Bioresource Technology 102 (2011) 78877895

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Production of bioethanol, methane and heat from sugarcane bagasse

in a biorenery concept
S.C. Rabelo a,b,, H. Carrere b, R. Maciel Filho a, A.C. Costa a
a
Faculdade de Engenharia Qumica, Universidade Estadual de Campinas (UNICAMP), Caixa Postal 6066, 13083-970 Campinas, So Paulo, Brazil
b
INRA, UR50, Laboratoire de Biotechnologie de lEnvironnement, Avenue des Etangs, Narbonne F-11100, France

a r t i c l e i n f o a b s t r a c t

Article history: The potential of biogas production from the residues of second generation bioethanol production was
Received 23 November 2010 investigated taking into consideration two types of pretreatment: lime or alkaline hydrogen peroxide.
Received in revised form 26 May 2011 Bagasse was pretreated, enzymatically hydrolyzed and the wastes from pretreatment and hydrolysis
Accepted 30 May 2011
were used to produce biogas. Results have shown that if pretreatment is carried out at a bagasse concen-
tration of 4% DM, the highest global methane production is obtained with the peroxide pretreatment:
72.1 L methane/kg bagasse. The recovery of lignin from the peroxide pretreatment liquor was also the
Keywords:
highest, 112.7 0.01 g/kg of bagasse. Evaluation of four different biofuel production scenarios has shown
Sugarcane bagasse
Bioethanol
that 6365% of the energy that would be produced by bagasse incineration can be recovered by combin-
Anaerobic digestion ing ethanol production with the combustion of lignin and hydrolysis residues, along with the anaerobic
Lignin digestion of pretreatment liquors, while only 3233% of the energy is recovered by bioethanol production
Biorenery alone.
2011 Elsevier Ltd. Open access under the Elsevier OA license.

1. Introduction remainder is stockpiled (UNICA, 2011). Due to the amount of this

biomass as an industrial waste, there is great interest in developing
In recent years, efforts have increased toward the commercial in a biorenery concept, methods for the production of fuels and
production of ethanol, considered the most promising biofuel from chemicals that offer economic, environmental, and strategic
renewable resources. The development of second generation bio- advantages.
ethanol made from lignocellulosic biomass can increase the sus- A biorenery integrates biomass conversion processes to pro-
tainability of feedstock production without competing with food duce fuels, electrical power and chemicals from biomass and, as
production or the cultivation of farmland. Unfortunately, this pro- such, is analogous to a petroleum renery (Cherubini, 2010). By
cess is very costly in terms of the energy input required. However, producing multiple products, a biorenery can take advantage of
the possibility of using the lignocellulosic material in its entirety, the differences in biomass constituents and intermediates and
thus linking bioethanol production with the coproduction of ther- maximize the value derived from the biomass feedstock according
mochemical fuels and/or power, can reduce production costs by to the market situation and biomass availability (Luo et al., 2011).
minimizing the utilization of fossil energy sources and reusing Bagasse, like all lignocellulosic materials, has three major constitu-
the excess material and by-products of the technology employed ents: cellulose, hemicellulose and lignin which, due to its recalci-
(Laser et al., 2009). trant nature, cannot be easily separated into readily utilizable
Among the various agricultural crop residues, sugarcane ba- components. Thus, second generation bioethanol production in-
gasse is the most abundant lignocellulosic material in tropical volves four steps: pretreatment, to render the cellulose accessible;
countries. In Brazil, as a general rule, 1 ton of sugarcane generates hydrolysis with the addition of enzymes or an acid catalyst to re-
280 kg of bagasse and the estimate for the 2011/2012 sugarcane lease the monomeric sugars; fermentation to convert sugars into
harvest is of 568.50 million tons (UNICA, 2011). About 50% of this ethanol; and nally, distillation for product recovery (Margeot
residue is used in distillery plants as a source of energy; the et al., 2009).
Pretreatment is one of the most expensive and least technolog-
ically-mature steps in the process of converting biomass to
fermentable sugars. Hence, it offers a great potential for improve-
Corresponding author. Present address: Laboratrio Nacional de Cincia e
ment in efciency and the reduction of costs through research
Tecnologia do Bioetanol (CTBE), Caixa Postal 6170, 13083-970 Campinas, So Paulo,
Brazil. Tel.: +55 19 3518 3191; fax: +55 19 3518 3164.
and development (Mosier et al., 2005). In this work, two promising
E-mail address: [email protected] (S.C. Rabelo). pretreatment technologies were used: pretreatment with lime

0960-8524 2011 Elsevier Ltd. Open access under the Elsevier OA license.
doi:10.1016/j.biortech.2011.05.081
7888 S.C. Rabelo et al. / Bioresource Technology 102 (2011) 78877895

(Fuentes et al., 2011; Rabelo et al., 2008) and pretreatment with 2. Methods
alkaline hydrogen peroxide (Rabelo et al., 2008; Rivera et al.,
2010). They were chosen because they can be carried out in condi- 2.1. Process description
tions of moderate temperature and pressure and without acids.
Lime pretreatment has low formation of fermentation inhibi- Based on the above considerations, the bagasse biorenery con-
tors, increases pH and provides a low-cost alternative for lignin cept was developed to produce biofuel (bioethanol) and the pro-
solubilization, removing approximately 33% of lignin and 100% of cess efuents were characterized and utilized for the production
acetyl groups. The action of lime is slower than that of other pre- of additional biofuel (methane) and power (lignin and/or enzy-
treatment chemicals but its low cost and safe handling makes it matic hydrolysis residue) to improve the overall recovery of bioen-
attractive (Wyman et al., 2005). ergy from sugarcane. The scheme of the integrated process
Furthermore, lime can be easily recovered as calcium carbonate adopted in this paper can be seen in Fig. 1.
by neutralization with carbon dioxide, although this is an energy-
intensive process and as yet, is not economically feasible. The cal- 2.2. Raw material
cium hydroxide can be subsequently regenerated using established
kiln technology (Kaar and Holtzapple, 2000). Sugarcane bagasse (Saccharum ofcinarum) from a single har-
Hydrogen peroxide is a well-known reagent in the paper and vest was obtained from the Usina So Luiz-Dedini S/A sugar plant
cellulose industry, where it is used as a bleaching agent. It also (Pirassununga/SP, Brazil). It was dried at 45 C for 48 h, left for 48 h
has the great advantage of not leaving residues in the biomass as at room temperature, put into plastic bags and kept in a storage
it degrades into oxygen and water. Furthermore, the formation of room. The dry matter content (DM) was approximately 95%.
secondary products is practically inexistent. This agent has been
shown to be an excellent choice for the pretreatment of sugarcane 2.3. Pretreatment methods
bagasse, leading to almost 100% recovery of the cellulose as glu-
cose after enzymatic hydrolysis (Rivera et al., 2010). The pretreatments were carried out in optimal conditions as
Biomass ethanol production generates great amounts of residu- determined in previous studies (Fuentes et al., 2011; Rabelo
als, both liquid and solid. Among the liquid residuals are the pre- et al., 2008; Rivera et al., 2010). Biomass concentrations in each
treatment liquor, rich in pentoses, soluble and insoluble lignin; pretreatment were varied to determine the maximum solids con-
and the vinasse, the efuent after ethanol recovery. There are also centration compatible with good performance.
solid wastes from the hydrolysis process, composed mainly of lig-
nin and hemicelluloses that are not solubilized during pretreat- 2.3.1. Calcium hydroxide (lime) pretreatment
ment and hydrolysis. The studies were done with solids concentrations of 4%, 5%, 6%,
One of the great challenges of ethanol production from bio- 7% and 8% DM. The material was treated with a lime solution pre-
mass is the use of all wastes left over from the process. It is prob- pared by dissolving 0.47 g/g DM in 100.0 mL distilled water. In all
ably only in this way that the process can be made economically the assays, a certain amount of lime remained insoluble although it
viable as well as environmentally sustainable. The best use for the continued to dissolve during pretreatment. The asks were incu-
pentoses in the pretreatment liquor is ethanol production; how- bated in an orbital shaker MA-832 (Marconi, Piracicaba, SP, Brazil),
ever, current processes using existing microorganisms lead to ex- agitated at 150 rpm, 90 C for 90 h (Fuentes et al., 2011; Rabelo
tremely low yields (Kaparaju et al., 2009). The development of et al., 2008).
genetically-modied microorganisms that can convert pentoses
into ethanol is being assessed by different research groups but
2.3.2. Alkaline hydrogen peroxide pretreatment (AHP)
no viable industrial process has been developed up to now. There
The pretreatment was performed with solids concentrations of
exist some methods for vinasse re-use, such as thermal concen-
4%, 5%, 6%, 7%, 8%, 9%, 10% and 15% DM. A hydrogen peroxide solu-
tration, land application, for cattle feed, fungi production and
tion was prepared by dissolving 7.36% (v/v) of H2O2 in 100.0 mL
recycling of vinasse into partially dilute molasses in fermentation
distilled water and adjusting the pH to 11.5 with sodium hydrox-
(Andrade et al., 2009).
ide. The asks were incubated in an orbital shaker MA-832 (Mar-
Another alternative for the use of these wastes is biogas produc-
tion, which can be a sustainable solution for the organic matter re-
moval from efuents. Another further advantage here is the
possibility of using the new efuent resulting from biogas produc- Sugarcane
bagasse
tion as fertilizer on agricultural soils (Liu et al., 2006). Before using
the residual liquid, lignin can be precipitated and used for heat and
energy production (Sassner et al., 2008), though this requires addi-
Pretreatment
tional energy input to reduce the water content. In the current
study, several scenarios for bioethanol production and co-produc- Solid fraction Liquid fraction

tion of electrical power and biogas using dry bagasse as substrate

were investigated and compared in terms of gross energy output. Enzimac Hydrolysis
Anaerobic
Ethanol was produced from bagasse pretreated with either lime Solid fraction
digeson
Lignin
Liquid fraction
or hydrogen peroxide-pretreated bagasse, using sequential hydro-
lysis and fermentation (SHF). Methane and power were produced Fermentaon Methane

from the process residues of ethanol production. The primary

aim was to compare the energy output per unit of bagasse mass Boilers/energy
Disllaon
for the different scenarios (1) Combustion of untreated bagasse, Vinasse

(2) bioethanol from pretreated bagasse, (3) bioethanol from pre-

treated bagasse, lignin combustion, and biogas from pretreatment Second generaon Ferlizer
liquor and hydrolysis residue, (4) bioethanol from pretreated ba- bioehtanol
gasse, lignin and hydrolysis residue combustion, biogas from liquor
pretreatment. Fig. 1. Integrated process for second generation bioethanol and biogas production.
 S.C. Rabelo et al. / Bioresource Technology 102 (2011) 78877895 7889

coni, Piracicaba, SP, Brazil), agitated at 150 rpm, at 25 C for 1 h 2.6.2. Pretreatment liquors and solids residues (methane production)
(Rabelo et al., 2008; Rivera et al., 2010). 2.6.2.1. Chemical Oxygen Demand (COD). COD was determined
using Spectroquant test kits (Merck, Darmstadt, Germany) based
on the oxidation of the sample through a heated sulfuric acid solu-
2.4. Pretreatment liquor and precipitated lignin tion of potassium dichromate, using silver sulfate as the catalyst.
Samples were analyzed in triplicate by adding 2.0 mL of diluted
After the pretreatment step, the liquors were separated from pretreatment liquor to tubes containing the oxidant solution and
the solid fraction by ltration and then reserved for the precipita- heated for 2 h at 150 C. Thereafter, the tubes were cooled to ambi-
tion of lignin. ent temperature and the solution was analyzed in a spectropho-
The solids fraction was washed several times for removal of the tometer HACH DR/2000 (Hach Company, Loveland, CO, USA) at
water-soluble solids (WS) and used to determine the solids recov- 620 nm. The results obtained have been expressed in mg O2/L.
ery yield and chemical composition (Sluiter et al., 2008a,b). For the solid fraction, the COD was determined, as previously
The insoluble lignin was precipitated using a 1% (v/v) solution described, by adding 0.2 g of hydrolysis residue in the Spectro-
of hydrochloric acid down to pH 2 (Ibrahim and Chuah, 2004) quant test kits (Merck, Darmstadt, Germany).
and the liquor was centrifuged at 3000 rpm to obtain the liquid
fraction (soluble sugars) that was used for biogas production. 2.6.2.2. Sugar concentration. The pretreatment liquor was analyzed
by an HPLC equipped with a refractive index detector, in accor-
dance with the National Renewable Energy Laboratory (NREL)
2.5. Enzymatic hydrolysis standard procedure (Sluiter et al., 2008b). Approximately 5.0 mL
of liquor was subjected to acid hydrolysis after the addition of sul-
The enzymatic hydrolysis of the washed material was per- furic acid to a pH level of 2. The solution was hydrolyzed at 121 C
formed using a substrate concentration of 3.0% (w/w) WIS for 1 h. This step was necessary to ensure that all the oligosaccha-
(water-insoluble solids) in asks incubated in an orbital shaker rides present were hydrolyzed to monosaccharides and could thus
MA-832 (Marconi, Piracicaba, SP, Brazil) agitated at 100 rpm at be quantied.
50 C. The pH was adjusted to 4.8 with 0.05 mol/L sodium citrate The separation was performed in a Sugar-Pak I column (Waters
buffer. Corporation, Milford, MA, USA) at 70 C with a ow rate of 0.5 mL/
Cellulase from T. reesei (SigmaAldrich Corporation, St. Louis, min, using ltered deionized water as the mobile phase.
MO, USA) was added at a concentration corresponding to 50 FPU/
g WIS for the material pretreated with lime and 3.5 FPU/g WIS 2.6.2.3. Chemical composition. The chemical composition of the res-
for material pretreated with hydrogen peroxide. b-glucosidase idues resulting from the hydrolysis was determined by the Van So-
from Aspergillus niger (SigmaAldrich Corporation, St. Louis, MO, est method (Van Soest, 1963). The bagasse residue after enzymatic
USA) was added at a concentration corresponding to 25 UI/g WIS hydrolysis was analyzed by the FiberBag system (Gerhardt Analyt-
for both pretreated materials. ical Systems, Knigswinter, Germany). This involves a sequential
Cellulase activity was determined as lter paper units per mil- extraction under neutral and acid detergent, followed by strong
liliter, as recommended by the International Union of Pure and Ap- acid extraction. The different fractions were: (i) soluble in neutral
plied Chemistry (Ghose, 1987). b-glucosidase activity was detergent fraction (SND); (ii) hemicellulose (HEMI), which was ex-
determined through a solution of cellobiose 15 mmol/L and ex- tracted by acid detergent; (iii) cellulose (CELL), which was ex-
pressed in units per milliliter (IU/mL) (Wood and Bhat, 1988). En- tracted by 76% sulfuric acid; (iv) lignin (LIGN).
zyme activity was 47.44 FPU/mL for cellulases and 343.63 IU/mL
for b-glucosidase. 2.6.2.4. Total solids (TS), dry matter (DM) and volatile solids
After enzymatic hydrolysis, the liquid fraction rich in glucose (VS). Total solids (TS) or dry matter (DM) and volatile solids (VS)
was used for ethanol production by fermentation of the liquor by were determined in accordance with the Standard Methods (APHA,
Saccharomyces cerevisiae and the solid residue was used for biogas 1995). To determine TS or DM, approximately 20.0 mL of liquor or
production. In this study, only the residues remaining after enzy- 2.00 g of hydrolysis residue were dried at 105 C to constant
matic hydrolysis of bagasse pretreated at a solids concentration weight. After this, to quantitate VS, the material was burnt in an
of 4% DM were used to produce biogas, as this was the pretreat- oven at 550 C for two hours.
ment condition that led to the best glucose yields after hydrolysis.
2.6.3. Lignin characterization
Lignin samples obtained by the two pretreatments were ana-
2.6. Analytical methods lyzed for moisture content in accordance with the NREL standard
procedure (Sluiter et al., 2008a). Samples of lignin were also
2.6.1. Solid fraction after pretreatment (bioethanol production) analyzed by Differential Scanning Calorimetry (DSC) (Mettler-
Extractives, structural carbohydrates and lignin were analyzed Toledo, Columbus, OH, USA) and had their molar weight distribu-
in accordance with Sluiter et al. (2008a,b). Prior to compositional tion determined by Gel Permeation Chromatography (GPC) using
analysis the samples were nely ground in a knife mill MA-630/ a CLASS-LC10 data analyzer and a series of 500, 103 and 104
1/E (Marconi, Piracicaba, SP, Brazil). PLGel columns (Shimadzu, Kyoto, Japan).
Sugars and organic acids, HMF and furfural were analyzed with
an HPLC system (Waters Corporation, Milford, MA, USA) equipped 2.7. Biomethane potential (BMP) tests
with a refractive index detector. Cellobiose, glucose, xylose, arabi-
nose and mannose were separated on a Sugar-Pak I column Methane production was assessed by batch biochemical meth-
(Waters Corporation, Milford, MA, USA) run at a ow rate of ane potential (BMP) in mesophilic conditions (35 C) in 100 mL
0.5 mL/min at 70 C, with water as the eluent. Acetic acid, HMF, reactors, samples of the pretreatment liquor and solid residues
furfural, lactic acid, formic acid and levulinic acid were separated from hydrolysis were added to a solution of macroelements
using an Aminex HPX-87H column (Bio-Rad Laboratories Inc., Her- (sources of N, P, Mg, Ca, K), a solution of oligoelements, a solution
cules, CA, USA) at 65 C with 0.5 mL/min of H2SO4 at 5 mM as the of bicarbonate (buffer solution) and the innoculum (sludge from an
eluent. anaerobic digester at a sugar factory, Marseille, France; its compo-
7890 S.C. Rabelo et al. / Bioresource Technology 102 (2011) 78877895

sition was 91.3 0.002 g TS/L and 76.6 0.001 g VS/L). The samples that the residue from the hydrolysis of lime-pretreated bagasse
were added to obtain a concentration of 0.5 g COD/g TS of sludge. had more hemicellulose, since this pretreatment solubilizes only
Once the reactors had been prepared, a degasication with nitro- a small part of this polymer. Although peroxide pretreatment is
gen was carried out to obtain anaerobic conditions. more efcient in solubilizing hemicelluloses, 40% of the residue
Biogas production without the addition of liquor or hydrolysis after hydrolysis was hemicellulose, highlighting that it would be
residue was also carried out in order to quantitate endogenous advantageous to use the residue for biogas production. There was
production and served as a negative control. In addition, biogas almost no cellulose left in the residues; its amount was a little
production with the addition of ethanol, a completely degradable higher in the residue from lime-pretreated bagasse while the lignin
material replacing pretreatment liquor or hydrolysis residue, was composition was similar for the two residues.
used as a positive control. BMP tests were duplicated and lasted
up to 40 days. Liquor biodegradability was calculated on the basis 3.2. COD analysis and sugar concentration
of a theoretic production yield of 350 NmL CH4/g COD (Angelidaki
and Sanders, 2004). Results of COD in the samples of pretreatment liquor and
hydrolysis residues are presented in Table 1 In addition, the table
2.8. Volume and composition of the produced biogas shows sugar concentration in the pretreatment liquor.
It can be seen that COD values for the pretreatment liquor from
Biogas volume was measured by vertical displacement of water alkaline hydrogen peroxide were higher that those for the pretreat-
in a gasometer and its composition was measured by a gas chro- ment with lime. Besides the higher concentration of sugars in the
matograph Varian GC-CP4900 (Agilent Technologies, Santa Clara, liquor, which led to higher values of COD, the excess of hydrogen
CO, USA) equipped with two columns. The rst one, a Molsieve peroxide in the reaction medium could have further increased this
5A PLOT (SigmaAldrich Corporation, St. Louis, MO, USA), was used value. Indeed, the residual concentration of hydrogen peroxide
to separate O2, N2 and CH4 and the second, a HayeSep A column modies the COD value by consuming the oxidizing agent potas-
(Vici Valco Instruments Co. Inc., Houston, TX, USA), to separate sium dichromate (K2Cr2O7), as shown in Eq. (1) (Talinli and Ander-
CO2 from other gases. The detection of gaseous compounds was son, 1992).
done using a thermal conductivity detector with an injected vol- Cr2 O2- 3
7 3H2 O2 8H ! 2Cr 3O2 7H2 O 1
ume of 1.0 mL. Calibration was performed with a standard gas
composed of 25.00% of CO2, 1.98% of O2, 10.00% of N2 and 63.02% This interference can be corrected by knowing the residual con-
of CH4. centration of hydrogen peroxide in each pretreatment liquor.
According to Lin and Lo (1997), a concentration of 1.0 g/L of hydro-
gen peroxide is equivalent to 270 mg/L of COD. Similar results were
3. Results and discussion
obtained by Dantas (2005), who recorded 263 mg/L of COD in
1.0 g/L of hydrogen peroxide. Thus, even though there was residual
3.1. Chemical composition of solids fraction
peroxide in the pretreatment liquor, interference in the nal COD
was very low in comparison with the values given in Table 1.
Fig. 2 depicts the chemical composition of untreated bagasse,
In Brazil, vinasse, obtained after distillation, presents a COD
the solid fraction after pretreatment (4% DM) and solid residues
range from 15 to 33 g O2/L vinasse when the fermentation is car-
after hydrolysis with the two catalysts.
ried out with sugarcane juice, from 40 to 50 g O2/L vinasse when
One ton of bagasse (95% DM) gave a yield of 560.5 kg DM after
the juice and molasses are blended in fermentation and from 60
lime pretreatment. Much of the cellulose was preserved in the so-
to 75 g O2/L vinasse when fermentation is performed with molas-
lid fraction, representing 66.2% of pretreated material. Part of the
ses (Lamo, 1991).
hemicelluloses were solubilized, and this constituent represented
It can also be noted from Table 1 that higher sugar concentra-
20.4% of the chemical composition of pretreated material. For ba-
tions were observed for liquors from alkaline peroxide pretreat-
gasse pretreated with alkaline hydrogen peroxide, a yield of
ment as this pretreatment solubilizes more hemicelluloses than
433.2 kg DM was obtained, while the pretreated material was
the pretreatment with lime (Rabelo et al., 2008).
composed primarily of cellulose which represented 80.8% of the
material. Much of the hemicelluloses was solubilized during pre-
3.3. Volume and composition of the biogas produced
treatment such that after pretreatment this constituent repre-
sented only 11.7% of the solid fraction. It can be seen from Fig. 2
In this study, biomethane potential tests were carried out esti-
mate the potential production of biogas from the liquors from pre-
treatment and from the solid residues of hydrolysis. Methane
1.5 7.5
100 13.3 Table 1
25.0 25.1 25.1 COD values and sugar concentration.
80 7.2 3.2 Concentration of solids in COD (g O2/L or g O2/g Sugar
pretreatment (% DM) biomassa) concentration
60 38.4 66.2 80.8 39.5 (g/L)
50.8
40 Lime AHP Lime AHP
23.2 4 10.0 0.1 21.4 0.4 1.504 7.565
20 32.1 5 12.6 0.6 23.2 0.0 2.196 8.625
20.4 17.0
11.9 11.7 6 11.6 0.3 27.7 0.4 1.787 12.773
0 7 16.8 0.9 37.0 0.4 2.933 16.598
Untreated Lime* AHP* Lime** AHP** 8 21.2 0.1 43.5 0.4 2.904 19.731
9 45.2 0.7 16.920
10 58.2 0.3 20.330
15 67.4 0.9 21.759
Fig. 2. Chemical composition of untreated and pretreated bagasse () and 4 1.1 0.0 1.1 0.0
enzymatic hydrolysis residues (). SND represents the soluble matter in the
a
residue. Solid residues from hydrolysis.
 S.C. Rabelo et al. / Bioresource Technology 102 (2011) 78877895 7891

potential can be expressed specically per amount of waste (L CH4/ 4

kg-waste), per volume of waste (L CH4/L-waste), per mass of vola-
(A)
tile solids added (L CH4/kg-VS) or per COD added (L CH4/kg-COD).
The volume is usually expressed in standard pressure (1 atm) 3
and temperature (0 C) conditions (STP conditions) (Angelidaki
and Sanders, 2004). In this work, the theoretic methane yields
were based on the COD. To calculate the biodegradability, the the- 2
oretic methane production (350.0 NL/kg of COD) was considered.
Table 2 shows the composition of biogas and the methane yield
(NL/kg of COD). The highest methane yield was obtained by work- 1
ing with 4% DM solids for both types of pretreatment. Considering
the solid residue (residue after enzymatic hydrolysis), 162 2 L/kg
COD were produced for the hydrolysis residue of bagasse pre- 0
0 10 20 30 40
treated with lime in comparison to 152 1 L/g COD for the hydro-
Time (days)
lysis residue of bagasse pretreated with hydrogen peroxide.
Fig. 3 shows the amount of methane produced per liter of pre-
7 (B)

CH4/pretreatment liquor (NL/L) or

treatment liquor, considering pretreatments performed with dif-

hydrolysis residue (NL/kgx103)

ferent solids concentrations (415% DM) and the amount of 6
methane produced per ton of hydrolysis residue for the two types
of pretreatment considered. 5
It can be noted from Fig. 3 that, in general, higher solids concen- 4
trations in the pretreatment step lead to larger volumes of meth-
ane produced, as higher solids concentrations produce higher 3
concentrations of sugars and volatile acids in the reaction medium
2
(not shown).
The highest production of methane was observed for the pre- 1
treatments with the highest solids concentration: 15% DM (perox-
ide pretreatment) and 8% DM (lime pretreatment). 6.5 0.3 NL of 0
0 10 20 30 40
methane/L pretreatment liquor were produced using the liquor Time (days)
from the pretreatment with peroxide and 3.13 0.06 NL of meth-
ane/L pretreatment liquor using the liquor from the pretreatment Fig. 3. Volume of methane produced relative to the volume of pretreatment liquor
with lime. introduced (NL/L) or to the mass of hydrolysis residue from bagasse (NL/kg  103)
pretreated with lime (A) or alkaline hydrogen peroxide (B), with different solids
Fig. 3 also gives the volume of methane produced relative to the
concentrations (DM): 4% ( ), 5% ( ), 6% ( ), 7% ( ), 8% ( ), 9%
mass of hydrolysis residue put into the reactor for each pretreat- ( ), 10% ( ), 15% ( ) and solid residue ( ). N represents the Normal
ment. It can be seen that the results are similar for both pretreat- Temperature and Pressure (0 C and 1 atm).
ments: 167 1 NL of methane/kg of residue was obtained with the
residue of hydrolysis of bagasse pretreated with lime and
166 2 NL of methane/kg of when the pretreatment was done with treatment liquor and 0.0275 kg of hydrolysis residue. As 165.6 L of
alkaline hydrogen peroxide. methane were produced per kg of hydrolysis residues and 2.7 L
In this work, experiments to determine the potential of biogas biogas were produced per liter of pretreatment liquor, it is possible
production from vinasse were not carried out, although the pro- to calculate a volume of 4.5 L of methane/kg bagasse from
posed integrated process, shown in Fig. 1, considers this alterna- the hydrolysis residue and 67.5 L of methane/kg bagasse from
tive. Even using only the wastes from pretreatment and the pretreatment liquor. Thus, the total biogas production was
hydrolysis, however, the results are promising. In fact, nowadays 72.1 L methane/kg bagasse.
the vinasse is used in soil irrigation as a source of potassium but When the pretreatment was with lime (4% DM) 1 kg of bagasse
part of the process stream could be diverted to biogas production. resulted in the same volume of pretreatment liquor but 0.08225 kg
When pretreatment with alkaline hydrogen peroxide carried of hydrolysis residue. 167.2 L of methane/kg of hydrolysis residue
out with 4% DM is considered, 1 kg bagasse resulted in 25 L of pre- and 1.8 L methane/L of pretreatment liquor were obtained, which

Table 2
Biogas composition and methane yield from anaerobic digestion.

Concentration of solids in pretreatment (% DM) Biogas composition (%) CH4 yield (NL CH4/kg COD) Biodegradabilityb
Lime AHP Lime AHP Lime AHP
CO2 CH4 CO2 CH4
4 33.7 1.4 66.3 0.4 30.3 1.1 69.7 1.1 180 2 127 2 0.51 0.36
5 31.1 0.0 68.9 0.2 33.2 0.2 66.8 2.1 173 1 98 0 0.49 0.28
6 31.4 0.3 68.6 0.2 35.6 0.2 64.4 0.0 183 1 97 3 0.52 0.28
7 29.8 0.5 70.2 0.6 34.9 0.6 65.1 0.4 170 2 85 2 0.49 0.24
8 31.0 0.3 69.0 0.5 39.7 0.4 60.3 1.0 148 2 89 1 0.42 0.25
9 32.9 0.4 67.1 1.4 85 2 0.24
10 33.2 0.2 66.8 0.1 84 1 0.24
15 33.9 0.1 66.1 0.2 96 2 0.27
4a 38.2 0.0 61.8 0.2 37.4 0.9 62.6 0.3 162 2 152 1 0.46 0.43
a
Solid residues from hydrolysis.
b
The ratio of measured CH4 yield/theoretic CH4 yield.
7892 S.C. Rabelo et al. / Bioresource Technology 102 (2011) 78877895

resulted in a production of 13.7 L of methane/kg bagasse from the Lignin - lime

hydrolysis residue and 45.0 L of methane/kg bagasse from the pre- 1
Lignin - hydrogen peroxide
treatment liquor, making a total methane production of 58.7 L of
methane/kg bagasse. -1

Heat flux (mW)

Lu et al. (2009), assessing the co-production of hydrogen and
methane from cornstalks pretreated by steam explosion, obtained -3
63.7 and 114.6 L of hydrogen and biogas per kg of biomass, respec- EXO
tively. Bauer et al. (2009) evaluated a process for ethanol and
-5
methane production from steam-pretreated wheat straw. They
proposed optimizing the second generation ethanol production
-7
process by using vinasse after ethanol production to produce bio- 0 100 200 300 400 500
gas. They obtained 183 L of methane/kg of wheat straw from the Temperature (C)
vinasse. Kaparaju et al. (2009) evaluated the production of bioeth-
anol, biohydrogen and biogas from hydrothermally-pretreated Fig. 4. DSC prole representative of lignin degradation.
wheat straw. Pretreated biomass was enzymatically hydrolyzed
and fermented to ethanol and biohydrogen. The efuents from
the two processes were further used to produce methane with
is a simplied analysis in which the energy requirements involved
yields of 324 and 381 L/kg of volatile solids, respectively.
in biofuel production processes such as farming, harvesting, trans-
porting, feedstock processing, fermentation, alcohol recovery, alco-
3.4. Lignin characterization hol purication, biogas process, etc. were not considered. However,
the energy requirements of these unit operations can be assumed
The lignin mass precipitated and its moisture content are given to be the same whatever the pretreatment used.
in Table 3 for both types of pretreatment studied. Moisture content The combustion of the 1 ton of bagasse (DM) produced
directly inuences the caloric value of a compound. Table 3 also 15,130.6 MJ of useful energy and was considered as the reference
shows the values of molecular weight average Mw and number scenario (Scenario 1) (Dias et al., in press).
average Mn molecular weights for the lignin obtained by precipita- Among the scenarios studied (2, 3 and 4), the highest energy
tion of the pretreatment liquors from both reagents used. Polydis- output for both types of pretreatments was obtained with scenario
persity was calculated according to Eq. (2): 4. The total energy output for this scenario was 10,731.2 MJ/ton
Polydispersity Mw=Mn 2 bagasse and 10,281.7 MJ/ton bagasse when the pretreatment was
with respectively alkaline hydrogen peroxide and lime.
From Table 3 it can be seen that lignin isolated from peroxide In scenario 4, the hydrolyzed cellulose was used for bioethanol
pretreatment liquor presented fragments with higher weight aver- production leading to an energy generation of 5389.3 MJ/ton ba-
age molecular weights Mw and had more fragments with different gasse when bagasse was pretreated with peroxide and
molecular weights, which is evidenced by the higher polydispersity. 5278.4 MJ/ton bagasse for pretreatment with lime. Ethanol pro-
Fig. 4 shows the curves of differential scanning calorimetry for duction represented around 32% of the energy that can be pro-
samples of lignin precipitated from both pretreatment liquors. duced from bagasse combustion and 5051% of the total energy
The exotherm of degradation of the lignin from lime pretreatment recovered in scenario 4. Part of the lignin was precipitated from
occured around 413.5 C with enthalpy of 67.9 J/g and the exo- the pretreatment liquor and used together with the hydrolysis res-
therm for the lignin from the peroxide pretreatment occured at idue of each pretreatment process for energy generation through
400.6 C and enthalpy of 52.5 J/g. direct combustion in boilers. The energy provided by burning the
lignin was obtained through the mass of lignin recovered in each
3.5. Mass balance and energy output pretreatment, taking into account the caloric value of the lignin
of 25.689 kJ/kg (Baker, 1983). By burning lignin, it was possible
The mass ows for the two biorenery alternatives are pre- to obtain an energy output of 2895.9 MJ/ton bagasse for the lignin
sented in Fig. 5 for both the lime (A) and alkaline hydrogen perox- obtained from the pretreatment liquor from peroxide-pretreated
ide (B) pretreatments. Mass balance was expressed in terms of dry bagasse and 2633.4 MJ/ton bagasse when pretreatment was with
matter (DM) and the mass loss during chemical pretreatment of lime.
the biomass was considered in the calculation. This energy from lignin represented 1618% of the energy that
After pretreatment of one ton of bagasse with hydrogen perox- can be produced from bagasse combustion and 2627% of the en-
ide, 45.6% DM remained in the solid fraction and the remaining ergy recovered in scenario 4. Burning of the hydrolysis residue led
54.4% were extracted as hydrolysate. For lime pretreatment, to a generation of energy of 572.5 MJ/ton bagasse for peroxide pre-
59.0% of DM remained in the solid fraction and 41.0% were ex- treatment and 1458.5 MJ/ton bagasse for pretreatment with lime.
tracted as hydrolysate. Energy from burning hydrolysis residues represented 49% of en-
For biomass pretreated with peroxide and lime, Fig. 6 intro- ergy that can be produced from bagasse combustion and 514%
duces four possible scenarios proposed for optimizing bagasse of energy recovered in scenario 4. Finally, the soluble fraction of
use as well as the different contributions of individual biofuels in the pretreatment liquor was used for biogas production, generating
each scenario. It should be noted that the present study was car- 1873.5 MJ/ton bagasse and 911.4 MJ/ton bagasse of energy for the
ried out with the aim of comparing two pretreatment methods. It liquor of bagasse pretreated with peroxide and lime, respectively.

Table 3
Lignin characterization.
 
Lignin Lignin recovered (g/kg of bagasse) Moisture content (%) Weight average Mw Number average Mn Polydispersity Mw=Mn

Lime 139.63 0.01 22.72 0.5 907 460 3.66

AHP 147.28 0.01 19.43 0.1 952 502 3.88
 S.C. Rabelo et al. / Bioresource Technology 102 (2011) 78877895 7893

Fermentation
Hydrolyzate
Hydrolysis liquor (DM kg) fermentation (DM kg)
469.0 Saccharomyces
377.5
Enzymatic hydrolysis Glucose 372.5 cerevisiae
Xylose 75.5
Xylose 75.5 CO2 182.1
Fibers (DM kg)
50.0 FPU/g Glycerol 9.3
560.5
Sugarcane bagasse (DM kg) 25.0 CBU/g Ethanol 190.3
Cellulose 369.2
1000 950 Lime pretreatment Hydrolysis residue (DM kg)
Hemicellulose 112.9 Anaerobic digestion (DM kg)
Cellulose 384.8 Lignin 75.1 91.5
Innoculum 22.3
Hemicellulose 232.8 90 h Others 3.4 Cellulose 6.6
Methane 9.3
Lignin 250.8 90C Hemicellulose 46.5
Carbonic gas 13.0
Ash 15.2 0.4 (g/g) Lignin 22.9
Pretreatment liquor (DM kg)
Other 66.5 Others 15.5
389.5
Cellulose 14.8
Anaerobic digestion (DM kg)
Hemicellulose 119.9 Innoculum Fertilizer
73.5
Lignin 162.5
Methane 30.7
Others 92.3
Carbonic gas 42.8

102.5 kg precipitable lignin (recovered)

(A) Fermentation
Hydrolyzate
Hydrolysis liquor (DM kg) fermentation (DM kg)
390.7 Saccharomyces
348.2
Enzymatic hydrolysis Glucose 394.4 cerevisiae
Xylose 39.1
Xylose 39.1 CO2 192.9
Fibers (DM kg)
3.5 FPU/g Glycerol 9.9
433.2
Sugarcane bagasse (DM kg) Hydrogen peroxide 25.0 CBU/g Ethanol 201.5
Cellulose 356.3
1000 950 pretreatment Hydrolysis residue (DM kg)
Hemicellulose 51.3 Anaerobic digestion (DM kg)
Cellulose 384.8 Lignin 32.3 42.5
Innoculum 6.8
Hemicellulose 232.8 1h Cellulose 1.4
Methane 3.1
Lignin 250.8 25C Hemicellulose 16.8
Carbonic gas 3.7
Ash 15.2 7.35 % (v/v) Lignin 10.7
Pretreatment liquor (DM kg)
Other 66.5 Others 13.7
516.8
Cellulose 10.5
Anaerobic digestion (DM kg)
Hemicellulose 166.3 Innoculum Fertilizer
100.9
Lignin 184.3
Methane 46.0
Others 155.8
Carbonic gas 54.9

112.7 kg precipitable lignin (recovered)

(B)
Fig. 5. Mass ow in the biorenery process for pretreatment with lime (A) and hydrogen peroxide (B).

16000
Energy (MJ/ton bagasse)

14000
12000
10000
8000
6000
4000
2000
0
1 2* 2** 3* 3** 4* 4**
Scenario

Fig. 6. Energy output from each scenario studied in the biorenery concept using bagasse pretreated with hydrogen peroxide () or lime (). (1) Untreated bagasse
combustion, (2) pretreated bagasse bioethanol, (3) pretreated bagasse bioethanol, lignin combustion, biogas from pretreatment liquor and hydrolysis residue, (4)
pretreated bagasse bioethanol, lignin and hydrolysis residue combustion, biogas from liquor pretreatment.

This energy represented 611% of the overall potential from ba- which was more than twice the energy recovered by ethanol pro-
gasse combustion and 918% of energy recovered in scenario 4. duced alone from hexoses: 44005100 MJ/ton processed.
Kaparaju et al. (2009), studying the hydrothermal pretreatment Comparing scenarios 3 and 4, there is a decrease in the energy
of wheat straw, obtained their best result for energy generation, on production of 3.93% and 11.22%, respectively when using bagasse
the basis of multiple biofuels, of 9364 MJ/ton wheat straw of en- pretreated with peroxide and lime, in scenario 3, the residue of
ergy during the production of biogas, bioethanol and hydrogen. enzymatic hydrolysis was directed to the production of biogas,
The authors also showed that when using biomass to produce only which generated less energy than the straightforward burning of
bioethanol, the energy output was 3572 MJ/ton wheat straw. this residue. It is thus more advantageous to burn hydrolysis resi-
Using steam-pretreated hemp as the substrate, Kreuger et al. dues rather than to use them to produce biogas. Production of bio-
(2011) studied the co-production of ethanol and methane, obtain- ethanol alone (scenario 2) was of course the least energy-efcient
ing a high yield: 11,10011,700 MJ/ton processed dry matter process.
7894 S.C. Rabelo et al. / Bioresource Technology 102 (2011) 78877895

This study has not considered using vinasse for biogas produc- vectors, in particular liquid fuels, are of great interest and justify
tion. According to Salomon and Lora (2009), for each liter of rst the development of these processes.
generation bioethanol obtained after distillation, approximately
13 L of vinasse are produced and this is, in fact, the amount usually
Acknowledgements
found in large-scale production plants in Brazil. This efuent is
highly polluting but as high fertilizing value due to being rich in or-
The authors acknowledge the nancial support provided by Fa-
ganic matter and having high BOD (Biochemical Oxygen Demand).
pesp, CNPq and INRA.
The chemical composition of vinasse depends on the characteris-
tics of the soil, the variety of sugarcane, the period of the harvest
and the industrial process used for the production of ethanol. References
According to Lettinga and Haandel (1993), 1 L of vinasse yields
about 14.23 L of methane, which could represent a signicant in- Andrade, R.R., Rivera, E.A.C., Atala, D.I.P., Maugeri Filho, F., Maciel Filho, R., Costa,
A.C., 2009. Development of extractive processes and robust mathematical
crease in useful energy production. Along with the methane pro-
model for bioethanol production. In: Jason, B.E. (Ed.), Bioethanol: Production.
duced by the anaerobic digestion of vinasse, several other Benets and Economics. Nova Science Publishers, New York, pp. 7592.
products are released. The production of 68 kg granular sludge Angelidaki, I., Sanders, W., 2004. Assessment of the anaerobic biodegradability of
per m3 of bioethanol can be sold as inoculum of UASB digesters macropollutants. Rev. Environ. Sci. Biotechnol. 3, 117129.
APHA, 1995. Standard Methods for the Examination of Water and Wastewater, 19th
(Upow Anaerobic Sludge Blanket). Also, there is a release of a sig- ed. American Public Health Association, New York, USA.
nicant amount of solids which settle spontaneously (10 g/L of Baker, A.J., 1983. Wood fuel properties and fuel products from woods. In: Fuel Wood
vinasse). Management and Utilization Seminar: Proceedings, November 911, 1982.
Michigan State Univ., East Lansing, MI, pp. 1419.
In Brazil, the vinasse is directly applied to the soil as a fertilizer Bauer, A., Mayr, H., Hopfner-Sixt, K., Amon, T., 2009. Detailed monitoring of two
and source of potassium. But for this practice to be efcient, a thor- biogas plants and mechanical solidliquid separation of fermentation residues.
ough soil analysis is necessary, so that the appropriate amount of J. Biotechnol. 142, 5663.
Cherubini, F., 2010. The biorenery concept: using biomass instead of oil for
vinasse can be used. Environmental agencies are limiting the quan- producing energy and chemicals. Energy Convers. Manag. 51, 14121421.
tity of vinasse per hectare (Salomon and Lora, 2009). In the case of Dantas, T.L.P., 2005. Decomposition of hydrogen peroxide in a hybrid catalyst and
vinasse obtained from second generation bioethanol production, it advanced oxidation of textile wastewater by Fenton reagent modied. Master of
Science Thesis. Federal University of Santa Catarina, Brazil (in Portuguese).
will not contain potassium and phosphorus. Thus, ultimately it will Dias, M.O.S., Cunha, M.P.C., Maciel Filho, R., Bonomi, A., Jesus, C.D.F., Rossell, C.,
lose its advantage as a fertilizer, both in the case of direct applica- Rossell, E.V., in press. Simulation of integrated rst and second generation
tion of the vinasse to the soil or when the solid residue after biogas bioethanol production from sugarcane: comparison between different biomass
pretreatment methods. J. Ind. Microbiol. Biotechnol.
production is used.
Fuentes, L.L.G., Rabelo, S.C., Maciel Filho, R., Costa, A.C., 2011. Kinetics of lime
Despite the low energy output from the production of biofuels pretreatment of sugarcane bagasse to enhance enzymatic hydrolysis. Appl.
such as ethanol, methane and burning lignin when compared to Biochem. Biotechnol. 163, 612625.
the incineration of biomass, there are advantages in the conversion Ghose, T.K., 1987. Measurement of cellulase activities. Pure Appl. Chem. 59, 257
268.
of biomass to biofuels, mainly related to environmental gains Ibrahim, M.N.M., Chuah, S.B., 2004. Characterization of lignin precipitated from the
resulting from greenhouse gas reduction. soda black liquor of oil palm empty fruit bunch bers by various mineral acids.
Brazil has a great interest in cellulosic ethanol, the aims being to AJSTD 21 (1), 5767.
Kaar, W.E., Holtzapple, M.T., 2000. Using lime pretreatment to facilitate the
making ethanol from sugarcane more competitive and expanding enzymatic hydrolysis of corn stover. Biomass Bioenerg. 18, 189199.
production without having to increase the area under sugarcane. Kaparaju, P., Serrano, M., Thomsen, A.B., Kongjan, P., Angelidaki, I., 2009. Bioethanol,
By combining rst and second generations, more ethanol is pro- biohydrogen and biogas production from wheat straw in a biorenery concept.
Bioresour. Technol. 100, 25622568.
duced per planted area. Studies conducted indicate that a distillery Kreuger, E., Sipos, B., Zacchi, G., Svensson, S.E., Bjrnsson, L., 2011. Bioconversion of
that today produces 1 million liters of ethanol per day using sugar- industrial hemp to ethanol and methane: the benets of steam pretreatment
cane juice could, with lignocellulose biorenery technology, gener- and co-production. Bioresour. Technol. 102, 34573465.
Lamo, P., 1991. Methane-producing system through industrial wastewater
ate an additional 150,000 L of ethanol from bagasse. In 2025, with treatment. METHAX/BIOPAQ-CODISTIL, Piracicaba, SP (in Portuguese).
the technique improved, it could show an increase in production of Laser, M., Jin, H., Jayawardhana, K., Dale, B.E., Lynd, L.R., 2009. Projected mature
400,000 L, recovered from bagasse (Revista Pesquisa Fapesp, 2007). technology scenarios for conversion of cellulosic biomass to ethanol with
coproduction thermochemical fuels, power, and/or animal feed protein.
Sugarcane straw is another potential source for bioethanol produc-
Biofuels, Bioprod. Bioref. 3, 231246.
tion and, with the prohibition of burning, it has the potential for Lettinga, G., Haandel, A.C.V., 1993. Anaerobic digestion for energy production and
use as a source of cellulose. environmental protection. In: Johansson, T.B., Kelly, H., Reddy, A.K.N., Williams,
In the future, with many options for the construction of an ener- R.H. (Eds.), Renewable Energy Sources for Fuels and Electricity. Island Press,
Washington, pp. 817840.
getic matrix, bagasse could either be used for burning to generate Lin, S.H., Lo, C.C., 1997. Fenton process for treatment of desizing wastewater. Water
bioelectricity or for the production of cellulosic ethanol, and the Res. 31, 20502056.
decision will be the market, emphasizing the most protable Liu, D., Zeng, R., Angelidaki, I., 2006. Hydrogen and methane production from
household solid waste in the two-stage fermentation process. Water Res. 40
option. (11), 22302236.
Lu, Y., Lai, Q., Zhang, C., Zhao, H., Ma, K., Zhao, X., Chen, H., Liu, D., Xing, X.H., 2009.
Characteristics of hydrogen and methane production from cornstalks by an
augmented two- or three-stage anaerobic fermentation process. Bioresour.
Technol. 100, 28892895.
4. Conclusions Luo, G., Talebnia, F., Karakashev, D., Xie, L., Zhou, Q., Angelidaki, I., 2011. Enhanced
bioenergy recovery from rapeseed plant in a biorenery concept. Bioresour.
Technol. 102, 14331439.
The residues from second generation bioethanol production Margeot, A., Han-Hangerdal, B., Edlund, M., Slade, R., Monot, F., 2009. New
were used for biogas production and for heat generation using improvements for lignocellulosic ethanol. Curr. Opin. Biotechnol. 20, 372380.
two different types of pretreatment: alkaline hydrogen peroxide Mosier, N., Wyman, C., Dale, B., Elander, R., Lee, Y.Y., Holtzapple, M., Ladisch, M.,
2005. Features of promising technologies for pretreatment of lignocellulosic
and lime, and different solids concentrations.
biomass. Bioresour. Technol. 96, 673686.
Simulation of different scenarios has shown that the use of ba- Rabelo, S.C., Maciel Filho, R., Costa, A.C., 2008. A comparison between lime and
gasse for energy generation by incineration was more efcient, the alkaline hydrogen peroxide pretreatments of sugarcane bagasse for ethanol
energy recovery being about 1.6 times higher than for the best production. Appl. Biochem. Biotechnol. 144, 87100.
Revista Pesquisa FAPESP: Poltica de C & T, 2007. Cellulose ethanol: sugarcane
multiple fuels production scenario. However, this greater energy bagasse and straw are quoted to increase ethanol production. In: Ereno, D. (Ed.),
potential is realised only by the production of heat. Other energy 133rd ed. Brazil (in Portuguese).
 S.C. Rabelo et al. / Bioresource Technology 102 (2011) 78877895 7895

Rivera, E.C., Rabelo, S.C., Garcia, D.R., Maciel Filho, R., Costa, A.C., 2010. Enzymatic Talinli, I., Anderson, G.K., 1992. Interference of hydrogen peroxide on the standard
hydrolysis of sugarcane bagasse for bioethanol production: determining COD test. Water Res. 26, 107110.
optimal enzyme loading using neural networks. J. Chem. Technol. Biot. 85, UNICA (Sao Paulo Sugarcane Agroindustry Union), 2011. Projected sugarcane
983992. crushing for 2011/2012 harvest in South-Central Brazil set at 568.5 million tons.
Salomon, K.R., Lora, E.E.S., 2009. Estimate of the electric energy generating potential Available at <http://english.unica.com.br/noticias/show.asp?nwsCode=%7BB6
for different sources of biogas in Brazil. Biomass Bioenerg. 33, 11011107. F7067D-DF2A-4CFD-B1A2-8FFAA32E6D60%7D>. Last accessed: 01-04-2011.
Sassner, P., Martensson, C.G., Galbe, M., Zacchi, G., 2008. Steam pretreatment of Van Soest, P.J., 1963. Use of detergent in the analysis of brous feeds. A rapid
H2SO4-impregnated salix for the production of bioethanol. Bioresour. Technol. method for the determination of bre and lignin. J. Assoc. Ofce Anal. Chem. 46
99, 137145. (5), 829835.
Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., Crocker, D. Wood, T.M., Bhat, K.M., 1988. Methods for measuring cellulase activities. In: Wood,
2008a. Determination of Structural Carbohydrates and Lignin in Biomass. W.A., Kellog, S.T. (Eds.), Methods in Enzymology, vol. 160. Academic Press, San
National Renewable Energy Laboratory, Midwest Research Institute, Golden, Diego, CA, pp. 81112.
CO. NREL/TP-510-42618. Wyman, C.E., Dale, B.E., Elander, R.T., Holtzapple, M., Ladisch, M.R., Lee, Y.Y., 2005.
Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D. 2008b. Comparative sugar recovery data from laboratory scale application of leading
Determination of Sugars, By-products, and Degradation Products in Liquid pretreatment technologies to corn stover. Bioresour. Technol. 96, 20262032.
Fraction Process Samples. National Renewable Energy Laboratory, Midwest
Research Institute, Golden, CO. NREL/TP-510-42623.