REVIEW www.rsc.

org/ees | Energy & Environmental Science

Biofuels and synthetic fuels in the US and China: A review of Well-to-Wheel

energy use and greenhouse gas emissions with the impact of land-use change
Xiaoyu Yan,* Oliver R. Inderwildi and David A. King
Received 3rd August 2009, Accepted 14th October 2009
First published as an Advance Article on the web 12th November 2009
DOI: 10.1039/b915801d

Alternative transportation fuels are projected to grow substantially due to energy security concerns
especially in the US and China. Moreover, some of these fuels can potentially reduce greenhouse
gas (GHG) emissions from the transportation sector and hence, can help to mitigate climate
change. We present a comprehensive review on Well-to-Wheel fossil fuel use and GHG emissions of
biofuels and synthetic fuels in the US and China including emissions from land-use change. Our results
are carefully benchmarked to the emissions caused by crude oil-derived fuels as well as synthetic
fuels from fossil feedstocks in order to estimate the potential emission reduction or increase. The
review strongly suggests that biofuels and synthetic fuels can contribute to GHG mitigation in the
transport sector only if appropriate feedstocks are used and emissions from land-use change are
minimised.

1. Introduction mitigation benefits in order to gain a more holistic view of the

US and China are the two biggest energy consumers and carbon
dioxide (CO2) emitters in the world, together accounting for
more than 40% of the world total and both depend heavily on oil
import to meet demand (see Table 1).1,2 Alternative fuels are
therefore of rapidly-growing interest especially in those two
countries due to energy security concerns and anthropogenic
climate change associated with greenhouse gas (GHG) emis-
sions.3 Well-to-Wheel (WtW) energy and GHG studies for bio-
fuels and synthetic fuels (so-called Fischer–Tropsch (FT) fuels4)
in the US and China are reviewed. Moreover, the impact of the
associated land-use change is analysed in context with the GHG
Low-Carbon Mobility Centre, Smith School of Enterprise and the
Environment, University of Oxford, Hayes House, 75 George Street,
Oxford, UK OX1 2BQ. E-mail: xiaoyu.yan@smithschool.ox.ac.uk; Tel:
+44 (0)186 561 4923
potential emission savings.
Reported values for WtW fossil fuel use (FWtW) and GHG
emissions (GWtW) are first normalised to ‘per MJ final fuel
product’ if they have not already been normalised in the reviewed
literature. Fuel properties listed in Table 2 are used when they are
not provided in the original studies. Ethanol in the US is assumed
to be used in the form of E10 (10% ethanol and 90% gasoline
blend by volume) for normal gasoline vehicles and E85 (85%
ethanol and 15% gasoline blend by volume) for flexible fuel
vehicles, while ethanol in China is solely used in the form of E10.
Biodiesel in both countries is assumed to be used in the form of
B20 (20% biodiesel and diesel blend by volume) while FT diesel
in its pure form in normal diesel vehicles. The Tank-to-Wheel
(TtW) fuel economy for passenger cars using different fuels is
shown in Table 3 and from these values FWtW and GWtW as ‘per
vehicle kilometre driven’ can be derived. The GHG emissions
induced by land-use change are allocated to ‘per MJ final fuel

Dr Xiaoyu Yan is a post- Dr Oliver Inderwildi is

doctoral researcher at Smith
School of Enterprise and the
Environment at the University of
Oxford. He received his PhD in
mechanical engineering from
Queen Mary, University of
London. His research interests
include: energy demand and
supply, environmental impacts
and policies related to the road
transportation sector; life cycle
assessment of road vehicle fuel/
Xiaoyu Yan propulsion options.
a research fellow at the Smith
School of Enterprise and the
Environment at the University of
Oxford, where he heads the
School’s Centre for Low-
Carbon Mobility. He is a Senior
Member of Linacre College,
Oxford University, and associ-
ated researcher at the Depart-
ment of Chemical Engineering,
University of Cambridge.
Oliver R: Inderwildi

190 | Energy Environ. Sci., 2010, 3, 190–197 This journal is ª The Royal Society of Chemistry 2010
Table 1 Primary energy demand in the US and China in 2008 (Million Table 4 Fuel ethanol production in the US in 20067
tonnes of oil equivalent)1
Ethanol Ethanol Feedstock Arable Ethanol
US China feedstock yield l/ha area/Mha land/Mha share (%)

Oil 885 376 Corn (98%) 3770 6.35 174 3.8

Natural gas 601 73 Sorghum (2%) 1365 0.28
Coal 565 1406
Nuclear 192 15
Hydro 57 132
Total 2299 2003 mainly produced by fermentation of corn (maize) on an
Net oil import 542 199 unprecedented scale—6.4 billion gallons or 19 million tonnes
(Mt) in 2007! Approximately 14% of US corn output was used
Table 2 Properties for fuels to produce ethanol, and yet this enormous volume of ethanol
represents only about 4% of US gasoline demand on an energy
Lower heating value TtW CO2 emissions basis.8 Biodiesel production in the US is growing rapidly
MJ kg 1
MJ l 1
g MJ 1
gl 1 and currently about 0.45 billion gallons (1.5 Mt) are produced.
Biofuel use in the US is projected to increase significantly in
Gasoline 44.8 33.15 69 2295 the future, driven largely by policies, regulations and incentives.
Diesel 43.4 37.76 73 2759 Further increases in corn ethanol are anticipated, but this
Ethanol 26.8 21.18 71 1513
Biodiesel 38.0 33.44 75 2494 growth will be restrained by the availability of arable land,
FT diesel 43.6 34.01 71 2431 water resources and the food vs. fuel trade-off.9 Substantial future
growth of ethanol production will thus depend on the develop-
ment of conversion routes for cellulosic feedstocks (see Fig. 1).8
Table 3 TtW fuel economy for passenger cars using different fuels (MJ The life cycle of corn ethanol produced in the US has been
km 1)a studied extensively in recent years.10–16 The majority of these
studies found that corn ethanol has a modest net energy gain and
Gasoline Diesel E10 E85 B20 FTD can reduce FWtW and GWtW compared with gasoline. On the other
2.600 2.166 2.544 2.455 2.148 2.166 hand Pimentel and Patzek found no net gain at all11 and these
results were challenged as the data used were outdated and
a
Note: average fuel economy for new gasoline passenger cars sold in the unrepresentative of current practices.14 It was for instance found
US and China were found to be nearly the same––2.6 MJ km 4; diesel
cars are assumed to be ‘functional-equivalent’ with gasoline cars;5 fuel that no-tillage practice in corn cultivation would reduce FWtW and
economy for cars fuelled with E10,5 E856 and B205 are based on GWtW because of reductions in fertilizer and fuel use as well as
relative changes to their gasoline or diesel counterparts; fuel economy organic carbon release from soil.10 Moreover, Wang et al. recently
for cars fuelled with FTD is assumed to be the same with diesel.
showed that GWtW can be significantly reduced if electricity and
heat for the ethanol plant were generated from biomass instead of
fossil fuels.16 In this review, only values that can represent the
product’ so that the impact of land-use change can be incorpo-
average current practice are included. In the worst case, GWtW for
rated into the GWtW analysis.
corn ethanol are marginally lower than gasoline and still only 35%
lower in the most optimistic case. The life cycle of sorghum
2. US ethanol on the other hand, was much less studied. A recent study
Ethanol is by far the most widely used biofuel for transportation has shown that sorghum ethanol is a viable option in the US with
worldwide and the US accounts for nearly half of the global a positive energy return.17 Biodiesel in the US is mainly produced
ethanol production (See Table 4).7 In the US, fuel ethanol is from soybean by transesterification. The existing literature
suggests that soybean biodiesel in the US can reduce FWtW and
GWtW considerably compared with diesel.12,18,19 Nevertheless, the
Sir David King is the Director of
impact of land-use change is not included in these studies.
the Smith School of Enterprise
Ethanol produced from cellulosic feedstocks is regarded as one
and the Environment at the
of the most promising biofuels but yet to be commercialised.
University of Oxford. He was
Cellulosic ethanol can reduce FWtW and GWtW significantly
the UK Government’s Chief
compared with gasoline, mainly because of the lower agricultural
Scientific Adviser and Head of
inputs and the electricity generated by burning the lignin
the Government Office of
part.13,14,10–26 Ethanol produced from agricultural residue such as
Science from 2000 to 2007. He is
corn stover can achieve negative FWtW and GWtW, indicating the
also Director of Research in
system is a net energy generator and removes carbon from the
Physical Chemistry at Cam-
atmosphere.13,22,27 Additionally, ethanol produced from dedi-
bridge University, President of
cated perennial plants such as switchgrass is also desirable due to
the Collegio Carlo Alberto in
the low FWtW and GWtW associated with fossil fuels.14,21,25,26
Turin, Italy, and Senior Science
David A: King
Another promising choice is ethanol derived from the cellulosic
Adviser to the investment bank,
portion of municipal solid waste (MSW) not only because it has
UBS.
much lower FWtW than gasoline and negative GWtW but also

This journal is ª The Royal Society of Chemistry 2010 Energy Environ. Sci., 2010, 3, 190–197 | 191
 Fig. 1 US renewable fuels production from 1999 to 2007 and requirements by 2022.8
because it provides an alternative to conventional disposal of
MSW such as incineration and landfill.23
FT fuels are receiving more attention in recent years because
they can be produced from a wide range of feedstocks such as
coal, natural gas and biomass, allowing for a gradual transition
between energy carriers.28 Compared with diesel, however, FT
diesel derived from coal (FTD-Coal) and natural gas (FTD-NG)
would have much higher FWtW and GWtW without carbon
capture and storage (CCS).21,28–30 Even with CCS and electricity
generated from non-fossil sources, GWtW for FTD-Coal and
FTD-NG would still be slightly higher than those for diesel.29 FT
diesel derived from biomass such as forestry residue (FTD-
Forestry residue) has GWtW that are considerably lower than
those for diesel, and if combined with CCS, it can effectively
remove carbon from the atmosphere.28 However, results incor-
porating CCS are excluded in the current review because it is still
a technology in its very early research stage and the energy-
penalty estimates for using it are often optimistic due to lack of
experimental research.31
FWtW and GWtW ‘per unit final fuel product’ for each type of
fuel in the US reported in the studies mentioned above are
Fig. 2 WtW fossil fuel use and GHG emissions ‘per unit final fuel product’ for each fuel in the US.
192 | Energy Environ. Sci., 2010, 3, 190–197
schematically shown in Fig. 2, where the data points represent
average values and the error bars represent possible ranges.
Estimates for gasoline and diesel are obtained and then used as
reference values for comparison.14,20,30,32,33
Since the expansion of biofuel production requires land to
grow the biomass feedstocks, land-use change is usually inev-
itable. There is a growing concern over the GHG emissions
from biofuel production through land-use change in recent
years as it can alter both the aboveground and soil carbon
stock of that land.34–36 Two types of land-use change caused by
biofuel production have to be considered: direct land-use
change (DLUC) and indirect land-use change (ILUC). DLUC
occurs when new agricultural land is taken into production for
biofuel purposes from a previous use patterns such as grass-
lands or forests. ILUC occurs when land currently used for
food or feed production is used for biofuel production while
the demand for the previous land-use remains. In this case, the
displaced agricultural production will be moved to other places
where unfavourable land-use change may occur.36 Converting
grasslands in the US to corn fields for bioethanol production
could cause huge carbon debt through DLUC.34 Diverting US
This journal is ª The Royal Society of Chemistry 2010
corn export to bioethanol production could potentially cause
even greater carbon debt through ILUC.35 However, DLUC
can also be beneficial if for example, abandoned or marginal
agricultural lands are converted to perennial grasses farmed for
bioethanol production.37 The results from two recent
studies34,35 were used to allocate carbon debt from land-use
change to ‘per MJ biofuel’ produced (see Table 5). Biofuels
from non-dedicated feedstocks are assumed not to be associ-
ated with any major land-use change. Together with the TtW
fuel economy data, the FWtW and GWtW ‘per vehicle kilometre
travelled’ for each fuel in the US incorporating land-use change
impact are derived and shown in Fig. 3. It can be observed that
even with the slight higher TtW energy efficiency, low-level
biofuel blends such as E10 and B20-soybean can only achieve
marginal reduction in FWtW and GWtW as opposed to gasoline
and diesel, respectively. E85-Corn can effectively reduce FWtW,
while the GHG reduction potential is still marginal. In
contrast, E85 containing cellulosic ethanol would substantially
reduce both. Compared with diesel, FTD-Coal has much
higher FWtW and GWtW while FTD-NG has much higher FWtW
and slightly higher GWtW. FTD-Forestry residue has nearly
zero FWtW and considerably lower GWtW. When biofuels from
dedicated feedstocks show dramatic changes. E85-Corn would
have higher GWtW than gasoline due to DLUC if abandoned
cropland or grassland were used to grow the corn. If the US do
not increase the land area for corn production but divert its
corn export to ethanol production, the ILUC effect would raise
the GWtW level for E85 to a similar level for FTD-Coal. On the
contrary, E85-Switchgrass can be carbon negative if the
switchgrass were grown on abandoned or marginal cropland.
However, E85-Switchgrass can also have higher GWtW than
gasoline if the switchgrass were grown on corn fields. B20-
soybean would have higher GWtW than diesel if the soybean
were grown on grassland. The impact of ILUC on US soybean
biodiesel is not yet available from existing literature but it is
expected to be significant, since the US is the largest soybean
exporter in the world.38 FTD derived from dedicated perennial
plants grown on abandoned or marginal cropland could also
be carbon negative due to DLUC, although future work is
needed to assess the potential.
Table 5 CO2 emissions for biofuels due to land-use change in the US
Carbon debt/t Debt allocated
Type of land-use change CO2 ha 1 to biofuel (%)
DLUC—grassland to corn ethanol 134a 85a
DLUC—abandoned cropland to 69a 85a
corn ethanol
DLUC—grassland to soybean 134c 39a
biodiesel
DLUC—abandoned cropland to 69a 100a
cellulosic ethanol
DLUC—marginal cropland to 108a 100a
cellulosic ethanol
ILUC—corn ethanol
ILUC—corn field to cellulosic
ethanol
Data sources: a ref. 34; b ref. 35; c assumed to be the same with grasslands converted to corn ethanol; d 30 years is chosen, after ref. 35, as the allocation
period because reductions of GHG in that period would be both difficult to achieve and important to avoid irreversible adverse effects from climate
change.
This journal is ª The Royal Society of Chemistry 2010
3. China
The transformation into a market economy and participation in
world trade has led China to sustained high rates of economic
growth for more than two decades, accompanied by rapidly
rising energy demands and CO2 emissions.39 Driven largely by
the soaring oil demand and the consequently increased imports,
China has been promoting biofuel production since 2002.5 Due
to the strong support from the government such as direct
subsidies, tax exemption and low interest loans, China’s fuel
ethanol production capacity reached 1.94 Mt by 2008, with four
of the five licensed fuel ethanol plants in operation using corn
and wheat as feedstocks and the remaining one using cassava.40
The Chinese government recently announced its intention to halt
building grain-based ethanol plants and to develop methods that
use non-food crops as feedstocks. This intention is most likely
based on the potential threat fuel conversion poses on the
national food security (so-called food vs. fuel trade-off).40,41
There is no reliable data for China’s biodiesel production due to
an information monopoly, but it is so far believed to be insig-
nificant compared to fuel ethanol production.41–43 Targets set for
fuel ethanol and biodiesel production in the ‘Medium- and Long-
Term Development Plan for Renewable Energy in China’ are 10
and 2 Mt by 2020, respectively. These official targets were
considered to be conservative and may very well be excee-
ded.40,41,43
Although a wide range of estimated FWtW and GWtW for both
corn ethanol44–48 and wheat ethanol46,48–50 have been reported,
they were generally found to be unfavourable compared with
gasoline. Sweet sorghum is considered to be a high-potential
feedstock in China’s ethanol industry40 and a 5000 t year 1
demonstration project has been established.43 However, sorghum
ethanol produced by current practices appears to have higher
FWtW and GWtW than gasoline.47 Sorghum ethanol will thus not
be able to contribute to GHG mitigation and is hence only
advantageous from the energy security point of view. Cassava is
identified as one of the most promising biofuel crops in China
because it can be grown on mountainous land or agriculturally
marginal land which is unsuitable for food crops and conse-
quently the feedstock farming would not compete with food
production.43,51 A number of recent life cycle studies have
Biofuel productivity/ Allocation periodd CO2 emissions g
MJ ha 1 year 1 years CO2/MJ
78317a 30 48
78317a 30 25
18141a 30 96
25861a 30 89
53818a 30 67
30 104b
30 111b
Energy Environ. Sci., 2010, 3, 190–197 | 193
 Fig. 3 WtW fossil fuel use and GHG emissions ‘per vehicle kilometre travelled’ for each fuel in the US.
confirmed that cassava ethanol can achieve reasonable reduc-
tions of GWtW up to 25% compared with gasoline.47,48,52–55
Sugarcane ethanol could be another option since its FWtW and
GWtW are much lower than those of gasoline.50 China’s sugarcane
production increased from 81 Mt in 1997 to 106 Mt in 2008, with
half of the increase from productivity improvements and the
other half from land expansion.56 However, further increase in
production will largely depend on productivity gains on existing
fields, because additional land suitable for sugarcane production
is extremely scarce.56 The potential for fuel ethanol production
from sugarcane is therefore considered to be limited and other
options have to be contemplated.
Two recent studies found quite differing results for soybean
biodiesel in terms of FWtW and GWtW,47,57 suggesting that further
clarifying work is still needed. Rapeseed biodiesel was shown to
have very low FWtW and GWtW.50,58 However, being a large net
importer of soybean and rapeseed,38 China may not be in
a position to produce biodiesel from these feedstocks. Jatropha is
regarded as a very promising biodiesel feedstock, because it
requires only minimal water and nutrients and is able to grow on
marginal, degraded, even desertified land.43 Life cycle studies
suggest that Jatropha biodiesel has a great potential in reducing
FWtW and GWtW compared with diesel.47,58 There are currently
a few small scale biodiesel plants in China with used cooking oil
as feedstock.43 Although used cooking oil derived biodiesel
shows promise to reduce GWtW,47 the potential for future
production is believed to be less than 2 Mt year 1.43
FT fuels produced from coal have been proposed as feasible
and practical options to replace petroleum-based fuels in China
both in the short and medium term mainly because of relatively
abundant coal resources, and commercial scale plants are already
in the design and construction phase.59,60 Unfortunately, FWtW
194 | Energy Environ. Sci., 2010, 3, 190–197
and GWtW for FTD-Coal in China61 are even higher than those in
the US29 most likely due to the higher grade of carbonisation of
grid electricity. FTD-NG is much better than FTD-Coal but still
worse than diesel.61 Furthermore, China has relatively limited
natural gas resources and most of the reserves are located far
away from demand39 which rules them out as major part of the
solution.
On the other hand, China has abundant cellulosic biomass
resources, most of which are crop straws and forestry residues
with an estimate of around 220–380 million tonnes of oil
equivalent available for bioenergy production each year.39,62 If
these resources can be successfully utilized for cellulosic ethanol
or FT fuel production, more than half of the energy demand in
China’s road transport sector in 20305 could be met by these
low-carbon fuels. China has been active in cellulosic ethanol
research for decades and the first collaborative research venture
to systematically study cellulosic ethanol technology is on-going
in collaboration with researchers at Oxford University, UK.40
Although some major technological difficulties in reducing the
cost of cellulosic ethanol have been solved, other problems such
as low conversion rate and high energy inputs still exist.43
Neither the biological route nor the chemical route from
cellulosic materials to ethanol is yet feasible compared with
fossil fuels.27 Future work should also assess the WtW fossil fuel
and GHG benefits of different fuels produced from cellulosic
feedstocks.
FWtW and GWtW ‘per unit final fuel product’ for each type of
fuel in China reported in the studies mentioned above are sum-
marised and shown in Fig. 4, where the data points represent
average values and the error bars represent possible ranges.
Estimates for gasoline and diesel are also obtained and then used
as reference values for comparison.46,47,55,57,61,69–71
This journal is ª The Royal Society of Chemistry 2010
 Fig. 4 WtW fossil fuel use and GHG emissions ‘per unit final fuel product’ for each fuel in China.

China has been experiencing intense land-use change for Table 6 Cultivated land-use change in China from 1991 to 2001 (in
decades. A comprehensive assessment of the land-use change millions of ha)63
patterns in China was conducted in a recent study.63 Table 6 Decrease Increase Net loss
shows the pattern for cultivated land-use change in China from
1991 to 2001. The main causes of cultivated land loss in China are Ecological landa 5.78 3.26
degradation due to over-intensive agriculture (converted to Desertified land 1.28 1.31
Salinised land 0.06 0.12
ecological land for conservation), non-agricultural land expan- Non-agricultural landb 2.68 0.24
sion and desertification. Although large areas of ecological land Total 9.80 4.93 4.87
such as forests and grasslands were converted to cropland in a
Forests, grasslands and wetlands. b Land areas such as urban built-up
order to offset the loss, there was still a net loss of nearly 5 Mha. areas and land used for transportation infrastructures.
In addition, the productivity of the newly cultivated lands was
much lower than those lost to non-agricultural use.63,72 There-
Table 7 Carbon debt induced by converting forest and grassland into
fore, biofuel production increase in China will most likely be
cropland in China (in t carbon ha 1)
associated with large GHG emissions through DLUC. The
carbon debt induced by converting forest and grassland into Forest Grassland
cropland in China is presented in Table 7. CO2 emissions for
biofuels due to DLUC are calculated and shown in Table 8. Biomass carbon 49a 10b
Soil carbon 136c 135d
FWtW and GWtW ‘per vehicle kilometre travelled’ for each fuel Soil carbon lost 45.8%e 9%f
in China incorporating land-use change impact are then derived Total carbon debt 111.3 22.2
and shown in Fig. 5. Compared with gasoline, E10-Sorghum has
Data sources: a ref. 64; b ref. 65; c ref. 66 & 67; d ref. 67; e ref. 34; f ref. 68.
higher FWtW and GWtW while E10-Corn and E10-Wheat both
have lower FWtW and slightly lower GWtW. E10-Cassava and
E10-Sugarcane are clearly more favourable than gasoline, while
B20 seems to be superior to diesel, with B20-Rapeseed and B20- blends considered—518 g CO2-eq km 1 (too high to show in
Jatropha outperforming B20-Soybean and B20-Used cooking Fig. 5) if the soybean were grown on deforested farmland and
oil. FTD-Coal and FTD-NG are way off the chart, with FWtW as still much higher than diesel if grown on grassland. GHG benefits
high as 5.8 and 5.0 MJ km 1 and GWtW being 569 and 263 g CO2- for B20-Rapeseed and B20-Jatropha would be gone when the
eq km 1, respectively. After land-use change impact is included, feedstocks were grown on deforested farmland and largely
GWtW for E10-Sorghum, E10-Corn and E10-Wheat are much diminished on grassland. On the other hand, GWtW for
higher than gasoline, whereas those for E10-Sugarcane and E10- E10-Cassava and B20-Jatropha could be much lower if the
Cassava would be still lower than gasoline if the feedstock were feedstocks were grown on marginal cropland or degraded land
grown on grassland but higher on deforested farmland. B20- because of the potential soil carbon sequestration. These should
Soybean would have the highest GWtW among all the biofuel be carefully assessed in future studies.

This journal is ª The Royal Society of Chemistry 2010 Energy Environ. Sci., 2010, 3, 190–197 | 195
Table 8 CO2 emissions for biofuels in China due to direct land-use change

CO2 emissions/g CO2-

eq MJ 1
Productivity/t Conversion rate Biofuel productivity/ Debt allocated Allocation
Feedstock ha 1 year 1 feedstock/fuel MJ ha 1 year 1 to biofuel (%) period/years Forest Grassland

Corn 7a 3.25a 57723 73a 30 172 34

Wheat 4.3b 3.65b 31573 66b 30 284 57
Sweet sorghum 64.5c 18.8c 91947 80c 30 118 24
Cassava 13.2d 2.8d 126343 84d 30 90 18
Sugarcane 69e 15e 123280 100f 30 110 22
Soybean 1.8c 5.9c 11593 72c 30 845 169
Rapeseed 1.8g 3.3g 20727 50g 30 328 65
Jatropha 5c 3.3c 57576 60c 30 142 28
Data sources: a average values in ref. 44, 47 & 48; b ref. 48 & 49; c ref. 47; d average values in ref. 47, 48, 54 & 55; e ref. 41 & 56; f assumed to be 100 due to
lack of data; g ref. 73 & 74.

Fig. 5 WtW fossil fuel use and GHG emissions ‘per vehicle kilometre travelled’ for each fuel in China.

4. Conclusions
We present a comprehensive review on Well-to-Wheel fossil fuel
use (FWtW) and GHG emissions (GWtW) of biofuels and
synthetic fuels in the US and China including emissions from
land-use change. Ethanol produced from grains offers very
limited FWtW and GWtW reductions compared with gasoline.
When the effect of land-use change is included GWtW could even
be much higher! Ethanol derived from high-yielding crops such
as cassava or sugarcane on the other hand has the potential to
reduce FWtW and GWtW, however, this potential is diminished if
forests were cleared to grow the feedstocks. Although soybean,
rapeseed and jatropha biodiesel appear to be favourable
compared with diesel at first glance, their GWtW are increased
significantly through land-use change. The potential for bio-
diesel production from used cooking oil will certainly remain
marginal. Fischer–Tropsch diesel derived from coal (FTD-
Coal) is the most undesirable alternative to diesel, with up to
double the GWtW. FT diesel derived from natural gas (FTD-
NG) is somewhat better than FTD-Coal but still considerably
worse than diesel. Cellulosic ethanol and FT fuels produced
from cellulosic biomass could offer opportunities for reducing
fossil fuel use and GHG emissions substantially without dis-
turbing land-use patterns and affecting food supply adversely.
It is however extremely unlikely that a major fraction of the
current liquid fossil fuel demand can be catered for by biofuels
and synthetic fuels in general.
The present review strongly suggests that biofuels and
synthetic fuels in the US and China can contribute to GHG
mitigation in the transport sector only if appropriate feedstocks
are used and emissions from land-use change are minimised. It is
thus highly recommended that future studies on the life cycle of
biofuels and synthetic fuels should include the impact of direct
and indirect land-use change as in most cases it determines to
a large extent the GHG mitigation potentials.

196 | Energy Environ. Sci., 2010, 3, 190–197 This journal is ª The Royal Society of Chemistry 2010
Acknowledgements
This work is part of the Future of Mobility project, conducted at
the Smith School of Enterprise and the Environment, University
of Oxford, and funded by Shell International Petroleum Co. Ltd.
We are grateful to Jack Jacometti, VP Future Fuels, Stewart
Kempsell, GM Business Development, Future Fuels & CO2,
Steve Skippon, Scientist, Shell Technology Centre, and Sylvia
Williams, Business Development Manager, Global XTL Devel-
opment, for fruitful discussions.
Notes and references
1 BP, BP Statistical Review of World Energy June 2009, BP, London,
2009.
2 IEA, Key World Energy Statistics 2008, OECD, Paris, 2008.
3 G. Walker and D. A. King, The Hot Topic: What We Can Do about
Global Warming, Harvest Books, London, 2008.
4 O. R. Inderwildi, S. J. Jenkins and D. A. King, Angew. Chem., Int.
Ed., 2008, 47, 5253–5255.
5 X. Yan and R. J. Crookes, Energy Policy, 2009, 37, 658–668.
6 M. C. Roberts, Energy Policy, 2008, 36, 1233–1235.
7 M. Balat and H. Balat, Appl. Energy, 2009, 86, 2273–2282.
8 S. K. Hoekman, Renewable Energy, 2009, 34, 14–22.
9 T. Shirvani, O. R. Inderwildi and D. A. King, SSEE Review 0002,
Oxford University, ISSN 2041–5028, DOI: 10.4210/ssee.res.2009.
0002.
10 S. Kim and B. E. Dale, Biomass Bioenergy, 2005, 28, 475–489.
11 D. Pimentel and T. W. Patzek, Nat. Resour. Res., 2005, 14, 65–76.
12 J. Hill, E. Nelson, D. Tilman, S. Polasky and D. Tiffany, Proc. Natl.
Acad. Sci. U. S. A., 2006, 103, 11206–11210.
13 A. Lavigne and S. E. Powers, Energy Policy, 2007, 35, 5918–5930.
14 A. E. Farrell, R. J. Plevin, B. T. Turner, A. D. Jones, M. O’Hare and
D. M. Kammen, Science, 2006, 311, 506–508.
15 S. Kim and B. E. Dale, Bioresour. Technol., 2008, 99, 5250–5260.
16 M. Wang, M. Wu and H. Huo, Environ. Res. Lett., 2007, 2, 024001.
17 D. Wang, S. Bean, J. McLaren, P. Seib, R. Madl, M. Tuinstra, Y. Shi,
M. Lenz, X. Wu and R. Zhao, J. Ind. Microbiol. Biotechnol., 2008, 35,
313–320.
18 J. Sheehan, V. Camobreco, J. Duffield, G. M.H. Shapouri, Life Cycle
Inventory of Biodiesel and Petroleum Diesel for Use in an Urban Bus,
National Renewable Energy Laboratory, 1998.
19 H. Huo, M. Wang, C. Bloyd and V. Putsche, Environ. Sci. Technol.,
2009, 43, 750–756.
20 GM, Well-to-Wheel energy use and greenhouse gas emissions of
advanced fuel/vehicle systems – North American analysis, 2001.
21 J. S. Fleming, S. Habibi and H. L. MacLean, Transp. Res., Part D:
Transp. Environ., 2006, 11, 146–159.
22 J. Sheehan, A. Aden, K. Paustian, K. Killian, J. Brenner, M. Walsh
and R. Nelson, J. Ind. Ecol., 2003, 7, 117–146.
23 Y. Kalogo, S. Habibi, H. L. Maclean and S. V. Joshi, Environ. Sci.
Technol., 2007, 41, 35–41.
24 M. Wu, Y. Wu and M. Wang, Biotechnol. Prog., 2006, 22, 1012–1024.
25 M. R. Schmer, K. P. Vogel, R. B. Mitchell and R. K. Perrin, Proc.
Natl. Acad. Sci. U. S. A., 2008, 105, 464–469.
26 S. Spatari, Y. M. Zhang and H. L. MacLean, Environ. Sci. Technol.,
2005, 39, 9750–9758.
27 O. R. Inderwildi and D. A. King, Energy Environ. Sci., 2009, 2, 343–
346.
28 O. P. R. van Vliet, A. P. C. Faaij and W. C. Turkenburg, Energy
Convers. Manage., 2009, 50, 855–876.
29 P. Jaramillo, W. M. Griffin and H. S. Matthews, Environ. Sci.
Technol., 2008, 42, 7559–7565.
30 M. P. Hekkert, F. Hendriks, A. P. C. Faaij and M. L. Neelis, Energy
Policy, 2005, 33, 579–594.
31 S. Page, A. Williamson and I. Mason, Energy Policy, 2009, 37, 3314–
3324.
32 M. Wang, H. Lee and J. Molburg, Int. J. Life Cycle Assess., 2004, 9,
34–44.
33 A. Schafer, J. B. Heywood and M. A. Weiss, Energy, 2006, 31, 2064–
2087.
This journal is ª The Royal Society of Chemistry 2010
34 J. Fargione, J. Hill, D. Tilman, S. Polasky and P. Hawthorne, Science,
2008, 319, 1235–1238.
35 T. Searchinger, R. Heimlich, R. A. Houghton, F. X. Dong,
A. Elobeid, J. Fabiosa, S. Tokgoz, D. Hayes and T. H. Yu, Science,
2008, 319, 1238–1240.
36 F. Cherubini, N. Bird, A. Cowie, G. Jungmeier, B. Schlamadinger and
S. Woess-Gallasch, Resour., Conserv. Recycl., 2009, 53, 434–447.
37 D. Tilman, R. Socolow, J. A. Foley, J. Hill, E. Larson, L. Lynd,
S. Pacala, J. Reilly, T. Searchinger, C. Somerville and R. Williams,
Science, 2009, 325, 270–271.
38 US Department of Agriculture, Oilseeds: World markets and trade,
http://www.fas.usda.gov/oilseeds/circular/2009/June/oilseedsfull0609.
pdf.
39 X. Yan and R. J. Crookes, J. Energy Inst., 2007, 80, 110–115.
40 S. Li and C. Chan-Halbrendt, Appl. Energy, 2009, 86, S162–S169.
41 H. Yang, Y. Zhou and J. G. Liu, Energy Policy, 2009, 37, 1876–1885.
42 A. Zhou and E. Thomson, Appl. Energy, 2009, 86, S11–S20.
43 C. Wu, X. Yin, Z. Yuan, Z. Zhou and X. Zhuang, Energy, 2009, DOI:
10.1016/j.energy.2009.04.006.
44 Z. Zhang and X. Yuan, Huanjing Kexue, 2006, 27, 437–441.
45 Z. Zhang and X. Yuan, Huanjing Kexue, 2006, 27, 616–619.
46 Z. Hu, P. Tan and D. Lou, J. Tongji Univ. (Nat. Sci.), 2007, 35, 1099–
1103.
47 X. Ou, X. Zhang, S. Chang and Q. Guo, Appl. Energy, 2009, 86,
S197–S208.
48 S. R. Yu and J. Tao, Energy, 2009, 34, 476–484.
49 X. B. Dong, S. Ulgiati, M. C. Yan, X. S. Zhang and W. S. Gao,
Energy Policy, 2008, 36, 3882–3892.
50 X. Yan and R. Crookes, Renewable Sustainable Energy Rev., 2009, 13,
2505–2514.
51 C. Jansson, A. Westerbergh, J. Zhang, X. Hu and C. Sun,
Appl.Energy, 2009, 86, S95–S99.
52 S. R. Yu and J. Tao, Energy, 2009, 34, 22–31.
53 Z. Y. Hu, F. Fang, D. F. Ben, G. Q. Pu and C. T. Wang, Appl. Energy,
2004, 78, 247–256.
54 R. Leng, C. T. Wang, C. Zhang, D. Dai and G. Q. Pu, J. Clean. Prod.,
2008, 16, 374–384.
55 D. Dai, Z. Y. Hu, G. Q. Pu, H. Li and C. T. Wang, Energy Convers.
Manage., 2006, 47, 1686–1699.
56 G. Kostka, C. Polzin and J. Scharrer, Appl. Energy, 2009, 86, S100–
S107.
57 Z. Y. Hu, P. Q. Tan, X. Y. Yan and D. M. Lou, Energy, 2008, 33,
1654–1658.
58 Z. Hu, P. Tan, D. Lou and D. Y., Trans. CSAE, 2006, 22, 141–146.
59 X. Hao, G. Q. Dong, Y. Yang, Y. Y. Xu and Y. W. Li, Chem. Eng.
Technol., 2007, 30, 1157–1165.
60 L. Zhang and Z. Huang, Energy, 2007, 32, 1896–1904.
61 W. Shen, A. L. Zhang and W. J. Han, in 13th International Conference
on Management Science and Engineering, Lille, France, Editon edn.,
2006, pp. 1735–1739.
62 X. Li, Y. Huang, J. Gong and X. Zhang, Energy, 2009, DOI: 10.1016/
j.energy.2009.04.018.
63 K. Zhang, Z. Yu, X. Li, W. Zhou and D. Zhang, Land Degrad. Dev.,
2007, 18, 209–219.
64 J. Y. Fang, A. P. Chen, C. H. Peng, S. Q. Zhao and L. Ci, Science,
2001, 292, 2320–2322.
65 J. W. Fan, H. P. Zhong, W. Harris, G. R. Yu, S. Q. Wang, Z. M. Hu
and Y. Z. Yue, Clim. Change, 2008, 86, 375–396.
66 X. Q. Zhang and D. Y. Xu, Environ. Sci. Policy, 2003, 6, 421–432.
67 Z. B. Xie, J. G. Zhu, G. Liu, G. Cadisch, T. Hasegawa, C. M. Chen,
H. F. Sun, H. Y. Tang and Q. Zeng, Global Change Biol., 2007, 13,
1989–2007.
68 Z. P. Wang, X. G. Han and L. H. Li, J. Environ. Manage., 2008, 86,
529–534.
69 Z. Huang and X. Zhang, J. Tongji Univ., 2003, 31, 1472–1476.
70 W. Shen, A. Zhang and W. Han, Natural Gas Ind., 2006, 26, 148–152.
71 D. Karman, IEEE EIC Climate Change Conference, Ottawa, Canada,
2006.
72 H. M. Yan, J. Y. Liu, H. Q. Huang, B. Tao and M. K. Cao, Fall
Annual Meeting of the American-Geophysical-Union, San Francisco,
CA, 2009.
73 X. Yan, PhD Thesis, University of London, 2008.
74 P. Janulis, Renewable Energy, 2004, 29, 861–871.
Energy Environ. Sci., 2010, 3, 190–197 | 197