Fuel 332 (2023) 126022

Contents lists available at ScienceDirect

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

Full Length Article

Techno-economic assessment of an integrated biorefinery producing

bio-oil, ethanol, and hydrogen
Temitayo Giwa a, Maryam Akbari a, b, Amit Kumar a, *
a
Department of Mechanical Engineering, 10-263 Donadeo Innovation Centre for Engineering, University of Alberta, Edmonton, Alberta T6G 1H9, Canada
b
CanmetENERGY, Natural Resource Canada (NRCan), 1615 Lionel-Boulet Boulevard, Varennes, Quebec J3X 1P7, Canada

A R T I C L E I N F O A B S T R A C T

Keywords: The production of multiple products, in an integrated manner, encourages the full valorization of biomass
Multi-product feedstock and reduces waste streams. This may improve the economic outlook of the biorefinery and provide a
Biorefinery hedge for investors. There is limited understanding of the techno-economics of an integrated, multi-product
Techno-economic assessment
biorefinery. This study addresses this gap. In this study, we investigated an integrated, multi-product bio­
Ethanol
refinery. In addition to bio-oil, the valuable products ethanol and hydrogen are produced from fast pyrolysis by-
Fermentation
Pyrolysis products. We developed process models for bio-oil, ethanol, and hydrogen production from spruce woodchips in
an integrated syngas fermentation and fast pyrolysis plant. Six scenarios set against a base fast pyrolysis plant
were investigated, and the economic performance of each scenario was evaluated based on the rate of return. The
scenarios differ in terms of the use of the non-condensable gases and the type of external fuel used for process
heat. In every scenario, biochar is used for ethanol production and bio-oil is assumed to be a final product. The
base case had a return rate of 7.0%, while the return rate ranged between 7.46 and 13.01% among the six
scenarios. The scenario producing hydrogen from hydrogen-rich non-condensable gases and ethanol from bio­
char was the most profitable. The profitability was found to be most sensitive to product prices, capital cost, and
feedstock cost. The study’s results show that an integrated, multi-product approach to biorefining has the po­
tential to improve the economics of biorefineries and biofuel production.

The traditional approach for biomass valorization focuses on pro­

1. Introduction ducing a single product of interest [3], with by-products going to less
valuable applications, the most common of which is combustion [4]. A
Petroleum has contributed immensely to the prosperity we experi­ potential solution is to follow an integrated, multi-product approach to
ence today. A large portion of global energy demand is satisfied by pe­ add value to by-products or co-products. Having multiple value-added
troleum [1,2]. However, burning petroleum-based fuels has adverse products increases the revenue of the biorefinery and thus can be a
effects on the climate. Moreover, petroleum resources are finite. A sus­ potential means to improve the economics of renewable fuels and
tainable energy future requires supply security and CO2 emissions chemicals production from lignocellulosic feedstocks. Moreover, adding
reduction. Bio-fuels can play an important role in the global energy mix more conversion technologies to a biorefinery to convert different ma­
and are especially important for hard-to-decarbonize sectors, like the terial streams such as intermediate products and/or by-products into
transportation and petrochemical sectors. Today, food-based biomass is more valuable and marketable products encourages the full use of the
the major source of commercially available bio-fuels. Using these food- original feedstock. Hence, waste streams are channeled for valuable
based bio-fuels, mainly ethanol, challenges food security and raises product production. This improves the overall rate of return from the
ethical questions. Lignocellulosic biomass, the so-called second-gener­ investment. Moreover, having multiple products from a biorefinery
ation biomass, which is used in this study, has been a major subject of ensures that investors are protected against market uncertainties. This is
research in the bio-energy domain as a potential solution to the prob­ because the products serve as a hedge for each other against unfav­
lems of food-based bio-fuels. The production of bio-fuels from ligno­ ourable market conditions. This is, in fact, the approach in the petro­
cellulosic feedstock is plagued by a lack of economic competitiveness, leum industry [5].
especially when compared with the petroleum-based fuels. Gasification and pyrolysis technologies are the two most common

* Corresponding author at: 10-263 Donadeo Innovation Centre for Engineering, 9211-116 St, Edmonton, AB T6G 2H5, Canada.
E-mail address: Amit.Kumar@ualberta.ca (A. Kumar).

https://doi.org/10.1016/j.fuel.2022.126022
Received 2 June 2022; Received in revised form 30 August 2022; Accepted 13 September 2022
Available online 24 September 2022
0016-2361/© 2022 Elsevier Ltd. All rights reserved.
T. Giwa et al. Fuel 332 (2023) 126022

Nomenclature LF Location factor

M Maintenance cost
APEA Aspen Process Economic Analyzer NCGs Non-condensable gases
CAPEX Capital expenditure OC Operating charges
CO Carbon monoxide OPEX Operating expenditure
CO2 Carbon dioxide PO Plant overhead
FCI Fixed capital investment PSA Pressure swing adsorption
G&A General and administrative RUST Regression, Uncertainty, and Sensitivity Tool
GHG Greenhouse gas TDIC Total direct and indirect cost
H2 Hydrogen TEC Total equipment cost
IC Indirect cost TIC Total installed cost
IRR Internal rate of return

thermochemical technologies for biomass valorization. In gasification, predominantly from grains [20], though lignocellulosic feedstock is also
the production of multiple products relies solely on the conversion of the used. Both rely on the use of fermentable substrates derived from
syngas generated during the gasification process. In the fast pyrolysis biomass. However, this approach cannot be used to derive ethanol from
process, biomass is heated in an inert atmosphere to a temperature in the biochar. The gasification of biochar, however, allows ethanol to be
range of 400–550 ◦ C, producing bio-oil, non-condensable gases and produced from biochar as converts it to fermentable gases. The syngas
biochar [6]. In the case of pyrolysis (fast pyrolysis in particular), several produced through biochar gasification can then be used for the synthesis
products have been derived from bio-oil, the major product of interest in of ethanol. This synthesis can be either thermochemical or biochemical.
the fast pyrolysis process [6]. Bio-oil has been produced with the aim of Biochemical processes like syngas fermentation have some advantages
replacing heavy fuel oil (HFO) or coal, light fuel oil (LFO), gas oil, and over the thermochemical process. While fermentation occurs at near
vacuum gas oil [7]. Because bio-oil can significantly reduce carbon ambient temperature and pressure, the thermochemical process requires
emissions, the market for bio-oil, especially for use in boilers or furnaces elevated temperature and pressure [21,22]. Also, stringent gas cleaning
(in place of natural gas and heavy and light fuel oils), gas turbines, and is required in the thermochemical process as the catalysts are prone to
diesel engines, is expected to grow rapidly in the coming years (with deactivation by even very small levels of impurities in the gas stream.
North America dominating) [8]. Bio-oil is also likened to crude-oil and is Microbes, on the other hand, can cope under higher levels of impurities.
subjected to further processing for the simultaneous production of gas­ Another advantage to the use microbes in general is that there is no fixed
oline and diesel [9,10]. Another emerging approach is to fractionate the ratio for H2/CO as there is in the catalytic process [22].
bio-oil into different fractions based on the functional groups and po­ The gasification-syngas fermentation process has been modelled in
tential application and then further process each fraction into valuable the existing literature. Pardo-Planas et al. [23], using switchgrass as
end products [11]. Although this simultaneous production of gasoline feedstock, studied the impacts of the gas uptake rate and ethanol con­
and diesel from bio-oil is a multi-product approach, it focuses solely on centration achieved in the fermentation unit on the volume of the
bio-oil and neglects the potential to valorize pyrolysis by-products into fermenter and the energy requirement of the system, respectively. De
valuable end products. Instead, the non-condensable gases and, in most Medeiros et al. [24] carried out the modelling and economic analysis of
cases, biochar produced in the fast pyrolysis process, in addition to bio- an energy self-sufficient gasification-syngas fermentation process with
oil, are usually used to generate process heat, although there are other sugarcane bagasse as feedstock. Ro et al. [25] assessed the economics of
suggested uses for biochar such as soil amendment [12], activated car­ ethanol production from syngas derived from wood chips, corn stover,
bon production [13], and carbon nanomaterial production [14,15]. wheat straw, swine manure mixed with wheat straw, and oil seed rape
However, these applications are not strategically integrated into the meal. In addition to a techno-economic assessment, Roy et al. [21]
existing biorefinery. Valuable products like ethanol, which fits strate­ calculated the GHG emissions of the process. They considered mis­
gically into the product portfolio of a biorefinery, can be produced from canthus as the feedstock with four pretreatment options for an ethanol
biochar. Non-condensable gas streams are hydrogen-rich and can be production rate of 22 million liters per year. Benalcazar et al. [26]
purified to hydrogen. Converting pyrolysis by-products to value-added conducted environmental and techno-economic assessments of ethanol
products encourages new investments in pyrolysis plants, which face production from sugarcane bagasse, pine wood, corn stover, and euca­
challenges due to the lack of standards for pyrolysis by-products [16]. lyptus wood in three countries, the Netherlands, Brazil, and the United
As an energy carrier hydrogen is very versatile, as it can be stored, States. Although these studies assessed the gasification-syngas fermen­
used, and transported as gas or liquid and can produce electricity or fuel tation process from technical, economic, and environmental perspec­
[17]. Hydrogen has many applications and is used globally in ammonia tives, they all considered standalone systems. There is no study, to the
production oil refining, chemical and fertilizer production, steel best of the authors’ knowledge, that has considered ethanol production
manufacturing, metallurgy, food processing, etc. Experts have advised from the fermentation of biochar-derived syngas integrated with fast
that meeting the goals of the Paris Agreement is not possible without pyrolysis.
green hydrogen, as eliminating emissions in some sector is difficult. The In this study, biochemical and thermochemical technologies were
infrastructure for hydrogen delivery exists in parts of the world such as integrated to increase the products in the portfolio of the fast pyrolysis
the U.S., where there are hydrogen pipelines, which is another advan­ biorefinery as a potential way to achieve more efficient use of the
tage for the large-scale production of hydrogen [17]. original biomass. Different scenarios of this integration were assessed to
Ethanol is versatile; it can serve as a solvent, beverage, antifreeze, produce ethanol and hydrogen from the non-condensable gases and
and fuel. Perhaps the most important application is as fuel. Today biochar, in addition to bio-oil, by evaluating the economic implications
ethanol serves as a blend for gasoline and improves its octane number, and comparing them with the conventional fast pyrolysis process. The
replacing tetraethyl lead. Apart from rising consumption of alcoholic following objectives were set:
beverages, several countries have mandated the use of ethanol in
transportation as a clean fuel, and this has led to a significant growth in • To develop process models of an integrated, multi-product bio­
its demand [18,19]. The production of ethanol from biomass sources is refinery producing bio-oil, ethanol, and hydrogen.

2
T. Giwa et al. Fuel 332 (2023) 126022

• To develop a techno-economic model to assess the rate of return for sub-scenarios each) were developed to carry out this assessment. The
each scenario and compare the results with conventional fast feedstock considered in this study is spruce woodchips and is processed
pyrolysis. at a rate of 2000 dry tonnes/day, in all cases. To have a good reference
• To assess six scenarios that consider different uses of non- point for the comparison of the pathways developed, the conventional
condensable gases (NCGs) and two types of external fuel (natural pyrolysis process (the base case) was also modelled. Fig. 1 (a) is a
gas and woodchips). simplified process flow diagram of the base case. The feedstock is dried
• To perform sensitivity analysis to understand the impact of input and ground to meet the requirements for pyrolysis. During pyrolysis,
parameters on the rate of return. bio-oil, biochar, and NCGs are produced. Bio-oil is considered a final
• To carry out uncertainty analysis to understand the effect of the product. NCGs are burned to satisfy process heat requirements, and
variability of the most impactful parameters on the rate of return. some of the produced biochar is used to supplement the heat demand. In
scenario 1, the production of ethanol from biochar through gasification
2. Method and subsequent syngas fermentation was assessed. NCGs are considered
in this scenario as fuel, and an external fuel (either natural gas or
This section discusses the approach used in this study. The scenarios woodchip) is used as a supplement. In every scenario, “a” indicates that
and the unit processes involved in the process modelling are described. natural gas is the supplemental fuel and “b” that woodchip is the sup­
plemental fuel. The process flow diagram of scenario 1 is shown in Fig. 1
(b). Scenario 2 (see Fig. 1 (c)) is like scenario 1 in terms of the use of
2.1. Scenario description biochar, but here, the NCGs are combined with the syngas generated
from the gasification of biochar to produce more ethanol. Scenario 3 (see
This study examines the potential to improve the economics of the Fig. 1 (d)) also uses biochar for ethanol production like scenarios 1 and
pyrolysis process. Six process scenarios (three major scenarios with two

Fig. 1. (a) Simplified process flow diagram of the base case (conventional fast pyrolysis process), (b) Simplified process flow diagram of scenario 1 (in addition to
bio-oil, ethanol is produced from biochar), (c) Simplified process flow diagram of scenario 2 (in addition to bio-oil, ethanol is produced from biochar and NCGs), and
(d) Simplified process flow diagram of scenario 3 (in addition to bio-oil, ethanol is produced from biochar and hydrogen is produced from NCGs).

3
T. Giwa et al. Fuel 332 (2023) 126022

Fig. 1. (continued).

2; however, the NCGs are purified for hydrogen production. Table 1 2.2. Description of process units and modelling
summarizes the scenarios.
The process modelling carried out in this study was done in Aspen
Plus and has three sections. The first is the modelling of the conventional
fast pyrolysis process to yield bio-oil (the main product), NCGs, and

4
T. Giwa et al. Fuel 332 (2023) 126022

Table 1 The experimental yields used for this model are taken from an earlier
Description of scenarios. study [28]. The outputs of the pyrolyzer are volatile components and
Scenario Description biochar. The volatile component contains bio-oil compounds, which are
ID a liquid at room temperature, and NCGs, which are a gas at room tem­
Base case Conventional fast pyrolysis process for production of bio-oil, biochar perature. These volatile compounds and the solid biochar particles were
and NCGs separated with the help of cyclones with a solid removal efficiency of 90
1a NCGs are combusted to generate process heat, and natural gas is used %. To extract the bio-oil, the volatile stream first needs to be cooled.
as supplemental fuel Cooling the volatiles from 490 ◦ C to 50 ◦ C condenses the bio-oil com­
1b NCGs are combusted to generate process heat, and biomass is used as
supplemental fuel
pounds. This extracted heat was used to generate some of the steam
2a NCGs are converted to bioethanol, and natural gas is the sole fuel for required for the gasification process. Bio-oil is collected and stored, and
process heat generation NCGs are used for heat generation, ethanol production, or hydrogen
2b NCGs are converted to bioethanol, and biomass is the sole fuel for production, depending on the biorefinery scenario.
process heat generation
3a Hydrogen is produced from the NCGs, and natural gas is the sole fuel
for heat generation 2.2.3. Biochar gasification
3b Hydrogen is produced from the NCGs, and biomass is the sole fuel for The gasification of biochar was modelled through Gibbs energy
heat generation minimization. An RGibbs reactor was used to simulate the gasification
process and predict syngas composition. The gasification temperature
was 800 ◦ C and atmospheric pressure was considered [31]. The steam-
biochar. The pyrolysis section of the model, which is common to all
to-char ratio was set such that biochar conversion was 80 %, as re­
scenarios, consists of feedstock pretreatment, the pyrolysis unit, and
ported by Chaudhari et al. [31]. Gasification was assumed to be carried
product separation. In this study, one of the major goals is to use the
out in an indirectly heated dual fluidized bed gasifier. This gasifier
biochar to produce ethanol that can be sold, thereby improving the
configuration has two reactors, a gasification reactor, and a combustion
economics of bio-oil production. The production of ethanol from biochar
reactor. Steam gasification, an endothermic reaction, takes place in the
is the second section of the modelling. This section, also common to all
gasification reactor. The heat required for steam gasification is supplied
scenarios, consists of biochar gasification to produce syngas, syngas
by the second reactor. A solid heat carrier, olivine, circulates between
fermentation to produce ethanol, and product separation for the con­
the two reactors. The olivine circulation rating was taken as 12.3 kg/kg
centration of ethanol. The last section is the purification of pyrolysis gas
of bone-dry feed (biochar), as reported in an earlier study [32]. Biochar
or NCGs. For scenario 1, NCGs were burned in a combustor along with
is a non-conventional compound in Aspen Plus. Hence, it was specified
supplemental fuel to provide process heat. For scenario 2, NCGs were
as a non-conventional compound in the process model using its ultimate
mixed with biochar syngas to be fermented. For scenario 3, the sepa­
analysis as in an earlier study [33].
ration of hydrogen from the NCGs using a pressure swing adsorption
system was considered. To ensure that the modelling developed in this
2.2.4. Syngas fermentation
study is accurate, each process unit was developed and validated with
Microorganisms capable of fermenting syngas are known as ace­
results found in other studies. Gasification, syngas fermentation, and
togens [34]. Acetogens are anaerobic micro-organisms (including bac­
ethanol purification were validated with Medeiros et al.’s work in terms
teria and archaea) that use the acetyl-CoA pathway to fix CO2 [35] and
of ethanol yield [24]. The rest of the unit processes including feedstock
other substrates like glucose [36]. The products from syngas fermenta­
pretreatment and pyrolysis units were validated with the study by Patel
tion generally fall into two major categories: carboxylic acids and cor­
et al. in terms of products (biochar, bio-oil, syngas) yield and compo­
responding alcohols. Usually, carboxylic acid is readily produced, unlike
sition [6]. The following sections discuss the process units considered in
alcohol. Acetic acid, ethanol, butyric acid, butanol, hexanoic acid,
this study.
hexanol, and 2,3-butanediol can be produced from syngas fermentation
[34]. Of these, acetic acid and ethanol are the prominent ones [34]; both
2.2.1. Feedstock and pretreatment
are readily produced by many acetogens [36]. The production of ethanol
The feedstock studied here is spruce wood chips. Spruce is an
from syngas is shown by Eqs. (1) and (2), as given by Pardo-Planas et al.
important tree species globally and in Canada; it makes up about 47 % of
[23].
Canada’s total forest inventory [27]. The ultimate and proximate ana­
lyses of the feedstock are taken from an earlier study [28]. 6CO + 3H2 O = C2 H5 OH + 4CO2 (1)
The particle sizes of the as-received feedstock is not suitable for the
pyrolysis process and contains 50 % moisture [28]. For fast pyrolysis, it 6H2 + 2CO2 = C2 H5 COOH + 3H2 O (2)
is recommended that the particle size is at most 2 mm [29]; this size was
The fermentation process was modelled, with Eqs. (1) and (2) as the
adopted in this study. Consequently, the feedstock was ground in a mill
stoichiometric equation inputs. The process conditions for syngas
to a particle size not more than 2 mm. The grinding operation was
fermentation were based on experimental data from the work by Gaddy
modelled using the crusher block in Aspen Plus. The grinding energy
et al. [37], who used Clostridium ljundhalii as the acetogen. Their work
consumption was taken as 33 kWh/t [30]. The grinding operation is
considered 20 scenarios with different process conditions to see which
followed by drying, modelled using a RStoic reactor and a flash sepa­
process would produce the most ethanol. This study is based on the
rator. Rogers and Brammer recommend that the moisture content of the
approach that recirculates the bottom product (i.e., acetic acid and
feed entering the pyrolysis reactor be less than 10 wt% [29]. Drying the
water) of the ethanol purification stage into the fermenter, as demon­
feedstock using the heat recovered from the flue gas was done to a 7 wt%
strated in earlier work by Gaddy et al. [38]. This approach eliminated
moisture content.
the need to separate the acetic acid produced during fermentation; the
water stream containing acetic acid can be recycled. Recycling the acetic
2.2.2. Pyrolysis
acid led to an equilibrium concentration in the system and thus the
The modelling of the pyrolysis process is based on previous experi­
microbes did not produce more acetic acid. DeMedeiros et al. [24] used a
mental work conducted on the pyrolysis of spruce wood chips in a batch
similar approach. Off-gas from the fermentation unit was combusted to
fluidized bed pyrolysis reactor [28]. The process was carried out at a
generate process heat.
temperature of 490 ◦ C, under atmospheric pressure. The pyrolysis
reactor was modelled using the RYield block. This block requires py­
2.2.5. Ethanol separation and purification
rolysis product yield and, with it, estimates process energy consumption.
The ethanol produced in the fermentation reactor is highly diluted

5
T. Giwa et al. Fuel 332 (2023) 126022

(2 wt%). This stream must be concentrated to a purity of 99.9 % if the Table 2

ethanol is to be used as fuel. The separation and purification of ethanol Economic parameters.
occur in two different units to achieve the purity specified. First, the Parameter Value Source
broth from the fermentation unit is concentrated to 93 wt% in the
Base year 2019
distillation column. The distillation column was simulated. The column Currency USD
has 25 stages and the feed to the column enters at stage 11. A reflux ratio
of 5 (on a mole basis) was specified. Second, the concentrated ethanol
Plant characteristics
stream exiting the distillation column was sent into a molecular sieve Lifetime 20 years
separation unit. Purification of ethanol beyond 93 wt% could not be Location Alberta, Canada
achieved because of the azeotropic nature of the ethanol–water mixture Operating hours 8000 hr/yr
[38]. In the molecular sieve separation unit, ethanol was further purified
to a purity of 99.9 %. The molecular sieve separation dehydrates the Material Market prices
ethanol through adsorption. This system consists of packed columns of Hydrogen 1.50 $/kg [46]
Bio-oil* 0.32 $/L [47]
adsorption materials. The molecular sieve separation process was
Ethanol 0.36 $/L [48,49]
modelled using a separator block in the process model. Electricity 0.068 $/kWh [50]
Natural gas 1.51 $/GJ [51]
2.2.6. Hydrogen production Spruce woodchips 43.23 $/t [28]
Olivine 232.70 $/t [52,53]
Hydrogen was produced only in scenario 3 and was done in a pres­
sure swing adsorption (PSA) unit to separate hydrogen from the NCG
stream. The PSA is a commonly used technology for the separation of Labour rate
Operator labour 30.77 $/hr [54,55]
hydrogen from other gases [39,40]. One of the advantages of the PSA
Supervisor labour 38.46 $/hr [54,55]
process is that it can achieve high purity hydrogen, as high as 99.9 %
*
[41]. In this study, the hydrogen content in the NCG stream was about There is no clear market for the large-scale production of bio-oil at the
75 % by volume. The PSA functions through adsorption, which occurs in moment, and one needs to be demonstrated so that all the bio-oil produced can
be used.
the PSA beds. These beds consist of solid adsorption materials like
zeolite and activated carbon [42]. The type of adsorbents used depends
on the composition of the hydrogen stream that needs to be purified.
Table 3
This is because the affinity for each impurity differs with the type of Base case equipment cost.
adsorbent [42]. Separating hydrogen from the impure hydrogen stream
Unit Base Scaling Base Scale Source
is possible because the desirable adsorbent has less affinity for hydrogen.
equipment parameter capacity factor
The adsorption of impurities within the beds occurs at high pressure, cost ($k)
in the range of 10–50 bar [43,44]. A pressure of 30 bar was selected in
Gasification $4,760 Solid flow 500 0.60 [32]
this study. The NCGs are thus compressed from atmospheric pressure to rate tonnes/
30 bar. A compression ratio of 2.8 was selected for each compressor. The day
PSA process, however, works near ambient temperature [44]. Once a Fermentation $1,759 NA* 1000 m3 NA [37,56]
bed of adsorbent becomes saturated with impurities, it is taken offline Molecular $2,987 Purge stream 22,687 0.60 [32]
sieve flow rate kg/hr
for regeneration (desorption) using purge gases such as nitrogen [45].
PSA bed $7,063 Impurities 0.294 0.74 [57,58]
The desorption process takes place at low pressure. Hence, the pressure stream flow kmol/s
in the bed must be lowered to near atmospheric pressure. The desorbed rate
impurities’ stream is also used for combustion. To ensure continuous *
A bubble column reactor is assumed. The given cost is the cost of one reactor
operation, several PSA beds are used in the PSA process [44]. Hydrogen of 1000 m3 volume, which is the practical maximum volume currently [58]. The
recovery in the PSA process has been reported to be between 80 and 90 number of vessels was determined by combining the maximum reactor volume,
% [44]. A hydrogen recovery of 90 % was adopted in this study. this study’s syngas flow rate, and reaction residence time.

2.3. Techno-economics assessment

Table 4
2.3.1. Economic evaluation of the base case Cost component of capital investment.
The assessment of the profitability of each scenario developed in this Component Formula
study was based on the internal rate of return (IRR) generated by the Total equipment cost (TEC) Obtained from equipment sizing
scenario. This rate of return is the minimum return that the plant must Total installed cost (TIC) 302 % of TEC
generate for the investors to just recuperate all their expenses. The ex­ Indirect cost (IC) 89 % of TEC
penses are twofold: the capital expenditure (CAPEX) and the operating Total direct and indirect cost (TDIC) TIC + IC
Contingency (Con.) 20 % of TDIC
expenditure (OPEX). A discounted cash flow model was used to assess Fixed capital investment (FCI) TDIC + contingency
the rate of return from the combined CAPEX and OPEX. The assumptions Location factor (LF) 10 % of FCI
for our discounted cash flow analysis are given in Table 2. Total capital investment (TCI) FCI + LF
The capital expenditure is estimated from the cost of each piece of
equipment used with the plant. The purchase equipment costs for con­
operating cost. The variable operating cost includes the costs of the
ventional equipment were determined using the process model, and the
feedstock, utilities, and raw materials. The fixed operating cost includes
purchase equipment costs of other non-conventional equipment were
labour and supervision, maintenance, and other general and adminis­
taken from published sources.
trative costs.
Table 3 gives the base equipment cost derived from the literature.
Table 5 gives the fixed operating cost schedule used in this study.
In addition to the purchase equipment cost, costs are incurred to put
the equipment in place at the plant site. Table 4 gives the breakdown of
2.3.2. Sensitivity and uncertainty analysis
the factors used in estimating the CAPEX from the equipment purchase
A Morris sensitivity analysis was conducted to determine the input
cost.
parameters that have the most impact on the rate of return of the
The OPEX has two parts: the variable operating cost and the fixed

6
T. Giwa et al. Fuel 332 (2023) 126022

Table 5 the base case is about 52 %.

Operating cost schedule. Energy consumption for the major process units is given in Table 7.
Component Formular The energy consumption values shown for the common process units are
the same in all scenarios, except for the drying and distillation unit. In
Maintenance cost (M) 3 % of TCI
Operating charges (OC) 25 % of labour cost the case of drying, the calculated energy demand is about 60.8 MW in all
Plant overhead (PO) 50 % of (labour cost + maintenance cost) cases where natural gas is combusted. However, in the cases where
Sub-total operating cost M + OC + PO + labour cost + raw materials cost + woodchips are combusted, additional woodchips, hence additional
utilities cost drying energy, are required. These additional energies are about 10.6,
General and administration (G 8 % of sub-total operating cost
& A)
22.4, and 20.7 MW more than the base drying requirement of 60.8 MW
Total operating cost G & A + sub-total operating cost for just the feedstock. The distillation operation consumes more energy
in scenario 2 than in scenarios 1 and 3 because it processes a larger
amount of ethanol. In the base case, all the NCGs are consumed to satisfy
process. The sensitivity analysis was done using the Regression, Un­ 86 % of the heat demand, with supplemental heat being satisfied by 16
certainty and Sensitivity Tool (RUST) [59]. RUST was further used with % of the biochar generated. In scenario 1, 68 % of the heat requirement
the parameters of greatest influence to carry out a Monte Carlo simu­ is met by combusting all the NCGs and the off-gas (produced at the rate
lation to determine the uncertainty in the profitability of the processes. of 375 tonnes/day) from the syngas fermenter. In scenario 2, off-gas
from the fermenter produced at the rate of 421 tonnes/day satisfies
3. Results 38 % of the heat demand, while in scenario 3, off-gas from the PSA and
the fermenter is combined to satisfy 37 % of the heat demand. The
3.1. Production rates and energy consumption amount of external fuel used to satisfy the energy balance in all the
scenarios is shown in Table 6.
Table 6 presents the production rate of the three final products and
the amount of external fuel required for each scenario. This study
considered a fast pyrolysis plant processing 4000 tonnes/day of biomass 3.2. Techno-economic assessment
with 50 % moisture which was reduced to 7 % by drying. This is
equivalent to 2000 dry tonnes of biomass/day. The production rates of 3.2.1. Capital cost
bio-oil, biochar, and NCGs in the pyrolysis unit are 1023, 303, and 718 The capital costs of the base case and scenarios 1, 2, and 3 are broken
tonnes/day, respectively. In all three scenarios, ethanol is produced down in Table 8 by capital cost categories. As expected, the capital costs
from biochar through biochar gasification with a syngas production rate of the scenarios are higher than those of the base case because of the
of 555 tonnes/day. additional capital investment in the extra equipment in all three sce­
The production rate of ethanol is the same in scenarios 1 and 3. This narios. Scenario 2 is the most capital-intensive plant, with a cost that is
is because they process syngas derived from biochar gasification only. about 2.6 times that of the base case. However, scenario 3 is only slightly
Scenario 2 processes more gases (NCGs from pyrolysis + syngas from less costly than scenario 2. Scenario 1, the least costly among the sce­
biochar gasification) and produces nearly double the amount of ethanol narios, is also almost double the base case. Scenario 1 is the least costly
produced in scenarios 1 and 3. The ethanol yield in scenarios 1 and 3 was of all the scenarios because it processes a lesser amount of gas, and thus
derived to be 857 L/tonne of biochar and in scenario 743 L/tonne of produces lesser amount of ethanol than scenario 2. Also, in addition to
biochar plus NCGs. The existing literature on syngas fermentation re­ bio-oil, scenario 1 produces only ethanol. Meanwhile, scenario 3 pro­
ports ethanol yields of 212–546 L/tonne of biomass feedstock duces ethanol and hydrogen in addition to bio-oil. For these mentioned
[21,23–26,60,61]. The relatively higher yield reported in this study is reasons, scenario 1 has correspondingly smaller and fewer equipment
due to the higher carbon content of biochar compared to whole biomass. than scenarios 2 and 3, respectively.
The carbon content of spruce woodchips biochar considered in this study Fig. 2 presents the breakdown of the capital cost for the base case and
is 92 %, while in the cited literature, the carbon content of the feedstocks the scenarios considered, according to the major processing units in the
is between 45 % and 53 %. Hydrogen production is only possible in biorefineries. The difference between the capital cost of the biorefinery
scenario 3 at about 18,000 tonnes/year. The overall mass yields were 61 scenarios and the base case has been established. Hence, the capital cost
%, 69 %, and 64 % in scenarios 1, 2, and 3, respectively. Meanwhile, the breakdown discussed hereafter is focused on the biorefinery scenarios.
base case has an overall mass yield of 64 %. Hence, only scenario 2 has a In every scenario, the pyrolysis plant accounts for a significant share of
better feedstock conversion compared to the base case. However, the the total capital cost and the largest share in both scenarios 1 and 3, at
market outlook of ethanol and hydrogen is better than biochar. Hence,
from an economic standpoint, if biochar is excluded, the mass yield of Table 7
Energy consumption of major unit operations.
Unit Energy consumption (MW)
Table 6
Production rates and external fuel consumption. Heat
Biomass drying 60.80α
Bio-oil Ethanol H2 Ext. fuel Fuel Pyrolysis 60.45
(m3/yr) (m3/yr) (tonne/ (tonne/yr) type Gasification 42.01
yr) Distillation 39.73* (74.00**)
Scenario 289,000 85,000 – 45,504 NG
1a Electricity
Scenario 289,000 85,000 – 231,647 Biomass PSA compression unit 6.19
1b Grinding 5.5
Scenario 289,000 158,860 – 97,848 NG
2a
α
This the energy consumption for scenarios consuming natural gas. An
Scenario 289,000 158,860 – 492,557 Biomass additional 10.6, 22.4 and 20.7 should be added to scenarios 1b, 2b, and 3b,
2b respectively (when woodchips are the external fuel).
Scenario 289,000 85,000 18,170 90,608 NG *
This is the distillation unit energy consumption associated with sce­
3a narios 1 and 3.
Scenario 289,000 85,000 18,170 454,667 Biomass **
This is the distillation unit energy consumption associated with sce­
3b
nario 2.

7
T. Giwa et al. Fuel 332 (2023) 126022

Table 8
Capital cost (costs are in million $) for a 2000 tonne per day biorefineries.
Base case Scenario 1a Scenario 1b Scenario 2a Scenario 2b Scenario 3a Scenario 3b

Total equipment cost $27.95 $54.19 $54.75 $72.08 $73.23 $71.72 $72.78
Total installed cost $84.41 $163.65 $165.34 $217.67 $221.14 $216.59 $219.80
Indirect cost $24.88 $48.23 $48.73 $64.15 $65.17 $63.83 $64.78
Total indirect cost $109.28 $211.88 $214.07 $281.82 $286.31 $280.41 $284.58
Contingency $21.86 $42.38 $42.81 $56.36 $57.26 $56.08 $56.92
Fixed capital investment $131.14 $254.26 $256.88 $338.18 $343.57 $336.50 $341.50
Location cost $13.11 $25.43 $25.69 $33.82 $34.36 $33.65 $34.15

Total capital investment $144.25 $279.69 $282.57 $372.00 $377.93 $370.15 $375.65

Fig. 2. Capital cost distribution of the base case and biorefinery scenarios (Scenario 1a: Ethanol from biochar; natural gas for supplementary heat | Scenario 1b:
Ethanol from biochar; woodchips for supplementary heat | Scenario 2a: Ethanol from biochar and NCG; natural gas for supplementary heat | Scenario 2b: Ethanol
from biochar and NCG; woodchips for supplementary heat | Scenario 3a: Ethanol from biochar; Hydrogen from NCG; natural gas for supplementary heat | Scenario
3b: Ethanol from biochar; Hydrogen from NCG; woodchips for supplementary heat).

about 39 % and 30 %, respectively. It also accounts for about 30 % of the are burnt instead. Furthermore, scenarios 1 and 3 share similar dollar
total capital cost in scenario 2. However, in scenario 2, the cost of the values for the fermentation and ethanol purification units (because they
fermentation unit is the largest cost contributor at about 42 %. The cost process the same amount of syngas), while those of scenario 2 are about
of the fermentation unit is also significant in scenarios 1 and 3, ac­ double the costs of scenarios 1 and 3.
counting for 28 % and 21 %, respectively. The fermentation cost is
higher (double) in scenario 2 than in scenarios 1 and 3 because a larger 3.2.2. Operating cost
amount of gases is processed into ethanol. Both scenarios 1 and 3 have a The total operating cost for each case is constituted by fixed and
gas input of 648 m3/min compared to 1323 m3/min in scenario 2. variable operating costs. Fig. 3 gives the operating costs of the scenarios
However, for scenario 3, the cost of the PSA unit is higher than that of and base case. It should be noted that the operating costs presented are
the fermentation unit, constituting nearly 24 % of the capital cost of for the first year of biorefinery operation. However, a yearly escalation
scenario 3. This high cost of the PSA unit is largely caused by the in each operating cost category was considered in the discounted cash
compression requirement of the PSA process, which needs expensive flow analysis.
compressors. As shown in Fig. 3, the base case has the lowest operating cost. This is
It is important to note the cost similarities and differences in the because in the six biorefinery scenarios there are additional costs to
scenarios. As shown in Fig. 2, the dollar contribution values of the py­ operate the extra equipment. The total delivered feedstock cost is the
rolysis and gasification units are the same in all scenarios. The pre­ parameter that most influences the operating cost; it is responsible for
processing cost values are the same in all cases where natural gas is about 74 % of the base case. The percentage contribution of the feed­
burnt; however, it is slightly higher in the scenarios where woodchips stock cost in the operating cost is between 51 % and 64 % in the

8
T. Giwa et al. Fuel 332 (2023) 126022

Fig. 3. Base case and scenarios operating costs (Scenario 1a: Ethanol from biochar; natural gas for supplementary heat | Scenario 1b: Ethanol from biochar;
woodchips for supplementary heat | Scenario 2a: Ethanol from biochar and NCG; natural gas for supplementary heat | Scenario 2b: Ethanol from biochar and NCG;
woodchips for supplementary heat | Scenario 3a: Ethanol from biochar; Hydrogen from NCG; natural gas for supplementary heat | Scenario 3b: Ethanol from biochar;
Hydrogen from NCG; woodchips for supplementary heat).

Fig. 4. Internal rate of return of biorefinery scenarios (Scenario 1a: Ethanol from biochar; natural gas for supplementary heat | Scenario 1b: Ethanol from biochar;
woodchips for supplementary heat | Scenario 2a: Ethanol from biochar and NCG; natural gas for supplementary heat | Scenario 2b: Ethanol from biochar and NCG;
woodchips for supplementary heat | Scenario 3a: Ethanol from biochar; Hydrogen from NCG; natural gas for supplementary heat | Scenario 3b: Ethanol from biochar;
Hydrogen from NCG; woodchips for supplementary heat).

9
T. Giwa et al. Fuel 332 (2023) 126022

biorefinery scenarios. Following the feedstock cost, the costs of main­ has slightly lower capital and operating costs and higher revenue than
tenance and utilities are influential. The maintenance cost is more sig­ scenario 2. However, both scenarios 2 and 3 have higher profitability
nificant than the utility cost in the cases where natural gas is burnt, than 1, despite having higher capital and operating costs. This higher
while the cost of utilities is more significant when the fuel of choice is profitability is a result of higher net revenue.
biomass. The potential of the base case fast pyrolysis process matching the
The lower operating cost of the “a” sub-scenarios compared to the return rate of scenario 3a, in which ethanol and hydrogen were pro­
“b” sub-scenarios is because natural gas is cheaper and has a higher duced in addition to bio-oil, was assessed, based on the sales of biochar.
heating value than biomass. The difference in operating costs for sub- It was derived that the base case could match this profitability of 13.01
scenarios “a” and “b” for all the scenarios considered is less pro­ % if biochar could be sold at $236/tonne. Hence, at this price point and
nounced for scenario 1 than scenarios 2 and 3. The cost differences above, the base case fast pyrolysis process is preferable to scenario 3a.
between “a” and “b” in all scenarios are about $7.61 M, $15.08 M, and Shabangu et al. [62] reported a breakeven selling price for biochar of
$13.01 M, for scenarios 1, 2, and 3, respectively. This cost difference is $280/tonne for a slow pyrolysis plant co-producing methanol at a dis­
more pronounced in scenarios 2 and 3 because both cases are more count rate of 10 %. Although this was slow pyrolysis, their product
energy-intensive than scenario 1. distribution is near that of fast pyrolysis. It could be expected that at a
similar product distribution and discount rate, the result obtained in this
3.2.3. Rate of return study is reasonably within Shabangu et al.’s reported cost. Dickson et al.
The profitability of each scenario was also studied using rate of re­ [63] also cited biochar breakeven prices of $173-$320/tonne. The bio­
turn as the metric. The rate of return for each scenario is presented in char market price is between $726 and $3,080 in the US market, the
Fig. 4. Generally, compared to the base case, the scenarios studied are European market, and globally [64–66]. These high prices seem
more profitable. This indicates that there is an economic benefit to attractive and suggest that biochar should be sold instead of further
producing valuable products from non-condensable gases and biochar. It processing it. However, the market for biochar is small. Market research
is important to point out the trend related to the fuel type. As shown, the puts the 2018 global demand for biochar at about 350 kt per year [67],
“a” scenarios, which use natural gas as a fuel, are more profitable than which could be easily met by four fast pyrolysis plants of the capacity
the “b” scenarios, in which woodchips are used as fuel. This difference in assumed in this study. The high market price may be due to the lower
profitability between the “a” and “b” scenarios is an indication of the scale of production, and prices should decline as supply gets higher.
lower projected net income (before taxes and depreciation) associated
with the use of woodchips for the same projected revenue. For instance, 3.2.4. Influence of scale on the capital cost per unit of processed biomass
the second year’s projected net incomes (before taxes and depreciation) The effect of capacity on the capital cost per unit of biomass pro­
for scenario 1 with natural gas and woodchips as fuel are $12.5 and $4.6 cessed was assessed in this study. Fig. 5 gives the plot of the capital cost
million, respectively. It was also seen that the difference in the profit­ as a function of plant capacity for scenario 3a. Scenario 3a was chosen
ability of the “a” and “b” scenarios is higher in scenarios 2 and 3. These for brevity as it is the most profitable scenario. The capital cost increases
differences stem from the higher operating cost associated with burning with capacity, as expected. However, a diminishing trend in the increase
woodchips than natural gas, as pointed out in Section 2.2.2.2. was observed at a scale factor of 0.598. This diminishing trend is seen in
Furthermore, when natural gas is burnt, scenario 3 is the most the plot of capital cost per unit of processed biomass vs capacity, as
profitable scenario, with a return of about 13 %. Likewise, when shown in Fig. 6. The capital cost per unit capacity reduces as the capacity
woodchips are burnt, scenario 3 is the most profitable scenario. The increases, showing the presence of economies of scale. However, the
higher profitability of scenario 3 over scenario 2 is because scenario 3 slope of the curve decreases as the capacity is increased, indicating a

Fig. 5. Determination of scale factor for scenario 3a (scenario where ethanol is produced from biochar, hydrogen is produced from NCG, and woodchips is used for
supplementary heat supply).

10
T. Giwa et al. Fuel 332 (2023) 126022

Fig. 6. Influence of capacity on capital cost for scenario 3a (scenario where ethanol is produced from biochar, hydrogen is produced from NCG, and woodchips is
used for supplementary heat supply).

Fig. 7. Sensitivity analysis of scenario 3a (scenario where ethanol is produced from biochar, hydrogen is produced from NCG, and woodchips is used for supple­
mentary heat supply).

11
T. Giwa et al. Fuel 332 (2023) 126022

diminishing return with added capacity. volume inside a single reactor are essential. In the case of the PSA unit,
the high cost is largely from the compression needed for the purification
3.3. Sensitivity analysis and uncertainty analysis process.
The analysis carried out in this study is deterministic in nature. In
Sensitivity analysis was conducted for scenario 3a for brevity and reality, the values of the input parameters would vary and are unpre­
because it was the most profitable among the scenarios studied. The dictable and will result in variation in the output parameter (the rate of
profitability was assessed by varying most of the input parameters by return in this case). To determine the variability of the rate of return due
+/− 30 % of their base values, except the market prices of bio-oil and to the potential variation in the input parameter, an uncertainty analysis
ethanol. This range was chosen as preliminary estimates are reported to was conducted. 10,000 Monte Carlo simulations to derive the uncer­
be in this range [68]. A range of +/− 50 % was used for the market prices tainty in the rate of return of scenario 3a were run. Uncertainty analysis
of bio-oil and ethanol because of the high uncertainty in the market was also conducted for the input parameters identified in the previous
prices. Fig. 7 presents the results of the Morris sensitivity analysis. The section as having the most impact on the rate of return. Fig. 8 is a his­
horizontal axis of the plot gives the mean of the change in the rate of togram plot showing the range of the rate of return (between − 9% and
return as the input parameters change from their minimum to maximum 33 %).
values [59]. The higher a parameter’s mean, the greater its impact on
the rate of return. The standard deviation on the vertical axis indicates 4. Conclusion
the interaction of a parameter with other parameters or a nonlinear
influence on the rate of return [69]. As shown, bio-oil price, capital cost, In this study, the prospect of improving the profitability of bio­
hydrogen price, feedstock cost, and ethanol price are the most important refineries was assessed through an integrated, multi-product approach,
parameters, in that order. The importance of the prices of the products where valuable products, ethanol and hydrogen, were produced from
emphasizes the importance of having a good market price for the the by-products (biochar and non-condensable gases) of a traditional
products, notwithstanding any process improvement. However, the fast pyrolysis biorefinery. Six scenarios, based on the NCG stream and
market for bio-oil is not yet established. Meanwhile, the hydrogen the type of external fuel (natural gas or biomass) used to supplement
market has gained traction recently. Compared to bio-oil and hydrogen, heat, were assessed. A fast pyrolysis biorefinery processing biomass at a
the market for ethanol is well established and can be expected to grow in rate of 2000 dry tonne/day was considered. A data-intensive techno-
future. Focus should be given in improving the yield of ethanol from the economic model was developed to assess the internal rate of return of
fermentation process as ethanol sales will also boost revenue. Further the scenarios relative to the base fast pyrolysis plant.
improvements in biomass harvesting and transportation are needed to It was observed that the production of valuable products like ethanol
reduce the delivered cost of the feedstock. For the capital investment, and hydrogen from the by-products of fast pyrolysis is a profitable
the pyrolysis, fermentation, and PSA units contribute significantly to venture, especially in an uncertain biochar market. This is because if
capital cost. As more pyrolysis projects are built and the learning rate biochar cannot be sold, the six scenarios assessed show a higher rate of
increases, the capital investment in the pyrolysis reactor is expected to return than the base case, which has a rate of return of 7.00 %. When
decrease. The maximum practical volume of a 1000 m3 fermentation natural gas is the external fuel, the internal rate of return is between 10
reactor does not show economies of scale. Increasing ethanol yield from and 13 %. When woodchips are burnt, the internal rate of return is be­
the fermentation process and better reactor design to allow for a high tween 7.46 and 9.80 %. In every scenario, it is most profitable to use

Fig. 8. Uncertainty analysis of scenario 3a (scenario where ethanol is produced from biochar, hydrogen is produced from NCG, and woodchips is used for sup­
plementary heat supply).

12
T. Giwa et al. Fuel 332 (2023) 126022

biochar for ethanol production and the hydrogen-rich non-condensable [6] Patel M, Zhang X, Kumar A. Techno-economic and life cycle assessment on
lignocellulosic biomass thermochemical conversion technologies: A review. Renew
gases stream for hydrogen production. Overall, using natural gas as the
Sustain Energy Rev 2016;53:1486–99.
external fuel is cheaper than burning woodchips because of the higher [7] Oasmaa A, Lehto J, Solantausta Y, Kallio S. Historical review on VTT fast pyrolysis
heating value and lower cost of natural gas. Although scenario 3a, in bio-oil production and upgrading. Energy Fuels 2021;35(7):5683–95.
which both ethanol and hydrogen were produced in addition to bio-oil, [8] Global Pyrolysis Oil Market 2020-2025 - Growing Applications in Biorefineries to
Offer Lucrative Opportunities 2020. Available from: https://www.businesswire.
has the highest return rate (13 %), this rate of return could be matched com/news/home/20200811005742/en/Global-Pyrolysis-Oil-Market-2020-202
by the fast pyrolysis plant if biochar can be sold at $236/tonne. The 5—Growing-Applications-in-Biorefineries-to-Offer-Lucrative-Opportunities—Rese
assessment of the influence of capacity shows that higher capacities are archAndMarkets.com. [Accessed August 15, 2022].
[9] Patel M, Oyedun AO, Kumar A, Gupta R. A Techno-economic assessment of
favoured. However, fluctuating feedstock supply and high trans­ renewable diesel and gasoline production from aspen hardwood. Waste Biomass
portation distances may limit plant capacities. Sensitivity analysis con­ Valorization 2019;10(10):2745–60.
ducted on scenario 3a (the most profitable scenario) showed that the [10] Pootakham T, Kumar A. Bio-oil transport by pipeline: A techno-economic
assessment. Bioresour Technol 2010;101(18):7137–43.
price of the products is an important parameter affecting the profit­ [11] Hu W, Dang Q, Rover M, Brown RC, Wright MM. Comparative techno-economic
ability, in addition to feedstock prices and capital investment. The re­ analysis of advanced biofuels, biochemicals, and hydrocarbon chemicals via the
sults of this study provide helpful insights on the potential of improving fast pyrolysis platform. Biofuels 2016;7(1):57–67.
[12] Brassard P, Godbout S, Lévesque V, Palacios JH, Raghavan V, Ahmed A, et al. 4 -
the economics of the fast pyrolysis process by producing value-added Biochar for soil amendment. In: Jeguirim M, Limousy L, editors. Char and carbon
products from the by-products. Overall, from these analyses, there is materials derived from biomass. Elsevier; 2019. p. 109–46.
an indication that an integrated, multi-product concept, has the poten­ [13] Tan X-f, Liu S-b, Liu Y-g, Gu Y-l, Zeng G-m, Hu X-j, et al. Biochar as potential
sustainable precursors for activated carbon production: Multiple applications in
tial to improve the economic outlook of biorefining processes.
environmental protection and energy storage. Bioresour Technol 2017;227:
359–72.
CRediT authorship contribution statement [14] Hidalgo P, Navia R, Hunter R, Coronado G, Gonzalez M. Synthesis of carbon
nanotubes using biochar as precursor material under microwave irradiation.
J Environ Manage 2019;244:83–91.
Temitayo Giwa: Conceptualization, Methodology, Software, Data [15] Akbari M, Oyedun AO, Kumar A. Techno-economic assessment of wet and dry
curation, Validation, Writing – original draft. Maryam Akbari: Meth­ torrefaction of biomass feedstock. Energy 2020;207:118287.
[16] Gradual Increase of Bio-oil Production Expected as Commercialization of Fast
odology, Validation, Writing – review & editing. Amit Kumar: Pyrolysis Gains Momentum; 2021. Available from: https://stratasadvisors.com/Ins
Conceptualization, Methodology, Validation, Writing – review & edit­ ights/2021/11242021-Bio-oil-Production. [Accessed August 19, 2022].
ing, Supervision. [17] Cho R. Why We Need Green Hydrogen; 2021. Available from: https://news.clim
ate.columbia.edu/2021/01/07/need-green-hydrogen/. [Accessed July 28, 2022].
[18] Industrial Alcohol Market Size, Share & Trends Analysis Report Available from: htt
ps://www.grandviewresearch.com/industry-analysis/industrial-alcohol-market.
Declaration of Competing Interest [Accessed August 10, 2022].
[19] Ethanol Market Size, Share & Trends Analysis Report; Available from: https://
www.grandviewresearch.com/industry-analysis/ethanol-market. [Accessed
The authors declare that they have no known competing financial August 14, 2022].
interests or personal relationships that could have appeared to influence [20] Zabed H, Sahu JN, Boyce AN, Faruq G. Fuel ethanol production from
lignocellulosic biomass: An overview on feedstocks and technological approaches.
the work reported in this paper.
Renew Sustain Energy Rev 2016;66:751–74.
[21] Roy P, Dutta A, Deen B. Greenhouse gas emissions and production cost of ethanol
Data availability produced from biosyngas fermentation process. Bioresour Technol 2015;192:
185–91.
[22] Stoll IK, Boukis N, Sauer J. Syngas Fermentation to Alcohols: Reactor Technology
Data will be made available on request. and Application Perspective. Chem Ing Tech 2020;92(1–2):125–36.
[23] Pardo-Planas O, Atiyeh HK, Phillips JR, Aichele CP, Mohammad S. Process
simulation of ethanol production from biomass gasification and syngas
Acknowledgements fermentation. Bioresour Technol 2017;245:925–32.
[24] de Medeiros EM, Posada JA, Noorman H, Osseweijer P, Filho RM. Hydrous
As a part of the University of Alberta’s Future Energy Systems bioethanol production from sugarcane bagasse via energy self-sufficient
gasification-fermentation hybrid route: Simulation and financial analysis. J Cleaner
research initiative, this research was made possible in part thanks to Prod 2017;168:1625–35.
funding from the Canada First Research Excellence Fund. The authors [25] Modisha PM, Ouma CNM, Garidzirai R, Wasserscheid P, Bessarabov D. The
are grateful to the NSERC/Cenovus/Alberta Innovates Associate Indus­ prospect of hydrogen storage using liquid organic hydrogen carriers. Energy Fuels
2019;33(4):2778–96.
trial Research Chair Program in Energy and Environmental Systems
[26] Benalcázar EA, Deynoot BG, Noorman H, Osseweijer P, Posada JA. Production of
Engineering and the Cenovus Energy Endowed Chair in Environmental bulk chemicals from lignocellulosic biomass via thermochemical conversion and
Engineering at the University of Alberta for financial support for this syngas fermentation: a comparative techno-economic and environmental
assessment of different site-specific supply chain configurations. Biofuels, Bioprod
research. Astrid Blodgett is thanked for editing this paper.
Biorefin 2017;11(5):861–86.
[27] Statistical Data: Forest Inventory Canada; Available from: https://cfs.nrcan.gc.ca/
Appendix A. Supplementary data statsprofile/inventory/canada. [Accessed July 23, 2021].
[28] Patel M, Oyedun AO, Kumar A, Gupta R. What is the production cost of renewable
diesel from woody biomass and agricultural residue based on experimentation? A
Supplementary data to this article can be found online at https://doi. comparative assessment. Fuel Process Technol 2019;191:79–92.
org/10.1016/j.fuel.2022.126022. [29] Rogers JG, Brammer JG. Estimation of the production cost of fast pyrolysis bio-oil.
Biomass Bioenergy 2012;36:208–17.
[30] Jafari NL. A study of cellulosic biomass size reduction. University of British
References Columbia; 2016.
[31] Chaudhari ST, Bej SK, Bakhshi NN, Dalai AK. Steam gasification of biomass-derived
char for the production of carbon monoxide-rich synthesis gas. Energy Fuels 2001;
[1] IEA Data and statistic: Total energy supply (TES) by source, World 1990-2018;
15(3):736–42.
Available from: https://www.iea.org/data-and-statistics?country=WORLD&fue
[32] Dutta A, Talmadge M, Hensley J, Worley M, Dudgeon D, Barton D, et al. Process
l=Energy%20supply&indicator=TPESbySource. [Accessed February, 17, 2021 ].
design and economics for conversion of lignocellulosic biomass to ethanol:
[2] Hannah R. Energy Production and Consumption; Available from: https://ourworldi
thermochemical pathway by indirect gasification and mixed alcohol synthesis.
ndata.org/energy-production-consumption. [Accessed February 17, 2021].
Golden, CO (United States): National Renewable Energy Lab (NREL); 2011.
[3] Morales M, Quintero J, Conejeros R, Aroca G. Life cycle assessment of
[33] Burhenne L, Damiani M, Aicher T. Effect of feedstock water content and pyrolysis
lignocellulosic bioethanol: Environmental impacts and energy balance. Renew
temperature on the structure and reactivity of spruce wood char produced in fixed
Sustain Energy Rev 2015;42:1349–61.
bed pyrolysis. Fuel 2013;107:836–47.
[4] Dassanayake GDM, Kumar A. Techno-economic assessment of triticale straw for
[34] Phillips JR, Huhnke RL, Atiyeh HK. Syngas fermentation: a microbial conversion
power generation. Appl Energy 2012;98:236–45.
process of gaseous substrates to various products. 2017;3(2):28.
[5] Zhang YHP. Reviving the carbohydrate economy via multi-product lignocellulose
biorefineries. J Ind Microbiol Biotechnol 2008;35(5):367–75.

13
T. Giwa et al. Fuel 332 (2023) 126022

[35] Liberato V, Benevenuti C, Coelho F, Botelho A, Amaral P, Pereira N, et al. [53] Zhu Y, Jones SB. Techno-economic analysis for the thermochemical conversion of
Clostridium sp. as Bio-catalyst for fuels and chemicals production in a biorefinery lignocellulosic biomass to ethanol via acetic acid synthesis. Richland, WA (United
context. Catalysts 2019;9(11). States): Pacific Northwest National Lab (PNNL); 2009.
[36] Drake HL, Gößner AS, Daniel SL. Old acetogens, new light. Ann N Y Acad Sci 2008; [54] Canada salary calculator: CanadaVisa; 2016. Available from: http://www.canadavi
1125(1):100–28. sa.com/canada-salary-wizard.html. [Accessed November 20, 2020].
[37] Gaddy JL, Arora DK, Ko C-W, Phillips JR, Basu R, Wikstrom CV, et al. Methods for [55] Akbari M, Oyedun AO, Kumar A. Comparative energy and techno-economic
increasing the production of ethanol from microbial fermentation. Bioengineering analyses of two different configurations for hydrothermal carbonization of yard
Resources, Inc.; 2007. waste. Bioresource Technol Reports 2019;7:100210.
[38] Peng Y, Lu X, Liu B, Zhu J. Separation of azeotropic mixtures (ethanol and water) [56] Humbird D, Davis R, McMillan JD. Aeration costs in stirred-tank and bubble
enhanced by deep eutectic solvents. Fluid Phase Equilib 2017;448:128–34. column bioreactors. Biochem Eng J 2017;127:161–6.
[39] Bernardo G, Araújo T, da Silva LT, Sousa J, Mendes A. Recent advances in [57] Kreutz T, Williams R, Consonni S, Chiesa P. Co-production of hydrogen, electricity
membrane technologies for hydrogen purification. Int J Hydrogen Energy 2020;45 and CO2 from coal with commercially ready technology. Part B: Economic analysis.
(12):7313–38. Int J Hydrogen Energy 2005;30(7):769–84.
[40] Nordio M, Wassie SA, Van Sint AM, Pacheco Tanaka DA, Viviente Sole JL, [58] Larson ED, Jin H, Celik FE. Large-scale gasification-based coproduction of fuels and
Gallucci F. Techno-economic evaluation on a hybrid technology for low hydrogen electricity from switchgrass. 2009;3(2):174-94.
concentration separation and purification from natural gas grid. Int J Hydrogen [59] Di Lullo G, Gemechu E, Oni AO, Kumar A. Extending sensitivity analysis using
Energy 2020. regression to effectively disseminate life cycle assessment results. Int J Life Cycle
[41] Fahim MA, Alsahhaf TA, Elkilani A. Chapter 11 - Hydrogen production. In: Assess 2020;25(2):222–39.
Fahim MA, Alsahhaf TA, Elkilani A, editors. Fundamentals of petroleum refining. [60] Piccolo C, Bezzo F. A techno-economic comparison between two technologies for
Amsterdam: Elsevier; 2010. p. 285–302. bioethanol production from lignocellulose. Biomass Bioenergy 2009;33(3):478–91.
[42] Banu A-M, Friedrich D, Brandani S, Düren T. A multiscale study of MOFs as [61] Michailos S, Parker D, Webb C. Design, Sustainability analysis and multiobjective
sdsorbents in H2 PSA purification. Ind Eng Chem Res 2013;52(29):9946–57. optimisation of ethanol production via syngas fermentation. Waste Biomass
[43] Speight JG. Chapter 15 - Hydrogen production. In: Speight JG, editor. Heavy oil Valorization 2019;10(4):865–76.
recovery and upgrading. Gulf Professional Publishing; 2019. p. 657–97. [62] Shabangu S, Woolf D, Fisher EM, Angenent LT, Lehmann J. Techno-economic
[44] Allevi C, Collodi G. 12 - Hydrogen production in IGCC systems. In: Wang T, assessment of biomass slow pyrolysis into different biochar and methanol concepts.
Stiegel G, editors. Integrated Gasification Combined Cycle (IGCC) Technologies. Fuel 2014;117:742–8.
Woodhead Publishing; 2017. p. 419–43. [63] Dickinson D, Balduccio L, Buysse J, Ronsse F, van Huylenbroeck G, Prins W. Cost-
[45] Akbari M, Oyedun AO, Kumar A. Ammonia production from black liquor benefit analysis of using biochar to improve cereals agriculture. GCB Bioenergy
gasification and co-gasification with pulp and waste sludges: A techno-economic 2015;7(4):850–64.
assessment. Energy 2018;151:133–43. [64] Campbell RM, Anderson NM, Daugaard DE, Naughton HT. Financial viability of
[46] Recharge: Green hydrogen’ on sale in open market at 80% higher price than grey biofuel and biochar production from forest biomass in the face of market price
H2; 2020. [Accessed October 21, 2020 2020]. volatility and uncertainty. Appl Energy 2018;230:330–43.
[47] Argusmedia: US VGO touches six-month high; 2020. Available from: https://www. [65] Shemfe MB, Gu S, Ranganathan P. Techno-economic performance analysis of
argusmedia.com/en/news/2138029-us-vgo-touches-sixmonth-high. [Accessed biofuel production and miniature electric power generation from biomass fast
December 13, 2020]. pyrolysis and bio-oil upgrading. Fuel 2015;143:361–72.
[48] Ethanol Commodity: Markets Insider Available from: https://markets.busine [66] Draper K, Groot H, Miles T, Twer M. Survey and analysis of the US biochar industry
ssinsider.com/commodities/ethanol-price. [Accessed 21 October, 2020]. preliminary report draft; August 16, 2018, WERC Project MN17-DG-230.
[49] Akbari M, Oyedun AO, Jain S, Kumar A. Options for the conversion of pulp and [67] Biochar market size, share & trends analysis report by technology (gasification,
paper mill by-products in Western Canada. Sustainable Energy Technol Assess pyrolysis), by application (agriculture (farming, livestock)), by region, and
2018;26:83–92. segment forecasts, 2019 - 2025; Available from: https://www.grandviewresearch.
[50] Multi-residential and commercial rates: EPCOR; 2019. Available from: https com/industry-analysis/biochar-market#:~:text=The%20global%20biochar%
://www.epcor.com/products-services/water/rates-terms-conditions/Pages/co 20market%20size,be%20the%20key%20growth%20driver. [Accessed February
mmercial-multi-residential-rates.aspx. [Accessed November 1, 2019]. 19, 2021].
[51] Canada Energy Regulator. Canada’s Energy Future Data Appendices; 2017. [68] Towler G, Sinnott R. Chapter 7 - Capital cost estimating. In: Towler G, Sinnott R,
Available from: https://apps.cer-rec.gc.ca/ftrppndc/dflt.aspx?GoCTemplateCultur editors. Chemical engineering design. Second Edition. Boston: Butterworth-
e=en-CA. [Accessed January 1, 2020]. Heinemann; 2013. p. 307–54.
[52] Spath P, Aden A, Eggeman T, Ringer M, Wallace B, Jechura J. Biomass to hydrogen [69] Franczyk A. Using the Morris sensitivity analysis method to assess the importance
production detailed design and economics utilizing the Battelle Columbus of input variables on time-reversal imaging of seismic sources. Acta Geophys 2019;
Laboratory indirectly-heated gasifier. Golden, CO (US): National Renewable 67(6):1525–33.
Energy Lab; 2005.

14