Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Estimation of Liquid Fuel Yields From Biomass: Environ. Sci. Technol. 2010, 44, 5298-5305

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

Environ. Sci. Technol.

2010, 44, 52985305

Estimation of Liquid Fuel Yields from


Biomass
NAVNEET R. SINGH,
W. NICHOLAS DELGASS,
FABIO H. RIBEIRO, AND
RAKESH AGRAWAL*
School of Chemical Engineering and Energy Center at
Discovery Park, Purdue University, West
Lafayette, Indiana 47907

Received January 29, 2010. Revised manuscript received


April 18, 2010. Accepted May 7, 2010.

We have estimated sun-to-fuel yields for the cases when


dedicated fuel crops are grown and harvested to produce liquid
fuel. The stand-alone biomass to liquid fuel processes, that
use biomass as the main source of energy, are estimated to
produce one-and-one-half to three times less sun-to-fuel yield
than the augmented processes. In an augmented process,
solar energy from a fraction of the available land area is used
to produce other forms of energy such as H2, heat etc.,
which are then used to increase biomass carbon recovery in
the conversion process. However, even at the highest
biomass growth rate of 6.25 kg/m2 y considered in this study,
the much improved augmented processes are estimated to
have sun-to-fuel yield of about 2%. We also propose a novel standalone H2Bioil-B process, where a portion of the biomass is
gasified to provide H2 for the fast-hydropyrolysis/hydrodeoxygenation of the remaining biomass. This process is estimated to
be able to produce 125-146 ethanol gallon equivalents (ege)/
ton of biomass of high energy density oil but needs experimental
development. The augmented version of fast-hydropyrolysis/
hydrodeoxygenation, where H2 is generated from a nonbiomass
energy source, is estimated to provide liquid fuel yields as
high as 215 ege/ton of biomass. These estimated yields provide
reasonable targets for the development of efficient biomass
conversion processes to provide liquid fuel for a sustainable
transport sector.

Introduction
The liquid hydrocarbons currently used by cars, buses, trucks,
trains, and airplanes provide high volumetric energy density
fuels along with the associated ease of use. However, for the
era when fossil fuels will either be limited or when the
transition will have to be made to a sustainable economy in
which energy needs are primarily met with renewable
resources, we need to critically examine the viability of
meeting the needs of the entire transportation sector with
resources such as biomass, electricity, hydrogen, etc.
For both electricity and H2, the high energy density
methods for on-board storage are currently unavailable, and
the search for solutions is proving to be quite challenging
(1, 2). In the meantime, with the given energy density of
batteries, partial use of electricity through plug-in hybrid
electric vehicles (PHEVs) still requires large quantities of liquid
* Corresponding author telephone: (765) 494-2257; fax: (765) 4940805; e-mail: agrawalr@purdue.edu.
5298

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 13, 2010

hydrocarbon fuels to satisfy the remainder of the transportation sector needs (3). For example, replacement of the current
internal combustion engine based light duty vehicles (LDVs)
in the United States with PHEVs containing batteries that
can provide a driving distance of about 64 km (PHEV40)
between successive charges would decrease the liquid fuel
required for the LDVs from 8.9 to 3.4 million barrels per
day (Mbbl/d) (3). In spite of this huge reduction, the
remaining oil demand for the total U.S. transportation sector,
including heavy duty vehicles (HDVs) such as air planes,
trucks, trains, etc. will still be quite large at 8.3 Mbbl/d (4, 5).
Replacing this massive remaining oil need with electricity or
H2 would require huge development in alternate infrastructure along with major technical innovations.
A simple fact that favors the use of biomass for transportation fuel is that a sizable quantity is available as
sustainable waste biomass, e.g. waste coproduct from food
crop plants, forest residues, etc. Recent studies estimate that
such sustainably available waste (SAW) biomass in the USA
could be 349 million Tons/y or more (6, 7). A recent NRC
report estimates that, with the appropriate use of the land
under the U.S. Conservation Reserve Program, 149 million
Tons/y of dedicated fuel crops could be produced by 2020
(6). If this land is readily available as spare land with no other
competing demand, and the dedicated fuel crops could be
collected with minimal energy input, then a total of 498
million Tons/y of biomass could potentially be sustainably
available in the USA. However, this amount of biomass will
provide only about one-fifth to one-fourth of the 8.3 Mbbl/d
of oil with the currently practiced biomass to liquid fuel
technologies (6-8). Clearly, compared to the quantity of oil
needed, the amount of sustainably available biomass is
limited.
If additional land is to be used to grow dedicated biomass
for liquid fuel, then it is informative to compare the efficiency
at which incident solar energy is harvested in dedicated fuel
crops vs other secondary forms of energy such as H2, heat,
and electricity that could be recovered from the same land
area. In most cases, less than 1% of the incident solar energy
is stored in biomass (1-3). Even at the upper expected range
of biomass growth rate of 6 kg/m2 y, the efficiency of
collection of solar energy as biomass energy will be in the
neighborhood of 2% (9). Inspection of the efficiencies of
the currently available technologies tells us that generally
the recovery of solar energy as biomass is an order of
magnitude lower than that of sun to heat (50-70%),
electricity (10-42%), or hydrogen (5-27%) (10-14). Therefore, when compared to other solar energy conversion and
recovery processes such as electricity or H2, this translates
into a need for larger land area to recover the same amount
of solar energy as biomass (8).
Due to the relatively low efficiencies for growing dedicated
biomass for transportation fuel and the limited availability
of SAW biomass, there is a need to critically examine the
amount of liquid fuel that can be produced from a given
quantity of biomass. First, we need to understand the
potential of self-contained process where biomass is the main
source of energy and liquid fuel production is maximized.
Second, we need to develop augmented processes that will
synergistically use supplemental energy available at much
higher efficiencies to increase liquid fuel production. This
will provide us the potential to efficiently increase liquid fuel
production from a given quantity of biomass and thereby an
assessment of the degree to which biomass can play a role
in a transportation fuel infrastructure.
10.1021/es100316z

2010 American Chemical Society

Published on Web 06/07/2010

FIGURE 1. Estimated values of the overall annual biofuel yield from 1 m2 of land area with annual solar incident energy of 6307 MJ/
m2 y. For each process, literature reported or estimated conversion efficiency values are shown in the parentheses. All yield
numbers are for high energy density liquid fuel with the exception of fast-pyrolysis which is for low-energy density bio-oil.
In this paper, we estimate the annual solar incident energy
on a given land area that could be recovered as high energy
density liquid fuel for dedicated fuel crop cases. These
calculated sun-to-fuel (S2F) yields for different processes
provide us the much needed overall efficiencies of converting
solar energy to liquid fuel. We evaluate self-contained as
well as augmented thermochemical processes, where supplementary energy is used to increase the S2F yield from a given
land area (8, 15). This information is useful in assessing how
to maximize liquid fuel production from a given land area.
We also propose a novel self-contained biomass to liquid
fuel thermochemical process. This process, when experimentally developed, has the potential to provide significantly
greater S2F yield of high energy density oil than any current
self-contained thermochemical process.
An additional benefit of this S2F yield study is that the
potential of different processes can also be compared in terms
of the quantity of liquid fuel that could be produced from
a unit quantity of biomass fed to each conversion process.
This is helpful for understanding of how to increase the liquid
fuel production from a given quantity of SAW biomass.
Even though we discuss increased fuel production in the
context of thermochemical processes, it should be emphasized that the calculated liquid yields provide us with potential
liquid fuel yields against which all suitable processes,
including those based on biochemical routes, can be
measured. It is not our purpose to imply that thermochemical
routes are superior to other alternative biomass conversion
routes.
Liquid Fuel Yield from Self-contained Biofuel Processes.
We briefly discuss here the existing as well as some novel
self-contained biomass to liquid fuel processes for their liquid

fuel yield. We define a self-contained process as the one


where very little, if any, energy is used from a nonbiomass
source. The calculation details for all the processes are given
in the Supporting Information (SI).
In Figure 1, we show estimated values of the overall annual
oil yield by different processes using dedicated fuel crops
and the total sunlight falling on 1 m2 of land area. The results
are based on an average annual incidence of 6307 MJ/m2 y
(1752 kW h/m2 y) of solar energy in the USA (8, 16, 17).
While calculations were done for four different collection
rates of biomass: 1.5, 3, 5, and 6.25 kg/m2 y, the results
reported in Figure 1 are for the biomass collection rate of 3
kg/m2 y. By collection rate, we refer to the annual amount
of biomass that arrives at a processing plant from 1 m2 of
land area. In one scenario, a fraction of the land area may
be assigned to directly recover additional energy needed to
grow, harvest, and transport the biomass to the plant gate.
If this additional energy is a small fraction of the biomass
energy arriving for processing, then the biomass collection
rate will be close to the biomass growth rate. Alternatively,
if most of the additional energy is utilized as electricity and
heat, then since these forms of energy are recovered from
solar energy at an order of magnitude higher efficiency than
biomass, the fraction of the land area dedicated for the
additional energy will be relatively small. In this case, the
biomass collection rate will again be nearly equal to the biomass
growth rate. Thus, in Figure 1, for a biomass energy content
of 17 MJ/kg, a collection rate of 3 kg/m2 y implies that from
the total 6307 MJ/m2 y of solar energy, 51 MJ/m2 y are
available as biomass energy at the plant for further processing
VOL. 44, NO. 13, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

5299

TABLE 1. Estimated Carbon Loss for Biomass to High Energy Density Liquid Fuela
biomass

energy content
(MJ/kg)

energy content
(kJ/mol C)

carbon loss with 100%


efficient conversion (%)

assumed conversion
efficiency (%)

total carbon
loss (%)

switchgrass
poplar
sugar

17.2
19.6
14.1

485
455
423

24.6
32.8
42.8

75
75
97

43.4
49.6
44.5

The energy content of gasoline ) 604.1 kJ/mol C.

to produce liquid fuel. In this paper, we report all energy


contents as lower heating values (LHV) unless specified
otherwise.
From the available biomass energy, the liquid fuel yield
is calculated based on overall process conversion efficiency.
This means that in our S2F calculations, we do not account
for any auxiliary power that the plant may use. It is assumed
that such auxiliary power is either absent or a small fraction
of the total rate at which energy is processed within a plant.
Thus, in a thermochemical plant, major energy inputs such
as the heat needed to dry the biomass, electricity needed to
circulate hot gases, etc. are produced from the biomass feed
to the plant. Similarly, for a fermentation based biochemical
plant, the heat energy needed for separation of ethanol or
butanol is assumed to be supplied from the lignin portion
of the biomass. If within a chemical process, excess energy
is generated which cannot be converted to liquid fuel, we do
not take credit for such energy in the S2F calculations. Our
S2F calculations do account for all the major energy
conversion steps, and the numbers in Figure 1 provide a
good representation of the efficiency of each process in
recovering annual incident solar energy as liquid fuel.
Another useful piece of information that can also be
derived from the S2F calculations in Figure 1 is the amount
of liquid produced per ton of biomass. For each process,
information regarding the amount of biomass used and the
corresponding energy content of the fuel produced is
available. Since all processes do not produce the same
product, a convenient unit to compare processes on the same
basis is ethanol gallon equivalent (ege) per ton of biomass
(18). For this calculation, we divide the energy content of the
fuel produced in Figure 1 by the lower heating value (LHV)
of 80.14 MJ/gal of ethanol to get equivalent gallons of ethanol.
This number is then divided by the amount of biomass to
get ege per ton of biomass. This information is useful as it
allows us to compare different processes on the basis of
biomass conversion to liquid fuel. This is particularly
important when the available biomass is limited.
When biomass is processed through a conventional
gasifier followed by a Fischer-Tropsch diesel (FTD) process,
the overall process efficiency of the chemical conversion
process can vary between 41 and 50% (19). This leads to a
recovery of 20.9-25.5 MJ/m2 y of energy as biofuel, which
also translates into 87-106 ege of liquid fuel per ton of
biomass. For the biochemical routes using enzymes and
microbes, the reported efficiency values for the lignocellulosic
biomass is in the 35-50% range with 75-105 ege/ton of
biomass (18, 20, 21). This implies that energy content of the
recovered ethanol is similar to that of diesel produced in the
gasification/FT route.
On the other hand, conventional fast-pyrolysis is quite
efficient (65-77%) (22, 23) in converting biomass to a lowenergy density liquid fuel referred to as bio-oil and recovers
33.1-39.5 MJ/m2 y of solar incidence energy. The bio-oil
from fast-pyrolysis has extremely high oxygen content
(35-40 wt %), however, and its energy content is only half
that of petroleum and similar to that of the original biomass
(17 MJ/kg). Bio-oils do not easily blend with petroleum
products and tend to polymerize and condense with time
during shipment and lead to gumming in downstream
5300

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 13, 2010

reactors. This necessitates the subsequent upgrading of this


bio-oil by hydrodeoxygenation (HDO) using H2 in the
presence of a catalyst (15, 24, 25). A recent study by Pacific
Northwest National Laboratory (PNNL) reports modeling
results for a three step process of fast-pyrolysis, hydrodeoxygenation, and hydrocracking to produce gasoline and
diesel-like liquid fuel from lignocellulosic biomass (26).
One of our goals in this paper is to study processes that
have a potential to eliminate three-step processing and the
handling challenges associated with the fast-pyrolysis based
route for producing upgraded oil. For this purpose, we will
first discuss augmented processes where H2 needed to remove
oxygen from the biomass to directly produce high energy
density oil is supplied from a nonbiomass source.
Liquid Fuel Yields from Augmented Biofuel Processes
Using Supplementary Hydrogen. A key feature of all the
three self-contained processes discussed so far is that biomass
is the sole energy source being converted to liquid fuel. As
seen from Table 1, this inherently limits the biomass carbon
atoms that can be recovered as high energy density liquid
fuel.
The energy content on a per carbon basis for biomass
sources such as Switchgrass, Poplar, and sugars is only twothirds of the energy content of the molecules composing
gasoline. This means that when biomass carbon molecules
are upgraded to a high energy density liquid fuel such as
gasoline, even for a 100% energy efficient conversion process,
about one-third of the biomass carbon atoms will be rejected
as low energy molecules such as CO2. With reasonably
optimistic process conversion efficiencies, we find from Table
1 that roughly only half of the biomass carbon is recovered
as high energy density liquid fuel. Most likely, the lost carbon
will be in its low-energy state as CO2. Considering the fact
that solar energy to biomass carbon is a low-efficiency
process, it is attractive to find processes that will economically
either reduce or eliminate the loss of condensed biomass
carbon as CO2 during the biomass to liquid fuel conversion
process.
It is possible to envision thermochemical as well as
biochemical conversion processes using supplementary
sources of energy such as heat, H2, or electricity to increase
production of high energy density liquid fuel per ton of
biomass. We have focused here on thermochemical routes.
It is worth noting that for self-contained biomass conversion
processes both biochemical as well as gasification/FTD
processes were found to provide similar yields. Therefore,
we expect that the yields calculated from thermochemical
processes using supplemental energy should provide good
target estimates for the corresponding non-thermochemical
processes as well. Although synergistic augmented biofuel
processes using supplemental energy from fossil fuel or
nuclear are described elsewhere (8, 15), in this work we are
interested in the sun-to-fuel yield to create alternative
scenarios for a future where most transport needs are met
through renewable energy. Therefore, in this analysis, we
have used solar energy as the supplemental energy source.
In order to avoid difficulties associated with the handling
of bio-oil from a fast-pyrolysis reactor prior to its HDO
treatment, it will be best to pyrolyze biomass in the presence
of H2 and a catalyst as shown in a process alternative in

FIGURE 2. H2Bioil process where solar H2 is used for fast-hydropyrolysis of biomass and subsequent HDO of the pyrolyzer product
stream.
Figure 2. Ideally one would like to choose a reactor design,
catalyst, and operating conditions to directly produce
deoxygenated high energy density liquid fuel in one step.
However, it is likely that the much higher temperature, most
likely in excess of 400 C, and much shorter contact time (1
s) that will be needed for the fast-hydropyrolysis step will
not be the optimal conditions for the level of deoxygenation
required. Therefore, if needed, the exhaust from the fasthydropyrolysis reactor, after removal of any char present
and adjustment of the temperature, can be directly sent
through an HDO reactor to allow the needed contact time
with the HDO catalyst at its preferred operating conditions.
The thrust is to avoid some of the steps associated with fastpyrolysis such as total condensation of bio-oil, the associated
problems with subsequent handling, and then revaporization
for HDO. Bridgwater and Peacock describe a large number
of fast-pyrolysis reactor configurations, and many of them
can be adapted for fast-hydropyrolysis (23). We use the term
H2Bioil to refer the processes that are based on fasthydropyrolysis/HDO of biomass with supplemental H2 and
directly produce high energy density liquid fuel.
It should be noted that a few groups have reported labscale hydropyrolysis of biomass in fixed-bed reactor mode
(27-34). In some cases, lower char and higher oil yields were
observed when H2 is present during pyrolysis (29, 30, 34).
Also, less intense degradation reactions have been reported
under H2 than under an inert atmosphere (33). These early
experiments provide favorable conceptual support for hydropyrolysis to produce oil with low oxygen content.
However, unlike fast-pyrolysis experiments, all the reported
hydropyrolysis data in the literature are from fixed-bed
reactors. Fast pyrolysis experiments are generally not conducted in a fixed-bed mode due to the need for rapid
heating-cooling and the requirement of a very short
residence time in the reactor. Thus, there is a need to conduct
hydropyrolysis experiments in a mode similar to that for
fast-pyrolysis with a possibility of immediate downstream
HDO. However, unlike fast-pyrolysis, the fast-hydropyrolysis
reactors will possibly operate at a much higher pressure.
Such experiments in the presence of H2 could provide higher
yields of deoxygenated oil per unit of biomass.
In the absence of experimental data for a fast-hydropyrolysis operation mode, it is difficult to make an accurate
estimate for the oil yield. However, it is possible to make a
reasonable estimate for yields that could be achieved. The
fixed bed hydropyrolysis studies do report decrease in char
production that could lead to higher oil yield. Elimination
of the bio-oil revaporization step prior to the HDO reactor

and the possibility of better heat integration could also


contribute to improved process efficiency and yield. In light
of these factors, in our calculations, the carbon yield in the
oil from the H2Bioil process was taken to be 70%, which is
same as reported in the literature for the bio-oil from a fastpyrolyzer (35). When bio-oil from a fast-pyrolyzer is upgraded
through HDO, the experimentally observed energy content
of the upgraded oil is found to be in the range of 38-43
MJ/kg (26). For our calculations, we have assumed that, with
enough solar H2 and advances in catalysis and processing
conditions, the energy content is likely to be closer to the
upper end value of 42 MJ/kg of oil. Furthermore, the carbon
and oxygen in the oil from the H2Bioil process were taken
to be 86.5 and 1.25 wt % with the rest being hydrogen. On
this basis, the estimated yield of oil from the H2Bioil process
is 215 ege/ton of biomass (see the SI for details). Note that
this number is slightly lower than the 230 ege/ton of biomass
reported earlier (15) as a slightly lower and more probable
value of the oil LHV is used in this work. For the estimation
of H2 consumption, we assumed that two H2 molecules are
needed to remove an oxygen atom. As shown in the detailed
calculations provided in the SI, this leads to an estimated H2
requirement of 1240 L at STP/L oil (0.11 kg H2/L oil), which
is higher than the highest experimental value reported in the
literature to upgrade bio-oil from a fast-pyrolyzer (1144 L
at STP/L upgraded oil) (25). This gives us a confidence that
our estimate of H2 consumption is conservative, and in an
actual process, the H2 consumption may be lower.
From the calculations for Figure 1, out of 6307 MJ/m2 y
of solar energy, the H2Bioil process is estimated to use 231
MJ/m2 y of energy to make H2 and the rest to grow biomass.
Using a sun to electricity conversion efficiency of 15% and
an electrolyzer efficiency (based on LHV of H2) of 50.7%, 17.6
MJ/m2 y worth of energy is estimated in solar H2. It is
estimated that from this H2Bioil process one can obtain 49.8
MJ/m2 y of oil which, of course, corresponds to the 215 ege/
ton biomass.
It is worth noting that if H2 were to be produced through
a thermochemical process rather than electrolysis, the overall
yield from solar energy to fuel will still be in the neighborhood
of 49.8 MJ/m2 y. A recent study reports 30% high heating
value (HHV) based efficiency to H2 conversion from the net
solar energy soaked up by the thermochemical reactor (14).
When one accounts for radiation losses etc., the solar energy
to H2 efficiency turns out to be 16% based on LHV of H2
(see the SI). This implies that only one-half of the 231 MJ/
m2 y of solar energy shown in Figure 1 will be needed to
VOL. 44, NO. 13, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

5301

FIGURE 3. H2Bioil-B process where a portion of the feed biomass (32-42%) is fed to the gasification zone to provide H2 for
fast-hydropyrolysis and HDO of the remaining biomass fed to the hydropyrolysis zone.
supply H2, but this decrease will have a small incremental
impact on the overall yield of 49.8 MJ/m2 y as oil.
Results are also available for a gasification/FT based hybrid
hydrogen-carbon (H2CAR) process where almost all the
biomass carbon is recovered as liquid fuel by using supplementary H2 (8). Figure 1 shows that the estimated recovery
of solar energy as liquid fuel from the H2CAR process is indeed
the highest of any process. However, this increase in oil
production comes at a steep increase in demand for solar H2
at about 0.33 kg H2/L oil. In fact, the inefficiency of the H2CAR
process is reflected in the expected production of only 66.9
MJ/m2 y of liquid fuel from the input of 72 MJ/m2 y of solar
H2 as seen from Figure 1. If energy efficiency were the only
criteria, one would be better off using H2 via fuel cell vehicles
which are more efficient than the oil based internal combustion engines.
It is of interest to compare the performance of H2Bioil,
which is an augmented process, with a self-contained process.
However, rather than comparing it with a gasification/FT or
a biochemical process, it will be more informative to compare
it with a self-contained version that is also based on fasthydropyrolysis.
Liquid Fuel Yields from a Self-Contained Fast-Hydropyrolysis Based Process. The self-contained version of the
H2Bioil process can be created by gasifying a portion of the
biomass to supply the needed H2 for the fast-hydropyrolysis/
HDO of the remaining biomass portion. In order to consolidate biomass gasification/fast-hydropyrolysis in one
reactor unit, we propose the novel fast-hydropyrolysis reactor
configuration shown in Figure 3 (H2Bioil-B). In this process,
depending on the efficiency of gasification section, 32-42%
of the total biomass is gasified to produce an H2/CO mixture
which is sufficient to hydropyrolyze and hydrodeoxygenate
the remaining fraction of the biomass that is directly fed to
the hydropyrolysis zone. The hot gas from the gasifier is
directly injected in the pyrolyzer zone. If needed, the
temperature of the exhaust gas prior to its injection in the
pyrolyzer zone may be adjusted. Also, if required, a hot or
a cold recycle stream may be injected between the gasifier
and the pyrolyzer zone to provide better temperature control
in the pyrolyzer section of the reactor.
The majority of the CO from the gasification zone is
expected to provide H2 through the water-gas-shift (WGS)
reaction. Generally, the WGS reaction can be conducted at
high temperatures of 350-500 C using an iron oxide based
catalyst (36). In this temperature range, formation of H2 is
preferred and most of the CO can be converted. Of course,
the use of in situ water-gas-shift catalysts in the fast5302

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 13, 2010

hydropyrolysis and the HDO reactors can provide an effective


means to control the H2 partial pressure during biomass
conversion. As H2 is consumed, there will be driving force
for CO reaction with H2O that is formed as a byproduct of
the HDO reaction. Furthermore, an interesting but as yet
unanswered question is whether CO can play a direct role
in pyrolysis or deoxygenation.
It should be noted that for the H2Bioil-B process, in situ
biomass gasification will provide hot syngas for the biomass
fast-hydropyrolysis process, leading to an increase in energy
efficiency. An important advantage of this process over
conventional HDO is that it will avoid the need for pure
hydrogen and thus save the energy and capital cost associated
with hydrogen purification and recovery. Due to expected
improvements in energy efficiencies due to process integration, as compared to a process using three separate steps of
fast-pyrolysis, gasification and HDO, we expect higher liquid
fuel yield from the H2Bioil-B process (26).
In calculations for Figure 1, we find that for a biomass to
H2 conversion efficiency of 50% (1), nearly 42% of the biomass
will be gasified to provide the needed hydrogen to yield 30.1
MJ/m2 y of solar energy as liquid fuel. This translates into
a carbon efficiency, energy efficiency, and liquid fuel yield
of 40.6%, 59.1%, and 125.4 ege/ton of biomass. For a higher
biomass gasifier efficiency of 75% to hydrogen, the carbon
efficiency, energy efficiency, sun-to-fuel recovery, and liquid
fuel yield estimates are 47.2%, 68.6%, 35 MJ/m2 y, and 146
ege/ton of biomass, respectively (refer to the SI for calculation
details). The calculated yield of 125-146 ege/ton of biomass
is highest among the hitherto known self-contained biomass
to high energy density liquid fuel processes (roughly 50%
higher liquid fuel yield than the self-contained gasification/FT
or current biochemical processes). However, the use of the
organic residue from a biochemical process to produce
additional liquid fuel via gasification/FT process has been
suggested in the literature (20, 37) and has the potential to
produce a competitive 77 gallons of gasoline and 11 gallons
of diesel equivalent oil per ton of biomass (equivalent to
136 ege/ton of biomass) (20). The process by ZeaChem
first produces acetic acid by fermentation of sugars (38). The
acetic acid is then converted to an ester which is reacted
with H2 to form ethanol. The H2 is produced via gasification
of the lignin residue. It is claimed that the nth plant will yield
135 ege/ton of biomass. The proposed H2Bioil-B uses only
a thermochemical route and has the potential to be built on
small scale while providing high yields of a high energy density
liquid fuel product.

FIGURE 4. Sun-to-fuel yield for various conversion processes at biomass collection rates of (a) 1.5, (b) 3, (c) 5, and (d) 6.25 kg/m2 y.
The values shown in the bars are upper estimates.
Since we have S2F recovery estimates for various selfcontained as well as augmented processes, they provide us
an opportunity to make some interesting observations:
Observations from S2F Recovery Estimates. From the
various S2F recovery estimates shown in Figure 1 for a
biomass collection rate of 3 kg/m2 y, we note that
(1) While S2F recoveries shown for the self-contained fastpyrolysis and H2Bioil-B are similar, the energy content
on a per unit mass basis of the bio-oil from fastpyrolysis is less than half of the oil from H2Bioil-B and
it cannot be used without spending additional energy
to upgrade it.
(2) For the self-contained processes, the energy recovered
as high energy density oil is in the range of 18 to 35
MJ/m2 y. When compared to the energy content of 51
MJ/m2 y for original biomass, the energy efficiency
for conversion of biomass varies over a wide range of
35-68%. The efficiency range of 35-50% for the
gasification/FT and biochemical based processes is
similar. It seems that fast-hydropyrolysis/HDO based
processes have the potential to be among the most
efficient of the self-contained processes. This provides
an incentive for their experimental demonstration.
(3) On the basis of the incident solar energy of 6307 MJ/
m2 y, the recovery as liquid fuel for the self-contained
processes is less than 0.56%. This is not surprising as
the efficiency of collecting dedicated fuel crop was
0.81%. Nevertheless, it does point out that the solar
energy recovered as liquid fuel via the dedicated fuel
crop and self-contained processing is quite low.
(4) The augmented processes hold the promise to substantially improve S2F efficiencies (0.8-1.06%). As a
matter of fact, the energy content in the fuel can be
similar (49.8 vs 51 MJ/m2 y) or even higher (66.9 vs 51
MJ/m2 y) than that in the biomass. On the basis of the
S2F yields, the liquid fuel produced by the augmented

processes using the same total land area can be oneand-one-half to three times that of the self-contained
processes.
(5) It is interesting to compare incremental oil productions
and H2 consumption for the two routes: (i) the H2CAR
process as compared to the self-contained gasification/FT process and (ii) the augmented H2Bioil process
with the self-contained H2Bioil-B process. We find that
the energy content of the incremental oil produced
per unit of energy in the H2 used (MJ in incremental
oil produced/MJ in H2 used) is much more for the
H2Bioil based route (0.9-1.2) than for the H2CAR route
(0.65). This shows that the fast-hydropyrolysis/HDO
based process is very efficient in increasing the biomass
carbon recovery. However, its carbon recovery is
limited to 70%. In order to get nearly 100% carbon
recovery, the H2CAR process essentially converts CO2
to liquid fuel. Conversion of CO2 to liquid fuel is an
energy intensive process and is reflected in the high
H2 requirements for the H2CAR process. This also
demonstrates the energy penalty associated with very
high recoveries of biomass carbon.
(6) We can draw another important conclusion from
Figure 1. When additional liquid fuel is needed, it
should preferably be first produced by using augmented processes in conjunction with SAW biomass
rather than through growing dedicated fuel crops. Such
a step would make more efficient use of sunlight falling
on a given land area. This is possible because heat or
H2 are recovered from sunlight at a much higher
efficiency than a dedicated fuel crop, and at least 65%
of the energy in H2 can be recovered as the incremental
liquid from the SAW biomass.
In Figure 4, we show estimated S2F yields for four different
biomass collection rates. While the expected range for the
yields are calculated in the SI for a number of conversion
processes, we only show upper estimates in the bar graphs.
VOL. 44, NO. 13, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

5303

As expected, increased growth rates and hence collection


rates for biomass have a profound impact on S2F yields. For
all processes except H2CAR, the increase in liquid fuel rate
is directly proportional to the increase in biomass collection
rate. For the H2CAR process, large quantities of H2 are needed
and high biomass growth rates lead to relatively larger fraction
of the land area being needed for the H2 production. This
results in less than proportional increase in liquid fuel yield
with the increase in biomass growth rates. It is informative
to note that at even the highest collection rate of 6.25 kg/
m2 y, the S2F yield from the H2CAR process is 119.9 MJ/
m2 y. This corresponds to 1.9% of the annual incident solar
light being collected as liquid fuel. This tells us that, even
under a highly favorable process condition, the dedicated
biomass carbon route to liquid fuel provides only a small
portion of the incident solar energy as liquid fuel. This has
a strong implication on the land area requirements when
dedicated biomass is to be grown for liquid fuel production.

Discussion
In this work, we have estimated the sun-to-fuel yield for
liquid hydrocarbon fuel via the biomass carbon route. This
is relevant when dedicated biomass is to be grown and
harvested for the production of liquid fuel. In such a case,
one of the goals is to increase the collection of solar energy
incident on a given land area as liquid fuel. The S2F yield for
the self-contained processes that mainly rely on the biomass
to supply all the energy need is quite low. Even at a high
biomass collection rate of 6.25 kg/m2 y, only 1.16% of the
solar energy is estimated to be recovered as liquid fuel. Two
main contributing factors to the low S2F yield are (i) the
relatively low recovery of solar energy as biomass energy
and (ii) the fact that less than 50% of the biomass carbon
recovery in the liquid fuel is from self-contained processes.
We find that the S2F yield can potentially be increased by
a factor of 1.5-3 when all the available land area is not used
to grow dedicated fuel crop but the solar energy falling on
a portion of the land area is harnessed as hydrogen which
is then used in novel augmented biomass conversion
processes to increase biomass carbon yield as liquid fuel.
Such augmented processes are estimated to have higher S2F
yield because hydrogen is harvested from solar energy with
a much higher efficiency than biomass.
In this paper, we have examined a number of alternative
augmented processes that provide different levels of biomass
carbon recovery by using a wide range of supplemental H2
mass per unit of liquid fuel produced. A process such as
H2CAR, based on gasification/FT chemistry, can recover
nearly 100% biomass carbon but will need approximately
0.33 kg H2/L oil produced. On the other hand, fasthydropyrolysis/HDO based H2Bioil has a potential to recover
70% biomass carbon with 0.11 kg H2/L oil.
A side benefit of the development of the augmented
H2Bioil process is that it has led to an efficient self-contained
H2Bioil-B process concept. We have proposed a process
whereby a portion of the biomass is gasified to produce H2/
CO containing hot gas which in turn is used for the fasthydropyrolysis/HDO of the remaining biomass. This
H2Bioil-B process, after successful experimental demonstration, could result in a high energy density liquid fuel yield
that is greater than other known self-contained processes.
The model calculations indicate possible yields of 125-146
ege/ton biomass. This yield range can be taken to represent
a good achievement target for any efficient self-contained
conversion process.
The low S2F yields for the dedicated fuel crops point us
to the urgency of developing processes that are efficient and
increase the liquid fuel yield from an available quantity of
the SAW biomass. This stresses the need to decrease the
5304

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 13, 2010

release of SAW biomass carbon as CO2 during the conversion


process. Use of supplemental energy such as H2, heat, or
electricity, which are recovered at much higher efficiencies
from solar energy, to increase liquid fuel yield from SAW
biomass will decrease the land area needed to grow dedicated
biomass for the incremental liquid fuel. The development of
efficient augmented processes will not only help in increasing
S2F yields for the dedicated fuel crop cases but will also be
valuable in increasing liquid fuel from a given quantity of
SAW biomass.

Acknowledgments
We thank the US Department of Energy (Grant No. DE-FG
3608GO18087) and AFOSR (Grant No. FA 9550-08-1-0456)
for partial support of this work. We also thank Eric Smoldt
for his valuable help with the figures.

Abbreviations
ege
FT
FTD
H2Bioil
H2Bioil-B
H2CAR
HDO
WGS

ethanol gallon equivalent


Fischer-Tropsch
Fischer-Tropsch diesel
hydrogen bio-oil process
H2Bioil process where needed H2 is produced from
a portion of biomass
hybrid hydrogen-carbon process
hydrodeoxygenation
water-gas shift

Supporting Information Available


Assumptions, definitions, and basis for calculations along
with estimation and modeling details for both the selfcontained and augmented processes. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited
(1) NRC. The NRC report-The Hydrogen Economy-Opportunites,
Costs, Barriers, and R&D Needs; The National Academies Press:
Washington D C, 2004.
(2) Tarascon, J. M.; Armand, M. Issues and challenges facing
rechargeable lithium batteries. Nature 2001, 414 (6861), 359
367.
(3) Kintner-Meyer, M.; Schneider, K.; Pratt, R. Impact Assessment
of Plug-in Hybrid Vehicles on Electric Utilities and Regional U.S.
Power Grids, Part 1: Technical Analysis; Pacific National
Laboratory: Richland, WA, 2007; http://www.ferc.gov/about/
com-mem/wellinghoff/5-24-07-technical-analy-wellinghoff.
pdf.
(4) Davis, S. C.; Diegel, S. W. Transportation Energy Data Book,
edition 28 ORNL-6978; Oak Ridge National Laboratory: Oak
Ridge, TN, 2009; http://cta.ornl.gov/data/index.shtml.
(5) BP. Statistical Review of World Energy; London, June 2008.
(6) NRC. Americas Energy Future Panel on Alternate Liquid
Transportation Fuels-Liquid Transportation Fuels from Coal and
Biomass: Technological Status, Costs, and Environmental Impacts; National Academies Press: Washington, D.C., 2009; p 300.
(7) Perlack, R. D.; et al. Biomass as Feedstock for a Bioenergy and
Bioproducts Industry: The Technical Feasibility of a Billion-Ton
Annual Supply; DOE/GO-102995-2135 ORNL/TM-2005/66,
2005; http://feedstockreview.ornl.gov/pdf/billion_ton_vision.
pdf.
(8) Agrawal, R.; et al. Sustainable fuel for the transportation sector.
Proc. Natl. Acad. Sci. U.S.A. 2007, 104 (12), 48284833.
(9) Dismukes, G. C.; et al. Aquatic phototrophs: efficient alternatives
to land-based crops for biofuels. Curr. Opin. Biotechnol. 2008,
19 (3), 235240.
(10) King, R. R. Multijunction cells: Record breakers. Nat. Photon
2008, 2 (5), 284286.
(11) King, R. R.; et al. 40% efficient metamorphic GaInP/GaInAs/Ge
multijunction solar cells. Appl. Phys. Lett. 2007, 90 (18), 183516.
(12) Sargent and Lundy LLC Consulting Group Chicago, I. Assessment
of Parabolic Trough and Power Tower Solar Technology Cost
and Performance Forecasts; NREL/SR-550-34440, 2003; www.
osti.gov/servlets/purl/15005520-kLbVbt/native/.
(13) 1996 Status Report on Solar Trough Power Plants - Experience,
Prospects and Recommendations to Overcome Market Barriers
of Parabolic Trough Collector Power Plant technology; Pilkington

(14)
(15)
(16)

(17)

(18)
(19)
(20)

(21)

(22)

(23)

(24)

(25)
(26)

Solar International GMBH, Cologne, Germany, 1996; ISBN


3-9804901-0-6.
Diver, R. B.; et al. Solar Thermochemical Water-Splitting FerriteCycle Heat Engines. J. Solar Energy Eng. 2008, 130 (4), 041001.
Agrawal, R.; Singh, N. R. Synergistic Routes to Liquid Fuel for
a Petroleum deprived future. AIChE J. 2009, 55 (7), 18981905.
Knaupp, W. Power rating of photovoltaic modules from outdoor
measurements. Photovoltaic Specialists Conference, 1991, Conference Record of the Twenty Second IEEE, 1991; Vol. 1, pp 620624.
King, D. L. Photovoltaic module and array performance characterization methods for all system operating conditions; Proceedings of NREL/SNL Photovoltaics Program Review Meeting,
Nov 18-22, 1999, Lakewood, CO; AIP Press: New York, 1997;
SAND-96-2848C, 1996.
Schubert, C. Can biofuels finally take center stage. Nat.
Biotechnol. 2006, 24 (7), 777784.
Tijmensen, M. J. A.; et al. Exploration of the possibilities for
production of Fischer-Tropsch liquids and power via biomass
gasification. Biomass Bioenergy 2002, 23, 129152.
Greene, N.; et al. Growing Energy: How Biofuels Can Help End
Americas Oil Dependence; Natural Resources Defense Council:
New York, Dec 2004.
Mark Wright, M.; Brown, R. C. Comparative economics of
biorefineries based on the biochemical and thermochemical
platforms. Biofuels, Bioprod. Biorefining 2007, 1 (1), 4956.
Mohan, D.; Pittman, C. U.; Steele; P. H. Pyrolysis of Wood/
Biomass for Bio-oil: A Critical Review. Energy Fuels 2006, 20 (3),
848889.
Bridgwater, A. V.; Peacock, G. V. C. Fast pyrolysis processes for
biomass. Renewable Sustainable Energy Rev. 2000, 4 (1), 173.
Huber, G. W.; Iborra, S.; Corma, A. Synthesis of Transportation
Fuels from Biomass: Chemistry, Catalysts, and Engineering.
Chem. Rev. 2006, 106 (9), 40444098.
Elliott, D. C. Historical Developments in Hydroprocessing Biooils. Energy Fuels 2007, 21 (3), 17921815.
Jones, S. B.; et al. Production of Gasoline and Diesel from Biomass
via Fast Pyrolysis, Hydrotreating and Hydrocracking: A Design
Case; Pacific National Laboratory: Richland, WA, PNNL-18284
Rev. 1, 2009.

(27) Gercel, H. F.; Putun, A. E.; Putun; E. Hydropyrolysis of Extracted


Euphorbia rigida in a Well-Swept Fixed-Bed Tubular Reactor.
Energy Sources 2002, 24, 423430.
(28) Pu
tu
n, A. E.; et al. Fixed-bed pyrolysis and hydropyrolysis of
sunflower bagasse: Product yields and compositions. Fuel
Process. Technol. 1996, 46 (1), 4962.
(29) Pu
tu
n, A. E.; et al. Oil production from an arid-land plant: fixedbed pyrolysis and hydropyrolysis of Euphorbia rigida. Fuel 1996,
75 (11), 13071312.
(30) Dilcio Rocha, J.; Luengo, C. A.; Snape, C. E. The scope for
generating bio-oils with relatively low oxygen contents via
hydropyrolysis - a versatile technique for solid fuel liquefaction,
sulphur speciation and biomarker release. Org. Geochem. 1999,
30, 15271534.
(31) Rocha, J. D.; et al. Hydropyrolysis: a versatile technique for solid
fuel liquefaction, sulphur speciation and biomarker release. J.
Anal. Appl. Pyrolysis 1997, 40-41, 91103.
(32) Pindoria, R. V.; et al. A two-stage fixed-bed reactor for direct
hydrotreatment of volatiles from the hydropyrolysis of biomass:
effect of catalyst temperature, pressure and catalyst ageing time
on product characteristics. Fuel 1998, 77 (15), 17151726.
(33) Pindoria, R. V.; et al. Structural characterization of biomass pyrolysis
tars/oils from eucalyptus wood waste: effect of H2 pressure and
sample configuration. Fuel 1997, 76 (11), 10131023.
(34) Pindoria, R. V.; et al. Hydropyrolysis of sugar cane bagasse:
effect of sample configuration on bio-oil yields and structures
from two bench-scale reactors. Fuel 1999, 78, 5563.
(35) Ringer, M.; Putsche, V., Scahill, J. Large Scale Pyrolysis Oil
Production: A Technology Assessment and Economic Analysis;
NREL/TP-510-37779, National Renewable Energy Laboratory:
Golden, CO, 2006; http://www.nrel.gov/docs/fy07osti/37779.pdf.
(36) Farrauto, R. J.; Bartholomew, C. H. Fundamentals to Industrial
Catalytic Processes; Chapman and Hall: London, 1997.
(37) Lynd, L. R. Energy Myth Three - High L and Requirements and
an Unfavorable Energy Balance Preclude Biomass Ethanol from
Playing a Large Role in Providing Energy Services, in Energy and
American Society- Thirteen Myths; Sovacool, B. K., Brown, M. A.,
Eds.; Springer: New York, 2007; p 75-102.
(38) Zeachem. Zeachem technology overview. www.Zeachem.com/
technology/overview.php(accessed January 2010).

ES100316Z

VOL. 44, NO. 13, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

5305

You might also like