Lorencini 2016
Lorencini 2016
DOI 10.1007/s12649-016-9494-4
ORIGINAL PAPER
Abstract This work determined the optimal conditions to 1423 mL of H2 L-1 d-1 for liq ? C and sol ? E, respec-
pretreat sucargane bagasse with HCl by using the liquid tively—are the highest ever reported in the literature for H2
and the solid fractions resulting from the bagasse pretrea- production from sugarcane bagasse by a microbial
ment as substrate for fermentative hydrogen production by consortium.
a mixed culture. A 23 full factorial central composite
design (star configuration) helped to determine how tem- Keywords Sugarcane bagasse Hydrochloric acid
perature, time, and acid concentration affected the total Pretreatment Fermentation Biohydrogen
monosaccharides (TM), total reducing sugars (TRS), and
total inhibitors (TI) concentrations in the liquid fraction.
Temperature, time, and acid concentration impacted the Introduction
TRS and TM concentrations, but these variables did not
influence the TI concentration significantly. The optimal The annual Brazilian sugarcane production is ca. 716
pretreatment conditions were HCl at 7.36 % (v/v), 96.8 °C, million tons [1]. This amount corresponds to ca. 43 % of
and 441.6 min, which afforded the highest TRS concen- the global output, which makes Brazil the major sugarcane
tration in the liquid hydrolysates. The liquid fraction world producer. Every ton of sugarcane used to obtain
obtained from the bagasse pretreated with acid under the sucrose and ethanol generates about 0.3 ton of bagasse, and
optimal conditions (designated liq) was not suitable for H2 burning of this residue produces electricity. Unfortunately,
production by the mixed culture before treatment of the bagasse burning releases high quantities of carbon dioxide
fraction with activated carbon. The solid residual bagasse and particulate material into the atmosphere [2], which
(designated sol) alone afforded 6.0 mL of H2/g of bagasse. makes environmentally friendly strategies to treat sugar-
Liq treated with 10 % (m/v) activated carbon, to give cane bagasse highly desirable.
liq ? C, and sol added with the enzyme CelluclastÒ 10 Hydrogen gas constitutes an attractive sustainable fuel
U g-1, to afford sol ? E, yielded 45.3 and 7.8 mL of H2/g because its combustion produces water only. Besides, H2 has
of bagasse respectively, which amounted to 53.1 mL of H2/ high energy density (about 144 MJ kg-1) and is an excellent
g of bagasse. The volumetric productivities—1450 and energy carrier. Although H2 production currently relies on
fossil fuel-based methods, this fuel can also originate from
& Valeria Reginatto
fermentation of carbohydrate-rich materials. In this context,
[email protected] H2 production from low-cost waste such as lignocellulosic
materials like sugarcane bagasse could become a sustainable
1
Departamento de Quı́mica, Faculdade de Filosofia, Ciências e and economically viable process [3–8].
Letras de Ribeirão Preto, Universidade de São Paulo,
Sugarcane bagasse is a carbohydrate-rich waste that
14040-901 Ribeirão Preto, SP, Brazil
2
consists of ca. 40 % cellulose and 35 % hemicellulose; it
Institute of Chemistry, Center for Monitoring and Research
also contains 15 % lignin, a non-carbohydrate constituent
of the Quality of Fuels, Biofuels, Crude Oils and Derivatives
– CEMPEQC, UNESP - São Paulo State University, Prof. [9]. This composition has motivated the recent use of
Francisco Degni Street 55, 14800-900 Araraquara, SP, Brazil sugarcane bagasse as a lignocellulosic substrate for
123
Waste Biomass Valor
123
Waste Biomass Valor
Table 1 Central composite design matrix showing the code and real g L-1 in the hydrolysates obtained from the pretreatment of
values of the variables temperature (T), time (t), and acid concen- sugarcane bagasse under the different conditions of the experimental
tration (A) as well as the responses for total monosaccharides (TM), design
total reducing sugars (TRS), and total inhibitors (TI) concentrations in
Run T (°C) t (min) A (%) Glu Xyl Ara TM TRS Ac F HMF TI
1 70.0 (-1) 120.0 (-1) 2.0 (-1) 0.02 0.04 0.46 0.52 0.90 0.63 – – 0.63
2 70.0 (-1) 120.0 (-1) 6.0 (?1) 0.05 2.62 0.51 3.18 5.81 0.89 – – 0.89
3 70.0 (-1) 360.0 (?1) 2.0 (-1) 0.04 1.31 0.57 1.92 2.98 0.91 – – 0.91
4 70.0 (-1) 360.0 (?1) 6.0 (?1) 0.20 4.93 0.72 5.85 9.57 1.00 0.008 – 1.01
5 90.0 (?1) 120.0 (-1) 2.0 (-1) 0.14 5.01 0.48 5.63 8.89 1.19 0.004 – 1.19
6 90.0 (?1) 120.0 (-1) 6.0 (?1) 0.65 5.55 0.66 6.86 9.76 0.94 0.006 – 0.95
7 90.0 (?1) 360.0 (?1) 2.0 (-1) 0.59 5.72 0.70 7.01 12.09 0.98 0.04 – 1.02
8 90.0 (?1) 360.0 (?1) 6.0 (?1) 1.22 6.07 0.71 8.00 13.88 0.97 0.26 0.0004 1.23
9 63.2 (-1.68) 240.0 (0) 4.0 (0) 0.04 0.73 0.56 1.33 8.22 0.90 – – 0.90
10 96.8 (?1.68) 240.0 (0) 4.0 (0) 0.90 6.05 0.71 7.66 13.52 1.01 0.21 0.02 1.24
11 80.0 (0) 38.4 (-1,68) 4.0 (0) 0.04 0.72 0.84 1.60 7.29 0.81 – – 0.81
12 80.0 (0) 441.6 (?1.68) 4.0 (0) 0.57 5.37 0.72 6.66 10.15 1.02 0.04 – 1.06
13 80.0 (0) 240.0 (0) 0.64 (-1.68) 0.03 0.43 0.55 1.01 5.46 0.78 – – 0.78
14 80.0 (0) 240.0 (0) 7.36 (?1.68) 0.73 5.76 0.68 7.17 13.65 0.99 0.07 0.05 1.11
15 80.0 (0) 240.0 (0) 4.0 (0) 0.35 5.37 0.74 6.46 9.57 1.05 0.01 – 1.06
16 80.0 (0) 240.0 (0) 4.0 (0) 0.43 5.35 0.74 6.52 9.83 1.03 0.02 – 1.05
17 80.0 (0) 240.0 (0) 4.0 (0) 0.37 5.23 0.71 6.31 9.34 1.02 0.02 – 1.04
Glu glucose, Xyl xylose, Ara arabinose, TM Glu ? Xyl ? Ara concentrations, Ac acetic acid, F furfural, HMF hydroxymethylfurfural, TI
Ac ? F ? HMF concentrations
Optimization of Sugarcane Bagasse Acid independent variables temperature, time, and acid con-
Pretreatment centration, respectively.
After attainment of the surface-response results, multi-
Surface-response methodology (SRM) and multi-response response analysis helped to optimize the process conditions
analysis [16] helped to optimize the experimental condi- [16]. This method transformed response variables (Yi) into
tions (temperature, time, and acid concentration) that led to an individual function of dimensionless desirability (gi)
the highest total reducing sugars (TRS) concentration (Eq. 2) that ranged from 0 (undesirable response) to 1
possible. A 23 full factorial central composite design (star (desired response). The geometric means of individual
configuration) aided in this purpose (Table 1). SRM also desires furnished the overall desirability function (G)
enabled determination of the effect of temperature (T), (Eq. 3). The software Mathematic 5.0 helped to maximize
time, and acid concentration on TRS. G.
The analysis of variance (ANOVA), the multiple com- Yi Ymin
parison test, and all the statistical analyses were performed gi ¼ ; ð2Þ
Ymax Ymin
with the aid of the software Statistica 6.0. The data were fit
to a second-order equation (Eq. 1) as a function of the G ¼ ðgn11 þ gn12 þ . . .. . . þ gn1k Þ1=K ; ð3Þ
independent variables. where Ymin and Ymax are the response minimum and
X
3 X
3 X X
3 maximum values, respectively; k is the number of con-
Yi ¼ bo þ bj Xj þ bjj Xj2 þ bjk Xj Xk sidered responses; and n is the weight of each response.
j¼1 j¼1 j\k¼2 Finally, sugarcane bagasse liquid hydrolysates prepared
Yi ¼ bo þ b1 X1 þ b2 X2 þ b3 X3 þ b12 X1 X2 þ b13 X1 X3 in the optimal conditions helped to validate the optimized
process conditions obtained by multi-response analysis.
þ b23 X2 X3 þ b11 X12 þ b22 X22 þ b33 X32 ð1Þ
The validation experiments were performed in triplicate,
where bn corresponds to constant regression coefficients; and the resulting hydrolysates were characterized with
Yi refers to the dependent variables (TRS, TM, and TI respect to TRS, TM, and TI (the sum of acetic acid, fur-
concentrations); and X1, X2, and X3 are the coded fural, and HMF concentrations) concentrations.
123
Waste Biomass Valor
Fermentative Assays for Hydrogen Production temperature. Stirring was conducted for 2 h, which was
followed by filtration in Whatmann paper filter under
Inoculum and Culture Medium vaccum to detoxify the liquid hydrolysate for use as sub-
strate in the fermentative assays [19].
The inoculum consisted of a mixed culture (sludge) collected The Filter Paper (FP) activity of the enzyme was mea-
from an upflow anaerobic sludge blanket (UASB) reactor used sured according to Ghose [20] before addition of the
to treat the effluent from a sugar and ethanol mill near the city commercial enzyme Celluclast Ò (Novozymes, USA) to the
of Ribeirão Preto-SP, Brazil. The sludge was maintained in residual solid sugarcane bagasse pretreated with acid in the
our laboratory by feeding with glucose (10 g L-1, as carbon optimal conditions (sol). One enzyme unit (1 U) was
source) and nutrient solution. The latter solution, adapted from defined as the amount of enzyme that released 1 lmol of
Gonzalez-Gil et al. [17], consisted of NH4Cl (0.11 g L-1), TRS per min. The measured FP activity of Celluclast Ò was
MgSO47H2O (0.1 g L-1), KH2PO4 (0.136 g L-1), and 63.2 U mL-1. Thus, 0.52 mL of the enzyme was added in
Na2HPO4 (0.148 g L-1) as macronutrients plus 1 mL L-1 of the fermentation flasks containing 3.3 g of the substrate S,
the trace elements FeCl2.4H2O (2.0 mg L-1), H3BO3 which gave 10 U of enzyme/g of acid-treated bagasse (sol).
(50.0 mg L-1), ZnCl2 (50.0 mg L-1), CuCl2.2H2O
(38.0 mg L-1), MnCl24H2O (500.0 mg L-1) (NH4)6- Fermentative Batch Assays
Mo7O244H2O (50.0 mg L-1), AlCl36H2O (90.0 mg L-1),
CoCl26H2O (2.0 mg L-1), NiCl26H2O (142.0 mg L-1), H2 production batch tests were carried out in triplicate, in
Na2SeO5H2O (164.0 mg L-1), EDTA (1.0 mg L-1), and 125-mL bioreactors containing 100 mL of one of the
HCl 36 % (1.0 mg L-1). All the chemicals were analytical substrates (Glu, sol, sol ? E, liq, or liq ? C) and nutrient
grade. solution. The dried sludge (3.3 g) containing 45 % of VS
Drying of the sludge at 105 °C for 12 h according to a was also added to the bioreactors. The initial pH of the
procedure adapted from Buitrón and Carvajal [18] ensured assays was adjusted to 6.0 with NaOH 50 % (w/v), if
its enrichment with H2-producing bacteria. The volatile necessary. Argon gas (0.1 m3 min-1) was bubbled through
solids (VS) content of the dry sludge was analyzed before the bioreactor for 2 min, to ensure anaerobic conditions.
its use in fermentative assays. The bioreactors were operated at 35 °C and 100 rpm in
a shaker (Dubnoff 304D, Nova Ética, Brasil). The shaker
Substrates was coupled to pipes that led to a gas measurement system
consisting of an inverted flask containing NaOH 5 % (w/v)
The following substrates were used for H2 production solution and a flask. This arrangement helped to determine
during batch fermentative assays: 3.3 g of the bagasse in the volume that the produced gas displaced (Fig. 1). The
natura in 100 mL of distilled water (B); solution of glucose volume and composition of the generated biogas were
at the same concentration as the concentration of TRS in monitored during the tests. Gas chromatography revealed
the optimal hydrolysate (ca. 20 g L-1) (Glu); liquid the composition of the gas, as will be described below.
hydrolysate (liq) from 3.3 g of sugarcane bagasse treated Glucose, xylose, arabinose, acetic acid, furfural, HMF,
with acid in the optimal pretreatment conditions; 3.3 g of and the soluble metabolites from fermentation (acetic acid,
solid sugarcane bagasse that remained from the acid pre- butyric acid, lactic acid, and ethanol) were quantified at the
treatment in the optimal conditions (sol) (which corre- beginning and end of the fermentation assays.
sponded to 4.62 g of dried sugarcane bagasse in natura) in
100 mL of distilled water; 3.3 g of sugarcane bagasse Estimation of H2 Production Kinetic Parameters
pretreated with acid in the optimal conditions (which cor-
responded to 4.62 g of dried bagasse without pretreatment) The modified Gompertz equation (Eq. 4) provided the
plus enzyme Celluclast Ò 10 U g-1 in 100 mL of distilled kinetic data on biohydrogen production from the different
water (sol ? E); bagasse in natura plus the enzyme Cel- substrates and control. The H2 volume accumulated along
luclast Ò 10 U g-1 in 100 mL of distilled water (B ? E); the assay was introduced in the program Statistica 7 and
and liquid hydrolysate from 3.3 g of sugarcane bagasse modeled according to Eq. 4, to give the parameters Rm,
pretreated with acid in the optimal conditions detoxified Hmax, and k.
with activated carbon (liq ? C). All the experiments were
Rm :e
supplemented with macronutrients plus 1 mL L-1 of the H ¼ Hmax : exp exp ðtÞ þ 1 ð4Þ
Hmax
trace elements as described above in the culture medium
composition. where H = cumulative H2 volume in tests; Hmax = maxi-
The activated carbon 10 % (w/v) was added to the liquid mum potential from H2 production (mL); Rm = Maximum
hydrolysate (liq ? C) under stirring at 200 rpm and room H2 production rate (mL h-1); k = lag phase or the time
123
Waste Biomass Valor
elapsed before H2 production started (h), and t = duration Results and Discussion
of tests (h).
Optimization of Pretreatment
Analytical Methods
Table 1 lists the TRS, TM (sum of the concentrations of
Analysis of cellulose, hemicellulose, and lignin contents in glucose, xylose, and arabinose), and TI (sum of the con-
bagasse in natura (B) and in bagasse after acid pretreat- centrations of acetic acid, HMF, and furfural) concentra-
ment under optimized conditions (sol) was performed tions determined in the liquid hydrolysates obtained from
according to [22, 23] and to a modification from [24]. The the pretreatment of sugarcane bagasse under the different
hemicellulose content was obtained on the basis of the conditions of the experimental design.
results achieved for holocellulose (cellulose ? hemicellu- Xylose and acetic acid were the main monosaccharide
lose). All the analyses were accomplished in triplicate. and inhibitor in the hydrolysates, respectively, which evi-
The VS concentration in the sludge was assayed denced higher hemicellulose hydrolysis. Furfural, a product
according to the Standard Methods for Examination of of pentose degradation [13, 28], emerged at low concen-
Water and Wastewater [25]. tration, but it did not arise under the mildest experimental
With the aid of spectrophotometry, the TRS concentra- conditions (assays 1, 2, 3, 9, 11, and 13).
tions in the hydrolysate and at the beginning and at the end Glucose occurred at low concentrations (between 0.02
of fermentations were analyzed by the 3.5-dininitrosalyci- and 1.22 g L-1) in the hydrolysates, which indicated poor
lyc acid (DNS) method described by Miller [26]. High cellulose hydrolysis in the experimental conditions.
performance liquid chromatography (HPLC) provided the Because HMF originated from glucose dehydration, it also
concentration of glucose, xylose, and arabinose as well as emerged at very low concentration. According to Hendriks
the concentration of fermentation inhibitors such as acetic and Zeeman [30], treatment of lignocellulosic materials
acid, furfural, and HMF in the hydrolysates. The concen- with acid solubilizes hemicellulose, making cellulose more
tration of the soluble metabolites acetic acid, butyric acid, accessible for further enzymatic hydrolysis. Hence, the low
lactic acid, and ethanol in the samples kept in the biore- concentration of glucose and its degradation derivative
actors were also analyzed by HPLC before the start and at (HMF) in the hydrolysates does not mean that the cellulose
the end of the kinetic assays. An Aminex HPX-87H col- structure remained intact.
umn was used at 55 °C; the mobile phase was H2SO4 According to the results displayed in Table 1, pretreat-
0.005 mol L-1 at a flow of 1.0 mL min-1. All the samples ment with HCl at 6 %, at 90 °C, for 360 min (run 8)
were filtered through 0.45-lm acetate-cellulose membranes afforded the highest TRS concentration. Statistical data
123
Waste Biomass Valor
analysis by SRM helped to fit these data to a second-order According to Fig. 2b, temperatures between 63.2 and
model. Table 2 depicts the analysis of variance (ANOVA) 75 °C and acid concentrations between 0.64 and 3 %
and the regression coefficients of the second-order poly- afforded hydrolysates with the lowest TRS concentration.
nomials for TM, TRS, and TI concentrations in the Higher temperatures and acid concentrations increased
hydrolysates. ANOVA showed that only the models fit for TRS in the hydrolysates. Temperatures between 90 and
TM and TRS were statistically significant (p \ 0.05), 96.8 °C and acid concentrations between 6 and 7.36 %
because Fcalculated values [ Flisted. These models were also produced hydrolysates with the highest TRS.
predictive—their F ratio (Fcalculated/Flisted) was higher than Both acid concentration and time (Fig. 2c) significantly
the corresponding Flisted. In these models, only the linear influenced the TRS concentration. Acid concentrations
parameters of temperature, time, and acid concentration ranging from 6 to 7.36 % and time from 360 to 441.6 min
were statistically significant for TM and TRS (p \ 0.05). elicited the highest TRS.
The response-surface curves for TRS were generated for The models calculated for TRS concentration (Table 2)
further fermentation (Fig. 2). furnished the desirability function (G). The minimum and
Figure 2a–c present the surface-response curves for TRS maximum values of each response variable derived from
concentration as a function of temperature, time, and acid the experimental results obtained in the experimental
concentration. Temperature affected the TRS concentration design (Table 1) enabled determination of the gi function.
in the hydrolysate more than time (Fig. 2a). The effect of The optimized conditions afforded hydrolysates with
time was only important at low temperatures, but it became higher TRS. The process conditions that led to the maxi-
less pronounced at intermediate temperatures. Tempera- mum global desirability of the G function were 96.8 °C,
tures and times ranging between 90 and 96.8 °C and 441.6 min, and acid at 7.36 %.
between 320 and 441.6 min, respectively, gave a higher Table 3 compares the predicted and the experimental
TRS. data obtained from triplicate assays that helped to validate
the process conditions optimized by the multi-response
analysis. The relative deviation values revealed that the
predicted and experimental data correlated well.
Table 2 Regression coefficients and analysis of variance for TM,
TRS, and TI concentrations in the hydrolysates obtained from the
In this work, the hydrolysate composition obtained
pretreatment of sugarcane bagasse under the different conditions of under the optimized conditions was as follows: monosac-
the experimental design charides—xylose = 7.43 g L-1, glucose = 3.75 g L-1,
Coefficients TM TRS TI and arabinose = 2.73 g L-1; inhibitors—fur-
Y1 Y2 Y3 fural = 0.08 g L-1, HMF = 0.009 g L-1, and acetic
acid = 2.3 g L-1. The TM concentration was similar to
bo 4.92* 8.88* 1.13*
TM concentration values obtained for other bagasse
Linear
hydrolysates submitted to treatment with H2SO4 at the
b1 1.95* 2.51* 0.38*
same bagasse/acid solution ratio and final volume used
b2 1.10* 1.31* 0.095* herein. For example, Rai et al. [31] hydrolyzed sugarcane
b3 1.40* 2.04* 0.70* bagasse with H2SO4 for dark H2 production integrated with
Quadratic H2 production via photofermentation by pure cultures of
b11 -0.51 0.07 0.095* Enterobacter aerogenes MTCC 2822 and Rhodopseu-
b22 -0.64 -0.69 -0.30* domonas BHU 01, respectively. Sugarcane bagasse pre-
b33 -0.66 -0.39 0.58* treatment at 2 % H2SO4 (v/v) and at 121 °C, for 60 min,
Interactions yielded a hydrolysate that contained glucose (3.41 g L-1),
b12 -0.19 0.18 0.009 xylose (8.36 g L-1), and arabinose (0.55 g L-1). However,
b13 -0.55 -1.10 0.005 authors observed lower inhibitors concentrations than
b23 0.13 0.32 0.092* ours—acetic acid and furfural concentrations were 1.49
R2 0.85 0.81 0.55 and 0.19 g L-1, respectively.
Facalculated 25.40 18.15 1.57 Table 4 presents the lignocelluloses composition of the
Fblisted 3.41 3.41 3.29 bagasse before and after acid pretreatment under the opti-
Fcalculated/Flisted 7.44 5.32 0.48 mized conditions.
* Significant at 5 % level Acid pretreatment diminished the hemicellulose content
a
Ratio between the mean square of regression and the mean square
the most: from 26.92 to 1.31 %, which corresponded to
of residuals total removal of 95 %. The cellulose content reduced by
b
Tabulated for Fisher test using the significance level and degrees of 22 %, from 46.7 to 36.35 %. The soluble lignin content
freedom decreased from 0.1 to 0.03 %, but the insoluble lignin
123
Waste Biomass Valor
Fig. 2 Surface-response of TRS concentration as a function of a temperature versus time, b temperature versus acid concentration, and c time
versus acid concentration
Table 3 Predicted and experimental responses of TRS and TM promising process to hydrolyze hemicelluloses to sugars in
concentrations before and after acid pretreatment under the optimized high yields, and therefore to change the lignin structure and
conditions increase the cellulosic surface area. The changes promoted
Responses Predicted Experimental RD (%)b by acid pretreatment in the cellulose structure enhanced
enzymatic cellulose hydrolysis and proved to be a very
TRS 18.75 20.2 7.2
efficient method to saccharify lignocellulosic substrates
TMa 12.40 13.91 11.04
[28].
a
TM = total monosaccharides content (glucose 3.75; xylose 7.43;
arabinose 2.73 g L-1) H2 Production Tests
b
Relative deviation (RD): [(experimental value—predicted
value)/experimental value] 9 100
All the literature studies on the acid hydrolysis and/or
pretreatment of sugarcane bagasse for biohydrogen pro-
duction have used sulfuric acid [3, 6, 7, 31]. However,
content did not lower significantly. These results agreed when it comes to fermentative hydrogen production by a
with the data depicted in Table 1, which shows the highest mixed culture, it is especially important to avoid intro-
concentration of xylose and arabinose in all hydrolysates. duction of sulfate and nitrate into the hydrolysates—sul-
Glucose from cellulose appeared at much lower concen- fate- and nitrate-reducing bacteria can potentially consume
tration than xylose (Table 1), and phenolic compounds did H2 from the medium, which could reduce the H2 yield [15].
not arise in the hydrolysates, as corroborated by low lignin The next step after pretreatment of sugarcane bagasse
solubilization in the pretreatment conditions used here with HCl under the optimized conditions (96.8 °C,
(Table 4). According to Monlau et al. [28], hemicelluloses 441.6 min, and acid at 7.36 %) was to conduct fermenta-
are the most thermochemically sensitive of all the ligno- tive H2 production assays using the sugarcane derived
cellulosic components. During thermochemical pretreat- substrates (Fig. 3).
ment, the hemicelluloses side groups react first, followed Bagasse in natura (B) and bagasse in natura added with
by the hemicelluloses. Under neutral conditions, lignin Celluclast Ò 10 U g-1 (B ? E) did not produce H2, prob-
normally starts to dissolve in water at around 180 °C. ably because the H2-producing microbial consortium, was
Therefore, dilute acid pretreatment appears to be a more not able to attack the lignocellulose of untreated even with
Table 4 Lignocellulosic content (cellulose, holocellulose, hemicellulose, and soluble and insoluble lignin) in bagasse in natura (B) and after
acid pretreatment under optimized conditions (sol)
Bagasse Cellulose (%) Holocellulose Hemicellulose Soluble lignin Insoluble lignin Total lignin
(%) (%) (%) (%) (%)
In natura (B) 46.70 ± 0.46b 73.62 ± 1.43a 26.92 0.10 ± 0.01a 20.65 ± 0.61b 20.75
a b b
After acid pretreatment 36.35 ± 1.15 37.66 ± 0.29 1.31 0.03 ± 0.002 19.43 ± 0.47b 19.46
(sol)
Holocellulose = Hemicellulose ? Cellulose. The letters a and b in the same column represent a significant difference in the samples as
calculated by Tukey Test (p \ 0.05)
123
Waste Biomass Valor
the aid of CelluclastÒ. All the researchers that have used the basis of the results from the fermentative H2 production
CelluclastÒ to saccharify sugarcane bagasse employed a assays. The control gave the highest maximum H2 pro-
pretreatment method that made cellulose available; e.g., duction rate (7.25 mL h-1), followed by the detoxified
acid, alkaline, or steam explosion treatment [6, 7, 12, 32, hydrolysate liq ? C (6.04 mL h-1). However, liq ? C
33]. For example, Martin et al. [32] achieved the highest afforded the largest H2 volume (114.6 mL) and yield
convertibility of sugarcane bagasse, 74.9 %, when they (0.72 mol of H2/mol of consumed TRS) as compared with
used CelluclastÒ at 25 FPU/g of dry after pretreatment by Glu (91.2 mL and 0.54 mol of H2/mol of consumed TRS,
wet oxidation at 195 Æ C, for 15 min, in alkaline pH. respectively). The higher H2 yield obtained with liq ? C as
Pretreatment was necessary even when an enzymatic compared with glucose (Table 5) was probably due to the
cocktail containing components with different enzymatic sugar concentration. The TRS concentration in liq ? C
activities was employed [33]. In the present work, addition included disaccharides, which were probably used for H2
of only CelluclastÒ at 10 U g-1 of bagasse was not enough production by the mixed culture, increasing H2 yield.
to make bagasse in natura (B) fermentable. Detoxification of lignocellulosic hydrolysates with
Without treatment with activated carbon, the liquid activated charcoal is a well-known cost-effective procedure
hydrolysate (liq) was also not suitable for H2 production, that can absorb inhibitor compounds [34]. The hydrolysate
probably due to its inhibitors content. that did not undergo treatment with activated carbon (liq)
Detoxification with activated carbon led the hydrolysate did not constitute a feasible substrate for H2 production.
liq ? C to produce a higher H2 volume as compared with Inhibitors emerged during acid pretreatment, even at the
the control (Glu), which relied on glucose as substrate only. relative mild conditions applied in this work. Indeed, liquid
However, the time elapsed until the beginning of H2 pro- hydrolysate (liq) obtained at the optimal conditions con-
duction was longer in the case of sol, sol ? E, and liq ? C tained furfural, HMF, and acetic acid at 0.08, 0.009, and
as compared with glucose. For example, it took liq ? C 2.3 g L-1, respectively (Table 5). Quéméneur et al. [35]
73 h to begin H2 production, whilst 23 h were necessary reported that addition of furfural at 1 g L-1 (a concentra-
for glucose. This indicated that substrate adaptation by the tion that was much higher than the concentration of
microbial consortium was necessary (Fig. 3). In addition, 0.08 g L-1 in the liquid hydrolysate) to xylose, as the main
the use of dried sludge enhanced the time elapsed until the substrate, inhibited 76 % of maximum H2 production by a
beginning of H2 production probably to allow the spores to mixed culture. As for acetic acid, it had lower inhibitory
return to their vegetative form. effect on H2 production as compared with furfural. Wang
Table 5 summarizes the modified Gompertz model et al. [36] verified that addition of acetic acid at 5 and
kinetic parameters, yields, and productivities obtained on 10 g L-1 (which are higher concentrations than the one
verified herein, 2.3 g L-1) inhibited maximum H2 pro-
duction by 29 and 64 %, respectively. In addition, ligno-
120
Glu
cellulose degradation products, like phenolic compounds,
sol are also known to be fermentation inhibitors [14], but they
100 sol+E
liq
were not detected here.
liq+C The substrates sol and sol ? E afforded similar H2 pro-
Hydrogen (mL)
80
B
B+E
duction rates (5.6 and 5.9 mL of H2/h, respectively), but
60
addition of Celluclast Ò to sol ? E enhanced the volume of
40 produced H2 from 27.6 to 35.9 mL; i.e., a 23 % increase.
The acid pretreatment of bagasse enhanced the ability of the
20
enzyme to convert cellulose. Indeed, addition of CelluclastÒ
0 to the untreated bagasse (B ? E) had no effect on H2 pro-
0 20 40 60 80 100 120 140
Time (h)
duction. In other words, the enzyme helped to degrade the
cellulose content of pretreated bagasse (36.35 %, Table 4)
Fig. 3 Fermentative H2 production from bagasse-derived substrates. and to solubilize sugars, thereby raising the H2 volume.
Glu: glucose in the same initial TRS concentration as the optimal CelluclastÒ acts mainly as endoglucanase, hydrolyzing the
hydrolysate (20 g L-1); liq: liquid hydrolysate obtained from pre-
b-1,4 glycoside bonds of cellulose and releasing oligosac-
treatment of the sugarcane bagasse in the optimized conditions;
liq ? C: liquid hydrolysate obtained from pretreatment of the charides, which are mostly non-fermentable. If a further
sugarcane bagasse in the optimized conditions, after treatment with enzyme like a b-glucosidase, which act mainly on cel-
activated carbon; sol: solid bagasse that remained after pretreatment lobiose, had been used, the fermentable sugar concentration
of the sugarcane bagasse in the optimized conditions; sol ? E: solid
would enhance even more, and, probably, also H2 volume.
bagasse that remained after pretreatment of the sugarcane bagasse in
the optimized conditions plus the enzyme Celluclast Ò 10 U g-1. B: The combination of different enzymatic activities in enzyme
bagasse in natura; B ? E: bagasse in natura ? Celluclast Ò blends or cocktails, with FPase, endoglucanase, xylanase,
123
Waste Biomass Valor
Table 5 Modified Gompertz model parameters, yields, and productivity of fermentative H2 production assays using glucose and sugarcane
bagasse derivatives as substrates
Substrate k (h) Rm (mL h-1) Hmax (mL) YH2/GLU (mol of H2/mol of YH2/g (mL of Productivity (mL of
consumed TRS as glucose) H2/g of bagasse) H2 L-1 h-1) or (mL
of H2 L-1 d-1)*
and b-glucosidase activities has been successfully employed activated carbon. Overall, our data pointed out that the
to saccharify pretreated sugarcane bagasse for fermentations pretreatment of sugarcane bagasse with HCl enabled
[33, 37]. application of whole bagasse; that is, the liquid and the
Together, the substrates sol ? E and liq ? C afforded solid fractions, as substrate for H2 production. It is note-
53.13 mL of H2/g of bagasse or 2.10 mmol of H2/g of worthy that the high yields obtained herein were only
bagasse. The volumetric productivities found in the present possible after detoxification of the liquid hydrolysate.
work—1450 and 1423 mL of H2 L-1 d-1 for liq ? C and Authors who applied pure cultures to produce H2 from
sol ? E, respectively—are the highest ever reported in the bagasse reported similar or even higher volumetric rates
literature for H2 production from sugarcane bagasse by a than ours. For example, Pattra et al. [3] produced 1611 mL
microbial consortium. of H2 L-1 d-1 when they used a pure culture of Clostridium
Few literature works have studied sugarcane bagasse as butyricum as inoculum. Lai et al. [7] achieved the highest
substrate for H2 production by microbial consortia. volumetric rates for H2 production ever described in the
Chairattanamanokorn et al. [4] pretreated sugarcane literature—0.52 L of H2 L-1 h-1—when they employed
bagasse with NaOH 0.1 mol L-1 at 100 °C, for 2 h, and Thermoanaerobacterium aotearoense SCUT27/Dldh as
added cellulolytic enzymes to improve hydrolysis and H2 pure culture. However, for practical reasons, studies that
production from bagasse in the presence of a mixed cul- apply mixed cultures are more realistic than those that use
ture. The addition of enzyme increased TRS from 0.25 to pure cultures. When it is desirable to produce a fuel such as
0.79 g of reducing sugars/g of bagasse, which raised H2 H2 continually and in huge amounts, the sterile conditions
production from 1.4 to 13.39 mmol of H2/g of Total required by pure cultures would be impracticable.
Volatile Solids in the bagasse. Here, addition of Cellu- Table 6 presents the concentrations of TRS, monosac-
clastÒ to the solid substrate (sol) increased the volume of charides, inhibitors, and the main fermentation soluble
produced H2 by ca. 25 % and raised the yield by 60 %, metabolites including acetate, butyrate, lactate, and etha-
from 0.20 to 0.33 mol H2/mol of consumed glucose. nol, at the beginning and at the end of the fermentative
Fangkum and Reunsang [6] worked on H2 production assays. According to Table 6, 56.8 % of the initial bagasse
from sugarcane bagasse pretreated with H2SO4, to achieve (3.3 g/100 mL) was converted to TRS, as detected in the
0.84 mmol of H2/mmol of consumed sugar when they used liquid fraction (18.75 g L-1), and 20.33 % remained in the
elephant dung as inoculum. Here, the yields were 0.20 and solid fraction (6.71 g L-1). The enzymatic treatment of the
0.72 mmol of H2/mmol of consumed TRS as glucose for solid fraction (sol) helped to solubilize ca. 5 % more TRS
sol and liq ? C, respectively (Table 5). Fangkum and (from 6.71 to 7.70 g L-1).
Reunsang (2011) obtained 109.55 mL of H2 L-1 d-1, In terms of total mL of H2/g of bagasse (Table 6) obtained
which was ca. 13 times lower as compared with our data from the liquid and solid fractions of bagasse, ca. 85 % of the
for liq ? C (1450 mL of H2 L-1 d-1, Table 5). This dif- H2 volume originated from liq ? C and 15 % from sol ? E.
ference probably resulted from the type of acid applied to In addition, 2.42 and 1.0 mL of H2/g of initial TRS concen-
pretreat and detoxify the liquid hydrolysate (liq) with tration originated from liq ? C and sol ? E, respectively.
123
Waste Biomass Valor
During the fermentation assays conducted with the On the other hand, the initial acetate concentration in the
bagasse-derived substrates, the mixed culture consumed liquid hydrolysate (liq) diminished, and no H2 evolved
the main monosaccharides detected in the bagasse hydro- from this substrate. Most likely, microorganisms present in
lysate (glucose, xylose, and arabinose). The fermentation the mixed culture were able to use acetate. In a literature
assays that used liquid hydrolysate (liq) as substrate con- review, Monlau et al. [39] described that inhibition of
sumed TRS and monosaccharides, but this uptake did not biohydrogen production in the presence of lignocellulosic
result in H2 formation (Fig. 2; Table 6) or in the metabo- byproducts does not imply that bacterial activity is inex-
lites acetate and butyrate. TRS were converted mainly into istent, because carbohydrates can undergo degradation
lactate (Table 6). The liquid hydrolysate treated with through non-hydrogen-producing pathways such as lactate,
activated carbon (liq ? C) contained reduced TRS and ethanol, and propionate pathways. In addition, the presence
monosaccharides concentrations, mainly xylose and ara- of inhibitors during dark fermentation could shift the
binose, at the beginning of the assays as compared with the metabolism from H2-producing pathways (i.e., acetate and
untreated liquid hydrolisate (liq). This fact could be due to butyrate) to non-H2-producing pathways (i.e., lactate,
the treatment of liq with activated carbon. Xylose from ethanol, and propionate). Indeed, an increase in lactate and
hemicellulose was the main monosaccharide detected in ethanol occurred when liquid hydrolisate (liq) was the
the solid fraction of bagasse (sol) as well as in sol ? E. substrate (Table 6).
This was probably because the acid pretreatment acted The presence of ethanol and lactate did not impact the H2
mainly on the hemicellulose, and washing of the pretreated balance, because these metabolites originated from con-
bagasse was not enough to remove all the solubilized sumption of an additional electron from NADH. Low
xylose. The presence of a pentose degradation product in amounts of ethanol emerged in all cases, except for liq ? C.
the sol fraction, acetic acid, corroborated this. Butyrate is also a metabolite related to the fermentative
The furfural and HMF inhibitors emerged at the begin- H2 production, because the consumption of 1 mol of glu-
ning of the assays in liq, sol, and sol ? E. At the end of the cose furnishes 1 mol of butyrate and 2 mols of hydrogen
assays, the concentrations of these inhibitors were recov- [15]. Butyrate emerged in all assays with H2 production,
ered in liq. However, in sol and sol ? E, the concentrations but this emergence was more pronounced in the assays that
of the inhibitors were lower than the detection limit of the afforded the highest H2 production (Glu and liq ? C).
method (10-3 g L-1). Treatment of liq with activated Compared with the mmol of H2 produced (Table 5) in
carbon removed the inhibitors (furfural and HMF), as the assays Glu, liq ? C, sol, sol ? E, the butyrate con-
verified by their initial concentration in the liq ? C assay. centration (Table 6) represented 63.5, 75.4, 72.1, and
This removal was probably the reason why it was possible 75.6 %, respectively, of the maximum theoretical H2 yield
to use liq ? C as substrate (Fig. 2). Treatment of the liquid based on the stoichiometric ratio (1 mols of butyrate for
hydrolysate with activated carbon may also have removed 2 mols of H2). These results showed that H2 production by
other fermentation inhibitors that went undetected here. this mixed culture using either glucose or other substrates
Acetate was the metabolite from fermentative H2 pro- preferentially followed the butyrate pathway. Use of the
duction that furnished the highest H2 yield from glucose— liquid hydrolysate and the sugarcane derived substrates
consumption of 1 mol of glucose formed 2 mols of acetate, enhanced this effect. Fangkum and Reusang [6] studied the
and 4 mols of hydrogen [15]. However, acetic acid could also effects of sugarcane bagasse hydrolysate as substrate for H2
be an inhibitor or even a substrate for H2 production [36–39]. production, and they also detected butyrate as the main
During hydrolysis of lignocellulosic materials, acetic acid soluble product. These authors did not mention the pres-
originates from hydrolysis of acetyl groups in hemicellulose ence of lactic acid during fermentation, but they attributed
and, to some extent, lignin [28]. In the undissociated form, the low H2 yield to the existence of Sporolactobacillus sp.,
acetic acid can penetrate the cell membrane and inhibit acid lactic bacteria, in the microbial community. Although
product formation, disrupting the pH balance at high con- in the present work lactate emerged as a metabolite, it is
centration, inhibiting cell growth, or even killing cells [29, not possible to affirm that it originated from liquid
37]. However, some strains, mainly those belonging to hydrolysate fermentation, because it also appeared as the
Clostridia species, can use acetic acid and other organic main metabolite in the assay that used glucose as substrate.
acids as substrate to produce H2 [38]. In all the assays that
produced H2 (Glu, liq ? C, sol, and sol ? E), the acetate
concentration rose. Compared with the mmol of H2 produced Conclusions
(Table 5) in the assays Glu, liq ? C, sol, sol ? E, the acetate
concentration (Table 6) represented 12, 3, 8.5, and 7.8 % of The surface-response methodology used to optimize tem-
the maximum theoretical H2 yield based on the stoichio- perature, time, and HCl concentration conditions to pretreat
metric ratio (2 mols of acetate for 4 mols of H2). sugarcane bagasse revealed that temperature, time, and
123
Waste Biomass Valor
Table 6 Concentration of TRS, monosaccharides, inhibitors, and soluble fermentation metabolites at the beginning and at the end of the fermentation assays that used the sugarcane bagasse
derivatives as substrate
Concentration (g L-1) Substrate in the fermentative assay
Glu Liq Liq ? C Sol Sol ? E
I F I F I F I F I F
Saccharide
TRS 20.81 ± 1.93 8.91 ± 0.66 18.75 ± 2.35 15.8 ± 1.96 16.82 ± 2.08 5.6 ± 0.87 6.71 ± 2.17 – 7.7 ± 2.51 –
Glucose 19.93 ± 1.94 8.87 ± 0.25 3.75 ± 0.8 0.78 ± 0.07 3.26 ± 0.03 0.67 ± 0.02 0.64 ± 0.09 0.02 0.57 ± 0.01 0.02
Xylose – – 7.43 ± 0.73 – 5.98 ± 0.41 0.64 ± 0.08 1.49 ± 0.03 – 1.185 ± 0.05 0.30 ± 0.04
Arabinose – – 2.73 ± 0.80 1.4 ± 0.70 0.88 ± 0.02 0.012 ± 0.01 – – – –
Inhibitor
5-HMF – – 0.009 0.009 – – 0.002 – 0.002 –
Furfural – – 0.083 ± 0.006 0.067 ± 0.007 – – 0.007 – 0.008 –
Metabolite
Acetate – 0.013 ± 0.01 1.072 ± 0.06 0.67 ± 0.03 0.032 ± 0.002 0.074 ± 0.004 0.003 0.006 0.003 0.073 ± 0.01
Butyrate – 0.131 ± 0.22 – – – 0.151 ± 0.04 – 0.034 ± 0.03 – 0.062 ± 0.02
Lactate – 0.002 0.011 ± 0.005 0.065 ± 0.004 – 0.092 ± 0.03 – 0.001 – –
Ethanol – 0.002 0.012 ± 0.006 0.045 ± 0.002 – – – 0.005 – 0.003
I: initial concentration, F: final concentration. Concentrations without value or deviation means that they are less than 10-3 (except for TRS analysis). Glu: glucose in the same initial TRS
concentration as the optimal hydrolysate (20 g L-1); liq: liquid hydrolysate obtained from pretreatment of the sugarcane bagasse in the optimized conditions; liq ? C: liquid hydrolysate
obtained from pretreatment of the sugarcane bagasse in the optimized conditions treated with activated carbon; sol: solid bagasse that remained after pretreatment of the sugarcane bagasse in the
optimized conditions; sol ? E: solid bagasse that remained after pretreatment of the sugarcane bagasse in the optimized conditions plus enzyme CelluclastÒ 10 U g-1
123
Waste Biomass Valor
acid concentration affected the TRS and total monosac- 11. Monlau, F., Barakat, A., Trably, E., Dumas, C., Steyer, J.-P.,
charides (TM) concentrations, but these variables did not Carrere, H.: Lignocellulosic materials into biohydrogen and
biomethane: impact of structural features and pretreatment. Crit.
impact the TI concentration significantly. The condi- Rev. Environ. Sci. Technol. 43, 260–322 (2013)
tions 96.8 °C, 7.36 % HCl, and 441.6 min afforded the 12. Saripan, A.F., Reungsang, A.: Biohydrogen production by
highest TRS in the hydrolysates. However, the hydrolysate Thermoanaerobacterium thermosaccharolyticum KKU-ED1:
only became a suitable substrate to produce H2 after culture conditions optimization using xylan as the substrate. Int.
J. Hydrog. Energy 38, 6167–6173 (2013)
detoxification with activated carbon, and the H2 yield was 13. Lavarack, B.P., Griffin, G.J., Rodman, D.: The acid hydrolysis of
higher as compared with a control containing glucose only. sugarcane bagasse hemicellulose to produce xylose, arabinose,
Pretreatment with HCl also made the remaining solid glucose and other products. Biomass Bioenergy 23, 367–380
bagasse suitable for H2 production. Subsequent enzymatic (2002)
14. Jonsson, L., Alriksson, B.: Nilvebrant N–O bioconversion of
hydrolysis of this substrate increased the volume of H2 lignocellulose: inhibitors and detoxification. Biotechnol. Biofuels
produced from the initial bagasse even further. The 6, 16–22 (2013)
appropriate pretreatment of sugarcane bagasse with HCl 15. Valdez-Vazquez, I., Poggi-Varaldo, H.M.: Hydrogen production
can transform the whole bagasse into a renewable substrate by fermentative consortia. Renew. Sustain. Energy Rev. 13,
1000–1013 (2009)
for H2 production. 16. Derringer, G., Suich, R.: Simultaneous optimization of several
response variables. J. Qual. Technol. 12, 214–219 (1980)
Acknowledgments The authors wish to thank ‘Programa de Aper- 17. Gonzalez-Gil, G., Kleerebezem, R., Lettinga, G.: Assessment of
feiçoamento de Pessoal de Nı́vel Superior’ (CAPES) as well as metabolic properties and kinetic parameters of methanogenic
‘‘Fundação de Amaro à Pesquisa do Estado de São Paulo’’ (FAPESP) sludge by on-line methane production rate measurements. Appl.
for financial support. Microbiol. Biotechnol. 58, 248–254 (2002)
18. Buitrón, G., Carvajal, C.: Biohydrogen production from Tequila
vinasses in an anaerobic sequencing batch reactor: effect of initial
substrate concentration, temperature and hydraulic retention time.
References Bioresour. Technol. 101, 9071–9077 (2010)
19. Hargraeves, P.I., Barcelos, C.A., Costa, A.C.A., Pereira Jr, N.:
1. Instituto Brasileiro de Geografia e Estatı́stica - IBGE. Anuário Production of ethanol 3G from Kappaphycus alvarezii: evalua-
Estatı́stico do Brasil, v.73, ISSN 0100-1299 (2013) tion of different process strategies. Bioresour. Technol. 134,
2. Hofsetz, K., Silva, M.A.: Brazilian sugarcane bagasse: energy 257–263 (2013)
and non-energyconsumption. Biomass Bioenergy 46, 564–573 20. Ghose, T.K.: Measurement of cellulase activities. Pure Appl.
(2012) Chem. 59, 257–268 (1987)
3. Pattra, S., Sangyoka, S., Boonmee, M., Reungsang, A.: Bio-hy- 21. Garcı́a-Morales, J.L., Nebot, E., Romero, L.I., Sales, D.: Com-
drogen production from the fermentation of sugarcane bagasse parison between acidogenic and methanogenic inhibition caused
hydrolysate by Clostridium butyricum. Int. J. Hydrog. Energy 33, by liner alkylbenzzene-sulfonate (LAS). Chem. Biochem. Eng.
5256–5265 (2008) Q. 15, 13–19 (2001)
4. Chairattanamanokorn, P., Penthamkeerati, P., Reungsang, A., Lo, 22. Sun, J.: Isolation and characterization of cellulose from sugarcane
Y.-C., Lu, W.B., Chang, J.S.: Production of biohydrogen from bagasse. Polym. Degrad Stabil. 84, 331–339 (2004)
hydrolyzed bagasse with thermally preheated sludge. Int. J. Hy- 23. TAPPI. TAPPI Standard. Method T19 om-54. TAPPI test meth-
drog. Energy 34, 7612–7617 (2009) ods (1991)
5. Cheng, C.L., Chang, J.S.: Hydrolysis of lignocellulosic feedstock 24. Sluiter A, Hames B, Ruiz R, Scarlata C, Sluiter J, Templeton D,
by novel cellulases originating from Pseudomonas sp. CL3 for Crocker D. Determination of structural carbohydrates and lignin
fermentative hydrogen production. Bioresour. Technol. 102, in biomass. Technical Report. NREL/TP-510-42618 (2012)
8628–8634 (2011) 25. APHA, AWWA, WEF. Standard methods for the examination of
6. Fangkum, A., Reungsang, A.: Biohydrogen production from water and wastewater. 19th. edn. American Public Health Asso-
sugarcane bagasse hydrolysate by elephant dung: effects of initial ciation. Washington, DC (1995)
pH and substrate concentration. Int. J. Hydrog. Energy 36, 26. Miller, G.L.: Use of dinitrosalicylic acid reagent for determina-
8687–8696 (2011) tion of reducing sugars. Anal. Chem. 30, 785–793 (1959)
7. Lai, Z., Zhu, M., Yang, X., Wang, J., Li, S.: Optimization of key 27. Sá, L.R.V., de Oliveira, M.A.L., Cammarota, M.C., Matos, A.,
factors affecting hydrogen production from sugarcane bagasse by Ferreira-Leitão, V.S.: Simultaneous analysis of carboydrates and
a thermophilic anaerobic pure culture. Biotechnol. Biofuels 7, volatile fatty acids by HPLC for monitoring fermentative bio-
119–130 (2014) hydrogen production. Int. J. Hydrog. Energy 36, 7543–7549
8. Ntaikou, I., Antonopoulou, G., Lyberatos, G.: Biohydrogen pro- (2011)
duction from biomass and wastes via dark fermentation: a review. 28. Monlau, F., Barakat, A., Trably, E., Dumas, C., Steyer, J.P.,
Waste Biomass Valor. 1, 21–39 (2010) Carrère, H.: Lignocellulosic materials into biohydrogen and
9. Rezende, C.A., de Lima, M.A., Maziero, P., de Azevedo, E.R., biomethane: impact of structural features and pretreatment. Crit.
Garcia, W., Polikarpov, I.: Chemical and morphological charac- Rev. Environ. Sci. Technol. 43, 260–322 (2013)
terization of sugarcane bagasse submitted to a delignification 29. Klinke, H.B., Thomsen, A.B., Ahring, B.K.: Inhibition of etha-
process for enhanced enzymatic digestibility. Biotechnol. Bio- nol-producing yeast and bacteria by degradation products pro-
fuels 4, 1–18 (2011) duced during pre-treatment of biomass. Appl. Microbiol.
10. Alvira, P., Tomás-Pejó, E., Ballesteros, M., Negro, M.J.: Pre- Biotechnol. 66, 10–26 (2004)
treatment technologies for an efficient bioethanol production 30. Hendriks, A.T.W.M., Zeeman, G.: Pretreatments to enhance the
process based on enzymatic hydrolysis: a review. Bioresour. digestibility of lignocellulosic biomass. Bioresour. Technol. 100,
Technol. 101, 4851–4861 (2010) 10–18 (2009)
123
Waste Biomass Valor
31. Rai, P.K., Singh, S.P., Asthana, R.K., Singh, S.: Biohydrogen 36. Wang, Y., Zhao, Q.B., Yang, M., Yu, H.Q., Harad, H., Li, Y.Y.:
production from sugarcane bagasse by integrating dark- and Biohydrogen production with mixed anaerobic cultures in the
photo-fermentation. Bioresour. Technol. 152, 140–146 (2014) presence of high-concentration acetate. Int. J. Hydrog. Energy 33,
32. Martin, C., Klinke, H., Thomsen, A.B.: Wet oxidation as a pre- 1164–1171 (2008)
treatment method for enhancing the enzymatic convertibility of 37. Hernández-Salas, J.M., Villa-Ramı́rez, M.S., Veloz-Rendón, J.S.,
sugarcane bagasse. Enzyme Microbial. Technol. 40, 426–432 Rivera-Hernández, K.N., González-César, R.A., Plascencia-
(2007) Espinosa, M.A., Trejo-Estrada, S.R.: Comparative hydrolysis and
33. Visser, E.M., Leal, T.F., Almeida, M.N., Guimarães, V.M.: fermentation of sugarcane and agave bagasse. Bioresour. Tech-
Increased enzymatic hydrolysis of sugarcane bagasse from nol. 100, 1238–1245 (2009)
enzyme recycling. Biotechnol. Biofuels 8, 5 (2015) 38. Matsumoto, M., Nishimura, Y.: Hydrogen production by fer-
34. Canilha, L., Carvalho, W., Felipe, M.G.A., Silva, J.B.A.: Xylitol mentation using acetic acid and lactic acid. J. Biosci. Bioeng.
production from wheat straw hemicellulosic hydrolysate: hydro- 103, 236–241 (2007)
lysate detoxification and carbon source used for inoculum 39. Monlau, F., Sambusiti, C., Barakat, A., Quéméneur, M., Trably,
preparation. Brazilian J. Microbiol. 39, 333–336 (2008) E., Steyer, J.-P., Carrère, H.: Do furanic and phenolic compounds
35. Quermeneur, M., Hamelin, J., Barakat, A., Steyer, J.P., Carrere, of lignocellulosic and algae biomass hydrolyzate inhibit anaero-
H., Trably, E.: Inhibition of fermentative hydrogen production by bic mixed cultures? A comprehensive review. Biotechnol. Adv.
lignocellulose-derived compounds in mixed cultures. Int. J. Hy- 32, 934–951 (2014)
drog. Energy 37, 3150–3159 (2012)
123