Industrial Crops and Products 46 (2013) 217–225

Contents lists available at SciVerse ScienceDirect

Industrial Crops and Products

journal homepage: www.elsevier.com/locate/indcrop

Bioconversion of rice straw to sugar using multizyme complex of fungal origin

and subsequent production of bioethanol by mixed fermentation of
Saccharomyces cerevisiae MTCC 173 and Zymomonas mobilis MTCC 2428
Arpan Das, Tanmay Paul, Arijit Jana, Suman K. Halder, Kuntal Ghosh, Chiranjit Maity,
Pradeep K. Das Mohapatra, Bikash R. Pati, Keshab C. Mondal ∗
Department of Microbiology, Vidyasagar University, Paschim Midnapore 721 102, West Bengal, India

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

Article history: The multidimensional approach for ethanol production from rice straw was opted by three distinct
Received 8 October 2012 phases: firstly, statistical optimization of ␤-glucosidase production from co-culture of Aspergillus fumiga-
Received in revised form tus ABK9 and Trichoderma reesei SAP3 through mixed substrate (wheat bran and rice straw) fermentation;
28 December 2012
secondly, enzymatic saccharification of pretreated rice straw for high yield of reducing sugar and finally,
Accepted 1 February 2013
statistical optimization of bioconversion of the sugar to ethanol by mixed fermentation of Saccharomyces
cerevisiae MTCC 173 and Zymomonas mobilis MTCC 2428. In optimized media, maximum ␤-glucosidase
Keywords:
yield of 265.4 U g−1 was achieved. Enzymatic treatment (40 U g−1 ) of NaOH pretreated rice straw produced
Mixed fungal fermentation
Pretreatment
maximum reducing sugar of 24.9 g L−1 . It also showed maximum enzyme adsorption (Emax ) by 2 fold and
Adsorption decreased the absorption coefficient (Kad ) by 37.64% relative to untreated straw. During ethanol fer-
Rice straw mentation, inoculum ratio became most influencing factor to maximize ethanol production of 40.1 g L−1 ,
Bioethanol indicating the influencing effect of the perpetrator strains.
© 2013 Elsevier B.V. All rights reserved.

1. Introduction with the use of potential food resources (Das and Ghosh, 2009; Yang
and Wyman, 2008).
One of the greatest challenges for society in the 21st century One barrier for the production of ethanol from biomass is that,
is to meet the increasing demand of fuel energy for transporta- in lignocellulosic materials cellulose is tightly bound to hemicellu-
tion and industrial processes in a sustainable way. Global energy loses and lignin. Lignin resists degradation and confers hydrolytic
consumption continues to increase while fossil-fuel reserves are stability and structural robustness to the cell walls of the plants. To
diminishing. To prevent a looming energy crisis, the development overcome lignocellulose recalcitrance, pretreatment is required to
of renewable energy sources is becoming a priority. Bioethanol is an alter the structure as well as its submicroscopic chemical compo-
increasingly important alternative for the replacement of fossil fuel, sition, so that enzymatic hydrolysis of the carbohydrate fraction
with a world production of 19,535 millions of gallons in 2009 (Ruiz to monomeric sugars can be achieved more rapidly and with
et al., 2012). Bioethanol made from lignocellulosic biomass includ- greater yields. To date, the most widely recognized pretreatment
ing agricultural, forestry residues and woody crops, is finally being techniques include steam explosion, treatment with alkali, dilute
recognized widely as a unique alternative fuel with powerful eco- acid, sulphur dioxide, hydrogen peroxide, organosolve, supercriti-
nomic, environmental and strategic attributes (Yang and Wyman, cal ammonia etc. But to qualify as effective, the pretreatment must
2008) and currently, it is a hotspot in the bioenergy research field. meet the following criteria: (i) maximize fermentable sugar yields,
Cellulosic ethanol has recently been produced from agricultural (ii) avoid, or minimize degradation of carbohydrates, (iii) avoid, or
wastes (straw and baggase), which are low cost feedstock and avail- minimize the formation of microbial growth-inhibiting byproducts,
able in plenty and also does not have the ethical concerns associated and (iv) minimize energy, capital and operating costs. Considering
these, alkali, acid and peroxide treatments are widely applied in
industries, but the suitability and the concentration of the chem-
icals vary with the type of substrate utilized (Chiaramonti et al.,
2011).
∗ Corresponding author. Tel.: +91 03222 276554/555x477; fax: +91 03222
Cellulolytic filamentous fungi belonging to the genera Tricho-
275329.
E-mail addresses: arpan das85@yahoo.co.in (A. Das), mondalkc@gmail.com (K.C.
derma have long been considered for the most productive and
Mondal). powerful degrader of crystalline cellulose. Cellulase preparations

0926-6690/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.indcrop.2013.02.003
218 A. Das et al. / Industrial Crops and Products 46 (2013) 217–225

from mutated or genetically manipulated strains of Trichoderma collected from the local market of Midnapore town, West Bengal,
reesei are preferred by many companies worldwide, but, its price is India. The wheat bran contained significant amounts of carbo-
high (Kova’cs et al., 2009). The strain of T. reesei is usually been char- hydrate (58.0%) along with other essential nutrients (Kar et al.,
acterized by high endoglucanase and cellobiohydrolase producer 2012). The major constituents of rice straw were 41.2% cellulose,
but it is poor in secretion of ␤-glucosidase (Lynd et al., 2002) and its 31.7% hemicellulose, and 21.8% lignin. In the current experiment,
activity level is insufficient to provide the fast and complete conver- wheat bran was utilized as it is, but rice straw was subjected to
sion of cellobiose (the intermediate product of enzymatic cellulose mild alkaline pretreatment to facilitate its biological utilization.
hydrolysis) to glucose. Thus, T. reesei cellulase preparations had Briefly, the rice straw (10%, w/v) was treated with NaOH solution
to be supplemented with additional ␤-glucosidase (usually from (molarity was adjusted according to the experimental model)
Aspergillus sp.) to provide the more efficient saccharification of cel- for 30 min under boiling condition. After cooling, the straw was
lulosic substrates. washed several times in tap water to neutralize the pH followed
Generally the production of cellulases accounts for about 40% of by a rinse in distilled water, and finally air dried. The pretreated
total cost in bioethanol production (Gray et al., 2006). To lessen it, feedstock was stored in airtight container until used.
on site production of crude enzyme is more viable than commercial
cellulase due to their reasonable cost, variation of stability during
2.3. Solid state fermentation using central composite design
storage and transportation, and substrate specificity.
(CCD)
Solid state fermentation (SSF) can be of special interest in pro-
cesses where crude fermented product could be used directly as
In the study, central composite design (CCD) (Jabasingh and
enzyme source. This technology is promising because processes
Nachiyar, 2011) was used to evaluate the main and interaction
involving SSF have lower energy requirements, high product con-
effects of the four fermentation factors for ␤-glucosidase produc-
centration, lower input of infrastructure and skill, produce less
tion namely (A) initial moisture (55, 65, 75%), (B) substrate amount
wastewater and are environmentally friendly as they resolve the
(10, 15, 20 g), C: NaOH (1.0, 1.5, 2.0 M) for pretreatment for rice
problem of solid waste disposal (Sukumaran et al., 2009). Currently,
bran and D: inoculum ratio of A. fumigatus ABK9 and T. reesei
industrial demand for cellulases is being met by production meth-
SAP 3 (1.0, 1.5, 2.0). Other fermentation parameters like WB to
ods using submerged fermentation (SmF) processes. The cost of
RS ratio, incubation time, temperature and pH of the medium
production in SmF systems is however high and it is uneconomical.
were kept constant to 1:1, 96 h, 30 ◦ C and 6.5, respectively. Prior
Therefore, it is necessary to reduce the production cost by deploying
fermentation, the substrates were soaked with salt medium [Com-
alternative methods such as SSF. Although, SSF suffers from certain
position: (NH4 )2 SO4 1.4%, KH2 PO4 2%, CaCl2 , 2H2 O 0.4%, MgSO4
drawbacks such as alteration of pH during growth, accumulation of
0.3%, Peptone 1%, NaCl 0.3%, FeSO4 0.005%, MnSO4 0.0016%, ZnSO4 ,
metabolites (e.g. protease) and operational problems during scale
7H2 O 0.0014%, CoCl2 0.002%] to attain 50% initial moisture con-
up (Kar et al., 2012).
tain.
Thus, the aims of the current study were to develop a pro-
The CCD used in this experiment had six replicates at the cen-
cess of co-culturing the strains of T. reesei SAP3 and Aspergillus
tral point as well as two replicates at the axial and factorial points
fumigatus ABK9 for ␤-glucosidase production through central
(˛ = 2) leading to 30 experiments. Statistical analysis was done by
composite design (CCD) by solid state fermentation, enzymatic
Design Expert 8.0.3. Both linear and quadratic effects as well as the
saccharification of the pretreated rice straw for better yield of fer-
possible interactions of the four variables were calculated and their
mentable sugars and then, optimization of ethanol production by
significances were evaluated by variance analysis (ANOVA). 3D sur-
the strains of Saccharomyces cerevisiae (MTCC 173) and Zymomonas
face plots were drawn to show the effects of independent variables
(MTCC 2428) following Box–Behnken response surface methodol-
on the response. The ‘fit of the model’ was evaluated by determi-
ogy.
nation of R2 coefficient. Fermentations were performed in static
trays (30 cm × 15 × 2 cm), placed in a closed humidity chamber
2. Materials and methods (REMI Environmental Chamber, CHM6S, India). After fermentation,
enzyme in the fermented pith was extracted with sterile distilled
2.1. Microorganism and cultural condition water (1:10, w/v) by shaking at 120 rpm for 1 h, followed by cen-
trifugation at 5000 × g for 10 min. The supernatant was used as a
Fungal strain, T. reesei SAP 3 (MTCC 4876) was obtained from crude enzyme source.
the culture collection of Department of Microbiology, Vidyasagar
University and A. fumigatus ABK9 (GenBank Acc. no- HM807348.1) 2.4. Pretreatment of rice straw for enzymatic saccharification
was isolated from the soil of the pulp industry. Both the strains were
grown on potato dextrose agar (PDA) slants at 30 ◦ C for 5 days (until Alkaline, acidic, hydrogen peroxide pretreatments were carried
good sporulation occurred) and stored at 4 ◦ C until use. out to make the substrate more accessible to enzymatic saccharifi-
S. cerevisiae (MTCC 173) and Zymomonas mobilis (MTCC 2428), cation. Briefly, rice straw (10%, w/v) was boiled with sulphuric acid
two distillery strains for ethanol production were collected from (1–5 M), sodium hydroxide (1–5 M) or hydrogen peroxide (1–5%,
the Microbial Type Culture Collection (MTCC), Chandigarh, India. w/v) for 60 min. After cooling, the samples were washed with tap
The S. cerevisiae culture was grown in YEP broth media [contained water and air dried. Water-insoluble residue of the pretreated
(w/v) yeast extract 0.3%, peptone 1.0%, dextrose 2%, pH 6.0] and rice straw was subjected to enzymatic hydrolysis by the crude
Z. mobilis was grown in nutrient rich medium containing dextrose ␤-glucosidase loading of 20–40 U g−1 of the substrate. Enzymatic
2%, yeast extract 1.0%, KH2 PO4 0.2% and pH 6.0. After incubation saccharification was performed at varying pH (4.0–8.0), tempera-
for 24 h at 120 rpm, they were used as inoculum for alcohol pro- tures (40–55 ◦ C) and substrate concentrations (5–30%, w/v) on a
duction. rotary shaker at 100 rpm for 48 h. Samples were withdrawn peri-
odically and the amount of reducing sugar released was estimated
2.2. Substrates for fungal growth and enzyme production colorimetrically by DNS method (Miller, 1959). All experiments
were performed in triplicate. The enzymatic hydrolyzate was con-
Two abundant and cheap agricultural wastes namely wheat centrated up to 10% reducing sugar content and used for ethanol
bran (WB) and rice straw (RS) (∼1–3 mm particle size) were production.
 A. Das et al. / Industrial Crops and Products 46 (2013) 217–225 219

2.5. Optimization of ethanol production 2.8. Scanning electron microscopy

Fermentation temperature (A: 25, 30, 35 ◦ C), pH (B: 5.5, 6.0, The structural changes in the morphology of rice straw before
6.5), inoculum amount (C: 1, 2, 3%) and ratio of S. cerevisiae to and after treatment of alkali and enzyme were studied by scanning
Z. mobilis (D: 1.0, 1.5, 2.0), were optimized using a Box–Behnken electron microscope (JEOL JSM-5600). Images of rice straw were
response surface design (Box and Behnken, 1960) for enhanced taken at a magnification of 1000×. The specimens were mounted on
ethanol production. The design matrix with 29 runs (5 replicates at a conductive tape and coated with gold palladium using a JEOL–JFC-
the midpoint) was used for the experiment. The model, constructed 1200 fine coater and observed using a voltage of 25 kV.
as a response function of the variables on ethanol production was
a second-order polynomial as follows: 2.9. Determination of ˇ-glucosidase adsorption on cellulose
4 4 4
Y = ˇ0 + ˇi xi + ˇii xi2 + ˇij xi xj (1) ␤-glucosidase adsorption on pretreated rice straw was mea-
i=1 i=1 i,i=1
sured in a 100 ml Erlenmeyer flask containing 1% of substrate and
where, Y is the predicted response; xi and xj are the independent
enzyme concentrations varying from 1 to 10 mg mL−1 (prepared in
variables in coded values that influence the response, Y; ˇo the
50 mM citrate buffer, pH 5.5). The reactions were performed for
offset term; ␤i represent the linier effect of xi ; ˇij represent the
10 min at 50 ◦ C and subsequently, a 1.0 ml sample was taken and
interaction effect between xi and xj ; ˇii represent the quadratic
centrifuged at 5000 × g for 5 min. The resulting supernatant was
effect of xi . Regression analysis and estimation of the coefficients
analyzed for free enzyme concentration.
were performed using Design Expert software (Stat ease Corp, USA).
The enzyme adsorption was calculated following a Langmuir-
type isotherm as described by Pierre et al. (2011)
2.6. Analytical methods
E 
max Kad EF S
EB = (2)
␤-Glucosidase (E.C. 3.2.1.21) activity was determined by incu- 1 + Kad EF
bating 0.5 mL of enzyme solution with 0.5 mL of 1.0% (w/v)
cellobiose in 0.05 M sodium citrate buffer (pH 5.0) at 50 ◦ C for where, EB is the bound enzyme concentration, EF is the free enzyme
30 min and the liberated glucose was estimated by the glucose concentration, S is the straw concentration, Emax is the maximum
oxidase-peroxidase (GOD-POD) method (Bergmeyer, 1974). enzyme adsorption in g enzyme/g straw and Kad is the adsorption
Endoglucanase or CMCase (E.C. 3.2.1.4) activity was determined coefficient. To determine Emax and Kad , a linearized form of Eq. (2)
by incubating 0.5 mL of enzyme solution with 0.5 mL of 1.0% (w/v) was used as:
CMC (carboxymethyl cellulose) (Sigma, St. Louis, MO, USA) in
S  1
  1 
= + (3)
0.05 M sodium citrate buffer (pH 5.0) (Mandels et al., 1976). After EB Emax Kad EF Emax
incubation at 50 ◦ C for 30 min, the reaction was stopped by adding
1.0 mL of 3, 5-dinitrosalicylic acid (DNS) reagent. Liberated reduc- Once the data were plotted with S/EB on the y-axis and 1/EF on
ing sugars were estimated colorimetrically according to Miller the x-axis, a linear regression was obtained using Microsoft Excel.
(1959) using glucose as standard. Emax and Kad were calculated from the intercept and slop of the
Filter paper degrading activity (FPase) was determined using generated plot.
processed Whatman No. 1 filter paper as substrate (50 mg mL−1 )
according to the method of Wood and Bhat (1988). 3. Results and discussion
Xylanase (E.C. 3.2.1.8) activity was determined according to
Bailey et al. (1992) by measuring the release of reducing sugars From the current literatures, it is found that the substrates like
from 1.0% (w/v) birch wood xylan under standard conditions (50 ◦ C, wheat bran and rice straw can be considered as good supporting
0.05 M phosphate buffer, pH 7.0). Free sugar released was analyzed substrate for fungal growth probably due to the presence of var-
by the DNS method. ious available nutrients, good porosity, suitable particle size and
One unit of enzymatic activity was defined as the amount of consistency required for fungal anchorage and enzyme excretion
enzyme required to produce 1 ␮mol of reducing sugars (measured (Das et al., 2013; Sun et al., 2008; Kar et al., 2012). Its texture
as monosaccharides content) by hydrolyzing respective raw sub- remains loose even in moist condition, thereby provides a large
strate per minute under specified assay conditions. Enzyme activity surface area and increased water holding capacity (Gawande and
was expressed as U g−1 of dry substrate. Kamat, 1999). Considering these properties, wheat bran and rice
Soluble protein content was determined following the Lowry straw were chosen to provide optimum support for enzyme pro-
method (Lowry et al., 1951) using bovine serum albumin (frac- duction from co-culture of A. fumigatus ABK9 and T. reesei SAP3.
tion V) as standard. The concentration of total reducing sugars
and ethanol was determined using DNS method (Miller, 1959) and 3.1. CCD for optimization of ˇ-glucosidase production
dichromate method (AOAC, 1990), respectively.
The optimum levels of the selected factors and the effects of their
2.7. Determination of the chemical composition of rice straw and interactions on ␤-glucosidase production were determined by CCD.
FTIR analysis Obtained results were analyzed by analysis of variance (ANOVA)
and the predicted as well as observed responses are presented in
The lignin, hemicellulose and cellulose content of the rice straw Table 1. The second order regression equation (as presented below)
were analyzed according to the method of Pierre et al. (2011). The provided the levels of ␤-glucosidase production as a function of
infrared spectral analysis of the untreated and treated (NaOH & initial values of the interacting factors (Jabasingh and Nachiyar,
enzyme) rice straw was studied using Fourier Transform Infrared 2011):
Spectroscopy (Schimadzu, Japan). For this, 3.0 mg of rice straw was
dispersed in 300 mg of spectroscopic grade KBr and subsequently Y = 258.25 − 4.93A + 0.72B − 4.92C + 18.32D − 3.07AB
pressed into disks at 10 MPa for 3 min (Binod et al., 2012). The spec-
− 13.07AC + 6.32AD + 20.19BC − 7.87BD + 16.98CD
tra were obtained in transmission mode with an average of 30 scans
and a resolution of 4 cm−1 in the range of 4000–400 cm−1 . −20.55A2 − 12.10B2 − 6.48C 2 − 19.68D2 (4)
220 A. Das et al. / Industrial Crops and Products 46 (2013) 217–225

Table 1
Central composite design and the analysis of variances (ANOVA) for optimization of ␤-glucosidase production by mixed fungal fermentation with A. fumigatus ABK9 and T.
reesei SAP 3 through solid state fermentation.

Run no. (A) Initial (B) Substrate (C) NaOH (D) Fungal ratio ␤-Glucosidase activity ␤-Glucosidase activity
moisture (%) amount (g) pretreatment (M) (A:T) (U/g) observed (U/g) predicted

1 75 (1) 20 (1) 1 (−1) 1 (−1) 187.5 190.1

2 55 (−1) 20 (1) 1 (−1) 1 (−1) 190.2 192.6
3 65 (0) 15 (0) 1.5 (0) 1.5 (0) 257.4 258.3
4 75 (1) 10 (−1) 2 (1) 1 (−1) 106.3 109.2
5 45 (−2) 15 (0) 1.5 (0) 1.5 (0) 199.3 185.9
6 75 (1) 20 (1) 2 (1) 1 (−1) 163.1 160.6
7 65 (0) 15 (0) 2.5 (2) 1.5 (0) 229.4 222.5
8 75 (1) 10 (−1) 2 (1) 2 (1) 209.5 208.1
9 65 (0) 15 (0) 1.5 (0) 0.5 (−2) 146.7 142.9
10 55 (−1) 20 (1) 2 (1) 1 (−1) 209.6 215.4
11 65 (0) 15 (0) 1.5 (0) 1.5 (0) 253.8 258.3
12 65 (0) 15 (0) 1.5 (0) 1.5 (0) 269.4 258.3
13 55 (−1) 20 (1) 1 (−1) 2 (1) 158.9 166.9
14 65 (0) 15 (0) 0.5 (−2) 1.5 (0) 248.3 242.2
15 65 (0) 5 (−2) 1.5 (0) 1.5 (0) 216.6 208.4
16 75 (2) 15 (0) 1.5 (0) 1.5 (0) 165.8 166.2
17 75 (1) 20 (1) 2 (1) 2 (1) 220.8 228.1
18 55 (−1) 10 (−1) 2 (1) 2 (1) 216.2 225.3
19 65 (0) 15 (0) 1.5 (0) 1.5 (0) 259.4 258.3
20 55 (−1) 10 (−1) 1 (−1) 1 (−1) 205.2 209.7
21 55 (−1) 10 (−1) 1 (−1) 2 (1) 212.0 215.5
22 75 (1) 20 (1) 1 (−1) 2 (1) 193.1 189.7
23 65 (0) 15 (0) 1.5 (0) 1.5 (0) 260.2 258.2
24 55 (−1) 20 (1) 2 (1) 2 (1) 256.7 257.6
25 55 (−1) 10 (−1) 2 (1) 1 (−1) 147.3 151.7
26 75 (1) 10 (−1) 1 (−1) 1 (−1) 219.3 219.5
27 65 (0) 15 (0) 1.5 (0) 2.5 (2) 225.4 216.2
28 65 (0) 15 (0) 1.5 (0) 1.5 (0) 249.5 258.3
29 65 (0) 25 (2) 1.5 (0) 1.5 (0) 216.1 211.3
30 75 (1) 10 (−1) 1 (−1) 2 (1) 244.5 250.7

ANOVA for quadratic model

Source SS df F value P value

Model 45,750.3 14 46.16 <0.0001 Model R2 value = 0.9773 Adj R2 value = 0.9561
A 583.12 1 8.24 0.0117 Pred R2 value = 0.8902 Adeq precision = 25.056
B 12.76 1 0.18 0.6772
C 581.15 1 8.21 0.0188
D 8055.67 1 113.8 <0.0001
AB 150.68 1 2.13 0.1652
AC 2732.68 1 38.6 <0.0001
AD 638.83 1 9.02 0.0089
BC 6524.60 1 92.17 <0.0001
BD 990.68 1 14.00 0.0020
CD 4613.81 1 65.18 <0.0001
A2 11,584.33 1 163.65 <0.0001
B2 4016.51 1 56.74 <0.0001
C2 1150.33 1 16.25 0.0011
D2 10,618.88 1 150.01 <0.0001
Residual 1061.81 15
Lack of fit 836.02 10 1.85 0.2576
Pure error 225.79 5
Cor total 46,812.12 29

where, Y represents ␤-glucosidase production (U g−1 ), and A, B, C 3.2. Optimization and confirmation experiments
and D are initial moisture (%), substrate amount (g), amount of
NaOH (M) and fungal inoculum ratio (A. fumigatus: T. reesei) respec- Using Design Expart 8.0.3, numerical optimization subroutine
tively. ANOVA for ␤-glucosidase production indicated ‘F value’ of design space was explored with a fitted quadratic model to arrive
46.16, which implied that the model was appropriate. Model terms at optimum factor concentration. The goals for the variables were
having ‘Prob > F’ values less than 0.05 are considered to be signif- set as “in range”, varied from −1 level to +1 level while for the yield
icant. According to Table 1, the R2 value of 0.9773 was in good of ␤-glucosidase production, it was set as “maximize.” The opti-
agreement with the adjusted R2 value of 0.9561. The ‘adequate pre- mized variables were found using a desirability objective function
cision’ value of 25.056 indicated an adequate signal and suggested that assigns relative importance to the responses. Solutions with
that the model can be used to navigate the design space (Table 1). higher desirability gave optimum initial moisture (60.7%), substrate
 A. Das et al. / Industrial Crops and Products 46 (2013) 217–225 221

Table 2 was the potential of yielding relatively non-toxic hydrolyzates with

Adsorption parameters of ␤-glucosidase to the Langmuir model.
higher sugar content.
Substrate Emax (mg mg−1 Kad Catalytic efficiency There was a significant difference in sugar content obtained
of substrate) (mg mL−1 ) (Emax /Kad (mg−1 ) from biomass pre-treated with different alkali concentrations
Pretreated rice straw 0.878 3.033 0.289 (1–5 M). The free reducing sugar obtained during hydrolysis of rice
Untreated rice straw 0.44 4.864 0.090 straw increased with increasing enzyme concentration during pre-
treatment with a maximum sugar concentration of 19.9 g L−1 at 2 M
NaOH concentration (Fig. 3A). The substrate analysis for residual
amount (13.8 g), NaOH concentration of 2 M for pretreatment and cellulose, hemicellulose and lignin content after pretreatment also
the ratio of A. fumigatus ABK9 and T. reesei SAP 3 of 1.86. Under these indicated that that alkaline hydrolysis was more effective in remov-
conditions, confirmation experiments were conducted in three ing lignin, from the lignocellulosic feedstock and improving the
replicates. The observed mean of ␤-glucosidase activity was found accessibility to cellulose during subsequent enzymatic hydrolysis
to be 265.4 U g−1 , which were largely consistent with the predicted (Table S1).
values. Brijwani et al. (2010) also reported that co-culturing of T.
reesei with A. oryzae in 1:1 ratio significantly improved the produc-
tion of ␤-glucosidase levels in SSF of soybean hulls supplemented 3.4. Effect of substrate concentration on reducing sugar yield
with wheat bran compared to mono culture of T. reesei. In our
experiment, along with the ␤-glucosidase activity, a high amount Studies on the effect of biomass loading and enzyme doses on
of endoglucanase (988.6 U g−1 ), FPase (120.2 U g−1 ) and xylanase reducing sugar yield showed that biomass loading (NaOH pre-
(1177.8 U g−1 ) were also found in the crude enzyme preparation treated) of 10% (w/v) gave maximum reducing sugar yield of
(Fig. 1). This high yield of xylanase was another important finding 21.4 g L−1 with the enzyme concentration of 40 U mg−1 . It was also
of the study which ensures the successful bioconversion of xylan, found that an increase in the rice straw concentration (beyond
a major component of plant hemicellulose into xylose monomers 10 g %) resulted in a decrease in the degree of cellulose con-
that also can be utilized as carbon source by Z. mobilis during sub- version (Fig. 3B). This may be due to product inhibition (Cara
sequent ethanol production (Rogers et al., 1982). et al., 2007; Hodge et al., 2008) or inhibition by other compounds
such as sugar derivatives (furfural and hydroxymethylfurfural) and
lignin (Panagiotou and Olsson, 2007; Pan, 2008). The limitation of
3.3. Effect of substrate pretreatment on reducing sugar yield
mass transfer or other effects may related to the increased con-
tent of insoluble solids and non-productive adsorption of enzymes
The effect of different oxidizing (H2 O2 ), acidic (H2 SO4 ) or alka-
(Sorensen et al., 2006; Rosgaard et al., 2007).
line (NaOH) agents on substrate pretreatment for maximum sugar
yield after subsequent enzymatic hydrolysis with 40 U g−1 of ␤-
glucosidase, was investigated over a time period of 48 h. Result 3.5. Optimization of enzyme concentration, incubation time and
indicated that NaOH pretreatment was superior than others with pH on reducing sugar yield
improved the saccharification yield (Fig. 2). In respect to other
chemical pretreatments, alkali pretreatment causes swelling, lead- Generally cellulases penetrate into the substrate to access
ing to an increase in internal surface area, and preserves at least part and hydrolyse, unlike many common enzymes which take their
of the hemicellulose and acts mainly to remove the lignin content substrates into the active site pockets. Cellulases have specific
(Sun and Cheng, 2002; Cao et al., 2012). This pretreatment pro- domains for binding with their substrate so that the enzyme sits
cess favors the enzymatic attack by increasing the surface area and on the polymer and causes a slow degradation (Lynd et al., 2002).
removing the enzyme inhibitor, lignin. Thus, alkali pretreatment For this purpose, initial experiments were conducted to select

1400
CMCase FPase β-glucosidase Xylanase

1200
Enzyme activity (U/g)

1000

800

600

400

200

0
0 24 48 72 96 120 144
Fermentation time (h)

Fig. 1. Profile of cellulolytic enzymes production after validation of the response surface design model.
222 A. Das et al. / Industrial Crops and Products 46 (2013) 217–225

120

100

Reducing sugar yield (%)

80

60

40

20

0
1 2 3 4 5 1 2 3 4 5 1 2 3 4 5
Acid pretreatm ent (M) Alkali pretreatm ent (M) Hydrogen peroxide
pretreatm ent (%)

Fig. 2. Yield of reducing sugar from rice straw after acid (H2 SO4 ), alkali (NaOH) and peroxide (H2 O2 ) treatment at different concentrations.

the optimum enzyme concentration for pretreatment method. medium is also an important factor as it regulates the three
Fig. 3C shows that the sugar yield from pretreated rice straw was dimensional structure of the active site of enzyme as well as the
increased sharply with the incubation time up to 24 h while its formation of enzyme substrate complex. In the current study,
rate of increment reduced considerably thereafter. The yield of mild acidic pH of 5.5 was found to be the best for enzymatic
reducing sugar was also improved with increase in ␤-glucosidase saccharification (24.9 g L−1 ) of the pretreated rice straw (Fig. 3D).
concentration from 20 to 40 U g−1 . pH of the saccharification This is due to the optimum pH of this enzyme was around 5.5.

Fig. 3. Combinational effects of different factor concentration on reducing sugar yield during enzymatic saccharification.
 A. Das et al. / Industrial Crops and Products 46 (2013) 217–225 223

Table 3
Box–Behnken response surface design and the analysis of variances (ANOVA) for optimization of ethanol production by mixed fungal fermentation with S. cerevisiae MTCC
173 and Z. mobilis MTCC 2428 through submerged fermentation.

Run no. (A) Fermentation (B) Fermentation (C) Inoculum (D) Inoculant Ethanol yield Ethanol yield
temperature (◦ C) pH (%) ratio (Y:Z) Observed (g L−1 ) Predicted (g L−1 )

1 30 (0) 6.0 (0) 2 (0) 1.5 (0) 41.1 41.0

2 30 (0) 6.0 (0) 2 (0) 1.5 (0) 40.6 41.0
3 30 (0) 6.0 (0) 2 (0) 1.5 (0) 41.1 41.0
4 30 (0) 6.5 (1) 2 (0) 1.0 (−1) 30.3 30.5
5 35 (1) 6.0 (0) 1 (−1) 1.5 (0) 14.1 14.6
6 25 (−1) 6.0 (0) 2 (0) 1.0 (−1) 15.6 16.5
7 35 (1) 6.0 (0) 3 (1) 1.5 (0) 17.0 17.7
8 30 (0) 5.5 (−1) 3 (1) 1.5 (0) 30.1 30.4
9 35 (1) 6.0 (0) 2 (0) 1.0 (−1) 17.6 17.2
10 30 (0) 5.5 (−1) 2 (0) 2.0 (1) 31.4 31.8
11 30 (0) 6.5 (1) 3 (1) 1.5 (0) 25.7 26.5
12 30 (0) 6.0 (0) 1 (−1) 1.0 (−1) 27.2 27.1
13 30 (0) 6.0 (0) 2 (0) 1.5 (0) 40.7 41.0
14 25 (−1) 6.0 (0) 2 (0) 2.0 (1) 16.6 17.5
15 25 (−1) 6.0 (0) 3 (1) 1.5 (0) 16.3 16.4
16 30 (0) 6.0 (0) 2 (0) 1.5 (0) 41.1 41.0
17 35 (1) 6.0 (0) 2 (0) 2.0 (1) 19.0 18.5
18 35 (1) 5.5 (−1) 2 (0) 1.5 (0) 15.9 15.4
19 30 (0) 6.0 (0) 3 (1) 1.0 (−1) 32.7 31.7
20 30 (0) 6.0 (0) 3 (1) 2.0 (1) 32.1 30.9
21 35 (1) 6.5 (1) 2 (0) 1.5 (0) 14.7 14.2
22 25 (−1) 5.5 (−1) 2 (0) 1.5 (0) 14.2 13.4
23 30 (0) 5.5 (−1) 2 (0) 1.0 (−1) 25.8 26.0
24 30 (0) 6.5 (1) 2 (0) 2.0 (1) 26.6 27.0
25 25 (−1) 6.5 (1) 2 (0) 1.5 (0) 16.1 14.8
26 30 (0) 6.5 (1) 1 (−1) 1.5 (0) 27.5 27.6
27 25 (−1) 6.0 (0) 1 (−1) 1.5 (0) 14.2 14.9
28 30 (0) 6.0 (0) 1 (−1) 2.0 (1) 30.6 30.3
29 30 (0) 5.5 (−1) 1 (−1) 1.5 (0) 24.4 24.0

ANOVA for quadratic model

Source SS df F value P value

Model 2530.09 14 257.37 <0.0001 Model R2 value = 0.9962 Adj R2 value = 0.9923
A 2.34 1 3.33 0.0893 Pred R2 value = 0.9790 Adeq precision = 45.760
B 0.068 1 0.096 0.7611
C 21.07 1 30.00 <0.0001
D 4.20 1 5.98 0.0283
AB 2.40 1 3.42 0.0856
AC 0.16 1 0.23 0.6405
AD 0.040 1 0.057 0.8148
BC 14.06 1 20.03 0.0005
BD 21.62 1 30.79 <0.0001
CD 4.00 1 5.70 0.0317
A2 2324.28 1 3310.42 <0.0001
B2 365.38 1 520.36 <0.0001
C2 261.99 1 373.12 <0.0001
D2 140.58 1 200.20 <0.0001
Residual 9.83 14
Lack of fit 9.05 10 4.65 0.0759
Pure error 0.78 4
Cor total 2539.92 28

3.6. Kinetics of enzyme adsorption on pretreated rice straw represented in Table 2. The alkaline pretreatment of the straw
increased the Emax by 2 fold and decreased the adsorption coeffi-
Elucidation of the adsorption behavior of cellulases onto cel- cient Kad by 37.64% in respect to the untreated straw. The catalytic
lulosic substrates made a significant contribution toward the efficiency of pretreated straw (0.289 mg−1 ) was also higher than
development of efficient cellulase utilizing processes. An adsorp- the untreated one (0.090 mg−1 ). These values are good indicator
tion isotherm in Fig. S1 showed the changes in mass of adsorbed of higher binding efficiency of the enzyme onto the pretreated
proteins on native and alkali treated straw. From the Langmuir rice straw (Mamma et al., 2009). These results also indicated that
adsorption model (Eq. (3)), the values for Emax (maximum enzyme pretreatments lead to higher cellulase adsorption onto cellulose
adsorption) and Kad (adsorption coeffecient) were determined and microfibrils that resulted in higher enzymatic hydrolysis of rice
224 A. Das et al. / Industrial Crops and Products 46 (2013) 217–225

(spectrum c). It indicated that partial hydrogen bonds of cellu-

lose were destroyed, leading to enhance accessibility of cellulose
to enzyme. The band at 2903 cm−1 represents the C H stretching,
the decrease of which indicated that methyl and methylene esters
of cellulose were also ruptured.

3.9. Optimization of ethanol production by S. cerevisiae and Z.

mobilis using rice straw hydrolysate

Box–Behnken response surface methodology is generally used

to investigate a combined effect of several variables and to find
the optimum conditions. The fitness of the models was expressed
by coefficient determination, R2 , which was found to be 0.9962
for ethanol production (Table 3). This value indicated 99.62% of
the response was due to interactions among studied variables.
The P-value < 0.0001 also indicated that the model was highly sig-
nificant. The responses reveal that C, D, BC, BD, A2 , B2 , C2 , and
D2 are significant model terms. The response surface plots indi-
Fig. 4. FTIR spectra of the untreated (a), pretreated (b), and pretreated+enzyme
treated (c) rice straw. cated the interaction between fermentation pH and inoculum (%),
fermentation pH and inoculum ratio (Yeast: Zymomonas) were
more significant for ethanol production (Table 3). The experimental
straw. Further, their effectiveness for hydrolysis is strongly influ- results showed that an optimum production of alcohol (40.1 g L−1 )
enced by pretreatment efficacy and its effect on substrate features was achieved at 30 ◦ C, medium pH of 5.9, inoculum concentration
(Kumar et al., 2009). of 2.1% and Yeast: Zymomonas ratio of 1.53. The synergistic effect
of S. cerevisiae and Z. mobilis was also mentioned by Szambelan
3.7. SEM analysis et al. (2004). These characteristics of the organisms are mutually
beneficial in fermenting sugars even under anaerobic condition
The changes in surface morphology of rice straw after alkali that may be created during growth phases. The amount of ethanol
and enzyme treatment were observed through scanning electron produced in the current experiment was quite higher than some
microscopy (Fig. S2). The lignin and hemicellulose were partially recent reports where the similar type of agro-wastes were utilized
removed by NaOH pretreatment indicated by the formation of lin- as the substrates (Ruiz et al., 2012; Dubey et al., 2012; Rodrigues
ear cavities along through the surface of straw. Further, a large et al., 2011). A critical comparison of lignocellulosic bioethanol pro-
fraction of cellulose was removed by enzymatic hydrolysis. Initially duction with other reports cannot be drawn because the chemical
rice straw was seen to smoothened surface and later enzymatic compositions of the solid substrates strongly influence both cellu-
treatment leads to swelling possibly due to removal of the portion lases and ethanol production. Further, the differences in ethanol
of micro fibrils and absorption of water (Anish et al., 2004). Thus, yields could be related to the type of organism, culture vessel,
the chemical treatments disrupted part of the lignocellulose back production technique and unit of operational cost for bioethanol
bone of rice straw, exposed the cell wall network and facilitated production (Suresh, 2012).
the catalysis of cellulase for release of high amount of sugars.

4. Conclusion
3.8. FTIR analysis
Results indicated that mixed fungal fermentation of A. fumiga-
FTIR was used to analyze the change in bonding pattern of
tus ABK9 and T. reesei SAP 3 produced an enzyme cocktail with
untreated (spectrum a), NaOH-pretreated (spectrum b) and NaOH,
high cellulolytic activities after statistical optimization of SSF. High
followed by cellulase pretreated (spectrum c) rice straws (Fig. 4).
sugar yield of 24.9 g L−1 was obtained from the NaOH pre-treated
In the structure of native straw, generally lignin remain associated
rice straw after enzymatic treatment of 40 U g−1 . Conversion of
with cellulose and hemicellulose by certain chemical bonds, such as
the straw hydrolysate to bio-ethanol (40.1 g L−1 ) by mixed fermen-
␣-ether bonds, phenyl glycosidic linkages, acetal linkages, and ester
tation of S. cerevisiae MTCC 173 and Z. mobilis MTCC 2428 (ratio
bonds (He et al., 2008). From the FTIR analysis, it was observed that
of 1.53) can be an alternative sustainable approach to utilize this
the absorption at 1167 cm−1 , which is an indication of ester bond
agricultural waste for reduction of the ever increasing demand of
stretching, disappeared after NaOH treatment (spectrum a and b).
biofuels.
Such a reaction had broken the ester bond linkage between lignin
and cellulose, and released lignin from the cellulose back bone and
made cellulose more accessible for enzymatic saccharification. Car- Acknowledgements
bonyls mainly occur in the side chains of lignin structural units
and are also an important functional group in the side chains that Authors are thankful to the University Grant Commission [Sanc-
are either aldehyde groups lying in C-␥ or keto lying in C-␤. The tion No: F- 3/2007 (BSR)/11-114/2008 (BSR) and CSIR, New Delhi,
disappearance of spectral band at 1720 cm−1 that assigned to the India, for the financial contribution in this study.
carbonyl (C O) stretching of unconjugated ketones indicated that
the side chain of lignin was also broken down during NaOH treat-
ment. The band at 1356 cm−1 was attributed to phenol hydroxyl Appendix A. Supplementary data
stretching. This band also disappeared after NaOH treatment due
to the reaction of phenol hydroxyl with NaOH (He et al., 2008). The Supplementary data associated with this article can be
absorption at 3415 cm−1 represents the stretching of–OH groups, found, in the online version, at http://dx.doi.org/10.1016/
which was reduced after NaOH followed by enzymatic treatment j.indcrop.2013.02.003.
 A. Das et al. / Industrial Crops and Products 46 (2013) 217–225 225

References Kumar, R., Mago, G., Balan, V., Wyman, C.E., 2009. Physical and chemical character-
izations of corn stover and poplar solids resulting from leading pretreatment
Anish, R., Ahmad, A., Rao, M.B., Rahman, M.S., Trivedi, C., 2004. A process for technologies. Bioresour. Technol. 100, 3948–3962.
the biofinishing of denims in textile industry. Indian Patent Application No. Lowry, O.H., Rosenbrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurements
1299/DEL. with the Folin phenol reagent. J. Biol. Chem. 193, 265–272.
AOAC, 1990. In: Helrick, K. (Ed.), Association of Official Analytical Chemists. Official Lynd, L.R., Weimer, P.J., van Zyl, W.H., Pretorious, I.S., 2002. Microbial cellulase
Methods of Analysis. , 15th ed. AOAC, Arlington, VA. utilization: fundamentals and biotechnology. Microbiol. Mol. Biol. Rev. 66,
Bailey, M.J., Biely, P., Poutanen, K., 1992. Interlaboratory testing of methods for assay 506–577.
of xylanase activity. J. Biotechnol. 23, 257–270. Mamma, D., Hatzinikolaou, D., Kekos, D., Stamatis, H., Kalogeris, E., 2009. Adsorption
Bergmeyer, H.U., 1974. Methods of Enzymatic Analysis, 2nd ed. Verlag Chemie, of major endoglucanase from Thermoascus aurantiacus on cellulosic substrates.
Weinheim/Berlin, p. 1025. World J. Microbiol. Biotechnol. 25, 781–788.
Binod, P., Kuttiraja, M., Archana, M., Janu, K.U., Sindhu, R., Sukumaran, R.K., Pandey, Mandels, M., Andreotti, R., Roche, C., 1976. Measurement of saccharifying cellulase.
A., 2012. High temperature pretreatment and hydrolysis of cotton stalk for pro- Biotechnol. Bioeng. Symp. 6, 21–33.
ducing sugars for bioethanol production. Fuel 92, 340–345. Miller, G.L., 1959. Use of dinitrosalicylic acid reagent for determination of reducing
Box, G.E.P., Behnken, D.W., 1960. Some new three level designs for the study of sugar. Anal. Chem. 3, 426–428.
quantitative variables. Technometrics 2, 455–475. Pan, X.J., 2008. Role of functional groups in lignin inhibition of enzymatic hydrolysis
Brijwani, K., Oberoi, H.S., Vadlani, P.V., 2010. Production of cellulolytic enzyme sys- of cellulose to glucose. J. Biobased Mater. Bioenergy 2, 25–32.
tem inmixed culture solid-state fermentation of soybean hulls supplemented Panagiotou, G., Olsson, L., 2007. Effect of compounds released during pretreatment
with wheat bran. Process Biochem. 45, 120–128. of wheat straw on microbial growth and enzymatic hydrolysis rates. Biotechnol.
Cao, W., Sun, C., Liu, R., Yin, R., Wu, X., 2012. Comparison of the effects of five pretreat- Bioeng. 96, 250–258.
mentmethods on enhancing the enzymatic digestibility and ethanol production Pierre, G., Maache-Rezzoug, Z., Sannier, F., Rezzoug, S.A., Maugard, T., 2011. High-
from sweet sorghum bagasse. Bioresour. Technol. 111, 215–221. performance hydrolysis of wheat straw using cellulase and thermomechanical
Cara, C., Moya, M., Ballesteros, I., Negro, M.J., Gonzalez, A., Ruiz, E., 2007. Influence pretreatment. Proc. Biochem. 46, 2194–2200.
of solid loading on enzymatic hydrolysis of steam exploded or liquid hot water Rodrigues, T.H.S., Rocha, M.V.P., Ribeiro de Macedo, G., Gonçalves, L.R.B., 2011.
pretreated olive tree biomass. Process Biochem. 42, 1003–1009. Ethanol production from cashew apple bagasse: improvement of enzymatic
Chiaramonti, D., Rizzo, A.M., Prussi, M., Tedeschi, S., Zimbardi, F., Braccio, G., Viola, hydrolysis by microwave-assisted alkali pretreatment. Appl. Biochem. Bio-
E., Pardelli, P.T., 2011. 2nd generation lignocellulosic bioethanol: is torrefaction technol. 164, 929–943.
a possible approach to biomass pretreatment? Biomass Conv. Bioref. 1, 9–15. Rogers, P.L., Lee, K.J., Skotnicki, M.L., Tribe, D.E., 1982. Ethanol production by
Das, A., Ghosh, U., 2009. Solid state fermentation of waste cabbage by Penicillium Zymomonas mobilis. Adv. Biochem. Eng. 23, 37–84.
notatum NCIM NO-923 for production and characterization of cellulases. J. Sci. Rosgaard, L., Andric, P., Dam-Johansen, K., Pedersen, S., Meyer, A.S., 2007. Effects
Ind. Res. 68, 714–718. of substrate loading on enzymatic hydrolysis and viscosity of pretreated barley
Das, A., Paul, T., Halder, S.K., Jana, A., Maity, C., Das Mohapatra, P.K., Pati, B.R., Mondal, straw. Appl. Biochem. Biotechnol. 143, 27–40.
K.C., 2013. Production of cellulolytic enzymes by Aspergillus fumigatus ABK9 in Ruiz, H.A., Silva, D.P., Ruzene, D.S., Lima, L.F., Vicente, A.A., Teixeira, J.A., 2012.
wheat bran-rice straw mixed substrate and use of cocktail enzymes for deinking Bioethanol production from hydrothermal pretreated wheat straw by a floc-
of waste office paper pulp. Bioresour. Technol. 128, 290–296. culating Saccharomyces cerevisiae strain – effect of process conditions. Fuel 95,
Dubey, A.K., Gupta, P.K., Garg, N., Naithani, S., 2012. Bioethanol production from 528–536.
waste paper acid pretreated hydrolyzate with xylose fermenting Pichia stipites. Sorensen, I., Pedersen, S., Meyer, A.S., 2006. Optimization of reaction conditions
Carbohydr. Polym. 88, 825–829. for enzymatic viscosity reduction and hydrolysis of wheat arabinoxylan in an
Gawande, P.V., Kamat, M.Y., 1999. Production of Aspergillus xylanase by lignocellu- industrial ethanol fermentation residue. Biotechnol. Prog. 22, 505–513.
losic waste fermentation and its application. J. Appl. Microbiol. 87, 511–519. Sukumaran, R.K., Singhania, R.R., Mathew, G.M., Pandey, A., 2009. Cellulase
Gray, K.A., Zhao, L., Emptage, M., 2006. Bioethanol. Curr. Opin. Chem. Biol. 10, production using biomass feed stock and its application in lignocellulose sac-
141–146. charification for bio-ethanol production. Renew. Energy 34, 421–424.
He, Y., Pang, Y., Liu, Y., Li, X., Wang, K., 2008. Physicochemical characterization of rice Sun, Y., Cheng, J., 2002. Hydrolysis of lignocellulosic materials for ethanol produc-
straw pretreated with sodium hydroxide in the solid state for enhancing biogas tion: a review. Bioresour. Technol. 83, 1–11.
production. Energy Fuels 22, 2775–2781. Sun, X., Liu, Z., Qu, Y., Li, X., 2008. The Effects of wheat bran composition on the
Hodge, D.B., Karim, M.N., Schell, D.J., McMillan, J.D., 2008. Soluble and insoluble production of biomass-hydrolyzing enzymes by Penicillium decumbens. Appl.
solids contributions to high-solids enzymatic hydrolysis of lignocellulose. Biore- Biochem. Biotechnol. 146, 119–128.
sour. Technol. 99, 8940–8948. Suresh, P.V., 2012. Biodegradation of shrimp processing bio-waste and concomitant
Jabasingh, S.A., Nachiyar, C.V., 2011. Utilization of pretreated bagasse for the sustain- production of chitinase enzyme and N-acetyl-d-glucosamine by marine bacte-
able bioproduction of cellulase by Aspergillus nidulans MTCC344 using response ria: production and process optimization. World J. Microbiol. Biotechnol. 28,
surface methodology. Ind. Crops Prod. 34, 1564–1571. 2945–2962.
Kar, S., Gauri, S.S., Das, A., Jana, J., Mandal, A., 2012. Process optimization of xylanase Szambelan, K., Nowak, J., Czarnecki, Z., 2004. Use of Zymomonas mobilis and Sac-
production using cheap solid substrate by Trichoderma reesei SAF3 and study on charomyces cerevisiae mixed with Kluyveromyces fragilis for improved ethanol
the alteration of behavioral properties of enzyme obtained from SSF and SmF. production from Jerusalem artichoke tubers. Biotechnol. Lett. 26, 845–848.
Bioprocess. Biosyst. Eng., http://dx.doi.org/10.1007/s00449-012-0761-x Wood, T.M., Bhat, K.M., 1988. Methods for measuring cellulase activity. In: Wood,
Kova’cs, K., Szakacs, G., Zacchi, G., 2009. Comparative enzymatic hydrolysis of T.M., Kellogg, S.T. (Eds.), Methods in Enzymology, vol. 160. Academic Press Inc.,
pretreated spruce by supernatants, whole fermentation broths and washed London, pp. 87–112.
mycelia of Trichoderma reesei and Trichoderma atroviride. Bioresour. Technol. Yang, B., Wyman, C.E., 2008. Pretreatment: the key to unlocking low-cost cellulosic
100, 1350–1357. ethanol. Biofuels Bioprod. Bioref. 2, 26–40.