Enzymatic Hydrolysis of Chemithermomechanically Pretreated Sugarcane

Bagasse and Samples with Reduced Initial Lignin Content

Fernanda M. Mendes, Germano Siqueira, Walter Carvalho, André Ferraz, and Adriane M. F. Milagres
Dept. de Biotecnologia, Escola de Engenharia de Lorena, Universidade de São Paulo, 12602-810 Lorena, SP, Brasil

DOI 10.1002/btpr.553
Published online February 22, 2011 in Wiley Online Library (wileyonlinelibrary.com).

Chemithermomechanical (CTM) processing was used to pretreat sugarcane bagasse with

the aim of increasing cell wall accessibility to hydrolytic enzymes. Yields of the pretreated
samples were in the range of 75–94%. Disk refining and alkaline-CTM and alkaline/sulfite-
CTM pretreatments yielded pretreated materials with 21.7, 17.8, and 15.3% of lignin,
respectively. Hemicellulose content was also decreased to some extent. Fibers of the pre-
treated materials presented some external fibrillation, fiber curling, increased swelling, and
high water retention capacity. Cellulose conversion of the alkaline-CTM- and alkaline/sul-
fite-CTM-pretreated samples reached 50 and 85%, respectively, after 96 h of enzymatic hy-
drolysis. Two samples with low initial lignin content were also evaluated after the mildest
alkaline-CTM pretreatment. One sample was a partially delignified mill-processed bagasse.
The other was a sugarcane hybrid selected in a breeding program. Samples with lower ini-
tial lignin content were hydrolyzed considerably faster in the first 24 h of enzymatic diges-
tion. For example, enzymatic hydrolysis of the sample with the lowest initial lignin content
(14.2%) reached 64% cellulose conversion after only 24 h of hydrolysis when compared
with the 30% observed for the mill-processed bagasse containing an initial lignin content of
24.4%. V C 2011 American Institute of Chemical Engineers Biotechnol. Prog., 27: 395–401,

2011
Keywords: enzymatic hydrolysis, disk refining, breeding, pretreatment, sugarcane bagasse

Introduction CTM pretreatments of soft and hardwood chips with sul-

Methods that convert plant biomass into fermentable sug-
ars are generally composed of two main steps: a pretreat-
ment to degrade the plant structure and an enzymatic or
chemical hydrolysis step to convert the polysaccharides into
monomeric sugars. Each pretreatment technology has a dif-
ferent mechanism of action on the plant structure inducing
physical and/or chemical modifications, which are necessary
because the presence of hemicellulose and lignin in the cell
walls prevents cellulases from accessing inner parts of the
substrate.1,2
Most of the experimental pretreatment technologies
require the development of commercially proven equipment
to operate on a large scale.3 In contrast, the pulp and paper
industry has well-developed equipment designed for wood
and occasionally nonwood pulping.4 Among the processes
used by the pulp and paper industry, mild chemithermome-
chanical (CTM) pulping is supposed to be suitable for the
pretreatment of sugarcane bagasse. CTM pulping processes
are not capital intensive and give pulp yields in the range of
75–95%. Disk refining disintegrates fiber bundles and fibril-
lates fiber surfaces, which increases the total available sur-
face area of the fibers. When the process is combined with
the use of alkaline/sulfite, there is also significant swelling
of the fibers.4,5
Correspondence concerning this article should be addressed to A. M.
F. Milagres at adriane@debiq.eel.usp.br.
fite with the aim of subsequent enzymatic saccharification
were recently reported.3,6,7 The authors named the process
SPORL (sulfite pretreatment to overcome recalcitrance of
lignocellulose) because this pretreatment is based on acid/
sulfite precooking at 180
C for 30 min, followed by size
reduction using a typical disk refiner. Under such conditions,
a 90% successful enzymatic conversion of glucan was
obtained from the pretreated material, although a portion of
the original glucan contained in the wood chips was dis-
solved during pretreatment. Under more selective alkaline/
sulfite conditions, partial lignin and minimal hemicellulose
removal was observed and resulted in nonefficient enzymatic
hydrolysis of the pretreated biomass.
In mild CTM processes, only partial lignin and hemicel-
lulose removal is achieved. Therefore, it is possible to
hypothesize that plant materials with lower initial lignin
content would be especially suitable for this type of pre-
treatment. Several breeding or transgenesis programs
focused on providing plants with reduced lignin content
have been reported in the literature.8–11 In some cases,
reducing the original lignin content in alfalfa plants proved
to be an efficient way to increase its enzymatic digestibil-
ity.9 Moreover, improved digestibility in less lignified
immature grasses when compared with mature lignified
plants has previously been demonstrated.12 Additionally,
studies by Lam et al.8 with 150 lines of forage grass (Pha-
laris aquatica) and the internodes of 100 lines of ryegrass
(Lolium perenne) demonstrated that the hydroxycinnamic

C 2011 American Institute of Chemical Engineers

V 395
396
acid content was also important to explain the recalcitrance
of such grasses.
CTM pretreatment of sugarcane bagasse for subsequent
enzymatic hydrolysis has not yet been attempted. In this
work, mild alkaline-CTM and alkaline/sulfite-CTM pretreat-
ments of sugarcane bagasse were evaluated for subsequent
enzymatic hydrolysis of the pretreated material. Process vari-
ables were adjusted to avoid excessive biomass solubilization
during pretreatment to simulate a high-yield CTM process.
This type of high-yield CTM process can easily be per-
formed in single impregnation/refining machinery that is of-
ten available in mechanical pulping mills. The primary goal
was to evaluate to what extent lignin and hemicellulose
should be removed from the material to cause a significant
increase in bagasse digestibility.
Materials and Methods
Raw material and biomass preparation
Sugarcane bagasse was collected in a sugar and alcohol
mill after sugarcane processing for sucrose extraction. Gross
received biomass including the fiber bundles and pith13 was
termed mill-processed bagasse in this work.
One sample of the mill-processed bagasse was partially
delignified by sodium chlorite in aqueous acetic acid solu-
tion.14 Then, 600 g (dry weight) of bagasse was suspended
in 15 L of water and heated to 70
C. Sodium chlorite (180
g) and acetic acid (60 mL) were added to the bagasse sus-
pension. The mixture was then agitated with a glass rod for
10 min. To avoid fines (particles shorter than 0.2 mm)
losses, washing of the chlorite-treated material was per-
formed inside a 1-m long, 150-mm-diameter PVC column
set with a 200-mesh screen at the bottom. Fines passing
through the screen were pumped back to the column top. Fil-
trate recirculation permitted the formation of a fiber mat at
the column base that retained fines. Water recirculation was
stopped when wash water was free of turbidity. After this
point, additional bagasse biomass was applied to the column
and fresh water was passed through the biomass until the
wash reached a neutral pH. The bagasse was then washed
with 6 L of acetone.
An experimental sugarcane hybrid was also used in some
experiments. This plant was obtained from 286 clones in a
breeding program aimed at selecting plants with low lignin
contents.15 The 12-month-old plant was cut from field
experiments and passed through a cutter machine that
released 5- to 10-mm fragments. The cut material was
washed in a blender to remove residual sucrose. The
obtained biomass material was frozen and termed experimen-
tal hybrid in this work.
CTM pretreatment of sugarcane bagasse
Bagasse samples were treated in a two-step, laboratory-
scale process that simulates the industrial refining of CTM
pulping.16,17 Approximately 500 g of bagasse was impreg-
nated with alkaline or alkaline/sulfite liquors at a bagasse/liq-
uor ratio of 1:10 (w/v). Impregnation was performed by
applying a vacuum to the air-dried biomass contained in a
stainless steel reactor for 30 min. The liquor was then dis-
placed into the reactor, and an additional 15 min of vacuum
was applied. The alkaline liquor corresponded to a load of
5 g of NaOH per 100 g of bagasse, whereas the alkaline/sul-
Biotechnol. Prog., 2011, Vol. 27, No. 2
fite liquor corresponded to a load of 5 g of NaOH and 10 g
of Na2SO3 per 100 g of bagasse. Impregnated biomass was
cooked at 120
C for 2 h. Cooked biomass was washed with
tap water. To avoid fine loss, washing was performed inside
1-m long, 150-mm-diameter PVC columns as previously
described. The washed material was centrifuged to a final
consistency of 30% (w/w). Water released during the centrif-
ugation step was collected and used as diluting agent in the
subsequent disk refining process. Washed material was sus-
pended in water to a final volume of 25 L (2.0% consis-
tency) and refined in a Bauer MD-300 disk refiner with a
disk clearance of 0.1 mm (REGMED, Brazil) and up to 250
W h of energy consumption by the disk refiner. Refined sam-
ples were assayed for the fibrillation degree18 and further
centrifuged to a final consistency of 30%.
Chemical characterization of biomass samples
Chemical composition of the lignocellulosic samples was
determined in the ethanol-extracted material. Approximately
3 g of milled sample was extracted with 95% ethanol for 6 h
in a Soxhlet apparatus. Extracted samples were hydrolyzed
with 72% (w/w) sulfuric acid at 30
C for 1 h (300 mg of
sample and 3 mL of sulfuric acid). The acid was diluted to a
final concentration of 3% (addition of 79 mL of water), and
the mixture was heated at 121
C and 1 atm for 1 h. The
resulting material was cooled and filtered through a porous
glass filter number 3. The solids were dried to a constant
weight at 105
C, which was determined as the insoluble lig-
nin. The soluble lignin in the filtrate was read in a standard
UV cuvette (1-cm path length) at 205 nm. An absorptivity
(extinction coefficient) value of 105 L/g cm was used to cal-
culate the amount of acid-soluble lignin present in the hydro-
lyzate. The 205-nm absorptivities reported for most
lignocelluloses fall in the range of 88–113 L/g cm. The
value of 105 L/g cm used in this protocol represents an aver-
age of values found for different materials. Concentrations
of monomeric sugars in the soluble fraction were determined
by HPLC using a BIO-RAD HPX87H column at 45
C and
eluting at a rate of 0.6 mL/min with 5 mM sulfuric acid.
Sugars were detected with a temperature-controlled RI detec-
tor.19 Standard deviations from the analysis of triplicate
repeats were in the range of 3–5% of the average values.
Enzymatic hydrolysis
Enzymatic hydrolysis experiments used a mixture of com-
mercial enzyme preparations (Celluclast and Novozym 188,
Novozymes, Denmark) at a dosage of 8.8 filter paper units
(FPU) per gram of bagasse (d.w.) and 13.3 International
Units (IU) of b-glucosidase per gram of bagasse. Each hy-
drolysis experiment was carried out in 50-mL centrifuge
tubes containing 200 mg of lignocellulosic material (d.w.)
and 10 mL of 50 mM sodium acetate buffer at pH 5.0 plus
the enzyme solution (final consistency of 2%). The tubes
were incubated at 45
C under reciprocal agitation of 120
cycles per minute. The reaction was stopped at defined peri-
ods from 8 to 96 h by heating the reaction tube to 100
C for
5 min, followed by centrifugation at 7800g for 15 min. For
each reaction time, two replicate experiments were run.
Hydrolyzates were assayed for glucose, cellobiose, and
xylose content using the previously described HPLC proce-
dure. Cellulose conversion was calculated as (0.9  glucose
þ 0.95  cellobiose) released from glucan contained in the
Biotechnol. Prog., 2011, Vol. 27, No. 2 397

Figure 1. Light microscopy of fibers from sugarcane bagasse submitted to various pretreatments and enzymatic hydrolysis.
Sugarcane bagasse submitted to disk refining (a), alkaline-CTM (b), and alkaline/sulfite-CTM (c) pretreatments. (d–f) The condition of the residual
fibers after 96 h of enzymatic hydrolysis. Light microscopy of mill-processed bagasse partially delignified by sodium chlorite in aqueous acetic acid
solution (g) and a sugarcane bagasse hybrid (h) precooked with NaOH.

sample. Only minor amounts of cellobiose were detected in
the hydrolyzates. Hemicellulose conversion was calculated
as (0.88  xylose) released from xylan contained in the sam-
ple. Variation between hydrolysis replicates is shown as
error bars in the Results and Discussion section. Plots for the
hydrolysis process were based on the best fit assuming a
curved, time-course dependence for product formation as a
function of reaction time.20
Some selected samples were hydrolyzed with two subse-
quent enzyme loads in the reaction tubes. In this case, after
the initial load of 8.8 FPU of Celluclast and 13.3 IU b-glu-
cosidase per gram of bagasse, an equal amount of the
enzymes was added a second time to each tube after 48 h of
reaction.
mended procedure.22 Endoglucanase and xylanase activities
were assayed against carboxymethyl cellulose23 and xylan,24
respectively. b-Glucosidase and b-xylosidase activities were
determined by monitoring the hydrolysis of p-nitrophenyl-b-
23
D-glucopyranoside and p-nitrophenyl-b-xylopyranoside,25
respectively. Enzyme activities were expressed in IU.
Light microscopy analysis
Fibers were visualized by light microscopy using a Cole-
man M103 microscope. Fiber suspensions were placed
directly on microscope glass slides and visualized at 400
magnification. At least 10 fibers from each sample were
evaluated; the results presented in Figure 1 contain one fiber
that is representative of the average observations.

Protein and enzyme assays

Protein content of the samples was determined by the
Folin protein determination using BSA standards.21 Overall
cellulase activity of the samples was measured as filter paper
activity and expressed in FPU using the IUPAC recom-
Results and Discussion
CTM processing was used to pretreat sugarcane bagasse
with the aim of increasing cell wall accessibility to hydro-
lytic enzymes. In all pretreatments, the disk refining was set
398 Biotechnol. Prog., 2011, Vol. 27, No. 2

Table 1. Process Variables, Yield, Fibrillation Level, and Chemical Composition of Sugarcane Bagasse Pretreated in Alkaline- and
Alkaline/Sulfite-Chemithermomechanical Pretreatment
Yield of
Treated CSF after Bagasse Components Bagasse Components
NaOH Na2SO3 Material 250 W h (g/100 g of Original Bagasse) (% on Pulp Basis)
(g/100 g of (g/100 g (g/100 g Refining
Bagasse Sample Bagasse) of Bagasse) of Bagasse) (mL) Lignin Hemicellulose Glucan Lignin Hemicellulose Glucan
Mill-processed sugarcane bagasse
Untreated 0 0 100 nd 24.4 27.4 43.7 24.4 27.4 43.7
Refined 0 0 94.0 670 20.4 27.2 41.2 21.7 28.9 43.8
Precooked with NaOH 5 0 91.4 460 16.3 23.7 43.0 17.8 25.9 47
and refined
Precooked with NaOH/ 5 10 74.9 180 11.4 19.5 40.5 15.3 26.9 54.5
Na2SO3 and refined
Mill-processed sugarcane bagasse previously submitted to partial delignification
Untreated 0 0 100 nd 14.2 30.8 42.5 14.2 30.8 42.5
Precooked with NaOH 5 0 88.6 150 9.3 26.5 43.0 10.5 29.9 43.0
and refined
Experimental hybrid
Untreated 0 0 100 nd 19.1 27.0 42.0 19.1 27.0 42.0
Precooked with NaOH 5 0 78.6 340 12.8 22.6 37.3 16.3 28.8 47.5
and refined
CSF, Canadian Standard Freeness (mL).

laboratory samples with reduced initial lignin content were

Figure 2. Cellulose (a) and hemicellulose (b) conversion after
enzymatic hydrolysis of mill-processed sugarcane ba-
gasse submitted to various pretreatments.
The error bars represent variation between the two hydrolysis
replicates. When not visible, the error bars were smaller than
the symbol size. (-l-) untreated; (-^-) refined; (-h-) pre-
cooked with NaOH and refined; and (-*-) precooked with
NaOH/sulfite and refined.
to low energy consumption with the aim of merely preparing
a homogeneously fibrillated material suitable for subsequent
enzymatic hydrolysis. One mill-processed bagasse and two
evaluated.
Table 1 summarizes the process variables and chemical
compositions of the studied lignocellulosic materials. Mill-
processed bagasse (crude bagasse from the sugar and alcohol
industry) had a chemical composition of 24.4% lignin,
27.4% hemicellulose, and 43.7% cellulose, which is in
agreement with the data reported for several sugarcane ba-
gasse samples.13,26,27 After various CTM treatments, the
yields of pretreated mill-processed bagasse were in the same
range as those observed in similar industrial pulping proc-
esses and varied between 75 and 94%.5 In addition to chemi-
cal compositions of pretreated materials, Table 1 also
includes mass balances for each component. All procedures
adopted in this study avoided losses of insoluble materials
after pretreatment (see Materials and Methods section
describing filtration procedures designed to avoid losses of
the fine fraction). Therefore, mass losses observed during the
pretreatments relate to dissolution (or degradation to soluble
compounds) of any component. It is relevant because disso-
lution of the bagasse components could be related with po-
rosity increase in the cell walls, given that the overall
fibrillar structure was maintained in pretreated materials
(Figures 1a–c,g,h). Based on mass balance data, disk refining
and alkaline-CTM and alkaline/sulfite-CTM pretreatments
yielded lignin removal of 16, 33, and 53%, respectively.
Hemicellulose was also removed to some extent and demon-
strated removal of 0.7, 13, and 29%, respectively, for the
same samples. Fibers of the pretreated materials presented
some external fibrillation as illustrated in Figures 1a–c.
When alkaline or alkaline/sulfite cooking was used during
pretreatment, fiber curling and increased swelling were also
observed, suggesting the presence of more flexible fibers af-
ter these treatments (Figures 1b,c). These samples also pre-
sented low Canadian Standard Freeness values, especially
the alkaline/sulfite-pretreated sample, indicating increased
water retention capacity due to improved fibrillation and
hydrophilicity.4
Figure 2a shows the data for enzymatic hydrolysis of the
cellulose contained in the mill-processed bagasse samples
prepared with the various pretreatments. The low hydrolysis
levels observed for untreated mill-processed bagasse
Biotechnol. Prog., 2011, Vol. 27, No. 2
Table 2. Hydrolytic Activities in Commercial Enzyme Preparations
Enzyme Protein FPA (FPU/mg
Preparation (mg/mL) Protein)
Celluclast 1.5 L 177.2 0.52
Novozym 188 63.3 0 0.06
Figure 3. Correlation between lignin (-*-) and hemicellulose
(-~-) removal during the pretreatment of mill-proc-
essed sugarcane bagasse and the efficiency of enzy-
matic hydrolysis by Celluclast and Novozym.
Error bars represent variation between the two hydrolysis repli-
cates. When not visible, the error bars were smaller than the
symbol size.
confirmed that accessibility of the fiber cell walls by the
enzymes was very low and was similar to that observed by
Donohoe et al.28 Disk refining of uncooked material did not
improve enzyme accessibility as the hydrolysis of untreated
and refined bagasse samples resulted in only 13–15% cellu-
lose conversion after 96 h of digestion. However, it is rele-
vant to mention that disk refining efficiently provided a
homogenous material for subsequent enzymatic hydrolysis.
Additionally, it can be hypothesized that the industrial facili-
ties commonly used for CTM pulping could simplify the
development of a large-scale pretreatment into a single,
short-step process designed to provide high yields of pre-
treated bagasse. Removal of 33% of lignin and 13% of hem-
icellulose using alkaline precooking followed by refining
improved cellulose hydrolysis to 50%. This improvement
was even more pronounced when the alkaline-sulfite pre-
cooking was followed by refining, which increased cellulose
conversion to 85% after 96 h of hydrolysis. A time-course
evaluation of cellulose conversion indicates that for the alka-
line-CTM pretreatment most of the cellulose hydrolysis was
completed after 48 h of reaction. In contrast, cellulose hy-
drolysis continued at significant rates for up to 96 h of reac-
tion after alkaline/sulfite-CTM pretreatment. When
comparing these two pretreated substrates, fiber swelling was
greater (Figures 1b,c) and lignin removal was significantly
higher in the alkaline/sulfite-CTM pretreatment (Table 1).
Additionally, residual lignin in the alkaline/sulfite-CTM-pre-
treated material should contain sulfonic groups that confer
more hydrophilicity to the material.29 As a consequence, less
nonproductive cellulase adsorption should occur on this sul-
fonated residual lignin because cellulase adsorption is basi-
cally dependent on hydrophobic interactions.3,30,31 When
taken together, these data suggest that the digested material
becomes enriched in lignin and progressively more recalci-
trant. This characteristic would limit the ability of cellulose
hydrolysis to continue smoothly.32 It is well known that re-
sidual lignin on the surface of amorphous cellulose slows hy-
drolysis rates and decreases digestibility.33 On the other
399
Endoglucanase b-Glucosidase Xylanase b-Xylosidase
(IU/mg Protein) (IU/mg Protein) (IU/mg Protein) (IU/mg Protein)
29.5 0.27 6.2 0.3
17.9 0.55 0.12
hand, if hydrophobic interactions are one of the main causes
of nonproductive binding of cellulases,34 the lower lignin
content and increased hydrophilicity in the alkaline/sulfite-
CTM-pretreated material permitted enzymatic hydrolysis to
continue up to 96 h and reach 85% cellulose conversion.
Figures 1d–f illustrate how residual fibers appeared after
96 h of enzymatic hydrolysis. The occurrence of a large
number of microvoids in the remaining cell walls of the
alkaline/sulfite-CTM-pretreated material contrasted with the
small number of microvoids in the alkaline-CTM- and
refined-only pretreated materials, which corroborates the data
obtained from the chemical analysis of released sugars (Fig-
ure 2a). Hydrolysis of the hemicellulose fraction of the ba-
gasse samples by the enzymatic cocktail was also evaluated
(Figure 2b). The behavior of hemicellulose conversion was
similar to that already discussed for cellulose conversion;
however, in general, lower amounts of hemicellulose were
hydrolyzed. This hemicellulose hydrolysis was certainly a
result of the significant xylanase and b-xylosidase activities
detected in commercially available enzyme preparations
(Table 2).
The correlation between lignin and hemicellulose removal
during pretreatment and the efficiency of saccharification is
shown in Figure 3. The results would fit an S-curve, where
the efficiency of enzymatic hydrolysis increases with the re-
moval of lignin and hemicellulose during pretreatment. Simi-
lar results were already observed in other studies where the
lignin content was reduced either in transgenic alfalfa9 or by
selective delignification.35 Based on these data, it is clear
that high lignin and hemicellulose content in the substrate
prevents efficient enzymatic hydrolysis of cellulose contained
in the lignocellulosic materials. The apparent S-curve format
indicates that there is a critical level of lignin and hemicellu-
lose removal that would provide a significant increase in en-
zymatic hydrolysis efficiency. In the case of the currently
developed pretreatments for sugarcane bagasse, such critical
values were in the range of 18–25% for hemicellulose re-
moval and 40–50% for lignin removal.
It is known that the properties of lignocellulosic materials
are particularly affected by mild alkaline pretreatments. Al-
kali treatment increases the content of ionic groups and pro-
motes the cleavage of some interpolymeric crosslinks,
including lignin–carbohydrate linkages.36 The main chemical
reactions occurring in the hemicellulose component are
deacetylation and dissolution of low-molecular-weight frac-
tions. In lignin, the cleavage of a-ether linkages occurs. If
the pretreatment includes sulfite ions, sulfonation of lignin is
observed, which increases fiber swelling and improves lignin
dissolution.4,5 In the case of sugarcane bagasse and other
grasses, alkaline pretreatment also efficiently cleaves the
esters of hydroxycinnamic acids attached to either lignin or
hemicellulose hydroxyl groups.8,37,38
As shown in the former results, mild CTM processes
removed only part of the original lignin and hemicellulose.
To test the hypothesis that plants with lower original lignin
content would be suitable for these types of pretreatments,
two additional samples were evaluated. One sample was a
400
Figure 4. Cellulose (a) and hemicellulose (b) conversion after
enzymatic hydrolysis of three sugarcane samples
pretreated by alkaline-chemithermomechanical
process.
Unfilled symbols refer to experiments in which enzymes were
added a second time after 48 h of reaction. Error bars represent
variation between the two hydrolysis replicates. When not visi-
ble, the error bars were smaller than the symbol size. (-l-)
Mill-processed sugarcane bagasse precooked with NaOH and
refined; (-~-) experimental hybrid with reduced initial lignin
content precooked with NaOH and refined; and (-n-) mill-proc-
essed sugarcane bagasse previously subjected to partial
delignification.
partially delignified mill-processed bagasse. The other was
an experimental hybrid sugarcane selected in a previous
breeding program.15 The chemical compositions and the
responses of these samples to the alkaline-CTM pretreatment
are shown in Table 1. The mill-processed delignified bagasse
and the experimental hybrid presented lignin contents of
14.2 and 19.1%, respectively. These two experimental sam-
ples contained lower initial lignin content than the previously
evaluated mill-processed bagasse (24.4%). Therefore, the
mildest alkaline-CTM pretreatment was used in an attempt
to remove additional lignin and hemicellulose before enzy-
matic hydrolysis. Fibrillation, fiber curling, and swelling of
these pretreated materials were similar to those characteris-
tics observed for the mill-processed bagasse precooked with
NaOH (Figures 1g,h). Lignin and hemicellulose removal was
34 and 14% for the partially predelignified sample and 33
and 16%, respectively, for the experimental hybrid; those
Biotechnol. Prog., 2011, Vol. 27, No. 2
values were similar to the lignin and hemicellulose removal
observed for mill-processed bagasse. However, as the start-
ing materials contained less original lignin, the alkaline-
CTM-pretreated samples also displayed a lower final lignin
content. A small amount of cellulose loss (11%) was
observed after alkaline pretreatment of the experimental
hybrid, which differs from the results obtained with the mill-
processed bagasse and the partially predelignified bagasse.
These data suggest that the cellulose contained in the experi-
mental hybrid with lower original lignin content was more
susceptible to alkaline pretreatment, probably because of the
decreased protection of this component against the peeling
reactions.39 This hypothesis was also evidenced by the low
yield of pretreated material recovered from this sample.
Figure 4 shows the data for enzymatic hydrolysis of the
two samples with reduced original lignin content pretreated
by the CTM process when compared with those generated
from mill-processed bagasse. It was worth noting that the
samples with less lignin hydrolyzed considerably faster dur-
ing the first 24 h of enzymatic digestion. Enzymatic hydroly-
sis of the sample with the lower initial lignin content
(14.2%) reached 64% cellulose conversion after only 24 h of
hydrolysis, even when using the mild alkaline-CTM pretreat-
ment. This value was of the same order of magnitude as that
obtained after 96 h of hydrolysis of the mill-processed ba-
gasse (24.4% initial lignin content). Together, these data sug-
gest that breeding programs aimed at selecting plants with
reduced lignin content would favor the use of mild pretreat-
ments in the context of modern biorefineries. However, in
this study, the benefits of plants with reduced lignin content
became more significant when the lignin content was
reduced to approximately half of its original value.
The two samples with reduced lignin content presented a
cellulose and hemicellulose conversion plateau after 24–48 h
of enzymatic hydrolysis (Figure 4). This plateau has already
been discussed when comparing the different pretreatments
using mill-processed bagasse (Figure 2). Such behavior cor-
roborates the idea that after a short digestion period the re-
sidual substrate becomes enriched in lignin and becomes
progressively more recalcitrant and supports more nonpro-
ductive enzyme adsorption. To test these possibilities, Cellu-
clast and Novozym 188 were added again to these reactions
after 48 h of hydrolysis (unfilled points in Figure 4). Cellu-
lose conversion was not enhanced after the second addition
of enzyme to the reaction medium (Figure 4a), and hemicel-
lulose conversion was only slightly increased (Figure 4b).
Conclusions
The alkaline or alkaline/sulfite thermomechanical pretreat-
ments of sugarcane bagasse provided fibers altered in their
physical structures and chemical composition by partly
removing some cell wall components such as lignin and
hemicellulose. CTM processing provided increased fiber
swelling and high water retention capacity. These character-
istics permitted an enhancement in the enzymatic hydrolysis
of cellulose and hemicellulose. For the alkaline-CTM pre-
treatment, most of the cellulose hydrolysis was completed af-
ter 48 h of reaction. Even the samples with originally lower
initial lignin contents presented a similar plateau during en-
zymatic hydrolysis. Such behavior corroborates the idea that
after a short digestion period the residual substrate becomes
enriched in lignin and progressively more recalcitrant. When
the cellulolytic enzymes were reloaded in the reaction
Biotechnol. Prog., 2011, Vol. 27, No. 2
medium after 48 h of hydrolysis, the cellulose conversion
was not enhanced. In contrast, cellulose hydrolysis continued
at significant rates for up to 96 h of reaction after alkaline/
sulfite-CTM pretreatment, suggesting that the lower lignin
content and increased hydrophilicity in the alkaline/sulfite-
CTM-pretreated material seem to be the main causes of the
enhanced enzymatic hydrolysis.
Acknowledgments
The authors thank J.M. Silva and J.C. Tavares for their tech-
nical assistance. This work was supported by FAPESP (contract
number 08/56256-5), CNPq, and CAPES. F.M. Mendes and G.
Siqueira thank CNPq and CAPES for their student fellowships.
Literature Cited
1. Mooney CA, Mansfield SD, Touhy MG, Saddler JN. The effect
of initial pore volume and lignin content on the enzymatic hy-
drolysis of softwood. Bioresour Technol. 1998;64:113–119.
2. Várnai1 A, Siika-aho M, Viikari L. Restriction of the enzymatic
hydrolysis of steam pretreated spruce by lignin and hemicellu-
lose. Enzyme Microb Technol. 2009;46:185–193.
3. Zhu JY, Pan XJ, Wang GS, Gleisner R. Sulfite pretreatment
(SPORL) for robust enzymatic saccharification of spruce and
red pine. Bioresour Technol. 2009;100:2411–2418.
4. Biermann CJ. Essentials of Pulping and Papermaking. San
Diego: Academic Press; 1993.
5. Konn J, Pranovich A, Holmbom B. Dissolution of fibre material
in alkaline pre-treatment and refining of spruce CTMP. Holzfor-
schung. 2006;60:32–39.
6. Wang GS, Pan XJ, Zhu JY, Gleisner R, Rockwood D. Sulfite
pretreatment to overcome recalcitrance of lignocellulose
(SPORL) for robust enzymatic saccharification of hardwoods.
Biotechnol Prog. 2009;25:1086–1093.
7. Zhu W, Zhu JY, Gleisner R, Pan XJ. On energy consumption
for size-reduction and yields from subsequent enzymatic saccha-
rification of pretreated lodgepole pine. Bioresour Technol. 2010;
101:2782–2792.
8. Lam TB, Iiyama K, Stone BA. Hot alkali-labile linkages in the
walls of the forage grass Phalaris aquatica and Lolium perenne
and their relation to in vitro wall digestibility. Phytochemistry.
2003;64:603–607.
9. Chen F, Dixon RA. Lignin modification improves fermentable
sugar yields for biofuel production. Nat. Biotechnol. 2007;25:
759–761.
10. Vermerris W, Saballos A, Ejeta G, Mosier NS, Ladisch MR,
Carpita NC. Molecular breeding to enhance ethanol production
from corn and sorghum stover. Crop Sci. 2007;47:S142–S153.
11. Jackson LA, Shadle GL, Zhou R, Nakashima J, Chen F, Dixon
RA. Improving saccharification efficiency of alfalfa stems
through modification of the terminal stages of monolignol bio-
synthesis. Bioenerg Res. 2008;1:180–192.
12. Grabber JH, Panciera MT, Hatfield RD. Chemical composition
and enzymatic degradability of xylem and nonxylem walls iso-
lated from alfalfa internodes. J Agric Food Chem. 2002;50:
2595–2600.
13. Sanjuan R, Anzaldo J, Vargas J, Turrado J, Patt R. Morphologi-
cal and chemical composition of pitch and fibers from Mexican
sugarcane bagasse. Holz Roh Werkst. 2001;59:447–450.
14. Browing B. Methods of Wood Chemistry. New York: Wiley;
1967.
15. Barbosa MHP, Resende MDV, Peternelli LA, Bressiani JA, Sil-
veira LCI, Silva FL, Figueiredo ICR. Use of REML/BLUP for
the selection of sugarcane families specialized in biomass pro-
duction. Crop Breed Appl Biotechnol. 2004;4:218–226.
16. Ruzinsky F, Kokta BV. High-yield pulping of switchgrass using
the Na2SO3-NaHCO3 system. Cell Chem Technol. 2000;34:299–
315.
401
17. Vicentim MP, Ferraz A. Alkaline-sulfite chemithermomechani-
cal pulping of Eucalyptus grandis biotreated by Ceriporiopsis
subvermispora under varied culture conditions. Holzforschung.
2008;62:375–382.
18. Tappi Test Methods. Laboratory Beating of Pulp (PFI Mill
Method-TAPPI T248). Atlanta, GA: TAPPI 2000–2001.
19. Ferraz A, Baeza J, Rodriguez J, Freer J. Estimating chemical
composition of biodegraded pine and eucalyptus by DRIFT
spectroscopy and multivariate analysis. Bioresour Technol.
2000;74:201–212.
20. Engel PC. Enzyme kinetics. In: Engel PC, editor. Enzymology
Labfax. San Diego: Academic Press; 1996:77–113.
21. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein mea-
surement with the folin phenol reagent. J Biol Chem. 1951;193:
265–275.
22. Ghose T. Measurement of cellulase activities. Pure Appl Chem.
1987;59:257–268.
23. Wood TM, Bhat KM. Methods for measuring cellulase activ-
ities. In: Wood WA, Kellogg ST, editors. Methods in Enzymol-
ogy, Vol. 160. London: Academic Press; 1988:87–112.
24. Bailey MJ, Nevalainen KMH. Induction, isolation and testing of
stable Trichoderma reesei mutants with improved production of
solubilising cellulose. Enzyme Microb Technol. 1981;3:153–157.
25. Tanaka M, Taniguchi M, Matsumo R, Kamikubo T. Purification
and properties of cellulases from Eupenicillium javanicum.
J Ferment Technol. 1981;59:177–183.
26. Gámez S, Gonzalez-Cabriales JJ, Ramı́rez JÁ, Garrote G, Vaz-
quez M. Study of the hydrolysis of sugarcane bagasse using
phosphoric acid. J Food Eng. 2006;74:78–88.
27. Pessoa A Jr, Mancilha IM, Sato S. Acid hydrolysis of hemicellu-
lose from sugarcane bagasse. Braz J Chem Eng. 1997;14:309–312.
28. Donohoe BS, Selig MJ, Viamajala S, Vinzant TB, Adney WS,
Himmel ME. Detecting cellulase penetration into corn stover
cell walls by immuno-electron microscopy. Biotechnol Bioeng.
2009;103:480–489.
29. Sjostrom E. Wood Chemistry, 2nd ed. San Diego, CA: Aca-
demic Press; 1993.
30. Ramos LP, Breuil C, Saddler JN. Comparison of steam pretreat-
ment of eucalyptus, aspen and spruce wood chips and their enzy-
matic hydrolysis. Appl Biochem Biotechnol. 1992;34/35:37–47.
31. Lu YP, Yang B, Gregg D, Saddler JN, Mansfield SD. Cellulase
adsorption and an evaluation of enzyme recycle during hydroly-
sis of steam-exploded softwood residues. Appl Biochem Biotech-
nol. 2002;98:641–654.
32. Yang B, Wyman CE. BSA treatment to enhance enzymatic hy-
drolysis of cellulose in lignin containing substrates. Biotechnol
Bioeng. 2006;94:611–617.
33. Zhang Y-HP, Ding S-Y, Mielenz JR, Cui J-B, Elander RT,
Laser M, Himmel ME, McMillen JR, Lynd LR. Fractionating
recalcitrance lignocellulose at modest reaction conditions. Bio-
technol Bioeng. 2007;97:214–223.
34. Nakagame S, Chandra RP, Saddler JN. The effect of isolated
lignin, obtained from a range of pretreated lignocellulosic sub-
strates, on enzymatic hydrolysis. Biotechnol Bioeng. 2010;105:
871–879.
35. Lee SH, Doherty TV, Linhardt RJ, Dordick JS. Ionic liquid-
mediated selective extraction of lignin from wood leading to
enhanced enzymatic cellulose hydrolysis. Biotechnol Bioeng.
2009;102:1369–1376.
36. Chen H-T, Funaoka M, Lai Y-Z. Characteristics of bagasse lig-
nin in situ and in alkaline delignification. Holzforschung. 1998;
52:635–639.
37. Tilay A, Bule M, Kishenkumar J, Annapure U. Preparation of
ferulic acid from agricultural wastes: its improved extraction
and purification. J Agric Food Chem. 2008;56:7644–7648.
38. Mousavioun P, Doherty WOS. Chemical and thermal properties
of fractionated bagasse soda lignin. Ind Crop Prod. 2010;31:52–
58.
39. Moigne N, Jardeby K, Navard P. Structural changes and alka-
line solubility of wood cellulose fibers after enzymatic peeling
treatment. Carbohydr Polym. 2010;79:325–332.
Manuscript received July 6, 2010, and revision received Oct. 21, 2010.