KenanZeynalov/petrol_data · Datasets at Hugging Face

text stringlengths 0 1.75k
etc. [12].
The ratio of the sum of energy burdens of a particular product to
its energy content is deﬁned as the energy intensity of that pro-
duct. Note that the inverse of energy intensity of a product repre-
sents its product-speciﬁc efﬁciency. In the current study, a process-
based energy allocation was employed, which was reported in
Elgowainy et al. [12].
LCA of petroleum products accounts for energy use and emis-
sions associated with all stages in the fuel cycle, including crude
recovery and transportation, fuel production, transportation, dis-
tribution and combustion of the fuel by end-use application
[9,12]. Furthermore, allocation of energy use and emissions bur-
dens among co-products was performed by utilizing product-
speciﬁc efﬁciencies and process fuel shares [9,12]. This protocol
was followed along each stage of the product life-cycle. Key
parameters for upstream energy efﬁciencies and emissions associ-
ated with recovery, processing and transportation of various crude
inputs, NG and electricity generation are presented in Table S4, as
well as the references of the parameters. Crude oil, NG, and elec-
tricity generation mixes for US reﬁneries are based on 2010 data
to match reﬁnery LP modeling data inputs. The EU parameters in
Table S4 are based on data reported by JRC and Eurostat of the
European Commission [3,13]. GREET was populated with these
US and EU parameters (Table S4) to compare life-cycle energy
and GHG intensities of petroleum products from US and EU
reﬁneries.
A notable difference between US and EU crude recovery GHG
emissions estimates is the magnitude of associated methane
(CH4) emissions. This is attributed to the difference in methodolo-
gies used to estimate CH4 emissions. For the US, the GREET model
estimates CH4 emissions based on the ﬂaring emissions from satel-
lite data using a 5:1 ratio of ﬂared to vented associated gas [22]. On
the other hand, the JRC study relies on a report by the International
Association of Oil and Gas Producers (OGP), which collected emis-
sions data from OGP members [23]. Another key difference is the
share of oil sands in crude feed to US reﬁneries since GHG intensi-
ties of oil sands crude are typically higher compared to conven-
tional crude (see Fig. S4). Electricity GHG intensity is decided
primarily by the electricity generation mix. Compared to the US,
the GHG emission intensity of EU-generated electricity is signiﬁ-
cantly lower, mostly due to the lower share of coal power genera-
tion and higher share of nuclear and renewable power generation
in the EU mix. GHG emission factors for fuel combustion are fairly
consistent between the US and EU, except for diesel. This differ-
ence is due to the lower carbon content (on a mass basis) of EU die-
sel compared to US diesel (Table S5).
3. Results
3.1. Overall reﬁnery efﬁciency
Fig. 1 presents the grouping of US and EU reﬁneries using the
parametric assumptions described above. HP yields and crude
input
API
gravity
are
plotted
to
show
their
relevance
in
gLHV ¼
P
nðPn LHVnÞ
P
mðCm LHVmÞ þ P
0ðOIo LHVoÞ þ NGpurchased;LHV þ H2;purchased;LHV þ Electricitypurchased
ð1Þ
Fig. 1. Crude API gravity and heavy product yield of the studied US and EU
reﬁneries (The yield of heavy products, such as residual fuel oil, pet coke, asphalt,
slurry oil and reduced crude, is calculated as a share of all energy products by
energy value).
294
J. Han et al. / Fuel 157 (2015) 292–298
categorizing reﬁneries. Filled shapes represent US reﬁneries, while
unﬁlled shapes represent EU reﬁneries. These results show that
almost all Low API and High API/Low HP reﬁneries are present
within the US, rather than the EU Conversely, almost all EU reﬁner-
ies form part of the High API/High HP group. For all subsequent
results, comparing the Low API group with the High API/Low HP
highlights the impact of crude API gravity, while comparing the
High API/Low HP group with the High API/High HP group highlights
the impact of heavy product yield.
Fig. 2 illustrates the overall reﬁnery efﬁciency in each of the
three reﬁnery groups. The bottom, mid and top of the boxes in
Fig. 2 represent the 25th percentile, production-weighted average
and 75th percentile, respectively, while the ends of the error bars
represent the 10th and 90th percentiles. These results suggest
strong impacts of API gravity and HP yield on overall reﬁnery efﬁ-
ciency. This can be rationalized by the installed capacity (MJ
throughput/MJ crude inputs) of deep conversion units, such as cok-
ers and catalytic crackers, in each group (see Table S3). The Low API
group has a much larger installed capacity of deep conversion units
then the other groups. On the other hand, the High API/High HP
group has a negligible capacity of cokers and hydrocrackers.
These conversion units are more energy-intensive than other pro-
cess units within reﬁneries, and thus consume signiﬁcant amount
of utilities (heat and electricity) and hydrogen.
Hydrogen is highly GHG-intensive, depending on the source.
Thus, the amount and source of hydrogen consumption are key
LCA parameters. Fig. 3 illustrates that on a MJ/MJ crude basis, the

etc. [12].

The ratio of the sum of energy burdens of a particular product to

its energy content is deﬁned as the energy intensity of that pro-

duct. Note that the inverse of energy intensity of a product repre-

sents its product-speciﬁc efﬁciency. In the current study, a process-

based energy allocation was employed, which was reported in

Elgowainy et al. [12].

LCA of petroleum products accounts for energy use and emis-

sions associated with all stages in the fuel cycle, including crude

recovery and transportation, fuel production, transportation, dis-

tribution and combustion of the fuel by end-use application

[9,12]. Furthermore, allocation of energy use and emissions bur-

dens among co-products was performed by utilizing product-

speciﬁc efﬁciencies and process fuel shares [9,12]. This protocol

was followed along each stage of the product life-cycle. Key

parameters for upstream energy efﬁciencies and emissions associ-

ated with recovery, processing and transportation of various crude

inputs, NG and electricity generation are presented in Table S4, as

well as the references of the parameters. Crude oil, NG, and elec-

tricity generation mixes for US reﬁneries are based on 2010 data

to match reﬁnery LP modeling data inputs. The EU parameters in

Table S4 are based on data reported by JRC and Eurostat of the

European Commission [3,13]. GREET was populated with these

US and EU parameters (Table S4) to compare life-cycle energy

and GHG intensities of petroleum products from US and EU

reﬁneries.

A notable difference between US and EU crude recovery GHG

emissions estimates is the magnitude of associated methane

(CH4) emissions. This is attributed to the difference in methodolo-

gies used to estimate CH4 emissions. For the US, the GREET model

estimates CH4 emissions based on the ﬂaring emissions from satel-

lite data using a 5:1 ratio of ﬂared to vented associated gas [22]. On

the other hand, the JRC study relies on a report by the International

Association of Oil and Gas Producers (OGP), which collected emis-

sions data from OGP members [23]. Another key difference is the

share of oil sands in crude feed to US reﬁneries since GHG intensi-

ties of oil sands crude are typically higher compared to conven-

tional crude (see Fig. S4). Electricity GHG intensity is decided

primarily by the electricity generation mix. Compared to the US,

the GHG emission intensity of EU-generated electricity is signiﬁ-

cantly lower, mostly due to the lower share of coal power genera-

tion and higher share of nuclear and renewable power generation

in the EU mix. GHG emission factors for fuel combustion are fairly

consistent between the US and EU, except for diesel. This differ-

ence is due to the lower carbon content (on a mass basis) of EU die-

sel compared to US diesel (Table S5).

3. Results

3.1. Overall reﬁnery efﬁciency

Fig. 1 presents the grouping of US and EU reﬁneries using the

parametric assumptions described above. HP yields and crude

input

API

gravity

are

plotted

to

show

their

relevance

in

gLHV ¼

P

nðPn LHVnÞ

P

mðCm LHVmÞ þ P

0ðOIo LHVoÞ þ NGpurchased;LHV þ H2;purchased;LHV þ Electricitypurchased

ð1Þ

Fig. 1. Crude API gravity and heavy product yield of the studied US and EU

reﬁneries (The yield of heavy products, such as residual fuel oil, pet coke, asphalt,

slurry oil and reduced crude, is calculated as a share of all energy products by

energy value).

294

J. Han et al. / Fuel 157 (2015) 292–298

categorizing reﬁneries. Filled shapes represent US reﬁneries, while

unﬁlled shapes represent EU reﬁneries. These results show that

almost all Low API and High API/Low HP reﬁneries are present

within the US, rather than the EU Conversely, almost all EU reﬁner-

ies form part of the High API/High HP group. For all subsequent

results, comparing the Low API group with the High API/Low HP

highlights the impact of crude API gravity, while comparing the

High API/Low HP group with the High API/High HP group highlights

the impact of heavy product yield.

Fig. 2 illustrates the overall reﬁnery efﬁciency in each of the

three reﬁnery groups. The bottom, mid and top of the boxes in

Fig. 2 represent the 25th percentile, production-weighted average

and 75th percentile, respectively, while the ends of the error bars

represent the 10th and 90th percentiles. These results suggest

strong impacts of API gravity and HP yield on overall reﬁnery efﬁ-

ciency. This can be rationalized by the installed capacity (MJ

throughput/MJ crude inputs) of deep conversion units, such as cok-

ers and catalytic crackers, in each group (see Table S3). The Low API

group has a much larger installed capacity of deep conversion units

then the other groups. On the other hand, the High API/High HP

group has a negligible capacity of cokers and hydrocrackers.

These conversion units are more energy-intensive than other pro-

cess units within reﬁneries, and thus consume signiﬁcant amount

of utilities (heat and electricity) and hydrogen.

Hydrogen is highly GHG-intensive, depending on the source.

Thus, the amount and source of hydrogen consumption are key

LCA parameters. Fig. 3 illustrates that on a MJ/MJ crude basis, the