J Ind Microbiol Biotechnol (2011) 38:955–966

DOI 10.1007/s10295-010-0867-6

ORIGINAL PAPER

Simulation of integrated first and second generation bioethanol

production from sugarcane: comparison between different
biomass pretreatment methods
Marina O. S. Dias • Marcelo Pereira da Cunha •

Rubens Maciel Filho • Antonio Bonomi •

Charles D. F. Jesus • Carlos E. V. Rossell

Downloaded from https://academic.oup.com/jimb/article/38/8/955/5994411 by guest on 30 March 2022

Received: 1 June 2010 / Accepted: 26 August 2010 / Published online: 14 September 2010
Ó Society for Industrial Microbiology 2010

Abstract Sugarcane bagasse is used as a fuel in con- pentose fermentation to ethanol) in order for the integrated
ventional bioethanol production, providing heat and power process to be more economically competitive.
for the plant; therefore, the amount of surplus bagasse
available for use as raw material for second generation Keywords Simulation  Sugarcane  Sugarcane bagasse 
bioethanol production is related to the energy consumption Ethanol  Pretreatment  Enzymatic hydrolysis
of the bioethanol production process. Pentoses and lignin,
byproducts of the second generation bioethanol production
process, may be used as fuels, increasing the amount of Introduction
surplus bagasse. In this work, simulations of the integrated
bioethanol production process from sugarcane, surplus Bioethanol has been produced from sugarcane in Brazil on
bagasse and trash were carried out. Selected pre-treatment a large scale for more than 30 years [14]. Until 2005,
methods followed, or not, by a delignification step were Brazil was the largest ethanol producer in the world, but
evaluated. The amount of lignocellulosic materials avail- international interest in ethanol as a fuel aroused and
able for hydrolysis in each configuration was calculated motivated an increase of ethanol production from corn in
assuming that 50% of sugarcane trash is recovered from the the USA. Nevertheless, sugarcane remains the most effi-
field. An economic risk analysis was carried out; the best cient raw material for bioethanol production, with lower
results for the integrated first and second generation etha- consumption of fossil energy for ethanol production, pro-
nol production process were obtained for steam explosion ducing an average of 9.3 units of renewable energy for one
pretreatment, high solids loading for hydrolysis and unit of fossil energy consumed [22].
24–48 h hydrolysis. The second generation ethanol pro- In the search for the expansion of bioethanol production
duction process must be improved (e.g., decreasing without compromising food security and assuring fuel
required investment, improving yields and developing supplies, the use of lignocellulosic materials such as sur-
plus bagasse, trash and other agricultural residues has been
encouraged; it is estimated that the use of these raw
This article is based on a presentation at the 32nd Symposium on
Biotechnology for Fuels and Chemicals.
materials for bioethanol production will improve process
efficiency as well as reduce environmental impacts [23,
M. O. S. Dias (&)  M. P. da Cunha  A. Bonomi  37]. Bioethanol production from lignocellulosic materials
C. D. F. Jesus  C. E. V. Rossell (second generation (2G) bioethanol), including pretreat-
Laboratório Nacional de Ciência e Tecnologia do Bioetanol
ment processes and enzymes technology for cellulose
(CTBE), Caixa Postal 6170, 13083-970 Campinas,
São Paulo, Brazil saccharification, has been investigated with increasing
e-mail: marina.dias@bioetanol.org.br; diasmos@feq.unicamp.br interest for the past few years [36], due to the growing
concerns about climate change, increased energy demand
M. O. S. Dias  R. Maciel Filho  C. E. V. Rossell
and the forecast depletion of fossil resources [10, 32].
Faculdade de Engenharia Quı́mica, Universidade Estadual
de Campinas (UNICAMP), Caixa Postal 6066, One of the major lignocellulosic materials that can be
CEP 13083-970 Campinas, São Paulo, Brazil employed for bioethanol production is sugarcane bagasse,

123
956 J Ind Microbiol Biotechnol (2011) 38:955–966

the fibrous lignocellulosic residue generated in large sugarcane trash is recovered from the field [7, 39] and used
amounts (approximately 140 kg/ton of sugarcane, dry either as fuel in the cogeneration system or as raw material
basis) during sugarcane processing for sugar and ethanol for bioethanol production, if the thermal requirements of
production [20]. Where bioethanol is produced from sug- the plant are low enough, resulting in surplus trash. Yields,
arcane (first generation (1G)), such as in Brazil, sugarcane process and conversion parameters were obtained in the
bagasse is already available at plant sites, and production literature and from the industry for the first generation
of second generation bioethanol can share part of the process, and bioethanol production was evaluated for each
infrastructure where first generation bioethanol is pro- process configuration.
duced, such as juice concentration, fermentation and dis- This work provides a basis for economic evaluations to
tillation [7], and the utilities sector (cogeneration and water determine the feasibility of integrated second generation
cooling systems). Another important residue that may be bioethanol production.
employed for bioethanol production in the sugarcane
industry is sugarcane trash, which includes sugarcane

Downloaded from https://academic.oup.com/jimb/article/38/8/955/5994411 by guest on 30 March 2022

leaves and tops, usually burnt or left in the field; trash is Process description
produced at a rate of approximately 140 kg/ton of sugar-
cane (dry basis). With the elimination of burning practices, Sugarcane is comprised of sugars (mainly sucrose and
large amounts of trash will be available for use as an small fractions of the reducing sugars fructose and glu-
energy source [34]. cose), water, fibers and impurities. Sugarcane composition
Since sugarcane bagasse is used as a fuel in conven- varies greatly among varieties, climate, soil and age. It is
tional bioethanol production, providing heat and power for expected that over the next few years sugarcane quality
the plant [11], the amount of surplus bagasse available for will be enhanced, with an increase in its sugars and fiber
use as raw material for second generation bioethanol pro- content [22]; thus, an improvement in these parameters is
duction is directly related to the energy consumption of considered in this work, in comparison with present sug-
bioethanol production processes (both first and second arcane quality in Brazil.
generation). Pentose and lignin, byproducts of the second In this work an autonomous distillery is considered, that
generation bioethanol production, may be used as fuels is, a plant in which all the processed sugarcane is diverted
[21] to increase the amount of surplus bagasse, along with for ethanol production. A typical large scale plant (crushing
improved technologies for cogeneration and optimization around 12,000 tons of sugarcane—TC per day) is simu-
of the conventional bioethanol production process [11]. lated. The main parameters of the plant are displayed in
In order to be used as raw materials for bioethanol Table 1.
production, lignocellulosic materials such as sugarcane
bagasse and trash must undergo a pretreatment process First generation bioethanol production from sugarcane
through which the hemicellulose is removed and the cel-
lulose is made more accessible to enzymatic attack during The conventional bioethanol production process in an
hydrolysis [28, 31]. Different pretreatment processes have autonomous distillery follows these main steps: sugarcane
been investigated including physical (comminution), reception and cleaning, extraction of sugars, juice treat-
chemical (alkali, acid and solvents attack), physico-chem- ment and concentration, fermentation, distillation and
ical (steam explosion) and biological (white-rot fungi dehydration.
attack) approaches [35], but none have yet upgraded sec- Upon reception in the factory, sugarcane must be
ond generation to levels of production competitive with cleaned to remove part of the dirt carried along from the
conventional first generation ethanol production.
In order to point toward the most adequate configuration Table 1 Plant design: main parameters
of the second generation ethanol production process, in a Parameter Value Unit
process integrated with conventional first generation from
sugarcane, simulations of the integrated bioethanol pro- Sugarcane crushing rate 500 TC/h
duction process from sugarcane, surplus bagasse and trash Days of operation 180 days/year
were carried out using the SuperPro Designer process Fiber content 14 wt%
simulator [17]. Selected pre-treatment methods (steam Sugars content 15 wt%
explosion, hydrogen peroxide and organosolv) followed, or Amount of sugarcane trash 140 kg/TC
not, by an alkaline delignification step were evaluated, produced in the field
along with the subsequent enzymatic hydrolysis step. The Fraction of sugarcane trash 50 %
recovered from the field
amount of lignocellulosic material available for hydrolysis
on each configuration was calculated assuming that 50% of TC tons of sugarcane

123
J Ind Microbiol Biotechnol (2011) 38:955–966 957

field. Sugarcane cleaning is usually carried out using wash Table 2 Unit operation design: main parameters
water, which is recycled to the cleaning process after Parameter Value Unit
removal of dirt and other impurities.
Extraction of sugars is done using mills, where sugarcane Efficiency of dirt removal on sugarcane cleaning 90 %
juice and bagasse are separated. Water at a rate of 28 wt% of Sugar losses on sugarcane cleaning 0.8 kg/TC
sugarcane flow is used to improve sugars recovery. Sugar- Sugarcane wash water 2.2 m3/TC
cane juice contains water, sucrose and reducing sugars, Efficiency of sugars extraction on the mills 97.5 %
along with impurities such as minerals, salts, organic acids, Sugarcane bagasse moisture content 50 wt%
dirt and fiber particles which must be removed prior to Recovery of sugars on juice treatment 99.5 %
fermentation. First, screens are used to remove sand and fine Fermentation yield 92 %
particles of bagasse [5]. Juice then receives a chemical Ethanol recovery on distillation and dehydration 99.7 %
treatment to remove other impurities. In the process con- Pressure of steam produced in the boilers 90 bar
sidered in this work, juice undergoes heating from 30 to

Downloaded from https://academic.oup.com/jimb/article/38/8/955/5994411 by guest on 30 March 2022

TC tons of sugarcane
70°C; addition of lime takes place afterwards along with a
second heating, up to 105°C. Hot juice is flashed to remove
dissolved air and, after addition of a flocculant polymer, polymers: cellulose, hemicelluloses (mainly xylan) and
impurities are removed in a settler, where mud and clarified lignin [1]. While the hemicellulosic fraction is converted
juice are obtained. A filter is used to recover some of the into monomeric sugars (mainly pentoses) under mild pro-
sugars carried along with the mud, and the filtrate is recy- cess conditions, hydrolysis of cellulose into glucose, which
cled to the process prior to the second heating operation. is easily fermented to ethanol, is considerably harder to
Clarified juice contains around 15 wt% solids; a fraction accomplish [30].
of the clarified juice is concentrated in a 5-stage multiple Due to the close association of the three components
effect evaporators (MEE) up to 65 wt% solids. The con- (cellulose, hemicellulose and lignin) in the structure of the
centrated juice is mixed with the remaining clarified frac- biomass, an intense pretreatment process is required to
tion to produce a final juice containing 22.5 wt% solids separate lignin and hemicellulose from the cellulose by
which is cooled and fed to the fermentors. promoting their solubilization or degradation, reducing
A fed-batch fermentation process with cell recycling cellulose crystallinity and increasing porosity of the lig-
(Melle-Boinot process) is used [33]. In this process, yeast nocellulosic material, thus improving accessibility of the
(recovered from a previous fermentation batch) is fed to the remaining cellulose to enzymatic attack [19, 26, 31].
fermentor prior to the juice; the mixture remains in the The process of production of bioethanol from lignocel-
reactor for a few hours, and the sugars (sucrose, glucose lulosic materials consists of the following operations:
and fructose) are converted into ethanol, carbon dioxide biomass pretreatment, hydrolysis, concentration, fermen-
and by-products (alcohols, organic acids, etc.). Since fer- tation, separation, purification and effluent treatment and
mentation reactions are exothermic, fermentation vats are disposal [36].
cooled in order to attain high ethanol concentration in the In Brazil, sugarcane bagasse is an important raw mate-
fermented liquor (also called wine) [9]. The wine produced rial to be considered for ethanol production, via hydrolysis,
in fermentation is centrifuged to remove yeast cells, which due to its low costs and the possibility of sharing infra-
undergo a chemical treatment using water and sulfuric acid structure in existing mills, thus reducing required invest-
to reduce bacterial contamination. Yeast cells are used in ments [39]. In addition, another lignocellulosic residue
the next batch, while centrifuged wine is purified in dis- obtained from sugarcane—sugarcane trash—may be
tillation and rectification columns, producing hydrous recovered from the fields and used as raw material for
(around 93 wt%) ethanol. Anhydrous ethanol (99.3 wt%) is second generation bioethanol production. However, a
produced in an adsorption process using molecular sieves. fraction of the trash produced must be left in the field in
The main parameters of the unit operations of the bio- order to provide weed and disease control as well as soil
ethanol production process that constitutes an autonomous conservation and fertilizer recycling. The exact amount of
distillery are represented in Table 2. The efficiencies and trash that must be left in the field depends on local con-
other parameters displayed in Table 2 represent equipment ditions such as variety, climate, pests and others [15], but
commercially available in Brazil. several authors consider that the use of 50% of the trash in
the mills is a feasible figure in the years to come [7, 39].
Second generation bioethanol production processes In this work, different pretreatment methods were con-
sidered and compared: steam explosion, hydrogen peroxide
Second generation biofuels are produced from lignocellu- and organosolv, followed, or not, by an alkaline delignifi-
losic biomass, which is basically composed of three cation step.

123
958 J Ind Microbiol Biotechnol (2011) 38:955–966

SO2 catalyzed steam explosion of lignin from the pretreated material improves cellulose
accessibility to enzymes in the hydrolysis step.
Experimental results of the steam pretreatment process of
sugarcane bagasse and the subsequent enzymatic hydro- Cogeneration: co-production of heat and power
lysis were obtained by Carrasco et al. [3]. The authors
performed a series of experiments in which 300 g of Fibrous sugarcane residues (bagasse produced in the mills
bagasse (dry weight) was steam pretreated in a 10 L and trash recovered from the fields) are used as fuels for
steam-pretreatment reactor, using SO2 as a catalyst, fol- production of steam and electric energy in the cogeneration
lowed by enzymatic hydrolysis (1 L reactors) and system, supplying the entire thermal and electrical
fermentation. requirements of the ethanol production process.
The highest values for soluble xylose recovery were In the production of first generation bioethanol, all the
obtained for SO2 catalyzed steam pretreatment at 190°C for bagasse and trash available are used as fuel in boilers,
5 min, which also provided the highest overall sugar yields producing high pressure steam. Steam is fed to extraction-

Downloaded from https://academic.oup.com/jimb/article/38/8/955/5994411 by guest on 30 March 2022

(including monosaccharides produced upon pretreatment condensation steam turbines, in which low pressure steam
and hydrolysis) after enzymatic hydrolysis [3]. is produced to supply the thermal requirements of the
process as well as electrical energy, while surplus steam is
Alkaline hydrogen peroxide condensed. When second generation bioethanol is pro-
duced, surplus bagasse and trash are used as raw materials
Alkaline processes are expected to cause less sugar deg- in the pretreatment and hydrolysis processes, and in this
radation than acid processes; since hydrogen peroxide case the process determines the amount of steam that is
degrades into water and oxygen, it does not contaminate produced in the boilers; thus, back-pressure steam turbines
biomass. Another advantage of this process is the fact that (no condensation) are employed [11]. A simplified scheme
it takes place under mild process conditions (pressure, of the cogeneration systems for first and second generation
temperature and no acid catalysts) [28]. In this work, we bioethanol production is illustrated in Fig. 1.
used results obtained by Garcia [13] for the enzymatic When second generation bioethanol production is inte-
hydrolysis of hydrogen peroxide pretreated bagasse at a grated with the first generation, the exact amount of lig-
solids content of 5%. The optimum pretreatment condi- nocellulosic material available for hydrolysis is determined
tions were defined as 1 h of reaction, 25°C, 11% H2O2, according to the thermal requirements of the entire process,
pH = 11.5 and 8% bagasse (wt/v). The subsequent since a portion of the material available must be burnt in
enzymatic hydrolysis step was carried out for up to 72 h the boilers to supply the energy requirements of the inte-
[13]. grated plant. Thus, lower energy demand of the production
process leads to higher amounts of bagasse and trash
Organosolv pretreatment available for hydrolysis. One of the tools that can be
employed to reduce the thermal energy requirements of the
Sugarcane bagasse was submitted to ethanol organosolv process is the Pinch Analysis [7], through which thermal
pretreatment using a 50 L pilot scale reactor [24], in which integration of the process streams is evaluated. Different
an aqueous ethanol solution (50% v/v) and a solid to liquid authors applied this method to analyze the thermal inte-
ratio of 1:5 were employed, at a constant temperature of gration of the bioethanol production process [7, 8, 11, 12,
175°C. Different concentrations of catalysts were evalu- 16, 27], promoting a significant reduction in steam con-
ated, with varying pretreatment times. The subsequent sumption of the process.
enzymatic hydrolysis step was performed on the washed
water-insoluble residue of pretreated bagasse at 5% (w/v)
substrate content on a laboratory scale (25 mL) for up to
24 h. The best pretreatment results in terms of overall
glucose yield (20.9 g glucose/100 g bagasse) were
obtained using 1.25% H2SO4 as catalyst for 60 min [24].

Alkaline delignification

Alkali pretreatment decreases both lignin and hemicellu-

lose content of the biomass; combination of dilute NaOH
Fig. 1 Simplified scheme of the cogeneration system for the first
pretreatment and other treatments seems more efficient (a) and second (b) generation bioethanol production. I: boiler; II:
than the exclusive use of alkali pretreatment [1]. Removal steam turbines; III: process; IV: condenser

123
J Ind Microbiol Biotechnol (2011) 38:955–966 959

Process simulation The parameters adopted for the simulation of the cogene-
ration system are listed in Table 3.
Simulations were carried out using the SuperPro Designer Simulation was performed considering no thermal
process simulator from Intelligen, Inc. An autonomous integration of process streams. However, a reduction of
distillery with modern commercial technologies was con- 30.8% on the 2.5 bar steam consumption calculated by the
sidered for the first generation bioethanol production. A process simulator was taken into account; this reduction
simplified scheme of the simulation of the first generation was obtained after applying Pinch Analysis on the simu-
ethanol production is displayed in Fig. 2. lation of an autonomous distillery by Dias et al. [8].
In order to represent the process more accurately, sev- Simulation of the second generation production process
eral components that are part of sugarcane composition was carried out considering conversion parameters calcu-
(cellulose, hemicellulose, lignin, aconitic acid, which rep- lated from the results published in Carrasco et al. [3] for the
resents the acids present in sugarcane structure, and steam explosion (SE), Garcia [13] for the hydrogen per-
potassium oxide, which represents all the minerals), oxide (H2O2) and Mesa et al. [24] for the organosolv

Downloaded from https://academic.oup.com/jimb/article/38/8/955/5994411 by guest on 30 March 2022

material dragged along with sugarcane (dirt), by-products pretreatment (OS). The main parameters for the 2G pro-
(such as aconitate, formed during juice treatment, and cesses analyzed are presented in Table 4.
pentoses produced during pretreatment and hydrolysis The values presented in Table 4 were obtained for the
reactions in the simulation of the second generation pro- hydrolysis of sugarcane bagasse at a solid load of 5%; these
cess) and input material (flocculant polymer, used on juice figures were used in the simulations. An advanced hydro-
treatment in the settler) were inserted into the simulator lysis scenario was also considered for each studied pre-
database. treatment process, in which enzymatic hydrolysis is carried
First, a mass balance of the first generation process was out at a solids load of 15%, assuming that the same
carried out using an electronic spreadsheet, using data hydrolysis yields illustrated in Table 4 are achieved. In
collected in the industry (Usina da Pedra, a mill producing addition, alkaline delignification of pretreated bagasse for
sugar and ethanol located in the state of São Paulo) and both steam explosion and hydrogen peroxide pretreatments
from the literature; the main parameters are the ones was considered; in this case, a hypothetical increase of 20%
reported in Table 2. Simulation of the production process on the hydrolysis yields displayed in Table 4 was assumed.
was carried out using SuperPro Designer; the current ver- The integrated 1G and 2G production process considers
sion of this software (v. 7.5) does not allow the simulation fermentation of mixed sugarcane juice and hydrolyzed
of the cogeneration system, which was calculated using a glucose liquor; a block-flow diagram of the integrated
spreadsheet and integrated with the SuperPro simulation. process is shown in Fig. 3.

Fig. 2 Simplified scheme of the

simulation of the first generation
ethanol production from
sugarcane in an autonomous
distillery

123
960 J Ind Microbiol Biotechnol (2011) 38:955–966

Table 3 Cogeneration design:

Parameter Value Unit
main parameters
Boiler pressure 90 bar
Boiler efficiency—LHV basis 86 %
Live steam temperature 520 °C
Isentropic efficiency—first stage of steam turbine 72 %
First extraction—steam pressure 22 bar
Isentropic efficiency—second stage of steam turbine 81 %
Generator efficiency 98 %
Isentropic efficiency—condensing turbine 70 %
Condensation pressure 0.11 bar
LHV sugarcane bagasse—50% moisture 7565 kJ/kg
Electricity demand—1G production process 30 kWh/TC

Downloaded from https://academic.oup.com/jimb/article/38/8/955/5994411 by guest on 30 March 2022

Electricity demand—2G production process 24 kWh/ta
LHV sugarcane trash—15% moisture 12960 kJ/kg
LHV biogas 21320 kJ/Nm3
TC tons of sugarcane Process steam pressure 2.5 bar
a Molecular sieves—steam pressure 6 bar
Electricity demand as a
function of the lignocellulosic Steam explosion pretreatment—steam pressure 12.5 bar
material processed in the second Organosolv pretreatment—steam pressure 10 bar
generation production process

Table 4 Second generation pretreatment process design: steam in which the higher heating value (HHV, kJ/g) is calculated
explosion (SE), hydrogen peroxide (H2O2) and organosolv (OS): based on the lignin content of the lignocellulosic material
main parameters
(L, wt%). The value calculated using this correlation pre-
Parameter SE H2O2 OS sented a difference from the experimental value of less than
2% for materials such as tobacco leaf, corncob, corn straw
Conversion of xylan during pretreatment (%) 57 67 57
and wheat straw, among others [6]. Lower heating values
Conversion of cellulose during pretreatment (%) 5 0.7 0
(LHV) for unreacted cellulignin (moisture content around
Conversion of cellulose—24 h hydrolysis (%) 50 61 38
50% and varying composition according to the pretreat-
Conversion of cellulose—48 h hydrolysis (%) 70 70 NA
ment and hydrolysis yields) around 10,000 kJ/kg were
Conversion of cellulose—72 h hydrolysis (%) 72 77 NA
obtained.
Enzymatic load—cellulase (FPU/g dry biomass) 15 10 15
Enzymatic load—b-glucosidase (IU/g dry biomass) 24 20 15
NA data not available Economic assumptions and parameters

A preliminary economic risk analysis was carried out,

The 2G bioethanol production in this work considered considering the basic parameters displayed in Table 5
exclusively the fermentation of the glucose obtained from and the most relevant costs and prices presented in
cellulose hydrolysis; the pentose fraction, obtained from Table 6.
the hemicellulose, is biodigested, producing biogas which Investment data for the first generation biorefinery was
is used as a fuel. Cellulignin, another by-product of the 2G calculated based on data provided by Dedini (one of the
production, which comprises the remaining solid material major equipment manufacturers for the ethanol industry in
after hydrolysis, is also used as fuel in boilers, besides the Brazil). Table 7 shows these data for an autonomous dis-
lignin obtained after alkaline delignification. The use of the tillery crushing 2,000,000 tons of sugarcane per year,
process by-products (pentose, lignin and unreacted cellu- operating with 22 bar boilers and with azeotropic distilla-
lose) as fuels allows processing of larger fractions of tion as the ethanol dehydration method.
bagasse, since they supply part of the energy required to Process improvements considered dehydration with
run the process. molecular sieves (increase of 40% on the distillation sec-
In order to calculate the energy provided by cellulignin tor), boilers for production of 90 bar steam (increase of
burning, Eq. 1 was employed [6]: 40% on the cogeneration sector) and a heat exchanger
HHV ¼ 0:0877ðLÞ þ 16:4951 ð1Þ network (increase of 10% on the distillation sector).

123
J Ind Microbiol Biotechnol (2011) 38:955–966 961

Fig. 3 Block-flow diagram of

the integrated 1G and 2G
production process

Downloaded from https://academic.oup.com/jimb/article/38/8/955/5994411 by guest on 30 March 2022

Table 5 Economic analysis: main adopted parameters available from SuperPro Designer [17], and values for
Parameter Value
individual equipment available at the ‘‘Corn Stover to
Ethanol Model’’ [18].
Project lifetime 25 years
Salvage value of equipment –
Construction and start-up 2 years Results and discussions
Depreciation (linear) 10 years
Tax rate (income and social contributions) 34.0% Nineteen Alternatives (hereafter named scenarios) were
simulated, considering production of 1G and 2G ethanol,
Changes in equipment capacity were correlated to costs different pretreatment methods, such as steam explosion
considering a coefficient of 0.6. (SE) and hydrogen peroxide (H2O2) followed, or not, by an
Investment calculation on the 2G process considered alkaline delignification step (NaOH) and organosolv
results obtained for equipment dimensions and data pretreatments, considering different length of hydrolysis

Table 6 Average costs and prices adopted in the economic risk analysis
Parameter Lower Medium Higher

Overall investment x - 3r x x ? 3r
Sugarcane average cost (US$/TC)a 18.37 19.44 20.50
Sugarcane trash average cost (US$/t) 10.02 15.02 20.03
Electricity average price (US$/MWh)b 59.09 70.61 82.14
Ethanol average price (US$/L)c 0.46 0.50 0.54
Enzyme average cost (US$/t)d 40.27 118.44 196.61
Hydrogen peroxide average cost (US$/t)e 308.45 462.44 616.44
Sulfur average cost (US$/t)e 43.00 126.46 209.92
Sulfuric acid average cost (US$/t)e 32.24 65.79 99.35
Sodium hydroxide average cost (US$/t)e 104.08 189.24 274.40
a
6-year moving average of sugarcane prices (Dec 2009 values) in São Paulo state (SP), from July 2000 to December 2009 [38]
b
Minimum, medium and maximum prices on renewable energy auctions, values for 2009
c
6-year moving average of anhydrous ethanol prices paid to the producer (Dec 2009 values) in SP, from July 2000 to December 2009 [4]
d
Enzyme prices correspond to approximately between US$ 0.50 and US$ 1.25/gal 2G ethanol [25], for a steam explosion pretreatment with 5%
solids loading on hydrolysis
e
Catalyst average prices obtained from Brazilian average import and export prices from 2000 to 2009 [2]; the exchange rate of US$ 1.00 = R$
1.99 was adopted

123
962 J Ind Microbiol Biotechnol (2011) 38:955–966

Table 7 Investment data for an autonomous distillery The organosolv pretreatment presented the overall lowest
Parameter Value
ethanol production in the integrated process, due to the fact
that it requires a larger amount of energy during pretreat-
Overall investment (million US$) 150 ment, leading to lower amounts of surplus lignocellulosic
Investment on sugarcane reception and juice extraction (%) 15 material for 2G ethanol production.
Investment on juice treatment, fermentation 17 On the other hand, increasing the solids load during
and distillation (%) hydrolysis leads to a decrease in energy consumption of the
Investment on the cogeneration system (%) 30 process, and so has a significant impact on the amount of
Investment on other sectors (%) 38 lignocellulosic materials available for hydrolysis, thus
improving 2G ethanol production.
(24, 48 and 72 h) and different solids loading on hydrolysis
(5 and 15%). The amount of lignocellulosic material Equipment size
available for hydrolysis in each scenario was calculated.

Downloaded from https://academic.oup.com/jimb/article/38/8/955/5994411 by guest on 30 March 2022

The main results are displayed in Table 8. While installed capacity in pretreatment reactors is a
function of the amount of lignocellulosic materials pro-
Ethanol production in the integrated process cessed and the solids content on the pretreatment operation,
hydrolysis reactor sizes also vary tremendously depending
It can be verified that for the same hydrolysis conditions, on the hydrolysis time (24, 48 or 72 h) and solids loading
the overall ethanol productions (L/TC liters per ton of on hydrolysis (5 or 15%), as can be verified in Fig. 4.
sugarcane processed) are very similar for both hydrogen Thus, increasing the time of hydrolysis leads to extre-
peroxide and steam explosion pretreatments, considering mely large hydrolysis reactors (up to 77,000 m3). As
the conversion parameters obtained in the literature mentioned before, increasing the time of hydrolysis from
(Table 4)—for instance, 108.6 and 107.5 for H2O2 and SE 24 to 72 h leads to small increments on overall ethanol
pretreatments at 5% solids for 24 h hydrolysis, respec- production: from 2.6% for the hydrogen peroxide (H2O2)
tively. In addition, the increase on overall ethanol pro- pretreatment with 5% solids to 6.2% for the steam explo-
duction obtained when hydrolysis is carried out at a 5% sion (SE) pretreatment with 15% solids.
solids load for 72 h is not significant (less than 3%) when In addition, performing an alkaline delignification step
compared to that obtained after only 24 h of hydrolysis. (NaOH) after pretreatment promotes a reduction in the size

Table 8 Overall anhydrous

Scenario Ethanol Surplus Lignocellulosic 2G ethanol
ethanol production, surplus
production electricity material processed (L/TC)
electricity and 2G ethanol
(L/TC) (kWh/TC) (t/day, dry basis)
production based on the amount
of sugarcane (TC—tons of 1G 92.7 173.4 0 0.00
sugarcane) and amount of
lignocellulosic materials used as 1G ? H2O2, 5%, 24 h 108.6 98.0 1,311 15.84
feedstock for second generation 1G ? H2O2, 5%, 48 h 110.3 95.9 1,260 17.58
ethanol production for each 1G ? H2O2, 5%, 72 h 111.4 94.5 1,226 18.69
scenario
1G ? H2O2, 15%, 24 h 116.0 62.8 1,923 23.31
1G ? H2O2, 15%, 48 h 119.0 62.7 1,879 26.29
1G ? H2O2, 15%, 72 h 120.8 62.4 1,835 28.04
1G ? SE, 5%, 24 h 107.5 115.2 1,532 14.74
1G ? SE, 5%, 48 h 111.3 108.7 1,396 18.57
1G ? SE, 5%, 72 h 111.6 107.9 1,379 18.92
1G ? SE, 15%, 24 h 113.5 72.7 2,143 20.73
1G ? SE, 15%, 48 h 119.9 71.8 2,033 27.20
1G, first generation ethanol 1G ? SE, 15%, 72 h 120.6 71.6 2,018 27.82
production; H2O2, hydrogen 1G ? OS, 5%, 24 h 105.4 107.3 766 12.68
peroxide pretreatment; SE, 1G ? OS, 15%, 24 h 110.2 98.5 1,056 17.51
steam explosion; OS,
1G ? H2O2, 5%, 24 h, NaOH 112.4 83.8 1,379 19.67
organosolv; NaOH, alkaline
delignification after 1G ? H2O2, 15%, 24 h, NaOH 118.5 59.7 1,805 25.80
pretreatment; 5 and 15%, solids 1G ? SE, 5%, 24 h, NaOH 111.7 97.5 1,685 18.97
loading on hydrolysis; 24, 48 1G ? SE, 15%, 24 h, NaOH 116.0 68.0 2,055 23.28
and 72 h, hydrolysis length

123
J Ind Microbiol Biotechnol (2011) 38:955–966 963

Fig. 4 Installed capacity of

hydrolysis and pretreatment
reactors for each process
alternative

Downloaded from https://academic.oup.com/jimb/article/38/8/955/5994411 by guest on 30 March 2022

of the hydrolysis reactors, due to the fact that most of the consumption of energy, which reduces the amount of sur-
lignin present in the pretreated bagasse (which comprises plus bagasse and trash available for 2G ethanol production
around 40% of the dry cellulignin) is removed. Conse- when compared to the other pretreatment methods ana-
quently, the best configurations for the production of 2G lyzed in this work. Sulfuric acid is used in this process as
seem to be those that consider 24 h hydrolysis with well.
increased solids load and delignification of the pretreated Sulfur dioxide for the steam explosion pretreatment is
material. produced on site by consuming up to 0.9 t/h of sulfur on
combustion chambers. This process is already used in some
Input materials and catalysts sugarcane mills, since sulfur dioxide is used as a bleach to
reduce color in the sugar production process.
Another issue that must be addressed when evaluating 2G The amount of enzymes required on the hydrolysis step
ethanol production is the consumption of catalysts and of each process was evaluated based on the values provided
other substances. Due to the fact that the hydrogen per- in the literature: 10 FPU/g and 10 CBU/g of dry biomass
oxide pretreatment takes place in a reactor with 8 wt% material (DM) for the hydrogen peroxide pretreatment
solids and 11% peroxide (v/v), enormous amounts of H2O2 [13], 15 FPU/g and 15 IU/g DM [24] for the organosolv
are required to run the process: from 60 t/h (for the pretreatment and around 15 FPU/g and 24 IU/g DM [3] for
hydrogen peroxide pretreatment with 5% solids and 72 h of the steam explosion pretreatment (these conditions were
hydrolysis) to 95 t/h (15% solids, 24 h hydrolysis) of H2O2 defined for 5% solids on hydrolysis). The amount of
(100%); the actual amount employed would be larger, since enzymes was estimated considering that the enzymes
industrial H2O2 is obtained at concentrations up to 70% (v/v). employed have the following activities, based on values
In fact, the amount of hydrogen peroxide consumed in this determined by Carrasco et al. [3]: 65 FPU/g for cellulase
process is larger than the amount of overall anhydrous and 376 IU/g of b-glucosidase.
ethanol produced, including 1G ethanol. Thus, unless the
amount of catalyst is reduced dramatically, this process Economic analysis
requires a large industrial plant for hydrogen peroxide pro-
duction to be placed next to the sugarcane processing plant. Results for the overall investment for each alternative are
Studies considering reduction on catalyst consumption are shown in Fig. 5.
under development [29]. A significant increase of investment is required for the
In the organosolv pretreatment, an aqueous ethanol integrated process where 2G considers hydrolysis at 5%
solution (50% v/v) is employed at a solid to liquid ratio of solids loading and longer hydrolysis times.
1:5; after pretreatment, the mixture is filtered and the liquid An economic risk analysis was carried out in order to
fraction is fed to a distillation column, where solvent evaluate the processes for 2G ethanol production employed
recovery takes place. In this column, some ethanol loss on in an integrated sugarcane plant. The values depicted in
the bottom stream occurs, thus decreasing the overall Tables 5 and 6 were employed. An electronic spreadsheet
production of ethanol. In addition, the extreme pretreat- was developed to evaluate the internal rate of return for the
ment conditions (175°C, around 20 bar) involve a large different alternatives of integrated 1G and 2G ethanol

123
964 J Ind Microbiol Biotechnol (2011) 38:955–966

Fig. 5 Investment for the

process alternatives evaluated

Downloaded from https://academic.oup.com/jimb/article/38/8/955/5994411 by guest on 30 March 2022

Fig. 6 Ethanol and surplus
electricity production and
average internal rate of return
(IRR) for the integrated 1G and
2G bioethanol—selected
scenarios

Fig. 7 Internal rate of return

(IRR) of selected scenarios:
99.7% confidence interval

production. A normal distribution of overall investment, are not presented. In addition, due to the massive installed
ethanol and electricity average prices, sugarcane, sugar- capacity of hydrolysis reactors required for the integrated
cane trash, chemicals and enzyme average costs, presented processes with 72 h of hydrolysis, their results of the
in Table 6, was assumed for each alternative, considering economic analyses are also omitted.
10,000 possible scenarios with random combinations of The results obtained for the average internal rate of
these variables. return (IRR) on each scenario is illustrated in Fig. 6, and in
It was verified that in order to achieve similar IRR Fig. 7 the IRR for a 99.7% confidence interval is presented.
values obtained for steam explosion or organosolv pre- Results in Figs. 6 and 7 for the steam explosion
treatments, the hydrogen peroxide pretreatment requires pretreatment with 5% solids load on hydrolysis (SE 5%
that the catalyst (H2O2) price be reduced to 5% of its ori- 24/48 h) present a decrease of the IRR value, in spite of an
ginal value, or the ethanol price paid to the producer must increase in hydrolysis time. This is due to the fact that, for
be raised extremely (at least double that of the other dilute solutions, an increase in hydrolysis time requires
alternatives). Thus, results for the economic evaluation of larger investments and provides few gains in overall
the integrated process with hydrogen peroxide pretreatment ethanol production. Thus, increasing hydrolysis time (and

123
J Ind Microbiol Biotechnol (2011) 38:955–966 965

Fig. 8 Ethanol and surplus

electricity costs for the
bioethanol production
process—selected scenarios

Downloaded from https://academic.oup.com/jimb/article/38/8/955/5994411 by guest on 30 March 2022

Fig. 9 Sensitivity analyses:
impact of sugarcane and
enzyme costs on the internal
rate of return (IRR)

consequently, its yields) does not necessarily lead to an different solids loading on hydrolysis (5 or 15%) and
increase in process profitability; other factors, such as hydrolysis length (24, 48 or 72 h). The amount of ligno-
solids loading on hydrolysis, must be addressed. cellulosic material available for hydrolysis was determined
Ethanol and surplus electricity production costs were for each configuration, considering that the thermal
evaluated, proportionally decreasing the prices of both requirements of the integrated plant are supplied by using
products until the internal rate of return is equal to zero. part of the lignocellulosic material as fuel in boilers. An
Results for calculated production costs are shown in Fig. 8. economic risk analysis was carried out, allowing evaluation
Several sensitivity analyses were performed as well. of the internal rate of return of each scenario. Similar
Sugarcane and enzyme costs have significant impact on the results for overall ethanol production for the integrated first
IRR of the integrated process. Results for these two sen- and second generation processes with both hydrogen per-
sitivity analyses are shown in Fig. 9. oxide and steam explosion pretreatments were obtained for
It can be observed that large (66%) changes in the the same process conditions (solids loading and hydrolysis
enzyme cost leads to similar impacts on the IRR as rela- time). Performing hydrolysis for long periods (72 h)
tively small (15%) changes on sugarcane costs. Therefore, requires extremely large hydrolysis reactors and provides
research towards development of sugarcane varieties that only small increments on yield; thus, the best conditions
lead to decreased sugarcane costs may eventually yield for 2G production integrated with the conventional 1G
greater economic gains than decreases in enzyme costs. process seems to be those that consider around 24–48 h of
hydrolysis and increased solids loading on the hydrolysis
reactor. The amount of catalyst used in the hydrogen per-
oxide pretreatment and its price must be reduced in order
Conclusions
for the process to become feasible. The 2G production
process must be improved (e.g. decreasing investment,
In this work, simulations of the integrated bioethanol
improving yields, developing pentose fermentation to eth-
production process from sugarcane, bagasse and trash were
anol) in order to improve competitiveness of 2G integrated
carried out using the SuperPro Designer process simulator.
with 1G production process, reaching higher forecasted
Selected pre-treatment methods (steam explosion, hydro-
IRR values.
gen peroxide and organosolv) followed, or not, by an
alkaline delignification step were evaluated, along with Acknowledgments The authors would like to thank Dedini Indús-
the subsequent enzymatic hydrolysis step, considering trias de Base S/A for supplying investment data.

123
966 J Ind Microbiol Biotechnol (2011) 38:955–966

References bioethanol and valorization of side streams by anaerobic diges-

tion and combustion. Biotechnol Biofuels 1:14. doi:10.1186/
1. Balat M, Balat H, Öz C (2008) Progress in bioethanol processing. 1754-6834-1-14
Prog Energy Combust Sci 34:551–573 22. Macedo IC, Seabra JEA, Silva JEAR (2008) Green house gases
2. Brasil (2010) Ministry of Development, Industry and Foreign emissions in the production and use of ethanol from sugarcane in
Trade. Available from http://www.mdic.gov.br. Accessed on Brazil: the 2005/2006 averages and a prediction for 2020. Bio-
March 20, 2010 mass Bioenergy 32(7):582–595
3. Carrasco C, Baudel HM, Sendelius J, Modig T, Roslander C, Galbe 23. Martı́n C, González Y, Fernández T, Thomsen AB (2006)
M, Hahn-Hägerdal B, Zacchi G, Lidén G (2010) SO2-catalyzed Investigation of cellulose convertibility and ethanolic fermenta-
steam pretreatment and fermentation of enzymatically hydrolyzed tion of sugarcane bagasse pretreated by wet oxidation and steam
sugarcane bagasse. Enzyme Microb Technol 46:64–73 explosion. J Chem Technol Biotechnol 81:1669–1677
4. CEPEA—Center for Advanced Studies on Applied Economics 24. Mesa L, González E, Ruiz E, Romero I, Cara C, Felissia F, Castro
(2010) Available from http://www.cepea.usp.br. Accessed February E (2010) Preliminary evaluation of organosolv pre-treatment of
20, 2010 sugar cane bagasse for glucose production: application of 23
5. Chen JCP, Chou CC (1993) Cane sugar handbook: a manual for experimental design. Appl Energy 87:109–114
cane sugar manufacturers and their chemists. Wiley, London 25. Novozymes (2009) Enabling cost efficient enzymatic hydrolysis

Downloaded from https://academic.oup.com/jimb/article/38/8/955/5994411 by guest on 30 March 2022

6. Demirbaş A (2001) Relationships between lignin contents and for biomass-to-ethanol conversion. In: 7th Global clean technology
heating values of Biomass. Energy Convers Manag 42:183–188 conference, New York. Available from http://www.novozymes.
7. Dias MOS, Ensinas AV, Nebra SA, Maciel Filho R, Rossell CEV, com/en/MainStructure/Investor/Events?and?presentations/
Maciel MRW (2009) Production of bioethanol and other bio-based Presentations. Accessed March 29, 2010
materials from sugarcane bagasse: integration to conventional 26. Pandey A, Soccol CR, Nigam P, Soccol VT (2000) Biotechno-
bioethanol production process. Chem Eng Res Des 87:1206–1216 logical potential of agro-industrial residues. I: Sugarcane
8. Dias MOS, Ensinas AV, Modesto M, Nebra SA, Maciel Filho R, Bagasse. Bioresour Technol 74:69–80
Rossell CEV (2009) Energy efficiency in anhydrous bioethanol 27. Pfeffer M, Wukovits W, Beckmann G, Friedl A (2007) Analysis
production from sugarcane. Part 1: Process simulation and ther- and decrease of the energy demand of bioethanol-production by
mal integration. Proc ECOS 2009:425–436 process integration. Appl Therm Eng 27:2657–2664
9. Dias MOS, Maciel Filho R, Rossell CEV (2007) Efficient cooling of 28. Rabelo SC, Maciel Filho R, Costa AC (2008) A comparison
fermentation vats in ethanol production—Part 1. Sugar J 70:11–17 between lime and alkaline hydrogen peroxide pretreatments of
10. Dodić S, Popov S, Dodić J, Ranković J, Zavargo Z, Mučibabić RJ sugarcane bagasse for ethanol production. Appl Biochem Bio-
(2009) Bioethanol production from thick juice as intermediate of technol 148:45–58
sugar beet processing. Biomass Bioenergy 33:822–827 29. Rabelo SC, Garzón Fuentes LL, Garcia DR, Maciel Filho R, Costa
11. Ensinas AV, Nebra SA, Lozano MA, Serra LM (2007) Analysis AC (2009) Influence of biomass concentration increase in the pre-
of process steam demand reduction and electricity generation in treatment stage of sugarcane bagasse in the enzymatic hydrolysis
sugar and ethanol production from sugarcane. Energy Convers profile. XXV Congreso Colombiano de Ingenierı́a Quı́mica, Medellı́n
Manag 48:2978–2987 30. Rao K, Chelikani S, Relue P, Varanasi S (2008) A novel tech-
12. Franceschin G, Zamboni A, Bezzo F, Bertucco A (2008) Ethanol nique that enables efficient conduct of simultaneous isomeriza-
from corn: a technical and economical assessment based on dif- tion and fermentation (SIF) of xylose. Appl Biochem Biotechnol
ferent scenarios. Chem Eng Res Des 86:488–498 146:101–117
13. Garcia DR (2009) Determination of kinetics data of the pre- 31. Rosgaard L, Pedersen S, Meyer AS (2007) Comparison of dif-
treatment of sugarcane bagasse with alkaline hydrogen peroxide ferent pretreatment strategies for enzymatic hydrolysis of wheat
and subsequent enzymatic hydrolysis. MSc Dissertation. School and barley straw. Appl Biochem Biotechnol 143:284–296
of Chemical Engineering, University of Campinas (in Portu- 32. Saxena RC, Adhikari DK, Goyal HB (2009) Biomass-based
guese). Available from http://cutter.unicamp.br/document/?code= energy fuel through biochemical routes: a review. Renew Sustain
000442925. Accessed March 29, 2010 Energy Rev 13:167–178
14. Goldemberg J (2007) Ethanol for a sustainable energy future. 33. Sánchez ÓJ, Cardona CA (2008) Trends in biotechnological
Science 315:808–810 production of fuel ethanol from different feedstocks. Bioresour
15. Hassuani SJ, Leal MRLV, Macedo IC (eds) (2005) Biomass Technol 99:5270–5295
power generation—sugarcane bagasse and trash. CTC and 34. Seabra JEA, Tao L, Chum HL, Macedo IC (2010) A techno-
PNUD, Piracicaba economic evaluation of the effects of centralized cellulosic
16. Hoch PM, Espinosa J (2008) Conceptual design and simulation ethanol and co-products refinery options with sugarcane mill
tools applied to the evolutionary optimization of a bioethanol clustering. Biomass Bioenergy 34:1065–1078
purification plant. Ind Eng Chem Res 47:7381–7389 35. Silverstein RA, Chen Y, Sharma-Shivappa RR, Boyette MD,
17. Intelligen, Inc. (2009) SuperPro Designer, v. 7.5 Osborne J (2007) A comparison of chemical pretreatment
18. JBEI (2010) Corn Stover to Ethanol Model. Available from methods for improving saccharification of cotton stalks. Biore-
http://www.econ.jbei.org. Accessed March 29, 2010 sour Technol 98:3000–3011
19. Kuo C-H, Lee C-K (2009) Enhanced enzymatic hydrolysis of 36. Soccol CR et al (2009) Bioethanol from lignocelluloses: status
sugarcane bagasse by N-methylmorpholine-N-oxide pretreat- and perspectives in Brazil. Bioresour Technol 101:4820–4825
ment. Bioresour Technol 100:866–871 37. Srinivasan S (2009) The food v. fuel debate: a nuanced view of
20. Larson ED, Williams RH, Leal MRLV (2001) A review of bio- incentive structures. Renewable Energy 34:950–954
mass integrated-gasifier/gas turbine combined cycle technology 38. UDOP (Union of Biofuel Producers) (2010) Available from
and its application in sugarcane industries, with an analysis for http://www.udop.com.br/index.php?item=cana. Accessed February
Cuba. Energy Sustain Dev 5(1):54–76 20, 2010
21. Maas RHW, Bakker RR, Boersma AR, Bisschops I, Pels JR, de 39. Walter A, Ensinas AV (2010) Combined production of second-
Jong E, Weusthuis RA, Reith H (2008) Pilot-scale conversion of generation biofuels and electricity from sugarcane residues.
lime-treated wheat straw into bioethanol: quality assessment of Energy 35:874–879

123