KenanZeynalov/petrol_data · Datasets at Hugging Face

text stringlengths 0 1.75k
total hydrogen consumption decreases signiﬁcantly as API gravity
and HP increase. As discussed above, Low API reﬁneries have much
larger hydrocrackers (HYK), one of the largest consumers of hydro-
gen in reﬁneries [12]. Removal of crude sulfur content also con-
sumes a large amount of hydrogen. Table S2 shows that average
crude sulfur content decreases monotonically from Low API to
High API reﬁneries, which drives the reduction in hydrogen con-
sumption. Hydrogen demand in the High API/High HP group of
reﬁneries is reduced further because the sulfur removal require-
ment via hydroprocessing in HP is low relative to gasoline and
distillate.
Fig. 3 also shows that the amount of hydrogen from SMR
decreases signiﬁcantly as API gravity and HP yield increase. A con-
sequence of this lower share of hydrogen from the SMR results in a
higher overall energy efﬁciency for the High API/High HP reﬁnery
group because hydrogen consumption via SMR is relatively inefﬁ-
cient (70% efﬁciency) compared to other reﬁnery units, resulting
in signiﬁcant energy burdens for products of hydrocracking and
hydrotreating units. Hydrogen is also a co-product of catalytic
reforming, which produces high-octane reformate that contributes
to the gasoline pool. Thus, hydrogen originating from catalytic
reformers has a signiﬁcantly lower energy burden relative to
hydrogen produced from the SMR.
3.2. Product-speciﬁc efﬁciency
Fig. 4 shows the calculated average and variation of product-
speciﬁc efﬁciencies for each group of reﬁneries using the energy
allocation method. The product-speciﬁc efﬁciency for all products
in the High API/High HP group are consistently higher than the
other two reﬁnery groups, mainly due to more favorable crude
quality, higher HP yields and lower complexity. These results are
consistent with those recently reported by Elgowainy et al. [12],
which showed (1) among reﬁnery products, gasoline has the low-
est efﬁciency, (2) RFO has the highest efﬁciency, and (3) diesel can
display a wide range of efﬁciencies. In the latter case, Forman et al.
[9] showed that tighter regulation of aromatics in CARB diesel
combined with reﬁneries that utilize multiple inefﬁcient units via
deep-conversion pathways can result in relatively low diesel efﬁ-
ciency. Although noted only for California reﬁneries, its impact
can exacerbate the already wide range of diesel efﬁciencies in
reﬁneries outside California [9], in general due to the relatively
inefﬁcient diesel reﬁning pathways. Interestingly, HP yield has a
much larger impact on the reﬁning efﬁciency of RFO compared to
the impact of API gravity. The lower reﬁning efﬁciency of RFO with
lower HP yield is likely due to the larger share of HP components
from downstream processes (e.g., HYK and coker), which carry lar-
ger energy and emission burdens.
It is important to note that the estimation of product-speciﬁc
efﬁciencies (as well as energy intensities) depends on allocation
approaches. As mentioned earlier, a marginal approach employed
in the JRC study results in a lower reﬁning efﬁciency (or higher
energy intensity) of diesel than of gasoline in the EU reﬁneries
because the EU reﬁneries operate at the diesel limit, while the US
reﬁneries operate at the gasoline limit. In this study, on the other
hand, an attributional approach is applied where process energy
in each process unit is allocated to its products based on the prod-
ucts’ energy content. One could argue that, on the other hand, a
market-value-based allocation could in principle be more consis-
tent with the LP modeling approach since reﬁneries operate to
maximize proﬁt rather than energy efﬁciency. Elgowainy et al.
(2014) compared the product-speciﬁc efﬁciencies by a market-
value-based allocation with those by an energy-based allocation,
and observed no statistically signiﬁcant differences between them
for all reﬁned products (except for coke). This study also conducted
a process level market-value allocation, and found a similar trend
as shown in Fig. S5.
Fig. 2. Overall reﬁnery efﬁciency.
Fig. 3. Hydrogen consumption in kJ of hydrogen/MJ crude (Each box represent the
hydrogen from each source. ‘‘Purchase’’ refers to hydrogen produced outside of the
gate of the reﬁnery, typically external steam methane reformers [SMR]) while
‘‘SMR’’ refers to internal production of H2 through SMR of NG within the reﬁnery.
‘‘Reformer’’ refers to H2 from catalytic reformers).
J. Han et al. / Fuel 157 (2015) 292–298
295
Fig. 5 illustrates the energy intensities of petroleum products
for each group. Each bar denotes the contribution of each input
into the particular petroleum product. The energy intensity of a
given product is simply the aggregation of energy burdens (allo-
cated at the processing unit level) along the pathways that lead
to that product pool. The derivatives of crude and purchased HP,
as well as purchase butane and purchased blendstocks, comprise
the product pool. For example, HP, purchased in the form of heavy
gas oil or vacuum gas oil as a feed for the FCC, is processed into the
components of gasoline, distillate and residual oil, while purchased
butane and other blendstocks, such as reformate, alkylate and iso-
merate, are blended directly into the gasoline pool. Therefore, we
noticed that the sum of crude inputs—purchased HP, purchased
butane and blendstocks—are generally consistent, although the
compositions of the individual product pools are different. The dif-
ferent inputs that contribute to the individual product pools are
likely driven by the reﬁnery complexity and installed capacity of
process units and determined through reﬁnery optimization. For
example, the relatively smaller FCC reﬁning capacities in the High
API/High HP group result in a smaller contribution from purchased
HP relative to other groups throughout all products (see Table S3).
Notably, the FCC capacities in the Low API and High API/Low HP
reﬁnery groups are similar, affording similar contributions of pur-
chased HP in each reﬁnery group.
Consistent with the discussion above related to hydrogen con-
sumption, higher product-speciﬁc efﬁciencies and lower energy
intensities are observed in the High API/High HP reﬁnery groups,

total hydrogen consumption decreases signiﬁcantly as API gravity

and HP increase. As discussed above, Low API reﬁneries have much

larger hydrocrackers (HYK), one of the largest consumers of hydro-

gen in reﬁneries [12]. Removal of crude sulfur content also con-

sumes a large amount of hydrogen. Table S2 shows that average

crude sulfur content decreases monotonically from Low API to

High API reﬁneries, which drives the reduction in hydrogen con-

sumption. Hydrogen demand in the High API/High HP group of

reﬁneries is reduced further because the sulfur removal require-

ment via hydroprocessing in HP is low relative to gasoline and

distillate.

Fig. 3 also shows that the amount of hydrogen from SMR

decreases signiﬁcantly as API gravity and HP yield increase. A con-

sequence of this lower share of hydrogen from the SMR results in a

higher overall energy efﬁciency for the High API/High HP reﬁnery

group because hydrogen consumption via SMR is relatively inefﬁ-

cient (70% efﬁciency) compared to other reﬁnery units, resulting

in signiﬁcant energy burdens for products of hydrocracking and

hydrotreating units. Hydrogen is also a co-product of catalytic

reforming, which produces high-octane reformate that contributes

to the gasoline pool. Thus, hydrogen originating from catalytic

reformers has a signiﬁcantly lower energy burden relative to

hydrogen produced from the SMR.

3.2. Product-speciﬁc efﬁciency

Fig. 4 shows the calculated average and variation of product-

speciﬁc efﬁciencies for each group of reﬁneries using the energy

allocation method. The product-speciﬁc efﬁciency for all products

in the High API/High HP group are consistently higher than the

other two reﬁnery groups, mainly due to more favorable crude

quality, higher HP yields and lower complexity. These results are

consistent with those recently reported by Elgowainy et al. [12],

which showed (1) among reﬁnery products, gasoline has the low-

est efﬁciency, (2) RFO has the highest efﬁciency, and (3) diesel can

display a wide range of efﬁciencies. In the latter case, Forman et al.

[9] showed that tighter regulation of aromatics in CARB diesel

combined with reﬁneries that utilize multiple inefﬁcient units via

deep-conversion pathways can result in relatively low diesel efﬁ-

ciency. Although noted only for California reﬁneries, its impact

can exacerbate the already wide range of diesel efﬁciencies in

reﬁneries outside California [9], in general due to the relatively

inefﬁcient diesel reﬁning pathways. Interestingly, HP yield has a

much larger impact on the reﬁning efﬁciency of RFO compared to

the impact of API gravity. The lower reﬁning efﬁciency of RFO with

lower HP yield is likely due to the larger share of HP components

from downstream processes (e.g., HYK and coker), which carry lar-

ger energy and emission burdens.

It is important to note that the estimation of product-speciﬁc

efﬁciencies (as well as energy intensities) depends on allocation

approaches. As mentioned earlier, a marginal approach employed

in the JRC study results in a lower reﬁning efﬁciency (or higher

energy intensity) of diesel than of gasoline in the EU reﬁneries

because the EU reﬁneries operate at the diesel limit, while the US

reﬁneries operate at the gasoline limit. In this study, on the other

hand, an attributional approach is applied where process energy

in each process unit is allocated to its products based on the prod-

ucts’ energy content. One could argue that, on the other hand, a

market-value-based allocation could in principle be more consis-

tent with the LP modeling approach since reﬁneries operate to

maximize proﬁt rather than energy efﬁciency. Elgowainy et al.

(2014) compared the product-speciﬁc efﬁciencies by a market-

value-based allocation with those by an energy-based allocation,

and observed no statistically signiﬁcant differences between them

for all reﬁned products (except for coke). This study also conducted

a process level market-value allocation, and found a similar trend

as shown in Fig. S5.

Fig. 2. Overall reﬁnery efﬁciency.

Fig. 3. Hydrogen consumption in kJ of hydrogen/MJ crude (Each box represent the

hydrogen from each source. ‘‘Purchase’’ refers to hydrogen produced outside of the

gate of the reﬁnery, typically external steam methane reformers [SMR]) while

‘‘SMR’’ refers to internal production of H2 through SMR of NG within the reﬁnery.

‘‘Reformer’’ refers to H2 from catalytic reformers).

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

295

Fig. 5 illustrates the energy intensities of petroleum products

for each group. Each bar denotes the contribution of each input

into the particular petroleum product. The energy intensity of a

given product is simply the aggregation of energy burdens (allo-

cated at the processing unit level) along the pathways that lead

to that product pool. The derivatives of crude and purchased HP,

as well as purchase butane and purchased blendstocks, comprise

the product pool. For example, HP, purchased in the form of heavy

gas oil or vacuum gas oil as a feed for the FCC, is processed into the

components of gasoline, distillate and residual oil, while purchased

butane and other blendstocks, such as reformate, alkylate and iso-

merate, are blended directly into the gasoline pool. Therefore, we

noticed that the sum of crude inputs—purchased HP, purchased

butane and blendstocks—are generally consistent, although the

compositions of the individual product pools are different. The dif-

ferent inputs that contribute to the individual product pools are

likely driven by the reﬁnery complexity and installed capacity of

process units and determined through reﬁnery optimization. For

example, the relatively smaller FCC reﬁning capacities in the High

API/High HP group result in a smaller contribution from purchased

HP relative to other groups throughout all products (see Table S3).

Notably, the FCC capacities in the Low API and High API/Low HP

reﬁnery groups are similar, affording similar contributions of pur-

chased HP in each reﬁnery group.

Consistent with the discussion above related to hydrogen con-

sumption, higher product-speciﬁc efﬁciencies and lower energy

intensities are observed in the High API/High HP reﬁnery groups,