Bioresource Technology 135 (2013) 116119

Contents lists available at SciVerse ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Short Communication

Comparative study of various pretreatment reagents on rice husk and structural changes assessment of the optimized pretreated rice husk
Teck Nam Ang, Gek Cheng Ngoh , Adeline Seak May Chua
Department of Chemical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia

h i g h l i g h t s
" Among the pretreatment reagents, HCl hydrolyzed rice husk (RH) the best. " The optimized pretreatment condition is mild compared with other similar studies. " The increased pore volume & size of pretreated RH favors fungal fermentation.

a r t i c l e

i n f o

a b s t r a c t
The performance of alkalis (NaOH and Ca(OH)2) and acids (H2SO4, HCl, H3PO4, CH3COOH, and HNO3) in the pretreatment of rice husk was screened, and a suitable reagent was assessed for subsequent optimization using response surface methodology. From the assessment, HCl that hydrolysed rice husk well was optimized with three parameters (HCl loading, pretreatment duration, and temperature) using Box Behnken Design. The optimized condition (0.5% (w/v) HCl loading, 125 C, 1.5 h) is relatively mild, and resulted in 22.3 mg TRS/ml hydrolysate. The reduced model developed has good predictability, where the predicted and experimental results differ by only 2%. The comprehensive structural characterization studies that involved FT-IR, XRD, SEM, and BET surface area determination showed that the pretreated rice husk consisted mainly of cellulose and lignin. Compared to untreated rice husk, pretreated rice husk possessed increased pore size and pore volume, which are expected to be benecial for fungal growth during fermentation. 2012 Elsevier Ltd. All rights reserved.

Article history: Available online 24 September 2012 Keywords: Response surface methodology Optimization Rice hulls Pretreatment Structural assessment

1. Introduction Rice husk is a by-product of rice milling industry, and it represents approximately 20% by weight of rough rice (Hashim et al., 1996). In 2008, there were estimated 125 Mt and 137 Mt rice husk generated in Asia and worldwide, respectively (FAOSTAT, 2010). Rice husk contains relatively high cellulose content (4060%), and it is widely available at relatively low cost. Attributing to its recalcitrant nature, direct conversion of untreated rice husk usually results in low product yield. Therefore, pretreatment is necessary to partially disrupt the recalcitrant structure to achieve delignication of the lignocellulosic biomass. This renders the solid substrate more accessible to enzyme or microorganism during bioconversion. Prevailing rice husk pretreatments reported in literatures include acid (Dagnino et al., 2013), alkaline (Saha and Cotta, 2008; Singh et al., 2011), hydrothermal (Zheng et al., 2007), alkaline peroxide (Saha and Cotta, 2007), and ionic liquid dissolution pretreatments (Ang et al., 2011; Lynam et al., 2012). Among them, acid and
Corresponding author. Tel.: +60 3 7967 5286; fax: +60 3 7967 5371.
E-mail address: ngoh@um.edu.my (G.C. Ngoh). 0960-8524/$ - see front matter 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2012.09.045

alkaline pretreatments have been extensively used for pretreating lignocellulosic biomass (Kaar and Holtzapple, 2000; Saha and Cotta, 2008). To date, no specically effective reagent in pretreating rice husk has been reported. To have a greater insight into the pretreatment of rice husk, alkalis and acids were screened and assessed in the study, and the best performing reagent was subjected to subsequent optimization study. Various structural analyses including Fourier transforminfrared spectroscopy (FT-IR), X-ray diffraction (XRD), scanning electron microscopy (SEM), and Brunauer EmmettTeller (BET) surface area determination were conducted to assess the compositional, chemical, and structural changes of the biomass that is imparted by the selected pretreatment.

2. Methodology 2.1. Materials Rice husk was collected from Ng Trading Company, Selangor, Malaysia. The sample was washed and dried at 55 C before being

T.N. Ang et al. / Bioresource Technology 135 (2013) 116119

117

milled to approximately 30 mesh size, and was stored in a dry cabinet prior to use. 2.2. Assessment of pretreatment reagents Screening was conducted on two alkalis and ve acids, namely sodium hydroxide (NaOH, Merck), calcium hydroxide (Ca(OH)2, SigmaAldrich), sulphuric acid (H2SO4, Fisher Scientic), hydrochloric acid (HCl, Merck), phosphoric acid (H3PO4, Ajax Chemicals), acetic acid (CH3COOH, Riedelde Haen) and nitric acid (HNO3, Scharlau Chemie). With reference to the literatures (Chang et al., 1997; Saha and Cotta, 2008), the screening were carried out at the following conditions: (i) concentration of reagent, 0.5% (w/v); (ii) rice husk loading, 10.0% (w/v); (iii) water loading, 10.0 ml/g; (iv) pretreatment temperature, 100 1 C; (v) pretreatment duration, 2 h. Total reducing sugars (TRS) content was employed as an indirect pretreatment indicator, and its content in hydrolysate was determined by DNS method (Miller, 1959). The main action of acids and alkalis is to dissolve hemicellulose, to some extent cellulose, and lignin. Thus, TRS released from the hydrolysis of hemicellulose/cellulose can suitably reect the extent of structure disruption in rice husk. Pretreatment reagent that releases the highest TRS in hydrolysate was selected for subsequent optimization studies. To investigate the compositional changes imparted by the pretreatment reagents, cellulose, hemicellulose, lignin, and ash contents of the pretreated rice husk samples were characterized by using the Association Ofcial Analytical Chemists (AOAC, (2005)) ofcial methods. 2.3. Optimization of pretreatment The pretreatment reagent selected in Section 2.2 was employed in the optimization of rice husk pretreatment. Three parameters known to have effect on pretreatment, namely reagent loading (X1), heating duration (X2) and heating temperature (X3), were investigated. To determine the low and high levels of the chosen parameters, preliminary tests were conducted at: (i) reagent loading, 04% (w/v); (ii) heating duration, 16 h; (iii) heating temperature, 60140 C. The tests were performed by one-factor-at-a-time approach, and TRS detected in hydrolysate was measured as response for optimization of pretreatment. The range of each parameter determined from the preliminary tests was applied in the optimization study using BoxBehnken experimental design (BBD). TRS detected in the hydrolysate was determined as response (Y). The maximum release of TRS in hydrolysate was determined using response surface methodology (RSM), and the regression analysis of optimization data was performed with the aid of Design-Expert Version 6.0.6 (Stat-Ease Inc., Minneapolis). 2.4. Analytical techniques The FT-IR spectra of the rice husk samples between 600 and 4000 cm1 at 4 cm1 nominal resolution were recorded at room temperature with a FT-IR/FTFIR spectrometer (Perkin Elmer, Spectrum 400, USA). The spectra were presented in relative transmittance percentage (%) of wave number (cm1) and the background was recorded with empty cell. The crystallinity of the rice husk samples was examined by XRD measurement performed with a D8 Advanced X-ray diffractometer (Bruker AXS, USA) using Cu Ka monochromatized radiation at 40 kV and 40 mA at ambient temperature. The samples were scanned and the intensities were recorded in 2h range from 10 to 80 with a step size of 0.02. The crystallinity index (CrI) of the rice husk samples was calculated by using equation as reported by Parikh et al. (2007).

The structural changes of the pretreated rice husk were assessed with scanning electron microscope Quanta 200 FESEM (FEI, USA) operated at 25 kV accelerating voltage under low vacuum. The surface area and average pore size of rice husk samples were determined by nitrogen adsorption isotherm at 77 K using a high-performance six-sample surface area and pore size analyzer Autosorb-6B (Quantachrome, Florida, USA). The nitrogen adsorptiondesorption isotherm was operated at relative pressure P/P0 of 0.3, where P is the system pressure and P0 is the initial pressure at 1 bar. 3. Results and discussion 3.1. Assessment of pretreatment reagent Based on the results of screening for suitable pretreatment reagent, each alkali and acid had pretreated rice husk to a varying extent. Low amount of TRS was detected in the hydrolysates of NaOH and Ca(OH)2 pretreatments. This is because the alkalis only fractionally hydrolysed hemicellulose and cellulose (Weil et al., 1994), but mainly delignied rice husk (Brannvall, 2004). Generally, ndings from the assessment show that acids were better pretreatment reagents than alkalis (Fig. A1). The highest TRS was detected in the hydrolysate of HCl pretreatment (15.0 0.6 mg/ml), followed by HNO3 (12.2 0.1 mg/ml) and H2SO4 (7.2 0.3 mg/ml). Pretreatment with CH3COOH, H3PO4, NaOH, and Ca(OH)2 produced less than 1 mg TRS/ml hydrolysate, which is similar to pretreatment using only water. The characterization of rice husk samples showed that all acidpretreated rice husk had reduced hemicellulose content (Fig. 1), which explains the acids main role were hydrolysing the amorphous hemicellulose in the substrate (Orozco et al., 2007). Besides, cellulose was also partially hydrolysed during the acid pretreatments of rice husk (Weil et al., 1994), particularly pretreatments with strong acids, such as HCl, HNO3 and H2SO4. The availability of more reactive protons disrupt hydrogen bonding of cellulose chain prior to hydrolysis resulting in higher TRS yield (Orozco et al., 2007). Among the reagents, HCl hydrolysed rice husk and released the highest amount of TRS during pretreatment. This signies the effectiveness of HCl pretreatment in disrupting rice husk structure, while retaining signicant amount of cellulose (60%) in pretreated rice husk. Thus, HCl pretreatment was further optimized. 3.2. Optimization of pretreatment The preliminary pretreatment study shows that TRS yield increased sharply in the rst 2 h of pretreatment and reached plateau after 3 h (Fig. A2a). In HCl loading study, TRS yield increased sharply with HCl loading range between 0.25 and 0.75% (w/v) as more protons are available for the hydrolysis of rice husk (Fig. A2b). However, little increment in TRS yield was observed with HCl loading higher than 1.0% (w/v) signies that the rate of hydrolysis is limited by the surface area of rice husk available for reaction. Furthermore, the TRS yield was found to increase proportionally with the pretreatment temperature (Fig. A2c). The low and high levels of HCl loading (X1), pretreatment duration (X2), and temperature (X3) were determined in the preliminary studies. The design matrix of BBD including the response (Y) is given in Table 1, where Y is the TRS detected in the hydrolysate. From the runs, the highest TRS (23.9 mg/ml) was obtained with pretreatment conditions at 0.75% (w/v) HCl loading, 120 C for 2 h. The least TRS (10.9 mg/ml) was detected when pretreatment was conducted at 0.25% (w/v) HCl loading, 100 C for 2 h. Only minute

118

T.N. Ang et al. / Bioresource Technology 135 (2013) 116119

Fig. 1. Chemical composition of rice husk after acid and alkaline pretreatments (based on same initial sample weight).

amount of furfural (5% of TRS) was detected in the hydrolysates, whereas 5-(hydroxymethyl)furfural was not detected. The Analysis of Variance (ANOVA) of the reduced quadratic model conrmed the signicance and goodness of t of the model (Table 2). The reduced model expressed in actual terms is given in Eq. (1). The reduced model explained 95.2% of the variability in the optimization pretreatment of rice husk.

Y 156:32 49:74X 1 9:89X 2 2:36X 3 7:39X 2 1 0:008X 2 3 0:31X 1 X 3 0:08X 2 X 3 1

Table 1 Design matrix of BBD and response. Run Parameter X1 (%, w/v) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 0.25 0.25 1.25 1.25 0.25 0.25 1.25 1.25 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 X2 (h) 1 3 1 3 2 2 2 2 1 1 3 3 2 2 2 2 2 X3 (C) 120 120 120 120 100 140 100 140 100 140 100 140 120 120 120 120 120 Response Y (mg/ml) 18.3 21.6 21.5 22.3 10.9 23.6 19.0 19.5 16.1 23.5 19.1 20.2 23.9 23.3 23.1 23.1 22.2

The optimization study shows that HCl loading and pretreatment temperature have greater effect compared to the pretreatment duration, and both these inuencing parameters interacted in the pretreatment of rice husk as shown in Fig. 2. Higher temperature is needed to achieve the maximal TRS yield when lower HCl loading is used, and vice versa. The TRS yield showed a quadratic dependence on the pretreatment temperature with high effect of term X32. Numerical optimization was conducted for the maximization of TRS release, and the optimum pretreatment conditions were 0.5% (w/v) HCl loading, 125 C for 1.5 h with a predicted TRS yield of 22.8 mg/ml. Verication of the optimized pretreatment condition conrmed the TRS released at 22.3 0.3 mg/ml, which is a mere 2% discrepancy from the predicted value that further suggests the accuracy of the reduced model.
Table 2 Analysis of variance (ANOVA) of quadratic model for optimization of rice husk pretreatment. Source Model X1 X2 X3 X12 X32 X1X3 X2X3 Residual Lack of t Pure error Corrected total Sum of square 173.04 7.88 1.85 59.46 14.43 39.43 37.33 9.83 8.75 7.28 1.47 181.79 Degree of freedom 7 1 1 1 1 1 1 1 9 3 4 16 Mean square 24.72 7.88 1.85 59.46 14.43 39.43 37.33 9.83 0.97 1.46 0.37 F-value 25.44 8.11 1.91 61.19 14.85 40.58 38.42 10.11 3.96 P-value <0.0001 0.0192 0.2007 <0.0001 0.0039 0.0001 0.0002 0.0112 0.1034

T.N. Ang et al. / Bioresource Technology 135 (2013) 116119

119

4. Conclusion HCl has been found to be more superior to the other investigated reagents, and its subsequent HCl pretreatment optimization gives relatively mild optimal condition involving low HCl loading, low pretreatment temperature and short duration. The reduced model exhibits 2% discrepancy between the predicted and experimental values that suggests its accurate predictability. Structural characterization analyses showed that the pretreated rice husk, which consisted mainly of cellulose and lignin, had increased in pore size and pore volume, which is favorable for the attachment and growth of fungus during fermentation. Acknowledgement The authors thank University of Malaya for the nancial support (UMRG RG006/09AET and PS161/2010A). The authors also thank Ng Trading Company, Selangor, Malaysia for providing the rice husk sample. Appendix A. Supplementary data
Fig. 2. Response surface plot of HCl loading and pretreatment temperature on TRS yield (pretreatment duration = 2 h).

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biortech.2012. 09.045. References

3.3. Characterization of pretreated rice husk FT-IR investigation revealed that the pretreated rice husk exhibits enhanced cellulosic content as shown by the intensied bands at 1046 and 2924 cm1 (Fig. A4). Besides, cellulose portion of the pretreated rice husk has more disordered structure, and lignin was also observed in the HCl pretreated rice husk. These ndings are complemented by chemical composition analysis of the rice husk samples. The pretreated rice husk consists of mainly cellulose (65.1%, w/w) and lignin (26.7%, w/w) which can be seen from the increased intensity of relative transmittance of cellulose- and lignin-related bands in the FT-IR spectrum, whereas hemicellulose that is easily hydrolysed was not detected. Pretreated rice husk was slightly more crystalline than the untreated rice husk after HCl pretreatment. The increased in the CrI from 51.5 of the untreated rice husk to 56.3 of the pretreated rice husk is due to the removal of amorphous hemicellulose and cellulose from rice husk by HCl during pretreatment. The evidence for the removal is conrmed by the SEM micrographs, whereby at 1500 magnication, pretreated rice husk exhibited a smoother surface than the untreated rice husk (Fig. A5). Furthermore, the BET surface area analysis demonstrated that the specic surface area of the pretreated rice husk was signicantly reduced from 21.2 to 9.6 m2/g, which is due to the hydrolysis of amorphous hemicellulose/cellulose in rice husk and merging of smaller pores into larger pores. The total pore volume and pore size of the rice husk after pretreatment had increased from 1.76 102 to 1.81 102 cc/g and 33.2 to 75.2 , respectively. The pretreated rice husk with higher pore volume and bigger pore size is of advantageous for fungal growth during fermentation (Hsu et al., 2010), in spite of its slightly higher CrI and lower specic surface area compared to the untreated rice husk. Nevertheless, the underlying relationship of structural features and biomass digestibility of the pretreated substrate depends on the size of enzyme used in the enzymatic hydrolysis, and the accessibility of microbes to substrate during fermentation.

Ang, T.N., Yoon, L.W., Lee, K.M., Ngoh, G.C., Chua, A.S.M., Lee, M.G., 2011. Efency of ionic liquids in the dissolution of rice husk. BioResources 6 (4), 47904800. AOAC, 2005. Ofcial Methods of Analysis of the Association of Ofcial Analytical Chemists. The Association Ofcial Analytical Chemists, Washington. Brannvall, E., 2004. Pulping technology. In: Ek, M., Gellerstedt, G., Henriksson, G. (Eds.), Pulp and Paper Chemistry and Technology, vol. 2. Walter de Gruyter GmbH & Co, Berlin. Chang, V.S., Burr, B., Holtzapple, M.T., 1997. Lime pretreatment of switchgrass. Appl. Biochem. Biotechnol. 6365, 319. Dagnino, E.P., Chamorro, E.R., Romano, S.D., Felissia, F.E., Area, M.C., 2013. Optimization of the acid pretreatment of rice hulls to obtain fermentable sugars for bioethanol production. Ind. Crop. Prod. 42, 363368. FAOSTAT, 2010. Food and Agricultural Commodities Production, vol. 2010, Food and Agriculture Organization of the United Nations. Hashim, A.B., Aminuddin, H., Siva, K.B., 1996. Nutrient content in rice husk ash of some Malaysian rice varieties. Pertanika J. Trop. Agric. Sci. 19 (1), 7780. Hsu, T.C., Guo, G.L., Chen, W.H., Hwang, W.S., 2010. Effect of dilute acid pretreatment of rice straw on structural properties and enzymatic hydrolysis. Bioresource Technol. 101, 49074913. Kaar, W.E., Holtzapple, M.T., 2000. Using lime pretreatment to facilitate the enzymic hydrolysis of corn stover. Biomass Bioenerg. 18, 189199. Lynam, J.G., Reza, M.T., Vasquez, V.R., Coronella, C.J., 2012. Pretreatment of rice hulls by ionic liquid dissolution. Bioresource Technol. 114, 629636. Miller, G.L., 1959. Use of dinitrosalicylic acid reagent for determination of reducing sugar. Anal. Chem. 31, 426428. Orozco, A., Ahmad, M., Rooney, D., Walker, G., 2007. Dilute acid hydrolysis of cellulose and cellulosic bio-waste using a microwave reactor system. Process Safety and Environmental Protection, Trans. I. Chem. E. Part B 85 (B5), 446449. Parikh, D.V., Thibodeaux, D.P., Condon, B., 2007. X-ray crystallinity of bleached and crosslinked cottons. Textil. Res. J. 77 (8), 612616. Saha, B.C., Cotta, M.A., 2007. Enzymatic saccharication and fermentation of alkaline peroxide pretreated rice hulls to ethanol. Enzyme Microb. Tech. 41, 528532. Saha, B.C., Cotta, M.A., 2008. Lime pretreatment, enzymatic saccharication and fermentation of rice hulls to ethanol. Biomass Bioenerg. 32, 971977. Singh, A., Tuteja, S., Singh, N., Bishnoi, N.R., 2011. Enhanced saccharication of rice straw and hull by microwave-alkali pretreatment and lignocellulolytic enzyme production. Bioresource Technol. 102, 17731782. Weil, J., Westgate, P., Kohlmann, K., Ladisch, M.R., 1994. Cellulose pretreatments of lignocellulosic substrates. Enzyme Microb. Tech. 16, 10021004. Zheng, G.J., Zhou, Y.J., Zhang, J., Cheng, K.K., Zhao, X.B., Zhang, T., Liu, D.H., 2007. Pretreatment of rice hulls for cellulase production by solid substrate fermentation. J. Wood. Chem. Tech. 27, 6571.