Fuel 106 (2013) 463469

Contents lists available at SciVerse ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Techno-economic analysis of biomass to transportation fuels and electricity

via fast pyrolysis and hydroprocessing
Tristan R. Brown a,, Rajeeva Thilakaratne b, Robert C. Brown a,b, Guiping Hu c
a

Bioeconomy Institute, Iowa State University, Ames, IA 50011, United States

Department of Mechanical Engineering, Iowa State University, Ames, IA 50011, United States
c
Industrial and Manufacturing Systems Engineering, Iowa State University, Ames, IA 50011, United States
b

h i g h l i g h t s
" We model an updated 2000 metric ton per day fast pyrolysis and hydroprocessing system.
" We incorporate recent commercialization and literature data on the pathway.
" We present the results of a techno-economic analysis of this system.
" We compare these results with those published in a 2010 Iowa State University study.

a r t i c l e

i n f o

Article history:
Received 13 April 2012
Received in revised form 6 November 2012
Accepted 8 November 2012
Available online 1 December 2012
Keywords:
Fast pyrolysis
Hydroprocessing
Catalytic pyrolysis
Techno-economic analysis

a b s t r a c t
A previous Iowa State University (ISU) analysis published in 2010 investigated the technical and economic feasibility of the fast pyrolysis and hydroprocessing of biomass, and concluded that the pathway
could produce cellulosic biofuels for a minimum fuel selling price (MFSP) of $2.11/gal. The 2010 ISU study
was largely theoretical in that no commercial-scale fast pyrolysis facilities were being constructed at the
time of publication.
The present analysis expands upon the 2010 ISU study by performing an updated techno-economic
analysis of the fast pyrolysis and hydroprocessing pathway. Recent advances in pathway technology
and commercialization and new parameters suggested by the recent literature are accounted for. The
MFSP for a 2000 MTPD facility employing fast pyrolysis and hydroprocessing to convert corn stover to
gasoline and diesel fuel is calculated to quantify the economic feasibility of the pathway.
The present analysis determines the MFSP of gasoline and diesel fuel produced via fast pyrolysis and
hydroprocessing to be $2.57/gal. This result indicates that the pathway could be competitive with petroleum, although not as competitive as suggested by the 2010 ISU study. The present analysis also demonstrates the sensitivity of the result to process assumptions.
2012 Elsevier Ltd. All rights reserved.

1. Introduction
The Renewable Fuel Standard (RFS) was created in 2005 by Congress to mandate the domestic consumption of ethanol, particularly grain ethanol [1]. Concerns that the use of grains to produce
ethanol was causing both rainforest destruction [2] and hunger
in the developing world [3] led Congress to direct the Environmental Protection Agency (EPA) to revise the RFS in a manner that
placed greater emphasis on cellulosic biofuels [4]. While grain ethanol production has had little difculty meeting its share of the revised mandate (RFS2), cellulosic biofuels have fallen far short. The
RFS2 initially mandated the production of 250 million gallons per
year (MGY) of cellulosic biofuels in 2011, although no commer Corresponding author.
E-mail address: trb6c4@iastate.edu (T.R. Brown).
0016-2361/$ - see front matter 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.fuel.2012.11.029

cial-scale production occurred [5]. The RFS2 mandates the production of 16,000 MGY of cellulosic biofuels by 2022 [4].
In 2010 Iowa State University (ISU) researchers assessed the
technical and economic feasibility of three different cellulosic biofuel pathways (cellulosic ethanol; gasication and FischerTropsch
synthesis; and fast pyrolysis and hydroprocessing) and identied
fast pyrolysis (FP) and hydroprocessing as being the most economically feasible of the three [68]. Specically, the researchers calculated the minimum fuel selling price (MFSP) of gasoline and diesel
fuel produced via the pathway to be as low as $2.11/gal. FP and
hydroprocessing was the only one of the three pathways to attain
a MFSP for an nth plant below $4/gal gasoline-equivalent (gge) [9].
FP is a thermochemical pathway that uses heat to rapidly
decompose lignocellulosic biomass into solid (char), gas (noncondensable gas, or NCG) and liquid (bio-oil). Char has potential
value as both a cropland soil amendment [10] and carbon sequestration

464

T.R. Brown et al. / Fuel 106 (2013) 463469

method [11] while NCG provides process heat to the FP facility.

Bio-oil can serve as a feedstock for electricity generation [12] or
production of ethanol [13], renewable gasoline and diesel fuel
[14], and commodity chemicals [15]. Techno-economic analyses
suggest that serving as a feedstock for the production of renewable
hydrocarbon fuels and chemicals is the most economically-feasible
use of bio-oil despite the substantially higher capital and operating
costs involved, although this is sensitive to input costs and output
values [8,16,17].
Raw bio-oil is corrosive, viscous, and highly oxygenated, characteristics that make it difcult to store, transport and rene without
prior upgrading [18]. Upgrading can be accomplished via either
uid catalytic cracking (FCC) [19] or hydroprocessing [20]. For
the purposes of the RFS2, hydroprocessing is more applicable because FCC yields alkenes and aromatics [15] while hydroprocessing
yields alkanes and aromatics [20] more suitable for fuel blending.
Hydroprocessing employs two steps: hydrotreating and hydrocracking. Hydrotreating reacts bio-oil with hydrogen (1.2
3.5 wt% on a bio-oil basis) in the presence of a catalyst and heat
[21], removing heteronuclear atoms such as chlorine, nitrogen,
oxygen and sulfur and reducing bio-oils viscosity and corrosiveness [20]. Hydrocracking reacts the hydrotreated bio-oil with
hydrogen in the presence of a catalyst under more severe reaction
conditions, with the objective of achieving complete deoxygenation and inducing the depolymerization of the oligomeric species
within the bio-oil into monomeric hydrocarbons that can be split
and blended into gasoline and diesel fuels.
Transportation fuel yields from hydroprocessing of bio-oil are
sensitive to a number of factors such as operating conditions and
hydrogen consumption [20]. For this reason the yields reported
in the literature vary. The rst factor affecting fuel yields is the
bio-oil yield from the FP reactor. Yields of 5570 wt% on a biomass
basis are typical [22], although yields above or below this range are
not uncommon, depending on the kind of feedstock and reactor
employed. The second factor affecting fuel yields is the volume of
monomeric hydrocarbons yielded by hydrotreating and hydrocracking. This varies according to pyrolysis feedstock type, reactor
operating conditions, and bio-oil fraction. Elliott et al. [20] report a
range of 31.660.7 wt% (bio-oil basis) for ve different bio-oils under identical hydroprocessor operating conditions.
Catalytic FP (CFP) combines FP and FCC upgrading processes
into an integrated pathway. Pyrolysis occurs in the presence of a
FCC or pure zeolite catalyst at moderate temperatures and high
heating rates [23]. Bio-oil produced via CFP has a lower oxygen
and higher aromatics content than that produced via FP as well
as lower liquid and higher coke yields [24]. Bio-oil produced via
CFP still contains oxygen requiring further upgrading via hydroprocessing to maximize transportation fuel yields. The advantage of
CFP over FP as a step in the production of transportation fuels is
that the resulting bio-oil is reported to have a lower oxygen content following hydroprocessing (5 wt%) than FP bio-oil following
the same treatment (12.4 wt%) [25]. This suggests that the hydroprocessing of bio-oil produced via CFP will consume less hydrogen
than bio-oil produced via FP.
CFP is being commercialized by KiOR, which completed an initial
public offering in 2011 [26]. The company is currently constructing
a 454 metric ton per day (MTPD) CFP and hydroprocessing facility
in Columbus, Mississippi and plans to construct four 1361 MTPD
facilities in Georgia, Mississippi, and Texas in the coming years
[26]. The Columbus facility represents the largest pyrolysis project
to date in the US, making it an important bellwether for the future
economic feasibility of the CFP pathway. KiOR published detailed
information on the expected capital and operating costs for its
CFP facilities as part of its initial public offering that is particularly
useful for estimating the commercial prospects for catalytic and
fast pyrolysis, and identifying facility specications.

The objective of this paper is to quantify the economic feasibility of gasoline and diesel fuel production from stover via FP and
hydroprocessing using updated information that has become available as a result of the successful construction of the KiOR Columbus facility. KiOR data on equipment requirements and facility
capital costs that are similar for both FP and CFP are incorporated
into this analysis. A 2000 MTPD FP and hydroprocessing facility is
modeled, and the total project investment (TPI) and operating
costs are estimated to determine a MFSP for the gasoline and diesel
fuel under a 10% internal rate of return (IRR). The results of this
analysis are compared to the results of the 2010 ISU study, and a
sensitivity analysis is employed to identify the impact of the updated assumptions on facility MFSP.

2. Materials and methods

The following steps are employed by the FP system to convert
biomass feedstock to bio-oil: pre-processing, pyrolysis reaction,
solids removal, bio-oil recovery, heat generation, hydrotreating,
hydrocracking, and rening (see Fig. 1) [8]. During the pre-processing
step the stover feedstock is dried to 7% moisture content,
chopped, and ground to 3 mm particles. The feedstock is then sent
to the uidized bed reactor where it is rapidly heated at atmospheric pressure to 480 C to obtain yields of 63 wt% bio-oil,
17 wt% char, and 20 wt% NCG (see Table 1). Cyclones with an assumed efciency of 90% separate most of the char and ash from
the pyrolysis vapors, which are then cooled, condensed and sent
to an electrostatic precipitator (ESP) for separation of aerosols from
the NCG. A small fraction of ue gas is recycled as carrier gas for
the pyrolysis reactor. The char and NCG are combusted in a waste
heat boiler at 450 C to generate high pressure (50 MPa) steam,
which enters a staged turbine system to provide process heat
and facility electricity. Excess electricity is assumed to be sold into
the grid for $0.054/kWh. Tables 2 and 3 provide the properties of
the stover feedstock and compositions of the product bio-oil and
NCG.
Hydroprocessing employs a mild hydrotreating stage followed
by a more severe hydrocracking stage. Both stages use xed-bed
jacketed reactors. Hydrotreating occurs in a hydrogen-rich environment (4 wt% hydrogen), 710 MPa pressure, and temperatures
of 300400 C in the presence of a cobalt-molybdenum catalyst.
A hydrogen compressor and pressure swing adsorption unit
(PSA) are employed to recycle any excess hydrogen. This scenario
assumes that the hydrogen is purchased from an external source
for $1.33/kg, the same value used by the 2010 ISU study [8].1
Hydrotreating removes impurities such as sulfur and nitrogen from
the bio-oil while also partially deoxygenating it. Off-gas produced
during the hydrotreating stage is combusted to provide process heat
and to fuel the boiler and turbogenerator system for co-generation of
steam and electricity. The hydrocracking stage occurs at higher pressures (1014 MPa) and temperatures (400500 C), also in the presence of a cobalt-molybdenum catalyst. Hydrocracking completes the
deoxygenation and depolymerization of the heavy molecules found
in bio-oil to produce lighter molecules within the diesel fuel and gasoline ranges (C8C12). These molecules are then split according to
respective ranges for blending into gasoline and diesel fuel. While
similar to the system modeled in the 2010 ISU study [8], the fuel
yield is reduced to 57.4 MGY to reect the hydroprocessing results
presented in Elliott et al. [20]. The fuel yield is assumed to be split
evenly between gasoline and diesel fuel [28].

1
While the 2010 ISU study reports the use of $1.50/kg for the hydrogen price, the
tables in Appendix D of the companion NREL study indicate that a value of $1.33/kg
was actually used in the analysis [27].

465

T.R. Brown et al. / Fuel 106 (2013) 463469

Purchased Hydrogen
Gasoline

Hydroprocessing
Upgraded
Bio Oil

Condensor
Storage
Tanks

Refining

Diesel
Plant
Cooling
Cyclone

Off Gas

Compressor

Cooling
Tower
Recycled
Water

Moisture

Recycled
Hydrogen

Pyrolyzer

Pressure Swing
Adsorption

Cooling
Water

Biomass
Char
Chopper

Dryer

Flue Gas

Grinder
Fluidizing
Gas

Off Gas

Non-condensable
Gases

Pyrolysis
Heat

Waste heat
boiler
Steam

Excess
Heat

Flue Gas
Combustor

Turbine
Electricity

Air

Plant Excess Heat

Air

Feed
Water

Plant Steam

Fig. 1. Schematic of FP and hydroprocessing system.

Table 1
Product yields from FP of corn stover. Source:
2010 ISU study [8].

Table 3
Composition of bio-oil and NCG. Source: 2010 ISU study [8].
Composition (g/100 g dry biomass)

Product yields

FP (wt%)

Bio-oil
Char
NCG

63
17
20

Table 2
Properties of corn stover feedstock. Source: 2010 ISU study [8].
Element
Ultimate analysis (dry basis)

Biomass value (wt%)

Ash
Carbon
Hydrogen
Nitrogen
Sulfur
Oxygen

6.0
47.3
5.1
0.8
0.2
40.6

Proximate analysis (dry basis)

Element value (wt%)

Moisture
Fixed content
Volatile matter
Ash

25.0
17.7
52.8
4.5

Equipment costs are estimated using Aspen Process Economic

Analyzer software. Total project investment (TPI) for the facility
is estimated using the Peters and Timmerhaus factors (see Table 4)
[29]. A 20-year average IRR is calculated using a modied version
of the discounted cash ow rate of return (DCFROR) spreadsheet
employed by the 2010 ISU study [8]. The nancial assumptions
in the 2010 ISU study have been updated in the present analysis
to represent a more realistic nancing scenario in which both debt
and equity nancing are used (see Table 5). Facility capital is no
longer assumed to be derived entirely from equity but is now split

Non-condensable gases
Carbon dioxide
Carbon monoxide
Ethane
Hydrogen
Propane
Bio-oil compoundsa
Acetic acid
Benzene
Ethylphenol
Formic acid
Furfural
Methoxyphenol
Phenol
Propionic acid
Propyl-benzoate
Toluene

5.4
6.6
0.1
0.6
0.2
5.9
0.8
3.8
3.4
19.0
0.6
0.5
7.3
16.4
2.3

Other compounds
Char/Ash
Water

16.4
10.8

a
Bio-oil compound values in the 2010 ISU study were adjusted to achieve mole
balance.

evenly between equity and debt. This incurs tax-deductible interest payments over the 10-year life of the debt. As a new technological pathway it is unlikely that investors would consider a debt
offering by a FP and hydroprocessing company to be investmentgrade, necessitating a higher interest rate on the debt to account
for this additional risk. This analysis therefore assumes that the
cost of debt is the same as the High Yield Constrained bond index
yield of 7.5% [30]. This factor is sensitive to the assumption that the
facility is located within the United States; an identical facility
located in Europe would pay an interest rate closer to 10% due to

466

T.R. Brown et al. / Fuel 106 (2013) 463469

Table 4
Methodology for nth plant capital cost estimation. Source: Peters and Timmerhaus
[29].
Parameter

Assumption

Total purchase equipment cost (TPEC)

Purchased equipment installation
Instrumentation and controls
Piping
Electrical systems
Buildings (including services)
Yard improvements
Service facilities
Total installed cost (TIC)
Indirect cost
Engineering
Construction
Legal and contractors fees
Total direct and indirect costs (TDIC)
Contingency
Fixed capital investment (FCI)
Working capital (WC)
Land use
Total project investment (TPI)

100%
39%
26%
10%
31%
29%
12%
55%
TPEC  installation factor (3.02)
0.89  TPEC
32%
34%
23%
TIC + IC
20% of TDIC
TDIC + contingency
15% of FCI
6% of TPEC
FCI + WC + Land

Table 5
Comparison of TEA assumptions. Source: Wright et al. [8] and WSJ [30].
Assumptions

2010 ISU study

Present analysis

Facility size (MTPD)

Facility fuel output (MGY)
Equity%
Bond yield%
TPI (million $)
Hydrogen source
Hydrogen cost ($/kg)
Electricity price ($/kWh)
IRR%
Feedstock cost ($/MT)

2000
58.2
100
N/A
200
Merchant
1.33
0.054
10
83

2000
57.4
50
7.5%
429
Merchant
1.33
0.054
10
83

that continents current scal troubles [30]. This assumption is also

more pessimistic than the KiOR scenario, as that company has received a large interest-free loan from the Mississippi Development
Authority [26].
While the process model in the present analysis is based on the
model presented in the 2010 ISU study [8], several important differences should be noted. First, the FP and hydroprocessing model
was designed using Chemcad software rather than Aspen Plus.
Aspen Energy Analyzer was used to model an expanded boiler
and turbogenerator system and a heat exchanger network, both
of which are new (the 2010 ISU study model used a smaller boiler
for the combustion of only a fraction of char). The facility modeled
here combusts all of the char and NCG for the production of process heat, steam, and facility electricity, with excess electricity
being sold to the grid. A hydrogen compressor and pressure swing
absorption unit are introduced for the purpose of recycling excess
hydrogen. Finally, a rening unit containing a debutanizer and fuel
splitters has been added to model the actual production of gasoline
and diesel fuel.
3. Assumptions in the 2010 ISU study
Table 5 compares the major assumptions used in the 2010 ISU
study and the fast pyrolysis (FP) and hydroprocessing system presented in this study. Assumptions for both studies are presented on
the basis of a pyrolysis facility consuming 2000 MTPD of biomass
(25 wt% moisture) on a dry basis. The primary differences between
the 2010 ISU study and the present analysis are TPI, fuel yield, and
equity percentage. The 2010 ISU study uses a fuel yield from UOP

for FP and hydroprocessing of 42 wt% (bio-oil basis) [28]. This

translates to 58.2 MGY for a 2000 MTPD facility when a bio-oil
yield of 63 wt% (biomass basis) is assumed. The present analysis
uses the slightly lower annual fuel output of 57.4 MGY, which is
based on the results of hydroprocessing experiments published
by Elliott et al. [20]. The present analysis also estimates a much
higher facility TPI than is used by the 2010 ISU study: $429 million
vs. $200 million. The 2010 ISU study TPI is calculated from Aspen
Plus and literature data [8] while the present analysis includes
capital cost data published by KiOR for its 454 MTPD facility in
Columbus, MS [26], scaled up to accommodate a 2000 MTPD commercial facility.
While the fuel output of 58.2 MGY used by the 2010 ISU study is
at the higher end of the range reported in the literature for a 2000
MTPD facility, it is not unreasonable. Assuming a gasoline weight
of 3 kg/gal, a PNNL study [16] reports a range in the literature of
49.255 MGY. Elliott et al. [20], who based their analysis on a more
recent and comprehensive experiment, reports a range of 43.7
83.9 MGY (assuming a bio-oil yield of 63 wt%). The average fuel
yield for the ve hydroprocessed bio-oils in the latter study is
26.1 wt% (biomass basis), or 57.4 MGY for a 2000 MTPD plant (assumes 90.2% online time).
The second major difference between the 2010 ISU study and
the present analysis is the TPI. The present analysis incorporates
equipment requirements from KiORs S-1 ling so it is important
to note the differences between KiORs Columbus facility and the
2010 ISU study. First, the 2010 ISU study assumes that all of the
NCG but only a fraction (1/3rd) of the char produced via FP are
used to provide process heat for the pyrolysis process and includes
a relatively small boiler for this purpose [8]. KiORs Columbus facility also meets its process heat demand in this way but employs a
much larger boiler in combination with a turbogenerator that consumes all NCG gas and char from pyrolysis as well as off-gases
from hydroprocessing to generate electricity for both facility demand and external sale [31].
While KiORs S-1 does not state the cost of the boiler and turbogenerator system, we estimate it to contribute an additional
$141.2 million to TPI using Aspen Process Economic Analyzer.
This includes a $90 million boiler unit that is large enough to
burn all of the NCG and char produced via FP as well as a $30 million turbine system for the generation of electricity from highpressure steam (for comparison, the 2010 ISU study used a $34
million boiler) [27]. While unnecessary from an operational perspective (the facility could also purchase power from the electric
grid), green house gas (GHG) regulations on renewable fuels
might justify inclusion of power generation equipment despite
its high cost. Hsu [32] of the National Renewable Energy Laboratory (NREL) recently published a life cycle analysis (LCA) of the FP
and hydroprocessing pathway that reports a GHG reduction over
gasoline of 53% when grid electricity is used and 65% when biomass-derived electricity is used. While a 53% reduction is sufcient to qualify the pathway as an advanced biofuel under the
RFS2, only the 65% reduction qualies it as a cellulosic biofuel
pathway [4].
The cost of the boiler and turbogenerator system is unlikely to
have a drastic impact on the economic feasibility of this biofuels
pathway when electricity prices are high and/or the cost of capital
is low, although it should be noted that its inclusion might not be
necessary for the fuel product to qualify as cellulosic biofuel under
the RFS2. Kauffman et al. [33] shows that even corn ethanol can
meet a 50% reduction if the stover residue from the corn crop is
pyrolyzed and the carbon-rich char product is sequestered rather
than combusted. It is therefore plausible that a FP facility could
qualify for the 60% GHG reduction without purchasing the boiler
and turbogenerator system if it sequesters the char rather than
combusts it for electricity.

T.R. Brown et al. / Fuel 106 (2013) 463469

467

equipment increased by 13.6% between December 2007 and April

2011 (the month that KiOR led its S-1), or an increase to TPI of
$27 million for a $200 million facility with a 2007 cost basis year
[35].
4. Results

Fig. 2. Installed equipment costs for the present analysis and the 2010 ISU study
(indirect costs, working capital, and land costs are excluded) [8].

Fig. 3. Annual operating costs for the present analysis and the 2010 ISU study
(capital depreciation, average income tax, and annual return on investment are
excluded) [8].

Table 6
Comparison of transportation fuel MFSPs with a 10% IRR. Source: 2010 ISU study [8].
Analysis

2010 ISU study

Present analysis

Transportation fuel MFSP ($/gal)

2.11

2.57

Differences in assumed TPI for hydroprocessing also account for

much of the discrepancy between the total facility capital costs calculated in the 2010 ISU study and the present analysis. The 2010
ISU study calculates the TPI of the hydroprocessing section to be
$26.8 million [27]. This is much lower than calculated in either
the PNNL study ($110 million) [16] or KiORs public cost data
($128 million when TPI for the KiOR Columbus facility is scaled
to a 2000 MTPD commercial facility) [34]. The present analysis calculates the hydroprocessing unit to cost $89.7 million. The 2010
ISU study modeled the hydroprocessing system as a one-stage
black box unit and underestimated the systems equipment costs
as a result. The present analysis includes separate costs for the
hydrotreater and hydrocracker, as well as costs for rening equipment not included in the 2010 ISU study (debutanizer, gasoline
column, diesel fuel column, and hydrocracker product separation
column). The present analysis also assumes hydrogen overpressurization of 50%, requiring an additional compressor and pressure
swing absorption (PSA) unit (with the consequent equipment
costs).
The third difference in TPI is the cost year basis assumed between the 2010 ISU study and the present analysis. The 2010 ISU
study uses a 2007 cost year basis while the present analysis uses
a 2011 basis year. The Chemical Engineering Plant Cost Index for

The process design in the present analysis models the FP and

hydroprocessing of 2000 MTPD of stover. This generates 57.4 million gallons of transportation fuel and 223 million kWh of electricity annually. TPI is $429 million. This is substantially higher than
the TPI of $200 million calculated in the 2010 ISU study primarily
due to the larger boiler, the inclusion of a turbogenerator and a
rening unit, and the accounting for equipment cost ination between 2007 and 2011. Installed equipment costs for the two analyses are compared in Fig. 2.
As shown in Fig. 3, annual operating costs for the FP and hydroprocessing facility are slightly lower in the present analysis than in
the 2010 ISU study after capital depreciation, average income tax,
and annual return on investment are excluded (as these are largely
driven by capital costs). The primary difference is in the type of coproduct credits offset against operating costs. The 2010 ISU study
analysis assumed that the syngas is sold as fuel gas while the char
incurs a disposal cost. The present analysis assumes that both are
combusted for process heat and electricity, with excess electricity
being sold to the grid. The electricity attracts a higher return than
the fuel gas and the char is treated as feedstock for a value-added
product (electricity) rather than as a waste product, both of which
reduce operating costs. Furthermore, electricity must be purchased
under the 2010 ISU study but does not under the present analysis.
Operating costs in the present analysis are slightly lower than under the 2010 ISU study as a result, at $83.5 million and $89.5 million respectively [8].
The 2010 ISU study [8] quantied the economic feasibility of its
system model by xing the 20-year IRR at 10% and calculating the
MFSP for product transportation fuels. The present analysis quanties economic feasibility in the same manner for the purpose of
a straightforward comparison between both analyses. The 2010
ISU study calculated a MFSP of $2.11/gal. The present analysis calculates a MFSP of $2.57/gal for a 2000 MTPD FP and hydroprocessing system (see Table 6). The major reason that the present
analysis yields a MFSP that is only 22% higher than the MFSP from
the 2010 ISU study despite employing a TPI that is 115% higher is
the inclusion of electricity revenues. While the present analysis includes the signicant expense of an expanded boiler large enough
to burn all char, NCG, and off-gas as well as a turbogenerator for
the production of electricity, which are estimated to increase TPI
by $141.2 million, this additional expense is offset by the additional revenue from selling the excess electric power, averaging
$12 million annually.
5. Discussion
The economics of a FP and hydroprocessing facility are sensitive
to a number of important variables such as process yields, input
costs, and output market values. This sensitivity analysis examines
the impact of biomass feedstock cost, bio-oil yield, electricity value, transportation fuel yield, TPI, and cost of debt on transportation fuel MFSP. These impacts are displayed in Fig. 4. Optimistic,
baseline, and pessimistic scenarios are presented for each variable.
Fuel yield is the most important variable, with a 61 wt% yield (the
highest reported by Elliott et al.) [20] resulting in an MFSP of $1.71/
gal and a 32 wt% yield (the lowest reported by Elliott et al.) resulting in $3.29/gal. MFSP is also very sensitive to bio-oil yield, with a
70 wt% yield generating a MFSP of $2.30/gal and a 49 wt% yield

468

T.R. Brown et al. / Fuel 106 (2013) 463469

Fig. 4. Sensitivity analysis for 2000 MTPD FP and hydroprocessing facility. To reach the baseline transportation fuel MFSP of $2.57, a 10% facility IRR is assumed.

generating a MFSP of $3.32/gal. The cost of debt and price of electricity have relatively little impact on product MFSP by comparison, although they are not inconsequential.
The results of this sensitivity analysis demonstrate the importance of the assumptions on the economic feasibility of a 2000
MTPD FP and hydroprocessing system. MFSP is highly sensitive
to bio-oil yield, and a reduction from 63 to 49 wt% on a feedstock
basis changes the result from a competitive facility at current
petroleum prices to a loss maker. While facility economic feasibility is not as sensitive to variables such as TPI and the price of electricity, they still have an impact. Raising TPI from $429 million to
$463 million increases MFSP by 4%. The offsetting effect of electricity revenues on the increased equipment costs incurred by the
large boiler and turbogenerator system is illustrated by the 9% decrease to MFSP that occurs when the electricity price is assumed to
be zero. Similarly, failing to account for the sale of excess electricity generated by the boiler and turbogenerator system increases
MFSP by 9%. Alternatively, forgoing electric power generation at
the facility reduces TPI to $288 million and MFSP falls by 9%.

is 22% higher than the MFSP calculated in the 2010 ISU study
($2.11/gal).
The results of the present analysis indicate that the FP and
hydroprocessing pathway might not be quite as competitive with
petroleum-based transportation fuels as originally suggested by
the 2010 ISU study. While the calculated increase to facility MFSP
is not especially large, the increase to facility TPI is and could pose
a challenge to pathway commercialization in an economic scenario
in which capital nancing is difcult to obtain. While much of the
increase to TPI is offset by electricity revenues resulting from the
inclusion of a large boiler and turbogenerator system, the cost increases attributed to the inclusion of a rening unit, a higher estimated hydroprocessor cost, and equipment cost ination are not
offset and act to increase facility MFSP. Further research is needed
to determine whether the large boiler and turbogenerator system
is necessary from a lifecycle analysis perspective to achieve the
GHG reduction threshold required by the revised Renewable Fuel
Standard (RFS2), as the systems high capital cost could hinder
the acquisition of capital nancing for the facility.

6. Conclusion

Acknowledgments

The present analysis calculates the minimum fuel selling price

(MFSP) of a facility producing gasoline and diesel fuel from stover
via fast pyrolysis (FP) and hydroprocessing. In doing so it makes
multiple changes to the assumptions used by the previous ISU
study on the same pathway based on publicly available information on a catalytic fast pyrolysis facility that KiOR has constructed
in Mississippi [8,27]. First, the present analysis includes the construction of a $141 million boiler and turbogenerator unit capable
of meeting the facilitys electricity needs via combustion of char,
non-condensable gases (NCG), and off-gas from the hydroprocessor, with excess electricity being sold to the grid. Second, the present analysis assumes that a more expensive hydroprocessing unit
is necessary for the upgrading of bio-oil and includes a rening
unit for the production of gasoline and diesel fuel from the hydroprocessed bio-oil. These changes increase facility total project
investment (TPI) by $79 million over the 2010 ISU study. A cost
year basis of 2011 is used to account for recent ination in equipment costs, increasing installed costs by $27.1 million. Finally, a
slightly lower bio-oil yield is employed in the present analysis
(57.4 wt%) than in the 2010 ISU study (58.2 wt%) to reect experimental results reported in the literature. The result of these
changes is a minimum fuel selling price (MFSP) of $2.57/gal, which

This work was supported by the Bioeconomy Institute. The

authors would like to thank Stephen McGovern and J. Andres Soria
for helpful discussions on pyrolysis commercialization assumptions. The authors would like to thank Mark M. Wright for his insights on the 2010 ISU study.

References
[1] Solomon BD, Barnes JR, Halvorsen KE. Grain and cellulosic ethanol: history,
economics, and energy policy. Biomass Bioenergy 2007;31:41625.
[2] Searchinger T, Heimlich R, Hougthon R, Dong F, Elobeid A, Fabiosa J, et al. Use
of US croplands for biofuels increases greenhouse gases through emissions
from land-use change. Science 2008;319:123840.
[3] Runge CF, Senauer B. How biofuels could starve the poor. In: Foreign affairs;
2007.
[4] EPA. Regulation of fuels and fuel additive changes to renewable fuel standard
program; nal rule. Fed Regist 2010;75:14790.
[5] Anon. EMTS data. EPA; 2011.
[6] Kazi FK, Fortman JA, Anex RP, Hsu DD, Aden A, Dutta A, et al. Techno-economic
comparison of process technologies for biochemical ethanol production from
corn stover. Fuel 2010;89:S208.
[7] Swanson RM, Platon A, Satrio JA, Brown RC. Techno-economic analysis of
biomass-to-liquids production based on gasication. Fuel 2010;89:S119.
[8] Wright MM, Daugaard DE, Satrio JA, Brown RC. Techno-economic analysis of
biomass fast pyrolysis to transportation fuels. Fuel 2010;89:S2S10.

T.R. Brown et al. / Fuel 106 (2013) 463469

[9] Anex R, Aden A, Kazi F, Fortman J, Swanson R, Wright M, et al. Technoeconomic comparison of biomass-to-transportation fuels via pyrolysis,
gasication, and biochemical pathways. Fuel 2010;89:S2935.
[10] Laird DA. The Charcoal vision: a win-win-win scenario for simultaneously
producing bioenergy, permanently sequestering carbon, while improving soil
and water quality. Agron J 2008;100:17881.
[11] Brown TR, Wright MM, Brown RC. Estimating protability of two biochar
production scenarios: slow pyrolysis vs fast pyrolysis. Biofuels Bioproducts
Biorening 2011;5:5468.
[12] McCarl BA, Peacocke C, Chrisman R, Kung C-C, Sands RD. Economics of biochar
production, utilization and greenhouse gas offsets. In: Lehmann J, Joseph S,
editors. Biochar for environmental management. London: Earthscan; 2009. p.
34158.
[13] Lian J, Chen S, Zhou S, Wang Z, OFallon J, Li C-Z, et al. Separation, hydrolysis
and fermentation of pyrolytic sugars to produce ethanol and lipids. Biores
Technol 2010;101:968899.
[14] Ahmad MM, Fitrir M, Nordin R, Azizan MT. Upgrading of bio-oil into highvalue hydrocarbons via hydrodeoxygenation. Am J Appl Sci 2010;7:74655.
[15] Vispute TP, Zhang H, Sanna A, Xiao R, Huber GW. Renewable chemical
commodity feedstocks from integrated catalytic processing of pyrolysis oils.
Science 2010;330:12227.
[16] Jones SB, Valkenburg C, Walton CW, Elliot DC, Holladay JE, Stevens DJ, et al.
Production of gasoline and diesel from biomass via fast pyrolysis,
hydrotreating and hydrocracking: a design case. Richland: Pacic Northwest
National Laboratory; 2009. p. 176.
[17] Brown TR, Zhang Y, Hu G, Brown RC. Techno-economic analysis of biobased
chemicals production via integrated catalytic processing. Biofuels Bioproducts
Bioreneries 2012;6:7387.
[18] Pootakham T, Kumar A. Bio-oil transport by pipeline: a techno-economic
assessment. Biores Technol 2010.
[19] Corma A, Huber GW, Sauvanaud L, OConnor P. Processing biomass-derived
oxygenates in the oil renery: catalytic cracking (FCC) reaction pathways and
role of catalyst. J Catal 2007;247:30727.
[20] Elliott DC, Hart TR, Neuenschwander GG, Rotness LJ, Zacher AH. Catalytic
hydroprocessing of biomass fast pyrolysis bio-oil to produce hydrocarbon
products. Environ Prog Sustain Energy 2009;28:4419.

469

[21] French RJ, Hrdlicka J, Baldwin R. Mild hydrotreating of biomass pyrolysis oils
to produce a suitable renery feedstock. Environ Prog Sustain Energy
2010;29:14250.
[22] Mullen CA, Boateng AA, Goldberg NM, Lima IM, Laird DA, Hicks KB. Bio-oil and
bio-char production from corn cobs and stover by fast pyrolysis. Biomass
Bioenergy 2009;34:6774.
[23] Bridgwater AV. Review of fast pyrolysis of biomass and product upgrading.
Biomass Bioenergy 2012;38:6894.
[24] Mullen CA, Boateng AA, Mihalcik DJ, Goldberg NM. Catalytic fast pyrolysis of
white oak wood in a bubbling uidized bed. Energy Fuels 2011;25:544451.
[25] Wang Y, He T, Liu K, Wu J, Fang Y. From biomass to advanced bio-fuel by
catalytic pyrolysis/hydro-processing: hydrodeoxygenation of bio-oil derived
from biomass catalytic pyrolysis. Biores Technol 2012;108:2804.
[26] Lane J. KiOR les $100M IPO the complete digest analysis. In: Biofuels Digest;
2011.
[27] Wright MM, Satrio JA, Brown RC. Techno-economic analysis of biomass fast
pyroysis to transportation fuels. Golden (CO): National Renewable Energy
Laboratory; 2010.
[28] Holmgren J, Marinangeli R, Nair P, Elliott D, Bain R. Consider upgrading
pyrolysis oils into renewable fuels. Anglais 2008;87:95103.
[29] Peters M, Timmerhaus K, West R. Plant design and economics for chemical
engineers. 5th ed. McGraw Hill; 2003.
[30] Anon. Tracking bond benchmarks. In: The wall street journal. New York
(NY): Merrill Lynch; 2012.
[31] C.o.E.a.E.I.o.I.B. Production, N.R. Council. Renewable fuel standard: potential
economic and environmental effects of US biofuel policy. The National
Academies Press; 2011.
[32] Hsu DD. Life cycle assessment of gasoline and diesel produced via fast
pyrolysis and hydroprocessing. Biomass Bioenergy 2012;45:417.
[33] Kauffman N, Hayes D, Brown R. A life cycle assessment of advanced biofuel
production from a hectare of corn. Fuel 2011;90:330614.
[34] KiOR. Form 10-Q quarterly report. In: Commission S.a.E., (Ed.), KiOR,
Washington, DC; 2011.
[35] Anon. Economic Indicators. Chem Eng 2011;118:63.