Enhanced lignin extraction from different species of oil

Industrial Crops & Products 116 (2018) 122–136

Contents lists available at ScienceDirect

Industrial Crops & Products journal homepage: www.elsevier.com/locate/indcrop

Enhanced lignin extraction from different species of oil palm biomass: Kinetics and optimization of extraction conditions

T

Tazien Rashida, Nirmala Gnanasundaramb, Arunagiri Appusamyc, Chong Fai Kaitd, ⁎ Murugesan Thanabalana, a

Department of Chemical Engineering, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, Tronoh, 32610, Perak Malaysia, Malaysia Department of Chemical Engineering, Vellore Institute of Technology University, Vellore, 632014, India, India c Department of Chemical Engineering, National Institute Of Technology, Trichy, 620015, India, India d Fundamental and Applied Sciences Department, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, Tronoh, 32610, Perak, Malaysia, Malaysia b

A R T I C L E I N F O

A B S T R A C T

Keywords: Lignin extraction Oil palm Biomass Protic ionic liquid Kinetics Pseudo secondorder Solvent recycling

Lignin from industrial crops is a renewable bioresource which can be used for a variety of value-added applications, however effective separation of lignin from lignocellulosic biomass is still an ongoing challenge. The present study involves the selective extraction of lignins from three different morphological parts of oil palm biomass plant, namely empty fruit bunches (EFB), palm mesocarp fiber (PMF) and palm kernel shells (PKS) using Pyridinium Formate [PyFor], a Protic ionic liquid under mild extraction conditions compared to the conventional chemical processes. The effect of initial lignin contents and parameters (namely, particle size range, extraction temperature, time and solid loading) influencing the lignin extraction efficiency were analyzed and optimized using response surface methodology. The experiments conducted at the estimated optimum conditions gave the maximum lignin extraction of 92.01%, 91.23% and 90.70% at lowest process temperatures 351.5 K, 361.9 K and 370.8 K for EFB, PMF and PKS respectively. In a second-stage the extraction experiments were conducted to study the kinetics of extraction process under selected conditions and results were well correlated using pseudo-second order kinetics model. The theoretical lignin concentration at saturation (Cs), rate constant (K) and initial rate of extraction (h) at various temperatures ranging from 323 to 373 K were determined. The nature of biomass source and initial lignin content were found to have major impact on the extraction kinetics. Furthermore, the present estimated activation energies of 12 kJ mol−1, 23 kJ mol−1 and 28 kJ mol−1 for the present extraction of lignins from EFB, PMF and PKS respectively are remarkably lower as compared to those reported in literature for traditional wood pulp processes. The extracted lignins were successfully characterized using FTIR and 1H NMR analysis. The regeneration and recyclability of [PyFor] is also tested and the sustainability of the solvent for commercial application is proved.

1. Introduction Oil palm biomass is considered as a renewable biomass resource for the biofuel production (Basiron, 2004; Putro et al., 2016). Lignocellulosic biomass consists of three basic macromolecular components namely cellulose, hemicellulose and lignin, which are the potential building blocks for the production of biofuels, biochemicals, and biodegradable products (Korotkova et al., 2015; Zhang, 2008). Among these lignin is the second most essential polymer in the plant world with distinctive properties (Korotkova et al., 2015; Sun and Cheng, 2002). Lignin is linear polymer containing branching points where chemical linkages to hemicellulose and cellulose do exist. (Ralph et al., 2007). Lignin can be considered as a potential source of many value-added

⁎

products such as lignin based carbon fibers, isocyanate binders, biodispersants, phenolic and thermosetting resins etc. (Kadla and Kubo, 2004; Lora and Glasser, 2002). Lignin extraction from wood is a complex phenomenon, it involves the breaking of intermolecular bonds present between lignin and the plant matrix along with diffusion of lignin into the solvent (Korotkova et al., 2015). The most common separation techniques for commercial delignification of wood are kraft pulping (Abdul-Karim et al., 1995), sulfite pulping (Mansouri and Salvado, 2006), acid hydrolysis (Kumar et al., 2009) and organosolv pulping (Lu and Ralph, 2010). Due to the severity of the process conditions involved during the delignification process, such as; high temperatures and pressures, long dissolution times induce severe changes to the original lignin structure (Rashid et al., 2016; Wang and Chen,

Corresponding author. E-mail address: [email protected] (M. Thanabalan).

https://doi.org/10.1016/j.indcrop.2018.02.056 Received 22 August 2017; Received in revised form 30 January 2018; Accepted 16 February 2018 0926-6690/ © 2018 Elsevier B.V. All rights reserved.


T. Rashid et al.

Canadian hemp. Two phases of delignification were identified by Ho et al. (2011) and respective equations for the transition times at various temperatures and catalyst dosages were reported. Korotkova et al. (2015) studied the isolation of lignin from spruce wood and analyzed the extraction kinetics, with respect to the alkali concentration. A comprehensive list of literature involving the kinetics of several extraction processes are compared in Table 9. From the available literature on the lignin extrtaction, it is evident that a there exists a gap in the fundamental understanding of the impact of the initial lignin concentration of lignin and the effect of several operating parameters such as; particle size, temperature, extraction time, solid loading on the lignin extraction from three different species of oil palm biomass (OPB) namely; Empty Fruit Bunch (EFB), Palm Mesocarp Fiber (PMF) and Palm Kernell Shells (PKS). In the present research Response Surface Methodology (RSM), a statistical tool, is used to determine an appropriate experimental design protocol. RSM was effectively utilized for experimental design, by examining the effects of multiple variables and optimizing these variables for the optimal response (Dehghani et al., 2016). The Box-Behnken design was selected and the experimental data were analyzed using Statgraphics Centurion 15.2.11.0 version 8.0. The optimum conditions for lignin extraction were identified. On the basis of these facts, the main objective of this work was to establish extraction kinetics of lignin removal from oil palm biomass using [PyFor], by exploring the influence of lignin contents and various operating parameters. From the experimental data, second order extraction model was employed to provide a quantitative representation of present lignin extraction process. In addition, the extraction activation energy (Ea) for the extraction process was also calculated. This can be helpful for both fundamental and applied perspective as it can contribute to the basic understanding of the lignin extraction process using protic ionic liquids.

2016). Thus, in spite of the value-addition and attractive applications, only 1–2% of lignin is considered for valuable applications and the rest is being burnt as a low-grade fuel (Lora and Glasser, 2002; Putro et al., 2016; Zhang, 2008). Hence, the effective separation technique to break the compact network structure of lignocellulosic biomass, while preserving its biopolymers (cellulose, hemicellulose and lignin) from degradation is an essential feature, prior to its further utilization for the bio-refinery concept (Ho et al., 2011; Korotkova et al., 2015). Recently protic ionic liquids (PIL’s) have emerged as prospective solvents as they possess a range of distinctive properties, namely negligible vapor pressure, high thermal stability, and low chemical reactivity (Achinivu et al., 2014; George et al., 2015; Rashid et al., 2016). These unique characteristics, along with the fine tunable properties favor their application in diverse fields such as cationic surfactants, polymer membrane fuel cells, non −aqueous electrolytes etc. (Belieres and Angell, 2007). In the present study an attempt has been made to use Pyridinium Formate [PyFor] as a solvent (which is cost effective, less viscous, non-corrosive and easy to synthesize) to selectively extract lignin from biomass,. The characteristics as well as the lignin contents of a lignocellulosic biomass would be expected to vary depending on the type and age of plant, weather and growing conditions etc. and above all the techniques employed for the separation (Esteves Costa et al., 2016; Tejado et al., 2007). Lignin is an amorphous tridimensional polymer consisting of three primary units: syringyl (S), guaiacyl (G), and p-hydroxyphenyl (H) units, joined together by ether and CeC linkages (Nada et al., 1998; Tejado et al., 2007). The structure of lignin obtained from oil palm biomass (OPB) is more complex than the structure of wood lignin due to a complex arrangement of these (S), (G) and (H) units in the OPB fiber and shells (Sidik et al., 2013; Singh, 1999). From our previous work it is evident that lignin extraction using [PyFor] is purely physical (Rashid et al., 2016) and the original lignin structure is not disturbed. To the best of our knowledge, there exists a gap in literature data regarding the kinetics of physical lignin extraction process (solid-liquid extraction) from biomass using [PyFor] as a solvent. An important engineering tool in order to design a process is the mathematical modeling to achieve reduced energy, time and solvent consumption (Kitanovic et al., 2008). Generally, kinetic models are categorized as physical and empirical. Physical models deal with the mass transfer phenomenon through plant tissues into the bulk of solvent, while empirical models explain the quantification of solute extracted by a specific solvent with respect to time (Kitanovic et al., 2008; Piwowarska and González-Alvarez, 2012). Though empirical models are simpler than physical models, but they still provide useful information related to the better understanding of the process and hence for upscaling (Abdul-Karim et al., 1995). Although there are a few literatures on empirical delignification kinetics using organic acids and soda pulping (Dang and Nguyen, 2006; Dapıia et al., 2002; Park et al., 1999), nevertheless these separation techniques are more concerned on efficient recovery of cellulose only. Few studies regarding delignification kinetics of wheat straw, bagasse and rice husk biomass have also been reported in literature (Abdul-Karim et al., 1995; Correia et al., 2001; Ho et al., 2011; Korotkova et al., 2015; Park et al., 1999; Tu et al., 2008; Yiwen, 1986). These studies can be summarized as following; Yiwen (1986) studied the mechanism and kinetics of delignification from rice husk and suggested that the whole phenomenon can be divided into three phases i.e. (i) bulk phase (90 °C); (ii) supplementary phase (90–150°C) and (iii) final phase at much higher temperature i.e. (≥150 °C). based on the studies on kraft pulping, Abdul-Karim et al. (1995) concluded that the delignification is governed by first order chemical reaction. Park et al. (1999) systematically analyzed the kinetics and mechanism of delignification as well as silica removal from rice straw and suggested that both process follows first order kinetics, which are similar to that of formic acid pulping of bagasse (Tu et al. (2008)). Correia et al. (2001) studied soda pulping delignification kinetics and calculated the activation energies for lignin removal from

2. Experimental section 2.1. Materials and methods The starting material used for the current study include, Formic Acid 99% (w/w) and Pyridine (used as precursors for [PyFor] synthesis), Karl Fischer titration standards, Dimethyl Sulphoxide (DMSO-d6), Benzene and Ethanol, (used as a solvent for extractives estimation using soxhlet extraction), Acetone (99% purity), Sulphuric acid 98% (w/w), Glacial Acetic acid 99% (w/w), 17.5% sodium hydroxide solution, sodium chlorite (used for the characterization of raw biomass) were purchased from Sigma Aldrich and used as received. Pyridinium formate was synthesized (Rashid et al., 2016). Fresh EFB, PMF and PKS samples were supplied by FELCRA, Nasaruddin Oil Palm Mill, Bota, Perak, Malaysia and stored at ≈5 °C. The raw biomass was washed with detergent solution (2%) in order to remove oil and greases. The washed biomass samples were dried under natural sun light for 24 h before being grinded and crushed using power cutting mill (Pulverizette 25). Subsequently, the grinded biomass samples were sieved to attain three different particle sizes of (0.1–0.3 mm), (0.3–0.5 mm) and (0.5–1.0 mm) respectively. Triple distilled water was used for the experiments.

2.2. Proximate analysis The composition of EFB, PMF and PKS was determined based on the respective standard procedures: (i) acid insoluble and soluble lignin (TAPPI T222 om-02); (ii) holocellulose and a- cellulose (Teramoto et al., 2009); (iii) extractives ((NREL) laboratory analytical procedure (LAP-010)); (iv) moisture content (LAP-001); (v) ash content (LAP005). All the present reported data were taken from the average of the duplicates.

123


T. Rashid et al.

Fig. 1. Extraction pathway of lignin from EFB, PMF and PKS.

extracted lignin was dissolved in DMSO-d6 and the one-dimensional 1H NMR analysis was carried out at 25 °C using Bruker Advance 400 spectrometer, operating at the frequency of 500 MHz.

2.3. Extraction of lignin from oil palm biomass species The extraction path way of lignin from three different oil palm biomass species is shown is Fig. 1. For the lignin extraction process, a predetermined amount of extractives free oil palm biomass species (EFB/PMF/PKS) were loaded in a 100 ml round bottom flask containing appropriate amount of pyridinium formate [PyFor] (Rashid et al., 2016). The flask was mounted in an oil bath and stirred for 2 h at room temperature. After the completion of the extraction process the regeneration of the dissolved lignin was carried out. The oil palm biomass species (EFB/PMF/PKS)/[PyFor] mixture was regenerated using a 1:2 mixture of acetone/water. The regenerated lignin subsequently precipitated out from the acetone/water mixture, whereas the cellulose rich residue settles down. The cellulose rich material was separated and washed several times by filtration through a ceramic funnel, using triple distilled water. While, the lignin was precipitated from the lignin rich filtrate by evaporating acetone in air leaving [PyFor]/water and lignin behind. The precipitated lignin was filtered out using vacuum filtration. The obtained filtrate was subjected to rotary evaporator to recover [PyFor] and the recovered [PyFor] was reused. The extracted lignin and cellulose rich residue was oven dried at 70 °C till constant weight. All the calculations were based on the amount of acid insoluble lignin contents present in the respective biomass species (Sun et al., 2009). The amount of lignin extracted in (%) and in (g/L) were estimated, respectively, as shown below:

3. Results and discussions 3.1. Optimization for lignin extraction parameters Preliminary studies on the initial screening of various parameters were conducted to study their effect on the extraction process. For this purpose, the effect of lignin contents and process variables namely particle size range ((0.1–0.3), (0.3–0.5), (0.5–1.0)) mm, extraction temperature (298, 323, 348, 358, 373, 388) K, time (60, 120, 180) mins, solid loading (2, 5, 10, 15, 20) % w/w respectively on the extraction efficiency of lignin using [PyFor] was investigated, and the details are discussed in the following sections. 3.1.1. Effect of initial lignin contents The compositional analysis of EFB, PMF and PKS used in this study for lignin extraction are listed in Table 1, which satisfactorily agree with those available in the reported literature (Hamzah et al., 2011; Singh, 1999). Table 1 indicates that PKS has the highest amount of initial lignin content. The initial lignin content (%) and lignin extraction efficiency (%wt) achieved from the extraction of lignin from EFB, PMF and PKS using [PyFor] at similar extraction conditions (particle

Weight of lignin extracted ⎞⎟ Lignin extraction %=⎛⎜ ⎝ Acid insoluble lignin content in biomass species ⎠ (1) × 100

Table 1 Composition of different species of Oil Palm Biomass. Oil Palm Biomass Species

Weight of lignin extracted ⎞ Lignin extraction g / L=⎛⎜ ⎟ × 100 ⎝ Amount of solvent used ⎠

(2)

2.4. Characterization of recovered lignin from EFB/PMF/PKS The characterization of the extracted lignin was performed by using FTIR and 1H NMR techniques for its functional groups and purity analysis. The Perkin Elmer spectrometer was employed to obtain the FTIR spectra, by using the wave number in the domain of 3600 cm−1 to 1000 cm−1 with KBr pellet consisting 1% of finely grounded lignin. The

a

124

Components

EFB % Composition

PMF

PKS

Acid insoluble lignin Acid soluble lignin Hemicellulose a-Cellulose Extractives Ash Othersa

17.8 ± 0.22 2.20 ± 0.09 32.11 ± 0.36 36.83 ± 0.47 4.12 ± 0.31 3.53 ± 0.44 2.25

32.2 ± 0.43 2.51 ± 0.21 23.53 ± 0.40 33.63 ± 0.22 3.67 ± 0.33 2.59 ± 0.31 1.19

45.5 ± 0.40 1.15 ± 0.16 17.49 ± 0.34 29.93 ± 0.29 1.59 ± 0.27 2.23 ± 0.28 1.45

Calculated by difference.


T. Rashid et al.

dissolution and fractionation processes. It can be speculated that the reduced extraction efficiency for PMF and PKS was mainly due to the inherent physical differences (lignin content) between the different species. Results in the later section (Fig. 3(b) and (c)) also further revealed that PKS required the highest extraction temperature for the same degree of extraction from EFB and PMF, also a longer extraction time was used for the PMF and PKS (240 min) compared to that for the EFB (180 min). The lignin content of the EFB was lower and has a looser physical structure than PMF and PKS which could allow more efficient extraction at reduced temperature and shorter extraction time (Akinosho et al., 2017; Correia et al., 2001). 3.1.2. Effect of particle size Lignin extraction was studied using three different ranges of particle sizes namely (0.1mm–0.3 mm), (0.3mm–0.5 mm) and (0.5mm–1.0 mm) respectively (Fig. 3a). As expected increasing the particle diameter from (0.1 mm–0.3 mm) to (0.5mm–1.0 mm) results in a decrease in extraction efficiency. A possible explanation for higher lignin extraction at smaller particle size may be attributable to the shorter mass transfer distance and higher surface area of the smaller particles (Muhammad et al., 2013). Also increased penetration of [PyFor] is required in case of larger particles so as to achieve comparable extraction efficiencies in comparison to smaller particles (Muhammad et al., 2013; Pinkert et al., 2011; Tan et al., 2009). As the lignin is trapped in the strong threedimensional polymeric network of lignocellulosic biomass, the penetration of [PyFor] is hindered in case of larger particle size. As there is a sharp decrease in the amount of extracted lignin with respect to the particle size, the smallest particle size range of (0.1 mm–0.3 mm) was selected for further experiments.

Fig. 2. Comparison of initial lignin contents (%) and extraction efficiency (%) for different biomass species (particle size (0.1–0.3 mm), solid loading (10%), extraction temperature (348 K) and extraction time (180 min)).

size (0.1–0.3 mm), solid loading (10%), extraction temperature (348 K) and extraction time (180 min)) are shown in Fig. 2. The lignin extraction efficiency showed the following order: EFB > PMF > PKS which corresponds to the lignin contents of the EFB (17.8%), PMF (32.2%) and PKS (45.5%) respectively, showing a decrease in lignin extraction efficiency with increasing lignin content. These results are in accordance with Correia et al. (2001), where high lignin contents are reported to produce low pulp yields. According to Boeriu and Bravo, (2004), the presence of higher lignin content in the biomass networking contributes to the more recalcitrant characteristics of the material for

3.1.3. Effect of temperature Based on the present results obtained for EFB and PMF biomass

Fig. 3. Effect of operating parameters on lignin extraction from EFB, PMF and PKS.

125


T. Rashid et al.

matrix into the [PyFor] solution which favored mass transfer and hence the extraction efficiency was increased (Qu et al., 2010). However, at a biomass loading of > 10%, the extraction efficiency decreased which is consistent with the reported literature (Li et al., 2015; Tan et al., 2009). The ability of a solvent to dissolve lignin is a function of the cohesive energy density and hydrogen bonding capability of the solvent (Lin and Dence, 1992). Two reasons can possibly explain these observations, i) the presence of excess solvent at a solid loading of < 10% leads to a less cohesive energy between the biomass sample and [PyFor] mixture that results in reduced lignin extraction (Li et al., 2015); ii) the increase in viscosity of the mixture of biomass sample and [PyFor] at a solid loading of > 10% results in a reduction in [PyFor] ion kinesis. Low ion kinesis results in reduced interactions amongst [PyFor] and biomass samples. Also, more lignin is available for a fixed amount of solvent at increased biomass sample loading (Pinkert et al., 2011). Based on the present experimental results, to avoid the usage of excessive amount of solvent the effective biomass loading in the range of 5% to 15% was selected for further optimization studies.

Table 2 The three level variables selected for Box-Behnken design. Factors

Symbol

Coded Levels −1

Temperature (K) Time (min) Solid Loading (%)

X1 X2 X3


X1 X2 X3


X1 X2 X3

EFB 348 60 5 PMF 353 120 5 PKS 358 120 5

0

+1

353 120 10

358 180 15

358 180 10

363 240 15

373 180 10

388 240 15

species the lignin extraction significantly increased within the temperature range of 348 K–358 K followed by a reduction (Fig. 3b). This might be due to an increased solubility of lignin at a high temperature. Beyond 358 K, the lignin extraction does not show any significant increase. An increase in lignin extraction was observed in the temperature range of 358 K–373 K for PKS. High operating temperature and prolonged extraction time had a negative effect on the lignin extraction that could break the cleavage of ether linkages present in lignin which further leads to the presence of other compounds and the degradation of lignin as well (Pan et al., 2006; Sun et al., 2009). The operating temperature required for the same degree of lignin extraction obtained using the same solvent [PyFor] was found to increase with an increase in the lignin content in the biomass species. A possible explanation for the higher operating temperature required for PMF and PKS may be due to the reason that more energy is required to break the strong intermolecular forces and cleavage present amongst the lignin molecules with increased lignin contents (Akinosho et al., 2017; Grabber et al., 2009; Mathiarasi and Partha, 2016). Based on the present experimental results, the effective temperature in the range of 348–373 K was selected for further optimization studies.

3.2. Optimization and statistical analysis of process variables Based on the preliminary experiments, a Box–Behnken design was applied to investigate the influence of three independent variables, i.e., temperature (K) X1, time (min) X2, and biomass loading (%) X3, on percentage extraction of lignin (Y). Based on the complete design matrix, 15 randomized experiments (with three central points) were used for the experiments for EFB, PMF and PKS separately. The three levels that were chosen based on the preliminary study and were coded as low (−1), middle (0) and high (+1) and are presented in Table 2 for EFB, PMF and PKS separately. By applying the Box- Behnken design the generated matrix design of the experiments for EFB, PMF and PKS are shown in Tables 3–5 respectively. The effect of the independent variables on the observed response was explained by using a randomized experimental order (Li et al., 2015), and the following form of polynomial expression was employed to correlate the extraction percentage with other independent variables: k

y = βo +

i=1

3.1.4. Effect of time The extraction time had remarkable impact on lignin extraction (Fig. 3c). The lignin extraction rapidly increased almost linearly with increase in time and then showed a reduction in extraction until reaching equilibrium at 180 mins for EFB, and 240 mins for PMF and PKS respectively (Fig. 3c). The present investigation shows that the increase in extraction time for PMF and PKS is attributable to the additional energy required by [PyFor] to penetrate into the species with high lignin content (Correia et al., 2001; Grabber et al., 2009). Even though, with extended extraction time, lignin extraction started to decrease slightly. The possible reason could be that, an increase in extraction time can possibly lead to a condensation reaction with the carbohydrates or lignin (which was still trapped in the biomass), which accelerates the degradation of lignin during the extraction process and resulted in lower lignin extraction. After that it continued to be constant, signifying that the maximum lignin had been extracted (Mathiarasi and Partha, 2016; Tan et al., 2009). Based on the present experimental results, the effective time in the range of 60–240 mins was selected for further optimization studies.

k

k

k

∑ βixi + ∑ βiixi2 + ∑ ∑ i=1

βijxixj (3)

i=1 j=i+1

Where Y is the response (% extraction); bo is the coefficient constant, b1, b2, b3, b11, b22,…. are the regression coefficients. Based on Eq. (3) a second-degree polynomial equation was used to correlate the percentage extraction of lignin with other independent variables. Table 3 Experimental design matrix of Box–Behnken with observed and fitted values for EFB. EFB Run

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

3.1.5. Effect of solid loading Individual extractives free biomass samples (EFB, PMF and PKS) of (2, 5, 10, 15 and 20) % w/w were loaded with [PyFor] to investigate the effect of sample loading on lignin extraction. It was observed that lignin extraction increased significantly with increasing biomass loading from 2% to 10% (Fig 3d). Higher solid loading resulted in a larger concentration gradient during the diffusion from internal cell 126

Temperature

Time

X1 (K)

X2 (min)

Solid Loading X3 (%)

353 353 348 348 353 353 358 353 353 348 348 358 358 358 353

120 180 120 60 180 120 120 60 120 120 180 60 180 120 60

10 15 5 10 5 10 15 5 10 15 10 10 10 5 15

Lignin Extraction (%) Observed Values

Predicted Values

86.02 80.59 75.25 64.43 81.73 86.02 73.32 48.26 86.02 67.96 92.28 57.68 88.03 74.44 46.46

85.78 81.01 74.22 64.99 80.64 85.78 74.87 49.54 85.78 86.22 92.89 56.59 88.45 73.52 45.99


T. Rashid et al.

X32, and linear effect of X2 on lignin extraction are the most significant at approximately 99.51% of significant level with the p-values < 0.01. In addition, the linear effect of X1 is significant, since the p-value is < 0.05 for all the species which are 0.0482, 0.0096 and 0.0233 for EFB, PMF and PKS extracted lignin respectively. The remaining model terms are not significant. Although these coefficients are not significant, their effect could not be excluded so as to support the ranking of the model because of the R2 for lignin yield shows up to 98.51% variability of the response (Fu and Mazza, 2011). Based on the ANOVA analysis, the factors affecting the response (% lignin extraction) vary in the order as: extraction time > extraction temperature > biomass loading. The coefficient of determination (R2) of the model was ≥0.9951 indicating that 99.51% of the experimental lignin extraction values matched the model predicted values. For validation purpose, experiments were conducted at the estimated optimum conditions (Table 7) obtained from contour plots. The experimental and the values predicted by the present developed model (Eqs. (4)–(6)) for EFB, PMF and PKS with an overall standard error of estimation of 1.793,2.131 and 1.982 respectively, justifies the precision and appropriateness of the present suggested model for the extraction of lignin from oil palm biomass. The results revealed that the proposed model is sufficient to represent the lignin extraction process.

Table 4 Experimental design matrix of Box–Behnken with observed and fitted values for PMF. PMF Run

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Temperature

Time

X1 (K)

X2 (min)


363 358 358 363 353 358 358 358 363 358 363 353 358 353 353

240 240 180 180 240 120 180 180 120 120 180 180 240 120 180

10 5 10 5 10 15 10 10 10 5 15 15 15 10 5


Predicted Values

91.00 77.35 83.50 78.20 85.64 69.96 83.50 83.50 80.89 70.83 80.81 69.81 73.34 65.69 63.20

92.04 76.69 83.91 77.67 85.31 68.67 83.91 83.91 81.46 71.09 81.62 70.59 73.55 64.43 62.11

Table 5 Experimental design matrix of Box–Behnken with observed and fitted values for PKS.

3.3. Kinetic model for lignin extraction process

PKS Run

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Temperature

Time

X1 (K)

X2 (min)


373 373 358 373 358 388 373 358 388 358 373 388 388 373 373

180 120 180 120 180 180 240 240 240 120 180 120 180 240 180

10 15 15 5 5 5 5 10 10 10 10 10 15 15 10

During lignin extraction process, the kinetic aspects should be studied in order to identify the factors affecting the extraction rate during the detachment of lignin from the plant matrix (Chan et al., 2014; Zhao and Liu, 2013). Lignin extraction kinetics is very useful in order to evaluate which type of process conditions might limit the lignin extraction and also for the large-scale design of equipment (Ghasemi et al., 2014; Wahab et al., 2010). In this research a kinetic analysis of the experimental data was performed by applying both first order and pseudo second order kinetic model to predict the lignin extraction phenomenon involved in this study. The results of the preliminary studies and statistical analysis have shown that extraction temperature and time are the only independent variable that had significant effect on lignin extraction and hence, should be considered for further kinetics studies (Piwowarska and González-Alvarez, 2012). In order to obtain kinetic data for lignin extraction, it was chosen to work with the smallest particle size of (0.1 mm–0.3 mm) and a fixed solid loading of 10%. The temperature range was studied from 323 K–373 K and time in the range from 60 mins–300 mins was selected. For the kinetic evaluation, the concentration (g/L) versus time (t) data was followed at five different temperatures (323, 333, 348, 358 and 373) K as shown in Fig. 4.


Predicted Values

86.74 56.96 65.81 69.83 65.20 77.20 82.55 78.83 79.25 60.55 86.74 62.30 63.81 79.52 86.74

85.38 57.44 66.43 70.18 66.04 77.92 81.99 79.56 80.87 61.40 85.38 61.87 62.99 80.57 85.38

YEFB (%) = −4472.47 + 26.2048X1 + 0.148098X2 − 12.2886X3 − 0.0385 75 X12 + 0.00209167 X1X2 + 0.06172 X1X3 − 0.0026242 X22+ 0.0005425 X2X3 − 0.492365 X32 (4)

3.3.1. Pseudo second order kinetics modeling of lignin extraction Analysis of the data on the present extraction process(Fig 4), proved that the extraction process does not follow the first order kinetics to explain the lignin extraction from oil palm biomass using [PyFor]. Kumar et al. (2010) suggested a pseudo second order equation to represent the dissolution kinetic profiles of various solids in solvent. Based on our preliminary research (Rashid et al., 2016) the extraction of lignin using [PyFor] is physical in nature, though a small amount of other components i.e. cellulose and hemicellulose and are also extracted, but that are considered to be negligible. Following to this it can be assumed that the lignin extraction follows a pseudo second order model as the extraction is only a function of lignin concentration in the [PyFor] solvent irrespective of the amount of raw biomass used in the extraction (Ghasemi et al., 2014; Ho, 2006). The pseudo second order kinetic model can be written as:

(%) = −5564.96 + 28.5002X1 + 2.26217X2 + 22.6062X3 − YPMF 0.0360805 X12 − 0.00570575 X1X2 − 0.040095X1X3− 0.00028847 6X22−0.00261725 X2X3 − 0.383612X32 (5) YPKS (%) = −6648.94 + 34.9808X1 + 0.818214X2 + 21.9154X3 −0.0460402X12 − 0.000370639X1X2 − 0.0466667X1X3 − 0.00170857X22 + 0.00819917 X2X3 − 0.33506 X32 (6) where, X1, X2 and X3 are the coded independent variables, the terms X12, X22 and X32 are the quadratic (squared) effects, and X1X2 X1X3 and X2X3 are the interaction effect respectively. The analysis of variance (ANOVA) was performed to identify the significant factors affecting the lignin extraction at P-value = 0.05 and the results are shown in Table 6. According to Guo et al. (2010), Sidik et al. (2013) and Tan et al. (2011) when the p-value is < 0.05 the corresponding coefficient is more significant in the model. In the present study, quadratic effects of X22 and

dCt = K2 (Cs − Ct )2 dt

(7)

The integrated form of rate law for a pseudo second order extraction 127


T. Rashid et al.

Table 6 Analysis of variance (ANOVA) for the EFB, PMF and PKS lignin extraction as the desired response. Source

Mean Square

X1: Temperature X2: Time X3: Loading X12 X22 X32 X1X2 X1X3 X2X3 Total error Total (corr.)

Df

EFB

PMF

PKS

5.18 1978.05 16.08 3.43 329.53 559.43 1.57 9.52 0.10 9.00 2884.67

306.80 230.65 2.36 3.00 3.98 339.59 11.72 4.01 2.46 18.52 990.68

18.52 621.27 102.70 396.22 139.69 259.07 0.44 49.0 24.20 1.77 1521.12

F-Ratio

1 1 1 1 1 1 1 1 1 5 14

P-Value

EFB

PMF

PKS

EFB

PMF

PKS

0.58 219.61 1.79 0.38 36.59 62.11 0.17 1.06 0.01

16.56 12.45 0.13 0.16 0.21 18.33 0.63 0.22 0.13

10.42 349.37 57.76 222.81 78.55 145.69 0.25 27.55 13.61

0.0482 0.0000 0.2390 0.5640 0.0018 0.0005 0.6932 0.3510 0.9179

0.0096 0.0168 0.7353 0.7038 0.0066 0.0079 0.4625 0.6610 0.7302

0.0233 0.0000 0.0006 0.0000 0.0003 0.0001 0.6381 0.0033 0.0142

Table 7 Lignin extraction efficiency and optimum operating conditions obtained by [PyFor] dissolution. % Extraction Oil Palm Biomass

EFB PMF PKS

% Extraction

Observed Values

Optimum Operating Conditions

92.01 91.23 90.70

Predicted Values

Temperature (K)

Time (min)

Loading (%)

X1

X2

X3

351.5 361.9 370.8

172 240 238

9.58 9.33 9.21

92.98 90.02 91.06

Fig. 4. Experimental data for the kinetic evaluation lignin extraction from (a) EFB, (b) PMF and(c) PKS biomass using [PyFor] at various studied temperatures.

128


T. Rashid et al.

studied species (EFB, PMF and PKS) exhibit different kinetic profiles, nevertheless irrespective of the different kinetic profiles the overall lignin extraction trend is the same. The temperature dependence of “Cs” is shown in Fig. 6(a), which indicates that the rate of extraction is fast at the initial stages of the operation and then become slower after reaching the saturation concentration for each studied temperature (Bogdanov and Svinyarov, 2013; Rakotondramasy-Rabesiaka et al., 2007). The lignin saturation concentration follows the following order: EFB < PMF < PKS (Fig. 6) Therefore it can be assumed that lignin content in biomass species may limit the extraction process. Akinosho et al. (2017) correlated the ethanol yield and fermentation accessibility with lignin contents and concluded that, an increase in lignin contents results in more intense and complex cross-linking resulting in a more recalcitrant nature of the biomass, as a result of which the extraction process decelerates as a whole (Studer et al., 2011). The temperature dependence of extraction rate constant “K2” and initial extraction rate “h” is shown in Fig 6(b) and (c), which indicates that the extraction rate constant “K2” is higher in EFB as compared to PMF and PKS respectively, at the same studied conditions. It also shows that the initial extraction rate follows the same trend as that of the extraction rate constant. It can be assumed that rate of extraction is directly dependent on the lignin contents of, showing an overall decrease in K and h value with increasing lignin content. As the lignin content increases the lignin macromolecules are packed closer to each other that reveal the major mass transfer resistance (Ansari and Gaikar, 2014; Lin and Dence, 1992; Stylianopoulos et al., 2010). It could also be assumed that using same solvent the detachment of lignin in EFB is easier than that of PMF and PKS and this may be attributed to a reduced initial resistance of the plant cell walls and hence better mass transfer (Bogdanov and Svinyarov, 2013). Similar trends for an increase in lignin contents were related to a lower rate and extent of cell wall fermentation (Grabber et al. (2009)).

model can be obtained by considering the initial and boundary conditions as; Ct = 0 at t = 0; Ct = Ct at t = t and

Ct =

K2 tCs 2 1 + K2 tCs

(8)

The linearized expression of Eq. (8) is given by;

Ct 1 = t (1/ K2 Cs 2) + (t/ Cs )

(9)

when t = 0, the initial extraction rate is given by;

h = K2 Cs 2 where Cs (g/L) and Ct (g/L) are the concentration of lignin in the solution at saturation and at any time t (min) respectively, and K2 is the second order extraction rate constant (L/g min). To study the applicability of the model, the experimental data are plotted between t/Ct vs t for EFB, PMF and PKS respectively (Fig. 5). The plot of t/Ct vs t is linear and the theoretical saturation concentration “Cs”, second order rate constant “K2”, initial extraction rate “h”, are determined from the slopes and intercepts of the plots along with the coefficient of correlation (R2) at all studied temperatures (Table 8). The coefficient of correlation (R2) obtained for pseudo second order kinetic model is in the range of 0.99–0.97 at all the studied temperatures. The higher R2 value clearly suggests that the proposed model successfully represents the extraction kinetics of lignin and it is assumed to proceed in two steps, i.e. surface detachment of lignin followed by diffusion step (Kumar et al., 2010) which is a second-order process. From Fig. 4 it is evident that the three

Fig. 5. Linearization of Second order kinetic model for the extraction of lignin from (a) EFB, (b) PMF and (c) PKS using [PyFor] at various temperatures.

129


T. Rashid et al.

Table 8 Pseudo second order kinetics parameters obtained for the [PyFor] supported extraction of lignin from EFB, PMF and PKS at different temperatures. Extraction Temperature (K) Biomass Species (EFB) 323 333 348 358 373 Biomass Species (PMF) 323 333 348 358 373 Biomass Species (PKS) 323 333 348 358 373

Slope (lit/gm)

Cs Calculated (g/L)

Cs Experimental (g/L)

Intercept (min lit/g)

K (L/g min × 104)

h (gm/lit min)

R2

0.0678 0.0611 0.0569 0.0592 0.0596

14.749 16.367 17.575 16.892 16.779

14.201 15.812 16.931 16.499 15.931

1.125 1.203 1.313 1.517 1.694

38.07 31.05 24.65 23.10 20.97

0.889 0.832 0.761 0.659 0.590

0.996 0.997 0.991 0.987 0.984

0.0537 0.0424 0.0360 0.0329 0.0360

18.609 23.607 27.749 30.367 27.778

19.51 24.1 27.1 30.36 28.2

1.314 1.505 1.675 1.710 1.795

21.97 11.92 7.75 6.34 7.22

0.761 0.664 0.597 0.585 0.557

0.997 0.992 0.985 0.985 0.973

0.0430 0.0370 0.0310 0.0275 0.0232

23.256 27.027 32.258 36.364 43.103

21.2 26.1 32.3 37.0 42.3

1.492 1.542 1.633 1.758 1.8101

12.390 8.880 5.880 4.300 2.970

0.670 0.649 0.612 0.569 0.552

0.988 0.992 0.992 0.984 0.982

Fig. 6. Relationships between the temperature and lignin concentration at (a) saturation (Cs), (b) extraction rate constant (K), (c) initial extraction rate (h) and (d) Relationship between 1/T and the second order rate constant ln k, for the extraction of lignin from EFB, PMF and PKS using [PyFor] at various temperatures.

lignification of the PKS in comparison to the PMF and EFB biomass. These results are consistent with the similar results reported where the nature of hemp (lignin proportions) was found to be a major contributor to the chemical pulping behavior (Correia et al., 2001). Therefore, it can be considered that the temperature effect on the lignin extraction comes from its diffusion phenomenon instead of its influence on lignin solubility in [PyFor] (Bogdanov and Svinyarov, 2013; d’Alessandro, 2012). Table 8 shows that the extraction rate decreases

3.3.2. Thermodynamic studies of lignin extraction An increase in temperature could break the lignin polymeric network and facilitate the release of extractable lignin and therefore facilitate the extraction process (Boonkird et al., 2008; d’Alessandro, 2012). The results (Fig. 3b) showed that the amount of lignin removed in each of the biomass species is affected by the extent of the lignin polymeric network present. The difficulty experienced in extracting lignin from PKS may partly result from the higher degree of 130


T. Rashid et al.

was assumed to be constant. In a simple manner the lignin extraction process functions in two different mechanistic routes: (i) diffusion of the [PyFor] into biomass matrix, subsequent lignin dissolution into the [PyFor], and (ii) convection of the lignin over the porous structure of the cellulosic rich residue and its transfer into the [PyFor] solution. The lignocellulosic components i.e. cellulose, hemicellulose and lignin are interlinked through hydrogen and covalent bonding (Rashid et al., 2016; Remsing et al., 2008). It is evident from the experimental and theoretical studies that H-bonding plays a vital role in the lignin dissolution process (Rashid et al., 2016); it depends on the capability of an anion to dislocate the H-bonding network of cellulose and lignin. From our previous work it is evident that lignin is more soluble in the solvent with a higher degree of protonation (e.g. more ionicity) (Glas et al., 2014; Hart et al., 2015; Rashid et al., 2016). It is noteworthy that using [PyFor] a protic ionic liquid this disruption of H-bonding is enhanced by formate anion and lignin is extracted efficiently, leading to an increase in the apparent extraction kinetics (Rashid et al., 2016). Following to this at the first stage, a very high concentration gradient exists between the easily available lignin present in the finely grinded biomass feedstock, resulting in a rapid penetration of fresh [PyFor], the H-bonding in lignin is attacked by formate anion and hence the apparent extraction proceeds efficiently. In this stage, [PyFor] solvent having high polarity and lower viscosity will enhance its penetration inside the biomass matrix, which helps to improve the diffusion and solubility of lignin with greater extraction efficiency and mass transfer rates. Regarding the viscosity, the conventional IL’s generally possess very high viscosity (compared to PIL’s) that reduces their (IL’s) significance in biomass extraction applications (Achinivu et al., 2014; Rashid et al., 2016). Moreover, although lignin is highly soluble in [PyFor], only a partial extraction of it can take place from the biomass feedstock, specifically at room temperature (293 K). The second phase of the extraction process is a constantly falling extraction rate. During the beginning of this phase, the biomass feedstock started swelling and the reverse flow of the [PyFor] solvent hindering the lignin–[PyFor] compound convection toward the external surface. Nonetheless, the process is still governed by the substantial concentration gradient. At the end of this phase, when the saturation point is nearly reached, only a slow interior diffusion plays a significant role in the lignin–[PyFor] complex transfer into the extract. Based on the above perceptive, and in view of the kinetic parameters achieved, it can be established that the actual lignin extraction kinetics using [PyFor] is attributable to Hbonding interactions between both ions [PyFor] and lignin, which results into cell tissue disruption and modifications. In order to elucidate the matrix–solvent interaction and to prove the above hypothesis the microstructure of untreated EFB samples, and those treated with [PyFor] were examined by scanning electron microscopy (SEM) (Fig. 7). The SEM analysis clearly showed the significant physical changes of the plant tissues after treatment with [PyFor]. From Fig. 7 it is evident that the [PyFor] treated residue has more agglomeration into macrostructure texture which is due to the dissolution of fibers and debris during the extraction and regeneration process (Achinivu et al., 2014). The change in texture during [PyFor] pretreatment ascertained the dissolution of oil palm biomass. Similar observations have been previously reported for the extraction of polyphenolic compounds from medicinal plants using ionic liquid supported mild extraction conditions (Du et al., 2009).

with an increase in temperature, this confirms that the extraction process is driven by concentration gradient and similar results were reported by Kumar et al. (2010). A reduced lignin concentration at higher temperature i.e. 358 K, 373 K and 388 K was observed for EFB, PMF and PKS respectively. At increased temperatures the release of side products increases and as a consequence the “Cs” began to decrease (Casas et al., 2012; Nada et al., 1998). This study is of great significance, as sometimes the increase in temperature can promote the degradation of the lignin contents during the extraction process resulting in reduced lignin concentration in [PyFor] (Kubo and Kadla, 2005). The extraction rate constant could be related to extraction temperature with Arrhenius equation;

K = K o e (−Ea/ RT )

(10) −1

−1

Where Ko is the temperature independent factor (L g min ), Ea is the activation energy for lignin extraction (J mol−1), R is the gas constant (8.314 J mol−1 K−1) and T is the temperature (K), Eq. (10) can be linearized between extraction rate constant (K) and the inverse of temperature as follows;

−E 1 lnk = lnk o + ⎛ a ⎞ ⎝ R ⎠T

(11)

Using Eq. (11) the activation energies Ea were calculated by plotting lnK vs 1/T for lignin extraction from EFB, PMF and PKS. The estimated activation energies for the lignin extraction from EFB, PMF and PKS are 12 kJ mol−1, 23 kJ mol−1 and 28 kJ mol−1 respectively, with the lowest lignin contents EFB (17.8%) produced the lowest activation energy. This can be suggested that the Ea is dependent on the resistance between matrix and medium through which lignin as a solute should overcome (Bogdanov and Svinyarov, 2013). These reported values of activation energy are in the range of 16–131 kJ mol−1 for the pulping of rice or wheat straws by employing different techniques (Abdul-Karim et al., 1995; Dang and Nguyen, 2006; Epelde et al., 1998; Ho et al., 2011; Huang et al., 2007; Yiwen, 1986). Yiwen (1986) applied soda cooking to rice straw reported the activation energy as 49.7 kJ mol−1. For kraft cooking of wheat straw Abdul-Karim et al. (1995) reported activation energy as 131 kJ mol−1, whereas Epelde et al. (1998) employed soda and kraft pulping and the reported activation energy for wheat straw was 93 kJ mol−1. Huang et al. (2007) reported the activation energy of lignin removal from rice straw by using NH4OH-KOH as 35.6 kJ mol−1, while using a variety of softwoods and hardwoods Dang and Nguyen (2006) reported the activation energies in the range of 71–136 kJ mol−1. From these, it is evident that [PyFor] for selective lignin extraction required considerably a very low activation energy than those of chemical pulping methods previously reported in literature. The delignification conditions, activation energies and the models employed for all of these studies are summarized and compared with that for the present work (Table 9). 3.4. Phenomenon of lignin extraction The lignocellulosic biomass cell wall comprises of three basic layers namely; primary wall, middle lamella and the secondary wall. Over 50% of total lignin contents are present in the primary wall and the middle lamella (Overend et al., 2012). The secondary wall is the major portion of the lignocellulosic biomass fiber and contains only 10% of lignin contents (Goto et al., 1990). In order to describe the lignin transfer from the lignocellulosic biomass to the solvent [PyFor], the following basic assumptions were considered: To explain the mechanism of lignin extraction from the oil palm biomass, a basic treatment of such complex phenomenon is required. (i) though a small amount of other components i.e. hemicellulose and extractives are also extracted, but it was assumed that only the lignin extraction is taking place (ii) the extraction was considered to proceed from the diffusion of lignin from the cell walls into the [PyFor] solution; (iii) the concentration of lignin at the solution saturation under the same conditions

3.5. Effect of solvent recycling on extraction efficiency In order to establish the environmental friendly and industrially economical pretreatment of biomass, the solvent must be recyclable without losing its extraction efficiency. To validate this, solvent recovered from the extraction process was subjected to a rotary vacuum evaporator for 6 h at 80 °C and 30 kpa, and more than 98% mass of [PyFor] was recovered, which indicates that [PyFor] is potentially a viable and recyclable solvent. The recovered solvent was tested for 131


T. Rashid et al.

Table 9 Lignin extraction kinetics studies and activation energies from literature. Biomass

Pulping Regime

Conditions

Process and Model Used

Activation energy (kJ mol−1)

Year

Reference

Rice straw

soda cooking

423 K

49.7

1986

Yiwen (1986)

Wheat straw

Kraft cooking

413 K

Three stage transition process First-order

131

1995

Wheat straw Rice straw Canadian hemp hemp, wheat straw, bamboo Rice straw Bagasse Rice straw

Soda and Kraft pulping soda pulping alkaline pulping Organosolv Pulping

403 413 K, 240 min 443 K 210 min (423–453) K

first order pseudo first order reaction first order first order kinetics

93, 127 56.6 76 71–136

1998 1999 2001 2006

NH4OH–KOH pulping Organosolv Pulping Tetrahydrofurfuryl Alcohol/HCl Pulping low concentration aqueous alkali, high pressure [PyFor]

428 K 418 K 393 K

pseudo-first-order first order two phase delignification process first order kinetics

35.6 na 32.2

2007 2008 2011

Abdul-Karim et al. (1995) Epelde et al. (1998) Park et al. (1999) Correia et al. (2001) Dang and Nguyen., (2006) Huang et al. (2007) Tu et al. (2008) Ho et al. (2011)

na

2015

Pseudo Second-order

12 23 28

2017

spruce wood EFB PMF PKS

443 K, 10.34 mPa. 240 mins 351 K, 172 min 361 K, 240 mins 370 K, 238 mins

Korotkova et al. (2015) Present Work

vary depending on the lignin source and the intensity of separation techniques employed (Mohtar et al., 2015). The FTIR spectra of the present EFB, PMF and PKS extracted lignin (Table 10) are compared with that of lignin extracted from oil palm empty fruit bunch reported by (Mohamad Ibrahim et al. (2011) which confirms the presence of lignin backbone in the fingerprint regions. An extensive absorption band appearing at 3300 cm−1− 3500 cm−1 is attributed to the OeH vibrations in the aromatic and aliphatic OeH groups of lignin (Lourençon et al., 2015). The CeH vibration of methoxyl group in lignin appears at 2929 cm−1, 2825 cm−1 and 1460 cm−1 while the aromatic vibration of phenylpropane skeleton appears at 1600 cm−1. The aromaticity of lignin structure lies in aromatic skeletal vibration of guaiacyl (G) and syringyl (S) units in the band width of 1600 cm−1 to 1000 cm−1 (Casas et al., 2012; Dodd et al., 2014). All specific peaks are detected in the spectrum of the three extracted lignins confirmed that the backbone of lignin is not disturbed during the extraction and regeneration process (Kubo and Kadla, 2005). Due to the difference in the morphological parts of the biomass, some significant individual characteristics were noticed. The syringyl (S) units appear at 1326 cm−1, whilst guiacyl (G) units appear at 1269 cm−1 specifically in the FTIR spectra of lignin (Tejado et al., 2007). It can be seen from Fig. 8(a) that all three lignins showed bands at 1326 cm−1 and 1269 cm−1, however the intensity of these units is higher in case of PKS lignin. The band at 1125 cm−1 which is normally attributed to the deformation of CeH bond for aromatics in syringyl (S) units (Medina et al., 2015), was found to be absent in PKS lignin. It can be speculated that EFB and PMF lignin have more (S)

three extraction cycles and the lignin extraction efficiencies for EFB are found to be 91.81%, 91.50%, and 91.15% for extraction cycles of 1, 2 and 3 respectively. The recycled [PyFor] showed relatively comparable extraction efficiency as that of fresh [PyFor]. The small reduction in extraction efficiencies (≤0.3%) might be due to the residual water content in the solvent (which was not reduced to ppm level during the recycling experiments). The extraction efficiencies could be enhanced by reducing the residual water content to an appropriate ppm level during the regeneration of [PyFor]. During the regeneration and recycling process, the solvent purity and stability after each cycle were determined using 1H NMR. It was observed that the spectra in case of first and third recycle are comparable to the pure [PyFor] which indicates that no side products are formed during the dissolution process and the molecular structure of the [PyFor] was not disturbed.

3.6. Confirmation and comparison of extracted lignin structure 3.6.1. Fourier transform infrared spectroscopy In order to further elucidate the results, chemical structure analysis of the extracted lignin samples was carried out by FTIR spectroscopy. The FTIR spectroscopy is a common technique used for establishing the functional groups and structural features of lignin at particular vibrations. The qualitative FTIR analysis for the confirmation of functional groups such as hydroxyl, carbonyl, methoxyl, and carboxyl was performed in the range from 3500 to 1000 cm−1 (Wang and Chen, 2016). In general, the FTIR spectra of lignin have common features and particular vibrations, although the intensities of the bands in lignin may

Fig. 7. SEM micrographs of the untreated EFB sample (left), and after treatment with [PyFor] (right).

132


T. Rashid et al.

Table 10 The FTIR band assignments of [PyFor] extracted EFB, PMF and PKS lignin. EFB (Mohamad Ibrahim et al., 2011)

EFB This Work

PMF This Work

PKS This Work

Band (cm−1) 3404 2937 2841 1709

1327 1269

Band (cm−1) 3419.79 2935.47 2850.11 1717.48 1651.09 1600.82 1557.40 1512.56 1458.13 1425.58 1371.31 1325.75 1270.77

Band (cm−1) 3419.79 2935.47 2850.10 1716.21 1651.09 1601.45 1554.53 1512.56 1459.57 1425.58 1372.59 1325.75 1269.49

1033

1034.38

1034.38

Band (cm−1) 3419.79 2935.25 2849.19 1717.48 1652.37 1599.88 1554.53 1513.83 1458.13 1424.30 1371.31 1326.48 1269.49 1173.09 1125.53 1029.04

1606 1513 1461 1426

Assignment

OeH Stretching CeH Stretching CeH stretching (OCH3) CeH vibration of aromatic ring (Unconjugated) CeH vibration of aromatic ring (Conjugated) CeC stretching (aromatic skeleton) Aromatic skeletal breathing with C]O stretching Aromatic skeletal vibration, G* > S** CeH deformations methyl and methylene CeH deformation and aromatic skeletal vibrations Phenolic OH and aliphatic CeH in methyl groups Syringyl unit vibrations Guaiacyl ring breathing with carbonyl stretching (G) units Aromatic CeH in-plane deformation in the guaiacyl ring CeO deformation (methoxyl group) CeO deformations of aliphatics and alcohols, aromatic CeH in-plane deformation (G > S)

* Guaiacyl unit. ** Syringyl unit.

units. The presence of guaiacyl (G) unit in all three lignins confirmed that they had potential active site for polymerization (Mohamad Ibrahim et al., 2011). Bands are observed in the non-conjugated and conjugated carbonyl region at 1705–1715 cm−1 and 1650 cm−1 respectively (Boeriu and Bravo, 2004; Tejado et al., 2007), and the spectra in these region (Fig. 8(a)) shows that [PyFor] extracted lignins are rich in aromatic rings and are chemically unmodified which will eventually promote the commercial applications of lignin for valuable product such as lignin based carbon fibers, isocyanate binders, bio dispersants, phenolic and thermosetting resins (Casas et al., 2012; Tejado et al., 2007). During lignin isolation (Kraft and soda pulping, organosolv, ionic liquids process), the b−O−4 and a−O−4 linkages are cleaved and non-etherified phenolic OH groups are left in lignin (Singh and Dhepe, 2016). These groups appear at 1375 cm−1 in the FTIR spectra of lignin which is visible for all three [PyFor] extracted lignins studied depicting that beOe4 and aeOe4 linkages are cleaved to the required extent. A fair degree of comparison of the FTIR spectra of EFB (Mohamad Ibrahim et al., 2011) (Table 10) and the present extracted lignins indicate the successful and selective extraction of lignin from oil palm biomass samples using [PyFor]. 3.6.2. 1H nuclear magnetic resonance analysis Analysis of signal intensity using 1H NMR provides an adequate means of monitoring the purity of the extracted lignin structure. The 1H NMR spectra of EFB, PMF and PKS lignin are shown in Fig. 9 where the signals are assigned according to the literature (Rashid et al., 2016) (Table 11). The aromatic protons in guaiacyl (G) units and the syringyl (S) units are indicated by an amplification of signals between 8.1 and 6.4 ppm (Singh and Dhepe, 2016; Zhou et al., 2012). The sharp peak at 2.7 ppm is due to the solvent (DMSO-d6) used to prepare the samples (Rashid et al., 2016). The chemical shift between 4.1 and 3.2 ppm are attributed to methoxyl protons (e-OCH3) (Esteves Costa et al., 2016). The beOe4 structures appear at a resonance of 6.6–5.9 ppm. The peaks between 2.7–2.1 ppm and 2.1–1.7 ppm are assigned to the phenolic proton and aliphatic protons respectively (Esteves Costa et al., 2016; Costa et al., 2014). Similar results for acid/alkali and kraft lignin from EFB were reported by Medina et al. (2015) and Mohamad Ibrahim et al. (2011). From the present comparison, it can be concluded that the amplification of the signals in case of [PyFor] extracted lignins are comparable to the literature (Mohamad Ibrahim et al., 2011; Medina et al., 2015) indicating that the “core” of the lignin is the same and lignin is extracted successfully from oil palm biomass.

Fig. 8. (A) FTIR spectra of (a) EFB lignin (b) PMF lignin (c) PKS lignin from 1800 cm−1 to 1100 cm−1, (B) Detail of finger print region of (a) EFB lignin (b) PMF lignin (c) PKS lignin from 3600 cm−1 to 1200 cm−1.

133


T. Rashid et al.

Fig. 9. 1H NMR spectra of Indulin kraft lignin and [PyFor] extracted lignins.

temperatures till date for lignin extraction, furthemore a short operational time (180 min, 240 min, 238 min) was used indicating the feasibility and milder conditions of the process. The activation energies for the lignin extraction using [PyFor] were found to be considerably lower as compared those reported in literature for traditional wood pulp. This is of great significance from an industrial viewpoint, as a significant amount of energy could be economized if such a process is implemented. The characterization results using FTIR and 1H NMR proved that lignin structure was preserved during the extraction process. The recyclability experiments on [PyFor] proved the sustainability of the extraction process. Pyridinium Formate [PyFor] as a solvent is a cost effective, less viscous, non-corrosive and hence its application for the enhanced extraction of lignin will contribute to the diversification of the biomass feedstock supply for bio-based products and biorefinery purposes.

Table 11 Shift assignment for 1H NMR spectrometry of the investigated lignin samples. Proton shift Assignments for 1HNMR spectrometry for lignin Signal (ppm)

Assignment

8.1–6.4 6.1–5.7 5.7–5.2 4.1–3.2 2.7–2.1 2.1–1.7 1.7–0.5

Aromatic H in S and G units beOe4 and be1 Phenylcoumaran Methoxyl H H in Phenolics H in Aliphatics −CH2 and −CH3

4. Conclusions The temperature dependence of extraction kinetics of lignin from oil palm biomass using [PyFor] was investigated in a comparative approach with respect to the type of biomass species. All the kinetic parameters of lignin extraction were estimated. Good agreement was found between the experimental and theoraticel lignin sturation concentration. Based on the present results it was found that the new solvent, Pyridinium Formate [PyFor], a protic ionic liquid, can be efficiently used for the extraction of lignin from empty fruit bunches. The extraction efficieny was found to be dependent on the initial lignin contents present in raw biomass. A maximum lignin extraction of 92.01%., 91.23% and 90.70% for EFB, PMF and PKS respectively was achieved at the optimum values. The current processing temperatures (351.5 K, 361.9 K,370.8 K) are the lowest amongst the reported

Acknowledgements The authors wish to highly acknowledge the financial support given by Universiti Teknologi PETRONAS Malaysia to Ms Tazien Rashid in the form of GA. The authors also wish to acknowledge the financial support by YUTP-FRG research scheme ((0153AA-H44) for the present research References Abdul-Karim, L.R., Polyanszky, A.E., 1995. Kinetics of delignification in kraft pulping of wheat straw and hemp. Tappi J. 8, 161–164. Achinivu, E.C., Howard, R.M., Li, G., Gracz, H., Henderson, W.A., 2014. Lignin extraction

134


T. Rashid et al.

isolation from spruce wood with low concentration aqueous alkali at high temperature and pressure: influence of hot-water pre-extraction. Green Chem. 17, 5058–5068. Kubo, S., Kadla, J.F., 2005. Hydrogen bonding in Lignin: a fourier transform infrared model compound study. Biomacromolecules 6, 2815–2821. Kumar, P., Barrett, D.M., Delwiche, M.J., Stroeve, P., 2009. Methods for pretreatment of lignocellulosic biomass for efficient hydrolysis and biofuel production. Ind. Eng. Chem. Res. 48, 3713–3729. Kumar, K.V., Khaddour, I.A., Gupta, V.K., 2010. A pseudo second-order kinetic expression for dissolution kinetic profiles of solids in solutions. Ind. Eng. Chem. Res. 49, 7257–7262. Li, H.-Z., Zhang, Z.-J., Hou, T.-Y., Li, X.-J., Chen, T., 2015. Optimization of ultrasoundassisted hexane extraction of perilla oil using response surface methodology. Ind. Crops. Prod. 76, 18–24. Lin, S.Y., Dence, C.W., 1992. Methods in Lignin Chemistry. Springer, pp. 33–61. Lora, J.H., Glasser, W.G., 2002. Recent industrial applications of lignin: a sustainable alternative to nonrenewable materials. J. Polym. Environ. 10, 39–48. Lourençon, T.V., Hansel, F.A., da Silva, T.A., Ramos, L.P., de Muniz, G.I.B., Magalhães, W.L.E., 2015. Hardwood and softwood kraft lignins fractionation by simple sequential acid precipitation. Sep. Purif. Technol. 154, 82–88. Lu, F., Ralph, J., 2010. Chapter 6 – lignin. In: Sun, R.-C. (Ed.), Cereal Straw as a Resource for Sustainable Biomaterials and Biofuels. Elsevier, Amsterdam, pp. 169–207. Mansouri, N.-E.E., Salvado, J., 2006. Structural characterization of technical lignins for the production of adhesives: application to lignosulfonate, kraft, soda-anthraquinone, organosolv and ethanol process lignins. Ind. Crops. Prod. 24, 8–16. Mathiarasi, R., Partha, N., 2016. Optimization: kinetics and thermodynamic studies on oil extraction from Daturametel Linn oil seed for biodiesel production. Renew. Energ. 96, 583–590. Medina, J.D.C., Woiciechowski, A., Zandona Filho, A., Noseda, M.D., Kaur, B.S., Soccol, C.R., 2015. Lignin preparation from oil palm empty fruit bunches by sequential acid/ alkaline treatment–a biorefinery approach. Bioresour. Technol. 194, 172–178. Mohamad Ibrahim, M.N., Zakaria, N., Sipaut, C.S., Sulaiman, O., Hashim, R., 2011. Chemical and thermal properties of lignins from oil palm biomass as a substitute for phenol in a phenol formaldehyde resin production. Carbohydr. Polym. 86, 112–119. Mohtar, S., Busu, T.T.M., Noor, A.M., Shaari, N., Yusoff, N., Bustam, M., 2015. Extraction and characterization of lignin from oil palm biomass via ionic liquid dissolution and non-toxic aluminium potassium sulfate dodecahydrate precipitation processes. Bioresour. Technol. 192, 212–218. Muhammad, N., Man, Z., Bustam, M.A., Mutalib, M.I.A., Rafiq, S., 2013. Investigations of novel nitrile-based ionic liquids as pre-treatment solvent for extraction of lignin from bamboo biomass. J. Ind. Eng. Chem. 19, 207–214. Nada, A.-A.M.A., Yousef, M.A., Shaffei, K.A., Salah, A.M., 1998. Infrared spectroscopy of some treated lignins. Polym. Degrad. Stab. 62, 157–163. Overend, R.P., Milne, T., Mudge, L., 2012. Lignin. Fundamentals of Thermochemical Biomass Conversion. Springer Science & Business Media, pp. 61–76. Pan, X., Kadla, J.F., Ehara, K., Gilkes, N., Saddler, J.N., 2006. Organosolv ethanol lignin from hybrid poplar as a radical scavenger: relationship between lignin structure, extraction conditions, and antioxidant activity. J. Agric. Food. Chem. 54, 5806–5813. Park, S.-Y., Koda, K., Matsumoto, Y., Meshitsuka, G., Iiyama, K., 1999. Kinetic comparison between delignification and silica removal during alkaline pulping of rice straw. Japan TAPPI J. 53, 1492–1499. Pinkert, A., Goeke, D.F., Marsh, K.N., Pang, S., 2011. Extracting wood lignin without dissolving or degrading cellulose: investigations on the use of food additive-derived ionic liquids. Green Chem. 13, 3124–3136. Piwowarska, N., González-Alvarez, J., 2012. Extraction of antioxidants from forestry biomass: kinetics and optimization of extraction conditions. Biomass. Bioenerg. 43, 42–51. Putro, J.N., Soetaredjo, F.E., Lin, S.-Y., Ju, Y.-H., Ismadji, S., 2016. Pretreatment and conversion of lignocellulose biomass into valuable chemicals. RSC Adv. 6, 46834–46852. Qu, W., Pan, Z., Ma, H., 2010. Extraction modeling and activities of antioxidants from pomegranate marc. J. Food Eng. 99, 16–23. Rakotondramasy-Rabesiaka, L., Havet, J.-L., Porte, C., Fauduet, H., 2007. Solid–liquid extraction of protopine from Fumaria officinalis L. – analysis determination: kinetic reaction and model building. Sep. Purif. Technol. 54, 253–261. Ralph, J., Brunow, G., Boerjan, W. Lignins, 2007. Encyclopedia of Life Sciences. John Wiley & Sonshttp://dx.doi.org/10.1002/9780470015902.a0020104. Rashid, T., Kait, C.F., Regupathi, I., Murugesan, T., 2016. Dissolution of kraft lignin using protic ionic liquids and characterization. Ind. Crops. Prod. 84, 284–293. Remsing, R.C., Hernandez, G., Swatloski, R.P., Massefski, W.W., Rogers, R.D., Moyna, G., 2008. Solvation of carbohydrates in N, N′-dialkylimidazolium ionic liquids: a multinuclear NMR spectroscopy study. J. Phys. Chem. B 112, 11071–11078. Sidik, D.A.B., Ngadi, N., Amin, N.A.S., 2013. Optimization of lignin production from empty fruit bunch via liquefaction with ionic liquid. Bioresour. Technol. 135, 690–696. Singh, S.K., Dhepe, P.L., 2016. Isolation of lignin by organosolv process from different varieties of rice husk: understanding their physical and chemical properties. Bioresour. Technol. 221, 310–317. Singh, G., 1999. The Malaysian oil palm industry: progress towards environmentally sound and sustainable crop production. Ind. Environ. 22, 45–48. Studer, M.H., DeMartini, J.D., Davis, M.F., Sykes, R.W., Davison, B., Keller, M., Tuskan, G.A., Wyman, C.E., 2011. Lignin content in natural Populus variants affects sugar release. Proc. Natl. Acad. Sci. 108, 6300–6305. Stylianopoulos, T., Poh, M.-Z., Insin, N., Bawendi, M.G., Fukumura, D., Munn Lance, L., Jain, R.K., 2010. Diffusion of particles in the extracellular matrix: the effect of repulsive electrostatic interactions. Biophys. J. 99, 1342–1349.

from biomass with protic ionic liquids. Green Chem. 16, 1114–1119. Akinosho, H., Dumitrache, A., Natzke, J., Muchero, W., Jawdy, S.S., Tuskan, G.A., 2017. Effects of biomass accessibility and klason lignin contents during consolidated bioprocessing in Populus trichocarpa. ACS Sustain. Chem. Eng. 6, 35–44. Ansari, K.B., Gaikar, V.G., 2014. Green hydrotropic extraction technology for delignification of sugarcane bagasse by using alkybenzene sulfonates as hydrotropes. Chem. Eng. Sci. 115, 57–166. Basiron, Y., 2004. Palm oil production through sustainable plantations. Eur. J. Lipid Sci. Tech. 109, 289–295. Belieres, J.-P., Angell, C.A., 2007. Protic ionic liquids: preparation, characterization, and proton free energy level representation. J. Phys. Chem. B 111, 4926–4937. Boeriu, C.G., Bravo, Dominique, 2004. Characterisation of structure-dependent functional properties of lignin with infrared spectroscopy. Ind. Crop. Prod. 20, 205–218. Bogdanov, M.G., Svinyarov, I., 2013. Ionic liquid-supported solid–liquid extraction of bioactive alkaloids. II. Kinetics, modeling and mechanism of glaucine extraction from Glaucium flavum Cr. (Papaveraceae). Sep. Purif. Technol. 103, 279–288. Boonkird, S., Phisalaphong, C., Phisalaphong, M., 2008. Ultrasound-assisted extraction of capsaicinoids from Capsicum frutescens on a lab-and pilot-plant scale. Ultrason. Sonochem. 15, 1075–1079. Casas, A., Oliet, M., Alonso, M., Rodriguez, F., 2012. Dissolution of Pinus radiata and Eucalyptus globulus woods in ionic liquids under microwave radiation: lignin regeneration and characterization. Sep. Purif. Technol. 97, 115–122. Chan, C.-H., Yusoff, R., Ngoh, G.-C., 2014. Modeling and kinetics study of conventional and assisted batch solvent extraction. Chem. Eng. Res. Des. 92, 1169–1186. Correia, F., Roy, D., Goel, K., 2001. Chemistry and delignification kinetics of Canadian industrial hemp. J. Wood Chem. Technol. 21, 97–111. Costa, C.A.E., Pinto, P.C.R., Rodrigues, A.E., 2014. Evaluation of chemical processing impact on E. globulus wood lignin and comparison with bark lignin. Ind. Crop. Prod. 61, 479–491. Dang, V.Q., Nguyen, K.L., 2006. A universal kinetic equation for characterising the fractal nature of delignification of lignocellulosic materials. Cellulose 14, 153. Dapıia, S., Santos, V., Parajo, J., 2002. Study of formic acid as an agent for biomass fractionation. Biomass. Bioenerg. 22, 213–221. Dehghani, M.H., Mostofi, M., Alimohammadi, M., McKay, G., Yetilmezsoy, K., 2016. High-performance removal of toxic phenol by single-walled and multi-walled carbon nanotubes: kinetics, adsorption, mechanism and optimization studies. J. Ind. Eng. Chem. 35, 63–74. Dodd, A.P., Kadla, J.F., Straus, S.K., 2014. Characterization of fractions obtained from two industrial softwood kraft lignins. ACS Sustain. Chem. Eng. 3, 103–110. Du, F.-Y., Xiao, X.-H., Luo, X.-J., Li, G.-K., 2009. Application of ionic liquids in the microwave-assisted extraction of polyphenolic compounds from medicinal plants. Talanta 78, 1177–1184. Epelde, I.G., Lindgren, C., Lindström, M.E., 1998. Kinetics of wheat straw delignification in soda and kraft pulping. J. Wood Chem. Technol. 18, 69–82. d’Alessandro, L.G., 2012. Ultrasound assisted extraction of polyphenols from black chokeberry. Sep. Purif. Technol. 93, 42–47. Esteves Costa, C.A., Coleman, W., Dube, M., Rodrigues, A.E., Rodrigues Pinto, P.C., 2016. Assessment of key features of lignin from lignocellulosic crops Stalks and roots of corn, cotton, sugarcane, and tobacco. Ind. Crops Prod. 92, 136–148. Fu, D., Mazza, G., 2011. Optimization of processing conditions for the pretreatment of wheat straw using aqueous ionic liquid. Bioresour. Technol. 102, 8003–8010. George, A., Brandt, A., Tran, K., Zahari, S.M.N.S., 2015. Design of low-cost ionic liquids for lignocellulosic biomass pretreatment. Green Chem. 17, 1728–1734. Ghasemi, M., Naushad, M., Ghasemi, N., Khosravi-fard, Y., 2014. A novel agricultural waste based adsorbent for the removal of Pb(II) from aqueous solution: kinetics, equilibrium and thermodynamic studies. J. Ind. Eng. Chem. 20, 454–461. Glas, D., Van Doorslaer, C., Depuydt, D., Liebner, F., Rosenau, T., Binnemans, K., De Vos, D.E., 2014. Lignin solubility in non-imidazolium ionic liquids. J. Chem. Technol. Biotechnol. 90, 21–27. Goto, M., Smith, J., McCoy, B.J., 1990. Kinetics and mass transfer for supercritical fluid extraction of wood. Ind. Eng. Chem. Res. 29, 282–289. Grabber, J.H., Mertens, D.R., Kim, H., Funk, C., Lu, F., Ralph, J., 2009. Cell wall fermentation kinetics are impacted more by lignin content and ferulate cross-linking than by lignin composition. J. Sci. Food Agric. 89, 122–129. Guo, X., Zou, X., Sun, M., 2010. Optimization of extraction process by response surface methodology and preliminary characterization of polysaccharides from Phellinus igniarius. Carbohydr. Polym. 80, 344–349. Hamzah, F., Idris, A., Shuan, T.K., 2011. Preliminary study on enzymatic hydrolysis of treated oil palm, empty fruit bunches fibre (EFB) by using combination of cellulase and β1-4 glucosidase. Biomass.Bioenerg 35, 1055–1059. Hart, W.E.S., Harper, J.B., Aldous, L., 2015. The effect of changing the components of an ionic liquid upon the solubility of lignin. Green Chem. 17, 214–218. Ho, C.-L., Wu, K.-T., Wang, E.I.C., Su, Y.-C., 2011. Delignification kinetics study on the tetrahydrofurfuryl Alcohol/HCl pulping of rice straw. Ind. Eng. Chem. Res. 50, 3849–3857. Ho, Y.-S., 2006. Review of second-order models for adsorption systems. J. Hazard. Mater. 136, 681–689. Huang, G., Shi, J.X., Langrish, T.A.G., 2007. NH4OH–KOH pulping mechanisms and kinetics of rice straw. Bioresour. Technol. 98, 1218–1223. Kadla, J.F., Kubo, S., 2004. Lignin-based polymer blends: analysis of intermolecular interactions in lignin–synthetic polymer blends. Compos. Part A Appl. Sci. Manuf. 35, 395–400. Kitanovic, S., Milenovic, D., Veljkovic, V.B., 2008. Empirical kinetic models for the resinoid extraction from aerial parts of St. John’s wort (Hypericum perforatum L.). Biochem. Eng. J. 41, 1–11. Korotkova, E., Pranovich, A., Wärna, J., Salmi, T., Murzin, D.Y., Willför, S., 2015. Lignin

135


T. Rashid et al.

Tu, Q., Fu, S., Zhan, H., Chai, X., Lucia, L.A., 2008. Kinetic modeling of formic acid pulping of bagasse. J. Agric. Food Chem. 56, 3097–3101. Wahab, M.A., Jellali, S., Jedidi, N., 2010. Ammonium biosorption onto sawdust FTIR analysis, kinetics and adsorption isotherms modeling. Bioresour. Technol. 101, 5070–5075. Wang, G., Chen, H., 2016. Enhanced lignin extraction process from steam exploded corn stalk. Sep. Purif. Technol. 157, 93–101. Yiwen, H.G.S., 1986. Studies on the mechanism and the kinetics of rice straw pulping. J. China Pulp Paper 3, 10–17. Zhang, Y.-H.P., 2008. Reviving the carbohydrate economy via multi-product lignocellulose biorefineries. J. Ind. Microbiol. Biotechnol. 35, 367–375. Zhao, X., Liu, D., 2013. Kinetic modeling and mechanisms of acid-catalyzed delignification of sugarcane bagasse by aqueous acetic acid. Bioenerg. Res. 6, 436–447. Zhou, S., Liu, L., Wang, B., Xu, F., Sun, R., 2012. Microwave-enhanced extraction of lignin from birch in formic acid: structural characterization and antioxidant activity study. Process Biochem. 47, 1799–1806.

Sun, Y., Cheng, J., 2002. Hydrolysis of lignocellulosic materials for ethanol production: a review. Bioresour. Technol. 83, 1–11. Sun, N., Rahman, M., Qin, Y., Maxim, M.L., Rodriguez, H., Rogers, R.D., 2009. Complete dissolution and partial delignification of wood in the ionic liquid 1-ethyl-3-methylimidazolium acetate. Green Chem. 11, 646–655. Tan, S.S.Y., MacFarlane, D.R., Upfal, J., Edye, L.A., Doherty, W.O.S., Patti, A.F., Pringle, J.M., Scott, J.L., 2009. Extraction of lignin from lignocellulose at atmospheric pressure using alkylbenzenesulfonate ionic liquid. Green Chem. 11, 339–345. Tan, H.T., Lee, K.T., Mohamed, A.R., 2011. Pretreatment of lignocellulosic palm biomass using a solvent-ionic liquid [BMIM]Cl for glucose recovery: an optimisation study using response surface methodology. Carbohydr. Polym. 83, 1862–1868. Tejado, A., Peña, C., Labidi, J., Echeverria, J.M., Mondragon, I., 2007. Physico-chemical characterization of lignins from different sources for use in phenol–formaldehyde resin synthesis. Bioresour. Technol. 98, 1655–1663. Teramoto, Y., Lee, S.-H., Endo, T., 2009. Cost reduction and feedstock diversity for sulfuric acid-free ethanol cooking of lignocellulosic biomass as a pretreatment to enzymatic saccharification. Bioresour. Technol. 100, 4783–4789.

136