KenanZeynalov/petrol_data · Datasets at Hugging Face

text stringlengths 0 1.75k
Among the major stages in the life-cycle of a petroleum fuel (crude
recovery, transportation, reﬁning and fuel transportation, dis-
tribution and combustion), the largest GHG emissions source is
fuel combustion, which can be accurately estimated from the car-
bon content of the fuel. The next largest GHG emissions source for
desirable fuels (e.g., gasoline, diesel and jet) is the petroleum reﬁn-
ing stage. Oil reﬁneries process a slate of crude oils of different
qualities into multiple fuel products for various applications. In
order to accurately estimate variations in petroleum reﬁnery efﬁ-
ciency and GHG emissions, reliable information relating to overall
energy inputs and outputs is required for different crude types,
reﬁnery conﬁgurations and product outputs. In addition, energy
inputs and GHG emissions at the reﬁnery level need to be allocated
systematically among petroleum products in order to develop
accurate product-speciﬁc GHG emissions intensities.
Both crude quality and ﬁnal production speciﬁcation are key
drivers for reﬁnery conﬁguration, operations and ultimately, reﬁn-
ing energy efﬁciency. For example, historically, crude inputs into
US reﬁneries have typically been heavier (average API gravity of
30–32) than EU reﬁneries (average API gravity of 36–37) [1,8].
In the former case, because crude inputs are heavier, more inten-
sive processing is required to produce gasoline and distillate. In
terms of market demands, non-transportation fuel oil demands
in the US (Fig. S1) are smaller than in the EU. Consequently, US
reﬁneries produce a smaller share of residual fuel oil (RFO) than
EU reﬁneries do, and thus are considered to be more resource efﬁ-
cient. Since gasoline and diesel require signiﬁcantly more process-
ing than heavy products, US reﬁneries in general are more complex
and energy-intensive than EU reﬁneries. On this basis, it is unsur-
prising that US reﬁneries have larger deep conversion units such as
cokers and ﬂuidized catalytic crackers (FCC) relative to other
regions
(Fig.
S3).
These
process
units
are
instrumental
in
converting heavy reﬁnery intermediate streams into gasoline and
diesel and are typically energy-intensive; hence their impacts on
reﬁning efﬁciency and life-cycle analysis GHG emissions can be
substantial [9].
Other studies have examined product-speciﬁc efﬁciencies and
GHG intensities of reﬁned products and there is a wide variation
in the potential emissions due to differences in modeling metho-
dology and input data. Furuholt used data of eight general reﬁning
processes in Norwegian reﬁneries to allocate reﬁning energy use
and emissions to gasoline and diesel [10]. Similarly, Wang et al.
used a detailed process-level approach for a notional reﬁnery and
demonstrated the difference between various allocation metrics
(energy, market-value and mass) [11]. Recently, Elgowainy and
Forman et al. used a reﬁnery Linear Programming (LP) model to
simulate operation of 43 large US reﬁneries in order to estimate
life-cycle GHG emissions of major petroleum products such as
gasoline, diesel and jet fuels [9,12]. By covering 70% of the total
US reﬁning capacity, they: (1) developed a correlation between
the overall efﬁciency of US reﬁneries and the corresponding crude
quality, reﬁnery complexity and product slate; (2) provided aver-
age and variations of product-speciﬁc efﬁciency and process fuel
shares for each reﬁned product; and (3) examined the possible
impacts relating to changing crude slates, regional and seasonal
variation, changing gasoline-to-diesel (G/D) ratios and Gas to
Liquid (GTL) diesel blending on reﬁnery and product-speciﬁc
efﬁciencies. A recent well-to-wheels study by the Joint Research
Centre (JRC) of the European Commission evaluated energy and
GHG emissions performance associated with various automotive
fuels and powertrains, including petroleum gasoline and diesel.
By considering a marginal approach (future reduction in gasoline
and diesel demand), JRC concluded that marginal diesel in the EU
is more energy- and GHG emission-intensive than marginal gaso-
line [13]. Other authors have performed individual reﬁnery analy-
ses and incorporated these results into life-cycle analysis (LCA) for
multiple notional reﬁnery conﬁgurations [14–18].
These studies only focused on a speciﬁc region or conﬁgurations
and considered only a limited range of crude quality and product
slates, which is not sufﬁcient to fully understand the complex
interaction between crude quality, reﬁnery conﬁguration and yield
of gasoline and distillate on one hand, and the consequent life-
cycle GHG emissions on the other hand. These disparities between
previous studies suggest a need to use a large pool of reﬁnery data
to potentially simplify general understanding of reﬁnery GHG
emissions. Noting the impact of key reﬁnery metrics such as API
gravity and heavy product (HP) yield (e.g., RFO, pet coke and
asphalt) could have on reﬁnery efﬁciency and GHG emissions, we
analyzed results from LP modeling of 17 large EU reﬁneries in addi-
tion to recently reported 43 US reﬁneries [9,12] and grouped them
according to their crude API gravity and HP yields. Note that these
two parameters (API gravity and HP yield) were recently identiﬁed
by Elgowainy et al. [12] as the key parameters that determine a US
reﬁnery’s overall energy efﬁciency [9,12]. In this study, API gravity
and HP are used to represent resource efﬁciency. By analyzing data
at the sub-process level in these 60 reﬁneries, this study correlates
the crude API gravity and HP yields of different groups of reﬁneries
with the product-speciﬁc energy efﬁciency for each reﬁnery pro-
duct and presents previously unavailable life-cycle impacts of
reﬁnery resource efﬁciency on product-speciﬁc and reﬁnery-level
GHG emissions. The life-cycle analysis of petroleum fuels from
various

Among the major stages in the life-cycle of a petroleum fuel (crude

recovery, transportation, reﬁning and fuel transportation, dis-

tribution and combustion), the largest GHG emissions source is

fuel combustion, which can be accurately estimated from the car-

bon content of the fuel. The next largest GHG emissions source for

desirable fuels (e.g., gasoline, diesel and jet) is the petroleum reﬁn-

ing stage. Oil reﬁneries process a slate of crude oils of different

qualities into multiple fuel products for various applications. In

order to accurately estimate variations in petroleum reﬁnery efﬁ-

ciency and GHG emissions, reliable information relating to overall

energy inputs and outputs is required for different crude types,

reﬁnery conﬁgurations and product outputs. In addition, energy

inputs and GHG emissions at the reﬁnery level need to be allocated

systematically among petroleum products in order to develop

accurate product-speciﬁc GHG emissions intensities.

Both crude quality and ﬁnal production speciﬁcation are key

drivers for reﬁnery conﬁguration, operations and ultimately, reﬁn-

ing energy efﬁciency. For example, historically, crude inputs into

US reﬁneries have typically been heavier (average API gravity of

30–32) than EU reﬁneries (average API gravity of 36–37) [1,8].

In the former case, because crude inputs are heavier, more inten-

sive processing is required to produce gasoline and distillate. In

terms of market demands, non-transportation fuel oil demands

in the US (Fig. S1) are smaller than in the EU. Consequently, US

reﬁneries produce a smaller share of residual fuel oil (RFO) than

EU reﬁneries do, and thus are considered to be more resource efﬁ-

cient. Since gasoline and diesel require signiﬁcantly more process-

ing than heavy products, US reﬁneries in general are more complex

and energy-intensive than EU reﬁneries. On this basis, it is unsur-

prising that US reﬁneries have larger deep conversion units such as

cokers and ﬂuidized catalytic crackers (FCC) relative to other

regions

(Fig.

S3).

These

process

units

are

instrumental

in

converting heavy reﬁnery intermediate streams into gasoline and

diesel and are typically energy-intensive; hence their impacts on

reﬁning efﬁciency and life-cycle analysis GHG emissions can be

substantial [9].

Other studies have examined product-speciﬁc efﬁciencies and

GHG intensities of reﬁned products and there is a wide variation

in the potential emissions due to differences in modeling metho-

dology and input data. Furuholt used data of eight general reﬁning

processes in Norwegian reﬁneries to allocate reﬁning energy use

and emissions to gasoline and diesel [10]. Similarly, Wang et al.

used a detailed process-level approach for a notional reﬁnery and

demonstrated the difference between various allocation metrics

(energy, market-value and mass) [11]. Recently, Elgowainy and

Forman et al. used a reﬁnery Linear Programming (LP) model to

simulate operation of 43 large US reﬁneries in order to estimate

life-cycle GHG emissions of major petroleum products such as

gasoline, diesel and jet fuels [9,12]. By covering 70% of the total

US reﬁning capacity, they: (1) developed a correlation between

the overall efﬁciency of US reﬁneries and the corresponding crude

quality, reﬁnery complexity and product slate; (2) provided aver-

age and variations of product-speciﬁc efﬁciency and process fuel

shares for each reﬁned product; and (3) examined the possible

impacts relating to changing crude slates, regional and seasonal

variation, changing gasoline-to-diesel (G/D) ratios and Gas to

Liquid (GTL) diesel blending on reﬁnery and product-speciﬁc

efﬁciencies. A recent well-to-wheels study by the Joint Research

Centre (JRC) of the European Commission evaluated energy and

GHG emissions performance associated with various automotive

fuels and powertrains, including petroleum gasoline and diesel.

By considering a marginal approach (future reduction in gasoline

and diesel demand), JRC concluded that marginal diesel in the EU

is more energy- and GHG emission-intensive than marginal gaso-

line [13]. Other authors have performed individual reﬁnery analy-

ses and incorporated these results into life-cycle analysis (LCA) for

multiple notional reﬁnery conﬁgurations [14–18].

These studies only focused on a speciﬁc region or conﬁgurations

and considered only a limited range of crude quality and product

slates, which is not sufﬁcient to fully understand the complex

interaction between crude quality, reﬁnery conﬁguration and yield

of gasoline and distillate on one hand, and the consequent life-

cycle GHG emissions on the other hand. These disparities between

previous studies suggest a need to use a large pool of reﬁnery data

to potentially simplify general understanding of reﬁnery GHG

emissions. Noting the impact of key reﬁnery metrics such as API

gravity and heavy product (HP) yield (e.g., RFO, pet coke and

asphalt) could have on reﬁnery efﬁciency and GHG emissions, we

analyzed results from LP modeling of 17 large EU reﬁneries in addi-

tion to recently reported 43 US reﬁneries [9,12] and grouped them

according to their crude API gravity and HP yields. Note that these

two parameters (API gravity and HP yield) were recently identiﬁed

by Elgowainy et al. [12] as the key parameters that determine a US

reﬁnery’s overall energy efﬁciency [9,12]. In this study, API gravity

and HP are used to represent resource efﬁciency. By analyzing data

at the sub-process level in these 60 reﬁneries, this study correlates

the crude API gravity and HP yields of different groups of reﬁneries

with the product-speciﬁc energy efﬁciency for each reﬁnery pro-

duct and presents previously unavailable life-cycle impacts of

reﬁnery resource efﬁciency on product-speciﬁc and reﬁnery-level

GHG emissions. The life-cycle analysis of petroleum fuels from

various