Subcritical water hydrolysis of spent Java Citronella biomass for production of reducing sugar

Available online at www.sciencedirect.com

ScienceDirect Materials Today: Proceedings 5 (2018) 23128–23135

www.materialstoday.com/proceedings

ICAER-2015

Subcritical water hydrolysis of spent Java Citronella biomass for production of reducing sugar Robinson Timunga, Vaibhav V. Gouda, * F0F

a

Department of Chemical Engineering, IIT Guwahati, Guwahati-781039, India

Abstract The present study investigate the production of reducing sugars from spent Java citronella (Cymbopogon winterianus Jowitt) biomass using subcritical water (SCW) treatment in a batch reactor. The effect of process parameters such as temperature (140220 °C) and reaction time (5-40 min) on total reducing sugar (TRS) yield was studied. The production of TRS increased with an increase in the temperature from 140-160 °C and decreased thereafter. The maximum amount of TRS 132.09 mg.g-1 biomass was obtained at 160°C and 30 min reaction time. The formation of sugar inhibitors (5-HMF or Furfural) was almost insignificant upto 160°C, but increased subsequently with an increase in the reaction temperature and time. The maximum inhibitors concentration of 16.70 mg.g-1 was obtained at 220°C and 30 min reaction time, which includes 5.77 mg.g-1 of 5-HMF and 10.93 mg.g-1 of furfural. The crystallinity index (CI) value of biomass sample increased from 35.30% to 52.68% with an increase in the temperature from 140-180°C, further increased in temperature to 220°C decreases the CI abruptly to 20.15% thereby altering the biomass to relatively amorphous. The porous external surface as observed in scanning electron microscopic analysis of the sample after SCW treatment showed significant removal of sugars and lignin. © 2018 Elsevier Ltd. All rights reserved. Selection and Peer-review under responsibility of the Conference Committee Members of International Conference on Advances in Energy Research 2015 (ICAER-2015). Keywords: Subcritical water; Java citronella; Reducing sugars; Inhibitors; X-ray diffraction.

1. Introduction Biofuels derived from lignocellulosic biomass being sustainable fuels are potential feedstock for bioenergy production to meet the increasing energy demand. There has been worldwide extensive research on the utilization of

* Corresponding author. E-mail address: [email protected] 2214-7853 © 2018 Elsevier Ltd. All rights reserved. Selection and Peer-review under responsibility of the Conference Committee Members of International Conference on Advances in Energy Research 2015 (ICAER-2015).

Timung and Goud / Materials Today: Proceedings 5 (2018) 23128–23135

23129

lignocellulosic biomass for the production of biofuels such as bioethanol, biodiesel and biogas, alongside reducing greenhouse-gas emissions [1]. Additionally aromatic spent biomasses such as citronella, mentha etc. are the most abundant and underutilized biological resources due to its cooked condition during essential oil extraction (mostly distillation), and also a non-cattle feed. The spent biomass of aromatic plants are generally burnt haphazardly or disposed as waste, leading to environmental problems. India alone produces more than 6.0 million tons per annum of aromatic spent biomass [2]. Lignocellulosic biomass are composed of structural polysaccharides; cellulose and hemicellulose wrapped inside lignin. The production of fuels from biomass requires pretreatment for conversion of holocellulose to their constituent sugars and removal of lignin prior to hydrolysis and fermentation. Pre-treatment is necessary before hydrolysis in order to provide easy access for enzymes to convert cellulose to chains of glucose and its isomers, and hemicellulose to xylose and arabinose [3]. The efficiency of conversion of biomass to reducing sugars depends on the choice of biomass and pretreatment method employed. An ideal biomass for the production of bioenergy should possess high holocellulose content and calorific value [4]. Different pretreatment methods such as chemical (acid, alkali, organosolv), physical (ultrasound, milling), physico-chemical (steam explosion, AFEX, liquid hot water) and biological (using microorganisms such as fungi and bacteria) techniques have been practiced since decades [5], and in this work subcritical water (SCW) extraction technique was investigated. SCW is defined as hot water in a liquid state at temperature between 100°C to 374°C maintained under high pressure. SCW hydrolysis technique is an efficient and one step process for converting biomass to reducing sugars. SCW technology is considered as a green technique and provides extensive advantages over conventional methods. It is a one-step auto hydrolysis process with less reaction time, non-toxic, non-corrosive etc. and provides wide range of dielectric constant, density and viscosity to extract desired polar and non-polar compounds by tuning the operational temperature and pressure. The water at subcritical condition targets the cellulose and hemicellulose fractions of biomass to solubilize and liberate hexose and pentose sugars, while disrupting lignin and crystalline structure of cellulose [6]. However the hydrolysis rate and sugar yield depends on the biomass cell structure and composition. SCW extraction has gained significant interest towards extraction of saccharides from different feedstocks such as cellulose, oil palm fronds, coconut meal, bamboo, sugarcane bagasse, cotton stalk, tobacco stalk, wheat stalk and sunflower stalk etc. [7-11]. In the SCW treatment of biomass, polysaccharides degrades to oligosaccharides (Degree of polymerization (DP) <10) or monosaccharides, and monosaccharides decompose to degradation products depending on the operational temperature and reaction time [12]. This work is an endeavour to evaluate the importance of spent biomass of aromatic plants. Spent citronella biomass has been chosen due to its abundance as raw material in the North Eastern India. The present work aims at optimizing the process parameters of SCW hydrolysis of spent citronella biomass to produce maximum reducing sugars. The physical and chemical changes in the biomass sample after SCW treatment were evaluated. The results obtained in this study will also helps in understanding the potential usage of spent aromatic biomass as a feedstock for TRS production. 2. Materials and Methods 2.1. Materials Spent Java citronella biomass (after citronella oil extraction) obtained from Karbi Anglong, Assam (India) were initially dried, chopped into small pieces, grinded, sieved to 1 mm sizes using 16 BSS mesh screen and then stored in zipped plastic bags at ambient temperature until used. High pressure bath reactor (Model: 1734, M/s Amar equipment Pvt. Ltd., India) equipped with stirrer and reactor vessel (500 ml) made of stainless steel with 160 mm height and 75 mm internal diameter was used. The reactor was equipped with pressure gauge and thermocouple. The standard sugars such as Maltohexaose, Maltopentaose, Raffinose, Cellobiose, Glucose, Xylose, Arabinose and Erythrose, as well as HMF and furfural (all ≥ 95% purity) were obtained from M/s Himedia and M/s Sigma Aldrich. Nitrogen gas with 98% purity was purchased from Guwahati, Assam. 2.2. Methodology Hydrolysis of biomass was carried out in a 500 ml capacity high pressure batch reactor (maximum working temperature and pressure of 500°C and 350 kg.cm-1respectively) using the procedure as adapted by Mohan et al.,

23130

Timung and Goud/ Materials Today: Proceedings 5 (2018) 23128–23135

(2015) [7], at various temperature ranging from 140-220°C, biomass loading (3 wt%) and reaction time 5-40 min. The nitrogen gas was used to maintain the required pressure such as 52, 89, 145, 225, 335 psi or slightly more for 140, 160, 180, 200 and 220 °C respectively, to keep water at liquid state and also to prevent unnecessary reaction in the process. Before achieving a set temperature, stirring was performed to disperse the biomass uniformly in the water. The extraction time was considered from the moment desired reaction temperature was achieved. The hydrolysate samples were collected through sample outlet at subsequent 5 min interval. At the completion of reaction, the whole system was cooled to room temperature; residual biomass was collected, dried at 65°C for 48 hr and then characterized using X-ray diffraction (XRD), Fourier transformed infrared spectroscopy (FTIR) and Scanning electron microscope (SEM). The collected liquid samples were filtered using 0.2 µm nylon filter paper and analysed for sugar content in High Performance Liquid Chromatography (HPLC). The experiments at specified conditions were repeated twice and the average values were reported. At the end of every experiment, the whole system was flashed using millipore water at approximately 220 psi to prevent choking of sample port and gas inlet port by residual biomass. 2.2.1. HPLC analysis of hydrolysate The hydrolysate were analysed for the estimation of sugars such as maltohexaose, maltopentaose, raffinose, cellobiose, glucose, xylose, arabinose etc. and inhibitors such as 5-HMF and furfural using HPLC (Perkin Elmer Series 200) equipped with RI detector, vacuum degasser and network chromatography interface (NCI) 900. SUPECOGEL Ca column (dimensions, 300 mm length and 7.8 mm inner diameter) connected to Ca* guard column purchased from Sigma-Aldrich was used. The system was operated at 80°C using a column oven. Milli-Q water was used as mobile phase at flow rate of 0.8 ml.min-1 and retention time of 30 min. The peaks obtained were identified by comparing the retention time of standard sugars and quantified using the calibration plots standardized with at least 5 points of different concentrations. 2.2.2. X-ray diffraction (XRD) analysis The biomass crystallinity was estimated using wide angle X-ray diffractometer (Bruker D8 Advanced X-ray diffraction measurement systems) at the radiation supplied by the accelerated current of 40 mA and voltage of 40 kV. The diffraction angle (2θ) scan range was 10-28° at a scan speed 1°.min-1 and step size of 0.05° [7]. Crystallinity Index (CI) was calculated based on the intensity of amorphous region at ~18.5±0.05° of 2θ (Iamp) and the maximum intensity obtained from crystalline fraction at ~ 22.2±0.1°of 2θ (I002) according to Eq. (1). I -I (1) C I (% ) = 0 0 2 a m p × 1 0 0 I 002 2.2.3. Fourier Transform Infrared (FTIR) analysis FTIR spectra of biomass samples were obtained from FTIR spectrometer “IR Affinity1” (Shimadzu Corporation, Japan), measured at the range between 4000-400 cm-1 with nominal resolution of 2 cm-1 and 30 scans per spectrum. IR grade potassium bromide (KBr) was used as reference. 20 mg of biomass sample was mixed with 30 mg of spectroscopic grade dried KBr, grinded well for proper mixing and directly taken for the FTIR analysis. The baseline corrections and variations of frequency was applied using Shimadzu IR solution 1.5 software supplied with the equipment [7, 13] 2.2.4. SEM analysis The dry biomass samples were fixed with a conducting adhesive carbon tape on an aluminum sample holder. Sputter gold coating was performed 2–3 times to prevent charging. All the specimens were qualitatively examined in scanning electron microscope (LEO 1430vp, Germany) [7], under vacuum condition at accelerating voltage of 8 kV and 10 kV, and the working distance of 15 mm or 13 mm. The image was taken at the magnification of 1.78 KX.

Timung and Goud / Materials Today: Proceedings 5 (2018) 23128–23135

23131

3. Results and Discussion The spent aromatic biomass with cellulose and hemicelluloses as major components in its compositions was hydrolysed using SCW treatment at varying conditions to produce TRS. Different oligo and mono saccharides such as maltohexaose, maltopentaose, raffinose, cellobiose, glucose, xylose, arabinose and erythrose were released at different reaction conditions. The release of total sugars initially increased with an increase in the hydrolysis time and temperature up to 30 min and 160°C, but decreased further. This may be attributed to the fact that oligosaccharides and monosaccharides degrades much faster than the cellulose or hemicellulose or lignin [14]. Hence before the complete conversion of total sugars from holocellulose, the oligosaccharides or monosaccharides which were released disintegrate to other degradation products [15]. The overall degradation profile can be explained as – polysaccharides (cellulose and hemicellulose) converts into oligosaccharides (maltohexaose, maltopentaose, raffinose, cellobiose etc.) and monosaccharides (glucose, xylose, arabinose etc.), afterwards as temperature increases the oligosaccharides degrades to di, tri or monosaccharides and later monosaccharides degrades to sugar inhibitors (5-HMF, Furfural etc.) with further increase in the reaction temperature and time (Fig.1).

Fig. 1. HPLC chromatogram of hydrolysate samples for different temperatures at 30 min reaction time; A: Maltohexaose, B: Maltopentaose, C: Raffinose, D: Cellobiose, E: Glucose, F: Xylose, G: Arabinose, H: Erythrose, I: 5-HMF, J: Furfural.

23132

Timung and Goud/ Materials Today: Proceedings 5 (2018) 23128–23135

Temperature remains the most influencing factor in the SCW hydrolysis process. The increase in reaction temperature provides higher mass transfer and reduces surface tension of water thereby increases the solubility of the targeted compounds and provides high diffusivity through the sample matrix [8, 16]. The increase in TRS production with an increase in the temperature from 140-160 °C revealed that hydrolysis rate increases with an increase in the temperature. However, with further increase (180°C up to 220°C) in the temperature TRS production decreased, which suggests that the degradation rate of monosaccharides was higher than the conversion rate of saccharides from the lignocellulosic material at higher temperature. Reaction time is another important factor which affects the TRS formation and is dependent on the temperature and matrix of targeted compounds [17]. The effect of reaction time (i.e. 5, 10, 15, 20, 25, 30, 35 and 40 min) on the release of reducing sugars was evaluated. The degradation of polysaccharide to oligosaccharide and monosaccharide at lower temperature increases the production of TRS, but further increase in the reaction time and temperature resulted to decrease in TRS production which might be due to the degradation of saccharides to inhibitors (i.e. 5-HMF and furfural) as can be seen from Table 1. Table 1. . Effect of time and temperature on release of TRS and total inhibitors (TI) Time (min)

140°C

160°C

180°C

200°C

220°C

TRS

TI

TRS

TI

TRS

TI

TRS

TI

TRS

TI

5

68.38

0

70.82

0

80.83

1.47

40.22

4.30

34.43

10.23

10

72.96

0

81.60

0

97.15

2.73

52.76

9.47

38.15

13.20

15

81.84

0

92.93

0

99.34

5.07

71.58

10.97

45.55

13.67

20

89.52

0

105.89

0

108.49

6.07

86.60

11.73

59.87

14.17

25

95.57

0

119.16

0

119.83

7.00

90.68

12.80

82.83

15.43

30

97.99

0

132.09

0

124.97

7.97

89.53

13.10

81.18

16.70

35

95.73

0

108.83

0

123.51

6.83

88.55

11.07

73.59

14.93

40

72.25

0

85.60

0

102.45

6.13

85.75

10.17

66.84

11.47

The maximum TRS (132.09 mg.g-1) was obtained at 160°C and 30 min, which includes 52.24 mg.g-1 of maltohexaose, 0.70 mg.g-1 of maltopentaose, 1.13 mg.g-1 of raffinose, 10.37 mg.g-1 of cellobiose, 26.01 mg.g-1 of glucose, 26.33 mg.g-1 of xylose (maximum release amount of xylose) and 15.33 mg.g-1 of arabinose. The release of glucose increases with temperature producing maximum (27.01 mg.g-1) at 200°C and 30 min. Maximum maltohexaose, maltopentaose, raffinose, cellobiose and arabinose obtained were 54.10 mg.g-1 (at 160°C, 25 min), 5.86 mg.g-1 (at 180°C, 15 min), 7.85 mg.g-1 (at 160°C, 5 min), 25.73 mg.g-1 (at 160°C, 30 min) and 20.67 mg.g-1 (at 160°C, 30 min) respectively. On the other hand, maximum inhibitors (16.70 mg.g-1) were produced at 220°C and 30 min reaction time, which includes 5.77 mg.g-1 of 5-HMF and 10.93 mg.g-1 of furfural. Decreased in the concentration of 5-HMF and furfural after 30 min reaction time suggested that these compounds are not stable up to longer time in the reaction mixture and degrades to levulinic acid and 1-2-4 benzenenitrol [18]. 3.1. XRD Analysis The XRD analysis was performed to estimate the crystallinity index of hydrolysed biomass. The degradation of hemicellulose and amorphous cellulose occurs at lower temperature followed by higher degradation of crystalline cellulose at higher temperature [19]. The XRD pattern and CI (%) of SCW treated sample at different temperature can be observed from Fig. 2. The CI (%) of spent citronella biomass was 34.91%. The CI of SCW treated biomass sample increased from 35.30% to 52.68% with an increase in the temperature from 140-180°C. This may be attributed to the effective removal of certain amount of lignin and hemicellulose from biomass and not necessarily due to changes in the crystalline structure of the spent biomass. Further increase in the temperature decreases the CI, which indicates the disruption of crystalline structure of cellulose. The CI at 200°C and 220°C were 29.57% and 20.15% respectively.

Timung and Goud / Materials Today: Proceedings 5 (2018) 23128–23135

23133

Fig. 2. XRD pattern of SCW treated sample at different temperature.

3.2. FTIR Analysis

Fig. 3. FTIR analysis of Citronella biomass (Untreated biomass, spent biomass and SCW treated biomass at 140°C, 180°C and 220°C).

FTIR of untreated, spent and SCW treated citronella biomass (at 140°C, 180°C and 220°C) was performed to compare the chemical changes in distribution of different functional groups which might have altered due to degradation of the biomass (Fig. 3). There was an intense strong and sharp peak in the range of 3800-3200 cm-1 particularly at 3764.42 in all samples, which mostly represents the polymeric free hydroxyl (O-H) stretch. Another intense and diverged peak observed in the range of 2935-2915 cm-1 corresponds to the aliphatic alkyl group C-H methyl and methylene asymmetric stretch. Terminal aldehydic C-H stretch was found at around 2830 cm-1. A characteristic medium peak was noticed at around 2372.28 cm-1 which may corresponds to certain cyano carbons. Another distinct and sharp peak at 1750-1705 cm-1 in all the samples signifies un-conjugated xylan as well as aldo, keto, estero and or acido (C=O) stretch. A medium peak at 1529.47 cm-1 corresponding to aromatic C=C ring and lignin ring stretch was also observed. On the other hand, methylene C-H stretch (1485-1445 cm-1) and methyl C-H symmetric band in cellulose and hemicellulose (1380-1371 cm-1) is present in all the samples. Moreover CH=CH trans-unsaturated (910-860 cm-1) functional group with sharp peaks are also observed. Boeriu et al., (2004) reported most of the above stretches for lignin, which indicates that lignin is dispersed in the residue [20]. The peak at ~1450

23134

Timung and Goud/ Materials Today: Proceedings 5 (2018) 23128–23135

and ~894 cm-1 represents the crystalline and amorphous fractions of cellulose respectively [13, 21]. Miniscule changes was observed in all the samples except the disappearance of the peaks at 899 cm-1 at 200°C and 220°C revealed the maximum degradation of cellulose fractions to sugars. 3.3. SEM Analysis

Fig. 4. SEM images of Citronella biomass; (a) Untreated (b) Spent biomass and (c) after SCW treatment.

SEM analysis of the samples showed significant morphological changes after subsequent treatment. SEM images of untreated plant material showed smooth surface, and the cells of untreated plant material are intact (Fig. 4 (a)). After the extraction of oil it was observed that the cellular content of plant tissues distorted due to extensive thermal stress on the oil glands causing rapid expansion extending to disruption and releasing the oil as represented in Fig. 4 (b). Comparing the structural changes of untreated sample image with the insoluble fraction of treated sample after SCW treatment, exposed pores can be seen on the surface of hydrolysed biomass as depicted in Fig. 4 (c), revealed de-polymerization of holocellulose and removal of significant amount of lignin [22]. 4. Conclusions In the SCW hydrolysis of spent citronella biomass, the released of oligosaccharides such as maltohexaose, maltopentaose, cellobiose etc. increased with reaction time at lower temperatures (140-160°C), but decreased at higher temperature, because of its disintegration to monosaccharides (glucose, xylose and arabinose). Further increased in the reaction temperature and time degraded the monosaccharides to sugar inhibitors such as 5-HMF, furfural etc. SCW technology which is relatively simple, efficient and environment friendly remains a promising process for the production of TRS from lignocellulosic biomass.

Timung and Goud / Materials Today: Proceedings 5 (2018) 23128–23135

23135

Acknowledgement The work reported in this article was financially supported by a research grant received under Fast Track Scheme from Department of Science and Technology (DST), Government of India (No. SR/FTP/ETA-0018/2010). Authors would like to thank Mr Mood Mohan and Mr Narendra Naik for their help during this work. The authors also wish to thank the Central Instrument Facilities of Indian Institute of Technology Guwahati for providing analytical facilities. References [1] M.A.Rostagno, J.M.Prado, R.Vardanega, M.A.Meireles, Chem. Eng. Trans.37(2014) 403–408. [2] P.K.Rout, A.D.Nannaware, R.Rajeasekharan, (2013) WO2013102911A1. [3] V.B.Agbor, N.Cicek, R.Sparling, A.Berlin, D.B.Levin, Biotechnology Advances. 29(2011) 675-685. [4] M.Carrier, A.Loppinet-Serani, D.Denux, J.M.Lasnier, F.Ham-Pichavant, F.Cansell, C.Aymonier,Biomass & Bioenergy. 35(2011) 298–307. [5] Y.Zheng, Z.Pan, R.Zhang, Int J Agric & Biol Eng. 2(2009) 51-65. [6]G.Zhu, Z.Xiao, X.Zhu, F.Yi, X.Wan,Clean Technologies and Environmental Policy.15(2013) 55–61 [7] M.Mohan, T.Banerjee, V.V.Goud, Bioresource Technology. 191 (2015) 244–252. [8] O.Akpinar, K.Erdogan, S.Bostanci, Carbohydrate Research. 344(2009) 660-666. [9] M.Sasaki, Z.Fang, Y.Fukushima, T.Adschiri, K.Arai, Ind. and Eng.Chem Res. 39 (2000) 2883-2890. [10] R.Norsyabilah, S.Hanim, M.Norsuhaila, A.Noraishah, S.Kartina, Malays J Anal Sci. 17 (2013) 272-275. [11] P.Khuwijitjaru, A.Pokpong, K.Klinchongkon, S.Adachi, Int J Food Sci Technol. 49 (2014) 1946-1952. [12] F.P.Cardenas-Toro, S.C.Alcazar-Alay, T.Forster-Carneiro, M.A.A.Meireles, Food and public health. 4 (2014) 123-139. [13] R.Timung, M.Mohan, B.Chilukoti, S.Sasmal, T.Banerjee, V.V.Goud, Biomass & Bioenergy. 81 (2015) 9-18. [14] M.Khajenoori, A.HaghighiAsl, F.Hormozi, M.H.Eikani, H.Noori, Journal of Food Process Engineering. 32 (2009)804–816. [15] S.M.Zakaria, S.M.M. Kamal, Food Eng Rev. (2015) ISSN: 1866-7910. [16]Y.Zhao, W.J.Lu, H.T.Wang, D.Li, Environmental Science & Technology. 43(2009) 1565–1570. [17]Z.Chao, Y.Ri-fu, Q.Tai-qiu, Separation and Purification energy. 120(2013)141-147. [18] R.Norsyabilah, S.Hanim, M.Norsuhaila, A.Noraishah, S.Kartina, Malays J Anal Sci, 17(2013)272-275. [19]E.Quintana, C.Valls, A.G.Barneto, M.B. Roncero, Carbohydrate polymers. 119 (2015) 53-61. [20] C.G. Boeriu, D. Bravol, R.J.A. Gosselink, J.E.G. Dam, Industrial crops and products. 20 (2004) 205-218. [21] S.Y. Oh, D.I.I. Yoo, Y. Shin, G. Seo, Carbohydrate research. 340 (2005) 417-428. [22] E.C. Giese, M. Pierozzi, K.J. Dussán, A.K.Chandel, S.S. Da Silva, J. Chem. Technol. Biotechnol. 88(2013) 1266–1272.