You are on page 1of 10

Journal of Cleaner Production 113 (2016) 1005e1014

Contents lists available at ScienceDirect

Journal of Cleaner Production


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

The cellulose structural transformation for higher enzymatic


hydrolysis by ionic liquids and predicting their solvating capabilities
Tirath Raj a, Manali Kapoor a, Surbhi Semwal a, Sunitha Sadula a, Vibhav Pandey b,
Ravi P. Gupta a, Ravindra Kumar a, *, Deepak K. Tuli a, B.P. Das c
a
DBT-IOC Centre for Advanced Bioenergy Research, Research & Development Centre, Indian Oil Corporation Limited, Sector-13, Faridabad 121007, India
b
Analytical Division, Research & Development Centre, Indian Oil Corporation Limited, Sector-13, Faridabad 121007, India
c
Indian Oil Corporation Limited, Research and Development Centre, Sector-13, Faridabad 121007, India

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

Article history: Lignocellulosic material (LCM) is promising alternative resource for sustainable energy production such
Received 10 June 2015 as ethanol and butanol and biohydrogen. Cellulose is an abundant renewable polymer of LCM found in
Received in revised form plant cell walls (30e50%). The high crystallinity of cellulose makes it recalcitrant to hydrolysis into its
31 October 2015
individual sugar subunits for biofuels production. Moreover, ionic liquids are considered as green sol-
Accepted 5 December 2015
Available online 22 December 2015
vents and have been used for biomass solubilization. The present work describes three properties:
KamleteTaft (KeT) parameters, viscosity and surface tension of ve imidazolium-based ionic liquids
(ILs); namely [C2mim][OAc], [C4mim][OAc], [C2mim][Cl], [C4mim][Cl] and [C4mims][BF4], and their ef-
Keywords:
Ionic liquids
ciency in the cellulose structural transformation for improved enzymatic glucose recovery. Crystalline
KeT parameter cellulose was treated with ILs at two different temperatures, i.e. 100 and 130  C for 5 and 2 h, respec-
Kinematic viscosity tively, with 10% solid loading followed by enzymatic saccharication using 10 and 20 FPU/g substrate of
Surface tension commercial cellulases. ILs treatment of crystalline cellulose signicantly reduces the crystallinity, which
Cellulose transformation resulted in a very sharp increase of sugar yields after enzymatic saccharication. Cellulose treated for
Saccharication 130  C/2 h resulted in better glucose yields as compared to 100  C/5 h. ILs comprising acetate anion
resulted in highest glucose yields and chloride based ILs performed moderately, whereas BF 4 based IL
was ineffective in transforming the cellulose structure. In order to decipher the possible reasons of
varying efciency of these ILs, the KeT parameters; hydrogen bond acidity (a), hydrogen bond basicity
(b), solvent polarizability (p*), kinematic viscosity (h) and surface tension (s) were calculated for 100 and
130  C. These results show that, among all the properties of ILs, hydrogen bond basicity (b) is relatively
more important than kinematic viscosity and surface tension for impacting the structural transformation
and subsequent enzymatic hydrolysis. [C2mim][OAc] with high b value (1.32) and lower viscosity (4.4 cSt/
s) and surface tension (30.3 mN/m) was found to be most efcient in cellulose transformation resulting
in higher glucose yields (89.8%) upon saccharication. Effect of size of cation and anion of ILs and
properties of regenerated cellulose is also examined by PXRD and FT-IR to further support the ndings.
2015 Elsevier Ltd. All rights reserved.

1. Introduction production, however, present sources, i.e. sugar cane molasses,


juice or corn are not sufcient to produce required quantities of
The development of green and cleaner techniques for biomass ethanol. Therefore, lignocellulosic material (LCM), which is abun-
processing and fractionation is crucial from the point of view of dantly available is being exploited for ethanol, butanol and biogas
sustainability and environmental perspective. Ethanol is a proven production, as it comprises 35e50% of cellulose, 20e35% hemicel-
cleaner biofuel, which imparts desirable properties when blended lulose and 5e30% lignin along with 1e15% of extractives (Jiang
with gasoline. Almost all countries have set targets for ethanol ki-Arvela et al., 2010; Raj et al.,
et al., 2011; Kumar et al., 2015; Ma
2015; Wada et al., 2010). In the US, for example, renewable fuel
standards have mandates that by 2022 out of 36 billion gallons of
* Corresponding author. Tel.: 91 0129 2294463; fax: 91 01292286221. ethanol, 16 billion gallons will come from LCM for cleaner and
E-mail address: kumarr3@indianoil.in (R. Kumar). sustainable development (Geng and Henderson, 2014). Production

http://dx.doi.org/10.1016/j.jclepro.2015.12.037
0959-6526/ 2015 Elsevier Ltd. All rights reserved.
1006 T. Raj et al. / Journal of Cleaner Production 113 (2016) 1005e1014

of ethanol from LCM follows three step process: pretreatment, cellulose microbrils. The imidazolium cation has been proposed to
enzymatic saccharication and fermentation (Hassan et al., 2013; have hydrophobic interactions with the hydrophobic face of the
Sharma et al., 2015). In lignocellulosic biomass, cellulose is tightly cellulose. The C-2 proton in the imidazolium has been simulated to
bound to lignin and hemicellulose through strong inter/intra- interact as a weak hydrogen bond donor with the cellulose hy-
molecular hydrogen and covalent bonding, which makes it recal- droxyl groups during dissolution and cation acidity has recently
citrant for chemical or biochemical degradation (Brandt et al., 2011; been suggested to be an important parameter for predicting cel-
Sissine, 2007; Virunanon et al., 2013). lulose solubility in certain cases (Brandt et al., 2013).
Pretreatment is the key step to overcome the lignocellulosic The ability of ILs to dissolve and transform cellulose efciently
stringent seal around cellulose and make them susceptible for hy- depends on their physicochemical and solvent properties such as,
drolysis (Haghighi Mood et al., 2013; Hou et al., 2015). Further steps size of cation and anion, KamleteTaft parameters: hydrogen bond
involve isolation and enzymatic hydrolysis of cellulose and hemi- acidity (a), hydrogen bond basicity (b) and polarizability (p*) and
cellulose to generate monomeric sugars followed by microbial viscosity (h) along with surface tension (s). There are few reports
fermentation and distillation to produce fuel ethanol (Bian et al., on the properties of ILs and their impact on cellulose structural
2014; Peleteiro et al., 2014). Over the years, many pretreatment transformation and enzymatic saccharication and are limited to
methods have been intensively studied and are discussed in liter- individual property such as KeT parameters (Parviainen et al.,
ature including chemical (e.g. acid and organosolv process), phys- 2013; Sun et al., 2014; Xia et al., 2014).
icochemical (e.g. steam or carbon dioxide explosion, AFEX), In this study, ve ILs namely; 1-ethyl-3-methylimidazolium
biological (e.g. fungi and actinomycete), and ozonolysis or hot acetate [C2mim][OAc], 1-butyl-3-methylimidazolium acetate
water treatment (Amiri et al., 2014; Jiang et al., 2013; Kapoor et al., [C4mim][OAc], 1-ethyl-3-methylimidazolium chloride [C2mim][Cl],
2015; Kim et al., 2011). Very often these methods require high 1-butyl-3-methylimidazolium chloride [C4mim][Cl] and 1-butyl-3-
temperature and pressure or alternatively strong acids and bases methylimidazolium tetrauoroborate [C4mim][BF4] were analyzed
(Goshadrou et al., 2013). Additionally, these pretreatment processes for their KeT parameters, viscosity and surface tension. We have
result in the formation of microbial inhibiting byproducts such as focused on Avicel PH 101 cellulose as a model substrate which re-
furfural (2-furaldehyde), 5-hydroxymethylfurfural (5-HMF), acetic ects the properties of biomass. After treatment and regeneration
acid along with lignin derived phenolic compounds (4- of cellulose (Avicel PH 101) using all these ILs, the regenerated
hydroxybenzaldehyde, syringaldehyde, etc.) (Bensah and Mensah, cellulose was subjected to enzymatic hydrolysis and the efforts
2013; Haghighi Mood et al., 2013; Kim et al., 2011; Larsson et al., were made to nd the co-relation of the properties of ILs with
1999). Moreover, all these treatment processes result in the envi- enzymatic hydrolysis. Cellulose, thus obtained was characterized
ronmental issues (Goshadrou et al., 2013; Ximenes et al., 2010, by PXRD and FT-IR to nd out structural modications. This is one
2011). To circumvent these drawbacks, ILs are being explored for of the few such studies which will help to choose best IL for
pretreatment of biomass (Dadi et al., 2007). biomass pretreatment, wherein a comprehensive analysis of ILs
ILs are molten salts below 100  C comprising organic cations properties, i.e. KeT parameters, viscosity and surface tension were
and organic or inorganic anions (Hou et al., 2012). ILs have proven conducted under treatment conditions and their impact on the
to be potential, more sustainable and environmentally responsible structure of cellulose and enzymatic hydrolysis was evaluated.
alternatives to organic solvents for many industrial applications
such as: in chemicals synthesis, catalysis, bio catalysis, electro- 2. Material and methods
chemical devices and as engineering uids (Brandt et al., 2013; da
Costa Lopes et al., 2013; Plechkova and Seddon, 2008). These can 2.1. Material
be recycled for consecutive cycles to reduce their high cost and
make the process more practicable (Weerachanchai and Lee, 2014; 1-Ethyl-3-methylimidazolium acetate [C2mim][OAc], (98%), 1-
Xia et al., 2014). Certain imidazolium-based ILs have demonstrated ethyl-3-methylimidazolium chloride [C2mim][Cl], (97%), and 1-
a promising ability for efcient dissolution of biomass and cellulose butyl-3-methylimidazolium acetate [C4mim][OAc], (98%), 1-
regeneration upon addition of an anti-solvent such as water/ butyl-3-methylimidazolium tetrauoroborate [C4mim][BF4],
acetone/ethanol (Brandt et al., 2013; Goshadrou et al., 2013; Xia (98%), 4-Nitroaniline (4NA) (99%) and Reichardt's dye (RD) (dye
et al., 2014). IL pretreatment does not produce degradation prod- content  90%), imidazole (97%), butyl chloride and Avicel (PH
ucts that inhibit enzymes or fermenting microorganisms (Dadi 101) (>98%) were purchased from SigmaeAldrich (India) and used
et al., 2007). The initial hydrolysis rate for regenerated cellulose without any further purication. N,N-Diethyl-4-nitroaniline
obtained after pretreatment with [C4mim][Cl ] was approximately (DENA) (97%), was purchased from Oakwood Chemical Products
50-fold higher as compared to untreated cellulose (Dadi et al., (West Columbia, USA). 1-Butyl-3-methylimidazolium chloride
2007). Li et al. (2010) have reported that, [C2mim][OAc] treatment [C4mim][Cl] was synthesized as per as the synthetic protocol
of switchgrass gave more than 90% of hydrolysis yield after 12 h (Dupont et al., 2003). All experiments were conducted using a
saccharication time as compared to dilute acid treatment, which single lot of Avicel (PH 101). Cellulase enzyme was obtained from
gave 80% of yield after 72 h of saccharication time. Similarly, Singh M/S Advanced Enzymes Technologies Ltd. (Mumbai, India).
et al. (2009) have examined the structural disturbance of switch-
grass after pretreatment with [C2mim][OAc], which enhance the 2.2. Ionic liquid treatment
enzymatic saccharication from 16.5% to 72.5%. In one previous
work, energy cane treated with [C2mim][OAc], resulted in 87.0% of A 10% (w/w) Avicel (PH 101) was prepared by combining 1 g of
enzymatic hydrolysis (Qiu et al., 2012). Table SI 1 shows the Avicel with 9 g of IL in a 50 ml round bottom ask under stirring.
comparative sugar yield from biomass pretreated with different ILs The mixture was then placed in a preheated silicone oil bath on a
(Sant'Ana da Silva et al., 2011; Xiao et al., 2012). hot plate (IKA C-MAG H7 basic, Germany) equipped with a stirrer at
ILs are able to interact with cellulose through a variety of 200e300 rpm for given temperature (100 and 130  C) for desired
different interactions, including dispersive, pep, nep, hydrogen time (5 and 2 h) under sealed conditions, so as to avoid moisture
bonding, dipolar, and ionic/chargeecharge interactions. These in- uptake. All experiments were conducted in triplicate. After treat-
teractions between the basic anion and the cellulose hydroxyl ment, the regeneration of the dissolved cellulose was carried out by
groups helps to break up the hydrogen bond network between the the precipitation of cellulose/IL slurry with 50 ml of hot deionized
T. Raj et al. / Journal of Cleaner Production 113 (2016) 1005e1014 1007

water (60  C) under stirring. The regenerated cellulose was


collected by ltering through a Whatman No. 41 lter paper thor- ET 30  30:7
EN
T (1)
oughly washed with deionized water (5  50 ml) to remove re- 32:4
sidual ILs. The residual ILs were checked after fth wash by
ymax  yo
measuring the absorbance at 240 nm using Shimadzu UVeVis 2401 p* (2)
spectrophotometer of recovered liquid (Li et al., 2013). Regenerated s
cellulose was then stored in a freezer at 4  C for further analysis and where, ET (30) in kcal mol1 is 28,591/lmax (nm); lmax is the
for enzymatic hydrolysis. A portion of regenerated cellulose was air wavelength of RD dye in nm (Doherty et al., 2010). KeT parameter
dried at 30  C and used for characterization. (p*) is obtained (eq. (2)) by measuring the wavelength of maximum
absorbance, ymax in kilo keyser (kK, 103 cm1), of N,N-diethyl-4-
nitroaniline dye (Lee et al., 2008). y0 27.52 kK and s 3.182;
2.3. Enzymatic saccharication here, y0 is the regression value for a reference solvent system and s
is the susceptibility of intensity of spectrum absorption due to the
Enzymatic saccharication of untreated and regenerated cellu- changing solvent polarity. p* provides the measure of polarizability.
lose was carried out using commercially available cellulase en- Hydrogen bond acidity, a was determined using eq. (3).
zymes. All reactions were conducted at 10% loading by placing 
100 mg of regenerated cellulose (on dry weight) in 15 ml glass vials. ET 30  14:6 p*  0:23  30:31
a (3)
Then 0.05 M sodium citrate buffer was added to maintain the pH of 16:5
the solution at 5.0 followed by the addition 10 ml of 0.02% sodium
azide to prevent the growth of any microorganism. The cellulase 1:035*yDENAmax  y4NAmax 2:64
b (4)
preparation SacchariSEBC6 (Advance Enzymes, India) was found to 2:8
contain, b-glucosidase activity: 5930 IU/g, lter paper unit: 578 KeT parameter, (b) hydrogen bond basicity was obtained by
FPU/g, endo-cellulases activity: 4625 IU/g and endo-xylanase ac- measuring the relative difference of solvatochromism between 4NA
tivity: 64867 IU/g. b-glucosidase activity was determined as and DENA. Where, y(DENA)max and y(4NA)max are the wavelength
described previously (J ager et al., 2001). Filter paper units (FPU) and of maximum absorbance of dissolved DENA and 4NA respectively
endoglucanase (CMCase) activity was determined according to the using eq. (4) (Kim et al., 2014; Lee et al., 2008).
method described (Panda et al., 1987). The reaction was carried out
at 50  C for 72 h in a rotatory incubator with a shaking speed of
150 rpm. Aliquot samples with a volume of 50 ml were withdrawn at 2.5. Kinematic viscosity
0, 1, 2, 4, 6, 24, 48 and 72 h. Each aliquot was sealed and incubated
for 5 min in a boiling water bath to denature the cellulase enzyme. Kinematic viscosity of ILs was measured using 20 ml of ILs at 100
The aliquot was then centrifuged at 10,000 g for 5 min. 20 ml of and 130  C directly with a Cannon-Fenske Viscometer (USA) with
aliquot was diluted to 50 times by the addition of deionized water. 300 and 400 bulb sizes and double bubble. The measurements were
The diluted supernatant was then ltered through a 0.45 mm of the performed in triplicates. The temperature of the sample was
nylon syringe lter. Sugar analysis was performed on Waters HPLC maintained to 0.1  C via an external temperature controller. The
(Switzerland) using Aminex HPX-87P column coupled with the Viscometer was placed into a silicone oil bath with a temperature
refractive index detector. Milli-Q water was used as mobile phase at controlled system within 0.2  C using a bath circulator. The
a ow rate of 0.6 ml/min, with a column temperature of 75  C. All equilibrium time was about 15 min. The efux time of the liquid
the enzymatic saccharication experiments were conducted in solution through the capillary was measured manually with a
triplicate. stopwatch (Greases, 2014).

2.6. Surface tension


2.4. KamleteTaft (KeT) parameters
Surface tension measurements were conducted by using a
KamleteTaft parameters were determined according to the model DCAT 11EC Tensiometer and Wilhelmy plate purchased from
previously reported protocol (Sun et al., 2014). The three dyes, 4- Dataphysics. Before every use, the plate was rst rinsed with
nitroaniline (4NA), N, N-diethyl-4-nitroaniline (DENA) and Reich- acetone to remove any material sticking to the plate and washed
ardt's dye (RD) solution were prepared in ethanol to a concentra- with Millipore water. Finally, the plate was heated to red hot with a
tion of 1 mg/ml 2 ml of 4NA, 2 ml of DENA and 20 ml of RD were Bunsen burner and then cooled. Instrument works from 25 to 80  C
pipetted into four separate glass vials and the ethanol was evapo- (Pal et al., 2014). Surface tensions of ILs were measured at 25, 50
rated under stream of dry nitrogen. Dye concentration of 12, 8 and and 75  C and the graph was plotted. The surface tension values for
29 mM respectively, were obtained by adding 1.25 ml of the ILs to 100 and 130  C was obtained by extrapolation of the linear t of the
each vial and mixing on shaker at 300 rpm at 25  C for 30 min. The data obtained.
absorbance spectra at 30, 50 and 60  C of each IL-dye solution was
measured from 350 to 700 nm using a UVeVis dual beam spec-
trophotometer, Spectra Max M5 molecular devices, S/N MV 06585 2.7. Powder X-ray diffractometry (PXRD)
(North America, US) equipped with a temperature controller. These
parameters are temperature dependent and were measured be- The regenerated cellulose was air dried and characterized with
tween 30 and 60  C, which were then extrapolated by linear t to horizontal Rigaku PXRD in Panalytical (Netherlands), X-pert pro
estimate the values at 100 and 130  C (Doherty et al., 2010). diffractometer with acceleration voltage at 40 kW and current at
EN
T scale expresses the solvent polarity arising from overall 30 mA. The radiation was Cu Ka at (l 1.54 ) and grade range
interaction between a solvent and the dye. Therefore, the EN T po- 2q 30e50 with a step size of 0.003 . Crystallinity index (CrI) of
larity is obtained by measuring the maximum absorption in a sol- cellulose was calculated according to the empirical method as given
vent by using following eq. (1) (Kim et al., 2014; Lee et al., 2008). in eq. (5) proposed by Segal et al. (1959).
1008 T. Raj et al. / Journal of Cleaner Production 113 (2016) 1005e1014

highest (97.8%) solid recovery, while [C4mim][OAc] at 130  C/2 h


I  Iamp:
CrI 002 (5) resulted in lowest (86.3%). Similarly, [C4mim][Cl] and [C2mim][Cl]
I002 resulted in 96.0 and 97.4% of solid recovery after 100  C/5 h treat-
ment respectively.
where, CrI is the crystallinity index, (I002) is the highest peak in- Enzymatic saccharication of untreated and treated avicel was
tensity at (002) plane at an angle of diffraction 2q 22.4 and (Iamp) carried out using 10 and 20 FPU/g substrate of cellulase enzymes at
is the intensity diffraction for amorphous cellulose at 2q 16.0 . 10% (w/w) solid loading. Table 1, indicates that after ILs treatment,
higher glucose yields were achieved as compared to untreated
2.8. FT-IR cellulose except in the case of [C4mim][BF4]. Cellulose treated at
100  C/5 h using a 10 FPU/g substrate of enzyme resulted in glucose
Fourier Transforms Infrared Spectroscopy (FT-IR) of cellulose yields in the order of: [C2mim][OAc] (83.3%) > [C4mim][OAc]
samples was recorded using Prestige-21, Model No. A21004802514, (64.9%) > [C2mim][Cl] (61.6%) > [C4mim][Cl] (45.9%) > [C4mim]
FT-IR instrument. All spectra were recorded in the absorbance [BF4] (25.5%). Higher treatment temperature, 130  C/2 h resulted in
mode from an accumulation of 128 scans at 4 cm1 resolution over higher glucose yields with a similar order, as was in the case of
4000400 cm1 range. Samples were analyzed by grinding with 100  C/5 h. Glucose yields of [C4mim][BF4] at both temperatures
KBr (1:100, w/w) and pressing into pellets. were found lower than the untreated cellulose, even with 20 FPU/g
substrate of enzyme loading. The glucose yields using 10 and 20
2.9. Statistical analysis FPU/g substrate of the enzyme and its time course at different in-
tervals are represented in Fig. 1 and Fig. SI 1. Fig. SI 1a, 1b clearly
Statistical analysis was performed by one-way ANOVA followed shows that, increase of the enzyme dose resulted in an increase of
Tukey's HSD post hoc tests using JMP software, trial version (SAS, glucose yields and faster conversion to glucose. Fig 1 and Fig. SI 1
US) and statistical signicance was determined at 0.05 levels show that, [OAc] containing ILs resulted in higher glucose yields
(p  0.05). and higher rate of reaction at both 10 and 20 FPU/g substrate of the
enzyme.
3. Results and discussion Perusal of Table 1 also indicates a clear direct relationship be-
tween lowering of crystallinity by ILs treatment and the extent of
3.1. Treatment and enzymatic saccharication enzymatic hydrolysis. Thus, [C2mim][OAc] was more effective in
lowering the crystallinity of Avicel, from 89.0 to 57.4% and also
ILs treatment is usually carried out between 80 and 150  C with showed best enzyme digestability of 94.4% at 20 FPU/g substrate
time interval of 1 h to few hrs, but beyond 150  C, cellulose solu- (discussed in later part). Results presented in Table 1 show that, ILs
bilizes rapidly, however this also results in loss of sugars. Therefore, having same cation with different anion, i.e. [C2mim][OAc] and
two treatment conditions, i.e. 100  C for 5 h and 130  C for 2 h were [C2mim][Cl] resulted in 83.3% and 61.7% glucose yields at 100  C/5 h
chosen based on literature (Sun et al., 2014). Crystalline cellulose treatment using 10 FPU/g substrate of enzyme respectively.
was treated with ve ILs at 10% loading (w/w) in a 50 ml round Whereas, [C2mim][OAc] and [C4mim][OAc] ILs comprising different
bottom ask and was regenerated by Milli Q water as anti-solvent. cation and same anion resulted in 83.3% and 64.9% glucose yields
Solid recovery was in the range of 86.3e97.8% and the data are respectively under similar process parameters. It is evident from
summarized in Table 1. Solid recovery refers to the percent mass of Table 1, that ILs having a different combination of cation with same
cellulose (on dry basis) recovered from native after treatment and anion resulted in different glucose yield. Under similar treatment
regeneration. Table 1, shows that treatment at 130  C/2 h, resulted and saccharication conditions, [Cl] based ILs performed moder-
in slightly lower solid recovery (86.3e95.5%) as compared to ately, while [BF4] containing IL was ineffective. Therefore, both
100  C/5 h (92.2e97.8%). [C4mim][BF4] at 100  C/5 h resulted in the anion and cation of ILs are the key players to inuence the

Table 1
Solid recovery and glucose yield % of regenerated cellulose obtained.

T/t Solid recovery (%)a Glucose yield (%)b Overall glucose recoveryc (g/g avicel) Crystallinity index (CrI (%)

10 FPU/g substrate 20 FPU/g substrate

Untreated avicel N/A N/A 33.86 0.5 46.19 0.9 0.46 0.01 89.0
[C2mim][OAc] 100/5 93.30 0.5A 83.28 0.9A 88.80 1.0A 0.83 0.02A 67.3
130/2 89.69 0.5A* 89.83 0.8A* 94.48 1.0A* 0.85 0.02A* 57.4
[C4mim][OAc] 100/5 92.23 0.2B 64.92 0.2B 77.77 0.9B 0.72 0.01B 75.3
130/2 86.32 0.9B* 73.87 1.2B* 80.10 0.7B* 0.69 0.03B* 70.6
[C2mim][Cl] 100/5 96.00 0.4C 61.67 0.7B 63.44 1.2C 0.61 0.02C 73.4
130/2 94.74 0.4C* 63.27 1.2C* 71.24 1.3C* 0.67 0.02C* 72.5
[C4mim][Cl] 100/5 97.41 0.1D 45.96 1.5C 50.54 1.0D 0.49 0.01D 75.0
130/2 94.50 0.5C* 50.82 1.0D* 63.44 1.6D* 0.60 0.02D* 73.0
[C4mim][BF4] 100/5 97.83 0.3E 25.53 2.0D 32.72 0.8E 0.32 0.02E 88.5
130/2 95.45 0.3C* 27.33 1.9E* 33.20 1.9E* 0.32 0.03E* 87.8

N/A means not applicable.


Reaction condition: cellulose sample (1 g) was incubated in ILs (9 g) under sealed condition at 100  C/5 h and 130  C/2 h under stirring.
Enzymatic hydrolysis condition: 100 mg untreated/regenerated cellulose, 0.8 ml of citrate buffer (0.05 M, 10 and 20 FPU/g substrate enzyme, 50  C and 150 rpm).
All experiments were done in triplicate and the mean is reported with S.D. Values in the same column for same treatment temperature with different superscript letters
indicate the signicance difference at P  0.05. Letter with star (*) superscript shows statistical signicance at 130  C.
a
Calculated based upon mass of regenerated cellulose divided by mass of cellulose multiply by 100 (dry weight based).
b
Calculated based upon the mg of glucose released after 72 h of enzymatic hydrolysis divided by initial cellulose (0.1 g) multiply by 100 (dry weight basis) using anhydro
correction of 0.9 for glucose.
c
Overall glucose recovery g/g avicel is calculated using solid recovery after treatment based upon total glucose released after 72 h of enzymatic hydrolysis of 1 g cellulose.
T. Raj et al. / Journal of Cleaner Production 113 (2016) 1005e1014 1009

100 100

130 C/ 2 h

Percent of theoritical glucose yield (%)


100 C/ 5 h
10 FPU/g susbstrate 10 FPU/g susbtrate

Percent of theoritical glucose yield (%)


80 80

60 60

40 40

20 20

0
0
0 10 20 30 40 50 60 70
0 10 20 30 40 50 60 70
Time (h)
Time (h)
a b
Fig. 1. Time courses for enzymatic hydrolysis and glucose yield (%) of untreated and ILs treated cellulose. Conditions: 100 mg untreated/regenerated cellulose, 0.8 ml of sodium
citrate buffer 0.05 M, 10 FPU/g substrate, 50  C, 150 rpm, time, 72 h.

treatment efciency. Hence, in order to nd out the role of cation were in the order of: [C2mim][OAc] > [C4mim][OAc] > [C2mim]
and anion, we have measured KeT parameters, viscosity and sur- [Cl] > [C4mim][Cl] > [C4mim][BF4]. In a plot of KeT parameters vs.
face tension to nd the correlation of these properties with glucose 1000/temperature (Fig. 2), the slop of [OAc] containing ILs is
yields. In order to differentiate better and for economic aspects, steeper, showing that b values of ILs increases with increase in
lower enzyme loading, i.e. 10 FPU/g substrate was considered for temperature as compared to p* and a values (Fig. 2a, b and SI 2).
further studies and in detailed discussion. Cellulose treatment in ILs with b > 0.72 resulted in more than
55% glucose yield after 24 h of hydrolysis and ILs with b  0.72,
neither decrease cellulose crystallinity (discussed in later part) nor
3.2. Impact of KamleteTaft (KeT) parameters on enzymatic improve glucose yields. For instance, higher hydrogen bond basicity
saccharication of [C2mim][OAc] (b 1.27) gave highest glucose yield (83.3%) as
compared to [C4mim][OAc] (b 1.19), which resulted in 64.9%
KeT solvent parameters have been in past used to measure the glucose yields. Similarly, [C4mim][Cl] with b 1.16 gave 45.9% of
ability of a solvent to donate a hydrogen bond (a), accept a glucose yield whereas, [C4mim][BF4] having lower b value (0.67)
hydrogen bond (b) (Kamlet and Taft, 1976). It has earlier been gave only 25.5% glucose yield at 100  C/5 h. Similar trends were also
shown (Brandt et al., 2010) that basicity (b) correlates well with an observed for 130  C/2 h (Tables 1 and 2). Fig. 3 shows the correla-
IL's ability to dissolve lignocellulose. A recent study on the range of tion between glucose yield and b values concluding that, cellulose
cations in combination with same anion demonstrated that cation treated with ILs with higher b value leads to higher glucose yields.
acidity is also important for cellulose solubility (Parviainen et al., Higher hydrogen bond basicity (b) of [C2mim][OAc] could stabilize
2013). The KeT parameters describes the hydrogen bond acidity the solvation ion causing further higher solvation of cellulose and
(a), hydrogen bond basicity (b) and polarizability (p*) of ILs (Lee lower in case of [C4mim][BF4] due to low b value. In literature,
et al., 2008). KeT parameters of all ILs were measured at 100 and (Crowhurst et al., 2003; Welton, 1999); it is mentioned that
130  C and presented in Table 2. Since, [C2mim][Cl] has a high hydrogen bond basicity (b) of anion attracts hydroxyl protons of
melting point (ca. 77  C), its KeT parameters could not be deter- cellulose, disrupting the crystal lattice and hydrogen bonding
mined. The polarizability (p*) of ILs slightly decreased with an in- interaction of crystalline cellulose.
crease in temperature except for [C4mim][BF4] and was found to be
in the range of 0.88e0.99. With an increasing alkyl chain length
attached to cation, decrease in p* values was observed. [BF4] 3.3. Impact of viscosity (h) on enzymatic saccharication
containing IL are considered as poor solvents due to low acidity and
basicity (Ma ki-Arvela et al., 2010). Hydrogen bond acidity (a) and Kinematic viscosity (h) of ILs could also be an important factor
hydrogen bond basicity (b) were found to increase with increasing which can inuence the glucose yields. Viscosity of all ve ILs was
temperature (Table 2). The b values of ILs at both temperatures analyzed at 100 and 130  C and the data is presented in Table 3. The

Table 2
Solvatochromic KeT parameter of the ILs.

ILs a b p*
100  C 130  C 100  C 130  C 100  C 130  C
A A* A A* A
[C2mim][OAc] 0.72 0.03 0.79 0.02 1.27 0.0 1.32 0.02 0.99 0.01 0.98 0.01A*
[C4mim][OAc] 0.60 0.04B 0.63 0.01B* 1.19 0.0B 1.23 0.01B* 0.95 0.01B 0.88 0.02B*
[C4mim][Cl] 0.72 0.01A 0.77 0.01A* 1.16 0.0B 1.22 0.01B* 0.91 0.01C 0.90 0.02B*
[C2mim][Cl] N/Aa N/Aa N/Aa N/Aa N/Aa N/Aa
[C4mim][BF4] 0.59 0.01B 0.63 0.2B* 0.67 0.0C 0.72 0.01C* 0.99 0.02A 0.99 0.01A*

KeT parameters using the following set of dye: Reichardt's dye, N,N-diethyl-4-nitroaniline and 4-nitroaniline: (a), hydrogen bond acidity, (b), hydrogen bond basicity, (p*),
polarizability. All experiments were done in triplicate and the mean is reported with S.D. Values in the same column for same temperature with different superscript letters
indicate the signicance difference at P  0.05. Letter with star (*) superscript shows statistical signicance at 130  C.
a
N/A means not applicable, KeT parameter for [C2mim][Cl] could not be determined since the melting point of 1-ethyl-3-methyl imidazolium chloride is 77e75  C.
1010 T. Raj et al. / Journal of Cleaner Production 113 (2016) 1005e1014

1.1

K-T parameter ( value

K-T parameter (* value)


0.9

[C2mim][OAc] [C2mim][OAc]
[C4mim][OAc] [C4mim][OAc]
[C4mim][Cl]
[C4mim][Cl]
[C4mim][BF4]
[C4mim][BF4]
0.7
2.4 2.6 2.8 3.0 3.2 3.4
1000/T(K) 1000/T(K)
a b
Fig. 2. KamleteTaft parameters of various ILs using the following set of dye: Reichardt's dye, N,N-diethyl-4-nitroaniline and 4-nitroaniline. (a) Hydrogen bond acidity (a), (b)
polarizability (p*).

data shows that, as expected an increase in temperature, a signif- hence almost no solubilization. It can be argued, based on the effect
icant decrease in viscosity of ILs was obtained, i.e. viscosity of of viscosity and b value on glucose yields, b value is preferred than
[C2mim][OAc] at 100  C is 8.4 cSt/s, whereas, 4.4 cSt/s at 130  C. the viscosity when comparing the ILs treatment efcacy. From this
Viscosity of ILs at 100  C and also at 130  C was following the same study it is also observed that, ILs having b > 0.72, with low viscosity
order as: [C2mim][OAc]  [C4mim][BF4] < [C4mim][OAc] < [C2mim] results in higher glucose yields.
[Cl] < [C4mim][Cl]. An increase in viscosity of ILs with various
anion/cation combinations may be attributed to the increased of 3.4. Impact of surface tension on enzymatic saccharication
van der Waals forces over hydrogen bonding. Viscosity of ILs was
also increased with increasing alkyl chain length of cation of ILs Surface tension of all the ILs was measured at 25, 50 and 75  C
with same anion. Similar results are reported by Cao et al. (2014) and extrapolated by the linear t to get values for 100 and 130  C
and Huddleston et al. (2001). (Fig. SI 3). As shown in Table 3, the surface tension decreases with
In order to nd any correlation between viscosity of ILs and an increase in temperature. ILs with same anion and cation of
glucose yields, Fig. 4 was drawn, which shows that, the glucose different alkyl chain length, show an increase in surface tension
yield was in reverse order of viscosity. Highest glucose yield (83.2 with increasing alkyl chain length. Coming to the effect of anion,
and 89.8%) was achieved with [C2mim][OAc], having a lower vis- comparing the surface tension of ILs with butyl substituted on
cosity (8.3 and 4.4 cSt/s) at 100 and 130  C respectively. Similarly, cation the order is [OAc] > [Cl] > [BF4].
[C4mim][Cl] and [C2mim][Cl] showed highest viscosity, i.e. 25.71 Glucose yields after enzymatic saccharication of ILs treated and
and 21.71 cSt/s at 100  C leading to 45.9 and 61.7% glucose yields regenerated cellulose was drawn against the surface tension in
respectively. This is in consistent with the previous results (Ma ki- Fig. 5 and it showed negative correlation, i.e. ILs with lower surface
Arvela et al., 2010) that, ILs with lower viscosity can dissolve tension gave higher glucose yields except that of [C4mim][BF4].
more amount of cellulose. However, [C4mim][BF4] having a very [C2mim][OAc] with low surface tension, i.e. 32.5 mN/m gave 83.3%,
low viscosity 8.2 and 4.5 cSt/s at 100 and 130  C does not improve whereas, [C4mim][OAc] (37.3 mN/m) gave 64.9% of glucose yield at
the enzymatic hydrolysis. This may be attributed to the low 100  C/5 h. Similarly, [C2mim][Cl] with a surface tension of
hydrogen bond basicity (b < 0.72), [BF4] being non coordinating 42.4 mN/m gave 61.7% and [C4mim][Cl] (45.1 mN/m) gave 45.9%
anion, which can not interrupt the hydrogen bonding in cellulose glucose yield at 100  C/5 h. Similar trends of surface tension vis-
a-vis glucose yields were observed at 130  C/2 h. However, surface

tension of [C4mim][BF4] is 41.7 mN/m, lower than [C4mim][Cl]
(45.17 mN/m), which gave poor glucose yield (25.5%). It may be
1.8
90

1.6
80 Table 3
Viscosity and surface tension of different ionic liquids.
Glucose yield (%) after 72 h with

70 1.4 ILs Kinematic viscosity (h, cSt/s)a Surface tension (s, mN/m)b
10 FPU/g substrate

Beta value ()

60 T/K (373.15) T/K (403.15) T/K (298.15) T/K (403.15)


1.2
[C2mim][OAc] 8.35 0.5A 4.41 0.3A* 32.45 0.1A 30.25 0.5A*
50 [C4mim][OAc] 14.08 0.4B 6.73 0.4B* 37.34 0.2B 36.42 0.4B*
1.0
[C2mim][Cl] 21.47 0.3C 9.31 0.1C* 42.42 0.3C 40.48 0.5C*
40 [C4mim][Cl] 25.71 0.2D 11.44 0.1D* 45.07 0.1D 43.81 0.4D*
Glucose yield (%) (130 C/ 2h) [C4mim][BF4] 8.15 0.4A 4.53 0.2A* 41.68 0.2E 40.71 0.2C*
Glucose yield (%) (100 C/ 5h) 0.8
30 a
Kinematic viscosity (h) was calculated by Cannon-Fenske Viscometer with 300
Value 130 C
and 400 bulb sizes directly at 100 and 130  C.
Value 100 C
20
b
Surface tension (s) was calculated at 25, 50 and 75  C and then extrapolated by
0.6
[C2mim][OAc] [C4mim][OAc] [C4mim][Cl] [C4mim][BF4] the linear t to get value for 100 and 130  C. All experiments were done in triplicate
and the mean is reported with S.D. Values in the same column for same temper-
Fig. 3. Effect of hydrogen bond basicity (b) of ILs and correlation between b vs. glucose ature with different superscript letters indicate the signicance difference at
yield (%) of the treated cellulose at 100  C/5 h and 130  C/2 h. P  0.05. Letter with star (*) superscript shows statistical signicance at 403.15 K.
T. Raj et al. / Journal of Cleaner Production 113 (2016) 1005e1014 1011

100 100 order of glucose yields, thus anion with small size gives higher
Glucose yield (%) (130 C/ 2h) yields in enzymatic hydrolysis. [C2mim][OAc] and [C4mim][OAc]
90 Glucose yield (%) (100 C/ 5h)
Viscosity (cSt/ sec) (130 C/ 2h)
with same anion, i.e. [OAc] with different cation, i.e. [C2mim] and
80 Viscosity (cSt/ sec) (130 C/ 5h) 80 [C4mim] results in 89.8 and 73.9% glucose yields after 72 h of
enzymatic hydrolysis respectively. Similarly, [C4mim][OAc],
Glucose yield (%) after 72 h with

Kinematic viscosity (, cSt/sec)


70
[C4mim][Cl] and [C4mim][BF4] with same cation [C4mim] with
different anion, i.e. [OAc], [Cl] and [BF4] results in 73.9, 50.8 and
10 FPU/ g substrate

60 60

50
27.3% of glucose yields respectively at 130  C/2 h. Thus, [C2mim]
[OAc] with the best combination of smaller size anion and smaller
40 40 chain length on cation is quite powerful in cellulose structural
30 modication leading to highest glucose yields in enzymatic hy-
drolysis. The size of cation or anion is connected to its degree of
20 20 interaction with cellulose chain and disrupting inter and intra-
10 molecular hydrogen bonding in cellulose causing higher glucose
yields.
0 0
[C2mim][OAc] [C4mim][OAc] [C2mim][Cl] C4mim][Cl] [C4mim][BF4]

Fig. 4. Effect of viscosity of ILs and correlation with glucose yield (%) of the treated 3.6. Structural changes in treated cellulose
cellulose at 100  C/5 h and 130  C/2 h.
PXRD and FT-IR analysis of native and treated cellulose was
carried out to determine the structural transformations with
attributed to the low hydrogen bond basicity (b 0.72) of IL. respect to treatment temperature, time and ILs. Native cellulose
Therefore, it is clear from the above discussion that, surface tension showed a characteristic peak of cellulose I polymorph at 2q 15.2
and viscosity, both are showing negative correlation with glucose (101) plane and 22.6 (002) plane. Peaks at 2q 15.2 (101), 16.4
yields and a positive correlation is obtained between natural log- 101 and 22.6 (002) are characteristic of cellulose I. Peaks at 12.1
arithms of surface tension and viscosity as given in Fig. SI 4 (Singh (110), 20.0 (021) plane represents amorphous cellulose, herein
and Singh, 2011). From the above data it can be noticed that, the called as cellulose II. After ILs treatment at 100  C/5 h, [C2mim]
viscosity and surface tension has a signicant impact on glucose [OAc] and [C4mim][OAc] displayed the reduction in CrI, which may
yield if the variation is induced either in anion or cation. And a be due to conversion of cellulose I to cellulose II crystal structure as
possible explanation for the efcacy of ILs with lower surface ten- shown in Fig. 6. Cellulose treated with [C2mim][OAc] at 130  C/2 h
sion is that, liquids with lower surface tension has low surface has the lowest CrI (57.4%), compared with [C4mim][OAc] (70.6),
forces and could easily penetrate into the material resulting in more [C2mim][Cl] (72.5), [C4mim][Cl] (73.0), [C4mim][BF4] (87.8) and
interaction with cellulose hydroxyls. native cellulose (89.0%) as well. This decrease in CrI clearly shows
the transformation of cellulose I to cellulose II. Similar trends were
observed at 100  C/5 h, although the crystallinity index is higher at
3.5. Effect of size of cation and anion 100  C/5 h. Cellulose treated with [C4mim][Cl] and [C4mim][BF4]
retains the crystalline cellulose I structure exhibiting high CrI value,
The impact of different anion [OAc], [Cl], [BF4] and cations i.e. 73.0 and 87.8%, even at 130  C/2 h (Table 1), showing no
having two different alkyl chain lengths, [C2mim] and [C4mim] structural changes and hence, resulted to lower enzymatic hydro-
were analyzed to nd out the impact on glucose yields. Results of lysis (Fig. 7). Decrease in the X-ray peak intensity with the shift of
enzymatic saccharication of regenerated cellulose (Table 1) indi- the peak position to the left is observed with an increase in the
cate that, with increasing alkyl chain length of cation the glucose treatment temperature. For cellulose treated at 130  C/2 h, the in-
yields decreases. The size of anion of ve ILs follows the order of tensity at 22.6 decreases signicantly and disappeared at 15.2 and
[OAc] < [Cl] < [BF4] (Hou et al., 2015); which is reverse to the 16.4 and appearing a peak at 20.0 (110) plane for cellulose II.
Cellulose treated with [C2mim][OAc] at 130  C/2 h shows highly
amorphous nature with a broad peak around 21.4 assign to the
50 (I002) plane of cellulose II lattice with peaks at ~15.2 , 21.5 . It may
90 be argued that, ILs treatment improves the enzymatic hydrolysis by
transformation of cellulose I to cellulose II which is amorphous in
80 nature. Similar observation for cellulose I and II is also reported in
literature (Cheng et al., 2011).
Glucose yield (%) after 72 h
with 10 FPU/g of substrate

Surface tension (, mN/m)

70 [C2mim][OAc] with high b value (1.32), low surface tension and


40 low viscosity rapidly disrupt the inter and intramolecular hydrogen
60 bonding and transform the cellulose I crystalline structure to cel-
lulose II resulting in high glucose yields (89.7%). The formation of
50 more amorphous cellulose (cellulose II) facilitated high glucose
yields after enzymatic hydrolysis. Similarly, [C4mim][OAc], [C2mim]
40 [Cl] and [C4mim][Cl] with moderate viscosity, surface tension and
Glucose yield (%) (130 C/ 2h)
30 hydrogen bond basicity (b > 0.72), are able to reduce cellulose
30 Glucose yield (%) (100 C/ 5h) crystallinity, resulting in moderate glucose yields (Table 1). How-
Surface tension (130 C/ 2h)
Surface tension (100 C/ 5h)
ever, [C4mim][BF4] with low viscosity (8.2 cSt/s) and high surface
20 tension (41.7 mN/m)) is unable to penetrate the cellulose structure.
[C2mim][OAc] [C4mim][OAc] [C2mim][Cl] [C4mim][Cl] [C4mim][BF4]
This may be attributed to that these ILs do not interrupt the intra
Fig. 5. Effect of surface tension (s) of ILs and the correlation between s vs. glucose and inter molecular hydrogen bonds of cellulose, which resulted
yield (%) at 100  C/5 h and 130  C/2 h. higher CrI with lower glucose yields after enzymatic hydrolysis.
1012 T. Raj et al. / Journal of Cleaner Production 113 (2016) 1005e1014

100 C/ 5 h (002) 130 C/ 2 h (002)

(101) (101)
(021) (021)

Intensity (a.u.)
Intensity (a.u.)

Untreated avicel Untreated avicel

[C mim][BF ]
[C mim][BF ]
[C mim][Cl] [C mim][Cl]
[C mim][Cl] [C mim][Cl]

[C mim][OAc] [C mim][OAc]

[C mim][OAc]
[C mim][OAc]

5 10 15 20 25 30 35 40
5 10 15 20 25 30 35 40
Diffraction angle (2 )
Diffraction angle (2)

Fig. 6. PXRD diffraction patterns for untreated and ILs treated cellulose treated by ve ILs at 100  C/5 h and 130  C/2 h.

FT-IR spectrum is given in Fig. 8 for untreated and IL treated modication, hence, only 27.3% of glucose yield was achieved.
cellulose. A strong absorption band at 3400 cm1 is assigned to However, the absorption band at 898 cm1 assigned to CeOeC
different OeH stretching modes due to inter and intramolecular stretching at the b-(1 / 4) glycosidic linkage is weak and broad in
hydrogen bonding. This band is very intense in untreated cellulose cellulose I but strong and sharp in cellulose II. Therefore, infrared
and the intensity of these bands decreases with ILs treatment and spectra in the range of 850e1500 cm1 were characteristic of cel-
regeneration. Two bands at 2920 and 2850 cm1 are related to lulose samples regenerated from different treatments.
asymmetric and symmetric methyl and methylene stretching. The It was observed that the untreated cellulose possesses cellulose I
overlapping features from 1480 to 1290 cm1 region consist of a crystal type and transformed into cellulose II crystal structure after
number of bands such as 1419, 1328 and 1312 cm1 mainly due to the treatment and regeneration of cellulose. The absorption bands
various deformation modes of CeH bonds in the methylene groups. at 1367 and 2900 cm1 are for CeH and CH2 stretching, respectively
The absorption band at 1419 cm1 assigned to the CH2 scissoring and the band at 1631 cm1 relates to the bending mode of the
motion was strong in untreated cellulose and is the characteristic water. The peak at 1382 cm1 is attributed to the OeH bending and
peak of crystalline cellulose I. This band is reduced in [C2mim][OAc] absorption band at 1122 and 1161 cm1 corresponds to CeO anti-
and [C4mim][OAc] treatment due to structural transformation from symmetric bridge stretching. A strong peak at 1082 and 1033 cm1
cellulose I to cellulose II. This conrms the reduction in the crys- arises from CeOeC pyranose ring skeletal vibration. Similar
tallinity after ILs treatment of cellulose, hence resulting increase in observation is also reported by Goshadrou et al. (2013) and Kapoor
glucose yield. As discussed above, [C2mim][OAc] and [C4mim][OAc] et al. (2015).
ILs, having higher b value, i.e. 1.27, 1.19 causes higher trans-
formation of cellulose than [C4mim][Cl] and [C4mim][BF4]. The 4. Conclusion
intensities at 3400, 2887, 1328, 1122, 1082, 1033 cm1 peaks get
reduced after treatment resulting to produce 89.8 and 73.9% Cellulose transformation efciencies, for improving the enzyme
glucose yield respectively (Table 1). The intensity of these peaks hydrolysis by ve ILs, i.e. [C2mim][OAc], [C4mim][OAc], [C2mim]
after [C4mim][BF4] treatment remains same as in the case of native, [Cl], [C4mim][Cl] and [C4mim][BF4] were investigated and a corre-
which conrms that [C4mim][BF4] do not causes cellulose structure lation between KeT parameters including hydrogen bond acidity,
hydrogen bond basicity and polarity, viscosity and surface tension
and their glucose yields was established. Among the KeT param-
eters, b value stands as the best predictor and rst parameter needs
100 100 to be considered while checking ILs treatment efciency. [OAc]
anion based ILs with b > 0.72 causes in the maximum reduction of
90
CrI and resulted in 89.7% glucose yield after 72 h of cellulase hy-
80 drolysis and ILs with b  0.72 like [C4mim][BF4] neither induced
80
CrI % of ILs treated cellulose
Glucose yield (%) after 72 h with

any structural transformation in cellulose nor improve glucose


10 FPU/ g substrate of enzyme

yields. Viscosity and surface tension have a negative correlation


70
60 with glucose yields, lower the viscosity/surface tension higher the
60 sugar yields in saccharication. This correlation is applicable only if
IL has b > 0.72, hence, could also be an important parameter to
40 predict the treatment efciency. ILs with smaller anion and alkyl
50
chain on cation is efcient for the treatment and enzymatic hy-
40 drolysis. If variation is induced either in anion or cation, keeping the
Glucose yield (%) (130 C/ 2h)
20 other constant, the surface tension and viscosity could be an indi-
Glucose yield (%) (100 C/ 5h)
30
CrI % (130 C/ 2h)
cator of treatment efciency. [C2mim][OAc] having higher b value,
CrI % (100 C/ 5h) lower viscosity, lower surface tension and smaller size of anion and
20 0 shorter alkyl chain gave higher glucose yields. The experimental
[C2mim][OAc] [C4mim][OAc] [C2mim][Cl] C4mim][Cl] [C4mim][BF4]
correlation and detailed knowledge of multiple properties pre-
Fig. 7. Correlation of CrI of treated cellulose vs. glucose yield (%) after 72 h enzymatic sented in this work can be used as a tool for the rational design of
hydrolysis. ILs for lignocellulosic biomass treatment. However, these initial
T. Raj et al. / Journal of Cleaner Production 113 (2016) 1005e1014 1013

100 C/ 5h 130 C/ 2h

Untreated avicel
Untreated avicel
[C mim][BF ]
Absorbance

Absorbance
[C mim][BF ]

[C mim][Cl] [C mim][Cl]

[C mim][Cl] [C mim][Cl]

[C mim][OAc] [C mim][OAc]

[C mim][OAc] [C mim][OAc]

4000 3500 3000 2500 2000 1500 1000 500 4000 3500 3000 2500 2000 1500 1000 500
Wavenumber (cm ) Wavenumber (cm )
a b
Fig. 8. FT-IR spectrum of untreated and regenerated cellulose treated by ve ILs at 100  C/5 h and 130  C/2 h.

laboratory scale evaluation of IL-based treatment technology can be Dupont, J., Consorti, C.S., Suarez, P.A.Z., de Souza, R.F., 2003. Preparation of 1-Butyl-
3-Methyl Imidazolium-Based Room Temperature Ionic Liquids, Organic Syn-
effectively scaled to larger operations. The pretreatment of biomass
theses. John Wiley & Sons, Inc.
using these ILs are being conducted now. Geng, X., Henderson, W.A., 2014. Impact of non-solvents on the structural features
and enzymatic digestibility of cellulose regenerated from an ionic liquid. RSC
Adv. 4, 31226e31229.
Acknowledgements Goshadrou, A., Karimi, K., Lefsrud, M., 2013. Characterization of ionic liquid pre-
treated aspen wood using semi-quantitative methods for ethanol production.
Carbohydr. Polym. 96, 440e449.
The authors are grateful to Indian Oil Corporation Ltd, Research Greases, C.S., 2014. Cannon Fenske Viscometer.
and Development Centre, Faridabad is D1106 and Department of Haghighi Mood, S., Hossein Golfeshan, A., Tabatabaei, M., Salehi Jouzani, G.,
Naja, G.H., Gholami, M., Ardjmand, M., 2013. Lignocellulosic biomass to bio-
Biotechnology, Ministry of Science and Technology, New Delhi is
ethanol, a comprehensive review with a focus on pretreatment. Renew. Sustain.
BT/PB/08/03/2007 for nancial support and work. Dr. E. Ramu from Energy Rev. 27, 77e93.
analytical division, Indian Oil Corporation Ltd. is thanked for FT-IR. Hassan, E.S.R.E., Mutelet, F., Moise, J.C., 2013. From the dissolution to the extraction
of carbohydrates using ionic liquids. RSC Adv. 3, 20219e20226.
Hou, X.D., Xu, J., Li, N., Zong, M.H., 2015. Effect of anion structures on cholinium
ionic liquids pretreatment of rice straw and the subsequent enzymatic hydro-
Appendix A. Supplementary data
lysis. Biotechnol. Bioeng. 112, 65e73.
Hou, X., Smith, T., Li, N., Zong, M., 2012. Novel renewable ionic liquids as highly
Supplementary data related to this article can be found at http:// effective solvents for pretreatment of rice straw biomass by selective removal of
dx.doi.org/10.1016/j.jclepro.2015.12.037. lignin. Biotechnol. Bioeng. 109, 2484e2493.
Huddleston, J.G., Visser, A.E., Reichert, W.M., Willauer, H.D., Broker, G.A.,
Rogers, R.D., 2001. Characterization and comparison of hydrophilic and hy-
drophobic room temperature ionic liquids incorporating the imidazolium
References cation. Green Chem. 3, 156e164.
ger, S., Brumbauer, A., Fehe
Ja r, E., Re
czey, K., Kiss, L., 2001. Production and charac-
Amiri, H., Karimi, K., Zilouei, H., 2014. Organosolv pretreatment of rice straw for terization of b-glucosidases from different Aspergillus strains. World J. Micro-
efcient acetone, butanol, and ethanol production. Bioresour. Technol. 152, biol. Biotechnol. 17, 455e461.
450e456. Jiang, L.Q., Fang, Z., Li, X.K., Luo, J., Fan, S.P., 2013. Combination of dilute acid and
Bensah, E.C., Mensah, M., 2013. Chemical pretreatment methods for the production ionic liquid pretreatments of sugarcane bagasse for glucose by enzymatic hy-
of cellulosic ethanol: technologies and innovations. Int. J. Chem. Eng. 2013, 21. drolysis. Process Biochem. 48, 1942e1946.
Bian, J., Peng, F., Peng, X.P., Xiao, X., Peng, P., Xu, F., Sun, R.C., 2014. Effect of [Emim] Jiang, M., Zhao, M., Zhou, Z., Huang, T., Chen, X., Wang, Y., 2011. Isolation of cellulose
Ac pretreatment on the structure and enzymatic hydrolysis of sugarcane with ionic liquid from steam exploded rice straw. Ind. Crops Prod. 33, 734e738.
bagasse cellulose. Carbohydr. Polym. 100, 211e217. Kamlet, M., Taft, R., 1976. Solvatochromic comparison method I. Beta-scale of sol-
Brandt, A., Grasvik, J., Hallett, J.P., Welton, T., 2013. Deconstruction of lignocellulosic vent hydrogen-bond acceptor (Hba) basicities. J. Am. Chem. Soc. 98, 377e383.
biomass with ionic liquids. Green Chem. 15, 550e583. Kapoor, M., Raj, T., Vijayaraj, M., Chopra, A., Gupta, R.P., Tuli, D.K., Kumar, R., 2015.
Brandt, A., Hallett, J., Leak, D., Murphy, R., Welton, T., 2010. The effect of the ionic Structural features of dilute acid, steam exploded, and alkali pretreated mustard
liquid anion in the pretreatment of pine wood chips. Green Chem. 12, 672e679. stalk and their impact on enzymatic hydrolysis. Carbohydr. Polym. 124, 265e273.
Brandt, A., Ray, M., To, T., Leak, D., Murphy, R., Welton, T., 2011. Ionic liquid pre- Kim, I., Lee, B., Song, D., Han, J.I., 2014. Effects of ammonium carbonate pretreat-
treatment of lignocellulosic biomass with ionic liquidewater mixtures. Green ment on the enzymatic digestibility and structural features of rice straw. Bio-
Chem. 13, 2489e2499. resour. Technol. 166, 353e357.
Cao, Y., Chen, Y., Wang, X., Mu, T., 2014. Predicting the hygroscopicity of Kim, Y., Ximenes, E., Mosier, N.S., Ladisch, M.R., 2011. Soluble inhibitors/deactivators
imidazolium-based ILs varying in anion by hydrogen-bonding basicity and of cellulase enzymes from lignocellulosic biomass. Enzyme Microb. Technol. 48,
acidity. RSC Adv. 4, 5169e5176. 408e415.
Cheng, G., Varanasi, P., Li, C., Liu, H., Melnichenko, Y.B., Simmons, B.A., Kent, M.S., Kumar, G., Bakonyi, P., Periyasamy, S., Kim, S.H., Nemesto  thy, N., Be
la-Bako
 , K.,
Singh, S., 2011. Transition of cellulose crystalline structure and surface 2015. Lignocellulose biohydrogen: practical challenges and recent progress.
morphology of biomass as a function of ionic liquid pretreatment and its Renew. Sustain. Energy Rev. 44, 728e737.
relation to enzymatic hydrolysis. Biomacromolecules 12, 933e941. Larsson, S., Palmqvist, E., Hahn-Ha gerdal, B., Tengborg, C., Stenberg, K., Zacchi, G.,
Crowhurst, L., Mawdsley, P.R., Perez-Arlandis, J.M., Salter, P.A., Welton, T., 2003. Nilvebrant, N.-O., 1999. The generation of fermentation inhibitors during dilute
Solventesolute interactions in ionic liquids. Phys. Chem. Chem. Phys. 5, acid hydrolysis of softwood. Enzyme Microb. Technol. 24, 151e159.
2790e2794. Lee, J.M., Ruckes, S., Prausnitz, J.M., 2008. Solvent polarities and KamleteTaft pa-
da Costa Lopes, A., Joao, K., Morais, A.R., Bogel-Lukasik, E., Bogel-Lukasik, R., 2013. rameters for ionic liquids containing a pyridinium cation. J. Phys. Chem. B 112,
Ionic liquids as a tool for lignocellulosic biomass fractionation. Sustain. Chem. 1473e1476.
Process. 1, 3. Li, C., Knierim, B., Manisseri, C., Arora, R., Scheller, H., Auer, M., Vogel, K.,
Dadi, A., Schall, C., Varanasi, S., 2007. Mitigation of cellulose recalcitrance to Simmons, B., Singh, S., 2010. Comparison of dilute acid and ionic liquid pre-
enzymatic hydrolysis by ionic liquid pretreatment. Appl. Biochem. Biotechnol. treatment of switchgrass: biomass recalcitrance, delignication and enzymatic
137e140, 407e421. saccharication. Bioresour. Technol. 101, 4900e4906.
Doherty, T., Mora-Pale, M., Foley, S., Linhardt, R., Dordick, J., 2010. Ionic liquid sol- Li, C., Tanjore, D., He, W., Wong, J., Gardner, J.L., Sale, K.L., Simmons, B.A., Singh, S.,
vent properties as predictors of lignocellulose pretreatment efcacy. Green 2013. Scale-up and evaluation of high solid ionic liquid pretreatment and
Chem. 12, 1967e1975. enzymatic hydrolysis of switchgrass. Biotechnol. Biofuels 6, 154.
1014 T. Raj et al. / Journal of Cleaner Production 113 (2016) 1005e1014

Maki-Arvela, P., Anugwom, I., Virtanen, P., Sjo holm, R., Mikkola, J.P., 2010. Dissolu- Singh, M.P., Singh, R.K., 2011. Correlation between ultrasonic velocity, surface
tion of lignocellulosic materials and its constituents using ionic liquidsda re- tension, density and viscosity of ionic liquids. Fluid Phase Equilib. 304,
view. Ind. Crops Prod. 32, 175e201. 1e6.
Pal, M., Rai, R., Yadav, A., Khanna, R., Baker, G.A., Pandey, S., 2014. Self-aggregation of Singh, S., Simmons, B.A., Vogel, K.P., 2009. Visualization of biomass solubilization
sodium dodecyl sulfate within (choline chloride urea) deep eutectic solvent. and cellulose regeneration during ionic liquid pretreatment of switchgrass.
Langmuir 30, 13191e13198. Biotechnol. Bioeng. 104, 68e75.
Panda, T., Bisaria, V.S., Ghose, T.K., 1987. Effect of culture phasing and a poly- Sissine, F., 2007. Energy Independence and Security Act of 2007: a Summary of
saccharide on production of xylanase by mixed culture of Trichoderma reesei Major Provisions. DTIC Document.
D1-6 and Aspergillus wentii Pt 2804. Biotechnol. Bioeng. 30, 868e874. Sun, N., Parthasarathi, R., Socha, A.M., Shi, J., Zhang, S., Stavila, V., Sale, K.L.,
Parviainen, A., King, A.W.T., Mutikainen, I., Hummel, M., Selg, C., Hauru, L.K.J., Simmons, B.A., Singh, S., 2014. Understanding pretreatment efcacy of four
Sixta, H., Kilpela inen, I., 2013. Predicting cellulose solvating capabilities of cholinium and imidazolium ionic liquids by chemistry and computation. Green
acidebase conjugate ionic liquids. ChemSusChem 6, 2161e2169. Chem. 16, 2546e2557.
Peleteiro, S., Garrote, G., Santos, V., Parajo , J.C., 2014. Furan manufacture from Virunanon, C., Ouephanit, C., Burapatana, V., Chulalaksananukul, W., 2013.
softwood hemicelluloses by aqueous fractionation and further reaction in a Cassava pulp enzymatic hydrolysis process as a preliminary step in bio-
catalyzed ionic liquid: a biorenery approach. J. Clean. Prod. 76, 200e203. alcohols production from waste starchy resources. J. Clean. Prod. 39,
Plechkova, N.V., Seddon, K.R., 2008. Applications of ionic liquids in the chemical 273e279.
industry. Chem. Soc. Rev. 37, 123e150. Wada, M., Ike, M., Tokuyasu, K., 2010. Enzymatic hydrolysis of cellulose I is greatly
Qiu, Z., Aita, G.M., Walker, M.S., 2012. Effect of ionic liquid pretreatment on the accelerated via its conversion to the cellulose II hydrate form. Polym. Degrad.
chemical composition, structure and enzymatic hydrolysis of energy cane Stab. 95, 543e548.
bagasse. Bioresour. Technol. 117, 251e256. Weerachanchai, P., Lee, J.M., 2014. Recyclability of an ionic liquid for biomass pre-
Raj, T., Kapoor, M., Gaur, R., Christopher, J., Lamba, B., Tuli, D.K., Kumar, R., 2015. treatment. Bioresour. Technol. 169, 336e343.
Physical and chemical characterization of various Indian agriculture residues Welton, T., 1999. Room-temperature ionic liquids. Solvents for synthesis and
for biofuels production. Energy Fuels 29 (5), 3111e3118. catalysis. Chem. Rev. 99, 2071e2084.
Sant'Ana da Silva, A., Lee, S.H., Endo, T., Bon, E.P.S., E, 2011. Major improvement in Xia, S., Baker, G.A., Li, H., Ravula, S., Zhao, H., 2014. Aqueous ionic liquids and deep
the rate and yield of enzymatic saccharication of sugarcane bagasse via pre- eutectic solvents for cellulosic biomass pretreatment and saccharication. RSC
treatment with the ionic liquid 1-ethyl-3-methylimidazolium acetate ([Emim] Adv. 4, 10586e10596.
[Ac]). Bioresour. Technol. 102, 10505e10509. Xiao, W., Yin, W., Xia, S., Ma, P., 2012. The study of factors affecting the enzymatic
Segal, L., Creely, J., Martin, A., Conrad, C., 1959. An empirical method for estimating hydrolysis of cellulose after ionic liquid pretreatment. Carbohydr. Polym. 87,
the degree of crystallinity of native cellulose using the X-ray diffractometer. 2019e2023.
Text. Res. J. 29, 786e794. Ximenes, E., Kim, Y., Mosier, N., Dien, B., Ladisch, M., 2010. Inhibition of cellulases by
Sharma, S., Kumar, R., Gaur, R., Agrawal, R., Gupta, R.P., Tuli, D.K., Das, B., 2015. Pilot phenols. Enzyme Microb. Technol. 46, 170e176.
scale study on steam explosion and mass balance for higher sugar recovery Ximenes, E., Kim, Y., Mosier, N., Dien, B., Ladisch, M., 2011. Deactivation of cellulases
from rice straw. Bioresour. Technol. 175, 350e357. by phenols. Enzyme Microb. Technol. 48, 54e60.

You might also like