss an alternative viewpoint on the stiffness. In section 6 we discuss our results, and outline possible limitations. The appendices provide derivations and proofs needed in the main text. | cond-mat.mtrl-sci |